Leica M11 / M11-M / M11-P: Firmware Update v.2.0.2

[Derbyshire Man]

Advertisement (gone after registration)

5 hours ago, douglas fry said:

The Citizens Advice Bureau here in the UK said that 5 months is well beyond what can be considered a 'reasonable repair time', which as a consumer is within my rights. So I will wait and see what their response is. 

I shoot nearly every day so to keep the business going I need my rangefinders, so resorted to buy an M11-P to keep the workflow and keep the clients. Once this camera is repaired I need to sell it ASAP, as it is a big debt to carry for a small business.

How are you finding the M11-P Douglas other than expensive?!

[douglas fry]

11 hours ago, Derbyshire Man said:

How are you finding the M11-P Douglas other than expensive?!

Its a good camera but not without its faults, I needed to get a camera to keep me going while the M11 is away, and if you are going to have to spend a. lot of money it might as well be the P, and enjoy the extra memory and toughend glass etc.

 

I used it extensively last night and it froze once, and had one flakey DNG (probably linked) but will persevere, I have no choice really. The files however are great, there was another photographer at the venue with Canon R6 and R5 and we compared images......Lets say the M11 files looked far far better even at 12500 ISO 

[Advertisement]

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

[Fritz Fritz]

On 12/6/2023 at 6:14 AM, douglas fry said:

The snag is, if you find yourself with a faulty M11 (in my case live view didnt work at all, and the camera would freeze very frequently), Leica are quoting 5 months repair even with the camera well within the warranty period, no loaner or refund is offered (I asked) even if registered as a professional photographer with Leica. That is pretty poor service.

My M11 is at Leica New Jersey currently being serviced for exactly what you describe. Leica USA quoted 3 weeks turnaround. It’s been just shy of that so far so we’ll see.

Not that it’s any excuse. The M11 shouldn’t have been released at all with the current list of issues. 

Edited December 11, 2023 by Fritz Fritz

[Stef63]

On 12/6/2023 at 2:14 PM, douglas fry said:

The snag is, if you find yourself with a faulty M11 (in my case live view didnt work at all, and the camera would freeze very frequently), Leica are quoting 5 months repair even with the camera well within the warranty period, no loaner or refund is offered (I asked) even if registered as a professional photographer with Leica. That is pretty poor service.

This is against the law, especially within the European Union (EU). According to EU regulations, warranties must be honored within a "reasonable" timeframe. And it would be deemed unacceptable if Leica, being based in the EU, were to argue that your location outside the EU impacts the fulfillment of that warranty timeframe.

[lct]

15 minutes ago, Stef63 said:

This is against the law, especially within the European Union (EU). According to EU regulations, warranties must be honored within a "reasonable" timeframe. And it would be deemed unacceptable if Leica, being based in the EU, were to argue that your location outside the EU impacts the fulfillment of that warranty timeframe.

Location may impact unfortunately. According to EU regulations, the applicable law is that of the country where the consumer has his habitual residence. See: https://www.lexisnexis.co.uk/legal/guidance/rome-i-consumer-contracts-art-6

[kiwidad]

Is anyone experiencing banding at high iso ie over 6400 with 2.0.2 on their M11?

[lct]

Advertisement (gone after registration)

1 hour ago, kiwidad said:

Is anyone experiencing banding at high iso ie over 6400 with 2.0.2 on their M11?

Not me but why the 2.0.2 firmware if i may ask?

[kiwidad]

2 hours ago, lct said:

Not me but why the 2.0.2 firmware if i may ask?

I shot in a jazz club last night and used iso 25400 and got bands across the frame and I was able to reproduce in a dimly lit room at work today, I shot similarly at the same club 2 weeks ago and saw no such banding. Only variables were a sensor clean and updating from 1.6 to 2.0.2.

seems more prominent in underexposed images but I never noticed it till last night.

[lct]

6 hours ago, kiwidad said:

I shot in a jazz club last night and used iso 25400 and got bands across the frame and I was able to reproduce in a dimly lit room at work today, I shot similarly at the same club 2 weeks ago and saw no such banding. Only variables were a sensor clean and updating from 1.6 to 2.0.2.

seems more prominent in underexposed images but I never noticed it till last night.

I have read another post of yours regarding such banding, thank you, but firmware 2.0.2 was supposed to fix other issues. In this post you explained that your M11 had two sensor cleanings and that streaks coming from the first one have been discovered during the later, if i'm not mistaken. Your sensor has not been damaged by those cleanings hopefully. Now underexposure is to avoid at high isos so it is probably a better explanation.

Edited December 12, 2023 by lct

[kiwidad]

2 hours ago, lct said:

I have read another post of yours regarding such banding, thank you, but firmware 2.0.2 was supposed to fix other issues. In this post you explained that your M11 had two sensor cleanings and that streaks coming from the first one have been discovered during the later, if i'm not mistaken. Your sensor has not been damaged by those cleanings hopefully. Now underexposure is to avoid at high isos so it is probably a better explanation.

I understand it was to fix other issues but updates can introduce other issues and hence my question. I do not see this as a cleaning issue since the banding isn’t consistent in what shows in the image and since several settings other than iso can affect what shows in the image. I  Plan to send an image to leica today for their thoughts.

[J S H]

6 hours ago, kiwidad said:

I understand it was to fix other issues but updates can introduce other issues and hence my question. I do not see this as a cleaning issue since the banding isn’t consistent in what shows in the image and since several settings other than iso can affect what shows in the image. I  Plan to send an image to leica today for their thoughts.

Assuming you are shooting near wide open for these high iso shots, it’s safe to say that any problem with sensor streaks would not be visible. If it’s bad enough to show up wide open, it would be visible on every single image you make with the camera. 

[Tailwagger]

Just to keep a running count. I took precisely 60 shots this morning.  One damaged file. That's now a total of three corrupt files out of a few less than 200 since I made the mistake of upgrading to 2.0.2 from 1.6.1.  1.5% failure rate. Can I live with it? Not much choice at the moment.  Can they?  Well, its hard to see the current state of affairs as anything but a total embarrassment.  Releasing alpha quality firmware two years in, right along side a new M variant, is in a word, shameful.  One is forced to wonder how much longer it will take for an update, let alone an apology. 

[lct]

Took 82 shots under firmware 2.0.2 today. All DNG + JPG files on both internal memory and Sandisk UHS2 card formatted with SD Card Formatter. Got zero freeze and zero corrupt file as usual since firmware 1.6.1. I feel sorry for @Tailwagger and other colleagues complaining about their camera but i can only repeat that i would not generalize individual issues personally. As far as color rendition is concerned, i just picked out the pics below to show differences between the M11 color profile (1st pic) and a generic one (2nd pic) out of Iridient Developer. Auto WB and default settings, zero tweaking in PP, details in exif data. The M11 profile looks more faithful to my eyes but i tend to prefer the generic profile or perhaps a mix of them two subjectively. In any case, i would not vote for a firmware update if it happened to changes the M11 color profile that i prefer to the M240 profile personally. YMMV.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

[Tailwagger]

... and this morning, I lost shot #4.  1 out of 36 today after a dozen or so yesterday with no issue.  This failure, as well as the last, was the first shot of the day.  While previously the losses were mid-pack, I now believe that they too were a first shot in a sequence after a power cycle.  In this latest instance, while the file was corrupt on the SD card, the internal memory was not corrupt and I was able to recover it. Hallelujah. Which leads to the following TL;DR...

I had, since first encountering the problem, switched to storing duplicates to the internal memory.  However, the other day (my last post) when I went to see if the damaged file was okay on the internal mem, I found that my storage mode had reverted as I was back on the default user despite not having changed it manually. I believe others have seen similar behavior with user profiles going poof but as I, up until now, hadn't employed any, I'm not up on the issue. Regardless, this morning immediately before shooting, I confirmed that I was set to SD=IN.  A side effect of this was that the camera was on for quite a while prior to the first shot (the one found to be corrupted on the SD card).  From this incident (and recalling that my previous failures where in no way quick to the shutter), I conclude that its not likely that the theories around shooting too quickly after wakeup are an explanation for corruption... or at least not the only one.

I find that the oddity of the profile change suggests a possibility that there is something occasionally going awry w.r.t. writing values to NVRAM (perhaps on shutdown) that adversely affects start/wake up in some way.  It is also important to note that once the first shot is fired, the software seems to set everything to a known, correct state as the issue isn't seen again.  ie. AFAIK no one has seen a series of corrupted files, just one every now and again which suggests that the ship is somehow righted after a failure. If we believe that the corruption is only (or mostly... there could easily be more than one bug) occurring after boot or wake up, then it seems possible that a bug in the shutdown/sleep operation stores a bad value into NVRAM that results in the camera coming up in a bad state... thinking it's in the middle of a shot when it isn't, for example... thats cleared once a full image cycle is performed. 

If its not a bad persistent value written on the way down, then paging back in my previous life in VM and OS software, I'm similarly forced to wonder if the system is running afoul of uninitialized areas of volatile memory on boot/restart.  This scenario indeed has a very familiar smell to it.  A classic problem induced by... and many of you have complained about this... the desire to minimize start up time. Software typically assumes that if it has yet to manipulate a given value in memory, that value is 0.  As RAM on startup can contain random non-zero values, classically before any operations can be allowed to run, the system has to fill the RAM with zeroes. If, for example, the memory in question is a critical flag that is assumed to be false (0), but randomly winds up true (non-zero) and said flag signifies that the system is in the middle of a critical operation... say writing the image to SD... well, you can imagine the sort of havoc that can result... lockups, the exposure meter has been read when in fact it hasn't, all of the file's bits have been written when they haven't... you get the idea.

Depending on how much memory you have (I recall several of my colleagues at my last gig having to deal with the headache of initializing multiple terabytes of memory), it can take quite a bit of time before the system can be allowed to touch it.  Given startup/wakeup times are always a pain point, the engineers are tempted/forced to get clever about how they initialize memory to minimize the delay. Typically they defer initialization of less critical areas until just before they are actually referenced.  But if they overdo the optimization or create an odd circumstance where they can fail to init a critical location, its possible every now and again for an unsuspecting bit of code to read a random value and, of course, bad stuff ensues.  And trust me, figuring out just how and where such things went wrong isn't at all easy.

As I've said previously, if the problems plaguing the M11 were straightforward, they'd all have been fixed long ago.  And again, having lived similar sorts of nightmares in my own  career, I have some personal sympathy for those involved. But I am forced to continue to pound the table around how Leica has handled this and similar problems with the M11.  They need to be more transparent about known issues, suggest workarounds when available and avoid going silent by issuing statements about timelines/progress.  That, IMO, is far better than the user base having to speculate in the dark ad nausem (guilty as charged) around what is going on and how to deal with it.

Edited December 20, 2023 by Tailwagger

[rcusick]

I’ve owned my M11 since the beginning.  For me all issues have been fixed - mostly freezing issues at first.  For people who have issues - try new memory cards…

[hdmesa]

3 hours ago, Tailwagger said:

... and this morning, I lost shot #4.  1 out of 36 today after a dozen or so yesterday with no issue.  This failure, as well as the last, was the first shot of the day.  While previously the losses were mid-pack, I now believe that they too were a first shot in a sequence after a power cycle.  In this latest instance, while the file was corrupt on the SD card, the internal memory was not corrupt and I was able to recover it. Hallelujah. Which leads to the following TL;DR...

I had, since first encountering the problem, switched to storing duplicates to the internal memory.  However, the other day (my last post) when I went to see if the damaged file was okay on the internal mem, I found that my storage mode had reverted as I was back on the default user despite not having changed it manually. I believe others have seen similar behavior with user profiles going poof but as I, up until now, hadn't employed any, I'm not up on the issue. Regardless, this morning immediately before shooting, I confirmed that I was set to SD=IN.  A side effect of this was that the camera was on for quite a while prior to the first shot (the one found to be corrupted on the SD card).  From this incident (and recalling that my previous failures where in no way quick to the shutter), I conclude that its not likely that the theories around shooting too quickly after wakeup are an explanation for corruption... or at least not the only one.

I find that the oddity of the profile change suggests a possibility that there is something occasionally going awry w.r.t. writing values to NVRAM (perhaps on shutdown) that adversely affects start/wake up in some way.  It is also important to note that once the first shot is fired, the software seems to set everything to a known, correct state as the issue isn't seen again.  ie. AFAIK no one has seen a series of corrupted files, just one every now and again which suggests that the ship is somehow righted after a failure. If we believe that the corruption is only (or mostly... there could easily be more than one bug) occurring after boot or wake up, then it seems possible that a bug in the shutdown/sleep operation stores a bad value into NVRAM that results in the camera coming up in a bad state... thinking it's in the middle of a shot when it isn't, for example... thats cleared once a full image cycle is performed. 

If its not a bad persistent value written on the way down, then paging back in my previous life in VM and OS software, I'm similarly forced to wonder if the system is running afoul of uninitialized areas of volatile memory on boot/restart.  This scenario indeed has a very familiar smell to it.  A classic problem induced by... and many of you have complained about this... the desire to minimize start up time. Software typically assumes that if it has yet to manipulate a given value in memory, that value is 0.  As RAM on startup can contain random non-zero values, classically before any operations can be allowed to run, the system has to fill the RAM with zeroes. If, for example, the memory in question is a critical flag that is assumed to be false (0), but randomly winds up true (non-zero) and said flag signifies that the system is in the middle of a critical operation... say writing the image to SD... well, you can imagine the sort of havoc that can result... lockups, the exposure meter has been read when in fact it hasn't, all of the file's bits have been written when they haven't... you get the idea.

Depending on how much memory you have (I recall several of my colleagues at my last gig having to deal with the headache of initializing multiple terabytes of memory), it can take quite a bit of time before the system can be allowed to touch it.  Given startup/wakeup times are always a pain point, the engineers are tempted/forced to get clever about how they initialize memory to minimize the delay. Typically they defer initialization of less critical areas until just before they are actually referenced.  But if they overdo the optimization or create an odd circumstance where they can fail to init a critical location, its possible every now and again for an unsuspecting bit of code to read a random value and, of course, bad stuff ensues.  And trust me, figuring out just how and where such things went wrong isn't at all easy.

As I've said previously, if the problems plaguing the M11 were straightforward, they'd all have been fixed long ago.  And again, having lived similar sorts of nightmares in my own  career, I have some personal sympathy for those involved. But I am forced to continue to pound the table around how Leica has handled this and similar problems with the M11.  They need to be more transparent about known issues, suggest workarounds when available and avoid going silent by issuing statements about timelines/progress.  That, IMO, is far better than the user base having to speculate in the dark ad nausem (guilty as charged) around what is going on and how to deal with it.

Sorry this is happening to you. I read earlier in this thread you're using Sandisk UHS-II cards. But are they also the v90 Sandisk with the 260 MB/s write speed? The v60s I had gave me headaches. I've had no card issues with 2.0.2 on the M11M while using v90 Sandisk and Delkin Black cards.

[Tailwagger]

3 hours ago, hdmesa said:

Sorry this is happening to you. I read earlier in this thread you're using Sandisk UHS-II cards. But are they also the v90 Sandisk with the 260 MB/s write speed? The v60s I had gave me headaches. I've had no card issues with 2.0.2 on the M11M while using v90 Sandisk and Delkin Black cards.

Appreciate the sentiment and genuinely glad that you aren't.  Current card is a v90 300 MB/s 64GB. This hopefully suggests to you, as it does to me (and as I've believed all along), that subtle timing alterations alternately expose or hide this and possibly other problems that have been reported with the M11.

6 hours ago, rcusick said:

I’ve owned my M11 since the beginning.  For me all issues have been fixed - mostly freezing issues at first.  For people who have issues - try new memory cards…

I purchased my M11 on day number one.  I had a few lockups, certainly fewer than some on these pages seemed to have experienced. After 1.6, I can't recall seeing any. I only saw the exposure issue when trying like hell to induce it.  As it had been reported by others, I was curious if I could reproduce it to understand how to avoid the issue. And indeed I was able to and have done so.  I found neither of these problems to be all that infuriating, but I certainly can understand how an unexpected blown exposure could ruin one's day.  Love or hate, much depends on what each individual shoots and how they go about it.  For my style and work, the 11 had no truly serious issues... until 2.0.2 introduced me to this one (I skipped 2.0.1).  And for the record all of this, good and bad, has taken place with the very same card referenced above installed on that same day one.

To repeat. The card has not changed; the software has. Never had a file corrupted until the latest firmware release. On any M. With any card. Ever. Logically, what is to blame?

Before any answering that question, perhaps its worth reminding ourselves that Leica has stated very clearly in their release notes for 2.0.2 that there is a known file corruption problem for which they expect to supply a fix 'shortly'. A problem which in their words, not mine, is 'rare'.  I've hit the issue 4 times in around 300 frames.  Rare perhaps for some, but not for me.  Additionally, I have not seen Leica state anything about certain cards being problematic w.r.t. to this file corruption. To the contrary, and to their credit, they are owning the issue as their problem to fix.

Buy/try a new card... the problem may indeed go into hibernation as with different read/write/controller access/etc speeds; if the timing landscape gets altered enough what might have been a head on collision now turns into a near miss. Just recognize that any success achieved is more a matter of voodoo than technology.  As with hdmesa, I'm happy for all those who can count themselves fortunate for not having encountered this particular failure. But it should be understood that when credible users report that they have, those that haven't shouldn't assume that problem isn't in there, nor that they wont eventually run into it.  New card or no.

 

[Spindrift]

On 11/29/2023 at 3:55 PM, bilbrown said:

It took a long time to connect from the app and I restarted the app a couple of times then it ran very quickly.

I could have waited for the download, but since the camera is new I wanted to push it and see if it did what it was supposed to do.

 

 

I have not had a chance to shoot much since the update but I have an even tomorrow and my wife's birthday Friday so will figure out.

Maybe already posted but I had to rename my file (to all lower case if I recall) in order to get the camera to recognize the 2.0.2 FW update file in the menus. I followed the instructions and noticed the filename I downloaded was all UPPER case and the filename in the doc was all lower case. YMMV.

[elmars]

vor 15 Stunden schrieb Tailwagger:

They need to be more transparent about known issues

Yes!

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Al Brown]

On 12/6/2023 at 2:14 PM, douglas fry said:

The snag is, if you find yourself with a faulty M11 (in my case live view didnt work at all, and the camera would freeze very frequently), Leica are quoting 5 months repair even with the camera well within the warranty period, no loaner or refund is offered (I asked) even if registered as a professional photographer with Leica. That is pretty poor service.

In EU the LEGAL obligation to repair a faulty camera within warranty is 45 days or the seller has to replace it. No questions asked, no doubt here.