OFFICIAL: Canon 5D Mark III Announced

HappySnapper90 wrote in post #14052667
Sounds like a focusing issue as an image sensor isn't "sharp" or "soft" by itself.

Actually they can be if you smash them with too may MP. Also, look at the AA filter issue that seems to be cropping up. Nikon is aware or this and has created the D800E.

But you're right, it could be a focus issue.

Yohan Pamudji wrote in post #14052334
If you didn't label which was which and I had to guess, I might as well flip a coin.

I guess we are both blind then. I'm not seeing much of a difference either. I suppose no improvement isn't necessarily a great thing, but some are saying the 5D3 looks worse. I am just not seeing it. Personally, I wanted a 5D2 with better AF, so I am quite happy, so whatever. Just put these up because I read some people claim a 1.5 to 2 stop difference. Not even close.

bacchanal wrote in post #14052414
Whoa, whoa, whoa buddy. This thread is only for complaining about the 5DIII.

I still can't stop laughing

palmor wrote in post #14052693
Those look very good IMO.

Heresy!!!

Mike55 wrote in post #14052699
Actually they can be if you smash them with too may MP. Also, look at the AA filter issue that seems to be cropping up. Nikon is aware or this and has created the D800E.

But you're right, it could be a focus issue.

The AA filter isn't part of the sensor. The sensor just records whatever image is projected onto it - it can't be 'sharp' or 'unsharp'. But the sharpness of the image is affected by a number of things. There is lens sharpness - poor-quality lenses often don't project the sharpest images even at optimal settings. There is diffraction-related unsharpness at small apertures - however, this is rarely a problem, except when shooting at very small apertures, and a high-resolution sensor will never be less sharp (at the same print size) as a lower-resolution sensor simply due to diffraction or lens unsharpness. Then there is unsharpness caused by the AA filter, which deliberately blurs the image projected onto the sensor.

bacchanal wrote in post #14052414
Whoa, whoa, whoa buddy. This thread is only for complaining about the 5DIII.

yes !!

nicksan wrote in post #14052735
Heresy!!!

For me the low ISO examples you posted don't show much difference between the two. I don't think that is a bad thing for these examples though. We aren't going to know if the IQ is better/worse until some reviews come out that really measure DR/noise etc.

I'm in a slightly different boat then most because I'm looking to upgrade to the 5d III not from the 5d II. I shoot a 7d and a 1d III now and the 5d III will replace my 7d. So for me I think there is going to be a big upgrade.. bigger then someone coming from the 5d II.

Currently conducting an informal survey of IQ. I've taken unsharpened crops of the points of sharpest focus of various images from the 5D3 and 5D2 and printed them as 4x6" prints (equivalent to 24x36" if I were to print the whole photo at the same magnification). I've laid them out next to each other in a random order, and will be asking non-photographers in my workplace to identify which photo was taken with which camera. I will be telling them that two cameras were used - one sharper than the other - without identifying the brand and model, and asking them to label each image as 'sharper' or 'softer' on a separate form, which will then be collected and the results tabulated.

I will post the results later.

Shadowblade wrote in post #14052780
Currently conducting an informal survey of IQ. I've taken unsharpened crops of the points of sharpest focus of various images from the 5D3 and 5D2 and printed them as 4x6" prints (equivalent to 24x36" if I were to print the whole photo at the same magnification). I've laid them out next to each other in a random order, and will be asking non-photographers in my workplace to identify which photo was taken with which camera. I will be telling them that two cameras were used - one sharper than the other - without identifying the brand and model, and asking them to label each image as 'sharper' or 'softer' on a separate form, which will then be collected and the results tabulated.

I will post the results later.

If you tell them that one is, in fact, sharper, your results will be invalid.

Simply ask them to pick which one looks sharper.

OK, since the "7D look" thing came up, I decided to do a side by side between it and the 5D3.

Same deal. Sharpening zeroed out. 7D on left. 5D3 on right. 5D3 resized to 7D resolution. (Not sure what implications that would have...)

ISO100 detail area
Link

ISO100 detail area #2
Link

ISO100 detail area #3
Link

ISO6400 comparo
Link

Yeah, I'm seeing the "7D look"...on the 7D photos. The 5D3 files are closer to the 5D2 than the 7D for sure. Of course at ISO6400, we get a pretty clear picture which camera is superior. Not surprising.

FlyingPhotog wrote in post #14052795
If you tell them that one is, in fact, sharper, your results will be invalid.

Simply ask them to pick which one looks sharper.

Uh yeah...because I bet we all can't wait to see the results.

nicksan wrote in post #14052253
OK, here are side-by-sides at ISO100 of some detailed areas of the IR images.
Sharpening and NR zeroed out in ACR.

Link

Link

Link

IMHO, these are inconclusive. Can't really see much of a difference actually.

The first two it's hard to tell. The third image it looks like the 5D3 has a little more detail. Definitely not noticeable without cropping and studying them.

nicksan wrote in post #14052808
OK, since the "7D look" thing came up, I decided to do a side by side between it and the 5D3.

Same deal. Sharpening zeroed out. 7D on left. 5D3 on right. 5D3 resized to 7D resolution. (Not sure what implications that would have...)

ISO100 detail area
Link

ISO100 detail area #2
Link

ISO100 detail area #3
Link

ISO6400 comparo
Link

Yeah, I'm seeing the "7D look"...on the 7D photos. The 5D3 files are closer to the 5D2 than the 7D for sure. Of course at ISO6400, we get a pretty clear picture which camera is superior. Not surprising.

Thanks for doing all of these btw.. not sure if anyone has said that yet.

I really can't wait to shoot indoor dog agility with the 5d III. While the 7d and 1d III have done the job I'm going to see an immediate improvement in my images at ISO6400 and 12,800. It is going to let me offer larger prints to my clients in the indoor events then I normally would.

nicksan wrote in post #14052808
OK, since the "7D look" thing came up, I decided to do a side by side between it and the 5D3.

Same deal. Sharpening zeroed out. 7D on left. 5D3 on right. 5D3 resized to 7D resolution. (Not sure what implications that would have...)

ISO100 detail area
Link

ISO100 detail area #2
Link

ISO100 detail area #3
Link

ISO6400 comparo
Link

Yeah, I'm seeing the "7D look"...on the 7D photos. The 5D3 files are closer to the 5D2 than the 7D for sure. Of course at ISO6400, we get a pretty clear picture which camera is superior. Not surprising.

Yikes. Easy to see the 7D "blur" there.

nicksan wrote in post #14052808
OK, since the "7D look" thing came up, I decided to do a side by side between it and the 5D3.

Same deal. Sharpening zeroed out. 7D on left. 5D3 on right. 5D3 resized to 7D resolution. (Not sure what implications that would have...)

ISO100 detail area
Link

ISO100 detail area #2
Link

ISO100 detail area #3
Link

ISO6400 comparo
Link

Yeah, I'm seeing the "7D look"...on the 7D photos. The 5D3 files are closer to the 5D2 than the 7D for sure. Of course at ISO6400, we get a pretty clear picture which camera is superior. Not surprising.

Not a valid comparison. After all, we're arguing that the 5D3's images look like the 7D's blown up to a full-frame sensor (i.e. that a 1.6x crop from the 5D3 looks like a 7D's image), not that the 5D3 doesn't have better overall IQ due to being a larger sensor. You've equalised for resolution, but not for sensor area. The 5D3 crops are taken using 2.56 times more sensor area than the 7D crops.

For a valid comparison, take a 1.6x crop from the 22MP file (resizing it down to 8.6MP), blow it up to the 7D's 18MP, then put crops of the two images side by side. That way, 1000 pixels of the resized 5D3 image will have been taken using exactly the same sensor area as 1000 pixels of the 7D image.