5D4 Sensor Defect Discovered

[tron]

I had to try since this would save both of us the frustration of a second failure.

The reason there are not many returns - and I am not in a position to know seller's internal details - is that specific tests have to be done to discover the issues. Also it's a new camera and we do not know if they had many orders (this is also seller internal details).

 

[Donkey_Kong]

You people have crossed the line into insanity. I own a 5DIV and I'm willing to bet I use it a lot more than the majority of you do. There is no problem at all with my unit. It makes stunning images, and it is a pleasure to shoot with. This talk of pushing images 5+ stops? Cmon guys, the time has come to salvage what's left of your unfulfilling lives.

 

[tron]

You people have crossed the line into insanity. I own a 5DIV and I'm willing to bet I use it a lot more than the majority of you do. There is no problem at all with my unit. It makes stunning images, and it is a pleasure to shoot with. This talk of pushing images 5+ stops? Cmon guys, the time has come to salvage what's left of your unfulfilling lives.

As you rightly said: YOUR UNIT. Also read some entries where 2 or 3 stops are reported. Not 5....

 

[Fatalv]

You people have crossed the line into insanity. I own a 5DIV and I'm willing to bet I use it a lot more than the majority of you do. There is no problem at all with my unit. It makes stunning images, and it is a pleasure to shoot with. This talk of pushing images 5+ stops? Cmon guys, the time has come to salvage what's left of your unfulfilling lives.

The only thing that's apparently crossed the line into insanity is your logic and arrogance. Just because YOUR copy doesn't experience any issues doesn't negate the possibility that a portion (however small or large) are exhibiting problems. On the previous page of this thread there is clearly a banding issue with an image that's pushed two stops.

Regardless, since you 'use it a lot more than the marjority' of us, why are you here instead of out shooting? :

 

[Memdroid]

Can somebody please explain to me why you would under expose for 5 stops (blackness!!!) and push the damn thing 6 stops? How is this related to photography and have artistic merit?
It is like trying to get to your top speed on the first gear. Crazy!
Basically looking for a problem where there is none.

 

[ZachOly]

Can somebody please explain to me why you would under expose for 5 stops (blackness!!!) and push the damn thing 6 stops? How is this related to photography and have artistic merit?
It is like trying to get to your top speed on the first gear. Crazy!
Basically looking for a problem where there is none.

It’s not that crazy when you think about it.

You have a 24-70 2.8 II, shooting it wide open at 2.8 and 24mm. It vignettes pretty good.

Now, let’s say you’re shooting with that lens to take a landscape/portrait/whatever with lot of sky (ex. on a beach around sunset).

You’re also in AV mode and have zeroed out exposure compensation (say you forgot to make an adjustment).

All that sounds totally reasonable, right?

Well guess what, in my experience, the camera will underexpose the person by ~2 stops to protect the highlights in the sky. To remove the vignetteing, you’ll need another 1+ stop recovery. So now you’re at a ~3 stop push which is right around where that banding (on some) 5D4s becomes present.

 

[privatebydesign]

Can somebody please explain to me why you would under expose for 5 stops (blackness!!!) and push the damn thing 6 stops? How is this related to photography and have artistic merit?
It is like trying to get to your top speed on the first gear. Crazy!
Basically looking for a problem where there is none.

It’s not that crazy when you think about it.

You have a 24-70 2.8 II, shooting it wide open at 2.8 and 24mm. It vignettes pretty good.

Now, let’s say you’re shooting with that lens to take a landscape/portrait/whatever with lot of sky (ex. on a beach around sunset).

You’re also in AV mode and have zeroed out exposure compensation (say you forgot to make an adjustment).

All that sounds totally reasonable, right?

Well guess what, in my experience, the camera will underexpose the person by ~2 stops to protect the highlights in the sky. To remove the vignetteing, you’ll need another 1+ stop recovery. So now you’re at a ~3 stop push which is right around where that banding (on some) 5D4s becomes present.

NO.

 

[tron]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

 

[privatebydesign]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

Yep, and some people might have defective cameras, I don't think anybody disagrees with that. But many of the 'tests' being done or scenarios suggested as problematic are fanciful nonsense that go far beyond the capabilities of any camera currently available.

 

[Fatalv]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

You give people too much credit. They see the title, grab the pitch forks, and forget to read the thread

 

[tron]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

You give people too much credit. They see the title, grab the pitch forks, and forget to read the thread

+1 Plus they don't have the specific camera ;D

 

[takesome1]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

Not as unreasonable as thinking Canon service would start opening boxes of new camera bodies and test for a fault that only a few people have reported.

 

[tron]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

Not as unreasonable as thinking Canon service would start opening boxes of new camera bodies and test for a fault that only a few people have reported.

Who said about Canon service? Do you even read the posts? I was talking about the seller!

 

[takesome1]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

Not as unreasonable as thinking Canon service would start opening boxes of new camera bodies and test for a fault that only a few people have reported.

Who said about Canon service? Do you even read the posts? I was talking about the seller!

Even more unreasonable

 

[tron]

Hasn't anyone read about the -3EV and even -2EV stops mentioned earlier in this thread? Are these unreasonable?

Not as unreasonable as thinking Canon service would start opening boxes of new camera bodies and test for a fault that only a few people have reported.

Who said about Canon service? Do you even read the posts? I was talking about the seller!

Even more unreasonable

It's better than having to get back a second camera or return the money...

 

[hthieu17]

Hello guys, I’m new here. I actually signed up for an account because I also wanted to share with everyone about the problem of Canon 5D Mark IV. I have known and read about the defect for a while, but I didn’t have time to check it. So I finally gave it a try.
The lens I used for the test was Canon 24-70mm f2.8 L II. At the beginning, I switched to MF at 24mm and adjusted other specifications to f22, ISO Low 50, 2 minutes. Then, I covered the camera with my jacket to make sure it was around 95% in the dark.
After transferring the file to the laptop, I converted it with DNG converter. I pulled Exposure to +1, and +100 for Highlights, Shadows, Whites and Blacks. The final result is this. What the heck are those streaks?

 

[TudorD]

The "defect" can be reproduced in extreme conditions with any camera but on 5DMiv the noise pattern is colored and linear. The experiment was done with extreme negative exposure compensation, aperture closed:
- 5DMiv f22, ISO100, 30sec, exposure -5
- 7D f22, ISO200, 10sec, exposure -5
- Sony RX100 f11, ISO 100, 3,2sec, exposure -3

and "amplified" with +5 exposure and 100%shadows recovery. Sony underexposure was limited by camera capabilities and the obtained noise pattern is mostly white.

After a month of 5DMiv and one week of tests it is for the first time I can obtain this colored noise pattern.

 

[Mikehit]

What the heck are those streaks?

Well seeing as you found this thread have you reading all of it?

 

[Alex_M]

it almost sounds as if: the longer the exposure ( 2 min. long exposure in this case) the less shadow push is required for the phenomenon to present itself?

 

[Mikehit]

it almost sounds as if: the longer the exposure ( 2 min. long exposure in this case) the less shadow push is required for the phenomenon to present itself?

Yep, that is because the longer exposure is needed because of lower light levels which means lower signal to noise ratios which means more noise in the images. If you have to have the shutter open for 2 minutes the light is really, really low. Add to that the longer the shutter is open the more thermal noise there is in the sensor which adds to the problems.

Which is why shooting a sunset scene at 1/20 sec and lifting 5 stops can be quite different to putting your camera under a coat and exposing for 2 minutes and lifting 5 stops.

And all this is why some recommend that rather than shooting at ISO1600 for X seconds it is sometimes better to shoot at ISO 3200 for X/2 seconds. Or, if feasible, shoot at ISO 3200 for X seconds to overexpose (to increase signal to noise ratio) and bring it down in post processing.
It is one thing to identify a problem (anyone can do that), another to know how the camera works so you can work within the camera's limitations.