For those of you that think the 7D is good with noise...

jase1125 wrote in post #12614564
I had to do the same thing with my 5d2 at higher ISO's. ETTR to minimize noise isn't exclusive to one model of camera. We all know you don't have to do it at ISO 3200 on a 5d2 - as expected since most tests show it is about a stop better noise wise. However, put your 5d2 at 6400 and you will have to do the same to minimize noise. Again this isn't specific to the 7D. I can choose to shoot ISO 3200 with proper exposure at ISO 3200 and get acceptable (to me) results just like the 5d2 at 6400. However, both will benefit with noise reduction if I do ETTR. I could make the argument the 5d2 ISO capabilities suck because I can get better results with a Nikon D3S.

Even with a 5Dii file resized to a D3S file size?

phreeky wrote in post #12614643
I think that's about what any sane person expected.

You may have distilled the crux of this thread, or at least the source of the conflict expressed here, into a single word. Congratulations!

phreeky wrote in post #12614643
It's similar, maybe a touch better, than all the xxD cameras that were released prior to the 7D. Is that "not good"? I think that's about what any sane person expected, as it's not like Canon were making remarkable promises about the 7Ds noise handling capabilities.

If you expected more then that's your fault. Learn to do more thorough research and buy the camera body appropriate to your line of work next time.

He did, that's why he ditched the 7D got a 5Dii, tbh though reviews in general seem more favourable than they should be imo, so I think he could be forgiven for that mistake, especially considering he might have though a 7D would get close to a 5D classic.

idsurfer wrote in post #12612449
What's the rest of the EXIF info? SS?

Camera Canon EOS 7D
Exposure 0.02 sec (1/50)
Aperture f/2.8
Focal Length 60 mm
ISO Speed 5000
Exposure Bias 0 EV
Flash Off, Did not fire

rhys216 wrote in post #12614927
He did, that's why he ditched the 7D got a 5Dii, tbh though reviews in general seem more favourable than they should be imo, so I think he could be forgiven for that mistake, especially considering he might have though a 7D would get close to a 5D classic.

The problem with review sites is that sometimes they are biased, unscientific and demonstrate results only under certain conditions. This is why you should never consider only reading reviews as your source of information. Go to a store, shoot both side by side, and then decide. This will save you a lot of frustration.

Not to do with the case, but when I was out for shopping my super-tele, I was thorn between the 100-400L and the Sigma 150-500. If you read 100 reviews on the net, 99 will tell you to get the 100-400L as it is much sharper and faster. I went to a store and came home with the 150-500. Never looked back. My 150-500 copy is equally sharp or sharper than the 100-400L copy I tested it against. Call me lucky if you want. Fact is, if I had the 100-400L and got a chance to see the 150-500 out of the lens that now is mine, I'd be disappointed. This is why I don't trust 100% review sites.

phreeky wrote in post #12609137
- Canon has learnt to fit more pixels onto sensors without increasing overall noise (not per pixel!), but hasn't improved it much either.

The 7D actually has about the same pixel-level read noise as the 30D, at high ISOs, and 1/3 stop less at base ISO. It had increased in the 30D -> 40D -> 50D progression at high ISOs, and then dropped to about 30D levels i the 7D. Pixel-level photon count has decreased each step of the way, but quantum efficiency has remained the same or slightly increased (20D was ahead of its time in QE for a DSLR sensor).

The 7D does have its issues, but it has less noise, per unit of sensor area, than any xxD camera before the 60D. Most of the noise complaints, I think, are from people who are over-sharpening their 7D images at the pixel level, and/or viewing at the pixel level. How many prints do people make that are 5 feet wide? How many 5-foot-wide photos in museums do you see, devoid of noise or grain? And if you're going to make a 5 foot print, you should realize that you may need better than 1/fl for shutter speed, and focus is critical.

magwai wrote in post #12609250
Serious question: how can pixel size affect whether a filter is needed or not?

The AA filter is an optical filter that turn a point light source into 4 points forming the corners of a square; it has nothing to do with color, but the implied square is designed to be at a certain ratio to the pixel spacing. The filter would be necessary for proper sampling even with a B&W sensor. When the lens is soft enough, relative to pixel spacing, the AA filter is not necessary.

bohdank wrote in post #12610974
If you can show me how to do that with a zoom, I'd be greatful

Let's not forget, whatever a 7D can do with noise a 5d2 can do it better.

Not always; the 7D has better deep shadows at low ISOs, and less banding at very high ISOs. The 5D2 excels in low-DR low-ISO shots, and medium-high ISOs shots that don't show the horizontal banding.

The 7D has less horizontal banding at all ISOs, and it's vertical banding, most visible at low ISOs, is partly removable if a converter chooses to do so, because it is the same in every frame. 5D2 banding is in random locations each frame, and can not be calibrated out like it can be with the 7D.

This discussion is interesting. Canon's own brochure claims
""""""""
The EOS 7D uses Dual “DIGIC 4” processors to move the data from
the sensor into the image processing pipeline faster and give the
power to deal with the data from the 18 Megapixel sensor, even
when shooting at 8 frames per second. “DIGIC 4” is able to remove
highly noticeable colour noise and luminance noise without loss
in image detail. This allows for higher ISO shooting without the
associated issues of noise. The EOS 7D has a standard ISO range of
100-6400 and an expansion H setting of ISO 12,800
AND
Developed and built in-house, the CMOS sensor in the EOS 7D
delivers high resolution, high sensitivity that provides outstanding
image quality.
AND
The next leap in the ongoing development of the EOS SLR family is the outstanding EOS 7D.
""""""
I think that after reading this, it would be reasonable to think that I should expect good quality images at 100 ISO, which would be as good as, or better than my 40D. They talk about development progression, outstanding image quality, etc etc, but the reality is, on some 7Ds (mine without any doubt) 100 ISO images are dreadful. Nowhere in the brochure does it say I have to do this that or the next thing to get quality images. They boast how technologically advanced this camera is, so It would be reasonable for me to expect that if I reset everything to defaults and set the camera to auto on a bright sunny day and take a picture with blue sky I would get a pretty good picture( This is what Canon told me to do to confirm the problem), after all they say """the CMOS sensor in the EOS 7D delivers high resolution, high sensitivity that provides outstanding image quality"""
There is no right and wrong in this argument, it's not about whether person A or B can use a camera, understand digital technology etc etc. It's a case of quality control, or the lack of it. Some cameras are noisy others aren't and it's about what the manufacturer does about the problem.
Interestingly when I enquired about my camera which is in to be checked, I took the opportuniity to explain the noise was not the same as simply re-mapping the sensor it was image noise at 100ASA and was told " Yes, we know the problem"???

bohdank wrote in post #12610936
ETTR is not always possible and rarely useable under conditions where noise is going to be an issue.

ETTR at high ISOs is only useful for JPEGs, anyway. For RAWs, it is just an illusion, as most of the apparent benefit is the fact that if you use ETTR, you are actually shooting at a lower ISO. The ISO setting is *not* the ISO of your conversion. A latent RAW image, yet unconverted, really has no ISO exposure index; that is determined by how much exposure occurred in an area that is going to be rendered as gray in the final product.

Shooting at ISO setting 12800 with +1 EC has the same noise as 6400 with 0 EC, and 3200 with -1EC. The latter has 2 stops more DR in the form of headroom, though, and the RAW files are smaller.

rhys216 wrote in post #12611852
"As an aside, I think this is probably going to put the last nail in the Megapixel war’s coffin. The 4/3 companies have already said 12 Mpix is as far as they intend to go. I suspect the full-frame manufacturers are going to call a halt at 30 Mpix or so, just because there’s no sense in it: they’re already out-resolving the quality control of their best lenses."

Sounds like a lot of BS, to me. My superzoom with 1.7 micron pixels aliases the RAW blue channel, at f/8.

There are a lot of heads screwed on backwards regarding this topic, many of them in high places.

idsurfer wrote in post #12614612
I'll be the first to admit I do not understand imaging science. Are you suggesting that I shoot at a lower res RAW setting during times that I want to use higher ISO's?

No, I'm saying be aware of the variables that change when you look at images from different MP cameras. Our viewing methods and hardware are full of illusory trappings.

sandyn666666 wrote in post #12615177
This discussion is interesting. Canon's own brochure claims
""""""""
The EOS 7D uses Dual “DIGIC 4” processors to move the data from
the sensor into the image processing pipeline faster and give the
power to deal with the data from the 18 Megapixel sensor, even
when shooting at 8 frames per second. “DIGIC 4” is able to remove
highly noticeable colour noise and luminance noise without loss
in image detail. This allows for higher ISO shooting without the
associated issues of noise. The EOS 7D has a standard ISO range of
100-6400 and an expansion H setting of ISO 12,800
AND
Developed and built in-house, the CMOS sensor in the EOS 7D
delivers high resolution, high sensitivity that provides outstanding
image quality.
AND
The next leap in the ongoing development of the EOS SLR family is the outstanding EOS 7D.
""""""
I think that after reading this, it would be reasonable to think that I should expect good quality images at 100 ISO, which would be as good as, or better than my 40D. They talk about development progression, outstanding image quality, etc etc, but the reality is, on some 7Ds (mine without any doubt) 100 ISO images are dreadful. Nowhere in the brochure does it say I have to do this that or the next thing to get quality images. They boast how technologically advanced this camera is, so It would be reasonable for me to expect that if I reset everything to defaults and set the camera to auto on a bright sunny day and take a picture with blue sky I would get a pretty good picture( This is what Canon told me to do to confirm the problem), after all they say """the CMOS sensor in the EOS 7D delivers high resolution, high sensitivity that provides outstanding image quality"""
There is no right and wrong in this argument, it's not about whether person A or B can use a camera, understand digital technology etc etc. It's a case of quality control, or the lack of it. Some cameras are noisy others aren't and it's about what the manufacturer does about the problem.
Interestingly when I enquired about my camera which is in to be checked, I took the opportuniity to explain the noise was not the same as simply re-mapping the sensor it was image noise at 100ASA and was told " Yes, we know the problem"???

So when BMW tells you you're buying "The Ultimate Driving Machine," do you actually believe it to be the ultimate driving machine? If you do, I have some bottled air to sell you. It'll be the best air you've ever inhaled.

John Sheehy wrote in post #12615180
ETTR at high ISOs is only useful for JPEGs, anyway. For RAWs, it is just an illusion, as most of the apparent benefit is the fact that if you use ETTR, you are actually shooting at a lower ISO. The ISO setting is *not* the ISO of your conversion. A latent RAW image, yet unconverted, really has no ISO exposure index; that is determined by how much exposure occurred in an area that is going to be rendered as gray in the final product.

Shooting at ISO setting 12800 with +1 EC has the same noise as 6400 with 0 EC, and 3200 with -1EC. The latter has 2 stops more DR in the form of headroom, though, and the RAW files are smaller.

i don't get what you are saying... ETTR is to take advantage of a deeper bit rate which is designed from a low bit rate on the dark end and mathematically increase as you go to all white/full saturation. How much bit depth is based on cmos architecture + accociated chip package.