Shows this mazing artifact? It's not out of a 7D, not even a 50D but a G10. Cause the stuff people were showing seemed very familiar to me so I went looking at my set of photos.
hmmm...Wow, I went to check out the last couple of pages of the 7D Master Thread to see what the latest experiences are with the camera. I have to say that I am a bit shocked that people are actually viewing images at 200, 300, & even 400% to point out flaws that are ruining images! There is NO IMAGE that can be ruined by a detail that is only be viewable at 400%! As another posted stated, "Next time I go eat in a restaurant I'm going to bring a microscope to make sure there is no bacterias at all in my food."
keithreeder wrote:
Could be, but I had to go to 300% to see anything...
This artifact is by no means unique to the 7D or to Canon, and is a known (and apparently manageable) commodity to RAW converter providers.
Exactly. For everyone's info that was converted from RAW with ACR.
I think it's normal. We are talking about a digital image here which if you dig down to it's basics is a bunch of electrical signals, noise and all, coming from each sensor site on the sensor chip. Chips are not uniform things but varies from one edge of a wafer to the other. I tend to think the "banding" issue everyone jumps on is actually just the nature of the chips. You have wires etc running over the top of these doesn't surprise me there are slight variations in the resulting images. Could be as simple as a slight shadow from a power trace on the chip.
ejmartin wrote:
It looks like DPP is producing the artifacts, at a lower level, in skibum's example. As I have always said, the fix is to do local averaging of the RAW data, so the artifacts can be suppressed at the cost of resolution. This is likely also the reason why some have been disappointed by the noise at lower ISO; the noise is a residual artifact of the hardware issues that the RAW converter was unable to filter out.
Agree.
I don’t’ think, though, that the issue limited only to certain (defective) units.
Canon seems to have intentionally made one of the green channels on the color filter more transmissive - as a way to combat noise.
ISO/noise is improved but the side effect are these artifacts that we are seeing.
With DPP the maze effect is not visible.
I’m also sure that the final versions of ACR and LR won’t produce the maze effect either.
However, the green channel imbalance seems to be causing the black & white dots in 7D images (the ‘digital dandruff’, as RG describes it) that no other Canon DSLR has.
After some experimentation with 7D images, my impression is that the black & white dots limit the amount of post-processing that can be applied to 7D images.
Sharpening and enhanced contrast make the black & white dots quite visible and it’s these dots that give the overall ‘harsh’ look to 7D photos.
omarlyn wrote:
hmmm...Wow, I went to check out the last couple of pages of the 7D Master Thread to see what the latest experiences are with the camera. I have to say that I am a bit shocked that people are actually viewing images at 200, 300, & even 400% to point out flaws that are ruining images! There is NO IMAGE that can be ruined by a detail that is only be viewable at 400%! As another posted stated, "Next time I go eat in a restaurant I'm going to bring a microscope to make sure there is no bacterias at all in my food."
Well for something completely different, I'm amazed at the level of AF customizability on the 7D. I have now set it up so I can have 4 different settings for AF at the touch of a button. It's better than the 1D II for AF customization. Now if it works nearly as well in the field I'll be very happy.
Pixel Perfect wrote:
Hey what about those Yankees!
Well for something completely different, I'm amazed at the level of AF customizability on the 7D. I have now set it up so I can have 4 different settings for AF at the touch of a button. It's better than the 1D II for AF customization. Now if it works nearly as well in the field I'll be very happy.
That's a great idea Whayne, now if you just remember which button is which setting.
Sounds like a good idea but I would forget I set it that way.
Currently I have set it up so with * button I have centre AF point chosen, AF-ON I can switch to a registered point which is currently off-centre and with expansion, I can press the DOF preview button and have a specialised AI-servo set-up, and have set C1 for another AI servo set-up. I hope after experimentation, I can cover nearly all situations with a touch of a button or dial. I love the MFn button to allow fast changing of AF area options.
kewlcanon wrote:
Well why not send it to Canon for adjustment ?.
i will although i know with the 5d2 they just replaced the bodies though in the end and were apparently never able to adjust them, very, very few 5D2 had mazing to a really bothersome degree
but it seems like an awful lot of 7Ds have this, maybe even most to this point
so many that it almost seems like they made them this way on purpose, but what purpose would outweigh the issues it causes?
abqnmusa wrote:
Seems rediculous to be examining pictures at 200% or 400%. All camera are going to have artifacts beyond 100% view.
seems ridiculous to talk about noise at all then since the mazing, at best, leaves behind more extra noise than the difference between any modern sensor
abqnmusa wrote:
Why would you return the camera. The problem is not a 7D issue given that DPP does not produce artifacts.
Adobe should have a non-beta Adobe Camera RAW update for the 7D before long.
Otherwise DPP gets the job done
actually DPP shows it too, just not as much and even on cams where it barely shows a mze it still leaves behind extra speckles and noise at low ISO, so DPP doesn't really handle it perfectly either
I don’t’ think, though, that the issue limited only to certain (defective) units.
Canon seems to have intentionally made one of the green channels on the color filter more transmissive - as a way to combat noise.
ISO/noise is improved but the side effect are these artifacts that we are seeing.
With DPP the maze effect is not visible.
I’m also sure that the final versions of ACR and LR won’t produce the maze effect either.
However, the green channel imbalance seems to be causing the black & white dots in 7D images (the ‘digital dandruff’, as RG describes it) that no other Canon DSLR has.
After some experimentation with 7D images, my impression is that the black & white dots limit the amount of post-processing that can be applied to 7D images.
Sharpening and enhanced contrast make the black & white dots quite visible and it’s these dots that give the overall ‘harsh’ look to 7D photos. ...Show more →
the ironic thing if that is the case, that they went to a more transmissive G1 or something, is that the artifacts make most converters, really all, introduce more noise than whatever modest gain a more transmissive G1 might give I'd think!
At best it seems to help maybe 15%, assuming that is even the factor that has boosted photon capture efficiency which is anything but certain, while the extra noise is much more than 15% so....
in fact if you look at very mazing prone patches you can notice that DPP definitely shows a maze-pattern itself, just weaker and somewhat broken up into dots by the tendency for DPP to get zipper effect
sometimes it almost looks like DPP tries to selectively smooth over areas of low detail with 7D raws, but i could very well be wrong about that
abqnmusa wrote:
Why would you return the camera. The problem is not a 7D issue given that DPP does not produce artifacts.
I have a 5DII that produces superb images. For me, focusing on producing images with the 5DII might yield better results and be a better use of my time. It doesn't matter now, I can't return the camera. So I will cross my fingers and hope for the best when the ACR/LR support arrives and hope Phase One updates v5 of C1. I am a big supporter of Canon and have a lot of respect for the company and their products. I can wait the month or so for better support from the software suppliers.