7D Experiment

[Oggie]

I have found that paradoxically if I take similar pictures in large, medium and small RAW, the small is the sharpest and both medium and small are significantly better than the large.

Could people try this and see if they find the same?

[DrJon]

Re: 7D Experiment

How do you compare them, as they need to be the same size and down-sizing removes sharpness? (Plus depends critically on the algorithm.)

[briansquibb]

Re: 7D Experiment

Glad it isn't only me then that has come to this conclusion!

It also gives a bigger buffer for continuous shooting, more photos per card and faster post processing

[Oggie]

Re: 7D Experiment

How do you compare them, as they need to be the same size and down-sizing removes sharpness? (Plus depends critically on the algorithm.)

Not ...... the algorithm! smilieshock.gif

Sorry.

When I view the images at similar size of area of interest on my screen, for example I took pictures of birds on my feeder then enlarged the view in DPP so that in each case the bird occupied about two thirds linear of my screen.

The reason I ask is that my 7D has just returned from Elstree where it went to have it's sharpness sorted and when it came back it seemed very much better. When I got a chance to try it more thoroughly, I found it set to medium JPG. When I set it back to large RAW, it is as bad as ever.

[Woolley]

Re: 7D Experiment

Not scientifically tested but my inclination is that MRAW give the best quality, but needs sharpening (as do all RAW images)

[DrJon]

Re: 7D Experiment

Then you are relying on how well your graphics card, or possibly DPP, resizes the image (yes, it is all about the algorithm), so it's a lottery on which ones it will do best. The best comparison is to print them both at the same size using a high-quality workflow (QImage, for example). Anything else and the differences will be submerged in what the software is doing to the image.

MRAW is just a scaled-down version of Raw, it really won't be any better, except the camera will scale it better than DPP will onto the screen.

[DrJon]

Re: 7D Experiment

P.S. graphics card drivers use really poor resizing. My favorite free resizer is the Lanczos one in Irfanview (Image->Resize, select Resample radio button and Lanczos from the drop-down). Note all resizing will lower sharpness, so you should sharpen afterwards. (DPP might even so that to the displayed image, again based on some unknown algorithm.)

P.P.S. I mention QImage as it resizes the images using a good quality algorithm before sending it to the printer driver, so it sends the image to the printer at the printer's resolution, so there will be no resizing in the printer driver to affect things. (Especially as it may vary with resolution.) Also there is a free trial (well, there used to be, I bought it a long time ago).

[briansquibb]

Re: 7D Experiment

Yes I would go with the camera resizing being the best. And when you print at 720dpi there is little of no resizing to be done by the printer driver either.

This use of mRaw and sRAW being sharper appears to be a common feeling over the various forums.

[DrJon]

Re: 7D Experiment

I'm really pretty sure it isn't, certain even, just needs better processing of the full-size raw image... (Plus no IMHO here, as I'm happy raw will ultimately be capable of sharper images AT THE SAME SIZE as any of the reduced raw formats). Just not with the graphics driver doing the resizing, and possible not the printer driver (BTW QImage will happily resize to 720dpi before printing).

I's say the idea mRaw or sRaw is better is in the Urban Myth category, or people not knowing what they are doing. Sorry, but we're somewhere where I really do understand what happens (could be a first :-)

[Oggie]

Re: 7D Experiment

What you say is quite interesting Dr Jon.

I have yet to get into printing. I view pictures on my monitor only.

Bottom line I do not get acceptable pictures in LRAW but I do in MRAW. If it were a quirk in hardware or software then I would expect similar issues with my 1D Mkiv. I have tried DPP, Silkypix and PSE and get similar results with all of them. I would be very disappointed is Canon's own software was not compatible with their cameras.

What it is not in my case is an myth. I have first hand evidence of something being wrong somewhere. I also found it out first hand as I mentioned rather than reading it.

I also have a Pentax K20 which behaves as you would expect, the quality of the picture being roughly proportional to the size of the file recorded. It also produces a much better picture than the 7D at full size (if the battery doesn't go flat while it is focussing). It has to be said that both of my other cameras have a lower picture count than 7D - I do not know if that is relevant.

I'll have to have a look at the programs you mention and see if I get any joy with them.

[briansquibb]

Re: 7D Experiment

Here is an article by Ken Rockwell

http://www.kenrockwell.com/canon/5d-mk-ii.htm#perf

Page down untill you get to the section about Sharpness of Smaller images

Brian

[Donphotography]

Re: 7D Experiment

well guy's considering the extra controls and the 18mp + focus adjustment also the praise you give the 7D there seem to be quite a lot of problems with this camera, now I see you saying that the sharpest image is on the smaller file.
I think you have found a new toy and and the game is to find and solve the problem with it, personally I think you have been stung by canon.

[Muscat]

Re: 7D Experiment

Here is an article by Ken Rockwell

http://www.kenrockwell.com/canon/5d-mk-ii.htm#perf

Page down untill you get to the section about Sharpness of Smaller images

Brian

An interesting article

[briansquibb]

Re: 7D Experiment

well guy's considering the extra controls and the 18mp + focus adjustment also the praise you give the 7D there seem to be quite a lot of problems with this camera, now I see you saying that the sharpest image is on the smaller file.
I think you have found a new toy and and the game is to find and solve the problem with it, personally I think you have been stung by canon.

I think you are totally wrong - all the evidence on this forum shows the 7D to be a very good. It is just that there is a way to improve it.

What problems do you think there are with the 7D? Is this your experience too? For one of Canon's best selling cameras to have numerous major fundemental flaws without a large recall is just beyond belief - especially as an owner of two 7D's that are giving excellent results.

[Donphotography]

Re: 7D Experiment

I think you are totally wrong - all the evidence on this forum shows the 7D to be a very good. It is just that there is a way to improve it.

What problems do you think there are with the 7D? Is this your experience too? For one of Canon's best selling cameras to have numerous major fundemental flaws without a large recall is just beyond belief - especially as an owner of two 7D's that are giving excellent results.

I must be dyslexic because everything I am reading seem to be a problem with this camera 7D.
Right so now your saying that you haven't had any issues with them, perhaps thats why you got two LOL , no I don't own nor do I want to own one.