Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: peterv on March 07, 2019, 12:13:20 pm
-
Hi everyone,
I have the opportunity to go from an S2-P to an S007 now, or wait another year or so and go for an S3. My question is not about what would be the wisest investment or whether these cameras are worth their money. I would just like to know what fellow S-users expect the difference in image quality will be between the two CMOS S cameras.
I just read this (https://www.reddotforum.com/content/2018/09/photokina-2018-the-leica-s3/) interview with Toni Felsner, product manager for the Leica S3, and the following text caught my eye:
Conventional wisdom says that as resolution increases and pixel size decreases, dynamic range and high ISO performance suffer. Is this the case with the new sensor in the S3?
Absolutely not. We stayed at 37.5 megapixels for as long as we did in order to focus on delivering maximum image quality. Top DR. Excellent high ISO. But, sensor technology has evolved and we are now able to achieve both an increase in resolution and, at the same time a two stop improvement in noise performance, and even greater dynamic range than the S007.
Wait. Better dynamic range than the S007? With over 15 stops of range, the S007 is incredible. It’s changed the way I approach landscape shooting.
Yes, we are going to be able to offer even more. Wait until you see it.
That’s incredible. Any other cool tech on the sensor side?
So the sensor features a dual gain structure, which allows us to maximize higher ISO performance. In our test shooting, ISO 6400 looks super clean. Almost no noise whatsoever. And, the noise profile on the files at even higher settings looks less like noise, and more like grain.
Also, we implemented a new color filter array, which is going to offer really amazing color. We will tune the output for skin tone, but it will work extremely well for a wide variety of applications like landscape, architecture, fine art, etc. You’ll especially see an improvement in the red channel, which has always been challenging for digital photographers when photographing bright reds, like automotive paint, or flowers in the sun.
This sounds pretty confident that the S3 is going to be a nice, meaningful upgrade as far as image quality is concerned.
What do you guys expect from the IQ improvement in the new S3?
-
Hi,
Hard to say, the S3 will have a new sensor, probably from TowerJazz. It belongs probably to the same family as the sensors in the new Panasonic S1r.
Dual gain conversion improves DR, above some ISO setting. It does not help with Signal/Noise ratio.
Reducing pixel size doesn't really matter, true each pixel can capture less photons but there are more pixels and that compensates with regard to shot noise (what you can see in mid tones). DR may suffer a bit, doubling the number of pixels would reduce DR a bit, like 1/2 EV, but smaller pixels have often lower readout noise.
It is hard to know how good the new sensors are.
Best regards
Erik
-
Hi,
Hard to say, the S3 will have a new sensor, probably from TowerJazz. It belongs probably to the same family as the sensors in the new Panasonic S1r.
Dual gain conversion improves DR, above some ISO setting. It does not help with Signal/Noise ratio.
Reducing pixel size doesn't really matter, true each pixel can capture less photons but there are more pixels and that compensates with regard to shot noise (what you can see in mid tones). DR may suffer a bit, doubling the number of pixels would reduce DR a bit, like 1/2 EV, but smaller pixels have often lower readout noise.
It is hard to know how good the new sensors are.
Best regards
Erik
Erik, Have you factored fixed pattern noise into your thoughts about the effects of pixels size?
Edmund
-
No Edmund,
I have not.
You are most welcome to elaborate.
Best regards
Erik
Erik, Have you factored fixed pattern noise into your thoughts about the effects of pixels size?
Edmund
-
Yes please do, I’d like to know more about dual gain structure sensor technology and its advantages and or disadvantages.
-
Hi,
Much simplified, a pixel has photodiode that converts incoming photons (particles of light) into free electrons. These electrons are contained in a capacitance. The voltage over the capacitance is the charge / capacitance.
So, a small capacitance yields a higher voltage, which is good to reduce readout noise. But we also want to be able collect as many electron as possible, as long we have plenty of light, as high electron count gives good signal to noise ratio (SNR). SNR is simply the square root of the number of electrons collected at high exposures.
So, high capacitance -> god SNR in good light, but noisy read out.
But if we raise ISO, we will not capture so many photons, which means that we cannot use the full capacitance. Aptina has patented a solution where part of the capacitance is disconnected using an extra transistor. So, that reduces readout noise, normally visible as a small bump in the DR vs ISO plots.
Best regards
Erik
Yes please do, I’d like to know more about dual gain structure sensor technology and its advantages and or disadvantages.
-
Thanks for the clear explanation Erik, much appreciated!
-
Don't know about IQ improvement overall but I expect great things in the new release of the S3 why don't you think it will be this year? The new expected SL looks more promising for IQ and future features.
-
I too expect great things in the new S3, that's why I'd like to buy it, the reason why I think that won't happen this year for me, is because I don't expect to be able to afford one as soon as the S3 hits the market, which is expected to be somewhere during the spring. The price of the 007 on the other hand seems to be within reach, hence my question.
-
Thanks BC, you're absolutely right, there's always something new just around the corner. As a matter of fact, the announcement of the S3 is the reason I started looking for an S007. I figured the S007 price would be coming down this spring and I have been using an S2-P now for six years and I think I'm ready for a new S system camera.
I have just finished testing the S007 for one week and I was quite pleased with the results I got and it really is a completely different camera. Not in the way it handles ergonomically, that is all very familiar, but once the S007 is in your hands you feel that this is the second, much more sophisticated iteration. The mirror and the shutter are a lot quieter than the S2 and they sound like they were made for hundreds of thousands of effortless exposures. I love my S2-P but the mechanics of this newer S007 feel and sound much more reassuring and the same goes for the autofocus which is noticeably faster.
When it comes to colours, I too have a feeling that the out of the box colours from the CCD sensor are more friendly to my eyes than the newer CMOS 007. For instance in my test shots blue skies have a tendency to be some more cyan-ish on the S007 than the S2-CCD. Maybe that's just my eyes, I don't know, but there was nothing going on color-wise, that could not be easily corrected or altered to taste in Lightroom, the files are extremely malleable.
You mentioned your wish for a little higher ISO, and I must say I was very pleasantly surprised by how good 3200 ISO is. I am used to the look of the high ISO shots I make with the Sony RX1M2 which I use for reportage work and that camera has the same 42-Mpix full-frame CMOS sensor as the Sony A7R II, but the noise on the Leica S007 looks a whole lot more friendly and grain-like.
As for for the choice between an S3 and an S007, I (try to) sell fine art prints of my work and I'd really like to be able to make larger prints and that's where the 64MP from the S3 will come in very handy, the reason why I am a little bit on the fence right now about which camera to choose. And although I must say that I think I could be very happy with the S007 because it is really a big step up from my S2-P, the extra 26,5 megapixels would be nice to play with. And at the moment there are rumors (just rumors I need to emphasize) that the all new S3 will not be as expensive as the first S system cameras when they came to market because of the strong competition from Hasselblad and Fuji, around 10 to 15 thousand. So right now I'm leaning towards waiting for just a little bit longer to find out what the price difference will be.
-
As for for the choice between an S3 and an S007, I (try to) sell fine art prints of my work and I'd really like to be able to make larger prints and that's where the 64MP from the S3 will come in very handy, the reason why I am a little bit on the fence right now about which camera to choose. And although I must say that I think I could be very happy with the S007 because it is really a big step up from my S2-P, the extra 26,5 megapixels would be nice to play with. And at the moment there are rumors (just rumors I need to emphasize) that the all new S3 will not be as expensive as the first S system cameras when they came to market because of the strong competition from Hasselblad and Fuji, around 10 to 15 thousand. So right now I'm leaning towards waiting for just a little bit longer to find out what the price difference will be.
The competition from the 100mp Fuji will be a major threat indeed.
Fitted as it will be with a good EVF, IBIS, a probably superior 100mp Sony sensor, an integrated vertical grip great for portrait,... the Fuji seems much more appealing than the Leica unless you have existing S lenses.
Cheers,
Bernard
-
If you already have the S2, then you have the lenses to go with it, which tilts the balance to the S3.
-
Erik, thank you, your description of advancing chip architecture was lucid and brilliant.
My gripe is how journalists pander unthinkingly to PR and insult us with ridiculous statements disguised as questions.
"Conventional wisdom says that as resolution increases and pixel size decreases, dynamic range and high ISO performance suffer. Is this the case with the new sensor in the S3?"
Humbug! What a patsy question scripted by some PR goof to let Leica launch into a sales spiel! Maybe for older CCD's, but these are irrelevant from 2016+. Phase One's IQ3 100 which I use and now the IQ4 too, are just one leading manufacturers latest steps in a product line that have successfully obviated this supposed 'convention'. So, it's not conventional at all, as the exception dis-proves the rule, and it's hardly wise either because it bespeaks of ignorance of every step forward since 2016.
I respect Leica deeply so why do they fall prey to this foolery?
-
I know nothing about the fuji medium format camera, but I doubt that most photographers shopping for an S series would be comparing them to the Fuji. Maybe a Phase One, or a Hasselblad, especially this one.
Why would they not if they are indeed photographers (and not just people looking to been seen with a big Leica)?
The Fuji is pretty much superior in every way, IMHO starting with the lenses... and those are a lot cheaper.
Besides you can use C1 Pro to convert the Fuji files.
Cheers,
Bernard
-
Last week I was at Wex's flagship store in London (think of it as a smaller B+H) and I was talking to the salesperson as we were standing next to the Fuji display. I asked how well are the Fujis selling and he said very well. I said that's good for medium format and he replied no not the larger fuji, the aps C cameras.
IMO
BC
The Fuji MF series are selling steadily is my impression, especially to the landscape and travel crowd.
Fuji tell their fans to buy them for image quality, and Fuji have a lot of fans :)
Obviously, no one with a functioning high end Leica or Phase setup will want to convert, but if you are running a crop-factor Phamiya setup then moving to Fuji will give you a modern EVF body with good fast focus, the same sensor as a Phase 50MP, and keep you working with C1, at below $10K for a body and 2 or three lenses. The only missing feature seems to be central shutter. There's a Fuji Facebook group and one can see that people are doing perfectly good work with these. Leica they are not, but I guess they are now the Nikon of MF.
In my opinion, selling a Phamiya body and back and lenses and moving to Fuji is now almost a no-brainer for anyone who wants to stay in crop-MF and enjoy an EVF and CMOS sensor. This could change if Phase re-entered the market with mirroless.
On the other hand, the new Canon is rumored at 70MP, so maybe the crop MF party is gonna be over soon. I guess the A7R2 also killed a big chunk of the market a couple of years ago - at some point enough pixels are just enough pixels.
Edmund
PS I have a Fuji X100, and it is a camera with a distinct personality, not a generic. I think this is something Fuji are very smart about, this feeling of tailored design. Contrary to companies who thrive on secrecy, Fuji shop mockups and prototypes around from photofair to photofair and get them into people's hands, and it does show in the final product. There's a video around of a Fuji demo in Dubai, and what is amazing is the way they exhibit and pass around these mockups and prototypes, even lenses, whatever the risk of damage or design theft. You can see how people's hands fit around the bodies, and I am sure that is what the designers really watch.
-
Why would they not if they are indeed photographers (and not just people looking to been seen with a big Leica)?
The Fuji is pretty much superior in every way, IMHO starting with the lenses... and those are a lot cheaper.
Besides you can use C1 Pro to convert the Fuji files.
Cheers,
Bernard
this
i love that leica still tries to compete, in a way the S was the first step in the niche that the fuji and hasselblad now entered....
but it is not a competition at this point, the fuji is just so much better in every regard and so is the X1D which should also satisfy the "want", "brand", "jewelry" itch.....
even if the new S is a major step up from the last model, it might at that point compete with existing fuji and hapselblad 50mpix models....
but with the GF100? 100mpix? PD AF? IBIS? a sensor that is being pushed hard from all sides and directions in terms of DR, high ISO by upcoming FF models at much lower price point? i don't see the S coming close anywhere....
i find it funny that the first Q review(s) are coming out, i always thought the first Q was the best digital leica and a real contender all around....so i was really excited about this one....the sensor is (finally) actually competing across the board but now it just looks like that the weak point might be the (the horror!) leica lens....
-
now it just looks like that the weak point might be the (the horror!) leica lens....
Another demonstration that everything can happen once 8)
Edmund
-
Bernard, sorry, I was not mocking Leica for advancing their technical quality, simply that they chose as their blinding entrée something that has been currency for a long time. (ref, "..convention says...") When convention for the last three years at least has stopped saying that!
-
On the other hand, the new Canon is rumored at 70MP, so maybe the crop MF party is gonna be over soon. I guess the A7R2 also killed a big chunk of the market a couple of years ago - at some point enough pixels are just enough pixels.
You keep ignoring the possible benefits of a larger sensor (and in that comparison to an imagined future Canon 36x24mm sensor, both larger pixels and likely better pixel technology (Sony vs Canon) along with their being more of them). Are you convinced that the size gap from 36x24 to 44x33 is insignificant for IQ? I find that strange, since it is about as big or bigger than:
- 4/3" to "APS-C" (especially Canon's smaller version of "APS-C")
- 44x33mm to 54x40mm (from the "crop MF" that you disparage to "full frame MF")
- 645 to 6x6
- 6x6 to 6x7, and even
- 645 to 6x7.
In all those cases, there have been plenty of people arguing that the larger format has a significant advantage due to that larger size, and even putting there money where their mouths are.
The size gap 36x24 to 44x33 is more so for photographs in "vertical/portrait" shapes like 5:4 or 4:3, which require cropping away a larger fraction from the image recorded by the more "horizontal/landscape" oriented 36x24 format.
P. S. I also find strange Cooter's suggestion that people interested in the 3:2 shaped 45x30mm Leica S series would be more interested in that fictitious square format Hasselblad mockup/vaporware camera that an actual 4:3 format camera!
-
The Fuji is pretty much superior in every way, IMHO starting with the lenses...
You said a similar thing about the Leica SL lenses a while ago...
May I ask on what your opinion is based?
Did you ever try any Leica S or SL lenses?
-
You said a similar thing about the Leica SL lenses a while ago...
May I ask on what your opinion is based?
Did you ever try any Leica S or SL lenses?
SL lenses I don’t have a direct experience with. S lenses a very limited one.
I loved the sharpness and bokeh, was less impressed by CA. But I only tried 2 lenses if I recall (70mm and the 120mm macro I think).
They are sweet lenses but IMHO the Fuji lenses are a bit superior. This shouldn’t come as a surprise since they are more than 8 years newer and these 8 years have seen tremendous progress in terms of lens design.
Cheers,
Bernard
-
Im selling my like new S007 with 3 lenses 30mm 70mm and 120mm if your interested
Neil
-
BJL, I have a feeling you might be an engineer, and as their Telomeres run shorter some engineers realise that what gets made is defined by fashion and marketing as much as by technology.
The wider market of digital cameras is defined by pixel number; cameras get compared by one single number, just like guys get compared by ... what they earn. ;D
Now if you ask *me* what I believe, I would argue that the dev work which goes into the components is significant -within the limits allowed by physics. The problem the larger sensors often face is that their tech is cycled more slowly than the tech of the mass produced cameras. As a result, the intrinsic advantage of 2/3 stop in 44x33 gets confronted eg with a basic cell design that is 3 to 5 years more modern.
Thus the older device with more limited data throughput must fight a modern sensor which can source the data for computational photography. This can be seen with the currently employed crop 50MP which competes with 35mm sensors with all the belles and whistles like focus pixels and video frame rates.
By the time one is done counting the plus and minus points, the crop sensor is fighting for its life, while as you point out a larger devices can comfortably cruise on. Phase did the maths and ditched the crop products.
Unfortunately for those of us who like taking pictures, there are no reasonably large-sensor options left.
Edmund
You keep ignoring the possible benefits of a larger sensor (and in that comparison to an imagined future Canon 36x24mm sensor, both larger pixels and likely better pixel technology (Sony vs Canon) along with their being more of them). Are you convinced that the size gap from 36x24 to 44x33 is insignificant for IQ? I find that strange, since it is about as big or bigger than:
- 4/3" to "APS-C" (especially Canon's smaller version of "APS-C")
- 44x33mm to 54x40mm (from the "crop MF" that you disparage to "full frame MF")
- 645 to 6x6
- 6x6 to 6x7, and even
- 645 to 6x7.
In all those cases, there have been plenty of people arguing that the larger format has a significant advantage due to that larger size, and even putting there money where their mouths are.
The size gap 36x24 to 44x33 is more so for photographs in "vertical/portrait" shapes like 5:4 or 4:3, which require cropping away a larger fraction from the image recorded by the more "horizontal/landscape" oriented 36x24 format.
P. S. I also find strange Cooter's suggestion that people interested in the 3:2 shaped 45x30mm Leica S series would be more interested in that fictitious square format Hasselblad mockup/vaporware camera that an actual 4:3 format camera!
-
The wider market of digital cameras is defined by pixel number; cameras get compared by one single number ...
That goes against almost everything ever said by very serious photographers when arguing the inherent, insuperable inferiority of smaller (noise afflicted and dynamic range challenged) formats, so I will assume that you are joking.
BJL, I have a feeling you might be an engineer ...
Now if you ask *me* what I believe, I would argue that the dev work which goes into the components is significant -within the limits allowed by physics.
I in turn have heard that you are indeed an engineer (I am an applied mathematician, so not very far off) and if so, you know that although that is a plausible hypotheses, it needs to be tested and quantified before stating as fact that technological age outweighs size in this comparison — especially since Sony's 100 MP 44x33mm does not seem to be 3 to 5 years old. Let us see at least one photograph from the forthcoming 100MP Fujifilm camera before we declare it doomed!
BTW, while Phase One has apparently decided that its bulky MF SLR system with lenses and mount still based around the 645 film format can only survive at the high end of the price and size range, it is also true that both Hasselblad and Fujifilm have dedicated their new mirrorless MF systems to the 44x33mm format, and by developing new lenses and such are probably investing and risking more than Phase One is with its incremental upgrade approach. So I would say that their "votes" in favor of 44x33 are worth at least as much as Phase One's "vote" against.
-
IMO, it makes sense to wait unless you absolutely need it right now.
The cost of the 007 is dropping fast and the more you wait the more it will drop.
-
BTW, while Phase One has apparently decided that its bulky MF SLR system with lenses and mount still based around the 645 film format can only survive at the high end of the price and size range, it is also true that both Hasselblad and Fujifilm have dedicated their new mirrorless MF systems to the 44x33mm format, and by developing new lenses and such are probably investing and risking more than Phase One is with its incremental upgrade approach. So I would say that their "votes" in favor of 44x33 are worth at least as much as Phase One's "vote" against.
Actually, I think Sony's 100Mp crop sensor is about 1 year in market, which means that whatever is launched in the fall/winter will be 18 months newer. I expect the 70MP 35mm chips to drop in the fall.
Phase has had 100MP crop in an aerial system since May last year:
https://petapixel.com/2018/05/01/phase-ones-new-drone-camera-is-first-to-pack-100mp-sony-bsi-sensor/
So my feeling is that in the fall the latest 35mm gear from Sony -A7RIV- will be in-market with the FUJI GFX-100, with 70MP competing against 100 and roughly equivalent per-cell performance, but a lot of tricks on the Sony thanks to some high memory bandwidth to the sensor. Apart from the MF "look" I don't think 1/2 stop total DR advantage over the sensor or 50% more pixels in the bigger camera are really going to make a difference to many these days. Not a $5K difference anyway. A fullframe 150MP sensor and thus more than 1 stop DR or 2x the number of pixels and a real "look" difference does have meaning.
I regret to have to say this, but I think the guys at Phase know what they're doing. And their decision to abandon prosumer crop MF and concentrate on fullframe and software makes sense. Fuji don't do 35mm so crop-MF is ok for them, a big step up from APS-C for happy customers who appreciate Fuji ergonomics and optics.
Edmund
-
I regret to have to say this, but I think the guys at Phase know what they're doing. And their decision to abandon prosumer crop MF and concentrate on fullframe and software makes sense.
I'll wrap this up by agreeing with you on that part: Phase One's current MF SLR system is based on a very deep lens mount as needed for the deep mirror box of a 645 format film camera, and with a lens system (mostly) built around the 70mm image diagonal of that format, and the combination of that with a back that crops to 44x33mm (55mm diagonal) would be at a severe competitive disadvantage against the Hasselblad and Fujifilm mirrorless systems designed from the ground up for the 44x33mm format and freed of the lens design constraints of the larger image circle and far deeper and wider lens mount — and is maybe also in a weak position against the Leica 45x30mm format S-mount DSLR system, also with roughly 55mm image diagonal.
Maybe the end game for Phase One will be software, plus some special purpose industrial cameras like drones.
I somewhat agree with your expectation that larger, lower volume sensor formats might suffer from slower development rates, with the largest and lowest volume 54x40mm niche at the greatest disadvantage, 44x33mm second worst, then 36x24mm, and with APS-C probably attracting the most vigorous attention (other than phone-sized sensors). Also, any innovation at one sensor size is very easy to downsize (e.g. same photosite size, just fewer of them), whereas upsizing can involve extra complications with signal speed, worse noise and attenuation with longer signal paths, lower fab. yields, amplified by the need to stitch more "exposures" on the wafer, and so on.
-
It's worth noting re the above that Phase One hasn't abandoned the entry-level price point. They just don't tackle it with their new flagship products.
For example, we have the Phase One IQ260 (https://www.dtcommercialphoto.com/product-category/refurbished-backs-bodies-lenses/pre-owned-phase-one-digital-backs/) right now for just under $10k with full warranty, support, and guaranteed excellent condition. It's not the IQ4 150mp, but it's an unbelievably good camera for someone who, for example, shoots fashion for the web. Great skintones, more than enough res, great lenses/body/features, and the things it lacks (e.g. no super fast live view like the IQ4) aren't (typically) important for that use case.
-
Phase One's... lens system (mostly) built around the 70mm image diagonal of that format.
Every single Phase One and Schneider lens was built for a 645 full frame sensor.
Maybe you're confusing it with the Hasselblad line that has several HCD lenses that were designed for a cropped 645 sensor?
-
Every single Phase One and Schneider lens was built for a 645 full frame sensor.
Maybe you're confusing it with the Hasselblad line that has several HCD lenses that were designed for a cropped 645 sensor?
Thanks Doug; I just put the “mostly” in there in case I was overlooking some smaller format lenses.
That does confirm my point that the Phase One system is far better off going with 54x40 format exclusively.
-
P1 is going from niche to super niche.
There aren’t many examples where such a contraction has resulted in more ability to serve better customer needs in technology centric domains.
Cutting edge technology simply needs volume to justify the development costs.
Hasselblad and Fuji have played smarter IMHO.
I wouldn’t be surprised if P1 were aquired by another photography player (Fuji?) and they would mostly be interested in C1 Pro IMHO.
Cheers,
Bernard
-
P1 is going from niche to super niche.
There aren’t many examples where such a contraction has resulted in more ability to serve better customer needs in technology centric domains.
Cutting edge technology simply needs volume to justify the development costs.
Hasselblad and Fuji have played smarter IMHO.
I wouldn’t be surprised if P1 were aquired by another photography player (Fuji?) and they would mostly be interested in C1 Pro IMHO.
Cheers,
Bernard
Bernard no one is preventing P1 from downsizing the price when they want to sell more. It's just the sensor which at the moment doesn't really make sense to downsize - come on, Phase sell an XF body with no sensor for more money than Fuji sell a body with a 50MP sensor; their lenscaps probably list for as much as a backup dSLR, although equally probably Dealer Doug will send you one of those for free. All of this only works because they have a MONOPOLY on single-click 150MP.
Edmund
Note - I am not a paid shill of Doug Peterson, nor do our views on anything involving Phase One usually coincide, but I do respect the art of Doug Peterson the photographer. Any views by me represented in this topic are intended for purposes of satire and should not inform a buying decision.
-
Hi Edmund,
I guess we need to wait and see. Could be that 70 MP 24x36 sensors are overdue, but I don't think they have announced any camera, yet.
The 100 MP 33x44 mm sensor cropped to 24x36 mm would correspond to like 60 MP.
Best regards
Erik
Actually, I think Sony's 100Mp crop sensor is about 1 year in market, which means that whatever is launched in the fall/winter will be 18 months newer. I expect the 70MP 35mm chips to drop in the fall.
Phase has had 100MP crop in an aerial system since May last year:
https://petapixel.com/2018/05/01/phase-ones-new-drone-camera-is-first-to-pack-100mp-sony-bsi-sensor/
So my feeling is that in the fall the latest 35mm gear from Sony -A7RIV- will be in-market with the FUJI GFX-100, with 70MP competing against 100 and roughly equivalent per-cell performance, but a lot of tricks on the Sony thanks to some high memory bandwidth to the sensor. Apart from the MF "look" I don't think 1/2 stop total DR advantage over the sensor or 50% more pixels in the bigger camera are really going to make a difference to many these days. Not a $5K difference anyway. A fullframe 150MP sensor and thus more than 1 stop DR or 2x the number of pixels and a real "look" difference does have meaning.
I regret to have to say this, but I think the guys at Phase know what they're doing. And their decision to abandon prosumer crop MF and concentrate on fullframe and software makes sense. Fuji don't do 35mm so crop-MF is ok for them, a big step up from APS-C for happy customers who appreciate Fuji ergonomics and optics.
Edmund
-
The 52 MP 33x44 mm sensor cropped to 24x36 mm would correspond to like 60 MP.
The 52MP 33x44 sensor crops to 31MP. It is the Phase One (Sony) 150MP sensor that crops to 60MP.
-
You are right!
Sorry!
Best regards
Erik
The 52MP 33x44 sensor crops to 31MP. It is the Phase One (Sony) 150MP sensor that crops to 60MP.
-
Right now, the Sony sensors are (roughly) at technological parity from APS-C on up to 54x40mm MF - all are modern, copper-wired BSI designs, and all perform very similarly per pixel. Several of them (26 MP APS-C, 100 MP 33x44mm, 150 MP 54x40mm) actually seem to be the same pixels in different sensor sizes). There are three blips in this trend...
1.) The 50MP 33x44mm sensor is NOT of this lineage - it's a full generation or more older (non-copper, non BSI), and it barely competes with the 40+ MP modern 24x36mm sensors despite its size advantage. The 100MP 33x44mm sensor is the representative of the modern Sony lineage, but we only see it on the market in one industrial camera (the GFX 100 will be the first "standard" camera to use it).
2.) The 40+MP 24x36mm sensors are of roughly the same technological generation, but they are not the same pixels - the pixels are a bit bigger (just over 4 micron, while the others are all 3.76). Sony datasheets have shown a 61MP 24x36mm sensor, which will use the "standard pixel", but it hasn't cropped up (sorry) in any camera yet.
3.) If the 20MP Micro 4/3 sensor is a Sony sensor, it is a much older generation, technologically closer to the 50MP 33x44 sensor, and it uses a smaller, nonstandard pixel size. A "standard pixel" Micro 4/3 sensor would be around 15MP.
Once we see those two sensors (100MP 33x44 is probably a significant improvement, 61MP 24x36 is probably incremental because the 40+MP generation is already modern), there will be a wide range of cameras with very similar per-pixel behavior - it's just a matter of how many pixels you want to carry and pay for.
Essentially all non Sony-derived (counting both XTrans and Nikon/Sony designs as Sony-derived) sensors underperform the Sonys. The initial Photons to Photos results for the Leica Q2 sensor (probably TowerJazz, and probably also the Panasonic S1R sensor) show that it behaves much more like a high-resolution Canon sensor than a modern Sony. The default assumption about the Canon 70ish MP sensor is that it'll probably perform like a Canon, but maybe they've made a breakthrough.
Beating the Sonys may well take a new technology, whether it is organic sensors, a Foveon with the bugs out, or something else.
-
The default assumption about the Canon 70ish MP sensor is that it'll probably perform like a Canon, but maybe they've made a breakthrough.
I have seen the70MP Canon sensor mentioned once on a rumor site. It is not a rumor; it is a WAG.
-
Once we see those two sensors (100MP 33x44 is probably a significant improvement, 61MP 24x36 is probably incremental because the 40+MP generation is already modern), there will be a wide range of cameras with very similar per-pixel behavior - it's just a matter of how many pixels you want to carry and pay for.
It's so weird that in the modern digital age so many people conflate "Sensors" and "Cameras"
Even if it were true that you get the same per-pixel performance out of a small-format camera as you get out of an IQ4 150mp (it is not remotely true, but put that aside for a minute...) the difference between two cameras is far, far more than the difference in sensors.
If an iPhone had identical "pixels" to an IQ4 150mp the two would still have their unique sets of pros and cons, and have almost nothing in common as cameras.
-
It's so weird that in the modern digital age so many people conflate "Sensors" and "Cameras"
Even if it were true that you get the same per-pixel performance out of a small-format camera as you get out of an IQ4 150mp (it is not remotely true, but put that aside for a minute...) the difference between two cameras is far, far more than the difference in sensors.
If an iPhone had identical "pixels" to an IQ4 150mp the two would still have their unique sets of pros and cons, and have almost nothing in common as cameras.
Doug,
Thank you for pointing out that not all manufacturers are equal when it comes to Sony sensors.
I think we all agree that Sony can do more with its cameras than anybody else using their sensors - as one can see from their low-latency no-blackout EVF, excellent phase detect, industry-leading eye-detect follow-focus AF, 4K movie and high-speed movie modes. Phase One does not of course consider these technologies significant for professional photographers, but there are more amateurs than pros out there.
Edmund
-
fuji interview at DPR (https://www.dpreview.com/interviews/8410636142/cp-2019-fujifilm-interview-we-want-to-show-photographers-the-future?utm_source=self-desktop&utm_medium=marquee&utm_campaign=traffic_source)
interesting to hear about %...I am probably reading between the lines here but my interpretation is that the GFX100 is (very much like the 50S) is just a "pushing the limits" camera body to help sell lenses and keep customers who want more then a crop sensor....and they are killing it...doing everything right from every angle....price, check....quality, check....what are professionals asking for? C1, check....
still doubt that there is a GFX 100 in my future, but that is not because of the Fuji, it more about what the next sony will probably be...and it will most likely cover my needs better and more completely....which seems to be fine with Fuji....
-
interesting to hear about %...I am probably reading between the lines here but my interpretation is that the GFX100 is (very much like the 50S) is just a "pushing the limits" camera body to help sell lenses and keep customers who want more then a crop sensor....and they are killing it...doing everything right from every angle....price, check....quality, check....what are professionals asking for? C1, check....
still doubt that there is a GFX 100 in my future, but that is not because of the Fuji, it more about what the next sony will probably be...and it will most likely cover my needs better and more completely....which seems to be fine with Fuji....
$10,000 for a GFX100 body is still a lot of money in absolute terms.
-
$10,000 for a GFX100 body is still a lot of money in absolute terms.
100mpix with 14 and 16bit capture, up to 6fps, IBIS, AF to be years ahead of what we have seen from medium format before, incredible (and affordable) lenses and an updated version of the ergonomically best finder i have ever worked with....
the only reason 10K seems high is that the next sony will probably be VERY VERY close in terms of IQ with better AF, faster shooting, smaller package and 1/3 of the price....
this (global shutter tech) is what i am really excited about (https://www.sonyalpharumors.com/press-text-sony-develops-a-stacked-cmos-global-shutter-image-sensor/)
-
this (global shutter tech) is what i am really excited about (https://www.sonyalpharumors.com/press-text-sony-develops-a-stacked-cmos-global-shutter-image-sensor/)
Unfortunately that Pregious S global shutter still halves dynamic range and well capacity by using half the sensor area for light-shielded memory during read-out. That is the reason Sony is only talking about industrial uses (where global shutters have been available for some time), not “regular” cameras.
Panasonic has a better global shutter idea, if it can ever be bought to market: an electronically switchable shutter in front of the photosites.
-
the only reason 10K seems high is that the next sony will probably be VERY VERY close in terms of IQ with better AF, faster shooting, smaller package and 1/3 of the price....
No, it seems high because $10,000 is a lot of money in absolute terms.
-
this (global shutter tech) is what i am really excited about[/url]
From my best understanding, we are at least five years away from cameras that have excellent photographic image quality and global shutters.
This sensor is for industrial purposes like machine vision inspection. It would make a pretty awful photographic camera sensor.
-
As the OP who's still following the discussion with interest, I have to ask: is there a consensus among the connoisseurs here that Sony sensors are simply the best and everyone else is playing catch-up?
If you look at the technical specifications and the scientific tests that have been done and you want the part of the entire imaging pipeline where the sensor is concerned, to have the most modern tech and the best electronics and the highest signal-to-noise ratio and also keep your creative options as open as possible with a nice thick malleable file, there's actually only one sensor factory to watch, and that would be Sony?
Is there any other maker that produces high quality sensors that are of interest in the field of photography? Maybe for creative reasons?
Foveon sensors come to mind.
-
As the OP who's still following the discussion with interest, I have to ask: is there a consensus among the connoisseurs here that Sony sensors are simply the best and everyone else is playing catch-up?
Yes, the only company that seems close is Nikon but it is unclear what part of their sensors is based on their own IP vs Sony's.
Cheers,
Bernard
-
From my best understanding, we are at least five years away from cameras that have excellent photographic image quality and global shutters.
This sensor is for industrial purposes like machine vision inspection. It would make a pretty awful photographic camera sensor.
5 years is closer then i thought!
who knows what else they will come up until then but global shutter will change everything IMO....
-
As the OP who's still following the discussion with interest, I have to ask: is there a consensus among the connoisseurs here that Sony sensors are simply the best and everyone else is playing catch-up?
If you look at the technical specifications and the scientific tests that have been done and you want the part of the entire imaging pipeline where the sensor is concerned, to have the most modern tech and the best electronics and the highest signal-to-noise ratio and also keep your creative options as open as possible with a nice thick malleable file, there's actually only one sensor factory to watch, and that would be Sony?
Is there any other maker that produces high quality sensors that are of interest in the field of photography? Maybe for creative reasons?
Foveon sensors come to mind.
i saw the Q2 sensor DR test today and was disappointed....i did have high hopes for the SR1 (which i am assuming shares the sensor with the Q2) , the S1 looks like a killer video camera, not sure about the AF system right now but from what i understand it has potential to get much better with firmware....the mount opens a lot of doors....looked like a promising system to challenge sony (and nikon) but the test showed it more then a stop behind in DR tests....what makes it worse that neither SR1 or Q2 are really shipping yet and we know sony will drop a flagship (within a year?)....
I just dont see sony even having any competition at this point....the A7RII sensor released in 2015 still beats this new Q2 sensor....canon is alive because a lot of people still have a lot of canon glass and lots of people buying their first camera simply are buying a canon.....although that is changing dramatically....
fuji, hasselblad, phase one are all using the same sony sensor....not sure who makes the leica S sensor but again: it just does not compare.....
i wish there was competition, sony just saw nikon, canon and panasonic entering the FF mirrorless market and in all honesty the only real challenger to the A7R is the nikon....which uses a sony sensor....so much for competition....
-
The Z7 sensors appears to be manufactured by Sony and to have some Sony IP in it, but Thom is pretty clear that it also has some Nikon IP.
It is clearly not an off the shelf Sony part.
Cheers,
Bernard
-
The Z7 sensors appears to be manufactured by Sony and to have some Sony IP in it, but Thom is pretty clear that it also has some Nikon IP.
It is clearly not an off the shelf Sony part.
Cheers,
Bernard
Same with Fuji
-
Same with Fuji
AKFAIK, Fujifilm uses special CFA's on some of its sensors (X-trans"), but the underlying silicon is a standard Sony product. Nikon instead seems to contribute some technology in the silicon itself, and ges sensor no in Sony's product catalog.
Aside: there is an interesting twist recently; Fujifilm has Sony Semiconductor's new 26MP 24x16mm sensor in two cameras (X-T3 and X-T30) before anyone else, including Sony Imaging.