M11 freeze issues [Merged]

[hdmesa]

Advertisement (gone after registration)

15 minutes ago, lct said:

The fact that they’re trying to find a firmware solution is at least evidence that there can be a firmware issue while the faulty hardware theory is based on no evidence unless i'm missing something.

When Leica told me they were working on the freezing issue, I took that to mean they are still trying to reproduce the issue reliably so that can figure out where in the code to look. I wasn’t told this directly, but it sounded like from my emails back and forth with them that once they can reproduce an issue, a firmware fix can be written very quickly, perhaps only a few days, which would of course then have to be tested. 

There are at least two or more freezing issues, those that occur at startup and those that occur during use. They do not appear to be related, but I don’t know. 

Basically, “working on a fix” can mean anything in the process from trying to figure out how to reproduce the issue to knowing how and trying to write the fix into firmware. Given how quickly it sounded like they can write a fix, I think they are still trying to reproduce the issues and find the right combination of conditions that cause freezes. 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[neonvoid]

3 hours ago, IkarusJohn said:

?

Did you misread my post, or were you just looking for something to disagree with?

Probably misread, sorry.

[elmars]

vor 7 Stunden schrieb hdmesa:

Are the beta testers discussing issues in a forum/group setting?

Of course we do! 

[crf59]

So far, knock on wood, I have had no freezing issues at all. Is there some common set of circumstances that cause the camera to freeze? Rapid shooting, leaving it powered on for a long time, using uncoded lenses, high tide? Just such an odd problem. Being an engineer this smacks of a hardware issue. Maybe clock/timing issue. I'm really curious to see what ultimately comes out of this.

Edited June 18, 2023 by crf59

[neonvoid]

22 minutes ago, crf59 said:

Being an engineer this smacks of a hardware issue. Maybe clock/timing issue.

Maybe a rare race-condition the software can't recover from properly, resulting in a lock up.
A possible solution in this case would be to detect this incorrect state and at the very least hard-reboot the camera automatically.
If it's a hardware issue (e.g. some sensors acting up) - it should also be possible to detect via software.

If the issue is faulty processor/mainboard/memory/etc, well, in his case a they need to recall all the cameras and replace whatever need replacing.
 

[Photoworks]

3 hours ago, crf59 said:

So far, knock on wood, I have had no freezing issues at all. Is there some common set of circumstances that cause the camera to freeze? Rapid shooting, leaving it powered on for a long time, using uncoded lenses, high tide? Just such an odd problem. Being an engineer this smacks of a hardware issue. Maybe clock/timing issue. I'm really curious to see what ultimately comes out of this.

Don't you think all the issues have been toned down since the latest update?

My camera became very stable, so I assume it is software issue. I think one more round can fix the remaining random issues.

[hdmesa]

Advertisement (gone after registration)

11 hours ago, elmars said:

Of course we do! 

That’s what I figured, but I didn’t know for sure. If there were no freezing issues at the end of testing, then either the testing pool was too small and/or something changed during production. 

[lct]

2 hours ago, hdmesa said:

That’s what I figured, but I didn’t know for sure. If there were no freezing issues at the end of testing, then either the testing pool was too small and/or something changed during production. 

And/or beta testers had no uncoded lenses available or did not test them under lens detection auto.

[hdmesa]

27 minutes ago, lct said:

And/or beta testers had no uncoded lenses available or did not test them under lens detection auto.

Those of us with freezes at startup ruled that out as a possible cause a while back in the thread. 6-bit, not 6-bit, Leica, Voigtlander, doesn’t matter. That said, it’s hard to believe they wouldn’t have had at least one beta tester with an older uncoded Leica lens. How many beta testers were there, three or four? 🫣

[elmars]

Most likely, we didn't use the camera in a way that would cause crashes. We are very different in the group and use the camera very differently. Nevertheless, the entire possible usage can never be depicted.

I've offered before (without much success) to make my camera settings available to anyone affected by freezes. If freezes stop then, you can systematically find out the differences to your own settings and possibly get closer to the cause.

[lct]

Another help could consist in finding settings that would cause freezes repeatedly if there are any. I have tried to get there several times already, to no avail unfortunately, but my M11 is not prone to freeze apparently. I seem to recall that some unlucky users did try too but i suspect they did not succeed either. In such circumstances, it doesn't come as a surprise if Leica considers freezes negligible, at least statistically. Leica isn't just any camera manufacturer, however. The large sums of money they're asking for should justify their looking for a solution, and if they really are looking for one, they should say so publicly.

[DadDadDaddyo]

The part of the automobile most prone to failure:

 

 

 

the nut that holds the wheel....

[marchyman]

On 6/18/2023 at 4:17 AM, neonvoid said:

A possible solution in this case would be to detect this incorrect state and at the very least hard-reboot the camera automatically.
I
 

That brings back memories. Communications hardware a company I worked for had a watchdog timer built into its devices. If the software didn’t “kick the dog” every so often the timer would expire and reset the device. Sometimes the user didn’t know that had happened, thinking only that the net was slow for a few seconds. It made whatever would otherwise have been catastrophic bugs somewhat bearable until we could track them down and kill them.

Don't know how well that would work in a camera, though. 

[Guest]

On 6/13/2023 at 2:18 PM, DAERASO said:

I can put my money on that broadcast being Leica Miami... I sadly have nothing nice to say about them.

They cannot speak on behalf of Leica ... esp. 'on the record' - in many industries (not just cameras) this is part of the dealership agreement, which they would jeopardize.

I don't think it is fair to put them on the spot. I also want to acknowledge the wealth of YouTube podcasts they have shared over the years. They don't pretend to be journalists, but freely share their daily experience with Leica cameras and lenses. This is a significant contribution to our community.

[Guest]

On 6/17/2023 at 4:16 PM, lct said:

The fact that they’re trying to find a firmware solution is at least evidence that there can be a firmware issue while the faulty hardware theory is based on no evidence unless i'm missing something.

I am deeply dissatisfied with Leica's communications around several aspects of the M11. I support your sentiment you articulated earlier not to buy another M11 until this changes.

I would also add two more nits to pick -

1. The whole marketing myth regarding 'pixel binning benefits' was never substantiated. We didn't see better DR or noise at MDNG or SDNG resolutions. I would have expected a 'binning' solution to result in a 15 MP DNG. I have not seen an explanation how there could be any benefit going from 60 to 36 ...

How is this related? Again, speculation about 'future firmware unlocking the full potential (based on the marketing claims?)' - and Leica never stepping up to clarify and communicate clearly. And no benefit IRL.

2. M11 shows more 'purple fringes' than any other Leica camera I have used.

Similar to 'binning gate', there were friendly people on this forum sharing that 'Leica knows about it, let's wait for future firmware'. Guess what - silence & no follow up from Leica. Again.

Call to action for Leica: It would be amazing to communicate on technical questions regularly, perhaps facilitated by the moderators of this forum (much preferable to a static FAQ on your website, which wouldn't allow for follow up). It would be good to have a check-in every 1-2 months, to ensure that concerns of this community are answered with sufficient technical clarity and credibility. If workarounds to issues are known, please make the experience of owning a Leica better by sharing those that you have confirmed and would recommend (e.g. I have learned to always format SD cards in camera, or else experience serious issues while shooting with M and SL cameras). Let us know if there are known issues. I can only imagine how painful it is for your dealers to make things right on a 1:1 basis, instead of getting some air cover from Leica to the benefit of all Leica owners (... many of us do not even live close to a Leica store).

Edited June 19, 2023 by mzbe

[Kiwimac]

I’m no software engineer or camera designer. 
 

However. These cameras are relatively straightforward in that they do not permit vast customisation like say Sony. 
 

Therefore surely there can’t be that many variations in how they are set up. 
 

Given that, the logical remainder is either a manufacturing fault (unlikely but always possible) and/or a random failure in either one or several components soldered to a board somewhere. 
 

If it’s only firmware, surely every camera could be made to fail? Since it seems that isn’t so, surely component failure is all that’s left?

 

I presume Leica don’t manufacture every board in the camera in house, outsourcing that to specialists? If so, perhaps that’s where the error crept in?

[lct]

1 hour ago, Kiwimac said:

If it’s only firmware, surely every camera could be made to fail?

Every camera used the exact same way yes i guess. But every camera using whatever uncoded lens with dirts, screw heads or misplaced or mispainted coding dots in the way of the 6-bit sensor, i suspect no.

[Kiwimac]

4 hours ago, lct said:

Every camera used the exact same way yes i guess. But every camera using whatever uncoded lens with dirts, screw heads or misplaced or mispainted coding dots in the way of the 6-bit sensor, i suspect no.

So a camera on latest firmware using new coded Leica glass ought to be reliable?

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[lct]

In complex matters where a mix of causes, known or unknown, are at play, it can be good practice to eliminate them as and when they are identified. My feeling, FWIW, is one of these causes is uncoded lenses treated as coded ones under lens detection auto. Firmware update 1.6.1 did nothing against that. A way of eliminating such a possible cause could be to disable or grey out lens detection auto when uncoded lenses are mounted on the camera.

[hdmesa]

3 hours ago, lct said:

In complex matters where a mix of causes, known or unknown, are at play, it can be good practice to eliminate them as and when they are identified. My feeling, FWIW, is one of these causes is uncoded lenses treated as coded ones under lens detection auto. Firmware update 1.6.1 did nothing against that. A way of eliminating such a possible cause could be to disable or grey out lens detection auto when uncoded lenses are mounted on the camera.

This is all irrelevant to the freeze on startup issue, remember? They still happen with new, 6-bit coded Leica lenses on cleaned 6-bit sensors that we visually checked.

I think the person’s point was if a group this large can’t figure out how to reliably  provoke the issue, then it might make sense for Leica to start asking for the problem cameras back to do a tear-down. They could hire a third party consultant company test every single electronic component for a point of failure. Or just keep sending those of us with the issue a new camera until we get one that doesn’t freeze randomly at startup. Surely the tear-down solution makes more sense than playing Russian roulette with $9K German cameras over and over.

I really don’t see why you keep coming back to this thread (someone without this issue now). Why does it bother you to consider that this could be a hardware issue? The only reason I can see people clinging to the firmware issue theory is because they get to tell people, “You’ll be fine, your disease will clear up on its own once the doctor figures out what’s wrong with you.” Either way, it’s extremely disconcerting to still have this issue on new cameras rolling off the line right up to the time where the P version gets announced.

I still swear some of you who got “good ones” are having to pull the battery every so often and just not admitting it.