5D4 Sensor Defect Discovered
- Thread starter LSXPhotog
- Start date
L
Loswr
Guest
JukkaS said:
Your conviction to your prior statements is sadly lacking.
JukkaS said:
So, those who do not see the problem you see either in their own 5DIV cameras or in RAW files posted by others, or those who disagree with you, are exhibiting 'sick behavior'?
JukkaS said:
If what you have to say is rude and insulting, not speaking would be preferable.
JukkaS said:
Hopefully you'll be able to hold to your commitment this time.
Upvote
0
Some facts after additional shooting . Same lighting, same composition, same settings. My 5D4 is better than my 5D3 except where there is horizontal color banding in deep shadows. In the same case 5D3 is not showing them but is is way worse in noise. My 5D3 with Magic Lantern (Dual ISO setting 100/1600) is way better than 5D3 and shows no color bands so it is more usable than 5D4. Tests have been made at ISO 100 for both cameras. The settings have been chosen according to histogram so as to shoot ETTR as possible without clipping.
I have done some additional tests at ISO 50 and shadows improve but it seems that highlights are clipped faster (less DR?) Since I set ISO 50 I should be able to double the exposure. It proved that I couldn't without clipping so some exposure shots have been made by decreasing exposure by -1/3 and -2/3 (relative to the theoretical new double exposure).
There is a CR3 that 5DV will solve these issues.
There is a CR2 that possibly ML will run in 5D4 in about 2 years from now.
There is a CR1 that Canon will improve the situation via firmware.
Take your pick.
In the meanwhile I again have to decide to keep or to return it. It seems that 5D4 is suitable for most of the shooting I do except of a specific one for which 5D3 with ML has produced some excellent keepers. I may chose in between and keep one of my 5D3 cameras. The 5D4 could be excellent for birds and general use.
I have done some additional tests at ISO 50 and shadows improve but it seems that highlights are clipped faster (less DR?) Since I set ISO 50 I should be able to double the exposure. It proved that I couldn't without clipping so some exposure shots have been made by decreasing exposure by -1/3 and -2/3 (relative to the theoretical new double exposure).
There is a CR3 that 5DV will solve these issues.
There is a CR2 that possibly ML will run in 5D4 in about 2 years from now.
There is a CR1 that Canon will improve the situation via firmware.
Take your pick.
In the meanwhile I again have to decide to keep or to return it. It seems that 5D4 is suitable for most of the shooting I do except of a specific one for which 5D3 with ML has produced some excellent keepers. I may chose in between and keep one of my 5D3 cameras. The 5D4 could be excellent for birds and general use.
Upvote
0
L
Loswr
Guest
tron said:
ISO 50 is an expanded ISO setting. That means digital gain is being applied (negative gain, in this case). In other words, when you set ISO 50, your camera is really exposing at ISO 100, then pulling the exposure down a stop. A highlight that's blown at ISO 100 will still be blown at ISO 50, and if you keep all else the same you actually lose a stop of highlights (as you found out). Although the ISO 50 setting may have some utility for jpg shooters (enabling a slower exposure or wider aperture), for RAW shooters there's really no point in using it.
Upvote
0
I remember you saying something similar in the past. Now I verified it in practice since this was a controlled test. But still shadows seem improved. This improvement shows on 5D3 as well. I realize it may be in expense of a little DR or maybe because it pushed the histogram further to the right. Since I took care to push it as much as possible without clipping at iso 100 I have a feeling that somehow iso 50 with a -1/3 correction results in a compression that may cause a slight loss of DR but improves shadows at the same time. What is your opinion on this?neuroanatomist said:
EDIT: You mentioned a loss in highlights of course but I was wondering if maybe this loss is spread somehow in the full range.
Upvote
0
Mikehit said:
What would be needed? A light or lights with controlled and stable luminance. A scene with controlled and stable color and reflectance, along with controlled and stable geometry (for example, shadow lines must not change). These controls over the scene must extend to the whole volume of the room, so it is probably best to construct a standard box to contain it all. To standardize this system (because anyone and everyone needs to be able to replicate the results), allow for another couple hundred pages on this thread. Better yet, assemble a few concerned citizens in an airport hotel somewhere and do a pretend ISO committee. You will blow many times the cost of a 5DIV. I am not making fun.
The facts presented in this thread clearly point to the effect existing within the near zero portion of the signal. Getting precise in that zone costs money. What would you really gain? A precise understanding of how suboptimal the 5DIV is at astrophotography? For a general purpose camera, with capabilities probably centered on wedding photography, I cannot advise expending much effort to quantify that.
For those who think that Neuro is being rude, remember where he comes from. Within a laboratory where one is expected to draw firm conclusions based on data, mucking around in the near zero range of a signal is just plain wrong. Major ethical violation kind of wrong. People die kind of wrong. When faced with such a scene, one must add light. Sunshine is a powerful disinfectant, after all. His failure to adequately contextualize that viewpoint has lead to some unnecessary battles with jrista, but such is the risk of text based societies.
Upvote
0
L
Loswr
Guest
tron said:
My opinion is that you're exactly correct. Because ISO 50 is pulling the exposure down, noise is reduced. With RAW, you can just pull down an ISO 100 exposure in post, and I suspect you'll get slightly more reduction in noise vs an in-camera jpg, because of better processing algorithms. My description of it as 'useless for RAW' is based on that – ISO 50 does nothing you can't do at least as well, and probably better, compared to shooting ISO 100 with appropriate exposure settings and post-processing.
RE spreading it over the range, sort of. There is a tone curve applied when the exposure is pulled down (either by the camera or by the RAW converter). In a way, ISO 50 is the opposite of HTP. With the latter, the camera underexposes by a stop (so setting ISO 200 with HTP is really an ISO 100 exposure, and that's why you can't set ISO 100 with HTP enabled); the reduced exposure preserves a stop of highlights, and the camera boosts the lower luminance range with a tone curve. In both cases, IMO, you can get better results with more control by making the adjustments yourself, rather than using the default tone curves. But it's more work that way.
Upvote
0
retroreflection said:
+1E3
There may, in fact, be a problem in some 5D4 sensors; however, as the saying goes, the plural of "anecdote" is not "data." Seeing bands when pushing near-zero data is to be expected, and its absence would be a surprise.
Upvote
0
Orangutan said:
Just a question, why would bands be expected? Theres no bands on the 5d3. I think the 5d3 is worse overall but the degradation noise is random with heavy pushes. If the streaks were random it would be much better.
I see FW 1.03 is now out also. It will be interesting to see if there are further improvements.
Upvote
0
Your comments are well taken and illustrate the problems inherent in internet discussions on technical matters where you have no idea of the competence of the person making the claims.retroreflection said:
retroreflection said:
I pretty much agree with you.
Upvote
0
And what white balance settings should be used: deep shade or day light? 
sorry I cannot resist... oh, dear
sorry I cannot resist... oh, dear
ZachOly said:RAW "lens cap" file from FW 1.03, please
Use the following settings:
f/8, 20s, tripod, liveview, 2sec delay, ISO 100, dark room, covered eye piece.
Upvote
0
PixelTrawler said:
A few pages ago I posted a demo of a 16-bit blue gradient that I generated in Photoshop and then pushed the "exposure." It showed bands in the shadow area. I gave instructions to replicate it. For an explanation, see the Wikipedia article on posterization. In the bottom few bits of an image (dark areas) there is so little gradation that pushing the "exposure" will cause posterization, which shows as bands in some images.
Why not on 5D3: I can only speculate, but my guess would be the greater shadow noise on 5D3 turns the bottom few bits into a mush that does not display posterization as much when pushed.
Try this experiment: set up your 5D4 and 5D3 to capture a test image that should display the problem. Be sure to use a zoom lens rather than a prime. Find the right exposure to create the problem, then dial it in manually on both cameras. Now take 2 photos with each camera. Each set of 2 will have the same exposure, but the FL changed by just a few millimeters. The FL change should be enough to recognize, but not enough to make a dramatic difference to the overall image. Now push the exposures on all 4 images identically. Look at the two 5D4 images: are the bands in the same spot of the scene or the same spot of sensor? If it's the scene (i.e., they've moved relative to the sensor) then you have posterization. If they're in the same position in the frame then you might have a sensor-related phenomenon, though it may be a limitation rather than a defect. Now compare to your 5D3. Do you have the same image details at both ends of the DR? Same bands?
I'm not saying (and can't say) that there are no 5D4's with sensor problems; what I am saying is that pushing 5 stops will give you posterization in deep shadow areas.
Upvote
0
Similar threads
