TeamSpeed wrote:
Yeah, I have a hard time thinking that about any kind of consumer electronics would work after sustained stints in my oven heated to the temperature of boiling water. Even my car keeps its coolant temps below that temperature and that is primarily mechanical.
Even at 70C, I have a hard time thinking that electronics constantly heated to temps that cook meat to over well done shoe leather consistency and gets rid of about any harmful pathogen in our food is good either.
And yet, if you use a Macbook to edit your photos or videos, it is going to constantly hit 98-100c and perform thermal throttling.
I don’t believe image sensors are allowed to get to such temps tho.
Image sensors can operate at high temperatures, but they are noisy at high temperatures. A sensor at 40° C (104° F) will be noisier than one at 20° C (68° F), and that will be noisier than the same sensor at 0° C (32° F). There's no way around this, which is why good thermal design is so important for stills.
Temperature doesn't need to get so high it causes the camera to shut down or an indicator to light for it to cause excess shadow noise (thermal noise) in still images. I don't know if Canon's baked-in RAW file noise reduction firmware increases NR (reducing shadow detail) as temperature increases, but if the camera can't dissipate heat well, using temperature proportionate NR in firmware to help mask the problem is an obvious, but less than ideal, solution. I'm surprised no one seems concerned about this - or maybe I've missed a revelation somewhere.
Nobody is concerned because nobody knows. Everybody is concerned that the main board may be operating at temperatures that can cook beef or even boil water. Right now that seems to be a bigger concern than how the ISO was processed.
dmcphoto wrote:
Image sensors can operate at high temperatures, but they are noisy at high temperatures. A sensor at 40° C (104° F) will be noisier than one at 20° C (68° F), and that will be noisier than the same sensor at 0° C (32° F). There's no way around this, which is why good thermal design is so important for stills.
Temperature doesn't need to get so high it causes the camera to shut down or an indicator to light for it to cause excess shadow noise (thermal noise) in still images. I don't know if Canon's baked-in RAW file noise reduction firmware increases NR (reducing shadow detail) as temperature increases, but if the camera can't dissipate heat well, using temperature proportionate NR in firmware to help mask the problem is an obvious, but less than ideal, solution. I'm surprised no one seems concerned about this - or maybe I've missed a revelation somewhere. ...Show more →
I am concerned about this, especially the long exposure thermal noise. It is in fact one of the main reasons why I got the 5DSR and not the A7rII in 2015. The mirrorless camera had much worse thermal noise. There is a workaround with using LENR, but that doubles the read noise and reduces valuable exposure time during good light, to half the time. Using a single black frame and subtract it in post-processing is not a good option due to varying exposure times and sensor temperatures during a single session.
It is a separate topic though, and I will raise the question at the time we get a higher resolution R body from Canon which will be a more specialist camera for landscape. The R5 is a camera with much broader ambitions, and I don't expect it to perform particularly well for thermal noise in the first place.
alundeb wrote:
I am concerned about this, especially the long exposure thermal noise. It is in fact one of the main reasons why I got the 5DSR and not the A7rII in 2015. The mirrorless camera had much worse thermal noise. There is a workaround with using LENR, but that doubles the read noise and reduces valuable exposure time during good light, to half the time. Using a single black frame and subtract it in post-processing is not a good option due to varying exposure times and sensor temperatures during a single session.
It is a separate topic though, and I will raise the question at the time we get a higher resolution R body from Canon which will be a more specialist camera for landscape. The R5 is a camera with much broader ambitions, and I don't expect it to perform particularly well for thermal noise in the first place.
I've been wondering if there are still plans for a higher resolution R5 (R5r?) and if so when it might happen. Literally years ago there was going to be a 5DSr-II. That was scrapped and the rumored 5DSR successor turned out to be the R5 instead. Now I don't even see rumors of a high MP Canon camera.
dmcphoto wrote:
I've been wondering if there are still plans for a higher resolution R5 (R5r?) and if so when it might happen. Literally years ago there was going to be a 5DSr-II. That was scrapped and the rumored 5DSR successor turned out to be the R5 instead. Now I don't even see rumors of a high MP Canon camera.
I've been thinking the same lately. The R5 has become to the 5DS what the 1DX was to the 1DS III. We will see. These thoughts won't rush me into buying the R5 prematurely.
dmcphoto wrote:
I've been wondering if there are still plans for a higher resolution R5 (R5r?) and if so when it might happen. Literally years ago there was going to be a 5DSr-II. That was scrapped and the rumored 5DSR successor turned out to be the R5 instead. Now I don't even see rumors of a high MP Canon camera.
Not something I would expect anytime soon. It seems like the limiting factor is simply the technology to make higher resolution sensors is not quite there yet. It can be done, but checking a few places it seems like going from 45 to 60 would more than double the cost. Maybe in a couple of years.
I call BS again. I went through the Chinese link. The timer kicked in and it shuts down. He waited a few minutes and then disassembled it. Then he puts the battery in and says - "see the time is reset and no heat warning" - but he does not run it after.
I'm no engineer but it cooled down because he had it apart. Also based that sham of a test he did not state how long he had the battery out.
Looks like some other people got caught with their pants down again.
dmcphoto wrote:
I've been wondering if there are still plans for a higher resolution R5 (R5r?) and if so when it might happen. Literally years ago there was going to be a 5DSr-II. That was scrapped and the rumored 5DSR successor turned out to be the R5 instead. Now I don't even see rumors of a high MP Canon camera.
alundeb wrote:
I've been thinking the same lately. The R5 has become to the 5DS what the 1DX was to the 1DS III. We will see. These thoughts won't rush me into buying the R5 prematurely.
Yeah, it really sucks. I kept using my 1Ds IIIs for a few years until the 5DsR. Unfortunately now the R series has the RF mount, so it's more difficult to avoid. I will still have hopes for high resolution for about two years.
I call BS again. I went through the Chinese link. The timer kicked in and it shuts down. He waited a few minutes and then disassembled it. Then he puts the battery in and says - "see the time is reset and no heat warning" - but he does not run it after.
Exactly, job half done. After taking the internal "date and time" battery out and inserting it again, the screen shows no overheating and 15 mins record time. HOWEVER, he did not try to record anything to check what the real record time would be. That would have been a very useful piece of information, hopefully someone else will fulfill the task.
He measures the temperature on a metal plate with an IR thermometer?
As said before, the emissivity of blank metal is very low. Emissivity has to be close to 1 for the IR image to show the temperature of what you are pointing at, and not the ambient temperature reflected by the metal. The temperature reading is surely way off, showing a much lower temperature than the surface actually has. Let alone the chip beneath the clip.
In this post I proposed an alternate way to prevent the R5's firmware from saving the thermal management state information to NVRAM and possibly achieving the same result as removing the internal NVRAM battery.
This morning I performed an experiment on my RP demonstrating how the camera settings are lost when the battery-door latch sensor is defeated. Normally opening the battery door causes the camera to perform an orderly shutdown, same as using the power switch. This commits any settings changes made during the power-on session to NVRAM. In this experiment I defeat that process by holding-in the battery-door latch sensor and removing the battery. This causes settings changes to be lost - in this case, my aperture setting.
Here's a video of my RP experiment:
Doing this on the R5 will likely require taping a plastic tab to the inset latch sensor, since holding the sensor for the duration of recording video for the experiment wouldn't be feasible.
snapsy wrote:
In this post I proposed an alternate way to prevent the R5's firmware from saving the thermal management state information to NVRAM and possibly achieving the same result as removing the internal NVRAM battery.
This morning I performed an experiment on my RP demonstrating how the camera settings are lost when the battery-door latch sensor is defeated. Normally opening the battery door causes the camera to perform an orderly shutdown, same as using the power switch. This commits any settings changes made during the power-on session to NVRAM. In this experiment I defeat that process by holding-in the battery-door latch sensor and removing the battery. This causes settings changes to be lost - in this case, my aperture setting.
Here's a video of my RP experiment:
Doing this on the R5 will likely require taping a plastic tab to the inset latch sensor, since holding the sensor for the duration of recording video for the experiment wouldn't be feasible....Show more →
I can't get more than 5 minutes 8k so I am not trying your trick now. Reset everything on the camera and that did nothing. If someone pms me with another firmware I believe that loading it will reset the timer. A camera like this most likely has gone through 100 FW revisions and I just need one.
Andrew J wrote:
I can't get more than 5 minutes 8k so I am not trying your trick now. Reset everything on the camera and that did nothing. If someone pms me with another firmware I believe that loading it will reset the timer. A camera like this most likely has gone through 100 FW revisions and I just need one.
Can you describe what steps you took and in what sequence, in precise detail?
snapsy wrote:
In this post I proposed an alternate way to prevent the R5's firmware from saving the thermal management state information to NVRAM and possibly achieving the same result as removing the internal NVRAM battery.
This morning I performed an experiment on my RP demonstrating how the camera settings are lost when the battery-door latch sensor is defeated. Normally opening the battery door causes the camera to perform an orderly shutdown, same as using the power switch. This commits any settings changes made during the power-on session to NVRAM. In this experiment I defeat that process by holding-in the battery-door latch sensor and removing the battery. This causes settings changes to be lost - in this case, my aperture setting.
Here's a video of my RP experiment:
Doing this on the R5 will likely require taping a plastic tab to the inset latch sensor, since holding the sensor for the duration of recording video for the experiment wouldn't be feasible....Show more →
I wonder if you can get the same effect by using an LP E6 AC adapter and simply cutting the power.
