The pixelologists are in this thread
Pictures are posted in Nature and Wildlife, and other forums. I posted some nice 7D bird pictures there. Mazing did not ruin those photos.
Yep. Hope it gives you a good chuckle and a feeling of smug superiority.
thedigitalbean wrote:
You can add the 1Ds3, 1D3 and 5D2 to that list. I pulled up 7D and 1Ds3 images taken with the same glass of the same subject and in the same RAW processor and I see absolutely no difference in the "mazing" between these two bodies.
well you saw above where i posted one from my 50D and it was a heck of a lot smoother (and a 50D could give you birding reach at half the price, of course it is not quite as fast and the AF is not quite as good for one shot and hopefully not nearly as good for AI servo)
maybe I just got lucky with getting a 'rare' good copy of two 20D's, a 40D, a 50D, a 5D2 and trying 'rare' good copies of a couple 1D2ns, a 1D2, a 10D, a 20D and a 30D and got unlucky with two 7Ds.
Now maybe ACR 5.6 will deal with it reasonably and maybe C1v% can be coaxed into something reasonable but the 7D definitely does act differently than the previous bodies and they do not at all get these sorts of issues as easily.
ejmartin wrote:
I dunno. Is the difference in std dev correlated with anything? For instance is it greater when there is a larger difference of G-R and/or G-B? I seem to recall someone having a theory that the column readout swinging back and forth between color channels could affect the noise.
I have to go back and check to see if it varies per shade or not (it might just vary per total green signal in a direct way though).
Basically if I use Rawanalyze to select a block and then take the avg value and std dev of only green 1 for the patch or only green 2 for the patch the differences for both are larger with the 7Ds than with all the other canon bodies I have ever used. With the 7Ds it's like 0.5-2% for the avg values and 10-20% for the std dev while for the other cameras (none of which maze easily it's more like 0.05-0.5% and 2-5%). Where the average is higher the StdDev is always lower. All the 7D seem to have higher StdDev for G2 than G1 (not random body to body where some have G1>G2 and some G2<G1 or G2=G1).
i'd have to go back and check to say more though.
wild speculation:
maybe G1 is a little more broad spectrum to let more light in and improve SNR while G2 is a bit tighter band so they don't lose so much color info, of course if the demosaic just turns the G1/G2 difference into mazing noise or left over artifacts that would seem to more than defeat what little the extra light through G1 would provide....
too bad the 7D white paper is not out yet, not that they would necesarily get into this sort of detail about what they may or may not have done
the new 1D4 white paper seems to imply a normal 3-color RGB filter like with previous bodies
i'll try to fire off a letter to canon tech within the next couple of days and drive down there as well, if need be
brainiac wrote:
People just don't understand why comparing 100% crops favours the lower resolution camera...
By that logic, the 7D should be sharper than the 5DII when comparing 100% crops, no .
------------
For those who think it's unfair to compare 1.6x to FF, take a look at these test shots of the 40D and the new G11.
The G11 has a 9x smaller sensor than the 40D.
jorkata wrote:
By that logic, the 7D should be sharper than the 5DII when comparing 100% crops, no .
the 7D also has a smaller sensor though so it is not like comparing 18MP with 21MP on a FF, the smaller photosites stress things more so any handshake, motion blur, lens problems show up more
I just read all 140 pages(NOT), so basically, Skibum and Jorkata are the only folks having issue with the image quality. UCSB is in the happy camp now...finally.
What's going on with this camera ?? Should I sell mine even though it's working fine ??
MikeRichards: you can buy mine for $5000, yea, that's how good it is
Sorry, guys, I love this camera, I just cannot believe the amateur reviews that are floating around, everyone is trying his 5 minutes of fame, it seems. First there was that guy on BPN that used Spot-AF on this camera for BIF, then all these random folks who did not even use the camera to its full potential. Everyone wants to be a Rob Galbraith (did I spell that right?) these days with indoor tests....
Shoot, I had said "adios" to this thread earlier...doh !!!
kirry007 wrote:
I just read all 140 pages(NOT), so basically, Skibum and Jorkata are the only folks having issue with the image quality. UCSB is in the happy camp now...finally.
What's going on with this camera ?? Should I sell mine even though it's working fine ??
MikeRichards: you can buy mine for $5000, yea, that's how good it is
not really even going back to the scene from IR and their pre-release 7D it has lots of mazing
but if you don't notice it, don't care, etc. of course keep it
I don't know too much at all about that organization but it really sounds to me like they are complaining a lot about the mazing issue in their review although they don't call it mazing. They mention seeing unexpectedly large amounts of noise at lower ISOs, surprisingly more in the upper tones and often taking on a patterned sort of look to it. They pass it off to too many pixels on an APS-C sensor (this last part I don't agree with since that itself doesn't cause this).
(to be fair, i'm not sure why he harps so much about the auto 19 point selection AF since I would tend to doubt most 7D users will be the type to jsut point at a scene and hope the camera picks out the subject for them, i mean i guess it is nice for the times you quickly hand over the camera for a friend to a take a quick snap of you or the group but he makes it sound like a huge deal)
jorkata wrote:
By that logic, the 7D should be sharper than the 5DII when comparing 100% crops, no .
Sorry - I meant output resolution, not absolute resolution. I'll rephrase it as follows: people don't understand why comparing 100% crops favours the camera with fewer pixels.
brainiac wrote:
Sorry - I meant output resolution, not absolute resolution. I'll rephrase it as follows: people don't understand why comparing 100% crops favours the camera with fewer pixels.
jorkata wrote:
For those who think it's unfair to compare 1.6x to FF, take a look at these test shots of the 40D and the new G11.
The G11 has a 9x smaller sensor than the 40D.
Yes. Imagine a G11 type sensor as big as the 40D sensor. It would have 85 Mpixels. You would need good lenses to see 85 Mpixels of detail, but noisewise it would blow the 40D away, even while the luddites scream "yikes, small pixels, run away, great image quality like that is unworthy".
keithreeder wrote:
"...per unit of sensor surface area", Richard?
...when comparing equal sized sensors.
I don't understand why people struggle with this.
Sensor size, and new technology are the two factors in combatting noise. Pixel density doesn't matter much. To compare two cameras, regardless of their absolute sensor sizes, take the same picture with each, then uprez the smaller file to the dimensions of the larger, and compare away.
The interesting point about noise/detail per unit area of sensor is it gives you an idea of how two sensors WOULD compare IF THEY WERE THE SAME SIZE. It's a thoroughly misleading measurement of comparitive image quality if the sensors are not the same size because a crap 10x8 inch sensor can take pretty amazing pictures, even if it has high noise per unit area.
brainiac wrote:
I'll rephrase it as follows: people don't understand why comparing 100% crops favours the camera with fewer pixels.
The 7D has fewer megapixels than the 5DII (18 vs 21).
Comparisons are showing that it's softer than the 5DII at the pixel level - when these comparisons are supposed to favor the 7D.
So, why is the 7D still softer? Not talking about noise here.
brainiac wrote:
Yes. Imagine a G11 type sensor as big as the 40D sensor. It would have 85 Mpixels. You would need good lenses to see 85 Mpixels of detail, but noisewise it would blow the 40D away, even while the luddites scream "yikes, small pixels, run away, great image quality like that is unworthy".
...and note that the reason that an APS-C version of the G11 sensor would blow away the 40D is because it is newer and better technology.
.