7D on the way & Resolution questions

Well just ordered a 7D & EF-S 15-85 (UPS hurry up)
I like to know the usefulness of M-Raw (or S-Raw) – as I do not have it on the Xsi or 5D
How does canon reduce the size?
Is it pixel averaging?
Does the noise level remains the same (or smears or reduces)?
For that matter how does the process work on Jpeg reduction e.g. from fine to standard ?
Does canon also change the dynamic range (I think Nikon change from 14bit to 12bit color depth for speed)
Hope you can shade some light on my sensor

My point of view is that I would never use it. You paid for that many megapixels, so why not take them? Memory is cheap enough these days.

i agree but wodered how does M-Raw works
I do alot of Bracketing and somtimes would like use M-Raw (but i like to know the trade offs)

These were shot at MRAW, done when I first had the 7D, I didn't have time to swap back to fullsize as this guy only appeared for 30 seconds!

MIME changed to 'text/html' | Content warning: script

MIME changed to 'text/html' | Content warning: script

Congrats on buying the 7D.The different MP's only limit you yo the size of print that you want to make-it's that simple.If you have the memory space then use it so if you get that great shot that you want to make a 30x40 print you'll be able to.It's too easy to forget the mode that you left the camera in (at least for me) and lose that moment so...it's up to you to not use all the MP's-I don't lower mine "just to be sure".

I can see how there would be times when shooting one of the smaller raw formats may be a good idea (e.g. taking stills for a time-lapse movie). But I'm not sure why it may be an advantage when using bracketing?

Tiberius47 wrote in post #10191347
My point of view is that I would never use it. You paid for that many megapixels, so why not take them? Memory is cheap enough these days.

Because I don't need 21 megapixels (5D2 here). I shoot at sRAW1 (which turns out to be 11 megapixels), which is more than enough. I don't crop my pictures and don't print gigantic posters.

Tiberius47 wrote in post #10191347
My point of view is that I would never use it. You paid for that many megapixels, so why not take them? Memory is cheap enough these days.

Shooting in mRAW or sRAW increases your Max Burst count.

In the 7D it works out to this ....
In RAW Max Burst is 15 ( 18 MP )
In mRAW Max Burst is 24 ( 10 MP )
In sRAW Max Burst is 38 ( 4.5 MP )

CF Memory may be cheap, but you cannot increase the size of the in-camera buffer memory. You can increase its efficiency by decreasing image size.

I didn't pay what I paid for the 7D to buy more megapixels, I paid for increased capability and flexibility that I did not have in my previous DSLR body.

harcosparky wrote in post #10194636
Shooting in mRAW or sRAW increases your Max Burst count.

In the 7D it works out to this ....
In RAW Max Burst is 15 ( 18 MP )
In mRAW Max Burst is 24 ( 10 MP )
In sRAW Max Burst is 38 ( 4.5 MP )

CF Memory may be cheap, but you cannot increase the size of the in-camera buffer memory. You can increase its efficiency by decreasing image size.

I didn't pay what I paid for the 7D to buy more megapixels, I paid for increased capability and flexibility that I did not have in my previous DSLR body.

Exactly. My 5D2 has a full auto mode. To be literal, I paid for that auto mode. Doesn't mean I bought it for auto mode. Just because it's there doesn't mean you have to use it.

Don't get me wrong, 18+ megapixels are great for large prints and cropping. But that aside, why?

This brings up a good point - how are the mRAW or sRAW files generated on the camera? Are all the pixels captured then down-sampled, or are only a subset of pixels captured (e.g. every second pixel)? It seems the latter method would be the best for increasing the buffer depth. However, this would (I'm guessing) give a direct decrease in SNR - same physical sensor size, less pixels, yet the photodiode size stays the same, so we would have more noise per pixel compared to if a full RAW file were generated then down-sampled (thereby increasing SNR via averaging).

This would probably be simple to test - shoot a picture in RAW and sRAW, then down-sample the RAW to the same size as the sRAW and observe if the noise changed.

I guess my question has been discussed before (big surprise!):

https://photography-on-the.net/forum/showthre​ad.php?t=762492
https://photography-on-the.net/forum/showthre​ad.php?t=730030
https://photography-on-the.net/forum/showthre​ad.php?t=746579

Also, I'd recommend reading about pixel binning for those interested - it's not that sRAW simply takes a subset of pixels, but rather combines several pixels into a "super-pixel".

Well now the discussion is getting somewhere
i bought the 7D for multiple usage
if i do BIF i want to crop and will use 18M & 8FPS
when i do ladnscape for web display i may want to use M-Raw

i wonder if someone knows how does canon reduce the resolution - than we can decide how to best use it

It will be nice to experiment and actually test noise level vs details vs dynamic range in these diffrent modes

I believe the only use for the lower raw sizes are:

1) to reduce your overall file transfer time
2) to increase total buffer image count and/or to fit more files on a card

In all other situations, you could very well create an action to process the pics (or just do a resize before you perform the rest of your post processing).

I see no other real improvements, since the 7D is not pixel-binning, but rather is doing a resize for you via raw. Just as many believe removing noise is best reserved for the post processing software instead of in-camera, resizing may be too. I have been bit one too many times in the past where I had one of those "wow" shots, and wished I hadn't handicapped myself by shooting at a lower file size. Now I just size down in my post processing actions, this way should I ever have a "wow" moment again, I have the full resolution image available.

Teamspeed - your logic is hard to dispute
so i guess i shall keep my setup on RAW
however it still intrigues me how does Canon perform the downsizing

omer wrote in post #10196179
Teamspeed - your logic is hard to dispute
so i guess i shall keep my setup on RAW
however it still intrigues me how does Canon perform the downsizing

A very long and periodically interesting paper on Canon raw, sraw and mraw. Not sure if it answers what you are inquiring on though.

http://dougkerr.net/Pu​mpkin/articles/sRaw.pd​f