[SOLVED] R5 firmware 1.4.0. - inconsistent light metering/histogram

[grztus]

The problem has been solved, see post #12.

I have upgraded the firmware from 1.3.1 to 1.4.0 and voila! The problem is here... Canon has broken something in light metering in still photos mode.

I'm using my R5 on the telescope (no aperture and focus info) for contrails spotting, like this one for example (taken about two weeks before upgrading the firmware):


But after upgrading to 1.4.0 I'm not able to do that any more! And what's worse - I'm not able to downgrade the firmware to 1.3.1 because Canon doesn't want me to do that... what a mess.

The problem is somewhere in the firmware definitely:
on the screen or EVF we have our picture with exposure simulation, the histogram shows everything OK, but after taking the shot, the picture is completely over exposed. When I put the under expose correction (for example full -3EV) to move a histogram completly to the left, after the shot it is completly to the right side - it is impossible to take a good exposure shot.

The problem is only in still pictures mode, in the movie mode everything is OK.
Also there is no problem in still pictures mode when using RF lens.
So the inconsistent exposure is only with let's say - third party optics, like my telescope. If it was mechanical failure of metering system, it would be in all modes and scenarios - here it isn't.

I've tried to reinstall the 1.4.0 firmware - no change (yes, you can reinstall the same version of FV, but you cannot downgrade it to the last known working).
I've switched beetwen all the still modes available (P, M, A, S and so on) - no change.
I've switched beetwenn CF and SD cards - no change.
I've switched beetwen RAW, JPG, DualPixel RAW - no change.
I've reseted all the camera settings - no change.
The problem is even in full Manual mode - I'm chosing the correct settings for exposure, and the picture is still very over exposed.
I'm talking of about 15-20 EV stops! Full spectrum of histogram from left to right!

I really don't have any more ideas to find the solution.

CANON - please repair it immediately! And give - us - photographers - the choice to downgrade the firmware when you broke something in the new version!

By the way - there is also some issue in timelapse mode with electronic shutter - camera is taking 2 - 3 shots, instead of one. But this bug is for some other topic.

 

[Czardoom]

CANON - please repair it immediately! And give - us - photographers - the choice to downgrade the firmware when you broke something in the new version!

By the way - there is also some issue in timelapse mode with electronic shutter - camera is taking 2 - 3 shots, instead of one. But this bug is for some other topic.

Have you contacted Canon? If not, that is who you should be contacting to let them know and to see if there is a possible solution.

 

[tron]

Maybe EOS Utility allows you to downgrade. At least this is being mentioned for Canon DSLRs.

 

[grztus]

Thank You, EOS Utility is not working with firmware downgrade on EOS R5.
I have contacted Canon Poland and waiting for their reply.

I have also tried remote shooting, still the same problem.
The main picture is downloaded final photo with histogram showing overexposure, and the small picture in top-right corner is prt screen from live view at the time of taking this picture (histogram showing underexposure).

 

[YuengLinger]

Have you considered the possibility that the problems have nothing to do with the firmware upgrade? That they started at or around the same time you upgraded?

I upgraded my R5 to 1.40 within a few days of the fw release. I've had no issues with Sigma or Canon lenses.

What kind of adapter are you using to fit the telescope to your RF mount? Do you have another camera to test the telescope on, to make sure no problem cropped up with it?

Did you try turning off (or on, as the case may be) EXPOSURE SIMULATION?

Finally, I'm not sure it is correct to say that Canon doesn't want us to revert to older firmware versions. You make it sound as if they have some selfish agenda. It could be that they have data indicating allowing reversion causes many more serious problems than it could potentially fix.

I hope you find the causes of your seemingly rare, if not unique, problem with the R5.

 

[Kit.]

If it overexposes when shooting RAW in fully manual mode, it is highly unlikely to have anything to do with the firmware.

 

[EricN]

What telescope do you use?

 

[grztus]

The telescope is 16" Newtonian, it has 1800mm focal length f4. I connect it via eos-rf adapter, but it has nothing to do with the problem, when the adapter is detached from body (taking pics without anything on body) still the final image is very much overexposed relative to monitor or EVF view and histogram displayed live view.

Of course I have checked exp sim, but the problem like I say is a little different, even the histogram on the screen shows correct exposure, but the final image is totally overexposed. It is in all photography modes, auto iso, manual iso and so on. I really check everything possible with total reset of camera and reinstalling fw 1.4.0 three times.

I didn't change anything else, just firmware, and the problem began, that's why I'm pretty sure that this is the cause. But I cannot confirm this, because I cannot downgrade the firmware to the previous working version.

Because it is only in photo mode without RF lense (in video mode everything is OK even with telescope attached) I don't think it's mechanical problem of light metering system or any other part. I will try to make a short movie and paste it on YT to present what I'm writing about.

Canon Poland has forwarded my submission to Canon Europe. Still waiting for contact.

 

[grztus]

I upgraded my R5 to 1.40 within a few days of the fw release. I've had no issues with Sigma or Canon lenses.

Sigma or canon are communicating with body (af, aperture). In my case the telescope mount doesn't have any pins so there is nothing to communicate. When I bought R5 I was forced to set in menu to make a pictures without lense, because shutter didn't release.
I've made about 15k pictures thru this telescope with R5. I have upgraded FW one evening, and since next morning I had this issue.

 

[YuengLinger]

Sigma or canon are communicating with body (af, aperture). In my case the telescope mount doesn't have any pins so there is nothing to communicate. When I bought R5 I was forced to set in menu to make a pictures without lense, because shutter didn't release.
I've made about 15k pictures thru this telescope with R5. I have upgraded FW one evening, and since next morning I had this issue.

After your patient, complete explanations, I understand it is frustrating for a highly advanced camera to have problems with an adaptor that, simple as it may be, was working before the fw update. I hope Canon can resolve this for you, or you find a workaround.

We've become accustomed, over the years, to more and more features and capabilities. The simplicity some might remember in strictly mechanical bodies from decades ago did have some advantages.

 

[grztus]

I've made some more tests today, the problem is still, but just for check:
1. I've changed the languages - no improvement
2. I've checked the EF-RF adapter with canon 50/1.4 - works OK both in photo and video (like I wrote yesterday - when there is a communication, there's no problem)
3. I've checked what is going on, when the lense is not fully connected: both RF 24-105 and 50/1.4USM have the issue.

Conclusion - when there is no info about aperture (F 00) there is an issue. You can check it in your bodies with FW 1.3.1 and 1.4.0 and report here in topic what's your obserbvations. You just need to loose the lense and rotate it a little (the pins cannot communicate with body).

If you have FW1.4.0 and not observing the issue, please save your R5 parameters (you can save them to the card) and send me (zip file or something), I will try tou use it and check if it helps.

 

[grztus]

Yes, Yes, Yes! I've found the solution !
Canon Poland and Canon Europe couldn't help me, but I was shore the problem is in aperture information.
I've bought the dandelion chip and it helps, but later I've found the cause of the problem, which is the menu card with minimum and maximum aperture setting.
When minimum aperture is set for example f 4.0 the problem is here. But when the minimum aperture is not set (f 1.0) everything works fine. Maximum aperture do not have influence for the issue so it is not important.

 

[YuengLinger]

Yes, Yes, Yes! I've found the solution !
Canon Poland and Canon Europe couldn't help me, but I was shore the problem is in aperture information.
I've bought the dandelion chip and it helps, but later I've found the cause of the problem, which is the menu card with minimum and maximum aperture setting.
When minimum aperture is set for example f 4.0 the problem is here. But when the minimum aperture is not set (f 1.0) everything works fine. Maximum aperture do not have influence for the issue so it is not important.

Great troubleshooting! Great persistence! Congratulations!