Canon EOS 3D X [CR1]

[Aglet]

But in fact if you play around with it & pixel peep I think you'll find that up to about ISO800 or ISO1600 (depending on the camera of course, for modern Canons it's usually around 800 for crop & 1600 for FF) it's actually beneficial to do what you're saying i.e. boost the ISO to get better sensitivity for ETTR.  Until the camera starts doing more linear amplification (which causes DR to drop above those ISOs), capturing more photons is better.

Unfortunately, Canon's performance curves are "normal" when it comes to signal to noise ratio from base ISO up. SNR decreases proportionally with increase in ISO.
So ETTR at mid ISO doesn't have much of a loss of DR but SNR drops noticeably.  Still, an effective method.. of sorts.  I'd prefer an uncompromised camera.

[RLPhoto]

But in fact if you play around with it & pixel peep I think you'll find that up to about ISO800 or ISO1600 (depending on the camera of course, for modern Canons it's usually around 800 for crop & 1600 for FF) it's actually beneficial to do what you're saying i.e. boost the ISO to get better sensitivity for ETTR.  Until the camera starts doing more linear amplification (which causes DR to drop above those ISOs), capturing more photons is better.

Unfortunately, Canon's performance curves are "normal" when it comes to signal to noise ratio from base ISO up. SNR decreases proportionally with increase in ISO.
So ETTR at mid ISO doesn't have much of a loss of DR but SNR drops noticeably.  Still, an effective method.. of sorts.  I'd prefer an uncompromised camera.

aglet, I've been reading your posts a bit now. Can you post some of your D800 Photos here?

[Skulker]

But in fact if you play around with it & pixel peep I think you'll find that up to about ISO800 or ISO1600 (depending on the camera of course, for modern Canons it's usually around 800 for crop & 1600 for FF) it's actually beneficial to do what you're saying i.e. boost the ISO to get better sensitivity for ETTR.  Until the camera starts doing more linear amplification (which causes DR to drop above those ISOs), capturing more photons is better.

Unfortunately, Canon's performance curves are "normal" when it comes to signal to noise ratio from base ISO up. SNR decreases proportionally with increase in ISO.
So ETTR at mid ISO doesn't have much of a loss of DR but SNR drops noticeably.  Still, an effective method.. of sorts.  I'd prefer an uncompromised camera.

If by normal you mean about the same as Nikon I would agree with you. (why do I get the sneaky feeling you may not agree with me  )

http://www.techradar.com/reviews/cameras-and-camcorders/cameras/digital-slrs-hybrids/canon-1dx-1091200/review/page:5#articleContent

I expect some people will "pixel peep" at the graphs and say "mines better at XYZ" but I think any reasonable person would say those are all great camera's

[art_d]

And the idea that few people understand all the tech behind it, I agree. I am one of them, trying to learn more. But I'm definitely not convinced that the progression to 16 bit is a bad idea. It needs to be implemented properly, and I wouldn't be surprised if it became something we all understood a bit better in the coming year or two...

Going to 16 bit wouldn't hurt anything in the imaging process. It just isn't going to help anything. It's not going to  record better color transitions, just extra random noise. All it would do is make RAW files larger, because the files would now contain extra information on the random noise which previously was not being recorded. Larger files without any actual benefit to images is not something we should be asking for.

Why are you so sure it will record just extra noise but no more useful information?

People have studied this issue. See the following technical explanation related to noise and bit depth: http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/noise-p3.html

Also, see this rather lengthy discussion: http://www.luminous-landscape.com/forum/index.php?topic=60672.0

In short, represntatives from a medium format digital back dealer essentially concede that when the term "16 bit" is applied to MF sensors, it is done so as marketing shorthand in order to convey to potential customers that the MF sensor will have better tonal qualities compared to DSLRs of equal megapixels, even though the actual reasons for those better tonal qualities lie elsewhere. This may be fine for MFDB buyers since they are not being misled and their camers do produce better tonal qualities. However the problem which has resulted from this seemingly innocent bit of marketing is that some people have been led to think that if you make DSLR sensors 16 bit they will produce the tonal qualities of MF.

I've read this article from Chicago some time ago and thanks - I've read it again :-) It's hard not to agree with it's contents written by prof Martinec (the more that I've got just a master and engineer degrees in computer science, not a professor but let me point to some circumstances:

1. Examples showing no difference between the original image and image with clipped 2 bits don't make sense in this discussion - there it was to indicate no difference on screen while viewing, we are talking about the useful information used later for image manipulation. If we intend to get one picture 14 bits deep and another 16 bits  deep, convert them directly to 8bits jpegs and display on a screen, then most probably we won't see too much difference, I'd say - no difference. But if you'd like to manipulate it in PS, then depending on how much you want to manipulate, you'll see the difference sooner or later.

2. The long part of the article regarding noise is based on real values measured in real devices like 1d3 or 40d and compared to other devices. The read noise in sensor plays an important part. How about changing it a little in next generation of sensors? I mean - what if? What if in a new sensor some other technology would be used? Let's assume they would find a method to read each pixel's value not once after exposing it to light but could introduce sampling with frequency let's say 1MHz, which could eliminate some read noise and improve DR? I'm not saying that such sampling would help but I indicate, that some conclusions might not be same true in such a new type of sensor. So what if those additional bits were not just to record more noise?

3. I don't think that 16bit RAWs would make my photos any better than 14bit RAWs because in most cases I wouldn't know what to do with this. At the same time I think that guys at Adobe could know :-)

4. I think that everyone here has heard that there was a world market for no more than 5 computers and that 640KB of computer memory was enough :-) So why not 32 bit RAWs?

BTW: In the era of 386s@40Mhz and 486s@50Mhz if I would say to my professor that in 20 years there will be processors working at 2.4GHz and a graphics memory would utilize 7GHz clock it would be the best joke he'd hear that week. In one cycle of 2Ghz clock light (or other electromagnetic wave) travels like 15cm in vacuum...

I think the thing to keep in mind is that as far as we know, right now there are no cameras used in traditional photography that in any practical way actually make use of the full 16 bits (I believe there are scientific cameras that do, but that's a different story!). The belief that simply introducing 16 bit parts into a DSLR would make it perform like a MFDB is just not true.

Now, if Canon is on the verge of a technological breakthrough where they've actually managed to actually fill a full 16 bits with useful information, I think such a camera would quite literally mean the demise of the medium format digital back market. But, I remain sketpical of that

[Aglet]

If by normal you mean about the same as Nikon I would agree with you. (why do I get the sneaky feeling you may not agree with me  )

I have to agree, from what I see published on some web sites testing these cameras, the SNR of all the top line cameras is pretty similar across their ISO range.

In an apparently stunning turn, the D800's SNR is even lower than that of the 5D2.
If you compare the curves on DxOmark's website tho, the results are different than the techradar site supposedly using DxO testing software. Big difference in DR results.  So who's doing it properly?  Neither completely explains their methodology to anyone's satisfaction.

That same article sparked some lively debate before about the merit and validity of DxO's testing.
It only tells me that anyone can buy DxO's testing software and rig but do they really apply it properly?

I'll take DxO's testing on their site, thank you.  They build the stuff, they otta know how to use it properly and consistently. There the Canon's have an edge in SNR, tho I still don't know how they're testing it, the specifics.

But I'll take the worse reported SNR of the D800 over the low ISO noise of any Canon product to date because it's less visible. MUCH less visible.  And any bloke with access to both cameras can determine that quite readily with little more than Photoshop and ACR (or Lightroom) and a lenscap. 

[Aglet]

aglet, I've been reading your posts a bit now. Can you post some of your D800 Photos here?

I was saving some good examples to post on my web site but, considering how laggard I've been updating it this summer, i could maybe post a little something here.
Altho, really, the most impressive is the inside of a black lens cap, shot at 1/200s, f/16, in a dark room and pushed 4 stops in post. 

Anything specific you'd like to see? I haven't been able to get out and shoot much with it, despite intents.

I've got an ISO 3200 close-up of and ant I took when I first got it and was playing around with it in the back yard.
It's kind of impressive considering very little NR was used.

I took a few colorful, post-storm, sunset-silhouette shots at Lesser Slave Lake AB last month.  They were done with a D800 and ancient 20mm lens at ISO 400 but that doesn't really showcase its abilities. Even Canon bodies can shoot great sunsets and I proceeded to shoot dozens more of them with my 60D after taking a few shots with the 800, all of them more impressive than those I shot with the D800.

I DO have a landscape of sorts tho, where extreme shadow pushing in post yields a nice HDR type effect covering a Dynamic Range from sunlit cumulus clouds to the texture of charcoal in shade.  With ZERO NR applied and no noise visible in the pushed shadows.  Similar shot from my 60D is already on this site in the HDR gallery, near the bottom of page 2 -
www.canonrumors.com/forum/index.php?topic=8065.15

The other one with the charcoal is a bit more extreme.  I intend to prep that one for a poster.

2 images posted in response to RLPhoto removed August 26th.

[funkboy]

Would love it if it comes out in october

unlikely in a Potatokina year...

[funkboy]

I'd prefer an uncompromised camera.

I'd really prefer if ANY camera company added a metering mode that figured out how to get the maximum quantity of photons into eachphotosite while avoiding color channel blowout for us so we didn't have to do it manually by screwing around with histograms & tricking the camera into making it happen.

Regarding SNR, to quote Reichmann:

Why?

Well, there is the story of Willy Sutton the famous American bank robber. When he was finally arrested, he was ask, "Willy, why do you rob banks?" Willy answered, "Because that's where they keep the money."

The reason why we want to expose every shot that we take with the data as far to the right of the histogram as possible is because that's where the data is! It also is where the visible noise isn't. The visible noise is lurking in the darker stops.

There is actually more noise in the brighter stops, but because there is such a high signal level any noise is rendered invisible because of the superior S/N ratio.

A colleague measured his Canon 5D MKII and reported the following...

"My 5D Mark II has a noise level of ~70 units at its maximum highlight level of 16,383 (on a 14-bit scale), and a noise level of ~30 units at a much darker signal level of 16 (i.e., 10 stops darker). The highlights appear clean because the SNR is good (16,383 vs 70). The shadows appear gross because the SNR is dismal (16 vs 30) – in fact, the signal is buried in the noise."

[RLPhoto]

aglet, I've been reading your posts a bit now. Can you post some of your D800 Photos here?

I was saving some good examples to post on my web site but, considering how laggard I've been updating it this summer, i could maybe post a little something here.
Altho, really, the most impressive is the inside of a black lens cap, shot at 1/200s, f/16, in a dark room and pushed 4 stops in post. 

Anything specific you'd like to see? I haven't been able to get out and shoot much with it, despite intents.

I've got an ISO 3200 close-up of and ant I took when I first got it and was playing around with it in the back yard.
It's kind of impressive considering very little NR was used.

I took a few colorful, post-storm, sunset-silhouette shots at Lesser Slave Lake AB last month.  They were done with a D800 and ancient 20mm lens at ISO 400 but that doesn't really showcase its abilities. Even Canon bodies can shoot great sunsets and I proceeded to shoot dozens more of them with my 60D after taking a few shots with the 800, all of them more impressive than those I shot with the D800.

I DO have a landscape of sorts tho, where extreme shadow pushing in post yields a nice HDR type effect covering a Dynamic Range from sunlit cumulus clouds to the texture of charcoal in shade.  With ZERO NR applied and no noise visible in the pushed shadows.  Similar shot from my 60D is already on this site in the HDR gallery, near the bottom of page 2 -
www.canonrumors.com/forum/index.php?topic=8065.15

The other one with the charcoal is a bit more extreme.  I intend to prep that one for a poster.

but then, here's another sunset shot, pushed well in post....
along with the original jpg from camera, reduced.

While these are interesting, They don't showcase anything that couldn't be done on a d30.

[Aglet]

I'd really prefer if ANY camera company added a metering mode that figured out how to get the maximum quantity of photons into eachphotosite while avoiding color channel blowout..

Canon's color matrix metering, starting in the 7D I think, I find works much better than the previous metering systems when it comes to tricky compositions with lots of color.  I find I don't often have to compensate at all, many times, and when I do, it's considerably less than I had to with the previous, non-color reading metering systems.  Force of habit often has me adjusting -2/3 EV which I end up putting back in as +2/3 in post.

Having an ETTR until one color channel starts to clip option in the metering menu would be a nice touch. I requested this from the Magic Lantern guys but don't know if it ever showed up.

[Skulker]

If by normal you mean about the same as Nikon I would agree with you. (why do I get the sneaky feeling you may not agree with me  )

I have to agree, from what I see published on some web sites testing these cameras, the SNR of all the top line cameras is pretty similar across their ISO range.

Well I didn't expect that!!!! 

But I'll take the worse reported SNR of the D800 over the low ISO noise of any Canon product to date because it's less visible. MUCH less visible.  And any bloke with access to both cameras can determine that quite readily with little more than Photoshop and ACR (or Lightroom) and a lenscap. 

Ahh that's more like it!      Of course I could be countering with pointing out the 1Dx noise and DR at mid ISO's advantages, where I take most of my wildlife shots.

No offence intended  - From my point of view I think we are all really lucky that current cameras are so good! I did consider a change to Nikon for the low light performance and the 200-400 lens. I'm please I stuck with Canon as they have now caught up on the low light issue. I would love the new Canon lens with the built in converted, realistically I'm not going to spend that much.

Happy shooting.

[V8Beast]

Meh. Unless this supposed 3D packs at least 47 megapixels, Canon shouldn't even bother with a high MP body.

[Aglet]

While these are interesting, They don't showcase anything that couldn't be done on a d30.

HA!
Nice try.

That's only +2 EV and 63% fill light on Adobe ACR 6.7
Either of those settings alone can show up banding noise in dark areas in some low ISO canon raw files. (I'm discovering there's sometimes more to it than that tho, can vary with Canon shutter speed)
Addenda: I can actually drag fill light to 100 to start to blow out the sand in the LR corner and it only shows a little chroma noise that could be removed with basic NR and still show plenty of detail.

I'd LOVE it if my 5D2 came out this clean at those settings.

check my tech blog for the dark noise comparison if you want to see impressive lens caps shots.

2 samples at ISO 3200 from D800 removed August 26th.

[Timothy_Bruce]

I think it would be named 1D S
There is the 1D X and 1D C  and then the 1D S

[marekjoz]

I think it would be named 1D S
There is the 1D X and 1D C  and then the 1D S

Luckily there is no 1D S (nor 1Ds) yet, so there will be no confusion :-)