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

Im still trying to work out why people are even comparing the 7D with the 5D2, two totally different cameras!

aladyforty wrote in post #12612667
Im still trying to work out why people are even comparing the 7D with the 5D2, two totally different cameras!

Yeah I don't get it either. If you bought a 7D and expected what the 5D2 is then insufficient research was done. Both cameras can do things the other cannot, not to mention the huge price difference.

The subject of the thread is

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

aladyforty wrote in post #12608990
yeah possibly, 100-400L handheld out of a car window in a bit of a hurry on the job.

A good rule of thumb for 1.6x FOVCF is 1 / (1.6 x focal length) for shutter speed. If the car is moving, then you need to have something faster, perhaps 2x or 4x as fast.

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

That's an odd statement, considering that ETTR reduces noise in the tones that you're giving positive EC.

No, it's not, if you do not take it of context.

This is what I wrote

ETTR is not always possible and rarely useable under conditions where noise is going to be an issue.

So, what do I do ? I'm shooting a concert ISO 3200, F2.8 and 1/125s to get a reasonably accurate exposure. Tell me how I can use ETTR ?

bohdank wrote in post #12612834
No, it's not, if you do not take it of context.

This is what I wrote

ETTR is not always possible and rarely useable under conditions where noise is going to be an issue.

So, what do I do ? I'm shooting a concert ISO 3200, F2.8 and 1/125s to get a reasonably accurate exposure. Tell me how I can use ETTR ?

Yes, it is. You use a slower shutter or faster aperture. Or if you have hardware support, use ISO 6400.

rhys216 wrote in post #12612179
While I agree, with these two points in principal, I seriously doubt we will be seeing camera's from Canon/Nikon without AA filters any time soon, unless they release a camera purely for stills, because moire is a serious issue for video...

You're not listening. There wouldn't be any moire. You only need AA filters when the lens is under-sampled, with a coarse pixel density.

Current Canon DSLRs with video are seriously aliased, BTW. You can't do low-res video from high-res stills without aliasing, even with a very strong AA filter, unless you downsample the full image, or at least the edges of the virtual pixels of the new resolution. Canon skips pixels.

I don't want aliasing, stills or video.

bohdank wrote in post #12612834
No, it's not, if you do not take it of context.

This is what I wrote

ETTR is not always possible and rarely useable under conditions where noise is going to be an issue.

So, what do I do ? I'm shooting a concert ISO 3200, F2.8 and 1/125s to get a reasonably accurate exposure. Tell me how I can use ETTR ?

That argument can be made with any camera operating at its upper ISO limit. The same could be said with a 5D2 at ISO 6400. However, I can ERRT at ISO 1600 with a faster shutter speed than trying to cram a shot in at ISO 800 under exposed or in some circumstances "proper exposure" and end up with less noise in the ETTR ISO 1600 shot.

I think the reality is if one really studies and learns a particular camera then outstanding pictures can be captured. We try to argue that one camera is better than the other but the camera in and of itself is of no value - only the pictures that result are and that requires a photographer as well. I have more people wanting to purchase a picture I took of Mount Rainier from an airplane with a S95 than other pics I have taken using a 5D2, 7D, 1D Mark III and a 1D Mark IV.

It really is kind of silly when you think about those who are on a crusade to prove their camera is better than someone else's. Every camera is bested by another model in very short order so what is the point really?

Poe wrote in post #12612850
Yes, it is. You use a slower shutter or faster aperture. Or if you have hardware support, use ISO 6400.

Yes, but slower SS will cause blurry moving subjects. Even apertures of 1.4 are not going to give you SS fast enough to capture people in a dimly lit setting.

phreeky wrote in post #12612730
Yeah I don't get it either. If you bought a 7D and expected what the 5D2 is then insufficient research was done. Both cameras can do things the other cannot, not to mention the huge price difference.

My best friend owns a 5D11 and wishes it had 7D focus, Id love the 7D to have the quality of image that the 5D11 has but it doesn't. the sooner people learn to work with what they own the better. I have just spent the last week working out the 7D, I think I've got a handle on how it reacts in different situations now, it is totally different to my 5Dc and previous cameras but I'm happy with the quality for what I can do with it.

As far a noise goes, I've seen worse. it is manageable. If you don't want to post process get a camera that does it all in camera for you, forget DSLRs. I dont mind post processing as long as the end image is ok. also I printed out a few so called noisy shot only to find they looked awesome in print. For the net I just have to apply noise reduction a little more.

I think you could spend way too much time going through all the technological stuff that is on this thread and much is written by people who don't even own the 7D, Id rather just get on with learning how to use the tool that is the 7D, after all the camera is just a tool. its what you do with the images after you shoot them that counts

idsurfer wrote in post #12612917
Yes, but slower SS will cause blurry moving subjects. Even apertures of 1.4 are not going to give you SS fast enough to capture people in a dimly lit setting.

He never said anything about getting a particular DOF or stopping motion. He just said that it was going to get him a fair exposure. If there was more information to the scenario that needed to be taken into account, then it should have been stated up front else how are we suppose to give the best answer when we lack critical information?

The argument to use a faster shutter speed or larger aperture is stupid, otherwise why would you be that high ISO to begin with?

What you CAN do is increase your ISO (i.e. use the next ISO setting up, assuming it's still a true ISO setting), and then make sure you're not blowing your highlights too much with the full stop increase (a very slightly faster shutter speed might help, but if you can leave the shutter speed + aperture alone then it's best). Now you'll find the image looks over-exposed and noisier, but you may in fact find that when you now decrease the exposure in post that it has less noise than the lower ISO image.

phreeky wrote in post #12612994
The argument to use a faster shutter speed or larger aperture is stupid, otherwise why would you be that high ISO to begin with?

It is not stupid when you require a particular dynamic range to be captured.

phreeky wrote in post #12612239
As for the comments about higher densities but smaller output files, we already have lower resolution RAW files from the recent models. I can't say I've used them so I'm not sure if they're junk or something, but the idea has already been implemented.

The way sRAW and mRAW is done is not what I had in mind. They basically convert the RAW data to a color space like JPEG, but without the lossey compression. There are many ways to reduce RAW file size, including using a LUT and lower bit depth, without losing any spatial resolution, or using lossey compression at full resolution (lossey loses less at higher sampling rates). You can downsample to a CFA structure at another resolution, or downsample the interleaved color planes to non-interleaved, full RAWRGB pixels, using the same color transforms to *RGB, but without the demosaicing.

Nikon ruined the reputation of LUTs with their horrible lossy compression, which was designed by someone apparently unaware that their bad math caused color shifts.

magwai wrote in post #12609942
Thanks, that is interesting. I mistakenly thought that the AA filter was part of the low-level digital processing done internally by the camera.

The AA filter is two sheets of a (birefringent) material that turns a point light source into two point light sources, with one sheet 90 degrees rotated from the other, forming four dots from a single one. If you had a mismatched (fine) pixel pitch and (coarse) AA filter strength, you could see four images superimposed.

Perhaps you were thinking of demosaicing.

Demosaicing is the code which turns the RAW data in a CFA-filtered capture into an image of full-RAWRGB pixels, which are then converted to *RGB (sRGB, aRGB, or whatever). If the lens is sufficiently sampled (IOW, the sensor has a high enough pixel density), demosaicing can be a lot simpler, with no need to try to get sharp luminance at every output pixel.