Latent Pixel Failure

[StoneNYC]

So I own a 5Dmk2.

On my old slow computer I noticed that for images with long exposure (30 seconds or more) when the RAW file was first loading into Lightroom and rendering I would see green and red pixel dots that were very bright and then when it finished loading they would disappear.

When pixels fail in a sensor, is there some kind of RAW data that photo software "fills in the blanks"?

Anyone know?


~Stone

Mamiya: 7 II, RZ67 Pro II / Canon: 1V, AE-1, 5DmkII / Kodak: No 1 Pocket Autographic, No 1A Pocket Autographic | Sent w/ iPhone using Tapatalk

 

[Eric Rose]

It's haunted. Either that or the other pixels are shy.

 

[L Gebhardt]

The cameras contain mappings to fix/ignore bad pixels when saving the files. See Ask Olympus: What is Pixel Mapping?

RAW converters also have methods of fixing bad pixels so you don't see them. I suspect your camera has developed more since it was mapped at the factory and Lightroom is also fixing them.

The method of fixing is most certainly that software "fills in the blanks", whether it's the camera or RAW processor that does it.

 

[StoneNYC]

The cameras contain mappings to fix/ignore bad pixels when saving the files. See Ask Olympus: What is Pixel Mapping?

RAW converters also have methods of fixing bad pixels so you don't see them. I suspect your camera has developed more since it was mapped at the factory and Lightroom is also fixing them.

The method of fixing is most certainly that software "fills in the blanks", whether it's the camera or RAW processor that does it.

Ok thanks. I just wanted to be sure there's nothing to worry about.


~Stone

Mamiya: 7 II, RZ67 Pro II / Canon: 1V, AE-1, 5DmkII / Kodak: No 1 Pocket Autographic, No 1A Pocket Autographic | Sent w/ iPhone using Tapatalk

 

[RalphLambrecht]

could that be pixel noise,below the signal /noise ratio?

 

[StoneNYC]

could that be pixel noise,below the signal /noise ratio?

I don't know what that means?

PS hey Ralph! Good to see you on the (dark) light side

 

[RalphLambrecht]

well,eventually,the digital devil gets all of us;back on topic;every electronic or other sensorneeds a certain minimal signal strength to sense a signal. any weaker signal is just recorded as noise.try thisut your lens cap on and make a 3-minute-exposure with the shutteropen(B)setting. the resulting image will show your pixel noise(all the pixels that did not get a signal strong enough to positively record light)it usually looks like the night sky.don't be alarmed ,this is normal;nothing wrong with your sensor;just typical signal to noise ratio behaviour

 

[StoneNYC]

well,eventually,the digital devil gets all of us;back on topic;every electronic or other sensorneeds a certain minimal signal strength to sense a signal. any weaker signal is just recorded as noise.try thisut your lens cap on and make a 3-minute-exposure with the shutteropen(B)setting. the resulting image will show your pixel noise(all the pixels that did not get a signal strong enough to positively record light)it usually looks like the night sky.don't be alarmed ,this is normal;nothing wrong with your sensor;just typical signal to noise ratio behaviour

I see Ralph, I'll try it. However I think canon is ... "Compensating for something" and that the RAW data somehow corrects this as the file is imported and that's why on my slow computer I would see the failures at first and then they would dissipear, but now my faster computer I so t see them, but I'll give it a go.

 

[RalphLambrecht]

I see Ralph, I'll try it. However I think canon is ... "Compensating for something" and that the RAW data somehow corrects this as the file is imported and that's why on my slow computer I would see the failures at first and then they would dissipear, but now my faster computer I so t see them, but I'll give it a go.

yes, that'slikely; Nikon calls it long exposure and high -ISO noise reduction(NR)if you can, turn them off for the test.

 

[StoneNYC]

yes, that'slikely; Nikon calls it long exposure and high -ISO noise reduction(NR)if you can, turn them off for the test.

No, that function is turned off, it's definitely one of the things I turned off immediately.

I think that behind the scenes ALL sensors' pixels die somewhere on the sensor and there is data saying "no signal" from that pixel so the surrounding pixels add interpolated data to where that dead pixel was to "fill in the blanks" and in the import, that info is applied to the image when it is "rendering" and I just caught it mid-render before the adjustments were applied.

That's my guess?

 

[RalphLambrecht]

No, that function is turned off, it's definitely one of the things I turned off immediately.

I think that behind the scenes ALL sensors' pixels die somewhere on the sensor and there is data saying "no signal" from that pixel so the surrounding pixels add interpolated data to where that dead pixel was to "fill in the blanks" and in the import, that info is applied to the image when it is "rendering" and I just caught it mid-render before the adjustments were applied.

That's my guess?

if that's your story, stick to it. Istill think it has to do with the signal to noise ratio.

 

[StoneNYC]

if that's your story, stick to it. Istill think it has to do with the signal to noise ratio.

You would know more than me...

 

[RalphLambrecht]

You would know more than me...

and yet,I make every mistake possible,and often more than once.

 

[StoneNYC]

and yet,I make every mistake possible,and often more than once.

How can you learn if you never fail

 

[RalphLambrecht]

How can you learn if you never fail

understood, but usually,Itry to avoid making all mistakes myself. Yet...

 

[StoneNYC]

understood, but usually,Itry to avoid making all mistakes myself. Yet...

Haha

I'm sure I'll make a few mistakes with a processor too but I've got about 50 rolls of C-41 and 25 rolls of E-6 waiting to be processed in 35mm and 120, and just shoot 20 sheets of E-6 waiting as well... So... I'll have plenty of options on which films I want to "screw up" lol

Hope there isn't latent "pixel" failure on any of these films, some are almost a year old! (Mostly because hand processing color in your sink with no temp control can be a bit ... Difficult... To say the least.

My plan was to shoot mostly in color on digital, an film in B&W, however it seems that I prefer color film as well, especially for landscape and long exposure images. Alas, I'll just have to keep shooting both and see how things go...

 

[Pioneer]

In the beginning I had the same idea Stone. Shoot my color on digital and b&w on film. From what I could tell digital was better at color while it still didn't quite meet my expectation with black and white. It is really what got me back into digital.

But then I ended up picking up a Jobo processor. Now things are backward. I am now shooting color on film and I am shooting black and white on digital, mostly because I am trying to train myself to be better at seeing in black and white.

Along the way I have kind of revised my opinion of digital. I actually find digital more useful for low light photography, whether I am using color or black and white. Funny how impressions can change with time when you keep your eyes open.

 

[StoneNYC]

In the beginning I had the same idea Stone. Shoot my color on digital and b&w on film. From what I could tell digital was better at color while it still didn't quite meet my expectation with black and white. It is really what got me back into digital.

But then I ended up picking up a Jobo processor. Now things are backward. I am now shooting color on film and I am shooting black and white on digital, mostly because I am trying to train myself to be better at seeing in black and white.

Along the way I have kind of revised my opinion of digital. I actually find digital more useful for low light photography, whether I am using color or black and white. Funny how impressions can change with time when you keep your eyes open.

I can see how that would make sense, actually usually when I shoot a color scene with my Digital I switch the camera to monochrome and actually shoot in monochrome so when I'm "chincing" the images, I can get a better idea of the exposure levels etc. I don't know why but that's just how my brain works.

I agree to an extent the digital are better in low light situations, however they are not better with lowlight long exposure images, anything that takes a tripod to shoot I prefer to shoot film it just comes out better with less noise.

 

[Pioneer]

I agree to an extent the digital are better in low light situations, however they are not better with lowlight long exposure images, anything that takes a tripod to shoot I prefer to shoot film it just comes out better with less noise.

+1 on that. If I know I will be shooting long exposure images I almost always pull out my Pentax LX. The off the film metering is astoundingly accurate. It has been my long exposure champion for 15 years. If I never used it again for anything else I doubt I will ever get rid of it because of that one advantage.

 

[StoneNYC]

+1 on that. If I know I will be shooting long exposure images I almost always pull out my Pentax LX. The off the film metering is astoundingly accurate. It has been my long exposure champion for 15 years. If I never used it again for anything else I doubt I will ever get rid of it because of that one advantage.

does it go longer than 30 seconds? As far as exposure evaluation?

 

[Pioneer]

It will monitor the light reaching the surface of the film until the set exposure is reached, even if that takes all night. Even today it is considered one of the top choices for astrophotography.

 

[StoneNYC]

It will monitor the light reaching the surface of the film until the set exposure is reached, even if that takes all night. Even today it is considered one of the top choices for astrophotography.

Oh wow! How interesting, however how does it calculate or adjust for reciprocity failure?

 

[Pioneer]

I don't think it does but it still seems to work out. I have personally taken shots out in the Nevada desert very late at night with no moon that ran over 45 minutes on color slide film and they turned out. The results can be awesome. My best shots have come more in the 2 to 5 minute range on Automatic Exposure.

I am just a raw amateur and I don't know why this works, but reciprocity failure does not seem as important for these types of photos as people make out. The pictures do look different. If I use the LX on auto the pictures look like they were taken at night. If I calculate for reciprocity failure I seem to lose that nightime flavor to the pictures and they look almost as if they were taken in the evening. All I know is it seems to work.

I have shot Kodak E100 and Fuji Provia 100 slide film with good results and Kodak Ektar seems to do a pretty good job. I don't remember working with black and white but I may have to give this a try with some TMX100 or Fuji Neopan.

 

[StoneNYC]

I don't think it does but it still seems to work out. I have personally taken shots out in the Nevada desert very late at night with no moon that ran over 45 minutes on color slide film and they turned out. The results can be awesome. My best shots have come more in the 2 to 5 minute range on Automatic Exposure.

I am just a raw amateur and I don't know why this works, but reciprocity failure does not seem as important for these types of photos as people make out. The pictures do look different. If I use the LX on auto the pictures look like they were taken at night. If I calculate for reciprocity failure I seem to lose that nightime flavor to the pictures and they look almost as if they were taken in the evening. All I know is it seems to work.

I have shot Kodak E100 and Fuji Provia 100 slide film with good results and Kodak Ektar seems to do a pretty good job. I don't remember working with black and white but I may have to give this a try with some TMX100 or Fuji Neopan.

NeopanAcros100 or Tmax400 works "best" (least reciprocity) and Fuji slide films I believe.

But don't try this with FOMA, or you may get no image at all, some of the worst reciprocity ever...

 

[RalphLambrecht]

NeopanAcros100 or Tmax400 works "best" (least reciprocity) and Fuji slide films I believe.

But don't try this with FOMA, or you may get no image at all, some of the worst reciprocity ever...

have you tried digital sensors?apparently,they have no reciprocity failure at all