50MP Canon 5DS/R spec leaked

[hjulenissen]

Canon have traditionally benefitted heavily as a company from their CMOS tech, in cameras, printers , scanners, copiers, and everything electronic. I'd be very surprised if they ditched the sensor part - it would be a major slap in the face for them, and de facto an indication that they have lost the ability to do R&D in their core business.

Edmund

Qualcomm has relied heavily on custom ARM designs in their products. Now they are (temporarily?) switching to a standard ARM implementation in their top-of-the-line product. That has got to hurt, but I guess it is a matter of getting the right set of features out at the right time (possibly being late to the 64-bit party):
http://www.anandtech.com/show/7925/qualcomms-snapdragon-808810-20nm-highend-64bit-socs-with-lte-category-67-support-in-2015

I guess there are many possible twists on Canon + Sony sharing patents, sharing technology, purchasing manufacture capability or simply purchasing ready-made sensors. It has been claimed that Canon has a couple of image sensor manufacturing plants (unlike everyone else) that use "stone-age" process technology, even for image sensors. Perhaps they see declining sales and need different process tech in order to satisfy DR/MP expectations of their customers? Perhaps Canon thinks that sensors are becoming a "commodity", while they can differentiate on things like ergonomy, flash control, lens selection, etc? Perhaps the image sensor industry is changing, and having a potent cellphone-camera-sensor division is necessary in order to do the R&D needed in larger sensors?

I think it is great if Canon makes a camera that is optimized for slow-moving still-photography with high resolution, high (?) DR at low ISO, great colors. It would be an interesting complimentary model to more generic "low-light + action-photo + 4k video" models. If that means dealing with Sony for sensor tech, then it just means that they are being pragmatic.

Or all of this could just be a rumor (even the cameras are not confirmed yet), based on the expectations of camera enthusiasts that make our a large portion of the internet posts about cameras, but a small portion of total camera sales...

-h

[BJL]

... olympus even makes a viewfinder that clips on top of the viewfinder.  

Think about it.  You could have an optical finder and a clip on evf in one camera.

Yes think about it: Olympus (accidentally) lets you attach an EVF on top of another EVF, yet no DSLR maker has bothered to offer the far more appealing EVF-on-OVF option!  Not even Canon, despite pushing hard the idea of using its DSLRs as video cameras, a usage that makes the OVF useless.

[Telecaster]

Yes think about it: Olympus (accidentally) lets you attach an EVF on top of another EVF, yet no DSLR maker has bothered to offer the far more appealing EVF-on-OVF option!  Not even Canon, despite pushing hard the idea of using its DSLRs as video cameras, a usage that makes the OVF useless.

I suspect this is neither accident nor oversight. It's also the main reason why I refuse to buy any camera with a "flipping mirror" in it.

-Dave-

[voidshatter]

+1.
Basically, not many camera systems out there which will give all of the above AND ISO 35 or ISO 50 when you want it.

ISO 35 on IQ280 and ISO 50 on IQ260 are fake (i.e. pull from ISO 100).

[voidshatter]

5. Because of volume,MF sensors are likely not to represent the last technology for long. The strength of medium format is the size of the sensor[2]. 48 x 36 mm is the point were size become a very strong advantage, enough to compensate for a generation delay on technology. 

Resolution-wise, yes. DR-wise, no, the 54x40mm IQ180/IQ280 did not even survive through the D800E.

[Steve Hendrix]

ISO 35 on IQ280 and ISO 50 on IQ260 are fake (i.e. pull from ISO 100).

It doesn't matter whether it is pulled (or not) or FAKE, as you say.

The results are better - so from that standpoint, as a user, I don't care how they do it, all I care about is does it give me a feature I can take advantage of.


Steve Hendrix
Capture Integration

[voidshatter]

It doesn't matter whether it is pulled (or not) or FAKE, as you say.

The results are better - so from that standpoint, as a user, I don't care how they do it, all I care about is does it give me a feature I can take advantage of.


Steve Hendrix
Capture Integration

I have hard evidence that at the same aperture and shutter speed, ISO 35 yields exactly the same information as ISO 100 does in the RAW file. You do NOT get better results than what you get from ISO 100. It's simply a meta data for marketing purposes (i.e. a hype). My tests are base on an Alpa camera so there is no proof of aperture used. You may upload any RAW files taken with the 645DF+ body (so that the EXIF records the aperture), and I could show you how ISO 35 performs exactly the same as ISO 100 does.

[ErikKaffehr]

Hi,

I don't know about the modern backs but DxO-mmrk has measured the IQ-180, and it ISO sensivity was ISO 29 at all settings up to ISO 100.

The general tendency is that MFDBs tend to underexpose, thus getting some extra margin for protecting highlights.

The enclosed screen dumps from DxO mark indicates the trends clearly. Note that Pentax 645D ISO-s were dead accurate. It is obvious that implementation of ISO-s varies wildly between different back. Some is done by preamping, some by bit shifting and some with just tagging raw files?

Best regards
Erik

I have hard evidence that at the same aperture and shutter speed, ISO 35 yields exactly the same information as ISO 100 does in the RAW file. You do NOT get better results than what you get from ISO 100. It's simply a meta data for marketing purposes (i.e. a hype). My tests are base on an Alpa camera so there is no proof of aperture used. You may upload any RAW files taken with the 645DF+ body (so that the EXIF records the aperture), and I could show you how ISO 35 performs exactly the same as ISO 100 does.

[voidshatter]

Hi,

I don't know about the modern backs but DxO-mmrk has measured the IQ-180, and it ISO sensivity was ISO 29 at all settings up to ISO 100.

The general tendency is that MFDBs tend to underexpose, thus getting some extra margin for protecting highlights.

The enclosed screen dumps from DxO mark indicates the trends clearly. Note that Pentax 645D ISO-s were dead accurate. It is obvious that implementation of ISO-s varies wildly between different back. Some is done by preamping, some by bit shifting and some with just tagging raw files?

Best regards
Erik

Here is an illustration of native ISO and extended ISO. For the same aperture and shutter speed, you do not get different image quality if the measured ISO stays the same. You can even verify this with Raw Digger (check the same spot - do they share the same levels?). It doesn't matter how far the blue line deviates from the red line. Some cameras do not offer the orange line. Some cameras do not offer the green line. Some cameras get the blue line closer to the red line. These do not affect the fact that only the blue line is the native ISO range.

The P45+ has two native ISO settings: 50 and 100. The IQ180 has four native ISO settings: 100, 200, 400 and 800 (or including 1600 and 3200 if you count sensor+). Shooting an IQ280 at ISO 35 is just like shooting a D810 at ISO 32, self-deception.

[Steve Hendrix]

I have hard evidence that at the same aperture and shutter speed, ISO 35 yields exactly the same information as ISO 100 does in the RAW file. You do NOT get better results than what you get from ISO 100. It's simply a meta data for marketing purposes (i.e. a hype). My tests are base on an Alpa camera so there is no proof of aperture used. You may upload any RAW files taken with the 645DF+ body (so that the EXIF records the aperture), and I could show you how ISO 35 performs exactly the same as ISO 100 does.

You do not get better results? I don't think what you are saying is accurate.

https://captureintegration.com/best-long-exposure-results-with-phase-one-leaf/

And why do you use terms like "hard evidence"? Are you going to be on 60 minutes?


Steve Hendrix
Capture Integration

[voidshatter]

You do not get better results? I don't think what you are saying is accurate.

https://captureintegration.com/best-long-exposure-results-with-phase-one-leaf/

And why do you use terms like "hard evidence"? Are you going to be on 60 minutes?


Steve Hendrix
Capture Integration

As I said, if you upload 2 raw files of the same aperture and at the same shutter speed but at different ISO settings in the extended ISO range (in this P65+ case, 20 seconds for ISO 50, and 20 seconds for ISO 100), I will be able to show you these are of the same image quality.

"Hard evidence" means levels (i.e. the data) in the raw files (e.g. amount of highlight details retained), as well as measured SNR in the shadow.

To elaborate for the P65+ case:

ISO 50 + 20s = ISO 100 + 20s (because for each pixel there is virtually no difference of level number between the two raw files)
ISO 100 + 20s > ISO 200 + 20s (because the latter would lose one stop of highlight details)
ISO 100 + 20s > ISO 200 + 10s (because even though they retain the same amount of highlight details, the latter has a noisier shadow)

[jduncan]

I think you missed my point: of course sensors bigger than 33x26mm exist, but they all need the more expensive, lower-yield process of on-wafer stitching (etching part of the sensor, then moving the wafer to a new position to etch another part of it, moving with sub-micron position accuracy, and so on).

Two is still be stitching, but anyway, note my use of the word "suitable" when I mentioned the "33x26mm limit of all suitable fab. equipment."
I know of only two models of stepper from any maker with a field size bigger than 33x24mm:
- The discontinued Canon FPA-3000 iW with 50x50mm field size and rather large 0.8 micron minimum feature size: http://www.ventexcorp.com/downloads/FPA-3000iW+Specification.pdf
- The newer FPA-5510 iV with 52x34mm field size, but an even coarser 1.5 micron minimum feature size: http://www.usa.canon.com/cusa/semiconductor/products/semiconductor_equipment/steppers/fpa_5510iv_stepper#Specifications

As far as I can tell, even the smaller 0.8 micron minimum feature size of the FPA-3000 iW mean is unsuitable for making the pixel sizes that the photographic market uses (putting aside special lower resolution uses like X-rays, machine vision, big telescopes and so on.) Canon itself has explained in several white-papers that its needs to use stitching to make its 36x24mm sensors, and that the "APS-H" CMOS sensors of some 1D models was the largest size that it could make without stitching, even though that FPA-3000 iW already existed when Canon made its first 36x24mm sensors.

These steppers might be used to fab. some large, lower resolution sensor for X-rays, machine vision, and such.  For example the KAF-4320 (now sold by Onsemi) with its 24 micron pixels and a 50x50mm size that fits the field size of the FPA-3000 iW perfectly: http://www.onsemi.com/pub_link/Collateral/KAF-4320-D.PDF

Hi,

No, I did not mis your point, what I miss was your reply until today.  I agree with you that it's more complex, and we not even mention yields.  I was extending your observation and presenting a set of possibilities for bigger sensors, within current technology.  I know, also the double exposure is just an other way, less precise to talk about stitching. I used it to avoid using the same word too many times.

The point about the example is to illustrate that even with low volume it is possible to create the product.  Maybe Sony did not make it bigger due to some business reason (managing risks,  the needs of a high volumen client etc)

Best regards,

[voidshatter]

2. I am saying that that it will be remove one justification, if the Dynamic range is there. The logic is pretty straightforward: Can people justify MF? of course they can[1]. but they will have one justification less.

7. If it's real and has the proper dynamic range the sensor on the Canon will be about 8660 x 5770 pixels vs 8280 x 6208 for the IQ250. Most people don't shoot Otus lenses, so the IQ should prevail. As I say is one justification less.

dpreview writes the following:

As far as dynamic range is concerned, we're told that the new 5DS and 5DS R should give the same performance as the current EOS 5D Mark III. If true, this means that the new cameras won't be able to offer the same industry-leading dynamic range of Sony's current APS-C and full-frame sensors, but at least it isn't a step backwards. And hey - 50MP!

http://www.dpreview.com/previews/canon-eos-5ds-sr/5

[torger]

The dynamic range thing is going to be massively disappointing for forum folks, but if you shoot studio stuff you don't really need that much, so maybe it will be a hit anyway. I think it's a bit boring that Canon is not able to compete on this popular metric though... but MFD salesmen will be happy, they can show that a digital back from 2004 outperforms a high MP DSLR from 2015(!).

Still a bit of tiny hope that DR is better than announced, we just need base ISO raw file to see!

[gerald.d]

Here is an illustration of native ISO and extended ISO. For the same aperture and shutter speed, you do not get different image quality if the measured ISO stays the same. You can even verify this with Raw Digger (check the same spot - do they share the same levels?). It doesn't matter how far the blue line deviates from the red line. Some cameras do not offer the orange line. Some cameras do not offer the green line. Some cameras get the blue line closer to the red line. These do not affect the fact that only the blue line is the native ISO range.

The P45+ has two native ISO settings: 50 and 100. The IQ180 has four native ISO settings: 100, 200, 400 and 800 (or including 1600 and 3200 if you count sensor+). Shooting an IQ280 at ISO 35 is just like shooting a D810 at ISO 32, self-deception.

Which way round is this though?

Is ISO100 shot at ISO35 (29), or is ISO35 shot at ISO100?

Erik mentions that DXO report "the IQ-180... ISO sensivity was ISO 29 at all settings up to ISO 100.", so presumably it's the former.

If the sensitivity of the back has (somehow - not really interested in the technical detail of it) been measured at ISO29, then surely it is absolutely correct to say that it has an ISO sensitivity of 29, and not 100?

Kind regards,


Gerald.

[synn]

When I heard about this launch, I cam here expecting this exact chain of posts.
Wasn't disappointed.

[pedro39photo]

I love the look of DMF, its not about resolution its about the aesthetics of the format and the big viewfinder.
I make my run in this medium - mamiya zd - H3D 22MP - H3DII 39MP but i feel that in the last 5-6 years just pentax was bold to bring something new to this medium, and with the "right" prices and best body technology.
Now the problem its not about resolution its about motion ! a saw in the last 2 years a swift in my clients to the motion-movies.
My industrial clients now don't have large prints on there fair booths they have LCDs with movies !!!
I go to the shopping mall today and the fashion shops don't have beautiful large prints ! they have large vertical LCDs with fashion videos!!!
My next buy sad to say, will be a Canon C100, and today i spend more time on cinematographer discussion forums and less on medium format.
There will hallways be room for big resolution photography, but sad in the corporate business i see a swift for more video, and this its a small kill for dmf, like the d800 was, and the 5dsr will be
Nice weekend
PN

[yaya]

I have hard evidence that at the same aperture and shutter speed, ISO 35 yields exactly the same information as ISO 100 does in the RAW file. You do NOT get better results than what you get from ISO 100. It's simply a meta data for marketing purposes (i.e. a hype). My tests are base on an Alpa camera so there is no proof of aperture used. You may upload any RAW files taken with the 645DF+ body (so that the EXIF records the aperture), and I could show you how ISO 35 performs exactly the same as ISO 100 does.

Was that a "One Mississippi, Two Mississippi" kind of hard evidence on the 2 seconds exposures? The back cannot tell the exact exposure time on a mechanical shutter so you could have anything up to ~1s difference between those two frames and that also counts for the dark calibration frame...so hardly a "hard evidence..."

BR

Yair

[voidshatter]

Was that a "One Mississippi, Two Mississippi" kind of hard evidence on the 2 seconds exposures? The back cannot tell the exact exposure time on a mechanical shutter so you could have anything up to ~1s difference between those two frames and that also counts for the dark calibration frame...so hardly a "hard evidence..."

BR

Yair

I just found other pictures but you could always question the mechanical shutter thing. As I said, upload your RAW files (same aperture, same shutter speed, different ISO in the said extended ISO range) and I will show you 

[voidshatter]

Which way round is this though?

Is ISO100 shot at ISO35 (29), or is ISO35 shot at ISO100?

Erik mentions that DXO report "the IQ-180... ISO sensivity was ISO 29 at all settings up to ISO 100.", so presumably it's the former.

If the sensitivity of the back has (somehow - not really interested in the technical detail of it) been measured at ISO29, then surely it is absolutely correct to say that it has an ISO sensitivity of 29, and not 100?

Kind regards,


Gerald.

There tend to be a difference between the manufacturer ISO and the measured ISO.

ISO 35 setting (manufacturer ISO 35) on the IQ180 is shooting at manufacturer ISO 100 but the measured ISO is 29.
ISO 100 setting (manufacturer ISO 100) on the IQ180 is shooting at manufacturer ISO 100 but the measured ISO is 29.
ISO 200 setting (manufacturer ISO 200) on the IQ180 is shooting at manufacturer ISO 200 but the measured ISO is 60.
...

ISO 100 is the turning point, and is the true lowest native manufacturer ISO.