Noise 7D ISO 12800

ttt

any NR or PP?

Looks fine to me.

I find that at ISO 12800 I don't get spectacular image quality, but at least I can get a shot where I would otherwise not be able to.

Use ISO 1600 and +3 EC.

are you really going to get annoyed at the ISO H1 on the 7d? really?
I think your doing pretty good here

Oh whoa is me. Again another look how great it is with such a high ISO. Here's how you test ISO.

1. shutter 1/60 (anything faster and there is too much light)
2. aperture 2.8 (anything smaller is too much light)

Please oh please, really test it.

Poe wrote in post #11406747
Use ISO 1600 and +3 EC.

Why? Doesn't the 7D have hardware supported amplification up to ISO 6400?

Has the shot been downsampled? Or is it an unedited crop from a larger image?

When you downsample an image, you're discarding pixels and then blending the remaining pixels together to fill in the 'gaps' created by the missing pixels. This by itself will reduce the noise in an image The more you shrink an image, then more you reduce the noise.

To see accurately how much noise there is in an image, and to display it within forum guidlines, you need to take a straight 1000 x 800 pixel crop out of the image and use that.

I routinely shoot at 6400 and 12800, but I won't post anything here, either folks are really tired of my ISO posts or will try to point to online reviews that say the 7D doesn't do well.

If you like the 7D, you would like the 5D2 as well, it is somewhere between 2/3 and 1 1/3 stop better.

shedberg wrote in post #11406601
Looks fine to me.

I find that at ISO 12800 I don't get spectacular image quality, but at least I can get a shot where I would otherwise not be able to.

+1...and anytime Team Speed implies he's happy with the quality, I know it must be good. I miss your tests, Team Speed!

godzakka wrote in post #11406867
+1...and anytime Team Speed implies he's happy with the quality, I know it must be good. I miss your tests, Team Speed!

Admit it, no you don't.

I am out of things to test. It is cold and snowy and I really don't want to go out. The kids have driven me crazy this week so I don't want to take pics of them right now. I have 2 18-55 kit lenses in my position currently, one IS and one not. Maybe I can do something with them.

Maybe a 7 way shoot out at 50mm at f5.6 between all the lenses I have currently that can do that FL:
- non IS 18-55
- IS 18-55
- 17-55
- 50 1.8
- 50 1.4
- 50-500 OS

Yikes.... I think I have the 50mm well covered.

davidfig wrote in post #11406761
Oh whoa is me. Again another look how great it is with such a high ISO. Here's how you test ISO.

1. shutter 1/60 (anything faster and there is too much light)
2. aperture 2.8 (anything smaller is too much light)

Please oh please, really test it.

Does 1/15 at f/4.5 count?

gjl711 wrote in post #11406770
Why? Doesn't the 7D have hardware supported amplification up to ISO 6400?

The amount of DR you keep at 1600 instead of 12800, versus the marginal increase in noise from the +3 EC is much preferred, IMHO.

Poe wrote in post #11407300
The amount of DR you keep at 1600 instead of 12800, versus the marginal increase in noise from the +3 EC is much preferred, IMHO.

So maybe I'm missing something here. True that the DR at 1600 is marginally better, but pushing it three stops is going to toss away any DR savings you have plus loads more. I would think that if DR is what your trying to save letting the hardware do the amplification at ISO6400 and pushed 1 stop to 12800 is going to result in the smallest loss of DR. (1 stop as opposed to 3 stops)

gjl711 wrote in post #11409309
So maybe I'm missing something here. True that the DR at 1600 is marginally better, but pushing it three stops is going to toss away any DR savings you have plus loads more. I would think that if DR is what your trying to save letting the hardware do the amplification at ISO6400 and pushed 1 stop to 12800 is going to result in the smallest loss of DR. (1 stop as opposed to 3 stops)

I tried this last night, and the differences are marginal, I seemed to get a tiny bit more detail with the 1600 ISO, but there are also alot of black/white specks everywhere, which after noise reduction could just end up causing more work when you are done, with maybe marginally better results that won't show up in prints. I am not sure it would be worth the effort, but I might try that a couple more times this week.