40d noise!

Bukka wrote in post #4035212
Shot with an 50mm 1.8 (nifty fifty)
This is gonna be my halloween costume haha.
Shot at iso 3200 with aperature of 2.0
slight unsharp mask applied.
No noise reduction.

Do you have a point to this post? Such extreme contrast from top of head to dark shadows under the eyes lead to noise at nearly any iso.

Plus you used iso3200 for the sake of using iso3200 since your shutter speed was 1/1000. Using iso3200 when you could have gotten the shot with the same appeture at a shorter shutter speed will make high iso noise look better. You could have gotten that shot also at 1/125 and iso400 which I suggest you use next time.

RedHot wrote in post #4051652
Do you have a point to this post? Such extreme contrast from top of head to dark shadows under the eyes lead to noise at nearly any iso.

Plus you used iso3200 for the sake of using iso3200 since your shutter speed was 1/1000. Using iso3200 when you could have gotten the shot with the same appeture at a shorter shutter speed will make high iso noise look better. You could have gotten that shot also at 1/125 and iso400 which I suggest you use next time.

I believe their point was just to show how clean the image looked even given the use of 3200. Of course, as SilverOneMi has proved there appears to be something else going on.

I dunno though, any time you use Ken Rockwell's name you seem to be asking for trouble in these parts.

RedHot wrote in post #4051652
Do you have a point to this post? Such extreme contrast from top of head to dark shadows under the eyes lead to noise at nearly any iso.

Plus you used iso3200 for the sake of using iso3200 since your shutter speed was 1/1000. Using iso3200 when you could have gotten the shot with the same appeture at a shorter shutter speed will make high iso noise look better. You could have gotten that shot also at 1/125 and iso400 which I suggest you use next time.

So... You are saying that because of the contrast, the noise will be worse, but because of the fast shutter it will look better. Which side of the fence are you on again?

Goes to take out his 40D and snap a couple pics...

Here's a quick test I did. For each setting I took 3 shots, then used the second one just to keep it kinda random.

HTTP response: NOT FOUND | MIME changed to 'image/png'

braduardo wrote in post #4052213
So... You are saying that because of the contrast, the noise will be worse, but because of the fast shutter it will look better. Which side of the fence are you on again?

I am dealing with no fence. Both my statements are valid. High contrast, also known as broad dynamic range, will yield noiser shadows if the dynamic range of the image is greater than the dynamic range of the sensor and the highlights were exposed for.

But also if you use iso3200 "prematurely", i.e. using it to get a faster shutter speed than regular use or for the sake of using iso3200 in good light and brighter, you will get lower noise than if you needed iso3200 becase available light was low and your shutter speed is slow.

SilverOnemi wrote in post #4051223
i got iso 1600 samples with chroma noise, and im expecting even more with iso 3200 samples, but since you guys seem not to believe me...

It's nothing to do with not believing, it's about not agreeing.

Those Ken Rockwell shots are so ridiculously over-saturated that they're bound to bring out colour that would simply not be there in any normal use of the camera.

Well, I posted mine up there, but they don't seem to have gotten any notice one way or the other. I did forget to note that those are 100% crops, but I figured that kinda went without speaking. Anyone have thoughts on a better way to get the comparison shots?

Maybe I'll actually do a fair test of the 20D vs. the 40D if I get bored tonight.

braduardo wrote in post #4058661
Well, I posted mine up there, but they don't seem to have gotten any notice one way or the other.

It looks like the focus on your camera changed from iso1600 to 3200 as the 3200 image is much sharper and in focus compared to the iso1600. You should try a more controlled test .

Yeah, I saw 'em, Brad, but I don't really know if they address the question of whether - as suggested above - the 40D's chroma noise is "obviously" worse than say, the 30D.

I'll re-state my experience that - in Real World use - I am repeatedly struck by how low chroma noise is in the pictures I'm getting from the 40D compared with the 30D: in (as near as dammit) identical shooting situations, cropped and processed in the same way, I'm seeing less chroma than I routinely get with the 30D.

In fact in general terms, the noise I'm seeing is very (pre 12mp sensors) Nikon like - finely grained, low in chroma clumping and (apparently) film-like.

Easy to deal with in PP NR too, without hurting detail.

this is at ISO 1600:

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

Keith R wrote in post #4061662
...I'll re-state my experience that - in Real World use - I am repeatedly struck by how low chroma noise is in the pictures I'm getting from the 40D ....

I've also been pleasantly surprised by the very low chroma noise on most of my pics, with the 40D. I usually have to turn the slider from the LR default to zero chroma NR and the view to 200% or even 400%, just to check that there is any chroma noise at all (and there's normally just a tad). I can often get away with much less chroma NR than the default setting in Lightroom. Not always, but much more often than I would have expected.

Good to see I'm not imagining things, BB!



Grinchy, what's your take - based on that picture, for example?

I can see some chroma, mainly in the darker BG areas (I've looked at the full-sized original), but again, it's much less than I believe a 30D would generate on the same shot.

Not that the 30D is bad, but the 40D seems obviously better...

Keith R wrote in post #4074445
Good to see I'm not imagining things, BB!



Grinchy, what's your take - based on that picture, for example?

I can see some chroma, mainly in the darker BG areas (I've looked at the full-sized original), but again, it's much less than I believe a 30D would generate on the same shot.

Not that the 30D is bad, but the 40D seems obviously better...

overall I am VERY happy with the 40D....