Q2 Monochrom "stuck" bright white pixels

[SrMi]

Advertisement (gone after registration)

It would be interesting to hear from Q2M owners who do not have stuck pixels. Anyone?

 

Edited December 22, 2020 by SrMi

[nwphil]

looking for image submittals - statistical test on raw data

https://www.photonstophotos.net/Collaborations/Fixed_Pattern_Noise_Collaboration.htm

please pm me so I can send you the email for downloading images on Dropbox - use file naming as follow below:

"first and last name initials"-"last 4 digits of camera's serial number"-"##"(picture number sequential from 1 to 16 as instructed in htm document)

- note: open to suggestions for file naming too

Thanks

ps: the document shows an email at the end - do not use that email - I am already in contact with Bill Claff, who was been very open and generous with his time, willing to help and investigate this issue - let's not get it messy and overwhelming please

Edited December 22, 2020 by nwphil

[Advertisement]

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

[nwphil]

1 minute ago, SrMi said:

It would be interesting to hear from Q2M owners who do not have stuck pixels. Anyone?

 

I am starting to believe everyone will have - something that will look like stuck pixels. To be determined yet what they are, but I was not even remotely aware till yesterday

[SrMi]

1 minute ago, nwphil said:

I am starting to believe everyone will have - something that will look like stuck pixels. To be determined yet what they are, but I was not even remotely aware till yesterday

That is, unfortunately, my suspicion as well.

[anuncastshadow]

I must admit I feel bad for pointing out this minor issue that most Q2M users had been fortunate enough to not notice until now, similar to how someone may point out an annoying sound you might not have noticed but once it's pointed out it becomes impossible to ignore.

I would like to point out again that Leica did at least partially respond to this issue by acknowledging there's no current firmware or software solution, but they did request (and I sent along) two DNG files yesterday to aid them in investigating this matter further.

Based on everyone's collective experience it does seem to be software related, but until Leica has further communication regarding the matter I can understand everyone's caution.

 

[nwphil]

54 minutes ago, anuncastshadow said:

I must admit I feel bad for pointing out this minor issue that most Q2M users had been fortunate enough to not notice until now, similar to how someone may point out an annoying sound you might not have noticed but once it's pointed out it becomes impossible to ignore.

I would like to point out again that Leica did at least partially respond to this issue by acknowledging there's no current firmware or software solution, but they did request (and I sent along) two DNG files yesterday to aid them in investigating this matter further.

Based on everyone's collective experience it does seem to be software related, but until Leica has further communication regarding the matter I can understand everyone's caution.

 

you have nothing to be apologizing for - that's why forums exist, and this is indeed the positive side of it.

Neither is a minor issue imo

not happy with Leica's reply:

I am very sorry to hear of the trouble you are experiencing with your Leica Q2 Monochrome camera.  We can certainly assist in getting this fixed for you. I have attached our Repair & Service form, please complete and send in with your equipment in for evaluation. As soon as we receive it, our technician will inspect it and contact you with our findings.  An estimate will be emailed to you if there are any repair charges incurred. If you have any questions or concerns, please do not hesitate to contact us."

I understand that they are not sure of what is going on -yet, hence the disclaimer-type reply, but would expect a pre-paid shipping label for a new camera under warranty.

If they are not providing one, I will offer to send image files, and wait a bit more - if they insist in this path, then a return is likely, which is very unfortunate as I really like the camera

- they even suggest to return camera to store in a reply I gave to their answer - it's not right! not determined yet if this is class lawsuit action, major recall, or just a firmware update away. Placing the burden on consumer ( return shipping, restocking fees) and store ( despite being an official Leica store) is not proper  imo

Edited December 22, 2020 by nwphil

[nwphil]

Advertisement (gone after registration)

1 hour ago, nwphil said:

looking for image submittals - statistical test on raw data

https://www.photonstophotos.net/Collaborations/Fixed_Pattern_Noise_Collaboration.htm

please pm me so I can send you the email for downloading images on Dropbox - use file naming as follow below:

"first and last name initials"-"last 4 digits of camera's serial number"-"##"(picture number sequential from 1 to 16 as instructed in htm document)

- note: open to suggestions for file naming too

Thanks

ps: the document shows an email at the end - do not use that email - I am already in contact with Bill Claff, who was been very open and generous with his time, willing to help and investigate this issue - let's not get it messy and overwhelming please

no partial serial numbers will be needed - I will not see them anyway, just the researcher. 

Edited December 22, 2020 by nwphil

[marchyman]

I wonder.   In the Q or the Q2 the raw image has to go through a de-mosaicing function.  That would not be the case on the Q2M, correct?  I wonder if stuck pixels are hidden on the sensors with a color bayer filter because the value of a pixel is determined by multiple photoreceptor sites.

Just guessing.

[nwphil]

9 minutes ago, marchyman said:

I wonder.   In the Q or the Q2 the raw image has to go through a de-mosaicing function.  That would not be the case on the Q2M, correct?  I wonder if stuck pixels are hidden on the sensors with a color bayer filter because the value of a pixel is determined by multiple photoreceptor sites.

Just guessing.

I am not techno-savy enough to even consider that, but might make sense an be a reasonable explanation. They had testers for weeks, and sure did their own tests, so how come no one noticed this so far? To me ii seems that it's either a major production snafu, or there is some firmware/software glitch that went under the radar 

[nwphil]

1 hour ago, nwphil said:

you have nothing to be apologizing for - that's why forums exist, and this is indeed the positive side of it.

Neither is a minor issue imo

not happy with Leica's reply:

I am very sorry to hear of the trouble you are experiencing with your Leica Q2 Monochrome camera.  We can certainly assist in getting this fixed for you. I have attached our Repair & Service form, please complete and send in with your equipment in for evaluation. As soon as we receive it, our technician will inspect it and contact you with our findings.  An estimate will be emailed to you if there are any repair charges incurred. If you have any questions or concerns, please do not hesitate to contact us."

I understand that they are not sure of what is going on -yet, hence the disclaimer-type reply, but would expect a pre-paid shipping label for a new camera under warranty.

If they are not providing one, I will offer to send image files, and wait a bit more - if they insist in this path, then a return is likely, which is very unfortunate as I really like the camera

- they even suggest to return camera to store in a reply I gave to their answer - it's not right! not determined yet if this is class lawsuit action, major recall, or just a firmware update away. Placing the burden on consumer ( return shipping, restocking fees) and store ( despite being an official Leica store) is not proper  imo

received another reply from Leica USA and this time I noticed the email was actually from the repair department - somehow technicalinfo department email got into my sent email ( still no excuse for a lousy reply)

of interest is indeed Repair dpt. reply:

 

"Phillip

The Q2 Monochrom  automatically remaps pixels via its own system. Unfortunately I cannot say how long that takes.

If it does no correct itself over the next week of picture taking, please send it to our Service Dept for evaluation(...)"

promissing I would say, but I don't recall to see similar statement on the manual or anywhere else, and yet if indeed true and rightful, it is of utmost importance to have been broadcast clearly.

Edited December 22, 2020 by nwphil

[SrMi]

3 hours ago, nwphil said:

received another reply from Leica USA and this time I noticed the email was actually from the repair department - somehow technicalinfo department email got into my sent email ( still no excuse for a lousy reply)

of interest is indeed Repair dpt. reply:

 

"Phillip

The Q2 Monochrom  automatically remaps pixels via its own system. Unfortunately I cannot say how long that takes.

If it does no correct itself over the next week of picture taking, please send it to our Service Dept for evaluation(...)"

promissing I would say, but I don't recall to see similar statement on the manual or anywhere else, and yet if indeed true and rightful, it is of utmost importance to have been broadcast clearly.

Thank you for that information. I will try to take lots of pictures with my Q2M over the weekend, focusing on ISO 800 and higher :). It is strange that the hot pixels appear only at higher ISOs. We see here that something happens at ISO 800 (dual gain?), which causes hot pixels to show.

[nwphil]

1 hour ago, SrMi said:

Thank you for that information. I will try to take lots of pictures with my Q2M over the weekend, focusing on ISO 800 and higher :). It is strange that the hot pixels appear only at higher ISOs. We see here that something happens at ISO 800 (dual gain?), which causes hot pixels to show.

I might not be setting things correctly, but in my case it seems thatt hey are more defined at lower iso, but not always and not as many - it varies a bit how many per iso setting. At higher iso they get a bit more blurred and it's hard to distinguish from what i assume it's noise  - LR with spot finder and visualizer slider being used; at higher iso, has to be moved to the left quite a bit.

The main thing here to me, it's to make Leica aware that owners are concerned about a potential issue/phenomenon and there is no official explanation...yet.

 If indeed just like the batteries, that a certain amount of use is needed to bring it to full performance (batteries is ofcourse full charge cycles), I don't see why that was not mentioned. What? Don't they know tat people are going to analyze the shots to death, at pixel level, above and beyond needed? It happens with every new camera.

I cannot actually see them in my normal pictures, but have not really look deeply into it, or tried to match  the white dots in the shots with cap on to other shots (as a mask overlay). Also so far I have been using mostly iso 100. If they are present  and somehow they don't show ....well, what can I say? sometimes one sees what it wants to see, and I don't want to see white dots 😂

- but the normal shots from the OP actually do show up, they are clearly present.

My testing shots were shared to the researcher, but I won't expect to hear anything today (he is on eastern time). will post his findings and any other feedback from Leica as they become available. 

Edited December 23, 2020 by nwphil

[SrMi]

3 hours ago, nwphil said:

I might not be setting things correctly, but in my case it seems thatt hey are more defined at lower iso, but not always and not as many - it varies a bit how many per iso setting. At higher iso they get a bit more blurred and it's hard to distinguish from what i assume it's noise  - LR with spot finder and visualizer slider being used; at higher iso, has to be moved to the left quite a bit.

The main thing here to me, it's to make Leica aware that owners are concerned about a potential issue/phenomenon and there is no official explanation...yet.

 If indeed just like the batteries, that a certain amount of use is needed to bring it to full performance (batteries is ofcourse full charge cycles), I don't see why that was not mentioned. What? Don't they know tat people are going to analyze the shots to death, at pixel level, above and beyond needed? It happens with every new camera.

I cannot actually see them in my normal pictures, but have not really look deeply into it, or tried to match  the white dots in the shots with cap on to other shots (as a mask overlay). Also so far I have been using mostly iso 100. If they are present  and somehow they don't show ....well, what can I say? sometimes one sees what it wants to see, and I don't want to see white dots 😂

- but the normal shots from the OP actually do show up, they are clearly present.

My testing shots were shared to the researcher, but I won't expect to hear anything today (he is on eastern time). will post his findings and any other feedback from Leica as they become available. 

You can look for stuck pixels by taking pictures with the lens cap on and varying the ISO (e.g., with 1/15, f/4). If you open those images in LrC and use dust visualization, the location of stuck pixels will be visible.
I have provided Bill with data for various Leica cameras, including Q2 and Q2M. He reported several stuck pixels both on Q2 and Q2M. They do not bother me in Q2 as they apparently get eliminated automatically by Adobe software.
Those who use C1: just noticed that Q2M's stuck pixels can be easily eliminated by using Noise Reduction -> Single Pixel -> 2.

[nwphil]

21 minutes ago, SrMi said:

You can look for stuck pixels by taking pictures with the lens cap on and varying the ISO (e.g., with 1/15, f/4). If you open those images in LrC and use dust visualization, the location of stuck pixels will be visible.
I have provided Bill with data for various Leica cameras, including Q2 and Q2M. He reported several stuck pixels both on Q2 and Q2M. They do not bother me in Q2 as they apparently get eliminated automatically by Adobe software.
Those who use C1: just noticed that Q2M's stuck pixels can be easily eliminated by using Noise Reduction -> Single Pixel -> 2.

 

 

Thanks SrMi 

Bill gave me the results on my sensor and he says there are no signs of defective pixels ( txt file) - so, what are these white spots then?

( i just realize that two of the screenshots are hard to see - if anyone needs/wants full sized ones, I can send you the googledrive sharelink)

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!

Q1000453_1_bvz.txt

Edited December 23, 2020 by nwphil

[robgo2]

On 12/21/2020 at 9:58 AM, nwphil said:

Do you have or the chance to try with another software? A bit busy right now, but eventually I will get to it...sometime today.

I have Niki collection installed but not sure if will be of help; an ON1 trial is not activated yet, but this could be a good time to do it.

I don't think either Canon or Olympus editing software would be relevant for this issue, but I might given them a try and see if it works - any other software worthy of experimenting with in your opinion? - trying to prepare ahead for the blame-game of Adobe vs Leica... 😁

I have edited the Dpreview samples in ON1, and the stuck pixels are quite visible.  This really concerns me, as I have a Q2M scheduled for delivery on Dec. 26.

[robgo2]

On 12/21/2020 at 8:25 PM, fieldafield said:

Hi All, another one here. At 1600% they are clearly a cluster of 4 pixels . . . I have 6 - camera also bought from RangefinderHK  thoughit only arrived today soI have a return period. . . . I would expect it to be fixed by firmware but wanted to add my vote to the campaign.

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!

In my book, any aberrancy that can only be seen at 1600% enlargement is not a problem.  Do you see them at 100%?

[nwphil]

1 minute ago, robgo2 said:

I have edited the Dpreview samples in ON1, and the stuck pixels are quite visible.  This really concerns me, as I have a Q2M scheduled for delivery on Dec. 26.

those white spots might be just noise or indeed somewhat defective pixels that eventually the Q2M will map out - hence the reply from Leica, and the results I got from a sensor diagnosing test perfumed on images from my camera.

Somewhat find hard to believe that official testers were not able to see/find them, not to mention Leica technicians... There is no official explanation yet, or as a matter of fact of any kind. Some people seems to be able to spot in regular frames; I can only see them in cap-on shots.

Very likely Leica is paying attention now, as they received several emails, and I expect an answer or explanation of sorts will come out soon.

I am going to have images analyzed in a different way next week by the same person that performed the pixel tests., but as I said, they are not being classified as defective pixels - as it seems to closest might be noise(?)

It's weird...I will keep on digging. Results seem to vary widely, so more time and analysis is needed 

[RoySmith]

Years ago, there was a theory that gamma radiation when flying at altitudes > 20,000 ft caused dead pixels. I have not heard about this for years so I am unsure if it’s true or just an urban myth. PetaPixel had an article on it. 

 

[anuncastshadow]

1 hour ago, RoySmith said:

Years ago, there was a theory that gamma radiation when flying at altitudes > 20,000 ft caused dead pixels. I have not heard about this for years so I am unsure if it’s true or just an urban myth. PetaPixel had an article on it. 

 

I think this may be apropos of nothing, but the Q2M manual does acknowledge this, but only in the context of discouraging use of the camera while at a high altitude. I think it is safe to assume, obviously, that this phenomena does not affect cameras merely in travel. 

[Advertisement]

Hallo,

Look here: Leica Q

[robgo2]

I have run some tests using software that I have available trying to find out which ones can remove the stuck white pixels.  These include ON1, Exposure X6, ACR, Capture One 21 (trial version), Photoshop, Neat Image 8 and Topaz Denoise AI.  The winners are PS (dust and scratches filter at the lowest level) and C1 (noise reduction > single pixel).  These are the only two programs in my limited testing that can do the job.  In my own case, I will use PS, because I already have a subscription. C1 is also a good option for those who own it and like it (I am not amongst them.).