Another mention of a 70+ megapixel EOS R camera

Jul 21, 2010
31,099
12,863
If I were making an AE system for a mirrorless, the sensor can see the image... so you can actually see how bright the brightest stuff is, and how dark the darkest stuff is. If you're getting 0's and max values, then by default try to equalize the number of pixels falling off each end.
The problem is that on current cameras (DSLRs in live view and MILCs), the live image, the histogram and the highlight warnings (‘blinkies’) are based on the jpg conversion algorithm. From the WYSIWYG standpoint, that makes sense – the live image in the VF/LCD should reflect your picture style, HTP, etc. But for judging exposure, it’s not that helpful because what appears blocked or blown on the jpg may be completely recoverable from the RAW file. It’s not a perfect solution, but Google ‘Canon uniWB’ for an interesting workaround.
 
  • Like
Reactions: 1 users
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
If I were making an AE system for a mirrorless, the sensor can see the image...

AE systems could absolutely be more sophisticated than they are. But as Neuro pointed out, the live data is generally (always?) based off a JPEG version. To do some of the clever things you mention would require judging exposure by the RAW.


A little off-subject but what about a special burst mode that stops shooting when it thinks the shot is good from the standpoint of camera motion, and only writes that to card? So even if a slow shutter only has a 10% hit rate, you don't need to take 10 photos then go back to lightroom and see if any of them are actually good. The camera just tells you that you got a good one, whether its the first or the 30th...

There's probably a smartphone that already does this. ILC's are behind smartphones when it comes to clever tricks like this.
 
Upvote 0
"A little off-subject but what about a special burst mode that stops shooting when it thinks the shot is good from the standpoint of camera motion, and only writes that to card? So even if a slow shutter only has a 10% hit rate, you don't need to take 10 photos then go back to lightroom and see if any of them are actually good. The camera just tells you that you got a good one, whether its the first or the 30th..."

There's probably a smartphone that already does this. ILC's are behind smartphones when it comes to clever tricks like this.

In burst mode, iPhones will automatically flag images with the best sharpness, although the others aren't automatically discarded (I'm sure other brands of phone do something similar).
 
Upvote 0
72d4ebf679662f93aba829d4445e30caab0f5e59797a28733f6db9102360ccf0.jpg

Nice clean shadows in the image, seems to be high DR, did you shoot it on Nikon or Sony?
 
Upvote 0
Apr 29, 2012
220
91
If you scroll around the image you can find spots where the 5Ds appears better, but you also quickly realize that at any spot it's splitting hairs with differences that could be due to very slight differences in other factors, even just the default color profiles.

The D810 noise is worse in the screenshot I posted, and that's made undeniably clear from scrolling around the image. Though again, the 5Ds @ +4 is not as good as the D810 @ +5, which is why I said between 5/6ev.

Show us. Show us a real world image you've shot where you pushed shadows +4.5ev, +2ev over my 7D shot, and the difference was huge. You've got a 5D4 which should be able to handle that so if DR is so important, so often, and so huge you should have landscape shots that illustrate it.

I'm tired of words from people on this topic. The lack of images very strongly suggest to me that the people who bring up DR on forums are not the people shooting wide DR scenes and producing wide DR prints. And that they therefore have no real idea of the relative impact, what it actually means to shoot two cameras 1ev apart, or 2ev apart, or 3ev part.

Again, you've posted nothing.

Worse, the response you got suggests that the DRones cant even tell when noise is better or worse!? That's probably a good point to stop responding and move on, your point has been made.

Thanks for your thoughtful and illuminating posts with actual data and images to support your claims. Its unfortunate that isnt how most of the internet works.
 
  • Like
Reactions: 2 users
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
Thanks for your thoughtful and illuminating posts with actual data and images to support your claims. Its unfortunate that isnt how most of the internet works.

Thank you for the compliment.

I think it's time for me to stop replying to the same points in the same circle, and to borrow my friend's D800E for a day for some comparison landscapes and perhaps an interior real estate shot. Images would illustrate very clearly just how huge or not huge the DR differences are in practical use. No one would have to trust my words, I could post everything to flickr and everyone could make up their own minds. (Note: I've done this casually before, but I would want to put more effort into the scenes and into getting the best out of each RAW with documented steps before making it public.)

My replies would be so easy at that point: "See this link." :LOL:
 
  • Like
Reactions: 1 users
Upvote 0
If you scroll around the image you can find spots where the 5Ds appears better, but you also quickly realize that at any spot it's splitting hairs with differences that could be due to very slight differences in other factors, even just the default color profiles.

Ok, I went to https://www.dpreview.com/reviews/canon-eos-5ds-sr/10 and added D810 to compare 5Dsr and D810 across the whole sample image. 5DSr lags more than 1 stop but less than 2 stops. 5DSr at +2ev is slightly better than D810 at +4ev but much worse than D810 at +3ev. Same for 5DSr at +4ev - it's slightly better than D810 at +6ev but much worse than D810 at +4ev. It's consistent with these measurements
http://www.photonstophotos.net/Charts/PDR.htm#Canon EOS 5DS R,Nikon D810 that give them 1.8 stop difference at base ISO.

And so you're saying the practical difference is nil? 1.8 stops is nothing?

Show us. Show us a real world image you've shot where you pushed shadows +4.5ev, +2ev over my 7D shot, and the difference was huge. You've got a 5D4 which should be able to handle that so if DR is so important, so often, and so huge you should have landscape shots that illustrate it.

But it's pointless unless I shoot both 5DSr and D810 side by side. Or 5DIV and 7D side by side. I don't know how a landscape shot from the 5DIV alone can illustrate anything about the 7D.
I'm happy enough with dpreview and photonstophotos where they did measurements in the same conditions.

Again, you've posted nothing.

It doesn't make the sample from your 7D more relevant. It's still irrelevant and proves nothing in the context of this conversation.
 
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
5DSr at +2ev is...much worse than D810 at +3ev.

183929


183930

5DSr at +4ev...much worse than D810 at +4ev.

Tempted to download the 5Dsr file and show this same clip but with a little NR.

183931

I don't know how a landscape shot from the 5DIV alone can illustrate anything about the 7D.

DR is so very important, and even 1ev of additional DR is so huge, that you can't shoot a scene to show the difference. :rolleyes:
 
Upvote 0
Mar 2, 2012
3,187
543
For the record, larger pixels should result in higher DR not because of read noise but because of well capacity. It's odd that this is not the case right now.

Here is a half-baked thought: maybe they aren’t actually saturating the sensor at native response (e.g., “base ISO”) before they hit the quantization limit (2^14 in most cases).
 
Upvote 0
And?.. In your sample 5DSr has more grain in the shadows next to the feathers in +2ev vs +3ev comparison.
+3 vs +4 and +4 vs +5 are even more indicative:

183933

The difference is quite striking and shows how quickly 5DSr images degrade when lifting the exposure - that means the noise was there from the beginning and gets amplified. D810 is much cleaner overall.

DR is so very important, and even 1ev of additional DR is so huge, that you can't shoot a scene to show the difference. :rolleyes:

To show the difference between what and what? I don't have 7D or 5DSr or A7rIII to compare them to my 5DIV. And for the purpose of this forum conversation, I wouldn't bother to go and borrow them from somewhere. A single image from 5DIV will not prove anything. At the same time the samples from dpreview are good enough.
 
Upvote 0
And?.. In your sample 5DSr has more grain in the shadows next to the feathers in +2ev vs +3ev comparison.
+3 vs +4 and +4 vs +5 are even more indicative:

View attachment 183933

The difference is quite striking and shows how quickly 5DSr images degrade when lifting the exposure - that means the noise was there from the beginning and gets amplified. D810 is much cleaner overall.

A typical example of a flawed comparison. Normalize the two samples to the exact same framing and size and make the comaprison. Pictures are not displayed in the same fashion as this comparison.
 
Upvote 0
Sep 3, 2014
305
10
It is worth mentioning that, in most cases, DR increases are due to slightly lower noise, not increases in well capacity. Therefore the gains are in extremely low levels. This is why people often push shadows in order to demonstrate it. Unfortunately, it’s a game of diminishing returns.

I posted this example a few years ago. It shows an extreme compression from a camera said to have wide DR (A7Rii); its highlights are pulled down and its shadows are pushed up, and then some. The result? Shadows look flat. They might not look noisy, but they have little detail by definition; those portions simply didn’t receive much exposure. To get good details there, multiple exposures would be required.
 

Attachments

  • 24B6489E-3694-4DA6-A456-7DBD75DF94E9.jpeg
    24B6489E-3694-4DA6-A456-7DBD75DF94E9.jpeg
    771.7 KB · Views: 209
  • 40BD4EFD-FF89-47C7-BD3F-D35304C677DE.jpeg
    40BD4EFD-FF89-47C7-BD3F-D35304C677DE.jpeg
    787.9 KB · Views: 217
  • 687B814E-8C8B-44DF-AFF2-B04C08F431A2.jpeg
    687B814E-8C8B-44DF-AFF2-B04C08F431A2.jpeg
    963.9 KB · Views: 207
  • Like
Reactions: 1 user
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
Here is a half-baked thought: maybe they aren’t actually saturating the sensor at native response (e.g., “base ISO”) before they hit the quantization limit (2^14 in most cases).

I've read that the ADC's aren't linear in which case that's probably not it. But I couldn't say for certain.
 
Upvote 0
A typical example of a flawed comparison. Normalize the two samples to the exact same framing and size and make the comaprison. Pictures are not displayed in the same fashion as this comparison.
You can do it, the results are the same. 5Ds gets slightly better but still lags behind D810.

Not when you're magnifying one more than the other.

As above, there's no magnification, there's 'comp' button which will downsample 5DSr image to match D810. The results are still not in favour of 5DSr.
 
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
Unfortunately, it’s a game of diminishing returns.

I posted this example a few years ago. It shows an extreme compression from a camera said to have wide DR (A7Rii); its highlights are pulled down and its shadows are pushed up, and then some. The result? Shadows look flat.

And that right there, ladies and gentlemen, is the difference between real world experience and graphs. I've also had shots where I blended two exposures even though one could handle the push in terms of noise because the tonality and fine detail was better with the blend.
 
  • Like
Reactions: 1 user
Upvote 0
Of course there is. When you pixel peep (100%; 1:1) higher resolution images are effectively magnified more.
But it's not a magnification. It's exactly what you said, a 1:1 view where 1 pixel from the image corresponds to 1 pixel on the screen. "Comp" option shows the 5DSr's image downsampled and D810's image as 1:1.
 
Upvote 0
So the EOS R was Canon's answer to the a7iii and we won't be seeing a REAL pro body? Just a budget turd and a 75 MP camera thats sure to cost over $4k that no one is asking for or wants? The 5Ds sold next to nothing. What makes Canon think a mirrorless version of it will sell any better? Where is the 30-35MP pro, mirrorless version of the 5DIV and the 20-25MP pro, mirrorless version of the 1DxII? Canon is a joke.
 
  • Like
Reactions: 1 user
Upvote 0

dtaylor

Canon 5Ds
Jul 26, 2011
1,805
1,433
But it's not a magnification. It's exactly what you said, a 1:1 view where 1 pixel from the image corresponds to 1 pixel on the screen. "Comp" option shows the 5DSr's image downsampled and D810's image as 1:1.

A typical FHD monitor resolution is 96 ppi. (4k is typically >200 ppi.)

8688 / 96 = 90.5"
7360 / 96 = 76"

Both cameras have a FF 35mm sensor, so...magnification.
 
  • Like
Reactions: 1 user
Upvote 0