Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: PaulSchneider on June 20, 2010, 05:39:51 pm
-
Hi guys,
does anyone have an idea as to what lies ahead in medium format land?
Does Phase One have some surprises for Photokina?
Different points to consider:
a) Dalsa apparently is able to create medium format sized ccds with 92 MPx resolution. They already manufacture them for aerial photogrammetry and cartographic applications ...
http://www.azooptics.com/details.asp?newsID=7421 (http://www.azooptics.com/details.asp?newsID=7421)
Optical limitations. I just received a Schneider 43 XL and must say the performance is incredible. It is another league compared to the base offerings of hasselblad (currently am on a H3D-50 loaner till H4D-60 is finally shipped). Apparently this is the bleeding cutting edge of lens technology and one can see that in the results. I have never seen such sharpness out of a digital back, even applying a little bit of sharpening can immediately create halos and jagged edges, that's how sharp this lens is. But nonetheless even these lenses are are apparently conceived to max out at 5-6 microns. I'm confidend such Schneider lenses will resolve a higher resolution back but what about let's say some stock 80mm 2.8 lens that's been around for years with the same optical design? I'm afraid the optics aren't up to the task and creating a bigger sensor is out of question short of creating a new kind of bigger system and replacing the H-System alltogether.
c) R&D considerations, market considerations: People who just recently coughed up the money for a 60MP system would be pissed if there would be a 90 MP system around the corner in this short time-frame and which would cost maybe another 20 000 USD to "upgrade". I'd be seriously enraged if Hasselblad came up with a higher-res system just one or two months later after they finally have been able after more than a year! to produce a useable H4D-60.
d) Phase One hasn't had a new product for some time and they are perceived as the key innovators in the MFDB area ... they have to have something new to show around Photokina ...
So ... what will happen, anyone care take a look into the crystal ball?
Regards
Paul
-
Hi,
For long time five microns have sort of been seen as a practical limit. A five my sensor on 55 by 45 mm geometry would yield 99 MPixels, so it's certainly possible.
Shrinking the size of the pixels puts increasing demands on the rest of the system, but less dramatically than one would think. Going from from 60 MP to 90 MP is a 22% increase in linear resolution, so optical alignment needs to be 22% better, lenses 22% better and so on. BTW, photographer needs to be 22% better, too.
Regarding price I'd suggest that the cost of a 90 MP chip is the same as for a 60 MP chip. Camera electronics would need to be 50% faster to keep performance at a given level.
If I recall it correctly, the P45 was 39MP while the P65 is 60MP, according to Michael Reichmann it was a significant improvement, but that improvement probably went beyond just increasing pixel count. If Phase wants to get P65+ owners to upgrade they need a a significant step forward in image quality.
On the other hand, Phase just got some new competition, the Leica S2 and the new Pentax 645. So Phase perhaps would be more focused on improving their camera platform regarding precision, autofocus and so on.
Best regards
Erik
Hi guys,
does anyone have an idea as to what lies ahead in medium format land?
Does Phase One have some surprises for Photokina?
Different points to consider:
a) Dalsa apparently is able to create medium format sized ccds with 92 MPx resolution. They already manufacture them for aerial photogrammetry and cartographic applications ...
http://www.azooptics.com/details.asp?newsID=7421 (http://www.azooptics.com/details.asp?newsID=7421)
Optical limitations. I just received a Schneider 43 XL and must say the performance is incredible. It is another league compared to the base offerings of hasselblad (currently am on a H3D-50 loaner till H4D-60 is finally shipped). Apparently this is the bleeding cutting edge of lens technology and one can see that in the results. I have never seen such sharpness out of a digital back, even applying a little bit of sharpening can immediately create halos and jagged edges, that's how sharp this lens is. But nonetheless even these lenses are are apparently conceived to max out at 5-6 microns. I'm confidend such Schneider lenses will resolve a higher resolution back but what about let's say some stock 80mm 2.8 lens that's been around for years with the same optical design? I'm afraid the optics aren't up to the task and creating a bigger sensor is out of question short of creating a new kind of bigger system and replacing the H-System alltogether.
c) R&D considerations, market considerations: People who just recently coughed up the money for a 60MP system would be pissed if there would be a 90 MP system around the corner in this short time-frame and which would cost maybe another 20 000 USD to "upgrade". I'd be seriously enraged if Hasselblad came up with a higher-res system just one or two months later after they finally have been able after more than a year! to produce a useable H4D-60.
d) Phase One hasn't had a new product for some time and they are perceived as the key innovators in the MFDB area ... they have to have something new to show around Photokina ...
So ... what will happen, anyone care take a look into the crystal ball?
Regards
Paul
-
Well I said it before and I just hope I'm wrong.
What we will see soon:
- a 80+ Mp back, which still has the great old LCD, or a sligly better one, but still far behind current phones which cost only thousand bucks or so.
- same DR, same noise and nothing new
What I would like to see and the only thing I'm really interested:
-60Mp or similar (Can be 80, but ONLY if other things are significantly improved.
- higher DR I would love to see for example something like sensor + but that not only offers higher ISOs, but a higher DR when bined
- Live View .... I don't believe we will ever get it until I see it
- Noise improvements in the ISO 200 and 800 range
- a LCD on the back, which is GOOD, large and can be viewed anywhere.
-
Optical limitations. I just received a Schneider 43 XL and must say the performance is incredible. It is another league compared to the base offerings of hasselblad (currently am on a H3D-50 loaner till H4D-60 is finally shipped). Apparently this is the bleeding cutting edge of lens technology and one can see that in the results. I have never seen such sharpness out of a digital back, even applying a little bit of sharpening can immediately create halos and jagged edges, that's how sharp this lens is. But nonetheless even these lenses are are apparently conceived to max out at 5-6 microns. I'm confidend such Schneider lenses will resolve a higher resolution back but what about let's say some stock 80mm 2.8 lens that's been around for years with the same optical design? I'm afraid the optics aren't up to the task and creating a bigger sensor is out of question short of creating a new kind of bigger system and replacing the H-System alltogether.
c) R&D considerations, market considerations: People who just recently coughed up the money for a 60MP system would be pissed if there would be a 90 MP system around the corner in this short time-frame and which would cost maybe another 20 000 USD to "upgrade". I'd be seriously enraged if Hasselblad came up with a higher-res system just one or two months later after they finally have been able after more than a year! to produce a useable H4D-60.
The three things they would have to do to get me seriously interested would be:
1. add a usable live view in back
2. devide the price of their high end +back by 2+,
3. garantee me 500+ images battery life at -15C
The cash is buried in the garden near the swimming pool.
Cheers,
Bernard
-
Hi guys,
does anyone have an idea as to what lies ahead in medium format land?
Does Phase One have some surprises for Photokina?
Different points to consider:
a) Dalsa apparently is able to create medium format sized ccds with 92 MPx resolution. They already manufacture them for aerial photogrammetry and cartographic applications ...
http://www.azooptics.com/details.asp?newsID=7421 (http://www.azooptics.com/details.asp?newsID=7421)
Diging into the sources on that aerial photography gear, it uses the same 6 micron pixel size as the current Dalsa and Kodak MF sensors, but over a larger frame:
http://www.microsoft.com/ultracam/en-us/UltraCamXp.aspx (http://www.microsoft.com/ultracam/en-us/UltraCamXp.aspx)
I cannot tell if these are scanning backs with linear sensors, but that seems likely.
Anyway, the Dalsa and Kodak 6 micron options have not been around for very long, so I doubt that smaller pixel sizes are due in DMF sensors for another year or two.
Here is some speculation: a Kodak sensor sometime this year in more or less full 645 format of 56x41.5mm, so 60 to 65 MP, to catch up with Dalsa.
-
Here is some speculation: a Kodak sensor sometime this year in more or less full 645 format of 56x41.5mm, so 60 to 65 MP, to catch up with Dalsa.
Hoping it'll have microlenses as H4d-40
-
Well I said it before and I just hope I'm wrong.
What we will see soon:
- a 80+ Mp back, which still has the great old LCD, or a sligly better one, but still far behind current phones which cost only thousand bucks or so.
- same DR, same noise and nothing new
What I would like to see and the only thing I'm really interested:
-60Mp or similar (Can be 80, but ONLY if other things are significantly improved.
- higher DR I would love to see for example something like sensor + but that not only offers higher ISOs, but a higher DR when bined
- Live View .... I don't believe we will ever get it until I see it
- Noise improvements in the ISO 200 and 800 range
- a LCD on the back, which is GOOD, large and can be viewed anywhere.
Amen!
Even if live view is not for tomorrow, if they just could bring a really really good lcd, and to be honest at the prices they sell them they should be obliged by the consumers,
so, even if they just would bring a great lcd, that would be a significant accomplishment.
-
Amen!
Even if live view is not for tomorrow, if they just could bring a really really good lcd, and to be honest at the prices they sell them they should be obliged by the consumers,
so, even if they just would bring a great lcd, that would be a significant accomplishment.
Or, somehow interface directly with iPhone 4 or iPad
-
Am I the only one who would be ok with a 30-50mp back, as long as it was 6x7? Because seriously, that's all I want or need.
-
Am I the only one who would be ok with a 30-50mp back, as long as it was 6x7? Because seriously, that's all I want or need.
+1 (for 6x6)
-
+1 (for 6x6)
Agree
-
Cropping to square does not bother me that much, it would be nice but I can understand on that the manufacturers because that would be a too small market for them.
But where I can't follow them is on the ridiculous LCD.
All these external devices p....s me off. They should offer a very good integrated solution first, and then get married with I.Pad or whoever if they want to.
-
I just heard from somebody who spoke to an engineer at Hasselblad Denmark that the bottleneck actually lies in the right image processing chips those manufacturers can buy from third party suppliers. Apparently there's only one NVIDIA chip on the market that could support substantially more processing than 60 MP in a small package and the problem is that that right now this circuit eats power in the ballpark of 60 watts ... so apparently this is a serious limitation that has to be overcome too ...
-
I just heard from somebody who spoke to an engineer at Hasselblad Denmark that the bottleneck actually lies in the right image processing chips those manufacturers can buy from third party suppliers. Apparently there's only one NVIDIA chip on the market that could support substantially more processing than 60 MP in a small package and the problem is that that right now this circuit eats power in the ballpark of 60 watts ... so apparently this is a serious limitation that has to be overcome too ...
The Leica S2's Maestro image processor from Fujitsu has a hardware JPG engine on chip that is capable of rendering 160MP/s. The chip is extremely low power, with both a RISC and DSP engine in a dual core configuration. This allows the S2 to render on-the-fly JPGs for image display at 360x480 (460K) pretty much instantaneously. Zooming to 100% view is also close to real-time with no hourglass or status bar. So, it's possible.
Just having a good LCD and long battery life can mean a lot to a pro. We provided an S2 for a photographer shooting hi-end magazine fashion editorial a few weeks ago. On the first day, he shot 160GB untethered (about 1900 shots with DNG+ hi-res JPG) in a five hour period and still had over 30% battery remaining. Instead of needing to go to the computer to double check lighting and focus, he worked exclusively off of the LCD which provided an accurate rendering of what the picture actually looked like. Having worked with other backs pretty regularly over the years, he couldn't get over how convenient this was and continued to be in disbelief over the battery performance over the course of his four day shoot.
I'll certainly be interested to see what everyone comes up with at Photokina and will be there firsthand to see for myself.
David
-
Hi,
I'd guess there is a concern about processing power, but you can add more chips. The Sony Alpha 900 uses two Bionz processor and AFAIK the Nikon D3 has 6 signal processing chips working in parallel. I'd suggest that you cannot use PC-chipsets for digital cameras. The BIONZ in my Alpha 900 processes 5 images per second while the Xeon in my MacPro handles perhaps one per second, using eight threads on four cores. So I guess that there is a need for ASICs for digital cameras.
Regarding dissipating 60 Watts, why don't you try try to hold a 60 W lightbulb in your hand, lightened of course, and you realize that there is a problem.
Best regards
Erik
I just heard from somebody who spoke to an engineer at Hasselblad Denmark that the bottleneck actually lies in the right image processing chips those manufacturers can buy from third party suppliers. Apparently there's only one NVIDIA chip on the market that could support substantially more processing than 60 MP in a small package and the problem is that that right now this circuit eats power in the ballpark of 60 watts ... so apparently this is a serious limitation that has to be overcome too ...
-
Thanks David,
Good info. Did that chap put an order on the S2?
Best regards
Erik
The Leica S2's Maestro image processor from Fujitsu has a hardware JPG engine on chip that is capable of rendering 160MP/s. The chip is extremely low power, with both a RISC and DSP engine in a dual core configuration. This allows the S2 to render on-the-fly JPGs for image display at 360x480 (460K) pretty much instantaneously. Zooming to 100% view is also close to real-time with no hourglass or status bar. So, it's possible.
Just having a good LCD and long battery life can mean a lot to a pro. We provided an S2 for a photographer shooting hi-end magazine fashion editorial a few weeks ago. On the first day, he shot 160GB untethered (about 1900 shots with DNG+ hi-res JPG) in a five hour period and still had over 30% battery remaining. Instead of needing to go to the computer to double check lighting and focus, he worked exclusively off of the LCD which provided an accurate rendering of what the picture actually looked like. Having worked with other backs pretty regularly over the years, he couldn't get over how convenient this was and continued to be in disbelief over the battery performance over the course of his four day shoot.
I'll certainly be interested to see what everyone comes up with at Photokina and will be there firsthand to see for myself.
David
-
+1 (for 6x6)
Yes please
-
i'd be interested in a phase p40+ afi with a revolving back. everything is there, all they have to do is put it together and make it available.
-
I'd be interested (back and once more) if Hasselblad introduces a CFV39 back with a revolving sensor.
Other than that, booooring!. Of course I expect another kick-butt dslr from both Canon and Nikon
Eduardo
i'd be interested in a phase p40+ afi with a revolving back. everything is there, all they have to do is put it together and make it available.
-
Am I the only one who would be ok with a 30-50mp back, as long as it was 6x7? Because seriously, that's all I want or need.
+1. and live view / good LCD/LED display along with it
-
The cost is directly related to the size of the chip, not the amount of megapixels. So, 6x7, even if it was only 10MP, the price would be through the roof (as if it's not already).
-
The cost is directly related to the size of the chip, not the amount of megapixels. So, 6x7, even if it was only 10MP, the price would be through the roof (as if it's not already).
I would argue that the biggest contributor to the cost of a sensor is how small the batch is.
Cheers,
Bernard
-
I'd be interested (back and once more) if Hasselblad introduces a CFV39 back with a revolving sensor.
Other than that, booooring!. Of course I expect another kick-butt dslr from both Canon and Nikon
Eduardo
yes booooring, i like boring. like boringly easy to use a wlf and shoot verticles, then switch to horizontal without raising, turning or even moving the camera.
would be boringly easy to do to: take the existing afi body and put a phase one back on it...
it's like some sick joke thinking of the thousands of fashion photographers out there either bent over or on there knees for hours, shooting verticles in 2:3 format with gaffer tape over the lcd so they know if they've cropped the head or feet off the model. who the hell thinks this stuff up?
what format the ipad? 4:3 and verticle, just like a magazine of course.
-
Hi guys,
does anyone have an idea as to what lies ahead in medium format land?
Does Phase One have some surprises for Photokina?
So ... what will happen, anyone care take a look into the crystal ball?
Regards
Paul
Now with Leaf in their business portfolio.. it's up to them to make a move up the hill or stay where they are and work on
alternatives.. acquisitions, marketing, software..
I think Phase is ambitious enough to lead the MF market.. so there maybe a new design with new features coming soon..
Obviously Hasselblad's attitude is they are the ones to follow..
we'll see..
-
Hi!
I don't think so. If you make the sensor in one single piece of silicon I guess that basic theory is that cost increases about (linear dimension)^4. Now, all large sensors (larger than APS-H) are stitched except possibly Canon's as Canon seem to have a larger format stepper.
So doubling size would increase cost by a factor of 16. If you double linear size the area will be four times larger, so you get a fourth of the number sensors from a certain size of wafer.
The number of error per chip will also increase by four, causing yield to go down.
Best regards
Erik
I would argue that the biggest contributor to the cost of a sensor is how small the batch is.
Cheers,
Bernard
-
Hi!
I don't think so. If you make the sensor in one single piece of silicon I guess that basic theory is that cost increases about (linear dimension)^4. Now, all large sensors (larger than APS-H) are stitched except possibly Canon's as Canon seem to have a larger format stepper.
So doubling size would increase cost by a factor of 16. If you double linear size the area will be four times larger, so you get a fourth of the number sensors from a certain size of wafer.
The number of error per chip will also increase by four, causing yield to go down.
Erik,
Yes, I am aware of the basic theory. Now that is based on an everything else being equal approach that makes total sense within a typical small scale business model assuming the usage of existing methods.
I believe that very disuptive things can happen once you enter the realm of very large series, with new tools, methods and technologies being developped that can produce quantum leaps within a few years.
Now, the funny thing is that we probably don't even need any quantum jump to produce at reasonnable prices 6x7 sensors.
The price of a typical raw 200mm sillicon wafer is less than 50US$ if I am not mistaken. Even if you can only make one sensor per wafer and need to use 10 to find a perfect sensor you are still only at 500 US$ raw material cost... A lot more things are required to deliver a usable sensor, but we would still only be at around 5.000 US$ end user price, which is still only 15% of the price of a P65+ back...
Cheers,
Bernard
-
Bernard,
I don't argue with your numbers, I got the impression that you have some insights in the industry and I don't.. Also as far as I know essentially all large sensors are stitched from smaller pieces. This is actually visible for instance on some Nikon D3X sensors and also on the Sony Alpha 900 sensors. So DALSA can probably make larger sensors by "stitching" smaller ones, and that circumvents the size^4 rule.
Just a few things:
1) The rest of the electronics is essentially the same independent of sensor size, but you need faster processing with more pixels
2) CCD production is not leading edge technology. Technology is lead by very small feature size, like 40 nano meters. The impression I have is that sensor production is done in old fabs.
3) MFDB CCD production is rather small scale I think, at least compared with DSLR sensors.
The final observation is that prices have much to do with what customers are willing to pay. Also a company probably prefers to have it's top product at a very high price so it doesn't compete with it's 'lesser' products.
Best regards
Erik
Erik,
Yes, I am aware of the basic theory. Now that is based on an everything else being equal approach that makes total sense within a typical small scale business model assuming the usage of existing methods.
I believe that very disuptive things can happen once you enter the realm of very large series, with new tools, methods and technologies being developped that can produce quantum leaps within a few years.
Now, the funny thing is that we probably don't even need any quantum jump to produce at reasonnable prices 6x7 sensors.
The price of a typical raw 200mm sillicon wafer is less than 50US$ if I am not mistaken. Even if you can only make one sensor per wafer and need to use 10 to find a perfect sensor you are still only at 500 US$ raw material cost... A lot more things are required to deliver a usable sensor, but we would still only be at around 5.000 US$ end user price, which is still only 15% of the price of a P65+ back...
Cheers,
Bernard
-
What do I expect? Not much, really.
Another year is going by and away and still MF (if we can call'em as such) sensors are closer to 35 dimensions than to 6X8, the ultimate MF film size. MF back prices are still only for the rich (dentists and such. No offense, my dad was a dentist) and famous (high-roller photographers).
It beats me. I don't get it. As far as I'm concerned just about every high-tech manufacturing line is being developing accordingly to Moore's law. Digital sensors seem to breaking this law for 7 years now. Suddenly they came to a complete screeching stop. I say sensors in general because even in lesser sizes is the same thing, they only have increase in pixel density.
What kind of limitation was reached? Scientific? Technological? Marketing? Political perhaps? Was it a powerful groupuscle of economic interests seizing this technology? National security?
Well, I sound like conspiracy freak now. But you get my idea. I'm starting to become startled by this phenomenom.
Best
Eduardo
-
Now, all large sensors (larger than APS-H) are stitched except possibly Canon's as Canon seem to have a larger format stepper.
This internet myth refuses to die.
1. Canon has said repeatedly that it stitches its 36x24mm sensors, and that the 1D sensor size of about 19x28mm is about the largest that can be made without stitching, and this fits with the known maximum field size of 26x33mm for all suitable steppers, and all steppers introduced for some years now.
2. Canon makes one larger format stepper about 50x50mm field size, but it is a quite old model of very low resolution, 0.5 micron minimum feature size. Rather clearly, that minimum feature size is too large for good modern CMOS pixels of about 6 microns: CMOS sensors are typically made on steppers with minimum feature size 1/15th to 1/30th of pixel pitch.
3. That 50x50mm format stepper is likely used by Kodak to make its largest sensors (for X-rays and such) which are of exactly that size ... but those have huge 24 micron pixel pitch. And since Canon sells that stepper to al comers, and for example Kodak seems to have access to one, there is no Canon exclusive on it: any major sensor maker that wants one can probably get one.
4. Nikon used to sell a stepper with field size large enough for 36x24mm sensors, but discontinued it a few years ago. I am rather sure that Nikon or Sony could get one of those if needed (sensors are typically made on older equipment), so again the claim of a Canon exclusive fails.
P.S. The joins visible on some Nikon sensors are in the toppings (AA filter), not the sensor chip itself.
-
As far as I'm concerned just about every high-tech manufacturing line is being developing accordingly to Moore's law. Digital sensors seem to breaking this law for 7 years now.
On the contrary, digital sensors follow Moore's Law nicely: the law is mostly driven by the benefits for digital electronic devices (which sensors are not really; they are analog or mixed analog-digital) of reducing feature size and chip size, and so is manifested in the increase in pixel density that you mention and in the ever-shrinking average sensor size in the mass market of compacts and phone-cameras. But Moore's Law scarcely touches the upsizing of low-volume specialty sensors for MF and such ... especially when the economies of scale will never be there, with the MF industry having more or less abandoned further development of formats larger than 645 even before the digital transition.
-
The three things they would have to do to get me seriously interested would be:
1. add a usable live view in back
2. devide the price of their high end +back by 2+,
3. garantee me 500+ images battery life at -15C
The cash is buried in the garden near the swimming pool.
Cheers,
Bernard
My 3 wishlist items would be:
1. Make "long-exposure processing" (taking & subtracting a dark frame internally) an option that can be switched off, to allow for long, consecutive, uninterrupted exposure sequences [Phase One take note!] - let the user take care of dark subtraction later in the software of their choice
2. Make the IR-cut filter easily user-removable, like the Mamiya ZD and old Kodak DCS backs [again, Phase One take note!]
3. Add an option for active peltier cooling of the sensor to at least 20 degrees C below ambient temperature.
Can anyone guess why I want these features?
-
Hi BJL.
According to your response the answer is "Marketing"
(Maybe they follow Morre's law except in size grow)
That and probably the slow-down of world economy of the last years.
Ed
On the contrary, digital sensors follow Moore's Law nicely: the law is mostly driven by the benefits for digital electronic devices (which sensors are not really; they are analog or mixed analog-digital) of reducing feature size and chip size, and so is manifested in the increase in pixel density that you mention and in the ever-shrinking average sensor size in the mass market of compacts and phone-cameras. But Moore's Law scarcely touches the upsizing of low-volume specialty sensors for MF and such ... especially when the economies of scale will never be there, with the MF industry having more or less abandoned further development of formats larger than 645 even before the digital transition.
-
Thanks for all info!
Best regards
Erik
This internet myth refuses to die.
1. Canon has said repeatedly that it stitches its 36x24mm sensors, and that the 1D sensor size of about 19x28mm is about the largest that can be made without stitching, and this fits with the known maximum field size of 26x33mm for all suitable steppers, and all steppers introduced for some years now.
2. Canon makes one larger format stepper about 50x50mm field size, but it is a quite old model of very low resolution, 0.5 micron minimum feature size. Rather clearly, that minimum feature size is too large for good modern CMOS pixels of about 6 microns: CMOS sensors are typically made on steppers with minimum feature size 1/15th to 1/30th of pixel pitch.
3. That 50x50mm format stepper is likely used by Kodak to make its largest sensors (for X-rays and such) which are of exactly that size ... but those have huge 24 micron pixel pitch. And since Canon sells that stepper to al comers, and for example Kodak seems to have access to one, there is no Canon exclusive on it: any major sensor maker that wants one can probably get one.
4. Nikon used to sell a stepper with field size large enough for 36x24mm sensors, but discontinued it a few years ago. I am rather sure that Nikon or Sony could get one of those if needed (sensors are typically made on older equipment), so again the claim of a Canon exclusive fails.
P.S. The joins visible on some Nikon sensors are in the toppings (AA filter), not the sensor chip itself.
-
I'd guess there is a concern about processing power, but you can add more chips. The Sony Alpha 900 uses two Bionz processor and AFAIK the Nikon D3 has 6 signal processing chips working in parallel.
The D3/s has 12 analog readout channels situated on 6 (expensive) low-noise dual-channel chips. The DSP is done by the EXPEED, which as far as I know is still single core.
-
..... 3. Add an option for active peltier cooling of the sensor to at least 20 degrees C below ambient temperature. ......
I'm afraid you'd get troubles with moisture precipitation here and the additional power consumption ...
-
Hi,
I was referring to this article:
http://www.chipworks.com/blogs.aspx?id=4626&blogid=86 (http://www.chipworks.com/blogs.aspx?id=4626&blogid=86)
It says: "On the Nikon D3 we found a whopping 6 total Analog Devices AD9974 signal processors – presumably two for each color."
http://www.chipworks.com/uploadedImages/Bl...g/DSLR-Blog.jpg (http://www.chipworks.com/uploadedImages/Blog/Test_Blog/DSLR-Blog.jpg)
Best regards
Erik
The D3/s has 12 analog readout channels situated on 6 (expensive) low-noise dual-channel chips. The DSP is done by the EXPEED, which as far as I know is still single core.
-
... very interesting discussion here ...
so what is consensus right now:
Will there realistically be a sensor largern than 645 or are the costs of production in the year 2010 still too prohibitive?
....
-
Same blog sees no stitching artefacts on the Canon sensor. I have a feeling Canon may be throwing some proprietary technology at their dSLR sensors, both at the lithography and and the process levels. Owning the fab and making some of the fab equipment is two key competitive advantages they have and I would be surprised that they have not found some way to take advantage of them.
Edmund
Hi,
I was referring to this article:
http://www.chipworks.com/blogs.aspx?id=4626&blogid=86 (http://www.chipworks.com/blogs.aspx?id=4626&blogid=86)
It says: "On the Nikon D3 we found a whopping 6 total Analog Devices AD9974 signal processors – presumably two for each color."
http://www.chipworks.com/uploadedImages/Bl...g/DSLR-Blog.jpg (http://www.chipworks.com/uploadedImages/Blog/Test_Blog/DSLR-Blog.jpg)
Best regards
Erik
-
... very interesting discussion here ...
so what is consensus right now:
Will there realistically be a sensor largern than 645 or are the costs of production in the year 2010 still too prohibitive?
We'd have to have feedback from pros about the full story, but the price of the silicon wafer that has been repeatedely pushed forward as the key issue doesn't seem to be that relevant relative to the final price of such cameras.
Now I suspect that we have 3 issues at hand:
1. The MF manufacturers will not do it because their business model is in fact similar to that of 35mm manufaturers, meaning that it is based on the assumption that enough customers will buy enough lenses. These lenses being mostly 645 lenses (Hassy and Mamiya), they have zero interest in going bigger,
2. Color cast issues might be difficult to overcome with existing lenses able to cover such a wide sensor. Most customers are simply not willing to deal with this on every single wide image. The R&D cost needed to overcome this might not be justified considering the size of the target market.
3. In all objectivity, as a famous landscape Japanese photographer was telling me last week when looking at my 300 megapixels panos, very few people need 40 megapixels and even fewer need more. The main value of a 6x7 sensor would be the ability to reach 100 megapixels with large photosites... but who needs that?
Cheers,
Bernard
-
We'd have to have feedback from pros about the full story, but the price of the silicon wafer that has been repeatedely pushed forward as the key issue doesn't seem to be that relevant relative to the final price of such cameras.
Now I suspect that we have 3 issues at hand:
1. The MF manufacturers will not do it because their business model is in fact similar to that of 35mm manufaturers, meaning that it is based on the assumption that enough customers will buy enough lenses. These lenses being mostly 645 lenses (Hassy and Mamiya), they have zero interest in going bigger,
2. Color cast issues might be difficult to overcome with existing lenses able to cover such a wide sensor. Most customers are simply not willing to deal with this on every single wide image. The R&D cost needed to overcome this might not be justified considering the size of the target market.
3. In all objectivity, as a famous landscape Japanese photographer was telling me last week when looking at my 300 megapixels panos, very few people need 40 megapixels and even fewer need more. The main value of a 6x7 sensor would be the ability to reach 100 megapixels with large photosites... but who needs that?
Cheers,
Bernard
Megapixels aside, there's still the allure of the special look one can achieve whilst approaching large format ... I mean the dof appearance ...
-
Megapixels aside, there's still the allure of the special look one can achieve whilst approaching large format ... I mean the dof appearance ...
If somebody made a 6x7 sensor, how large would the market really be and what cameras would you put them on anyway?
Used RZ;s and Fuji 680's sell for about a buck fifty so there is no money in the lenses or bodies and unless somebody wanted to develop a whole new medium format digital system (that would probably be very expensive) I doubt seriously if everyone that keeps saying they'd love to have a 6x7 sensor on their RZ is going to run out and buy one even if it existed.
Look at Pentax. They started almost clean sheet and could have gone any direction but they chose a cropped 645 camera, maybe for costs, maybe because they had a built in market of Japanese film photographers that still owned a lot of Pentax 645 glass.
IMO
BC
-
Thoughts re 6 x 7 sensors...
1) 5-6 micron sensors are hovering around the point of being lens limited. There aren't likely to be anything other than small incremental improvements in lens technology.
2) As lens coverage increases from cropped sensor 35mm to full-frame 35mm to 645 to 6 x 7 to large-format there is a gradual drop in achievable resolution from the best lenses but remarkably little. Recent tests I've done suggest something of the order of 20% less resolution from, say, a Schneider 120mm Super Symmar HM (20+ year old lens design) to a brand new Nikon 70-200mm F2.8. The Mamiya 7 lenses are the equal of top 35mm lenses.
The above two points suggest something that was true in the film days: if we want more resolution, in most circumstances it is better to move towards a larger format.
3) 6 x 7 would be significantly more useable on the back of a view camera.
4) Wides would be easier to come by with less of the problems associated with being so close to the sensor (vignetting, colour casts, mechanical issues etc.).
5) The RZ was the industry workhorse. It's a hugely versatile camera with it's close-focus, ability to hand-hold at 1/30 sec., excellent lenses etc. I used to run a pro darkroom in London and I reckon at least 3/4 of the film we dealt with came from RZs or RBs. If 6 x 7 backs were produced, an updated RZ with a gradual program of new lenses could be extremely popular.
6) In the ten years or so that I've been following digital photography it has always been said that the current largest sensor size is the limit. Sensors have always then got bigger. Because of 1) and 2) this process is likely to continue. The P65+, which was perceived as crazy expensive on launch, is quite widely used now. If Phase were to launch a back with a 6 x 7 90MB sensor at Photokina to fit an RZ then I think people like Michael Reichmann would buy it and a reasonably large number of pro photographers would rent it. (Even if the LCD remained crap and live-view sucked).
-
Same blog sees no stitching artefacts on the Canon sensor.
That is, the same blog sees no stitching artifacts on the AA filter, which is where the stitching artifacts are on the Nikon/Sony sensors.
-
Hi,
I was referring to this article:
http://www.chipworks.com/blogs.aspx?id=4626&blogid=86 (http://www.chipworks.com/blogs.aspx?id=4626&blogid=86)
It says: "On the Nikon D3 we found a whopping 6 total Analog Devices AD9974 signal processors – presumably two for each color."
http://www.chipworks.com/uploadedImages/Bl...g/DSLR-Blog.jpg (http://www.chipworks.com/uploadedImages/Blog/Test_Blog/DSLR-Blog.jpg)
Best regards
Erik
It is an integrated analog front end with A-D converter. If this was included among the devices you were referring to, I apologize. I thought you were referring to DSP processing, which is all handled by EXPEED. I don't think this is two processors per color, but a 12 column readout. You know when you see vertical banding at low ISO on a D3/D700/D3s it is generally because of the failure of one of these dedicated column units.
-
http://www.dcviews.com/press/dalsa-140.htm (http://www.dcviews.com/press/dalsa-140.htm)
Dalsa already makes sensors almost in a 9x9 format ...
so it remains a question of costs?
-
Hi,
My idea is just to say that sensor with more pixels need more CPU-power, but that can be achieved, either by having faster devices or by adding more devices. Sorry for being incorrect and thanks for straightening things out.
In my view, given a certain estate of silicon we can implement different number of pixels. Within reason cost will be approximately the same, but we need to add more processing power to handle all data if increase chip density.
Total number of collected photons depends mostly on silicon estate, so total noise and image quality is not really dependent on number of pixels. With increasing number of pixels, resolution will go up (unless limited by lens) and per pixel image quality (Noise and DR) will go down. Total image quality is not much affected by number of pixels, except DR (in the engineering sense) which will decrease with increasing number of pixels, due to reduced "full well capacity" and constant read noise.
My answer to the initial question is "Maybe or Maybe Not".
Best regards
Erik
It is an integrated analog front end with A-D converter. If this was included among the devices you were referring to, I apologize. I thought you were referring to DSP processing, which is all handled by EXPEED. I don't think this is two processors per color, but a 12 column readout. You know when you see vertical banding at low ISO on a D3/D700/D3s it is generally because of the failure of one of these dedicated column units.
-
Total number of collected photons depends mostly on silicon estate, so total noise and image quality is not really dependent on number of pixels. With increasing number of pixels, resolution will go up (unless limited by lens) and per pixel image quality (Noise and DR) will go down. Total image quality is not much affected by number of pixels, except DR (in the engineering sense) which will decrease with increasing number of pixels, due to reduced "full well capacity" and constant read noise
Hi Erik,
A very good summary.
The last part, the reduced DR per sensel surface area, is IMHO under-valued by the school that advocates more sensel density to the extreme! Downsampling, if done properly, will only recover a certain amount of noise but it is the absolute storage capacity (besides read noise and pattern noise) that limits the maximum DR that's achievable.
Cheers,
Bart
-
doubletake - I thought I was on DPReview or whatever its called...
-
Well I said it before and I just hope I'm wrong.
What we will see soon:
- a 80+ Mp back, which still has the great old LCD, or a sligly better one, but still far behind current phones which cost only thousand bucks or so.
- same DR, same noise and nothing new
What I would like to see and the only thing I'm really interested:
-60Mp or similar (Can be 80, but ONLY if other things are significantly improved.
- higher DR I would love to see for example something like sensor + but that not only offers higher ISOs, but a higher DR when bined
- Live View .... I don't believe we will ever get it until I see it
- Noise improvements in the ISO 200 and 800 range
- a LCD on the back, which is GOOD, large and can be viewed anywhere.
Hey Christopher
The rumors I hear is that phase will present an 80Mp back at photokina - Let's hope that not only is a larger sensor in the same old wrapping
If it's from Dalsa we can expect great color reproduction but probably also forget about improvements at high ISO!
/David
-
Hey Christopher
The rumors I hear is that phase will present an 80Mp back at photokina - Let's hope that not only is a larger sensor in the same old wrapping
If it's from Dalsa we can expect great color reproduction but probably also forget about improvements at high ISO!
/David
How much truth do you attribute to this rumors? 10%, 50 %, 100%? Reliable source?
Regards
Paul
-
How much truth do you attribute to this rumors? 10%, 50 %, 100%? Reliable source?
Regards
Paul
Hard to say obviously - but when you start hearing it from more than one source you start thinking.
Bets are open
/David
-
Am I the only one who would be ok with a 30-50mp back, as long as it was 6x7? Because seriously, that's all I want or need.
the P45 with 39mp was enough and most part of the time too much , that stupid pixel race will end some day
-
I'd be ok with a 31 mp or even 22 mp back as long as it would be a no-crop 6x7.. and let's not go back to the '6x7 back would be 120 not 20 mp' thing. Sensor dimensions > sensor resolution, I just want to stress that. And I'm quite sure I'm not alone in thinking that way. Unfortunately 67 back is as unrealistic as a usable live preview on MFDB. As long as there are no new 67 camera designs, there's not enough money to draw the industry's attention.
-
A 6x7 mm sensor with a res of 31 mp would have 11.6 micron square pixels. That would morie like hell.
-
A 6x7 mm sensor with a res of 31 mp would have 11.6 micron square pixels. That would morie like hell.
Yeah but it would be great for anything that needed big DOF and small apertures, and the bigger the sensor the more challenging that becomes. Also I may be wrong, but my guess is larger sensel pitch will allow better sensitivity and possible improvements in DR. There probably is no one best solution for all work.
-
Yeah but it would be great for anything that needed big DOF and small apertures, and the bigger the sensor the more challenging that becomes. Also I may be wrong, but my guess is larger sensel pitch will allow better sensitivity and possible improvements in DR. There probably is no one best solution for all work.
You make a good point. Personally I would like to see manufacturers stop the resolution race, stick with what they have, and increase quality with respect to noise, sensitivity, and dynamic range.
-
You make a good point. Personally I would like to see manufacturers stop the resolution race, stick with what they have, and increase quality with respect to noise, sensitivity, and dynamic range.
Hum... companies design for the benchmarks their products will be submited to. The only DR benchmark I know of is DxO and their results are discarded as soon as they don't match people expectations...
In the end DR claims will end up being perceived as marketing claims since we have de facto collectively decided to give up on DR measurments.
The logical consequence of our choices will be a reduced focus from manufacturers on DR.
Cheers,
Bernard
-
No one has even mentioned they cannot even make a decent LCD screen...
I am waiting for the 40 megapixel iPhone to come out until I invest anymore...
S.
-
Why not an iPad to view the pictures in a decent format ?
PdF
-
I believe there will never be a backing up of resolution for medium/large format.
I do hope they keep photosite size where it is or at least don't go too much smaller. I wouldn't rule out that it may go bigger, but if so, I would expect by very little. We have customers using digital backs with 9 and 12 micron photosites who can't shoot higher than 50 ISO without gaining unacceptable levels of noise, color artifacting, etc. Of course that is older technology, but other than Canon with the G11/S90, I can't think of anyone who has backed up on resolution.
The resolution issue is so interesting because I hear so often how "who needs so and so many megapixels, stop already". But I've been hearing this ever since the 22MP DB's came out. And yet, the P65+ has been a huge success.
Canon and Nikon will continue to up the megapixel count, and yet when it is discussed as in rumor about the next model, it's usually as a positive "The 1DS-MKIV is supposed to be 32MP's!". There's very little negativity associated with it....yet.
And for sure that is because Canon/Nikon do so much more with their cameras and in ways that more effectively benefit the demands of commercial photography today. So medium format is seen as concentrating too much on megapixels, size of file, etc, rather than making their cameras more flexible and versatile, ala 35mm. But currently, that is one of medium format's few remaining advantages over 35mm. I can guarantee you that a $28,000 6x7, 24MP digital back will fail in the market place. While some may see advantages in the lower resolution, the majority will pass at paying that premium for a product that captures only 2/3 the file size of 35mm.
I agree and believe that medium format has focused on resolution and chip size enhancement (though modestly), but more because other enhancements take a long time. I know they have been working on them. But in the meantime, you can't fault a company for putting out products that people buy and that do produce revenue. They do have to make money while they work on technology that may be 4-6-8 years away. It kind of puts a drain on R&D if you don't have revenue coming in.
Compared to 35mm, medium format has always had the advantage of bigger imaging area, larger image/file size. Those have been their primary advantages. And those advantages were traditionally necessary and a clear reason to choose them. With the changes in the commercial markets, the type of photography being demanded in general (even with wedding, portrait, architecture, etc), the need for speed, flexiblility, maneuverability, and with 35mm now producing (somewhere) in the range of traditional medium format image quality, (and not to mention crimped stills budgets) medium format's big advantages have lost their influence with commercial photographers (though upping it with other markets that value those assets and appreciate the evolution medium format has made from film-based to digital).
So...I expect that we will see more resolution announced this fall, hopefully spread over a larger sensor size, and hopefully progress will have been made on the issues medium format struggles with - usable LCD, in-camera functionality, etc. I see this as a benefit to those in the commercial field (and elsewhere) who still choose to use medium format for its strengths. It helps make that choice to remain with medium format easier, which is a very big plus. For those who don't benefit from medium format's strengths, it probably doesn't matter how improved in terms of "usability" medium format gets because it is, at least for the time being, not the right product for them and that choice has already been made.
We'll see.
Steve Hendrix
-
Hi Steve,
Excellent summary, seen from a technical/economical standpoint.
Best regards
Erik
I believe there will never be a backing up of resolution for medium/large format.
I do hope they keep photosite size where it is or at least don't go too much smaller. I wouldn't rule out that it may go bigger, but if so, I would expect by very little. We have customers using digital backs with 9 and 12 micron photosites who can't shoot higher than 50 ISO without gaining unacceptable levels of noise, color artifacting, etc. Of course that is older technology, but other than Canon with the G11/S90, I can't think of anyone who has backed up on resolution.
The resolution issue is so interesting because I hear so often how "who needs so and so many megapixels, stop already". But I've been hearing this ever since the 22MP DB's came out. And yet, the P65+ has been a huge success.
Canon and Nikon will continue to up the megapixel count, and yet when it is discussed as in rumor about the next model, it's usually as a positive "The 1DS-MKIV is supposed to be 32MP's!". There's very little negativity associated with it....yet.
And for sure that is because Canon/Nikon do so much more with their cameras and in ways that more effectively benefit the demands of commercial photography today. So medium format is seen as concentrating too much on megapixels, size of file, etc, rather than making their cameras more flexible and versatile, ala 35mm. But currently, that is one of medium format's few remaining advantages over 35mm. I can guarantee you that a $28,000 6x7, 24MP digital back will fail in the market place. While some may see advantages in the lower resolution, the majority will pass at paying that premium for a product that captures only 2/3 the file size of 35mm.
I agree and believe that medium format has focused on resolution and chip size enhancement (though modestly), but more because other enhancements take a long time. I know they have been working on them. But in the meantime, you can't fault a company for putting out products that people buy and that do produce revenue. They do have to make money while they work on technology that may be 4-6-8 years away. It kind of puts a drain on R&D if you don't have revenue coming in.
Compared to 35mm, medium format has always had the advantage of bigger imaging area, larger image/file size. Those have been their primary advantages. And those advantages were traditionally necessary and a clear reason to choose them. With the changes in the commercial markets, the type of photography being demanded in general (even with wedding, portrait, architecture, etc), the need for speed, flexiblility, maneuverability, and with 35mm now producing (somewhere) in the range of traditional medium format image quality, (and not to mention crimped stills budgets) medium format's big advantages have lost their influence with commercial photographers (though upping it with other markets that value those assets and appreciate the evolution medium format has made from film-based to digital).
So...I expect that we will see more resolution announced this fall, hopefully spread over a larger sensor size, and hopefully progress will have been made on the issues medium format struggles with - usable LCD, in-camera functionality, etc. I see this as a benefit to those in the commercial field (and elsewhere) who still choose to use medium format for its strengths. It helps make that choice to remain with medium format easier, which is a very big plus. For those who don't benefit from medium format's strengths, it probably doesn't matter how improved in terms of "usability" medium format gets because it is, at least for the time being, not the right product for them and that choice has already been made.
We'll see.
Steve Hendrix
-
i agree steve, very good written and analysed. and very honest.
-
One more thumbs up for Steve's discussion. It seems that the MF backs with higher pixel counts keep being the most profitable: Steve implies that they sell well by MF standards, and the price premium for more MP seems to exceed the extra component costs, so the markups are probably highest for the higher MP models. Given that message from the marketplace, my guess even more is that the main change in MF sensors this year is likely to be larger sensors at the top of the line, with the same 6 micron pixel size ... and this is the best that those who fear "excessive pixel counts" can hope for.
-
I like Steve's honesty -but have a minor remark: Nikon has had good success pitching the large-pixel D700 against the 5DII which outresolves it by a factor of 2 (2 generations of sensors).
Edmund
One more thumbs up for Steve's discussion. It seems that the MF backs with higher pixel counts keep being the most profitable: Steve implies that they sell well by MF standards, and the price premium for more MP seems to exceed the extra component costs, so the markups are probably highest for the higher MP models. Given that message from the marketplace, my guess even more is that the main change in MF sensors this year is likely to be larger sensors at the top of the line, with the same 6 micron pixel size ... and this is the best that those who fear "excessive pixel counts" can hope for.
-
Hum... companies design for the benchmarks their products will be submited to. The only DR benchmark I know of is DxO and their results are discarded as soon as they don't match people expectations...
In the end DR claims will end up being perceived as marketing claims since we have de facto collectively decided to give up on DR measurments.
The logical consequence of our choices will be a reduced focus from manufacturers on DR.
Cheers,
Bernard
DXO isn't doing a great job with respect to measuring DR for MF DB. As they sell their own software, it wouldn't be to their advantage to measure how much better a MFDB file gets when run through its own manufacturers software. That's the part they have left out and its an important part of the MFDB process. I don't take any stock in DXO measurements or those on other sites where they measure DR with the IEEE specs and not ones useful to photographers.
I still would like to see more DR from the next backs but wonder if its possible. There's hardly been any change in DR of the digital backs (at base ISO) for a number of generations.
-
Let's say the new size is it's P90+ with 120MP how far are we compared to 8x10 Format?
any info.. Dalsa or Kodak?
-
Just looked at the Kodak webpage, their largest sensor is 50 mp with a 6 micron pixel, same dimensions as the P45 sensor. Maybe we will see a P50? as a true upgrade from the P45?
-
DXO isn't doing a great job with respect to measuring DR for MF DB. As they sell their own software, it wouldn't be to their advantage to measure how much better a MFDB file gets when run through its own manufacturers software. That's the part they have left out and its an important part of the MFDB process. I don't take any stock in DXO measurements or those on other sites where they measure DR with the IEEE specs and not ones useful to photographers.
I still would like to see more DR from the next backs but wonder if its possible. There's hardly been any change in DR of the digital backs (at base ISO) for a number of generations.
DxO provides the information you seek, if you know where to look.
http://luminous-landscape.com/forum/index....c=42158&hl= (http://luminous-landscape.com/forum/index.php?showtopic=42158&hl=)
For image quality to improve in the most useful part of the range, sensors need to have higher quantum efficiency (the percentage of incident photons that are recorded by the sensor) and lower read noise. It is on the latter score that MFDB's are woeful relative to contemporary CMOS-based DSLR's.
-
Just looked at the Kodak webpage, their largest sensor is 50 mp with a 6 micron pixel, same dimensions as the P45 sensor. Maybe we will see a P50? as a true upgrade from the P45?
I like the idea of having good quality long exposures even though i don't use it daily.. I was looking at P65+ Dalsa backs and heard they are limited to 30 seconds or so..
Then I read about stacking Long Exposure images to make it look like 5, 10, 15, 30 minutes exposure.. but there is limited or no info/tutorial how to do it successfully..
-
For image quality to improve in the most useful part of the range, sensors need to have higher quantum efficiency (the percentage of incident photons that are recorded by the sensor) and lower read noise. It is on the latter score that MFDB's are woeful relative to contemporary CMOS-based DSLR's.
Well MFDB can't be all that woeful because they still have stops more DR than CMOS DSLR's in that kind of test. What's that chart show only 7.5 stops for the nikon D3x at base iso and about 4 at 6400 iso? Test any MFDB using the same S/N acceptance ratio and you'll have more like 10.5 or 11 at base ISO.
It looks like an increase in well size is going to help with read noise and that's what you see if you look through the sensor spec sheets for sensors designed for scientific uses (like astronomy) that have big wells. Some of these have a SNR spec of 90 db. But really, I was wondering if it would be possible to get much more DR without going to higher bit files and chip architecture. It seems possible but in practice it hasn't been happening.
-
Well MFDB can't be all that woeful because they still have stops more DR than CMOS DSLR's in that kind of test. What's that chart show only 7.5 stops for the nikon D3x at base iso and about 4 at 6400 iso? Test any MFDB using the same S/N acceptance ratio and you'll have more like 10.5 or 11 at base ISO.
Sorry, wishful thinking. If one uses the same criterion (range of EV over which the SNR is greater than 24dB) as in the example in the post I linked to above, the P65+ only gets about 7.0 stops DR at base ISO. Again this can be read off the relevant SNR plot at DxO's test site (http://www.dxomark.com/index.php/en/Camera-Sensor/All-tested-sensors/Phase-One/P65-Plus)
Now, that is DR per pixel. But DR is depends on scale in the image, since SNR is typically higher on coarse scales and lower on fine scales (eg, the more you magnify an image, the more apparent noise becomes, and thus the lower the S/N ratio). To fairly compare, one should look at DR at a fixed spatial scale, which means scaling the result by the square root of the pixel counts. sqrt[65/24]~1.65 or about 0.7 stops extra for the P65. In other words, the ISO 100 DR of the P65+ at the spatial scale of D3x pixels is about 0.3 stops better. A third of a stop, that's all. Not so great, consider it is gathering twice as much light to begin with (all other things being equal that would mean it should have a half stop advantage rather than a third stop).
It looks like an increase in well size is going to help with read noise and that's what you see if you look through the sensor spec sheets for sensors designed for scientific uses (like astronomy) that have big wells. Some of these have a SNR spec of 90 db. But really, I was wondering if it would be possible to get much more DR without going to higher bit files and chip architecture. It seems possible but in practice it hasn't been happening.
Read noise depends on all sorts of design choices, and is not necessarily correlated to well capacity. I could give examples if you like. I'm not sure it's germaine to bring in astro CCD specs, that's a quite different set of design criteria for image capture. As far as bit depth is concerned, one only needs as many bits as there are stops of engineering DR. Present-day MFDB's need no more than 12 bits per pixel, perhaps a tad more if oversampling confers an advantage (though I have yet to see such).
-
Hi,
The DxOmark data is based on raw data without conversion. If you look at the data Emil referenced you will find that the P65+ has a bit less DR at 24 DB SNR than the D3X.
Also, just realize that would MFDBs have a real advantage in DR it would also mean that they would perform well at high ISO. It seems obvious that they don't do that.
Emils explanation of pattern noise is interesting. Also note that in DxO tests Phase One P65 is a little bit better than Nikon D3X and Nikon D3X is much better than Canon 1DsIII, Canon 5DII or Sony Alpha 900. Also note that the D3X probably achieves it's rating in the slow 14 bit/pixel mode.
Check also this: http://www.dxomark.com/index.php/en/Our-pu...e-Phase-One-P65 (http://www.dxomark.com/index.php/en/Our-publications/DxOMark-reviews/DxOMark-review-for-the-Phase-One-P65).
That said, many renown photographers claim significant advantage of MFDBs over DSLRs.
Astronomy is more about collecting light than resolution. They normally photograph stars which are single point light sources. I guess that the sensors are cooled, probably with liquid nitrogen.
Best regards
Erik
Well MFDB can't be all that woeful because they still have stops more DR than CMOS DSLR's in that kind of test. What's that chart show only 7.5 stops for the nikon D3x at base iso and about 4 at 6400 iso? Test any MFDB using the same S/N acceptance ratio and you'll have more like 10.5 or 11 at base ISO.
It looks like an increase in well size is going to help with read noise and that's what you see if you look through the sensor spec sheets for sensors designed for scientific uses (like astronomy) that have big wells. Some of these have a SNR spec of 90 db. But really, I was wondering if it would be possible to get much more DR without going to higher bit files and chip architecture. It seems possible but in practice it hasn't been happening.
-
Hi,
Two reasons probably.
1) Nikon doesn't have a 24 MP model below USD 3000
2) Utilizing those 24 MPixels take a tripod and good technique
Best regards
Erik
I like Steve's honesty -but have a minor remark: Nikon has had good success pitching the large-pixel D700 against the 5DII which outresolves it by a factor of 2 (2 generations of sensors).
Edmund
-
I have another thought for you: if Nikon priced its 24MP product the same as the 12MP - which they almost could, same body, same chip size and yield, same memory bandwidth- their press customers get confused, buy it, and suddenly get noisy unfocused images at pixel level and the product wouldn't sell. This way they customers get a camera which is very fast in frame rates, focuses perfectly, and has ISO of 6400 no sweat - then the customers don't *try* to buy better.It's an astute marketing trick, steering customers to the camera they *need* by pricing it lower.
When I image a fashion show with the D3x, I know the guys with the D3 will get better results.
Edmund
Hi,
Two reasons probably.
1) Nikon doesn't have a 24 MP model below USD 3000
2) Utilizing those 24 MPixels take a tripod and good technique
Best regards
Erik
-
That said, many renown photographers claim significant advantage of MFDBs over DSLRs.
I hate to have to write this because I don't really care, but:
- Few, if any, use the D3x,
- These claims are never backed up by any data or sample images for that matter.
What many backs users are seeing are backs that are calibrated for under-exposure, which gives them the impression that they have more highlight headroom for a given "correct exposure" (as defined by a traditional film based approach). More apparent highlight headroom is materialized by the mythical ability to recover blown highlights. I write "mythical" because it doesn't exist with the linear sensors we have today.
Highlight recovery can only be made possible by the existence of data in the first place, which is only possbile because the back histogram doesn't show the real information available in the files. Put it otherwise, a digital camera equiped with a linear sensor (read 100% of sensors on the market today) can only show highlight recovery ability if the system is calibrated for under exposure.
In other words, any DR claim based on apparent highlight recovery ability alone is misguided in the first place.
Any DR comparison has to be made based on actual raw ETTR (regardless of exposure and of in camera histograms) and can only be made by looking at how noisy the shadows are.
Now this comparision can be done and might show some advantage to the backs, but we are getting awfully close to the DxO results and the only possible difference between their results and actual prints comparision is in the "look of the noise".
If this has been done and showed results differing significantly from DxO results, how come we don't see these results? Again, don't show me 5DII results, I know about the noise banding at ISO 100 in deep shadows.
More often than not things that are hard to find end up not existing at all...
Cheers,
Bernard
-
Sorry, wishful thinking. If one uses the same criterion (range of EV over which the SNR is greater than 24dB) as in the example in the post I linked to above, the P65+ only gets about 7.0 stops DR at base ISO. Again this can be read off the relevant SNR plot at DxO's test site (http://www.dxomark.com/index.php/en/Camera-Sensor/All-tested-sensors/Phase-One/P65-Plus)
Now, that is DR per pixel. But DR is depends on scale in the image, since SNR is typically higher on coarse scales and lower on fine scales (eg, the more you magnify an image, the more apparent noise becomes, and thus the lower the S/N ratio). To fairly compare, one should look at DR at a fixed spatial scale, which means scaling the result by the square root of the pixel counts. sqrt[65/24]~1.65 or about 0.7 stops extra for the P65. In other words, the ISO 100 DR of the P65+ at the spatial scale of D3x pixels is about 0.3 stops better. A third of a stop, that's all. Not so great, consider it is gathering twice as much light to begin with (all other things being equal that would mean it should have a half stop advantage rather than a third stop).
Read noise depends on all sorts of design choices, and is not necessarily correlated to well capacity. I could give examples if you like. I'm not sure it's germaine to bring in astro CCD specs, that's a quite different set of design criteria for image capture. As far as bit depth is concerned, one only needs as many bits as there are stops of engineering DR. Present-day MFDB's need no more than 12 bits per pixel, perhaps a tad more if oversampling confers an advantage (though I have yet to see such).
I'm suggesting that you check DXO's initial assumption before cranking out a lot of science based on their data. DXO is not using C1 to convert phase files first and that improves them significantly. DSLR do a fair bit of file processing in the camera before the RAW is written. MFDB handle this differently and do some final file processing in software such as black frame subtraction. You just can't look at these platforms as the same. If DXO makes the leap that they are then linking to DXO as reference isn't going to help. Don't forget they are not a scientific body, rather they are a for profit company and one that happens to make software only for DSLR and not for MFDB.
-
Hi,
The DxOmark data is based on raw data without conversion. If you look at the data Emil referenced you will find that the P65+ has a bit less DR at 24 DB SNR than the D3X.
Best regards
Erik
This is exactly my point - were DXO to repeat the test with the black frame subtracted and what ever else MFDB software is doing then the DR figures would change. In any case it would be really exciting to see a larger sensor at photokina.
-
This is exactly my point - were DXO to repeat the test with the black frame subtracted and what ever else MFDB software is doing then the DR figures would change. In any case it would be really exciting to see a larger sensor at photokina.
basically bernhard is 100% right which was he says. c1 is not changing the raw data, the teoretical dr is determined by the raw information and cant be increased by the sw, but it can be decreased by bad code.
highlight recovery is done by the software and can be done very different, this means the restoration of color- neutral data if one or two of the color channels still hold some information, meanwhile one channel (or two) are clipped.
this is not different if used mf data or 35mm data, how usefull it is depends on the sw and if it either clips the whole file if one channel is clipped or if it use the rest information and is able to get colorshift free highlights. if done bad this mostly results in cyan or magenta casts in these highlight zones.
it should not be valued by DXO, cause its not a real part of the DR, it depends too much on the kind of image if these infos can be restore and - as i wrote above - this has nothing to do with the sensor size.
-
Let's say the new size is it's P90+ with 120MP how far are we compared to 8x10 Format?
any info.. Dalsa or Kodak?
By what criteria do you want to make the comparison?
1) You get considerably more accurate colour from a P65+ than any form of scanned film, including 8x10.
2) You also get considerably less dynamic range (at the highlight end) than scanned colour negative (but significantly more than scanned colour transparency).
3) The P65+ will generate unpleasant moire under relevant conditions, it can also render neon and some other subjects in strange bad ways that are hard to correct.
In real world tests a P65+ with a good lens will generally yield similar or even slightly better levels of resolution than a scanned sheet of 4x5 film. However, 8x10 is still in another league. The increase in resolution going from 4x5 to 8x10 scanned film is only slightly less than a straightforward linear geometric increase. To equal 8x10 resolution, even going to a 5 micron sensor (about the limit with foreseeable lens technology), will require something of the order of 240MP and a sensor size of around 66mm x 90mm.
However, to put this in perspective, shooting 8x10 is a royal pain and very prone to technical errors which render its potential rather moot. Of course it also has very shallow depth of field so in most circumstances the resolution advantage only applies to shots focused at infinity without the requirement of sharp foreground information. Because of this and because of point 1) a digital back some way short of this size and resolution would still be effectively superior to even 8x10 film for virtually all purposes.
-
This is exactly my point - were DXO to repeat the test with the black frame subtracted and what ever else MFDB software is doing then the DR figures would change.
Hi Eric,
While a blackframe subtraction would change the DR, it is not for the better. Simple blackframe subtraction is only for the removal of non-random (pattern) noise, it increase random noise levels (by up to 40% for the shadows).
The DxO database is not such a bad startingpoint.
Cheers,
Bart
-
Hi!
I'd suggest that these are excellent points. I would just add that reducing Moiré without taking resort to AA-filtering is one of the advantages of increasing resolution. As Bernard always would point out, you can always increase resolution by stitching. That approach may be practical for landscape photography, studio may be another thing.
'Horses for the courses'
Best regards
Erik
By what criteria do you want to make the comparison?
1) You get considerably more accurate colour from a P65+ than any form of scanned film, including 8x10.
2) You also get considerably less dynamic range (at the highlight end) than scanned colour negative (but significantly more than scanned colour transparency).
3) The P65+ will generate unpleasant moire under relevant conditions, it can also render neon and some other subjects in strange bad ways that are hard to correct.
In real world tests a P65+ with a good lens will generally yield similar or even slightly better levels of resolution than a scanned sheet of 4x5 film. However, 8x10 is still in another league. The increase in resolution going from 4x5 to 8x10 scanned film is only slightly less than a straightforward linear geometric increase. To equal 8x10 resolution, even going to a 5 micron sensor (about the limit with foreseeable lens technology), will require something of the order of 240MP and a sensor size of around 66mm x 90mm.
However, to put this in perspective, shooting 8x10 is a royal pain and very prone to technical errors which render its potential rather moot. Of course it also has very shallow depth of field so in most circumstances the resolution advantage only applies to shots focused at infinity without the requirement of sharp foreground information. Because of this and because of point 1) a digital back some way short of this size and resolution would still be effectively superior to even 8x10 film for virtually all purposes.
-
Hi Eric,
While a blackframe subtraction would change the DR, it is not for the better. Simple blackframe subtraction is only for the removal of non-random (pattern) noise, it increase random noise levels (by up to 40% for the shadows).
The DxO database is not such a bad startingpoint.
Cheers,
Bart
Thanks Bart.
I don't know exactly what Phase is doing with their files in C1 in addition to the black frame subtraction but it seems like more. I guess this could be tested and the numbers compared to find out how much. I don't believe my imatest license is still active so I won't try this but it would be an easy thing to do to see how much improvement in at least DR can be had with or without C1 and also a good way to show the real DR differences between backs and DSLRs. If Imatest and DXO agreed on DR values for the stricter S/N ratio that's meaningful to photographers I'd be surprised but would of course be able to let this rest. Note how the DR of the D3x dropped down from just under 13 stops to only 7.5 stops DR at base ISO when changing the acceptable S/N ratio to something meaningful to photographers.... That's a big adjustment! I don't think DXO does any service by even using the high values since they are not applicable to photographers. They should only be using the set of data Emil posted. Anyhow last time I tested DSLR and CCD sensors, the drop in the DSLR was dramatic like the D3x to almost half, however the drop in the MFDB was only 1 stop or so - but then I used capture one to convert the raws... I'd expect the same to be true now. Looking at the usable DR not theoretical max MFDB's probably have about 10.5 to 11 stops while the DSLR's have something like the nikon d3x does - 7.5 to 8 stops.
It all has to do with how it's measured and DXO is not doing anyone any favors (except themselves).
-
Note how the DR of the D3x dropped down from just under 13 stops to only 7.5 stops DR at base ISO when changing the acceptable S/N ratio to something meaningful to photographers.... That's a big adjustment! I don't think DXO does any service by even using the high values since they are not applicable to photographers. They should only be using the set of data Emil posted.
Eric,
As explained to you by Emil a few posts above (here (http://luminous-landscape.com/forum/index.php?showtopic=44390&pid=374829&mode=threaded&start=#entry374829)), the P65+ shows 7 stops of DR in the same contiditons where the D3x shows 7.5.
Do you believe in their results?
Cheers,
Bernard
-
Hi Eric,
To begin with, Phase One raw files have a lot of extra calibration data that C1 can utilize to improve image quality.
Regarding DR I'd suggest that it is actually very simple. You measure it exposing something like a "Stoufer wedge" and measure noise in each area. DxO actually publishes all data. So anyone can apply any threshold on SNR. Keep in mind that DR is an engineering term and it's standard definition is what DxO is measuring. I actually would suggest that DxO does a service to the community by presenting accurate data. The fact that the community doesn't know how to interpret data doesn't invalidate DxO measurements. There is no data available on DR at all, except those published by DxO.
An interesting point by Emil Martinec is that the vision system is sensitive to different noise patterns. Minolta sensors used to have a patchy look in ACR based RAW-converters. Panopeeper has analyzed a lot of exposures I made with a Sony Alpha 900 and interpreted it as noise reduction, whereas other authors regarded it to be round of error in AD conversion. With LR3 the patchyness is gone and we have a nice tight grain pattern instead. DR is not increased, but the image is much more useful. Some authors say that the Canon D1s# cameras have ugly pattern noise and the Nikon D3X is much better in that sense.
I am much opposed to the approach to miss credit any finding that we don't like. DxO has a bunch of engineers and they measure DR according to established procedures, that certainly has a value.
Michael Reichmann has compared the Canon 5DII, the Sony Alpha 900 and the Nikon D3X and found that the image quality was similar at low ISO, but he didn't compare DR as he considered it to be difficult to measure. In my view that was a resaonable approach
Regarding DR we can reproduce about seven stops on paper. THe DR of papers, printer and ink in combination is about 7 stops. So whenever we discuss DR in prints the DR has been compressed by applying a lot of manipulations lake gradation curves and clarity. It's very possible that Phase One can pull out more information from an image than ACR using proprietary information from individual sensor calibration and other data.
One of the issues I see that terms are invented to explain differences we see, but there is no reference frame for those terms. There is talk about microcontrast, tonality and so on. DR is also like that, except it has a technical definition (and I think an ANSI standard for measuring it).
Best regards
Erik
Thanks Bart.
It all has to do with how it's measured and DXO is not doing anyone any favors (except themselves).
-
Bernard,
Many on this forums tend to discredit DxO data if they don't like it. In my view, discrediting measurements is not a scientific approach. It's like the old practice of shooting the messenger when you don't like the message.
It's OK to say that measurements are wrong but you need to state in what way they are wrong or produce evidence to the contrary. It's quite obvious that there is a lot of data at DxO if you care to dig into it, like Emil's comparison that you quoted.
Best regards
Erik
Eric,
As explained to you by Emil a few posts above, the P65+ shows 7 stops of DR in the same contiditons where the D3x shows 7.5.
Do you believe in their results?
Cheers,
Bernard
-
Thanks Bart.
I don't know exactly what Phase is doing with their files in C1 in addition to the black frame subtraction but it seems like more. I guess this could be tested and the numbers compared to find out how much.
I have not seen this documented anywhere, but C1 does seem to apply automatically noise reduction selectively to shadow areas.
This is true not just for back files, it also does it on D3x shadows.
Cheers,
Bernard
-
It's OK to say that measurements are wrong but you need to state in what way they are wrong or produce evidence to the contrary. It's quite obvious that there is a lot of data at DxO if you care to dig into it, like Emil's comparison that you quoted.
Erik,
Yes, indeed.
As was said several time, the lack of evidence that DxO is wrong, months after the release of their data, is probably the best proof that they are right...
Cheers,
Bernard
-
1) You get considerably more accurate colour from a P65+ than any form of scanned film, including 8x10.
2) You also get considerably less dynamic range (at the highlight end) than scanned colour negative (but significantly more than scanned colour transparency).
3) The P65+ will generate unpleasant moire under relevant conditions, it can also render neon and some other subjects in strange bad ways that are hard to correct.
I concur that the above post #60 by Steve Hendrix speaks very well indeed. Else it is remarkable that any post on MFDB appear to take turn in posts comparing to DSLR. Those tools are completely different, and not merely in image quality. It has been stated in film days already that stepping up from SLR to medium format causes an immediate improvement in ones photography. I believe the reason is that we see better and slow down more with medium format. It lends to a more deliberate and planned approach.
In contrast to medium format cameras, DSLRs are seemingly being stuffed with more and more wizbang and buttons by each new iteration. Personally I prefer more analog interface by medium format and am keen on technical camera with even more such. Simple and manual makes me focus more on photography rather than the interface. We are all different. While technological advancement is great, I for one hand the TV remote to my wife because it has many buttons! Simply my mind is more analogue. That is one advantage with MDFB and associated cameras that is worthy to mention as a great strength. Some people like all the electronic features and interface, while others like me much prefer an analogue interface.
The above list 1-3 is actually exact contrary to the list of where I would like to see MFDB improve:
1) Colors may be accurate with MFDB and more so than DSLR, however as a landscape shooter I do not want accurate colors. What I want is colors that pleasingly render nature and light in same manner or better as Velvia 50 do. No digital today enables you to reach that objective with any means of ease. And no, it is not possible today to replicate the beauty of Fuji Velvia 50. This is why I still also shoot film.
2) DR on MFDB is great and exceeds DSLR, however the problem with digital is that it is a linear capture and does not have the nature of film at highlights. That is one of the weaknesses with digital today. What happened to Fuji's patent of a new type of sensor? We need a new type of sensors with a response similar to film.
3) I use Aptus 65 28MP MFDB and have not encountered much worthy of problem with moire. With higher resolution the problem becomes smaller, unless I am mistaken. However, there are tools to deal with it, e.g. in Capture One. Thus it is nil issue for MFDB.
In regards to resolution, also my 28MP digital back is suffice for portraits. However, for landscapes I would prefer around 80MP for landscapes, both to get high resolution single captures in 4x5 proportions, but also single crops in decent size of 617 proportions. I think more than 80MP in FF 645 sensors would be around what todays sharpest lenses can resolve from Bayer type sensors and in order not not make more problem with tolerances on equipment. After that, yes... would be great with larger sensors instead of having to buy $$$ set of new higher resolving lenses. The question is perhaps if the industry will choose such path instead of trying make us buy new lenses and cameras?
And frankly why stick to 645 proportioned sensors (1.33:1) instead of 4x5 proportioned (1.25:1)? With same image circle, a 4x5 proportioned sensor will yield a larger area, as of course... by many considered a far more pleasant proportion of frame.
Regards
Anders
-
2) DR on MFDB is great and exceeds DSLR, however the problem with digital is that it is a linear capture and does not have the nature of film at highlights. That is one of the weaknesses with digital today. What happened to Fuji's patent of a new type of sensor? We need a new type of sensors with a response similar to film.
Anders,
Why is this a problem? It just requires a different approach to exposure.
The DR of backs is already way superior to that of Velvia and that is IMHO the main reason why we sometimes feel that digital is less pleasant colorwise. When shooting Velvia, especially when you need to scan it, you know that you have less than 5 stops of usable DR, and that forces you to concentrate on scenes where the sweet light hits the right spot.
When shooting digital there is much more freedom to address other types of scenes where the light isn't exactly how it should be. That works, but in the end the colors feel less nice just because the light was less sweet.
Just my 2 cents.
Cheers,
Bernard
-
Anders,
Why is this a problem? It just requires a different approach to exposure.
The DR of backs is already way superior to that of Velvia and that is IMHO the main reason why we sometimes feel that digital is less pleasant colorwise. When shooting Velvia, especially when you need to scan it, you know that you have less than 5 stops of usable DR, and that forces you to concentrate on scenes where the sweet light hits the right spot.
When shooting digital there is much more freedom to address other types of scenes where the light isn't exactly how it should be. That works, but in the end the colors feel less nice just because the light was less sweet.
Just my 2 cents.
Cheers,
Bernard
Bernard,
Simply the rendering is different in highlights no matter how adjust/expose. This is an issue primarily for transient light, where is a very large DR. Film simply has the transition towards blown out white that is more pleasing.
Rgds
-
Bernard,
Many on this forums tend to discredit DxO data if they don't like it. In my view, discrediting measurements is not a scientific approach. It's like the old practice of shooting the messenger when you don't like the message.
It's OK to say that measurements are wrong but you need to state in what way they are wrong or produce evidence to the contrary. It's quite obvious that there is a lot of data at DxO if you care to dig into it, like Emil's comparison that you quoted.
Best regards
Erik
Erik,
I'm not discrediting DXO because I don't like it. I'm pointing out a valid flaw in the way they approach MFDB. Your argument re:shooting the messenger is silly. Why do you defend them? I have a stoufer transmission step wedge and have used it to test DR with imatest software in the past. I know how it works so save yourself the trouble.
Bernard,
I didn't answer your question because I've had you set to ignore since I got tired of reading that stitching was the solution for everything. But your question was answered - DXO data for MFDB is not accurate for the reasons I've already pointed out. When I see
To all reading this point - the fuss about DR is because I'm hoping that any new sensors that come out at photokina will have more improvement in DR than in pixels. I'd rather see a bigger sensor preferably one with a more square ratio than a rectangle.
-
Bernard,
I didn't answer your question because I've had you set to ignore since I got tired of reading that stitching was the solution for everything. But your question was answered - DXO data for MFDB is not accurate for the reasons I've already pointed out. When I see
Fine Eric, whatever works for you sir.
Cheers,
Bernard
-
Bernard,
Simply the rendering is different in highlights no matter how adjust/expose. This is an issue primarily for transient light, where is a very large DR. Film simply has the transition towards blown out white that is more pleasing.
I see, it is true that the transition to the edges of the sun is often handled better by slides/nega.
Cheers,
Bernard
-
I'm suggesting that you check DXO's initial assumption before cranking out a lot of science based on their data. DXO is not using C1 to convert phase files first and that improves them significantly.
True, it does improve my D3x files also thanks to their automatic shadow cleaning algos. Very nice I have to say! I am sure that the D3x would gain some more DR stops if it were measured on C1 conversions.
Cheers,
Bernard
-
... Nikon has had good success pitching the large-pixel D700 against the 5DII which outresolves it by a factor of 2 (2 generations of sensors).
Edmund
True: the observations about evidence of a strong market preference for greater image detail (more MP, etc.) were addressed to the medium format market. The lower pixel count of the D700 compared to similarly priced alternatives arguably helps it to offer better low light performance and a higher frame rate, but neither of these is of much of a selling point for lower pixel count MF backs.
-
Hi,
The way I see it DxO does publish measurements, no one else does. So the only data we have freely available to compare cameras is coming from DxO. The reason i defend DxO is:
1) Their measurements are based on standards and comparable across systems
2) Generally, I got the impression that imaging experts regard the DxO data as relevant
3) Generally, I got the impression that DxO data is quite consistent with specifications of sensors
4) I have not seen any firm evidence of the several step advantage of MFDB CCDs over DSLRs sensors.
Now, it would be easy to post two correct exposures of Stoufer wedges as RAW files, so anyone could see the difference with their own eyes. To my knowledge no one has made such files downloadable. A few years ago a Swedish periodical did publish Stouffer wedge exposures with different sensors and there was about one stop difference between DSLRs and MFDBs in that test. That is the only wedge based comparison I have seen.
There is another observation regarding MFDBs.
Let's assume that we expose to the right, thus choosing to put highlights near saturation. Lets also assume that we shoot at say 100 ISO. Would an MFDB have 4 steps better DR than a DSLR it would need to achive the same performance at 1600 ISO as the DSLR at 100 ISO. This is clearly not the case. MFDBs do not perform well at high ISO. Don't know why. With correct exposure to the right DR would show up as latitude for underexposure.
So to sum up:
- I have not seen any theory explaining why MFDBs would have more than one stop advantage over DSLRs.
- DxO data seems to agree well with predictions based on theory
- Extended DR is in my view not consistent with bad high ISO performance on MFDBs.
That said, I have nothing against MFDBs. The only issue I have that I have only seen a single comparison between MFDBs and DSLRs based on Stouffer wedges and the difference was about one stop. So if the DR advantage with MFDBs really exists I would like to see it demonstrated in correctly made test shots, or have a physically correct explanation on the difference.
Best regards
Erik
Erik,
I'm not discrediting DXO because I don't like it. I'm pointing out a valid flaw in the way they approach MFDB. Your argument re:shooting the messenger is silly. Why do you defend them? I have a stoufer transmission step wedge and have used it to test DR with imatest software in the past. I know how it works so save yourself the trouble.
Bernard,
I didn't answer your question because I've had you set to ignore since I got tired of reading that stitching was the solution for everything. But your question was answered - DXO data for MFDB is not accurate for the reasons I've already pointed out. When I see
To all reading this point - the fuss about DR is because I'm hoping that any new sensors that come out at photokina will have more improvement in DR than in pixels. I'd rather see a bigger sensor preferably one with a more square ratio than a rectangle.
-
True: the observations about evidence of a strong market preference for greater image detail (more MP, etc.) were addressed to the medium format market. The lower pixel count of the D700 compared to similarly priced alternatives arguably helps it to offer better low light performance and a higher frame rate, but neither of these is of much of a selling point for lower pixel count MF backs.
How would we know? we still haven't got a single back with low pixel count made with a current pixel technology. I really think that a back shooting 2 frames/s would be a hit with the fashion crowd.
Edmund
-
How would we know? we still haven't got a single back with low pixel count made with a current pixel technology. I really think that a back shooting 2 frames/s would be a hit with the fashion crowd.
You're right it would be - and it is: The Phase One P40+ shoots 10 megapixel images at a sustained 1.8 frames per second (as opposed to most dSLRs that shoots 5-10 frames very quickly until it fills the buffer). At the flip of a switch it can also shoot 40 megapixel images at a sustained 1.2 frames per second. If you did a 10 megapixel sensor at that sensor size you would have huge moire issues, but since the P40+ 10 megapixel file is produced by a pre-raw pixel binning it has the very-low moire characteristics of a 6 micron pixel size.
Doug Peterson ()
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Cambo, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870 | Cell: 740.707.2183
Newsletter: Read Latest or Sign Up (http://www.captureintegration.com/our-company/newsletters/)
RSS Feed: Subscribe (http://www.captureintegration.com/2008/08/11/rss-feeds/)
Buy Capture One at 10% off (http://www.captureintegration.com/phase-one/buy-capture-one/)
Personal Work (http://www.doug-peterson.com/)
-
[quote name='dougpetersonci' date='Jul 9 2010, 03:45 PM' post='375292']
You're right it would be - and it is: The Phase One P40+ shoots 10 megapixel images at a sustained 1.8 frames per second (as opposed to most dSLRs that shoots 5-10 frames very quickly until it fills the buffer). At the flip of a switch it can also shoot 40 megapixel images at a sustained 1.2 frames per second. If you did a 10 megapixel sensor at that sensor size you would have huge moire issues, but since the P40+ 10 megapixel file is produced by a pre-raw pixel binning it has the very-low moire characteristics of a 6 micron pixel size.
[font="Arial"]Doug Peterson ()
__________________
A bit academic since there is no camera you can stick a P40+ on that will accurately focus fast on a moving subject (surely the main reason one needs a lot of frames per second). No, despite the best marketing efforts this Phase pixel binning thing is deeply, deeply uninteresting as competition for a Nikon or a Canon.
However, a 40 megapixel 6x7 back would be hugely useful, (significantly less demanding of lenses, engineering, focus and technique than its sub-645 equivalent and much easier to use on a view camera). I am absolutely sure of its popularity were it to be priced comparably to its smaller cousin.
(Better still a 160 megapixel 6x7 back with 40 megapixel pixel binning but we might have to pay a bit more for that...).
-
A bit academic since there is no camera you can stick a P40+ on that will accurately focus fast on a moving subject (surely the main reason one needs a lot of frames per second). No, despite the best marketing efforts this Phase pixel binning thing is deeply, deeply uninteresting as competition for a Nikon or a Canon.
However, a 40 megapixel 6x7 back would be hugely useful, (significantly less demanding of lenses, engineering, focus and technique than its sub-645 equivalent and much easier to use on a view camera). I am absolutely sure of its popularity were it to be priced comparably to its smaller cousin.
(Better still a 160 megapixel 6x7 back with 40 megapixel pixel binning but we might have to pay a bit more for that...).
I agree somewhat. But it depends on the movement of the subject - how much and how fast. Some fashion is not shot with a lot of continuous, dramatic movement. But I never really saw the Sensor Plus pixel binning as a competitor to Canon/Nikon as much as it provided an alternative. The vast majority of our customers who shoot medium format digital also shoot Canon/Nikon, depending on the situation (and we know what those situations usually are). The Sensor Plus technology allows these photographers to shoot medium format in more situations than they would otherwise. I see this as a significant benefit because if they shoot medium format, they clearly want to shoot medium format and often are restricted to the amount that they can, depending on the project. So, I don't see Sensor Plus providing the option of not using Canon/Nikon for most, but rather extending the times they would use medium format.
Steve Hendri
-
I agree somewhat. But it depends on the movement of the subject - how much and how fast. Some fashion is not shot with a lot of continuous, dramatic movement. But I never really saw the Sensor Plus pixel binning as a competitor to Canon/Nikon as much as it provided an alternative. The vast majority of our customers who shoot medium format digital also shoot Canon/Nikon, depending on the situation (and we know what those situations usually are). The Sensor Plus technology allows these photographers to shoot medium format in more situations than they would otherwise. I see this as a significant benefit because if they shoot medium format, they clearly want to shoot medium format and often are restricted to the amount that they can, depending on the project. So, I don't see Sensor Plus providing the option of not using Canon/Nikon for most, but rather extending the times they would use medium format.
Steve Hendri
Well said Steve and i am one of those folks although I push it into the 35mm world type stuff a lot more often than I thought i would but it REALLY does work great. The noise levels alone in the high ISO stuff is pretty darn impressive. I never go past ISO 1600 on anything so yes there are limits here. I did break down and bought a cheap canon for a upcoming gig but nothing to do with technology reasons more with security and damage I simply don't want to take the risk with my Phase gear and I just shot a few and processed them . Let me just say how much i really appreciate my MF gear. I actually have to work at processing with this thing. ROTFLMAO
-
I never really saw the Sensor Plus pixel binning as a competitor to Canon/Nikon as much as it provided an alternative.
Hi Steve - This is true enough today. But I have a feeling as N&C improve their per-pixel performance, the ability to bin pixels cleanly will figure more prominently in their strategies. For example, the D3x makes a pretty damn good low light camera that compares favorably with the D3 (but not the D3s) IF you downsample the 24.5mp files to 12mp judiciously.
In other words, I'm not putting bets on Nikon necessarily continuing its "low pixel/high sensitivity" versus "high pixel/low sensitivity" cameras strategy. Then again, the odds of them doing anything to alienate their Olympic shooting extravaganza photographers is next to zero, so they'd have to really deliver on performance.
But binning and sensor+ technology looks increasingly attractive to me these days. The cleaner the per pixel performance, the more a downsampled capture from a small-pixel camera can demonstrate the same benefits as captures from a large-pixel camera (where the sensor area is held constant).
There seems to me some reason to try to provide medium format cameras that are capable of handling fast action in low light, and they could produce through downsampling/sensor+/what-have-you results that would exceed the 35mm DSLRs even in low light situations. Give me an S2-sized body at max D3x prices, with a large sensor, fast capture streaming, CMOS, and able to switch roles depending upon what's asked of it. I'll be yours for life. And BTW, I've seen some models who can deliver 5 killer poses per second. [My pet theory is that the nerdy models deliver more cool facial expressions per second, and that theory is worth a laugh anyway.]
-
We provided an S2 for a photographer shooting hi-end magazine fashion editorial a few weeks ago. On the first day, he shot 160GB untethered (about 1900 shots with DNG+ hi-res JPG) in a five hour period and ...
David
One shot every 9.5 seconds for 5 hours nonstop. I am impressed by the photographer more than the camera.