Reading mode: Light Dark

question on ISO settings for the 50D

Go to last post
I have no problem with you posting a real world image nor do I mind that you did not include any additional information about noise at ISO 1600 on the 50D.

I do have a problem with you complaining when someone else tries to be helpful and takes the time to provide more information. That's what these forums are for. Your second message was unreasonable and if you weren't being so defensive you'd just admit it to yourself and move on. It's really no big deal; we've all clicked "post" at some time or other when we should have given more thought to our message first, and we've all caught a bit of flack when we've done it.

Having said that, I'm now done with this topic. This is getting too far from what the OP was asking about.

ppage
 
clknight said:
I could post images that show great ISO 1600 performance all day, and you could continue to discount them as being variable and illusive.
Click to expand...
But others could post messages all day showing lousy performance at ISO 1600 and I'm sure you would discount then as being variable and illusive.

Going beyond one or two sample images and including some discussion about noise is not only relevant, it's necessary.

ppage
 
John Sheehy said:
No. The XTi has relatively poor quantum efficiency compared to current cameras, collecting about 65% as many photons with the same real exposure. It has about 4x as much read noise at high ISOs as the 7D, with the same real exposure.
Click to expand...
No?

The theory is nice, but DXOMark's SNR tests beg to differ. At the pixel level, they give a tiny advantage to the 400D. At the image level, maybe a third of a stop to the 50D.

I consider a third of a stop less noise not to be particularly significant.

I'll be curious to see if the 7D has improved things further, but until actual production samples are in reviewers hands, I'm not making any evaluation of it one way or the other.
 
clknight said:
John Sheehy said:
They are examples of photography. There are usually too many variables for them to mean anything as far as camera qualities matter, at the level at which they actually vary.
Click to expand...
contains all of the variables. So why is it irrelevant?
Click to expand...
Because you can't quantify them, and you can hide the differences.
clknight said:
The only way to give the OP a cut and dry answer is if all he wanted to do was shoot test shots in the same conditions all the time. Some people seem to do that exclusively here, but I don't believe that's his intention.
Click to expand...
Then he can't really have what he wants. It is no challenge to produce a small, high-key ISO 1600 image without distracting noise.
clknight said:
I could post images that show great ISO 1600 performance all day, and you could continue to discount them as being variable and illusive. I guess to make it valid there should be "DXO" in the link, or have some subjective SNR results.
Click to expand...
I'm not intending to pick on you personally, you were just the person doing this when I felt it was time to bring up the issue again.

DxO is not perfect or complete, but it is a move in the right direction.

--
John

 
HotspurJr said:
HotspurJr said:
No. The XTi has relatively poor quantum efficiency compared to current cameras, collecting about 65% as many photons with the same real exposure. It has about 4x as much read noise at high ISOs as the 7D, with the same real exposure.
Click to expand...
No?

The theory is nice, but DXOMark's SNR tests beg to differ. At the pixel level, they give a tiny advantage to the 400D. At the image level, maybe a third of a stop to the 50D.
Click to expand...
This is where DxO is blind; 18% SNR is hardly affected at all by read noise. It is affected, statistically, mostly by shotnoise (although read noise is more visible than it is statistic). The linear trends on the log/log graph should make it clear that read noise is not affecting the statistics very much.

If you go to the DR charts, then they are also non-indicative of read noise in an absolute sense; only relative to saturation, and the 400D is only about 65% as sensitive as the 50D.

DxO never gives straight-out read noise relative to absolute signal. 0.5% SNR would be far more useful for that than 18% SNR (assuming external metering, and not trusting the camera).
HotspurJr said:
I consider a third of a stop less noise not to be particularly significant.
Click to expand...
That depends on where the two values lies relative to the threshold of visibility.
HotspurJr said:
I'll be curious to see if the 7D has improved things further, but until actual production samples are in reviewers hands, I'm not making any evaluation of it one way or the other.
Click to expand...
--
John

 
John Sheehy said:
This is where DxO is blind; 18% SNR is hardly affected at all by read noise. It is affected, statistically, mostly by shotnoise (although read noise is more visible than it is statistic). The linear trends on the log/log graph should make it clear that read noise is not affecting the statistics very much.
Click to expand...
Here's the problem:

When somebody asks - like the original poster did - if he's going to get noise improvements by upgrading cameras, he really doesn't care about the source of the noise. Read noise, shot noise, whatever. It really doesn't matter very much. Noise is, to most people in most applications, noise.

The only thing that matters is how clean the image is.

So you may well be perfectly right in theory when it comes to technological improvements and all that jazz. But it's pretty much irrelevant when it comes to the question:

"If I buy a new camera, will my high ISO results improve?"
 
Shoot the 50D at 1250 iso and don't underexpose...raw..it's pretty good.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top