MFlenses - A new life in a digital age
MFLenses 2026

Manual Focus Lenses Forum

Explore the world of vintage lenses through repair notes, real test shots, and detailed archive threads packed with images.

Explore the world of vintage lenses through repair notes, real test shots, and detailed archive threads packed with images.
Spotlight

Featured partner placement for the forum front page.

Hot weather and digital sensor noise
View previous topic :: View next topic  


Reply

Am I the only (visionary?) who seems to notice that with the current hot temperatures (around 35°C here) the CMOS sensor of my digital camera (in this case the 5DMkII, but I think it might apply to many cameras) produces more noise in situations (like 200 ISO for example) where it would produce little or none normally?


Reply

You love the camera and you are breathing on it to hard. Its reacting.


Reply

Orio wrote:
Am I the only (visionary?) who seems to notice that with the current hot temperatures (around 35°C here) the CMOS sensor of my digital camera (in this case the 5DMkII, but I think it might apply to many cameras) produces more noise in situations (like 200 ISO for example) where it would produce little or none normally?


I can't say that I've noticed anything but the temperatures here haven't reached 35. However noise is expected to increase with temperature. That's why low light detectors such as those in astronomical telescopes have cooled sensors. ( I think that some of the Medium Format sensors may have built in cooling but I can't remember where I got this from and I might be wrong). A quick Google search suggests that dark noise is expected to double for every 8-9 degrees rise in temperature. Another source suggests 6 degrees. Whatever the value it's a non-trivial effect.


Reply

sichko wrote:
However noise is expected to increase with temperature. That's why low light detectors such as those in astronomical telescopes have cooled sensors. ( I think that some of the Medium Format sensors may have built in cooling but I can't remember where I got this from and I might be wrong). A quick Google search suggests that dark noise is expected to double for every 8-9 degrees rise in temperature. Another source suggests 6 degrees. Whatever the value it's a non-trivial effect.


I have noticed. Today we hit 38°C (not 35 as I wrote before) and my 5DMkII performed sh** at 200 ISO - which has never happened before.

At least I'm happy to know it's not my hallucination.


Reply

I seem to recall that the early critics of Liveview made the comment that the energy consumption for this function would be likely to heat the sensor and cause excessive noise. I suspect the relationship between noise and temperature is well-known but the manufacturers don't think it worth publicising! What a surprise.


patrickh


Reply

You think you got problems? Where I process my colour negs in the back of the garage (at 38 deg C) is currently at 47 Confused

You digital guys got it easy Laughing


Reply

GrahamNR17 wrote:
You think you got problems? Where I process my colour negs in the back of the garage (at 38 deg C) is currently at 47 Confused
You digital guys got it easy Laughing


Graham I developed my first print the other day with a temperature of 24°C and the Ilford Multigrade developer went out of whack Sad

Please check in the accessory forum, I have a question for you about the alkali fixer Smile


Reply

Yes, thermal noise is a problem since early days of digital imaging. In fact, noise level will double for every 5 degrees Celcius.

There are various methods employed by digital back manufacturer when they introduce digital back at early days, for the two early digital back that I use:

1) Leaf DCB digital back - they use peltier cooling to cool down the sensor, however, Leaf do not hermatically sealed the sensor (unlike my other astronomical cooled CCD) and will fog when the temperature is lower than the dew point;

2) Megavision S2 digital back - the manufacturer suggest to shoot dark frame (multiple shots with with lens covered and shoot at the same temperature and duration of the actual shot) and bias frame (with lens covered and shoot at the fastest shutter speed). So that the actual shot will be reduced with the average dark frame and bias frame such as to compress noise.

In fact, the dark frame reduction method is used by DSLR camera manufacturers for long duration shots, say for more than 1 minute shot, as noise levels becomes unacceptable.


Reply

Heat is surely a problem. I had similar dissapointments with my Sigma SD14...the images were not to my satisfaction when it came to critical exposures.

Just last month i came back from a month long trekking expedition in the Himalayas...the images i took there are breathtaking and have given me immense satisfaction.

Now i was surprised why this happened...i used the same camera body, lens, tripod and shooting technique...but radically different image results!

And after much time the only answer that i figured out was the ambient temperature in which i had taken the images. In the mountains, the temperature was around 10 degrees celcius or lower...sometimes in minus. Compared to this, city temperatures in India usually are 35 degrees celcius and higher...in summers 40 degrees celcius wherever i have been living.

So, ambient temperature is a BIG factor in the image capture quality of DSLR's...particularly large image sensors. I did not notice such appreciable changes with my Panasonic Lumix FZ3.

Please have a look at one of the pictures i have posted on my Flickr page to see what i am talking about:

http://www.flickr.com/photos/sendhil76/3528465912/

This one image helped me in figuring out what was happenning as there is no noise at all in the image...something a lot of people observed and pointed out for such a low light levels image.


Reply

This is one reason I rarely use live view as it heats up the sensor.


Reply

GrahamNR17 wrote:
You think you got problems? Where I process my colour negs in the back of the garage (at 38 deg C) is currently at 47 Confused

You digital guys got it easy Laughing


I think I would die at 47 degrees.

If sweating was an Olympic event, I'd pick up a gold, easily!! Laughing