Canon 5D Mark III Light Leak?

[Alker]

Been testing inside the house (23:00 hours and pretty dark here)

With the VF covered and Lenscap on => Same issue as in the movie
With the VF covered and no Lenscap on => No issue
With the VF covered and a very very very strong led beam beamed directly on the LCd => No issue.

So in short....no issue (for me) without the lenscap.
Strange enough I never leave the lenscap on.

 

[Viggo]

Been testing inside the house (23:00 hours and pretty dark here)

With the VF covered and Lenscap on => Same issue as in the movie
With the VF covered and no Lenscap on => No issue
With the VF covered and a very very very strong led beam beamed directly on the LCd => No issue.

So in short....no issue (for me) without the lenscap.
Strange enough I never leave the lenscap on.

Yup, same here.. I'll try in the sun tomorrow.

 

[DavidRiesenberg]

Just tested in myself. With the lenscap on the issue is there (~2/3 stops) but uncovering the eyepiece has a much stronger impact (~2 stops).
Without the lenscap there is no problem, even if bombing it from up close with a strong flashlight.

 

[NAshby]

I tried this out on my 7D and didn't have an issue, although my friend said he had this happen with this 5D II and 5D II.

 

[Dylan777]

Been testing inside the house (23:00 hours and pretty dark here)

With the VF covered and Lenscap on => Same issue as in the movie
With the VF covered and no Lenscap on => No issue
With the VF covered and a very very very strong led beam beamed directly on the LCd => No issue.

So in short....no issue (for me) without the lenscap.
Strange enough I never leave the lenscap on.

Yup, same here.. I'll try in the sun tomorrow.

I'm California, right now is 2:17PM, temp at 82F, very sunny.

* Covered VF, with 50mm lens attached - NO ISSUE

 

[sky87]

I've just tested this with my mk3 and I have the same problem.
Also did a quick test with the lens attached and pointed it to the lamp, half pressed the shutter to focus and take a reading then shone a flashlight against the lcd and that affected the exposure.
Will do some proper tests in the daylight and see how it differs.
Its causing slight concerns with me, lets see if Canon do a recall. I'd hate to be without my camera!

 

[Alker]

Been testing inside the house (23:00 hours and pretty dark here)

With the VF covered and Lenscap on => Same issue as in the movie
With the VF covered and no Lenscap on => No issue
With the VF covered and a very very very strong led beam beamed directly on the LCd => No issue.

So in short....no issue (for me) without the lenscap.
Strange enough I never leave the lenscap on.

Yup, same here.. I'll try in the sun tomorrow.

I'm California, right now is 2:17PM, temp at 82F, very sunny.

* Covered VF, with 50mm lens attached - NO ISSUE

I'm California, right now is 2:17PM, temp at 82F, very sunny.

Lucky you......
Sometimes I have to wait days/weeks for a day with sunshine ....

 

[prestonpalmer]

I've just tested this with my mk3 and I have the same problem.
Also did a quick test with the lens attached and pointed it to the lamp, half pressed the shutter to focus and take a reading then shone a flashlight against the lcd and that affected the exposure.
Will do some proper tests in the daylight and see how it differs.
Its causing slight concerns with me, lets see if Canon do a recall. I'd hate to be without my camera!

This "issue" is exactly the same on my 5D2. Canon will not be doing a recall. I have taken tens of thousands of amazing wedding pictures with my 5D2's with the same aparrent problem. As far as I am concerned, this is a non issue with my 5D3.

 

[Dylan777]

Been testing inside the house (23:00 hours and pretty dark here)

With the VF covered and Lenscap on => Same issue as in the movie
With the VF covered and no Lenscap on => No issue
With the VF covered and a very very very strong led beam beamed directly on the LCd => No issue.

So in short....no issue (for me) without the lenscap.
Strange enough I never leave the lenscap on.

Yup, same here.. I'll try in the sun tomorrow.

I'm California, right now is 2:17PM, temp at 82F, very sunny.

* Covered VF, with 50mm lens attached - NO ISSUE

I'm California, right now is 2:17PM, temp at 82F, very sunny.

Lucky you......
Sometimes I have to wait days/weeks for a day with sunshine ....

If this is a "REAL" problem - I can bring my down to Canon Service Center, Irvine - CA. Only 15mins driving for me.

STILL let's hope this is NOT a major problem for 5D III

 

[AvTvM]

my 7D does not show this problem.

it cannot be seen clearly, -> http://photorumors.com/2012/03/02/whats-inside-the-guts-of-the-canon-eos-5d-mark-iii-camera/ but there seems to be a huge gaping hole in the magnesium chassis of the 5D III underneath the senseless top-deck LCD.

I hope the problem turns out to be real and leads Canon to just abolish the totally unnecessary analogue-days-artefact top-LCD on all future digital cameras. That would be the best, easiest and CHEAPEST solution.

 

[Alker]

Canon to just abolish the totally unnecessary analogue-days-artefact top-LCD on all future digital cameras

Unnecessary ?
For you maybe, NOT for me.

 

[Arkarch]

Yeah really,

I use the top LCD frequently.

Anyway, went outside in Phoenix AZ around 3:00 PM, no issue.

I suppose in the most light-tight situation, lens cap on, covered eyepiece, etc, maybe the LED could be a faint light source.

Someone mentioned tripod placement - yes I would agree. Even slight changes of aim during a test like this may change the metering.

 

[mmaher92037]

My 5D III exhibits the "light leak" behavior when using both the LCD light and a flashlight. My 7D doesn't act the same way-- its meter is steady with and without the light.

 

[RLPhoto]

This is why patience in adopting new equipment is always benifical. Impatience is an expensive commodity.

 

[Smith]

It seems that this might be due to manufacturing variations instead of a "design flaw." Some people can reproduce this effect on both the 5DII and 5DIII while others cannot . There also seems to be a lot of variation to how much the exposure is affected. This is important for various reasons:

1) Manufacturing variations are much harder to fix than large obvious flaws which uniformly affect everyone. There might be some improvement as the manufacturing process matures but there's no guarantee. There will be no recall for something like this as Canon will just fix it under warranty.

2) There's no need to debate whether or not this is a "non-issue." For most people it isn't a problem. However, some people may have it worse than others. Based on how people are using the camera it may or may not be a problem for them. The fact that it doesn't affect you doesn't mean it's not a problem for someone else.

3) If later units have less of a problem than earlier ones it will depreciate the value of the camera for people who buy now. Nobody is going to want to buy a 5DIII from the "faulty early batches."

This issue reminds me of the light leakage problem on the Nikon 24-70 f/2.8. Although this really wasn't an issue for most people Nikon owned up to the problem and addressed it. Let's hope Canon takes the same action. IMO although it may not be an issue for many people it doesn't look good for your brand. This is a $3500 DSLR. At this price people expect a trustworthy tool that they can rely on. Intel had the division bug in the Pentium that would not affect most people. However, it just made people feel uncomfortable with trusting their results and that alone was a PR nightmare. Canon management needs to get in front of this issue and address it or the Internet community will create its own ugly conclusions.

 

[canikon]

I have similar situations of under exposure when I walked around a hotel to shoot in an event, the light sources are varied, so it caused high rate of under exposures. my backup camera is nikon d7000, it seems working as normal as usual. my 5d3 has leakage condition as the forum reported. I did the similar tests. So I suspected the link between the LCD leakage and under exposures.

 

[canikon]

more than agree with your statement. I will return my unit within the 30 day period and wait for the next batches. hopefully canon will recall or fix the issue as nikon did in 24-70.

It seems that this might be due to manufacturing variations instead of a "design flaw." Some people can reproduce this effect on both the 5DII and 5DIII while others cannot . There also seems to be a lot of variation to how much the exposure is affected. This is important for various reasons:

1) Manufacturing variations are much harder to fix than large obvious flaws which uniformly affect everyone. There might be some improvement as the manufacturing process matures but there's no guarantee. There will be no recall for something like this as Canon will just fix it under warranty.

2) There's no need to debate whether or not this is a "non-issue." For most people it isn't a problem. However, some people may have it worse than others. Based on how people are using the camera it may or may not be a problem for them. The fact that it doesn't affect you doesn't mean it's not a problem for someone else.

3) If later units have less of a problem than earlier ones it will depreciate the value of the camera for people who buy now. Nobody is going to want to buy a 5DIII from the "faulty early batches."

This issue reminds me of the light leakage problem on the Nikon 24-70 f/2.8. Although this really wasn't an issue for most people Nikon owned up to the problem and addressed it. Let's hope Canon takes the same action. IMO although it may not be an issue for many people it doesn't look good for your brand. This is a $3500 DSLR. At this price people expect a trustworthy tool that they can rely on. Intel had the division bug in the Pentium that would not affect most people. However, it just made people feel uncomfortable with trusting their results and that alone was a PR nightmare. Canon management needs to get in front of this issue and address it or the Internet community will create its own ugly conclusions.

 

[Fleetie]

Yes, I agree it does look bad for the brand.

I bet there are some very worried people at Canon's DSLR design department today. Brown trousers all round.

I still don't think it's a real-world issue, but it is slightly careless, and a potential marketing/image disaster.

I also predict that this is going to get used as an excuse for poor exposures by those who don't like to admit to making mistakes themselves! "IT was my 5D3 that screwed up!"

(The "Smite" feature is gone, so I can say it!)

 

[kareldonk]

i did some tests with my 7d and here's what I found

case 1: 7d body with 24-70 f/2.8L, LENS CAP ON, VF BLOCKED, DARK ROOM WITH PC MONITOR ON.
in Av mode or P mode, the shutter speed changes (3"2 to 2"5 and back) when I half press the shutter button and change the orientation of the camera a little.

case 2: A DIFFERENT 7d body with 50mm f/1.2L, LENS CAP ON, VF BLOCKED, DARK ROOM WITH PC MONITOR ON.
in Av mode or P mode, the shutter speed changes (3"2 to 2"5 and back) when I half press the shutter button and change the orientation of the camera a little.

The LCD backlight doesn't affect the shutter speed/exposure when on. but why is the exposure fluctuating that much?

 

[smithy]

I've tried this test on my 40D and 1V, and neither has the 'issue'.

The LCD backlight doesn't affect the shutter speed/exposure when on. but why is the exposure fluctuating that much?

You need to take the lens off and just put the cover on the camera body. Lens caps are designed to protect the lens, not block out 100% of light.