So, there is a new full-frame CMOS sensor. Interesting. Is Sony the source of that sensor?
Damn, talk about dropping the Hammer. This will be the undisputed king of sensors. No if and or buts. Geezus. The guys at Phase must have been LoL while reading all the speculative posts on this and other forums. I am certainly quite surprised by the news of this new back. Much sooner than I expected.
PS. With the HDMI Output, is the end goal to also offer video (or does it already)?
So, there is a new full-frame CMOS sensor. Interesting. Is Sony the source of that sensor?
Interesting. This is probably my dream sensor. 100MP, CMOS for live view and ability to shoot high ISO's. I'm also a little upset. I just moved to Phase One in June when I ordered the XF and IQ380. Well I did not receive the IQ380 until first week of November (almost 5 months after I had paid for the system) and now they come out with this. I wonder if I could make a good argument to allow me to trade it in.
I'm also interested in learning about the sensor array's physical size, and resulting photo-site pitch. That will determine several aspects of lens suitability and postprocessing (Capture sharpening) needs.
The current documentation only mentions an IQ3 50MP, IQ3 60MP, and IQ3 80MP back. So does this mean that the 100MP comes only as a complete XF camera solution, or will there be an IQ3 100MP back as well?
What is the noise structure at higher ISOs like? I always found the noise on Canon sensors to look quite ugly once you got to ISO 250 ~ 320.
Phase One IQ3 100mp Technical Specifications (https://digitaltransitions.com/xf-100mp-camera-system/#specs)
I would to know how this back responds to lens cast with Rodenstock lens, especially the HR 40mm and HR 32mm.
Long exposures capability is defiantly of interest to me.
How does the color look compared to the CCD counterparts?
Is the base ISO 50, or is it 100 with the option to pull the ISO down to 50?
I already clicked that link, but maybe I need to clean my glasses, because I see no mention of the 100MP
You may need to clear your cache or switch browsers? Or maybe there is an issue on our end!
Anyway, to avoid confusion
IQ3 100mp
40.4 x 53.7mm
11608 x 8708 pixel dimensions
4.6 micron sensor size
As mentioned on the page that Doug linked to, it's a Sony.
Two people have now asked the price, with nary a response....
This has long been one of the things that annoys me about the medium format digital distributorship model. Questions regarding prices or upgrades seem to often be treated as irrelevant ("if you have to ask, you can't afford it") or contingent ("contact your dealer's uncle's cousin Joe, and he'll get you the best price.")
This isn't the Hope Diamond, or heroin.
What is the price?
I go to the gym for half an hour on Sunday morning after waking up early to post content to half a dozen sites, and I don't answer every question before hand and suddenly I'm a diamond thief or drug dealer? :-p
Price is listed above by Michael. Also available as a stand alone back, or as an upgrade from other digital backs (Phase or Competitive).
Now I'm gonna get back to sweating. I'll be back in 30 min :).
I would to know how this back responds to lens cast with Rodenstock lens, especially the HR 40mm and HR 32mm.
Joe, shifting wide angles has been the Achilles heel of MFDB with every back after the P45+. When I tested the 80mp back against the 60mp, I found the 80 to be totally unusable with my wide lenses (the Schneiders). I just couldn't get into the 32 and 40 HRs given how big, heavy and expensive they were combined with the distortion they exhibit.
I don't know, though... I'm really intrigued about this new back and I seem to find myself using smaller and smaller shifts too. It has me wondering about shooting the XF and just cropping for composition instead of shifting.
For that, the 28 would be essential... is it any good? I wanna see some tests and more info about the 16bit pipeline of course, but I'm very intrigued.
Congratulations with what most likely is a very nice camera solution for those who need the high resolution. I'm a bit weary about the '16-bit color' claim, but we'll see once the Raw files become available if that really means a (as advertised) 15-bit ADC pipeline converted to 16-bit/channel output. The new higher MP resolution IQ3 backs are advertised as 13-bit, and the CMOS 50MP as 14-bit ...
I'm also interested in learning about the sensor array's physical size, and resulting photo-site pitch. That will determine several aspects of lens suitability and postprocessing (Capture sharpening) needs.
The current documentation only mentions an IQ3 50MP, IQ3 60MP, and IQ3 80MP back. So does this mean that the 100MP comes only as a complete XF camera solution, or will there be an IQ3 100MP back as well?
Cheers,
Bart
Time to rob a bank, any tutorials?
Chris, come on down to Indy and let's give it a run for the money with your system and mine. I have the 23mm and I am anxious to see how that works. You know how to reach me.
Joe, shifting wide angles has been the Achilles heel of MFDB with every back after the P45+. When I tested the 80mp back against the 60mp, I found the 80 to be totally unusable with my wide lenses (the Schneiders). I just couldn't get into the 32 and 40 HRs given how big, heavy and expensive they were combined with the distortion they exhibit.
I don't know, though... I'm really intrigued about this new back and I seem to find myself using smaller and smaller shifts too. It has me wondering about shooting the XF and just cropping for composition instead of shifting. For that, the 28 would be essential... is it any good? I wanna see some tests and more info about the 16bit pipeline of course, but I'm very intrigued.
CB
.... I think the Ken & Don anti-workshop just became a lot more interesting.....
Congratulations with what most likely is a very nice camera solution for those who need the high resolution. I'm a bit weary about the '16-bit color' claim, but we'll see once the Raw files become available if that really means a (as advertised) 15-bit ADC pipeline converted to 16-bit/channel output.
The price of the Phase One XF 100MP Camera System (with Schneider Kreuznach 80mm LS lens) is 48,990 USD. .
Is the HDMI feed 1080p or 4k/UHD?
and the back alone, without camera/lens combo?
Iq180 back alone tradein will cost you $26,394
I look forward to your analysis! Here are a gig of raws to play with:
Phase One IQ3 100mp raw file download (https://digitaltransitions.com/xf-100mp-camera-system/#raw)
I guess we can expect a more reasonably priced update from Hassy soon, and a $12K announcement from Pentax at Photokina. My pro dealer tells me the 645Z has been selling well, and so Pentax will probably want to catch the new train leaving the station.
Edmund
To be honest I do not understand why we need even more MP when we have such good solutions already.
Wondering out loud, with the Electronic First Curtain, I wonder how the 1/1600th LS sync is going to change/improve.leaf shutters have shutter leaves closing in a manner that they can't use EFCS, where sensor starts exposure row by row with vertical shutter blade following that process with some delay and mechanically shielding the rows for which exposure is over...
I look forward to your analysis! Here are a gig of raws to play with:
Phase One IQ3 100mp raw file download (https://digitaltransitions.com/xf-100mp-camera-system/#raw)
Doug thanks, I'm downloading the updated C1 9.0.2, but the files link is perhaps sent by email (and I couldn't enter a non USA location, if that matters)? I'll probably also have to wait for RawDigger to recognize the new Raw file format/size.
but I suspect that there will be many more well healed doctors and dentists who cannot resist the temptation to buy.
Out of curiosity, how do the numbers from rawdigger compare to the sensor used in the IQ250?
It's slightly worse than the IQ250, despite that this IQ250's darkframe was shot at warmer temperature. I don't know how he calculated the standard deviation for the IQ3 100MP.
Hi,
The Read Noise you've shown, is after subtraction of the Black point, but the masked photosites are already without blackpoint offset bias. When you subtract the bias again, the left tail of the standard deviation gets clipped to 0, and the SD of the now clipped Gaussian distribution is under-represented.
Cheers,
Bart
Do you mean I should read the numbers when I manually set black level to 0 for each channel?
Correct, that's what I did anyway.
Oddly, when I do not subtract the blackpoint from the 100MP, the saturation values are all the same, and they are all different from subtracting a fixed amount from all color planes. MAybe RawDigger needs to be tuned a bit for this new file format.
Cheers,
Bart
In your way of calculations the IQ250 scores 13.2 stops of DR, which is slightly below your 100MP score.
Keep in mind that if you export the IIQ file with Capture One into dng format then re-open with RawDigger you could get a different distribution of darkframe noise. I haven't played with that yet. I will see to the real-world dynamic range tests. The first thing is to identify the highlight headroom for the 100MP then decide how much aperture to compensate for a comparable ISO. For example, ISO 35 and ISO 100 are essentially the same for the IQ180. Not sure whether ISO 50 and ISO 100 will be the same for the 100MP this time.
For those that aren't quite following the math-oriented digging around of raw file offsets...
DT will have real-world dynamic range tests available later this week. That is, photos, showing the visual difference.
For those that aren't quite following the math-oriented digging around of raw file offsets...
DT will have real-world dynamic range tests available later this week. That is, photos, showing the visual difference.
Now is the end of CCD, except for the poor people ;)
I wonder how this new sensor will do with the tech lenses though, the smaller pixel pitch is worrying, but on the other hand if it's BSI maybe it's less bad than the IQ250 anyway. The larger sensor size will require larger shift amounts too though to avoid the feel of limited movement range.
Its not like all of the sudden the older backs have stopped working. The were used to produce amazing images then and now. This new sensor however is a monster. I am very eager to see results with tech camera lenses.Ditto to this
That math stuff went way over my head! Looking forward to your examples.
Looks like a brilliant piece of gear for those who need it.
Shooting weddings I couldn't use a $50k body, but will look by enviously from the sidelines with my baby pentax :) Hopefully down the track though I'll get into a full frame MF of some sort.
Its not like all of the sudden the older backs have stopped working. The were used to produce amazing images then and now. This new sensor however is a monster. I am very eager to see results with tech camera lenses.
If it is as it usually is, the "real world DR" is relying heavily on software noise reduction in the raw converter. That is you reduce noise and virtually improve DR at the cost of detail and color fidelity. A trick that's quite hard to detect for the layman, but it's still a trick.
If they state 15 stops of DR in the data sheets, like they currently do, I hope it's something substantial behind it so it's not the "16 bit color" all over again. To start with the file format must have changed slightly so the files actually contain 15 bits, as the old IIQ is just 14 bit expanded to 16. If it had changed I'm surprised that rawdigger could read the files at all, so I'm very skeptical about this number.
The the DR thanks to aggressive noise reduction may appear to be better than it is I find most likely though. You probably can't turn off the base noise reduction in C1, "turning off" noise reduction usually just means that you turn off a second layer of noise reduction. The base layer can be quite aggressive too.
What I DO find interesting is Phase One claiming one stop MORE dynamic range than their previous highest. That's very promising given how very very good the previous highest was.
Now is the end of CCD, except for the poor people ;)
I wonder how this new sensor will do with the tech lenses though, the smaller pixel pitch is worrying, but on the other hand if it's BSI maybe it's less bad than the IQ250 anyway. The larger sensor size will require larger shift amounts too though to avoid the feel of limited movement range.
Yes it's very interesting, but I do mind how it's achieved. I've analyzed hifi stuff both audio and photography quite extensively and I know how easy it is to fool oneself. If you stare at one property alone you miss something else, classical example noise reduction, if you stare at the grain you miss that details are smeared or colors are pastel like.
It's also very subject dependent, it may look good in one image, and worse in another... if you really want to know how a system performs you need both, you need the hard engineering facts and you need real life tests. If you only go for a week with unstructured laid back "real life tests" it may grow on you over the next year that the system isn't actually performing as you initially thought. Pairing "real life tests" with independent expert testing gives you (at least me) better confidence that the product really delivers as advertised.
Personally I think this DR thing is a way over the top exaggerated "need", but I understand as much that it makes sales just like megapixel number does. I do expect this new back to be excellent just like any other recent Sony sensor, but if it's actually better than the already excellent IQ250/645z/CFV-50c remains to be seen.
MFD is afterall much about "keeping the distance" from 135 gear, and today this is done with two hard numbers, megapixels and DR. 100 and 15 are two really good numbers for now :)
Yes it's very interesting, but I do mind how it's achieved. I've analyzed hifi stuff both audio and photography quite extensively and I know how easy it is to fool oneself. If you stare at one property alone you miss something else, classical example noise reduction, if you stare at the grain you miss that details are smeared or colors are pastel like.
It's also very subject dependent, it may look good in one image, and worse in another... if you really want to know how a system performs you need both, you need the hard engineering facts and you need real life tests. If you only go for a week with unstructured laid back "real life tests" it may grow on you over the next year that the system isn't actually performing as you initially thought. Pairing "real life tests" with independent expert testing gives you (at least me) better confidence that the product really delivers as advertised.
Personally I think this DR thing is a way over the top exaggerated "need", but I understand as much that it makes sales just like megapixel number does. I do expect this new back to be excellent just like any other recent Sony sensor, but if it's actually better than the already excellent IQ250/645z/CFV-50c remains to be seen.
MFD is afterall much about "keeping the distance" from 135 gear, and today this is done with two hard numbers, megapixels and DR. 100 and 15 are two really good numbers for now :)
I generally take zero stock in the absolute number any company/website/person assigns to dynamic range. They can be using whatever definition they choose.
Do they mean where you can see the line between a black patch and a darker patch in a ramp target? I don't shoot targets. I shoot people. Moreover I don't care if I can see the patch. I care about things like how much grain/noise is there when I pull up the shadows, and just as importantly what is the structure and aesthetic of that grain/noise (is it clumpy? finely distributed? colorful? monochromatic?). How accurate and natural is the color in those shadows?
Moreover I don't give two figs whether it's accomplished in hardware, firmware, software, or sorcery. So if the dark frame and secret sauce of Capture One is (as it has been in the past) part of the final result, I really don't care.
I care how the final image looks. That's it. That's why I say we'll have real world comparisons up later this week.
What I DO find interesting is Phase One claiming one stop MORE dynamic range than their previous highest. That's very promising given how very very good the previous highest was.
But it is essential to understand how to do a proper comparison of dynamic range. Preferably I would hope for the same headroom for highlight recovery, with some details in deep shadows to push. If ISO 50 is indeed an extended ISO just as what is with the IQ3 80MP, then you might end up over-exposing the IQ3 100 MP by a stop, leaving less highlight recoverability and cleaner shadows.
Things are finally getting interesting in the MF world! From a spec standpoint, this may be the first back that gets it really right.
The price point remains a bit of an issue though... :(
Cheers,
Bernard
Time to throw a cat in to keep the pigeons off the ground: Is this the same CFA seen on the A7r2 ?
I wonder why P1 doesn't apply sensor + technology to their CMOS backs... Is there a technical reason for not doing so?
For those that aren't quite following the math-oriented digging around of raw file offsets...
DT will have real-world dynamic range tests available later this week. That is, photos, showing the visual difference.
IMO, while this is welcome advancement, Phase One have put themselves into a corner for a competitor to emerge. Sony has Phase One by the boot straps with sensor development as they have demonstrated they will sell the same sensor to anyone (Pentax rocked the 50MP implement for a lot cheaper)
No need to do so since high ISO on the Sony sensor is already very good?
Cheers,
Bernard
We sold 50mp CMOS backs at a faster pace after the introduction of Pentax 645Z than before. It was one of our most commonly traded-in cameras.
Medium format is a niche. It's a small small percent of all cameras sold. So any competition, marketing, or awareness raising, helps everyone in medium format with a strong product. Our hardest task, by far, is getting people to come take a picture with our camera; once they are looking into any medium format solution (of any brand) they are far more likely to do so.
FPS or SPF?
Stabilized lenses, or sensor?
Focus speed, Focus Tracking, Multi Point Focus or PDAF Track Focus?
With HDMI out, does this allow for motion imagery, and if so Specs?
Useable ASA?
Articulating LCD?
The DR that Bart measured is the engineering DR. Photographically useful DR is less, since a SNR of 1:1 with engineering DR is not useful photographically. Photographic DR depends on the noise floor used for its determination but is typically a couple of f/stops less.
Original 645 film frame = 60mm x 45mm
That's why I asked and that's what's missing for 50 grand. 18/20/35 mpx in focus is worth a million mpx out of focus, unless you've come up with the lytro thing where focus doesn't matter.
This is where you went astray.
No 645 film camera ever exposed an image on film which was 60 x 45mm.
The film gate of the Mamiya 645 (i.e. how big the picture, as exposed on the film) is the benchmark used by Phase One.
This is where you went astray.
No 645 film camera ever exposed an image on film which was 60 x 45mm.
The film gate of the Mamiya 645 (i.e. how big the picture, as exposed on the film) is the benchmark used by Phase One.
Doug,
So the sensor size of the 60, 80, 100 (53.7mm x 40.4mm) is equivalent to what has always been captured on film?
Original 645 film frame = 60mm x 45mm
Actually, film 645 is 56 x 41.5 mm.
So the sensor size of the 60, 80, 100 (53.7mm x 40.4mm) is equivalent to what has always been captured on film?
Close enough see my post above.
Geeezus Guys.
The handful of usual suspects are already on the prowl and looking to have PhaseOne's head on a platter. Ok, step back for a second. This is a 100MP CMOS full frame 645 (40.4mm x 53.7mm) back with the latest (or at the very least very recent) SONY CMOS sensor technology. It has twice the MP of the previous SONY MF sensor and a larger form factor.
You guys need to chill out and head out and make images for a change.
Yunli, you have some awesome images on you 500px. The ones using the D4s and the IQ260 are superb among others. At the very least the new Phase 100MP back is gonna be as good or most likely a lot better regardless of whether it has 13.1 stops or 14.99 stops of DR or whatever. It is gonna have much more color sampling due to more pixels and if they are as good pixels as the one in the 50MP Sony sensor (or even if they are sightly worse, possible better!) there is just a lot more of them. Its gonna be splitting hairs regarding per pixel quality.
At the very least the new Phase 100MP back is gonna be as good or most likely a lot better regardless of whether it has 13.1 stops or 14.99 stops of DR or whatever. It is gonna have much more color sampling due to more pixels and if they are as good pixels as the one in the 50MP Sony sensor (or even if they are sightly worse, possible better!) there is just a lot more of them. Its gonna be splitting hairs regarding per pixel quality.
Medium format is a niche. It's a small small percent of all cameras sold. So any competition, marketing, or awareness raising, helps everyone in medium format with a strong product. Our hardest task, by far, is getting people to come take a picture with our camera; once they are looking into any medium format solution (of any brand) they are far more likely to do so.
Doug, why sensor + isn't used on P1's Cmos backs?
One of the perks of the job is doing my wedding photography (http://www.doug-peterson.com/p829532778) with Phase One gear (along with Fuji or Canon as a second camera).Doug,
Having the leaf shutter lenses is absolutely game-changing for what kind of light and lighting you can use at a wedding. That's a big deal because even our 10k systems feature a full line of leaf shutter lenses. If you'd ever like to arrange a test just let me know :).
http://www.phaseone.com/apsis/160409Sensorplus.pdf
"The Phase One Sensor+ solution to the noise problem was to add the pixels together inside the image sensor, before the amplification process. "
Sony needs to be able to sell sensors to others... hence it can't make on chip (on die) solutions that can't be sold to others... and to make different chips = less savings from scale.
Do not get me wrong, I think this will be an incredible camera. 100MPx from Phase One. Awesome, really awesome. My point was that the door is wide open for competition in that they are using a Sony sourced sensor and they are implementing it on an Optical viewfinder camera. I really do believe Sony will produce a mirrorless MF camera. Time will tell.
You guys use to mention seconds per frame (spf) rather than frames per second.
Sunscreen? Apply it anywhere you feel you need it. That's more information than I need.
There is just one problem. Lenses. Several of the Pentax newer 645 lenses are made to cover the 33x44mm format, just. IIRC the 28-45mm zoom is for 33x44mm sensor only. Ironically the older 645 lenses might work nicely although I would not expect them to get the best out of a 100mp sensor. Even the Hasselblad HCD lenses, the 24mm, 28mm and the 30-90mm only work with 1.1x crop or smaller sensors.
Sounds like a perfect first road trip for the new Benz.
With my DCamProf color profiling work I've looked at a few Sony high end CMOS sensors and their CFAs are if not identical very highly similar. You can use a A7r2 profile on a Pentax 645z for example. I never made any real attempt to actually find out exactly how similar they are. I very much doubt that Sony make custom mixes for different manufacturers though so I think all products based on the sony sensor of a particular model have the same CFAs, but the IR filter response can differ of course.
The cited example is monochrome.
Actually, the film image of 645 is 56 x 41.5 mm, so the sensor is getting pretty close to "full 645 frame".While we are splitting hairs over dB of DR and format sizes: even if "645 film format" was up 56x42mm depending on brand, mounts for transparencies cost about a mm along each edge, and printing with enlargers usually costs a bit off the edges of the image that digital printing does not, so yes: 53.7x40.4mm is close enough that we can all stop worrying about the dreaded and despised "digital crop".
While we are splitting hairs over dB of DR and format sizes: even if it was up 56x42mm, depending on brand, mounts for transparencies cost about a mm along each edge and printing with enlargers usually costs a bit off the edges of the image, so yes, 53.7x40.4mm is close enough that we can all stop worrying about the dreaded "digital crop".
My question is whether this sensor is for Phase One only. With the previous Sony 44x33mm CMOS sensor, it seemed that Sony was selling to all possible users in order to get enough sales volume, so is it the same this time, or is Phase One dominant enough in the "full frame 645" that it has negotiated an exclusive, the way it did with several Dalsa sensors?
One of the perks of the job is doing my wedding photography (http://www.doug-peterson.com/p829532778) with Phase One gear (along with Fuji or Canon as a second camera).
Having the leaf shutter lenses is absolutely game-changing for what kind of light and lighting you can use at a wedding. That's a big deal because even our 10k systems feature a full line of leaf shutter lenses. If you'd ever like to arrange a test just let me know :).
Ok, so if this is true, then we can reverse that process, and use the new IQ3-100 profiles on a Sony A7R2 in C1 9.0.2 ?if you believe that CFA is the same for 100mp (or 50mp) as for A7R2 then you can simply rename the profiles (the file names) in C1 and use them right now...
My question is whether this sensor is for Phase One only. With the previous Sony 44x33mm CMOS sensor, it seemed that Sony was selling to all possible users in order to get enough sales volume, so is it the same this time, or is Phase One dominant enough in the "full frame 645" that it has negotiated an exclusive, the way it did with several Dalsa sensors?mmm ... maybe that explains the fire sale for Hasselblad backs in the past couple of months ... getting ready for a new line up.
I would expect the samenbase Iso as A7r2 which I think is 200.
Edmund
mmm ... maybe that explains the fire sale for Hasselblad backs in the past couple of months ... getting ready for a new line up.
Phase One IQ3 100mp Technical Specs (https://digitaltransitions.com/xf-100mp-camera-system/#specs)
1.4 frames per second in 14-bit mode
0.7 frames per second in 16-bit mode
Ok, so if this is true, then we can reverse that process, and use the new IQ3-100 profiles on a Sony A7R2 in C1 9.0.2 ?
Interesting, the 16 bit mode is probably new and may really produce 16 bit data in the raw files. Maybe those 15 bits of DR is true afterall :)
Doug, why sensor + isn't used on P1's Cmos backs? Is there a (technical) reason for that? Personally, I find it the most attractive feature of P1 backs...
For that I do not doubt :) I'm sure it's the king of sensors as of now!
There are so many "if"'s for me: if it's BSI, and if it works with the Schneider 28mm XL, and if I can find a reliable ND solution for that 112mm center filter of the 28XL (leakage-free), and if I can manage to find a good copy of the 28XL, or if Rodenstock makes something else that is compatible and compact and wide enough (doing a better job than the 32HR) then... it's going to be impossible to resist ;D Nevertheless, if the 32HR works with it then I'm sure many tech cam users will be happy with it.
I think you can forget the 28XL. The A7r-II is BSI with the same pixel pitch and it didn't perform as well as we hoped for tech wide angles.
You have a thread to that discussion. Just curious. What was the major issue? Crosstalk or shading?
Unfortunately I never got any raw files in that thread so I don't really know for sure, I've just heard several say that it's not solved the issues of the A7r. So I don't really know where the limits are.
As the sensor has microlenses I expect shading (microlens ripple) be an issue, but it can generally be fixed with some software tricks. Crosstalk is harder to fix. BSI should in theory have much lower such issues, but it depends on your design target on how you make the sensor. Tech lenses haven't been in anyone's design targets in recent years.
I think that to make it work well with tech lenses you'd need to make the photo diode a bit smaller and thus hurt DR/noise performance a little, and that is not a tradeoff the manufacturers want to do, just like we're probably not going to see microlens-free sensors again in the nearest future.
I think the tech lens ranges will have to adapt to the sensors because the other way around won't happen, unless we get a good match due to some coincidence. I hoped that BSI would be one such, but the A7r-II made me doubt.
I would not be surprised if Rodenstock is already working on some new retrofocus wide angle to work with modern CMOS sensors. Say if this new back won't work with the HR32 that would be a major blow to the tech cam genre for architecture and landscape and only Rodenstock can fix that, Sony won't care. With the lens design getting closer and closer the SLR lenses it remains to be seen if they can keep up with the image quality advantage, especially since they must do it without software lens corrections. If they succeed it would be a $10k+ lens (and probably 1kg+) for sure...
But let's first see what this back can do with the current lens ranges... I can't wait to hear/see about it :)
I thought the detail on the studio shot was jaw dropping.
The ice mountainy shot i thought was a bit soft but put it down to weather conditions.
All I heard was just astigmatism/curvature/smearing caused by sensor cover glass, not crosstalk. I have never been able to know why and how the A7R-II does not work well with the 28XL.
...maybe that explains the fire sale for Hasselblad backs in the past couple of months ... getting ready for a new line up.
...Hasselblad seems to be moving their 50MP sensor as to be their entry level product which means that they are making space for larger sensors... Whether that will be the same Sony 100mp sensor or other remains to be seen...
...this is a CMOS sensor, and it has been confirmed that it is a combined Phase One / Sony design, fabricated by Sony. My understanding is that this sensor will (over time) also become available to other camera makers...
Phase One had exclusiveness for the 60MP Dalsa sensor (originally in the P65+) for a few years, so it may be some similar deal here. That would be great for Phase One, and it's more likely this deal to happen for a 54x41mm sensor than a 44x33mm as the former don't have the same mass market potential so Sony doesn't lose as much letting Phase One have exclusivity.
For the users it would be an unfortunate monopoly situation though of course.
I believe that Kevin and Michael made it clear in their article that this sensor was not eclusive to them.
Now it doesn't that they can't make a better use of it than the competition.
Cheers,
Bernard
Wayne, Theodoros,
From here (https://luminous-landscape.com/100-fxxing-megapixels-phase-one-xf-camera-reaches-100-mp-early-hands-one-report/):
Last night Alpa posted a picture of the 100 MP mounted with the 32HR, but later posted only the sample image of 90HR-SW instead. This isn't looking good for the 32HR if it's not BSI. Maybe Alpa will be able to surprise us.
It's like T-Rex showed up after the meteor crash and it had nothing left to eat.
This back really is like the second coming.
Too bad it's too late.
Most photography has transitioned to motion content, and what is left is displayed on the web, or electronically on monitors at low ppi.
It's like T-Rex showed up after the meteor crash and it had nothing left to eat.
monitors at low ppi.
Last time I checked, video did not kill the radio star.
I doubt motion is going to kill stills.
Is there anybody who downloaded the example files on the phase one website and opened them in photoshop? I really was absolutely not impressed by the image quality: sure, they are large, big and heavy but i really don't like the rendering on the airplane and lack of detail in all images except the one with the girl. It must be me but really, not impressed :-)
Hi Peter,
There are a few reasons for that, explanations but not excuses.
1. The denser sampling will gradually have more difficulty resolving detail due to residual lens aberrations. No lens is perfect, and this is part of what you see, more clearly/accurately (and enlarged!). It shows as a loss of micro-contrast at best, or blur if it gets worse (e.g. in corners or near the edge of the image circle).
2. Depending on the aperture, there will be a mix of defocus and diffraction. The plane of best focus is a rather thin zone. There are restoration tools available that can improve that in postprocessing.
3. Improper sharpening. I don't want to sound like a broken record, but people really need to wake up and start using better tools for Capture sharpening. There is also no real excuse for the halos that the images show around edges, only the explanation that the wrong method and settings were used.
There may also be an effect of the Raw converter that was used, I haven't checked the EXIF data of those files yet, but e.g. Capture One Pro is able to extract higher resolution from a Raw file than e.g. Lightroom/ACR. Additionally, if there are automatic lens corrections applied, then resolution can suffer. Resampling always loses some signal strength and usually adds some artifacts. The negative effects of resampling can be minimal if proper algorithms are used.
And finally there is a psychological effect when looking at such (much more) magniified detail on low resolution displays. This all assumes that shooting technique was optimal to begin with, because subject movement, and camera shake, will show.
Having said that, there can be better quality output than some of the examples shown, so take these samples as indicative, but not as an ultimate demonstration of what is possible.
Cheers,
Bart
Internet killed the video and radio star. Stills will always be here but not in the same capacity or context. It's not a case of "if", it's already here.
I feel like there is a large majority of not so educated shooters screaming "more resolution!" that drowns out the rest.
Last time I checked, you only need 12 mega-pixels or so for a full spread. 100 mega-pixels sound great, but at what expense?
It seems to me that we are gaining more resolution at the expense of other (more) important aspects.
Personally, I don't get it.
I have been comparing the quality of the blacks from my P45+ files to my fiancées IQ260 files and feel that I may want a more modern solution, especially with my recent venture into still life and beverage (most of which are very dark images).
However, this super high resolution backs just have so many problems associated with them.
Sure, you get bigger files, but you have to worry about camera shake, lens cast, cross talk, micro lens shading, diffraction, circles of confusion that are not small enough, etc., etc.
I feel like there is a large majority of not so educated shooters screaming "more resolution!" that drowns out the rest.
Last time I checked, you only need 12 mega-pixels or so for a full spread. 100 mega-pixels sound great, but at what expense?
....he may find that there is no safe conclusion that P1 was involved in the development of the sensor... As we all know, Hasselblad was first to implement the 50mp Sony Cmos sensor in production...
The 32HR is the one lens to test now though with this back, I'm extremely curious on the result. I have no idea how it will perform. I would not be surprised if it's a huge disappointment, but the opposite would not be surprising either. Someone's gotta test this soon! I can't wait :)
Phase One shipped its 50mp CMOS solution months before Hasselblad did.
Maybe you're mis-remembering because Hasselblad did a hurried announcement-of-an-announcement the day before Phase One did after they heard Phase One was about to announce theirs?
Phase One was first to ship the 50mp CMOS. First to ship 60mp. First (and still only) to ship 80mp. First to ship 100mp.
We're on it. Just as soon as our demo units arrive! They are en route at the moment; we are tracking them eagerly!
Hi Joe,
Well, there's an opportunity for the software industry to help the less educated users in getting better results, and even the educated users can benefit from more intelligent and better quality tools.
There are many types of photography and some do benefit from these kinds of resolutions. Even when technique is lacking, more dense sampling can also help to get better modest size images.
Cheers,
Bart
Nope, the Internet just gave both of them another way to perform. ;)
Yep, but IR filter is almost certainly different so the response is not exactly the same even if the CFA would be identical.
It should be noted that even if CFA is quite different normal low saturation colors will look sort of right anyway, to really see if there's a match you need to test high saturation colors.
All modern cameras today have quite similar CFAs. I've tested a fair bit of cameras and only my own Kodak based H4D-50 was substantially different within the "normal" range, as Kodak has built in quite some subjectivity into the CFAs itself, while all modern cameras are more designed to make objective realistic colors. Regardless you can still make hugely different results depending on profile design, just compare Leaf and Phase One color, despite the same sensor they have quite different color. Leaf ICC profiles can't be used with other cameras as Leaf has it's own processing pipeline with its own way to format their ICC profiles, nowadays builtin to Capture One. Otherwise you could switch between Phase One and Leaf color easily, but due to the different software handling of color you can't which is a good thing for Phase One so the products are more clearly differentiated.
and about 7 Billion others. ;)
Like it or not, the world has changed dramatically. Investing $50K in a camera that does not shoot 6 or 8K footage is not a wise venture, IMO. Unless you have the specific purpose that you are 100% sure will see out the value of the investment. Coming from a high end market in London I do not think that is wise unless you are a hire company.
An increasingly large chunk of high end stills advertising is now a lifted from RED and Alexa footage. This is for ease of production, cost efficiency, but also increasingly because of cross platform continuity and the diminished relevancy of stills. Vloggers and Youtubers are gaining more gravitas than the staple old school print media brands. Print budget is now channelled towards instagram and youtube vloggers/bloggers and other new media like iPhone apps.
What is left in any sort of capacity that someone can make a decent living from is editorial and if you think that these budgets allow or even need a $50K 100MP investment you are deluded or have a rich family.
If one carefully reads this statement (from P1's site),...
...he may find that there is no safe conclusion that P1 was involved in the development of the sensor...
If we want to be negative the biggest threat is that 135 is becoming too good. It's similar to when MFD took over large format film. The break point was 39 megapixels with P45+, many thought LF film was still better, but the 39 megapixels was close enough, more practical and fulfilled all reasonable printing needs. But I think MFD can exist as a niche, having the best stuff even if it's meaningless in any quality aspect still has some nice prestige too it.
My P45+ produces a file big enough for any kind of commercial use. The only problem I have with it as of late is that the deep blacks are not up to par with current sensors.
Given the choice, I'd rather have a hassle free 40 or 50 mega-pixel full framed modern back then a 100 mega-pixel back with a host of additional issues due to smaller pixels.
Hi Joe,
Not that I want to create awareness of another problem you may or may not (sub)consciously suffer from (depends on subject and shooting settings), but higher resolution also prevents many aliasing issues and produces more accurate colors, especially in the shadow regions.
Many of those 'issues' are already there at lower resolutions, but are now just more easy to see, and more easy to solve with proper tools.
Cheers,
Bart
About five years ago I saw several estimates that seemed to agree that the total worldwide market for high-end digital backs was about 7,000 units. My guess is that it's not all that much different today. How the mix of markets breaks down is anyone's guess though.
And of course, not to be forgotten (because it is likely one of the largest constituencies), are wealthy amateurs. About five years ago I saw several estimates that seemed to agree that the total worldwide market for high-end digital backs was about 7,000 units. My guess is that it's not all that much different today. How the mix of markets breaks down is anyone's guess though.
I will say that when one runs high-end travel / photography workshops (the way we do) and sees the number of serious and often talented amateurs shooting with $40-$80K MF systems, I would not be surprised if as many as half the world market for high-end MF is to affluent individuals. These are the same people who buy top-end cars, boats and homes. People for whom purchasing a $50K camera system like the XF100MP requires little more thought then whether they should upgrade to this year's BMW 750, or wait another 6 months to see the latest models for 2017.
Michael
That's not much of an endorsement, Michael, for the need for these cameras. Sure, some wealthy amateurs are outstanding photographers, but for the rest of that constituency, its largely a vanity market.
We're on it. Just as soon as our demo units arrive! They are en route at the moment; we are tracking them eagerly!
However, I can not see better color being a product of higher resolution, maybe higher quality pixels that just happen to be smaller?
No, it's due to the denser sampling with a Bayer CFA. In the lower exposure regions, the shot noise plus read noise constitute a random brightness sample in one color band. That means that the more random an average sample is, the less accurate the demosaicing will be, and the less accurate the resulting color. This is, amongst others, a reason why deep shadows gain a magenta cast (there is a relative under sampling of Red and Blue noisy samples, but they get interpolated to fill in the missing pixel positions). Larger Red and Blue blotches are the result. A higher resolution sensor offers more of those random samples, and they are like all larger sampling quantities getting closer to the real mean value for those locations (larger sample quantities produce smaller standard deviation).
The fact that the individual samples are also less noisy also helps, they enforce eachother.
Cheers,
Bart
I think the thing is that the high pricing is a bit provoking, together with the marketing like
"I use Phase One because it’s the best. I just can’t afford not to." - David LaChapelle
like if you're not using Phase One you're not among the best photographers, and if you can't afford it then you're really crap. Doug's salesman style about how great P1 is, is pretty pushy too :)
There is just one problem. Lenses. Several of the Pentax newer 645 lenses are made to cover the 33x44mm format, just. IIRC the 28-45mm zoom is for 33x44mm sensor only. Ironically the older 645 lenses might work nicely although I would not expect them to get the best out of a 100mp sensor. Even the Hasselblad HCD lenses, the 24mm, 28mm and the 30-90mm only work with 1.1x crop or smaller sensors.
I think the thing is that the high pricing is a bit provoking, together with the marketing like
"I use Phase One because it’s the best. I just can’t afford not to." - David LaChapelle
like if you're not using Phase One you're not among the best photographers, and if you can't afford it then you're really crap. Doug's salesman style about how great P1 is, is pretty pushy too :)
Interesting…, statistics can be fascinating.
Hi Erik, yes they can if they work in our favor ...
Just imagine a 16x multi-exposure sensor shift version of this huge slab of silicon. Quadruple file size, but also double resolution, even better color, and virtually no aliasing artifacts (and huge artifact free sharpening potential). It would of course be very hard to get a vibration free shot, but the quality would be awesome !!! Heck, even a 4x multisampling sensor would be a sensation, but why stop there ... ;)
Cheers,
Bart
Doug, when testing the 32HR, could you please make a test with full shift up and to the sides of a clear blue sky - this will clearly show how this back is handling banding, tilling and other not so nice things for us AP - it's ok if the metal disc in the lens is showing.
Thank you very much in advance!
/adam
If I can get a blue sky and air temperature I can tolerate standing in for the 2-3 hour test you will get your wish!
In fashion, photographers getting the most attention and work, at least at the moment, are shooting film. There is a real resurgence in Mamiya RZ and Pentax 67, and scanning colour prints being printed by the photographer in the darkroom.
28º and Sunny, perfect weather! The back will be nice and cool. Even less noise? 8)
If I can get a blue sky and air temperature I can tolerate standing in for the 2-3 hour test you will get your wish!
I'm myself interested in art photography, and indeed film is big there too. But it's not so much the quality, it's the process, being more "real photography", and the also the look. I'd love to shoot film for my own work, but I'm just too lazy, or at least I say so, to be honest I'm probably shooting too little volume to keep up a stable skill to not do beginner's mistakes on important shots. Digital is afterall a lot more forgiving...
There's a lot of talk about the 100MP, but the big thing here is surely that it's CMOS and it's full-frame... you know live view, high ISO, long exposure. 100MP is not that much of an upgrade compared to 80MP.
I think 100MP is quite logical amount for this next top of the hill back though... the only drawback I see is that it's probably hurting tech cam wide performance, smaller pixels makes shading and crosstalk more likely to happen, but if we're lucky it will still perform well due to some new pixel design and that would be just awesome.
Thank you so much!+1 to this please doug
3 frames in portrait orientation will be enough - 1 full up - 1 full up full left - 1 full up full right. just at f/11 that will do it - so it could be done in 10-15 minutes ;-)
If it could be possible to download the 3 files and their corresponding LCC files it will be fantastic!
/a
If I can get a blue sky and air temperature I can tolerate standing in for the 2-3 hour test you will get your wish!
There's a lot of talk about the 100MP, but the big thing here is surely that it's CMOS and it's full-frame... you know live view, high ISO, long exposure. 100MP is not that much of an upgrade compared to 80MP.
I think 100MP is quite logical amount for this next top of the hill back though... the only drawback I see is that it's probably hurting tech cam wide performance, smaller pixels makes shading and crosstalk more likely to happen, but if we're lucky it will still perform well due to some new pixel design and that would be just awesome.
+1 Sure thing... I also think that there is more talk on the sensor than the back itself...
Surely 100mp is nice to have for some (very few) of cases... I do have a feeling though that if one offers this sensor with four pixel binning ability and 25mp as an alternative choice of use, people will find it much more attractive...
Hi,
Hasselblad used to make multishoot, but according to Doug their lenses are not up to the task, so why even try?
Another way to see it that Hasselblad had a 200 MP product for a long time and that actually worked well under benign conditions.
Best regards
Erik
Hi,
Hasselblad used to make multishoot, but according to Doug their lenses are not up to the task, so why even try?
Another way to see it that Hasselblad had a 200 MP product for a long time and that actually worked well under benign conditions.
Best regards
Erik
You have to understand that P1 is addressing their products to cases where color accuracy and extreme detail is not required... they are well behind for professional use in art reproduction or demanding still life photography...
Well I toss and turn about adding a Sony to my Canon gear, then Phaseone come out with this and I'm thinking..."I wonder how much the Pentax will be?"
Looking forward to seeing more images made using this back.
Hi,
Hasselblad used to make multishoot, but according to Doug their lenses are not up to the task, so why even try?
Another way to see it that Hasselblad had a 200 MP product for a long time and that actually worked well under benign conditions.
Best regards
Erik
It looks like our first units will arrive either tomorrow or Wednesday depending on customs.
In any business, it's important to perceive the totality of a market, not just to see it through the filter of one's own needs, abilities and experience.
Michael
Cultural Heritage Imaging (the broader category under which art reproduction falls) is one of the core markets for which Phase One makes solutions. It has a dedicated version of Capture One for just this market (Capture One CH (http://dtdch.com/capture-one-ch/)) with powerful tools and bespoke ICC profiles meticulously created and tested for pure color accuracy for a variety of supported lighting sources, including the new DT Photon Preservation LED Lighting (http://dtdch.com/dt-atom-entry-level-digitizer/#photon) which has a CRI of 98 and more importantly a CQS of 98 and strong spectral flatness.
I'm glad to help anyone make a comparison in quality for art reproduction of our system versus literally any other made. Just shoot me an email to arrange it.
Here is a partial list of US clients using Phase One for Art Reproduction and other Cultural Heritage imaging (http://dtdch.com/copy-stands/):
Academy of Motion Picture Arts and Sciences
American Museum of Natural History
Amon Carter Museum of American Art
Ancestry.com
Backstage Library Works
The Bernice Pauahi Bishop Museum
Brown University
Center For Creative Photography at The University of Arizona
The Church of Jesus Christ of Latter-day Saints
Columbia University
Dallas Museum of Art
Dartmouth
Denver Art Museum
Denver Public Library
Drexel University
Duke University
Florida State University
Folger Shakespeare Library
The Franklin Mint
The Frick Collection
Harvard College Library
Harvard University
Harvard University Graduate School of Design
Houston Public Library
Indiana University
Indianapolis Museum of Art
The J. Paul Getty Museum
Library of Congress
The Mariners’ Museum
Minneapolis Institute of Arts
The Morgan Library & Museum
The Museum of Fine Arts, Houston
The Museum of Modern Art
National Gallery of Art
The New York Botanical Garden
New York Public Library
New York University
The Ohio State University
Princeton University Library
Public Design Commission of the City of New York
Smithsonian Institution
Stanford University
Sterling and Francine Clark Art Institute
The University of Alabama
University of Illinois at Urbana-Champaign
The University of Kansas
University of Michigan
University of Minnesota
The University of North Carolina at Chapel Hill
The University of Pennsylvania (UPENN)
University of Virginia
Walt Disney Studios-Burbank
Yale University
FADGI Image Quality Analysis showing the four-star (highest possible) quality on an objective test:
FADGI Preservation Grade Imaging - Phase One (http://dtdch.com/fadgi-image-performance-report/)
I think Doug was referring to Hassy V mount lenses, not H mount, and indeed some of the old analog lenses are quite soft. If you shoot for look-and-feel rather than resolving power alone the V mount lenses can do fine results. I'd love to see a CFV back with this new sensor, but I'm not sure it will happen. At some point Hassy will have to end their V support...
I believe Doug was referring to the V glass, not to the H glass... The H glass is excellent.
And Hasselblad is still selling the 200MP MS back:
http://www.bhphotovideo.com/c/product/1078381-REG/hasselblad_3013708_h5d_200c_ms_digital_camera.html
For about the same price you actually get twice as many MP... :)
Congratulations to Phase One for once more delivering ahead of the competition!
Just to mention, I checked resolution on my Planar 100/3.5 on film (ADOX) and it was in excess of 200 lp/mm, but probably at very low contrast. Not that it is that impressive, Erwin Putt measured a bunch of 50 mm lenses some years ago and all had around 200 lp/mm but just a few % MTF. But H-lenses are generally better.
I guess we can expect a more reasonably priced update from Hassy soon, and a $12K announcement from Pentax at Photokina. My pro dealer tells me the 645Z has been selling well, and so Pentax will probably want to catch the new train leaving the station.
Edmund
I downloaded one of the (iso 50) Raw files from the IQ3 100mp and wow. About as perfect and clean a file as I have seen. It is of an outdoor town scene in bad light. Dynamic Range (shadow recovery) is stunning. I removed all of the (lens, noise, you name it, everything off or at zero) corrections and just boosted up exposure and looked at the darker areas and detail just keeps showing up, clean. Color looks really nice, comparable to what I have seen from the 50MP Sony sensor but even smoother and cleaner. Stunning.
Looking forward to seeing more images made using this back.
All that rubbish and complaining about whether anyone can afford it, need it, want it, whether Phase is gonna survive, whether dealers and the marketing dpt. use cheap sales techniques, whether you can achieve the same using much cheaper gear, etc. etc. just have nothing to do with how this back performs. That is just background noise. The back is amazing. The whole IQ series from the start has been impressive. (I own one, IQ160, and although it was expensive I never regretted buying it, still performs great).
Choosing gear is a very personal process and honestly it is getting tiring reading through all the complaints, rants and rubbish and technical stuff instead of more about how the gear works during photography which is the main purpose of such things. If you find serious faults using the gear in the discussion fine, post them in context of the photography but man, some threads in this forum are getting ridiculous. Just helps to drive away the photographers and technicians who actual deal with Medium Format Digital gear.
The limitation of the focal plane on the 645Z makes it a non-starter for many including myself...
I downloaded one of the (iso 50) Raw files from the IQ3 100mp and wow. About as perfect and clean a file as I have seen. It is of an outdoor town scene in bad light. Dynamic Range (shadow recovery) is stunning. I removed all of the (lens, noise, you name it, everything off or at zero) corrections and just boosted up exposure and looked at the darker areas and detail just keeps showing up, clean. Color looks really nice, comparable to what I have seen from the 50MP Sony sensor but even smoother and cleaner. Stunning.
But the IBIS? Maybe Phase left the IBIS on the drawing table to leave something to upgrade for in 2 years?
Yes, I agree with Ken. The image looks very nice. I could be satisfied with such a quality, with little extra (multi-shot would be nice) to wish for. It will be interesting to see how the images that are taken with shifts will hold up, although perspective correction in software is becoming a feasible alternative (depending on required output size).
By doing it in postprocessing, one can shoot with the higher quality center of the lens' image circle, and have less Color Cast and cross-talk to worry about. It is also much easier to get perfect leveling in software, than it is on tripod. Even with a geared head and a high accuracy digital level, there may be a bit of residual rotation due to tolerances in mounting the sensor in the back.
Cheers,
Bart
I'm sorry if I missed somthing in this 11 pages and counting thread, but does anybody know what P1 means with "exceptional wide angular response" in their IQ3 100mp press release? (http://www.dpreview.com/news/0998933437/phase-one-launches-100mp-medium-format-camera-co-developed-with-sony)
It does sound promising, but it may not necessarily mean much. Wide angular response is a term sensor manufacturers use about how low an angle a sensor can take before crosstalk happens. So there's some action taken in that regard, but they don't compare to anything so we don't know what it means. This could be a BSI sensor afterall although my inital guess is that it's not. They could simply refer to offset microlenses too.
One should know that modern image sensors usually have rather poor angular response and the old CCDs had better angular response than any recent sensor. So if you compare to the typical CMOS you can make a "big" improvement and still be much worse than an old Kodak CCD.
We really need to see those tests which Doug's is making for us as soon as he gets the back :)
Although I dream of a CMOS that can do the SK28XL without issues, if this just reaches Dalsa 6um class on the angular response it's hugely great news. If on the other hand the sensor makes the Rodenstock wides obsolete it's pretty bad news, so the wait for those tests is really tough :)
What I noticed in raw digger is that the file is 16 bits, but only contains 14-bit data. So the image is a 14 bit file.
.......I could be satisfied with such a quality, with little extra (multi-shot would be nice) to wish for.......
We really need to see those tests which Doug's is making for us as soon as he gets the back :)
Although I dream of a CMOS that can do the SK28XL without issues, if this just reaches Dalsa 6um class on the angular response it's hugely great news. If on the other hand the sensor makes the Rodenstock wides obsolete it's pretty bad news, so the wait for those tests is really tough :)
I'm myself interested in art photography, and indeed film is big there too. But it's not so much the quality, it's the process, being more "real photography", and the also the look. I'd love to shoot film for my own work, but I'm just too lazy, or at least I say so, to be honest I'm probably shooting too little volume to keep up a stable skill to not do beginner's mistakes on important shots. Digital is afterall a lot more forgiving...
To be honest, by my latest test, I'm less and less lazy. Many labs around me (and I'm country side) invested in new development machines specially suited for film. Those machines can even "chemically" print digital images on reactive paper. The bath are automatically controlled and are clean and filtered. I understand who give up about film. This is the guys who faced hardgore the digital transition (sorry Doug ^^). They started to digitalise thousands and thousands of negatives thus killing any joy in the process. Digital for them is a Deliverance, to say the least.
Today film is maxed and do not suffer WA smearing :) Digital isn't maxed yet but is a super complement. Digital is the new pola. It is a good period for a photographer to "taste" real MF with film at little expense. Take your roll in any labs (at least with modern frontier or equivalent) and you are done. Choose the pics you want digitalized or do it yourself at home with a lightbank and a Merrill sensor stitch. For ppl seeking a pleasure of photography it is ONE way to go, for sure (not the way).
Of course such expensive and top notch items will find a house somewhere on the planet. Michael said the market is 7k Unit but IHMO, in late 2015 and start 2016 it is more 5k units worldwide. Asia market will buy most of the units. Asia is now where USA where in the 80/90's but with modern tech. When they do something it is just huge.
But the IBIS? Maybe Phase left the IBIS on the drawing table to leave something to upgrade for in 2 years?
Hi,
I checked a bit on the image, in C9. Pushing 3.08 EV, with sharpening and noise reduction set to zero, I see some colour blotching in grey areas I would that I would not expect. So I checked in RawDigger and saw that the image is exposed pretty ETTR, with clouds close to clipping.
What I noticed in raw digger is that the file is 16 bits, but only contains 14-bit data. So the image is a 14 bit file.
Having never shot with a high mp back, I do wonder other than detail, does the high mp back give better gradation or is that entirely down to bit depth?
I would of thought the ability to capture gentle tonal variations is going to make a photograph more beautiful or real rather than being able to count a few more blades of grass.
Is there a difference between 50mp and 100mp in this area?
I thought I saw on their site that P1 was offering this in H mounts? Don't dash my hopes of owning a used one 10 years from now :-)
In short, given equal composition and similar per pixel performance, yes, it helps a lot since you are increasing your sampling of any given area in the frame with more pixels.Thanks Ken, Out of my price range, but that would be just about the only reason to buy one for me.
Im sure the resident tech crew in this forum will chime in with a bunch of charts, numbers and what not.
Having never shot with a high mp back, I do wonder other than detail, does the high mp back give better gradation or is that entirely down to bit depth?
I would of thought the ability to capture gentle tonal variations is going to make a photograph more beautiful or real rather than being able to count a few more blades of grass.
Is there a difference between 50mp and 100mp in this area?
I suggest you compare sample shots and make test prints, because there's many different opinions about these things. The thing with "better tonality" is one of those effects that's hard to show in any measurements but some experience it, if it's placebo or not I can't really say.I was just curious, not enough to start downloading files and making prints. I would hope the new back brings more to the party than resolution, the 50mp cmos looked to be a big step forward for MF.
To be considered is that the standard is moved forwards all the time, in all formats. It becomes more and more difficult to see issues in the images, and as such the differences become smaller and smaller. Personally I doubt that you would get a big tonality improvement from an IQ250 to an IQ3 100. In fact I think it would be difficult to tell prints apart on a normal viewing distance.
Higher sampling is good though, you get less aliasing in small details, which can be an advantage in certain shooting conditions.
I'd say that today's hardware is so good across the board that "tonality" is more up to profile design than actual hardware properties.
In short, given equal composition and similar per pixel performance, yes, it helps a lot since you are increasing your sampling of any given area in the frame with more pixels.
Im sure the resident tech crew in this forum will chime in with a bunch of charts, numbers and what not.
Higher sampling is good though, you get less aliasing in small details, which can be an advantage in certain shooting conditions.
I'd say that today's hardware is so good across the board that "tonality" is more up to profile design than actual hardware properties.
Resolution is improved and aliasing is reduced at the same time, both allow more accurate tonality and smoother gradients and better microcontrast. The physically larger sensor will allow to use a larger magnification factor on sensor for the same Field of View, compared to a physically smaller sensor, which will then result in a higher MTF, which also gives more solid data (higher S/N ratio) for postprocessing.
I don't really ever get into the deep technical issues, so let me know if my logic is flawed here...
The IQ3 100 and the A7r2 have a similar pixel pitch, 4.51 µ on the Sony and 4.6 µ on the Phase. Given that, lenses that resolve well enough on the Sony should be fine with the IQ3 100, yeah?
I don't really ever get into the deep technical issues, so let me know if my logic is flawed here...
The IQ3 100 and the A7r2 have a similar pixel pitch, 4.51 µ on the Sony and 4.6 µ on the Phase. Given that, lenses that resolve well enough on the Sony should be fine with the IQ3 100, yeah?
I've been using 'Blad CFi and CFe lenses (and the Canon TS-e's) since I built my Sony rig and have been pretty happy with their sharpness, so I'm guessing I'd be happy with them on the new back, too. Of course, I'd need to add the focal plane shutter to my Arca. Hmm... interesting.
CB
I don't really ever get into the deep technical issues, so let me know if my logic is flawed here...
The IQ3 100 and the A7r2 have a similar pixel pitch, 4.51 µ on the Sony and 4.6 µ on the Phase. Given that, lenses that resolve well enough on the Sony should be fine with the IQ3 100, yeah?
I've been using 'Blad CFi and CFe lenses (and the Canon TS-e's) since I built my Sony rig and have been pretty happy with their sharpness, so I'm guessing I'd be happy with them on the new back, too. Of course, I'd need to add the focal plane shutter to my Arca. Hmm... interesting.
CB
There is no "block".
The XF speaks a different language (aka "intercomponent communication protocol") than legacy Mamiya bodies. Using the Mamiya 645 digital interface spec would have been very limiting given it was developed more than a decade ago.
If Hassy or Sinar want to make a back for the XF they are welcome to. I think Hassy will have no interest as they want to support the M platform. Feel free to talk to your friends at Sinar.
First 32HR and SK43 examples.
The SK43 seems to be problematic....... :(
https://www.alpa.ch/en/site/new-phase-one-100-mp-back-on-the-alpa
First 32HR and SK43 examples.
The SK43 seems to be problematic....... :(
https://www.alpa.ch/en/site/new-phase-one-100-mp-back-on-the-alpa
I think this new achievement is opening a world of opportunity to see new product and some concurrence / price reduction in the months that come.
First maybe someone selling the old gear to buy the 100MP increasing used market on the net that is where second hand from professional, doctor & dentist gear gone in good hands of creative and jung photographer that simply can afford the buy of this MFD gear / back's.
Generally speaking all old gear have characteristics and IQ well beyond what it's possible to achieve with a DSLR (also the newest), so if also Pentax and Hasselblad enter in this run for the big 100 Mp i think some benefit and price reduction (like the one active in this holidays for H5D-50c) will arrive.
To return to the discussion about the new born, I admit to have some doubt about the performance, especially about the noise in low light, the 4,6 microns are at the level of some apsc dslr and no sony nor others brands perform brilliant in low light / high ISO (well, recently they have improved but after all i use always base ISO if I search the best IQ).
Dubt about the 16 bit.... it's not clear, and from the doc I have see there is a cloud around it, first SONY doesn't distribute a DATASHEET of the sensor perfrormanche and the schematic block of the multiplexer / ADC.... remember the DALSA and KODAK (now ON Semiconductor) data-sheets (I.E. http://www.onsemi.com/pub_link/Collateral/KAF-40000-D.PDF for who doesn't even see one in it's life)... Well, where is the data sheet of SONY 50Mp MF Cmos ?, and the 100Mp MF Cmos ??
Without evidence in any form and rumors that ALL are 14Bit and some one says also with a sort of compression I doubt the performance of the new born put it in a class od it's own, surely is larger that benefit low DOF and little more evident MF look on the image but said that, nothing more.
I regularly use the 50c camera from Hasselblad, and before I was using the CCD version, testing also other models and brand I have the feeling that introducing DSLR CMOS tech in this way, is wrong... Specially the race of more and more pixels, the 50 CMOS on the bench had a slight edge over the best professional DSLR IQ wise especially for the 5,3 microns photo site that slight larger... with the 100Mp and 4,6 micron photo site it possible to get better results ?
NO... You have only more pixels for peeping and bigger files to impress Your customer !!
So far so good, but for me an attractive game changing product would be a 10x10 micron 50Mp true full frame sensor with true 16 or more bit processing made without micro lenses and with strategy to avoid cross talk between pixel and with high angular response for tech cameras...
Something but NOT like the sensor SONY put in the A7S-II, something with more conservative CFA Bayer Matrix and a new ADC array designed for precision (not for speed perf) maybe 20Bit that after the on chip elaboration become true (not compressed) 16 bit !
I admit this is my dream, the real situation is that they sell a reworked type of sensor only scaled to squeeze the maximum profit from a type of design, using the same tooling, foundry and process...
To return on my start, I think this is not the kind of sensor for my photography workflow, but surely is an achievement and awakening of this segment of photography almost known to be conservative and quality oriented.
Best regards, Domenico.
Thx for posting. The 32HR shots look stunning. The 40HR should work even better.
The present 100 MP sensor gains it's resolution from increase in surface area over the 50 MP sensor. The pixels are not smaller there is just more of them.
First 32HR and SK43 examples.
The SK43 seems to be problematic....... :(
https://www.alpa.ch/en/site/new-phase-one-100-mp-back-on-the-alpa
Is it me or is there some strange vertical blotching in the sky areas of both 32HR images? The edges are not quite well defined to be called banding, but there is somethings strange going on there.
The whole appeal of the back as a universal solution is pretty much at stake here.
The 32HR can be pushed to almost 15mm +/- on the CCD backs, so would be nice to see more extreme samples.
Is it me or is there some strange vertical blotching in the sky areas of both 32HR images? The edges are not quite well defined to be called banding, but there is somethings strange going on there.
Patience please everyone.
@Jeffery, in the past, using the SK43 and the IQ260, it wasn't uncommon for me to shift up to 15mm. With the SK35, I could get away with maybe 12. So the 32mm HR may be a workable solution if it can go to 12 or 13mm. The problem with architecture, though, was there was never a wider solution that was really workable and I always had to carry the 17 TS-E and a dslr for those occasional superwide shots.
The 32HR can be pushed to almost 15mm +/- on the CCD backs, so would be nice to see more extreme samples.
The most likely scenario I think is a new set of wides, but how long will it take? It can be a few years, a period when tech cams may become known for their many issues on the wide end and lose popularity.
If I were Alpa, Cambo, Linhof, Arca-Swiss or Silvestri now I would be pretty disappointed because it will hurt their business that the highest end back will have quite serious issues with the wide angle lenses for their systems.
Hi Peter and Jsiva,
The blotchy look you are seeing in the extreme shift, seems to have been carried over from the 50MP chip. When I tested the 50MP on extreme shifts, with the 28mm and 40mm and 32mm (loaner), I saw this same issue. It only manifested in solids, like this blue sky. Made worse in B &W.
Opened a case with my dealer and Phase One on the issue back in March 2015 and nothing ever did come from it, however Phase did admit to seeing the effect and agreed it was not my lens or something in post I did to cause it.
I found that Topaz denoise can help smooth this out, as it has an actual de-banding selection.
Interesting to see this problem again. Note, this is not to be confused with microlens ripple which makes a much more solid stripping and is most times removed by the LCC. In my work the LCC did not effect this blotchy banding effect.
Edit: The 100MP does seem to handle the color cast issues on shift much better, if the smaller shot from the Alpa site is before the LCC was processed. With the small 50MP chip all of the left side would have been visibly red, and you don't see that in the Alpa shot. So good news there.
Paul C
Chris Barrett and I will be shooting tomorrow with the ALPA system and numerous lenses. I just did some tests shots with the 23mm Schneider and they are pretty damn nice. It's a problem putting RAWs up on the Luminous-Landscape site because of the size. Our download file system has a max of 100MB and the RAWs exceed that. Our IT guys are looking into seeing if it is possible. Chris and I will report back plus we'll have a more in depth report in the weeks to come. If you can make it to Indianapolis and I'm around you're welcome to give it a try with me.
I am really curious to see how the resolving power (or lack thereof) of the Rodie 55mm holds up with such small pixels. With my P45+, it is very sharp until about 15/18 mm of shift.
Our download file system has a max of 100MB and the RAWs exceed that. Our IT guys are looking into seeing if it is possible.
Hi,Hasselblad still does multi-shoot, and multi-shoot does not stress the lens any more than a single shoot.
Hasselblad used to make multishoot, but according to Doug their lenses are not up to the task, so why even try?
Another way to see it that Hasselblad had a 200 MP product for a long time and that actually worked well under benign conditions.
Best regards
Erik
Hasselblad still does multi-shoot, and multi-shoot does not stress the lens any more than a single shoot.
So if the lenses are good for a 50mpixel capture, they will be for the 200mpixel capture of the H5D-200c.
You insist that the product is not for sale when it is, based on the 50mpixel cmos sensor.
http://www.bhphotovideo.com/c/product/1078381-REG/hasselblad_3013708_h5d_200c_ms_digital_camera.html
The new machine performs far better than the old one.
Phase one has a very good camera, and as always fantastic software and backs. There is no longer a need to a need for this kind of stuff.
Best regards,
Chris Barrett and I will be shooting tomorrow with the ALPA system and numerous lenses. I just did some tests shots with the 23mm Schneider and they are pretty damn nice. It's a problem putting RAWs up on the Luminous-Landscape site because of the size. Our download file system has a max of 100MB and the RAWs exceed that. Our IT guys are looking into seeing if it is possible. Chris and I will report back plus we'll have a more in depth report in the weeks to come. If you can make it to Indianapolis and I'm around you're welcome to give it a try with me.
Team,
I have been playing with the city raw sample image in C1 Pro 9.02. I'd be curious to hear what settings you guys are using for the IQ3 100 files?
Personally, here is what I did:
- ICC profile: IQ3 100mp Outdoor daylight
- Curve: High contrast
- WB: 7750, Tint: 24
- exposure: +0.3
- Contrast: 15
- Brightness: 20
- Saturation: 5
- Hilight recovery: 60
- Shadow: 30
- Clarity: 30
- Structure: 5
- Sharpening: amount 400, Radius 0.3, Threshold: 0.8
I did also apply verticals correction and a few layers to selectively darken the sky.
It looks much better to me than the initially provided settings. What did you guys do if I may ask?
Thanks.
Cheers,
Bernard
so much contrast :o
Would you be able to post on Drop Box and simply provide the link for anyone interested?
Hi,
You take make posting the wrong way. Doug said that the Hasselblad lenses were not good enough for the new sensor. So I made the point that they worked fine with multishot, which yields 200 M. But you are dead wrong that if a lens is good enough for 50 MP it will also be good enough for 200 MP multishot. The other way around is quite true.
Just consider this, you cannot resolve something by multishoot that is not projected by the lens.
@DT
[\* hypothetical mode on]
Doug,
Assuming I'd bought your IQ380 XF a few months ago (packaged as a kit with the release of the XF) , how much would it cost me today to crossgrade /upgrade to a CMOS IQ100 ?
No... it's not (the other way around)...
You don't count the (physical) sampling... (meaning that you think of it as if it was a single shot)... With multishot you practically quadrable the Niquist limit... (because you are shooting on 4 image areas that are in different position to each other...) For example, if one is using a Sinarback 54H in 16X mode and uses the Zeiss Contax 120mm Apo micro with it, the result (if one masters the process well) should be the same as if shooting (a single shot) on a 10x7,5cm (aprox) image area, but with 9μm pixels of true color (88mp in total) and has used a 240mm lens that is of the same LPM resolution & contrast as the 120mm has (but with an image circle as to cover the whole 10x7.5cm image area)....
EDIT: Imagine the above? ...you then know what "jaw drop quality" really is....
How much shift is actually needed for the 32HR to keep its relevance as a 645-fullframe architecture lens? I guess Chris or any other experienced architecture photographer can give some good insights on that.
Some more user-contributed tech cam samples over at getdpi. Large shifts on the 32HR is certainly a no-go due to residual color cast. But due to the offset microlenses it's not possible to state an exact limit without stepping, eg try 5, 7, 10, 12, 15, 20mm. The limit may also be different in portrait and landscape orientation.
If you have a IQ260 now with your 32HR upgrading to a IQ3 100 will for sure strongly reduce the shift range.
How much shift is actually needed for the 32HR to keep its relevance as a 645-fullframe architecture lens? I guess Chris or any other experienced architecture photographer can give some good insights on that.
I remember others asking the same...
But less shift is required because of the wider angle of view on the IQ3 100MP, isn't it?
Well, I don't, and now I'm asking!
It's a perfectly reasonable question, pricing probably set by Phase One and certainly 'relevant' .
My inbox today is inundated with emails from dealers offering upgrades on the new Nikon D5/D500 with clear upgrade pricing - no reason why Phase or their dealers should be recalcitrant.
Over to you, Doug.
But less shift is required because of the wider angle of view on the IQ3 100MP, isn't it?
That would be true if the IQ3 100MP was a larger sensor but it is not. It is 54x40mm, same as the IQx60 and IQx80 backs.
Ah, I thought they were physically smaller, my bad.
Cheers,
Bart
It depends on how you see it. The way I see it a lens with a certain sensor size provides a certain field of view, and for that combination to be used to its fullest in terms of image composition it must be able to make certain amount of shifts relative to the sensor size so the "horizon" can be placed low/high enough => larger sensor larger shift, but you can also use longer focal lengths. The 32mm is very wide, I'd satisfy with 40mm as the widest for my landscape, but architecture photographers needs really wide stuff.
In architecture photographers use ultrawide not because they want to, but because they must due to limited space.
Hi,
I am not a Hasselblad Multishot expert, but my understanding is that for 200 MP you do 16 exposures while shifting the sensor a half pixel. Now shifting the sensor a half pixel between exposures gives you for point samplings per pixel. So, what happen is that you split the pixel in four subpixels, each half the size of the original pixel. You don't loose full well capacity as you shoot four different exposures.
But, what happens is that you effectively reduced pixel size. You actually gain a lot, because you also take 16 exposures and can eliminate much of the Bayer interpolation, definitively a good thing.
Yes, you double Nyquist limit, but it is by virtue of emulating half the pixel size. So, to deliver 200 MP, your lens actually need to resolve those 200 megapixels, it is as simple as that, and if they do that, they will work just fine with an 100 MP back.
Best regards
Erik
So maybe a slightly different workflow can also be an alternative.
Yes I understand, but there are other possibilities than using optics that create issues that cannot be accommodated for. A simple stitch of two or three images can sometimes solve a lot. It allows to use the center of the image circle (is better quality), perhaps use longer focal length (is usually better image quality), and the stitching software can allow to use a projection method that preserves certain straight lines/edges, yet compress the extreme stretching to a perceptually more normal proportion. Leveling and squaring surfaces is also much easier and more accurate than possible with a limited preview screen size, and saves time on location (less variation in outside light conditions).
So maybe a slightly different workflow can also be an alternative.
Cheers,
Bart
I think this is already a common workflow in certain situations for many AP's - though only through using a shift lens to stitch. ie - using the 17mm TSE for an ultra wide shot.
Whilst it is a viable method, it comes with it's own set of difficulties as compositing of exposures is also common practice. ie - time on the computer goes up!
In architecture photographers use ultrawide not because they want to, but because they must due to limited space.
I stitch a lot, direct stitch where you are shifting the lens or back, here you can, somehow, easily combined different frames of the images so you can place people or not for instance....but if you start stitching around the nodal point, without shifting, as suggested here, you are asking for all kind of troubles with lines not lining perfect up and so on. If you have all the time in the world you can maybe make it work, but in a commercial workflow - that would be a challenge.
...a highly effective stitching solution could be made...
It's all about lines... ;-)
There is zero issues stitching perfectly lines to the pixel when rotating a good lens around it's entrance pupil when using good pano software such as PTgui. This is simply a non issue.
IMO, the best camera to shoot architectural interiors is by far the Roundshot 220VR... One shoots 120mm (or 220 if he needs ultra panoramic look done with telephoto) film with it, but with a Nikkor 17-35mm f2.8 used at only two or three focal lengths (say 21, 27 & 35 - nodal point pre-found and noted for ever), one can shoot whatever wide angle he wants, at whatever height he wants and as for detail... well... you'll never (and I mean never) miss some... For exteriors one may need the Zeiss 15mm f2.8 additionally and the shift adjustment added (usually a standard), but then the result will be very hard to beat and there will be no scene that wouldn't prove "a piece of cake" for the system... Killer camera! It starts where hi-end photography ends....
What a useless discussion - this back will sell to people who know it will work for them, and amateurs. A pro with CA or eveness issues is not going to order one of these 50K units in the hope she might find a lens that works when shifted.
What a useful comment! ;)
Cheers,
Bernard
I feel so honored talking to people who buy $50K gear ;D
As of now, people who are considering buying 50K$ gear. Which is the whole point of this discussion.
You apparently think that the only people buying such equipement are super rich and therefore don't really care about the value delivered since the cost is insignificant, I have no problem with that being you opinion.
But I would just appreciate if you could avoid contributing turning this into a self realizing prophecy by not giving this camera the chance to prove its ability to perform up to its price tag. Which you are trying to do by killing a relevant technical discussion.
Cheers,
Bernard
Dear Bernard -
You don't need to tell others what I think, I can do the telling quite well.
I think institutions looking to do reproduction will buy the 100MP as a safe one-click hi rez digitizer, no surprises.
I think pro photographers will do a very careful check whether the back works for them; if they know about shifting, then they will know about testing the shift before they buy.
And lastly, I think that anyone who believes they can predict the shift properties, CA, vignetting crosstalk etc of an unseen sensor with no datasheet and unqualified coverglass,when combined as a photographic system with lenses yet unused on this sensor, and make relevant contributions on this topic that others will base a $50K buy decision on is ... overenthusiastic.
My take on the new Phase is that it is a safe buy but probably overkill for institutions, has in fact an affordable price for pros who have the customers who need the resolution - but they will need to check it out- and will make amateur buyers happy. My take on the tech forum contributions of late is a lot of people are talking in huge tech detail about topics they don't really know much about, and not enough about stuff they are good at, like in your case I would be interested in a discussion about how one gets good at stitching.
Edmund
The latest tests of the tech camera lens combo is over at getdpi, and it now looks more promising than before. A recent version of Capture One had LCC bug which made some of the initial tests look bad.
My current guess is that I think it will be popular among tech cam users, perhaps more so than the IQ250.
Thanks for the update. These look like good news. I guess that Doug's results will probably confirm these findings with the Rodenstock 40mm lens. Still interested in seeing how the 32mm behaves though.
I'd also be interested to see some results with lenses being tilted. Could such tests be made also?
I agree Torger. Bernard, tilting does not change the angle at which light rays hit the sensor pixels.
...I believe that tilting downwards for typically landscape application does move the sensor closer to the edge of the image circle...
Bernard,
It does, which slightly changes the composition. However, a rise/fall sdjustment will counter the effect, while preserving tilt.
Anders,
One concerning item in the new posted samples is that purple/green glow around blacks. Do you think this is also related to the C1 bug or something else?
Chris Barrett and I will be shooting tomorrow with the ALPA system and numerous lenses. I just did some tests shots with the 23mm Schneider and they are pretty damn nice. It's a problem putting RAWs up on the Luminous-Landscape site because of the size. Our download file system has a max of 100MB and the RAWs exceed that. Our IT guys are looking into seeing if it is possible. Chris and I will report back plus we'll have a more in depth report in the weeks to come. If you can make it to Indianapolis and I'm around you're welcome to give it a try with me.
Any samples with the above mentioned 23mm Schneider...I assume it is the 24mm?
Kevin has the 23mm Rodie HR. I don't have any samples. He didn't employ any shifts, but from what I saw, that lens is TOTALLY usable on the 100 when centered.
We'll have sample images from the 23mm to show in a follow up article soon. The LLC part of C1 has a small bug in it so waiting fro Phase One to correct that. We can then do a final process on the 23mm. Thanks Chris for the great day, as always was fun shooting and chatting with you.
Well, Kevin and I spent yesterday with the new back and my Sony grabbing some interiors of really beautiful spaces. It wasn't an ideal test, as neither of us had the lenses that you would really want to use with this back for shooting architecture.
I've also had a look at the tech cam tests on GetDPI and was glad to see those.
I'm going to do a write-up over the next few days with images, but wanted to drop my initial thoughts. The Sony holds up decently against the digital back, decently. I've always known that the A7r files we're not as pretty but felt that they could be worked in post to get where I needed them. When you start pulling up shadows, however, The Phase really seems to shine and exhibits less noise. It also has better color, micro-contrast, DR, and sharpness out of the box.
I don't have any great wides left from my tech cam kit but still had this Rodie apo-sironar digital 45mm kicking around so I used that for both these shots. I didn't employ any movements, because I didn't feel the lens would hold up to it.
So, you can definitely get the nicest files out of the digi-back. It is, however, way less convenient to work with and from what I've seen, there will be less shifting available than there was with my IQ260... no matter what lenses you buy. I'd really love to make this back work for me, but I just don't know if I can live with less movements. As a 20+ year veteran of professional architectural photography, I'm beginning to feel like my dream system may never exist.
More to come...
IMHO
-CB
Both images on the MF2, Rodie 45mm, f/11, 2s, base settings with 50/50 hilight shadow recovery (just for yucks)
The Phase feels sharper, but focus could have been off a touch on the Sony, so nothing conclusive there.
(https://scontent.ford1-1.fna.fbcdn.net/hphotos-xfp1/t31.0-8/920861_1043583955664143_7422124833252695318_o.jpg)
A problem for the tech camera genre is that pixel technology is merging to be the same in 135 and MFD.
Before medium format sensors where in a zone where they could both have a bit higher pixel count and at the same time a larger pixel, and they would also allow themselves to waste some high ISO performance by not having micro lenses.
This allowed medium format sensors to function with lens designs that was impossible on the smaller formats, which is symmetric or weak retrofocus wides, providing a unique feature set, near distortion-free lenses with a large range of movements.
Now with pixel sizes just as small on MFD like on 135, and the same desires regarding ISO performance, the same limitations regarding lens designs is there which means that lenses must be designed in the same way as for the smaller formats.
It's quite obvious that both Rodenstock Digarons and Schneider Digitars are both obsolete in terms of design when it comes to supporting the new MFD king, Sony, but thanks to the extremely good tonality and dynamic range of the Sony some of this over-capacity can be used to do the impossible. Due to subject- and condition-dependent gradual degradations of image quality in shifted areas its hard to quickly get a grip of exactly how these systems perform. I'm myself thrown between hope and despair. It will for sure make images on the wides with quite large shift range, but when does the subtle color degradations become "too much"? That's the hard part...
MFD surely has got better and more all-around with CMOS, but it has also lost some of its originality. It's not really CMOSs fault as such, but rather the inability to focus on other properties than megapixels and dynamic range. Had they made a 60MP version put in some light shields in the design, and perhaps dropping or at least reduce the aggressiveness of the microlenses I think it could have been the most tech cam friendly sensor ever. When BSI then arrives they could reduce well depth and then start increasing pixel count again without losing angular response (when you have deep wells you need large pixels to have good angular response, it's simple geometry)
Hi Anders,
I am not so sure about the thing about well depth thing.
I would also guess that Sony is going after the main market and that is probably MF DSLRs. Personally, I feel MF needs to go mirror less. After all, who needs that flipping mirror just to show a mediocre image on a mediocre ground glass. The idea I really like is a technical camera with a high quality EVF, magnified LV and focus peaking. But sorry, if I cannot afford I will not buy…
Yes it's a main market thing. Tech cam wide photography with large shifts, eg architecture photography, is just too small a niche to care much about. If you can only make one sensor, make it for the key segment and try to shoehorn it into the others. And it is very much shoehorning going on right now in the tech wide tests I can assure you... not sure yet how that will end.
While I agree that on tech cams ground glass image is mediocre due to the high vignetting lenses and low angle of incoming light, I don't think the mirrorless discussion apply to the MF DSLRs. The image you see in the optical viewfinder is good, just like on the 135 Pro DSLRs. It's even better in MF as it's larger. As you're using auto focus you don't have the problem of messing around with a loupe on it either. I think the typical studio photographer, which is the main customer segment, will want to use an optical viewfinder for at least a handful more years so I don't think we'll see mirrorless for the H or XF system anytime soon. I also think Canon/Nikon's flagship models will stay optical viewfinder for quite long, as the key market of pro customers simply like it, and there's some technical reasons too related to the possibility to have a dedicated auto focus sensor which is currently performing much better for telephoto than on chip phase detection.
and there's some technical reasons too related to the possibility to have a dedicated auto focus sensor which is currently performing much better for telephoto than on chip phase detection.
Could someone please make a new thread titled "Far Fetched Camera Ideas and Workflows That Have No Practical Value in the Real World" and move 2/3 of these posts there?
As a 20+ year veteran of professional architectural photography, I'm beginning to feel like my dream system may never exist.
I'd love to hear at some point what the properties of such a dream system is.
Mine is pretty simple - a CMOS digital back that is designed to work to the edge of the image circles of the currently available Rodenstocks Digarons and Schneider Digitars. I can live with the LCC and lack of integration mess, but I really would love to be able to just care about the optics, its image circle and sharpness and not worry that the sensor can't handle my movement settings regardless what I set them to. And then to dream further I would love if Schneider with proper sensor support had stayed in and continued with their more traditional designs for us that prefer that instead of maximum-sharpness-at-all-costs type of designs which Rodenstock represent. They were really complementary brands, but Schneider was let down by the sensors making it impossible to make reasonable business with their lenses.
I guess that now we're pushed towards making center frames and adjusting perspective in post-processing. With the ultra-high resolution we're getting it seems feasible from a quality perspective. I haven't really thought about if that is a less productive way to work or not, but it surely is less pleasing as I see it.
Could someone please make a new thread titled "Far Fetched Camera Ideas and Workflows That Have No Practical Value in the Real World" and move 2/3 of these posts there?
I guess that now we're pushed towards making center frames and adjusting perspective in post-processing. With the ultra-high resolution we're getting it seems feasible from a quality perspective. I haven't really thought about if that is a less productive way to work or not, but it surely is less pleasing as I see it.
I'm also not happy about Phase Ones solid silence about these issues and that they never mention crosstalk -- crosstalk is an uncontroversial standard term in image sensor design and it's obvious that the special issues we have here is related to that. It's easily explained too. I don’t understand why Doug and other representatives are so reluctant to using this term. It's almost like they want photographers not know so they don't know what to look for and therefore buy without knowing what compromises they make when shifting.
I'd love to hear at some point what the properties of such a dream system is.
Mine is pretty simple - a CMOS digital back that is designed to work to the edge of the image circles of the currently available Rodenstocks Digarons and Schneider Digitars. I can live with the LCC and lack of integration mess, but I really would love to be able to just care about the optics, its image circle and sharpness and not worry that the sensor can't handle my movement settings regardless what I set them to. And then to dream further I would love if Schneider with proper sensor support had stayed in and continued with their more traditional designs for us that prefer that instead of maximum-sharpness-at-all-costs type of designs which Rodenstock represent. They were really complementary brands, but Schneider was let down by the sensors making it impossible to make reasonable business with their lenses.
I guess that now we're pushed towards making center frames and adjusting perspective in post-processing. With the ultra-high resolution we're getting it seems feasible from a quality perspective. I haven't really thought about if that is a less productive way to work or not, but it surely is less pleasing as I see it.
Okay, if I assume that RawDigger (RawDigger-1.2.3.435-x64.exe) reads the ISO 50 sample file correctly, then this is what I find:
ReadNoise (standard deviation)
R=5.04
G=5.04
B=5.03
G2=5.02
Blacklevel
Offset=1024 (automatic, the mean of the masked pixels is 1025.7, 1025.2, 1025.4, and 1025.4)
Saturation
R=65532
G=65532
B=65532
G2=65532
So for Dynamic Range I get for the green channels Log((65532-1024) / 5.03) / Log(2) = 13.65 stops dynamic range, instead of 15 stops. I'm not sure if the camera's 14 or 16 bit mode was used for this shot, but 13.65 stops is still impressive.
Again, this assumes that Rawdigger reads the file data correctly, and that the masked photosites contain only read-noise (QED).
Cheers,
Bart
Now that we have received our unit and have done a shoot with 16-bit mode turned on I can confirm the previous raw files (ISO Sweep of the city) was done in the 14-bit mode.
You can download 16 bit raws from our shoot here: https://digitaltransitions.com/first-phase-one-100mp-fashion-raw-files/
Hi Anders,Possibly, guess we’ll see if Phase wants to make a new body and new line up of lens that eliminates the mirror box. As it stands now, I don’t know if any company has the interest in expending R&D to develop a mirrorless MF system to replace the current dSLR systems. Tech cameras are great, but for many they don’t fit their shooting needs (either from a personal or a practical point of view) Personally for some things an EVF still isn’t great ... the new XF viewfinder and fast focusing along with focus calibration is pretty sweet. With the LS lenses, i haven’t seen mirror vibration an issue, even with the 240 LS on a tripod but not mirror up mode. still need to do more testing on that though.
Your points are well taken, but the mirror causes vibrations. Shooting with my Planar 120/4 on my Hasselblad 555/ELD at say 1/125 without mirror lock up just produces crap even if I am shooting on top of an RRS Versa 3S with an ArcaSwiss D4 head. I also think that phase detection combined with contrast sensing AF can give superior results, unless shooting fast action, and I don't see a lot of MFD at Formula One or horse show jumping championships.
Now, it is very well possible that new cameras have a better damped mirrors. The new Phase One camera has an almost vibration free LV mode. But I think the SLR concept belongs to history. A magnificent piece of history, but history nevertheless.
Best regards
Erik
Possibly, guess we’ll see if Phase wants to make a new body and new line up of lens that eliminates the mirror box. As it stands now, I don’t know if any company has the interest in expending R&D to develop a mirrorless MF system to replace the current dSLR systems. Tech cameras are great, but for manu don’t fit their shooting needs (either from a personal or a practical point of view) Personally for some things an EVF still isn’t great ... the new XF viewfinder and fast focusing along with focus calibration is pretty sweet. With the LS lenses, i haven’t seen mirror vibration an issue, even with the 240 LS on a tripod but not mirror up mode. still need to do more testing on that though.
Hi,
It looks interesting, at least in my version of RawDigger. The low numbers definitively look to be 14-bit as only each fourth bit contains data. Going up to say 100-200 levels the gaps start to fill in.
I have the latest version of RawDigger, but Phase One cameras above IQ-280 are not officially supported it seems. So it can be that RawDigger would interpret something wrong. Still it is very odd.
A small comment to Doug, it seems that focus is bad on the f/10 image, as the f/14 is far more distinct.
Best regards
Erik
Hi,
It looks interesting, at least in my version of RawDigger. The low numbers definitively look to be 14-bit as only each fourth bit contains data. Going up to say 100-200 levels the gaps start to fill in.
I have the latest version of RawDigger, but Phase One cameras above IQ-280 are not officially supported it seems. So it can be that RawDigger would interpret something wrong. Still it is very odd.
A small comment to Doug, it seems that focus is bad on the f/10 image, as the f/14 is far more distinct.
Best regards
Erik
I recently added a Olympus OM-D E-M5 Mark II to my setup, and I really like its "digital keystone compensation" feature, which lets you correct perspective (both vertical and horizontal) live on the camera screen (or viewfinder) while shooting.
then it will end up in RawDigger eventually.RD people got a lot of P1 raw files and it seems looking into that now... at least that is what I read a day before yesterday blog-wise.
> RawDigger format reader is based on dcraw
Actually, on LibRaw, and that is very different thing :)
> Not really, as the IIQ decoder in libraw is based on dcraw, right?
In the same sense computers are based on silicon, yes, it is :)
Hi, you will need to use Capture One v9.0.2 or later to convert the 16-bit IIQ file into DNG first. RawDigger v1.2.3 does not support the 16-bit IIQ file.
See my post here: http://www.getdpi.com/forum/medium-format-systems-and-digital-backs/57312-iq3-100mp-technical-camera-tests-color-cast-mazing-artifact-tiling-issue-dr-etc.html#post676386
Thanks for info!
Raw digger is now fixed for IQ3 100MP!
Thanks for info, stuff from you is always good!
Best regards
Erik
Hope this is not too far off topic (and has not already been mentioned) but was just wondering if this sensor uses the same Sony coverglass that is found in the A7RII?
Also, I know that everyone seems most keen to know how the tech wides work on this back but I would be really interested for Alpa to showcase some examples with either the Hasselblad H lenses or Contax 645 with their new shift adapter...
Due to both of the above designs, a lot of the issues being discussed here (and most going right over my head to be honest) would be less of an issue right? As somebody who does not own a MFDB or any tech lenses, I was amazed to see the original files (before LCC correction) of those samples being displayed by Alpa..
As a newbie, I think using a lens that does not need a LCC correction would be appealing (at a loss of sharpness I am sure)..
Looking forward to all the future test that are sure to come :)
One last question, just wondering how the P45+ handles the discussed issues with tech wides? Would it be a good one to aim for if I was happy enough with the resolution it offers.. (which is enough by the way).
Thanks a lot for any advice or input!
Looking at shadow recovery on the IQ3100 vs the A7r2, I'd agree the Phase is cleaner.
CB
Looking at shadow recovery on the IQ3100 vs the A7r2, I'd agree the Phase is cleaner.
CB
Hope this is not too far off topic (and has not already been mentioned) but was just wondering if this sensor uses the same Sony coverglass that is found in the A7RII?
Also, I know that everyone seems most keen to know how the tech wides work on this back but I would be really interested for Alpa to showcase some examples with either the Hasselblad H lenses or Contax 645 with their new shift adapter...
Due to both of the above designs, a lot of the issues being discussed here (and most going right over my head to be honest) would be less of an issue right? As somebody who does not own a MFDB or any tech lenses, I was amazed to see the original files (before LCC correction) of those samples being displayed by Alpa..
As a newbie, I think using a lens that does not need a LCC correction would be appealing (at a loss of sharpness I am sure)..
Looking forward to all the future test that are sure to come :)
One last question, just wondering how the P45+ handles the discussed issues with tech wides? Would it be a good one to aim for if I was happy enough with the resolution it offers.. (which is enough by the way).
Thanks a lot for any advice or input!
Not only that, since the IQ3 has more than double the MP, at equal (very large) print sizes the IQ3 file is gonna look MUCH nicer.
Yes, it seems the D810 gets the most DR out of the sony 35mm sensor.
Low iso performance is definitely a priority on a back like the IQ3 100MP while the A7RII being a much more consumer oriented product is tuned more for high performance across the whole iso range with obvious bumps (compared to the D810) in performance at several iso settings.
I have no clue what each company does on top of what the sensor can deliver from the factory or if some changes are made inside the sensor itself or if its a combination of hardware and programming (on sensor).
I don't think anyone without a signed NDA really knows.
I though Sony improved the depth and DR of the A7RII with the new firmware? and got it to 14bit.
I've done a bit more Read-noise and Dynamic Range number crunching based on some more files, and at different ISOs. The conclusion is that the 16-bit lossless mode does add DR compared to the 14-bit 'lossless' mode, but it remains a bit below the claimed 15 stops. Also, there is no real difference in readnoise between nominal ISO 50 and ISO 100. Read noise for ISO 3200 and above remains unchanged, so becomes basically ISO-less.
Attached are the summarized 14-bit mode shot results of the 'Roundtower' sequence. The numerical differences between the R/G/B/G2 color planes are marginal, they all show very similar values.
I've used RawDigger Version 1.2.4 Build 437 (which is supposed to read the 16-bit mode correctly) to read the files. The DR of 13.65 stops at nominal ISO 50 in 14-bit mode, change to a DR of 14.39 stops in 16-bit mode (not quite the claimed 15-bits but still very high).
Read noise drops from a standard deviation of 5.03 DN to 3 DN, at ISO 50 (and I assume at ISO 100, but I have no file to verify yet). Base ISO therefore seems to be nominal ISO 100. We'll see when DxOlabs publish their analysis how they rate the real ISO sensitivity.
At ISO 800 the DR changes from 11.66 stops in 14-bit mode, to 12.06 stops in 16-bit mode.
The observed 4-DN combing of the histograms in 14-bit mode, indeed disappears in 16-bit mode, so the electronics do seem to produce 16-bit native signal levels, with a noise floor in the 1.61 LSBs (least significant bits).
Low ISO read-noise is very low, and DR is impressive, assuming these files are representative for a larger population of sensors.
Cheers,
Bart
Hi,
I guess that Sony/Phase One really pulled that rabbit out of that hat!
As far as I recall, Nikon D810 was the camera with the highest DR (at the pixel level), and DxO measured that camera to 13.67EV, so this reaches 14.39 (according to your calculations) and that is quite a bit more.
On the other hand, I have noticed that D810 was close to 14 bit limit and DR. Going to 16 bit may have been needed to fully utilising those sensor pixels.
Seems like Phase One put together a really impressive back!
Just to add, thanks for sharing your findings!
Something else I noticed earlier with a DCamProf profile for the EOS 5Ds that 'AlterEgo' made available in that application's main thread, is that the profile may have a significant (in that case positive) impact on the demosaiced and rendered noise. So the Raw conversion quality also makes a difference.
That was an "aha"-experience during the development of DCamProf. It's not a good idea for the profile to push saturation of darkest shadows a lot, as it just makes the chroma noise more visible....
Yes, it seems the D810 gets the most DR out of the sony 35mm sensor.
Low iso performance is definitely a priority on a back like the IQ3 100MP while the A7RII being a much more consumer oriented product is tuned more for high performance across the whole iso range with obvious bumps (compared to the D810) in performance at several iso settings.
I have no clue what each company does on top of what the sensor can deliver from the factory or if some changes are made inside the sensor itself or if its a combination of hardware and programming (on sensor).
I don't think anyone without a signed NDA really knows.
I though Sony improved the depth and DR of the A7RII with the new firmware? and got it to 14bit.
The new uncompressed raw mode does not improve the EDR of the a7RII.
http://blog.kasson.com/?p=12287
Even at ISO 100, the camera has sufficient read noise to adequately dither the ADCs at 13 bits precision.
Jim
That was an "aha"-experience during the development of DCamProf. It's not a good idea for the profile to push saturation of darkest shadows a lot, as it just makes the chroma noise more visible....Could some please provide a link to this thread? I had a look but couldn't seem to find it.
I've done a bit more Read-noise and Dynamic Range number crunching based on some more files, and at different ISOs. The conclusion is that the 16-bit lossless mode does add DR compared to the 14-bit 'lossless' mode, but it remains a bit below the claimed 15 stops. Also, there is no real difference in readnoise between nominal ISO 50 and ISO 100. Read noise for ISO 3200 and above remains unchanged, so becomes basically ISO-less.
Attached are the summarized 14-bit mode shot results of the 'Roundtower' sequence. The numerical differences between the R/G/B/G2 color planes are marginal, they all show very similar values.
I've used RawDigger Version 1.2.4 Build 437 (which is supposed to read the 16-bit mode correctly) to read the files. The DR of 13.65 stops at nominal ISO 50 in 14-bit mode, change to a DR of 14.39 stops in 16-bit mode (not quite the claimed 15-bits but still very high).
Read noise drops from a standard deviation of 5.03 DN to 3 DN, at ISO 50 (and I assume at ISO 100, but I have no file to verify yet). Base ISO therefore seems to be nominal ISO 100. We'll see when DxOlabs publish their analysis how they rate the real ISO sensitivity.
At ISO 800 the DR changes from 11.66 stops in 14-bit mode, to 12.06 stops in 16-bit mode.
The observed 4-DN combing of the histograms in 14-bit mode, indeed disappears in 16-bit mode, so the electronics do seem to produce 16-bit native signal levels, with a noise floor in the 1.61 LSBs (least significant bits).
Low ISO read-noise is very low, and DR is impressive, assuming these files are representative for a larger population of sensors.
Cheers,
Bart
Could some please provide a link to this thread?the main DCamProf topic = http://forum.luminous-landscape.com/index.php?topic=100015.0
Hi Jim,
In my simple tests I did not see an improvement in dark noise. But I have noticed that dark tones that were pushed 4-5 stops remained neutral grey with the new uncompressed formate but went like brownish with the old formate.
So I felt the new file format yielded more neutral colours in extreme pushing. I did not come up with an explanation for the effect.
I have some samples, if needed, but it would take some time to find.
In my simple tests I did not see an improvement in dark noise. But I have noticed that dark tones that were pushed 4-5 stops remained neutral grey with the new uncompressed formate but went like brownish with the old formate.
So I felt the new file format yielded more neutral colours in extreme pushing. I did not come up with an explanation for the effect.
A 4.6 micron pitch will mean that at f/5.0 or narrower, diffraction will start to function as an OLPF of sorts, to reduce aliasing effects even further than the dense sampling already does, and apertures of f/16 or narrower will lose effective resolution due to diffraction...
Bart,
Setting aside advice not to shoot at an aperture of f/16 or narrower for exactly that reason, what can one say if one does? Does deconvolution sharpening return zero benefit?
Does colour reflected by a sensor's physical surface visibly change as pixel density increases, similar to the way in which colour reflected by CD/DVD/Blu-Ray visibly changes as pit density increases?
When MTF at Nyquist reaches zero, no more detail can be recovered regardless of sharpening method. But, visual perception is dominated by low frequencies and they are still improved by sharpening.
That is angle dependent. So different spectral line will be have maxima at different angles.
Erik,
I did wonder. Thank you.
In layman's terms, is that a "Yes"?
That was an "aha"-experience during the development of DCamProf. It's not a good idea for the profile to push saturation of darkest shadows a lot, as it just makes the chroma noise more visible....
Hi Anders,
Re. chroma noise,
I attended a session of the phone camera quality group of the IEEE, and they did some user testing. Surprisingly, consumers prefer chroma noise to luma noise. Photographers do not, the result was viewed as correct but counterintuitive by group members after the user tests were repeated. This may influence a choice of profile creation.