5D Mk III Light Leak Issue (through top LCD)...

[Remorhaz]

Apparently there is a light leak issue which has surfaced on the new 5DMk3

http://fstoppers.com/news-5dmk3-light-l ... e-surfaces

Multiple people are reporting that there is a light leak through the top LCD panel that directly affects metering. The easiest way to detect this issue is to put the body cap on the camera body as well as cover the viewfinder. Then place the camera in “P” mode at ISO800 and press the LCD backlight button. Shutter speed changes by about a stop when the LCD is on vs off. Some users reporting sunlight or walking into a room with direct overhead light shining on the top panel also affects metering.

[Matt. K]

Leaking out????....................or in???????

[gstark]

Leaking out, obviously.

It's why we used to have darkrooms - to keep the dark in.

[chrisk]

i think it is leaking "in" for this case. ie: more light is affecting the meter from the top of the lcd as opposed to TTL metering. look at the exposure time, it drops from 10" to 5" when the lcd is activated indicating that more light is reaching the meter.

[surenj]

Leaking out, obviously.

It seems to be a leak from within the camera to the sensor. It would be hard to prove that these is no leak 'out' because the problem occurs when the LCD illumination is used WHILE taking the picture.

Leaking out would describe the incontinence (either urinary or faecal) that the Canon exec may be experiencing when they realise they may need to recall at least one batch.

[Wink]

From what I've read about people testing this it seems this issue is a non issue. Just internet hysteria. Quite a few people reported old models did the same thing.

Who shoots with the lens cap on anyway?

[surenj]

this issue is a non issue

Damn. Here I was worrying like hell. Oh well... I am sure we will find something wrong with something or rather.

[Wink]

Don't worry Suren. You're still the most anti-Canon Canon shooter I know.

[Remorhaz]

From what I've read about people testing this it seems this issue is a non issue. Just internet hysteria. Quite a few people reported old models did the same thing. Who shoots with the lens cap on anyway?

I think when people compared to old models they were showing the affect of light coming in the viewfinder affecting exposure through the mirror box (which can and does happen with any SLR camera). The test with the Mk3 in this instance is with both the lens cap on and viewfinder covered just to show it's not either of those sources affecting exposure.

[photohiker]

Is this a real issue, or is it another attack of the measurebators?

Does the camera underexpose in normal circumstances?

[surenj]

You're still the most anti-Canon Canon shooter I know.

To be fair, they've done a great job with the 5d3, radio TTL etc. Even the G1x is nifty.

Does the camera underexpose in normal circumstances?

I just watched the Digitalrev review. Kai didn't seem to complain about it; so it must be ok.

[chrisk]

Is this a real issue, or is it another attack of the measurebators?

Does the camera underexpose in normal circumstances?

It would appear to be a mixture of both.

Based on reports, its not appearing in all circumstances, its only in quite dark conditions and if you have a light source shining above your LCD panel. It's not something thats particularly common but it can occur and having light leaking into your camera and affecting your metering is not cool...especially when you just forked out $4k.

[photohiker]

Fair enough. I agree a $4k camera shouldn't have basic flaws like this, but I'm also having trouble remembering ever taking photos under those circumstances...

Not getting one anyway

[surenj]

Based on the scientific method, these guys tested one camera and concluded it's ok.

http://www.canonrumors.com/2012/04/the-light-leak-update/

[gstark]

I suppose I should add some rationality to this discussion. (Yes, I know ... )

First of all ...

Leaking out????....................or in???????

It seems that most of you might not have noticed the little smiley face in Matt's post? He was joking ...

Leaking out, obviously.

It's why we used to have darkrooms - to keep the dark in.

As was I.



Now, to get to the issue at hand: this is basically an extension of the situation whereby you are shooting in a darker environment, camera mounted on a tripod, set for a long time exposure, TTL metering .... and you are not using an eyepiece blind.

Theoretically - and depending upon the camera and the metering systems it uses - light may leak in through the eyepiece, and affect the meter reading taken by the camera's metering system. Typically, this would lead to a mild case of overexposure, as more light would be registering at the meter than what is coming through the lens.

This problem doesn't occur through normal use of the camera, as usually one of our eyes covers the viewfinder eyepiece; the job of the eyepiece blind is to replace our eye in this instance.

In the case of the 5D3, the demonstration would appear to show a valid potential problem, as, save for a problem with the voltage regulation system causing an issue, turning on the LCD's backlight should not be causing any change in the exposure being registered, and especially so with the viewfinder and lens mount both securely blacked out.

So, if there is light leaking in to the camera body and affecting the meter's reading, then yes, there is possibly a problem.

OTOH, if there is a voltage regulation problem that occurs when the LCD backlight is activated, then again we we have a possible issue. The demonstrations though also show that the problem occurs in sunlight; this kind of eliminates the possibility of a voltage regulation issue and points the finger back at light sealing between the LCD's mount and the exposure metering sensor.

And the general discussions seem to indicate that the issue occurs across a number of instances of the camera: any one instance may or may not demonstrate the issue, but across a broad sampling, if there's a large number displaying the problem, then perhas there may be something that Canon need to attend to.

[Remorhaz]

but I'm also having trouble remembering ever taking photos under those circumstances...

I don't know I reckon I could craft a few not half unreasonable situations...

You're shooting a scene where you're out in open sunlight but your subject is in the shade (you've either cropped in tight using a longer focal length on the shaded subject or you're using spot metering - e.g. shooting a subject in or through a doorway or under a tree, etc). The sunlight on the top LCD +1's your exposure and the camera underexposes your subject.

You're indoors in lower ambient light shooting a subject but you are standing under a light (e.g. downlight) - ditto as above.

NB: The exposure incorrectness thing you could probably learn to deal with - if however the light actually leaks into and onto part of the sensor during the shot this could be another thing. e.g. I'm using a big stopper to turn day to night for a long exposure or I'm taking stacked star trail images and I turn on the top LCD light (or shine a torch onto the top of the camera) to check how I'm going and it foobars that frame

[Wink]

I just tested my 7D (yes, I'm aware this is all about the 5D MkIII) for sh*ts and giggles...

Inside my study with the lens cap ON viewfinder facing the window:
- Viewfinder uncovered: 1/2 sec @ f/3.5 ISO3200.
- Viewfinder covered: 4 sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 4 sec @ f/3.5 ISO3200.

Inside my study with the lens cap OFF viewfinder facing the window:
- Viewfinder uncovered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 1/200th sec @ f/3.5 ISO3200.

[photohiker]

Theoretically - and depending upon the camera and the metering systems it uses - light may leak in through the eyepiece, and affect the meter reading taken by the camera's metering system. Typically, this would lead to a mild case of overexposure, as more light would be registering at the meter than what is coming through the lens.

Hang on, you sure about that Gary? Wouldn't the extra light hitting the meter cause the camera to reduce the exposure and therefore under expose the real scene?

[chrisk]

I just tested my 7D (yes, I'm aware this is all about the 5D MkIII) for sh*ts and giggles...

Inside my study with the lens cap ON viewfinder facing the window:
- Viewfinder uncovered: 1/2 sec @ f/3.5 ISO3200.
- Viewfinder covered: 4 sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 4 sec @ f/3.5 ISO3200.

Inside my study with the lens cap OFF viewfinder facing the window:
- Viewfinder uncovered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 1/200th sec @ f/3.5 ISO3200.

the VF light leaking is common to almost all slr's; as gary already alluded to, your eye/ face typically covers the VF when shooting so its a non issue. not to mention that it is manufactured as a big window so light is bound to affect things getting inside there, thats not a design or manufacturing flaw. that's not the matter for which people are getting overly excited.

although i have my doubts as to the severity of the problem, i doubt there is a fundamental design flaw at all here; however, if there is indeed light leaking into the camera from the lcd panel, this is not a good thing at all.it certainly gives me reason for pause about buying one and i'm not too far off making that decision.

[surenj]

I to one should seriously consider buying a backup body for tricky lighting. Perhaps a 7D. Just use two bodies. Easy.

[aim54x]

I just tested my 7D (yes, I'm aware this is all about the 5D MkIII) for sh*ts and giggles...

Inside my study with the lens cap ON viewfinder facing the window:
- Viewfinder uncovered: 1/2 sec @ f/3.5 ISO3200.
- Viewfinder covered: 4 sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 4 sec @ f/3.5 ISO3200.

Inside my study with the lens cap OFF viewfinder facing the window:
- Viewfinder uncovered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered: 1/200th sec @ f/3.5 ISO3200.
- Viewfinder covered with LCD light on: 1/200th sec @ f/3.5 ISO3200.

Interesting results....wonder why the variance between the lenscap on situation regarding the viewdinder.

[surenj]

Perhaps it's because the sensor becomes more sensitive during less light so more likely pick up smaller changes in light.

[Reschsmooth]

Ok, I am late to the joke party on this, but, it was worth the wait.....this is all Julian Assange's fault. Oh yeah, I went there.

[aim54x]

this is all Julian Assange's fault. Oh yeah, I went there.

[Wink]

Looks like Canon has acknowledged they're looking at this. There's a press release posted on Canon Rumors from Canon US.

[surenj]

“To Users of the Canon EOS 5D Mark III Digital SLR Camera
Thank you for using Canon products.
The phenomenon described below has been confirmed when using the Canon EOS 5D Mark III Digital SLR Camera.
Canon is now examining the countermeasures and once the countermeasures are decided, we will post the information on our Web site.
Phenomenon
In extremely dark environments, if the LCD panel illuminates, the displayed exposure value may change as a result of the AE sensor’s detection of light from the LCD panel.
Affected Product
Canon EOS 5D Mark III Digital SLR Camera
Support
Once the preparations are complete, we will be making an announcement on our Web site.
This information is for residents of the United States and Puerto Rico only. If you do not reside in the USA or Puerto Rico, please contact the Canon Customer Support Center in your region.
Please register the EOS 5D Mark III. By registering, we will be able to notify you via email when service updates are available. If you already registered, please ensure you are opted-in to receive the notification.
Thank you,
Customer Support Operations
Canon U.S.A., Inc
Contact Information for Inquiries
Canon Customer Support Center
Phone: 1-800-OK-CANON
1-800-652-2666
TDD: 1-866-251-3752
Email: carecenter@cits.canon.com
For additional support options: www.usa.canon.com/support”

[chrisk]

How could they screw that up ? I mean come on.

[aim54x]

How could they screw that up ? I mean come on.

,

We can ask but it probably has something to do with bean counters and rushing product to market. Stuff like this happens all the time.....D5000 repair then re-Repair because they found a better fix, mirror faking off the 5cm..Leica X1 with all the rubber falling off...

[devilla101]

I've got a 5D mk III that is exhibiting this issue. However as far as I know it has not affected my exposures and I've used it already on weddings (excellent test) and portrait work. I've taken it around on touristy shots and I haven't really encountered any 'weird' exposure issues that I've already experienced with my 5D mk II.

I shoot mostly Manual and switch to AV and use EC and no have no problems.

Hope this helps