EOSHD testing finds Canon EOS R5 overheating to be fake

[Andrew Reid]

3 hours ago, wondo said:

if andrew is right, the big question that remains is why did they do it??? and why did they do it in such an excessive way?

if the camera had a hardware flaw and they would do a lot of pr to downplay it i would totally understand that.

but crippling a camera by firmware to a point where it becomes unusable does not make any sense.

limiting it to protect cinema eos sales?, ok, limiting it to protect sales of the next model?, ok, but totally ruining it?

development of this thing must have cost tons of money, and even if they wanted to protect cinema eos sales, there is still a bit of math do be done to understand whether this is reasonable by any means.

cinema eos cams are pretty much a niche product and i don`t see why they would protect them at such a high cost?

andrews findings are shocking, and if true, i`d really like to know the reason they did it?

there are a lot of statements on here about how much of a scandal this is and how bad canon is behaving, and given the fact that canon is not a chinese two man brand run by lunatics makes me even more eager to get to know the "mechanics" behind this...

Hello Wondo

Hard to say isn't it? I cannot figure out the business side of it at all. Surely the EOS R5 at 4400 euros is an important release that makes more profit overall than a more niche C200. So either it is internal politics that have played out badly, or possibly a buggy cripple algorithm. But really I do not have the answer. Canon owes us that. It could even be internal sabotage for all I know.

The strangest things for me are the following facts:

- Canon marketed it as 'comfortable on C300 III productions' and made a big thing about the 8K RAW, so were marketing even aware of the limitation?

- The cool down times are unreasonable. After a shut down at 62C, I can leave my camera on a desk for 1 hour turned off and come back to find only 5 mins 8K. There is no chance in hell any silicon or metal stays above 60C for that period of time inside a device that has no power. I even removed the battery, took the lens off and opened the card door to vent the internals. No 'weather sealing' at play here.

- There is only a small possibility that the EXIF temp is a long way off the CPU core temps. Also the Chinese guys with the heat guns were able to read the real surface temp of the chip and back of camera, which correlate closely to the EXIF temp we read from the intervalometer shots

- Why go to so much effort to build this absolute cutting edge machine.... Only not to let us use it?

It's a mega camera. That they saw fit to do this is so far beyond any realm of understanding.

People talk about plastic melting or safety regulations... But none of it can explain why 46C after 30 mins causes the camera to shut down immediately when switched to video mode, but 46C after 2 mins allows you to record 15 mins of 8K.

46C or even 62C are so far below what is considered unsafe.

The camera is not even hot to the touch.

Pulling the Cfexpress card out after the max 8K recording time is up, and that is hot but not uncomfortably... all normal. It gets hotter in my ProGrade card reader.

[Brian Flint]

I am retired now but have experience over many years as an electronic design engineer.

If there are 'timers' employed in the design then it is likely they would 'reset' to a zero or starting count state when the power is removed and then returned to  electronics. This would occur when the battery is removed. for a short period..

However if Canon was clever they could continuously write the count value of the timer(s) to a non-volatile memory  and when the power is resumed ( after battery removal ) the values of the count stored in the non-volatile memory could be returned to the timer(s). To do this there would have to be some non-volatile memory in the camera.

 

[HansD]

17 minutes ago, Brian Flint said:

I am retired now but have experience over many years as an electronic design engineer.

If there are 'timers' employed in the design then it is likely they would 'reset' to a zero or starting count state when the power is removed and then returned to  electronics. This would occur when the battery is removed. for a short period..

However if Canon was clever they could continuously write the count value of the timer(s) to a non-volatile memory  and when the power is resumed ( after battery removal ) the values of the count stored in the non-volatile memory could be returned to the timer(s). To do this there would have to be some non-volatile memory in the camera.

 

Canon has been doing that for a while! How else can they keep track of the shutter count without an extra battery inside?

[Andrew Reid]

4 hours ago, KarlL said:

Very interesting stuff! I'm a hobbyist Canon shooter myself, but I'm still on the EOS 6D since until now I couldn't find anything that really pushes me towards a new camera. The R5/R6 might be the one that actually catch me. However, after reading all of this, I get doubtful.

Besides that I'm an engineer for embedded software / electronics (in automotive) and I would like to share some of my knowledge here.

1. As horshack already pointed out the temperature written in the EXIF might not be the temperature of the main controller, but the one of another chip that's supposed to be a "good"/"reasonable" representation for the inner camera temperature.

Even if it turns out that the EXIF "camera temperature" has no relation to the temperature of the DIGIC X chip, the test results still confirm fake overheat timers. Why - let me explain:

1. CDA TEK's (BTM_PIX)'s app reads the temperature inhibitors. If the EXIF temp sensor is on a separate board and is not reading the CPU core temps, it would still increase if the CPU got hotter. The CPU is the primary heat source in the camera body and heats up the entire unit. We show that the temp inhibitors toggle on over time, even as the EXIF reported temp stays level at 46C or even decreases (as well as camera body external temps to the touch).

2. After powering off an electronic device or CPU, the part dissipates most of the 90C+ heat within seconds. If opening the card door, the chipset even then has a venting hole. Hot air escapes and dissipates quickly. When I switched off the camera for 1 hour and opened the card door, removed battery and lens, the temp inhibitors were left on even after 1 hour. It is impossible for the CPU to maintain 60C+ temps (33C above ambient temperature!) in this situation with no power.

3. The temp inhibitors flag ON even when the camera is not recording. They flag ON in the Wifi menu. ON in live-view. ON during stills shooting. If these processes are causing problematic temperatures for the CPU, it should shut the camera down in stills mode, but it doesn't... it can be left on.

Quote

2. From our electronics I know that we usually measure 4-5 temperatures from different chips and the differences are quite astonishing sometimes. While the main chip under "heavy duty" might be at around 90°C within a few minutes, peripheral chips show the temperature increase with a (for me) surprisingly high delay of a few minutes. This is partly due to the fact that thermal operations take quite some time, especially if the heat distribution medium is air. This depends on the placing of the chips of course.

Sure, but see my point 1 & 2 above.

A separate temp sensor would not stay at 46C if a main CPU at 90C started heating the body and internal air.

And heat distribution via air does not take 1 hour between a hot silicon chip and the ambient air right next to it and in/around the card slot.

Quote

3. The issue might not be overheating of the main controller, but of another chip, e.g. of the sensor read out circuit

The Chinese heat gun shows this not to be the case, but anyway, if the sensor is the limiting part that gets too hot, it would have to shut down during an 8K (45 megapixel) photo shoot at the same time it shuts down for 8K video.

Quote

4. There has been some discussion about maximum temperatures of electronic chips. It is correct that 90°C is still in the comfort zone of most electronic chips. However, you have to take into account that there is a maximum allowed temperature for surfaces is 65°C (Metal) and 85°C (Plastic) due to safety reasons

Are you talking about consumer health & safety regulations regarding contact with hot surfaces?

Or stability and integrity internal materials?

This depends on the plastic. You can have plastic parts inside the camera that don't melt until 170C.

And as for metal....Well, even solder does not melt at 46C!

Quote

I saw some videos in which they managed to shoot 4k HQ continuously for hours, by removing the cards from the camera.

See?

The sensor readout isn't the heat limited part.

Quote

If I combine the "safety" stuff with this I get to the conclusion that the issue might be the memory card getting too hot in terms of safety regulations.

Nope.

The problem occurs even with an SD card, which can be used for unlimited 4K recording in other devices without issue.

They don't get very hot (as in 80C+), maybe 50C max.

You can read/write for hours on an SD card in a reader so it isn't about SD card safety regulations.

There are cameras like the S1H that write 6K or 4K at 400Mbit to SD cards. Higher data rates than EOS R5 4K HQ.

Quote

Here's my scenario: I record for let's say 45 minutes, before my CFexpress card is full. I want to switch cards quickly -> I directly open the card slot and pull out a card that has a temperature of ~80-90°C. This is not acceptable by means of safety.

They don't get that hot though. I know, I've tried. This is just speculative on your part I'm afraid.

The Cfexpress card I have does get 'a bit warm like a cake'... but not scolding!

You can use it in a device that records 4K RAW for hours.

So I hardly think that there's anything about the comparatively low-bitrate 4K HQ 8bit H.264 (with CLOG disabled) on the EOS R5 that causes the Cfexpress card to heat up beyond the safety regulations.

Quote

Something I can imagine here is that they cannot directly measure the temperature of the memory cards.

Yes they can.

Quote

I hope Canon will release detailed information about this. Because even though I'm a stills shooter only, I don't want to be tricked. And I must admit I'm on the edge to switching to Sony, anyway.

I agree. They owe us a full and frank explanation.

In my opinion lying to customers is not a good look!

[Ilkka Nissila]

An IR thermometer only records the surface temperature, it may be that components are hotter inside. Putting the camera in a freezer for 25 min reportedly restores the camera's video recording capabilities to baseline (see experiment and observations by Jesse Evans at https://www.fredmiranda.com/forum/topic/1658635/4). Thus the camera is clearly monitoring its temperature and reacting accordingly, but after use that leads to overheating, it seems to be difficult to get the internal temperature to decline sufficiently without extreme measures.

 

I think Canon could tweak the operation of the camera; for example, if it is sitting in a menu, there should not be any need to continue live view operation and they could program the camera so that it doesn't read and process sensor data during this time, and run the processor at lower speed to minimize heating when in a menu browsing and adjusting settings. Also there could be recommendations as to what kind of live view settings to use to maximize subsequent recording time, and so on. It can't be that the camera loses its ability to record high-quality video by merely being switched on.

[Andrew Reid]

15 minutes ago, Ilkka Nissila said:

An IR thermometer only records the surface temperature, it may be that components are hotter inside. Putting the camera in a freezer for 25 min reportedly restores the camera's video recording capabilities to baseline (see experiment and observations by Jesse Evans at https://www.fredmiranda.com/forum/topic/1658635/4). Thus the camera is clearly monitoring its temperature and reacting accordingly, but after use that leads to overheating, it seems to be difficult to get the internal temperature to decline sufficiently without extreme measures.

Why does it need 25 mins in the freezer to get back to baseline temp?

It defies logic.

You heat a metal baking tray to 60C and try putting it in the freezer! It's cold after a minute.

It may be that extreme temp changes do change the artificial timer. It's still fake though. Otherwise the camera would be ready to go again after a couple of mins in the freezer.

I have officially received permission from The Girlfriend for some time with the fridge later and some frozen peas.

But to be honest, writing is already on the wall as far as I'm concerned.

And Jesse did not tell us how long his measurements on all the parts and camera areas took after he stopped recording 8K. So the 25 mins isn't the total time of idle off-time and even if it is, it doesn't show that the timer isn't artificially locking us out from the camera for unnecessarily long periods.

[BTM_Pix]

20 minutes ago, Andrew Reid said:

I have officially received permission from The Girlfriend for some time with the fridge later and some frozen peas.

OK but after you've finished doing your 9 1/2 weeks remake perhaps you could do some more testing of the R5.

[KarlL]

44 minutes ago, Andrew Reid said:

The problem occurs even with an SD card, which can be used for unlimited 4K recording in other devices without issue.

They don't get very hot (as in 80C+), maybe 50C max.

You can read/write for hours on an SD card in a reader so it isn't about SD card safety regulations.

There are cameras like the S1H that write 6K or 4K at 400Mbit to SD cards. Higher data rates than EOS R5 4K HQ.

They don't get that hot though. I know, I've tried. This is just speculative on your part I'm afraid.

The Cfexpress card I have does get 'a bit warm like a cake'... but not scolding!it really seems to be a c

You can use it in a device that records 4K RAW for hours.

Yes they can.

I agree. They owe us a full and frank explanation.

In my opinion lying to customers is not a good look!

So I hardly think that there's anything about the comparatively low-bitrate 4K HQ 8bit H.264 (with CLOG disabled) on the EOS R5 that causes the Cfexpress card to heat up beyond the safety regulations.

Hi Andrew, thanks for the information you are providing.

I think I may have messed up my point about the memory card. I don't mean that the memory card overheats due to the writing, but because it's close to something hot.

My "idea" was, that
- something in the camera - e.g. the main controller - leads to internals heating up
-- no one cares about temperatures beyond the regulations of surfaces, because it's internal
- this also leads to the memory cards getting hot
-- memory cards can be removed
-- Removal parts apply to the temperature limits defined by safety regulations

From the external forum:

Quote

CFExpress card temperature upon ejection from the camera - Label / Top side: 134f

134F is about 56°C. Safety regulation requires a maximum temperature of 55° C for "prolonged" and 60° C for short periods only.

But this does not explain why it's taking so long to cool down.

 

44 minutes ago, Andrew Reid said:

Quote

Something I can imagine here is that they cannot directly measure the temperature of the memory cards.

Yes they can

Just out of interest: how do you know?

 

44 minutes ago, Andrew Reid said:

They owe us a full and frank explanation.

I fully agree with you on this point. Even if they say "The current generation of our DIGIC X can only manage 65°C, because XYZ" it would be OK for me.

But like this I feel fooled and that's not the feeling I should have as the customer.

[TheBoogieKnight]

5 minutes ago, KarlL said:

Hi Andrew, thanks for the information you are providing.

I think I may have messed up my point about the memory card. I don't mean that the memory card overheats due to the writing, but because it's close to something hot.

My "idea" was, that
- something in the camera - e.g. the main controller - leads to internals heating up
-- no one cares about temperatures beyond the regulations of surfaces, because it's internal
- this also leads to the memory cards getting hot
-- memory cards can be removed
-- Removal parts apply to the temperature limits defined by safety regulations

From the external forum:

134F is about 56°C. Safety regulation requires a maximum temperature of 55° C for "prolonged" and 60° C for short periods only.

But this does not explain why it's taking so long to cool down.

 

Just out of interest: how do you know?

 

I fully agree with you on this point. Even if they say "The current generation of our DIGIC X can only manage 65°C, because XYZ" it would be OK for me.

But like this I feel fooled and that's not the feeling I should have as the customer.

SMART which is featured in all CFExpress cards as far as I know allows reporting of temperatures.

[KarlL]

29 minutes ago, Andrew Reid said:

Why does it need 25 mins in the freezer to get back to baseline temp?

It defies logic.

You heat a metal baking tray to 60C and try putting it in the freezer! It's cold after a minute.

It may be that extreme temp changes do change the artificial timer. It's still fake though. Otherwise the camera would be ready to go again after a couple of mins in the freezer.

I think, the heat does not go down that quick because the camera body functions as an isolation.

I would compare it to putting beer in the freezer:

- if you put just the bottle in, it's frozen within 90 minutes

- if you put it in a plastic box it's taking quite longer.

[arant.joseph]

2 hours ago, Andrew Reid said:

Plastic melting point is 170C

Oh I'm not saying it will melt, simply saying that over time the plastic components will experience a loss in overall structural stability.  I had a cheap ASUS laptop that had a mostly plastic exterior and could have fried an egg, and the first part to go structurally was the section closest to the CPU.

[KarlL]

1 minute ago, TheBoogieKnight said:

SMART which is featured in all CFExpress cards as far as I know allows reporting of temperatures.

Ah, that's true! I didn't think about it, but yes it's like almost any memory nowadays.

Thanks!

[Andrew Reid]

It's not a plastic box though. It's alloy

[kingmouf]

So maybe it would be possible that in a few weeks or months time, we will see Canon posting a service note and state that if you take your camera to an authorized service center, they can perform "an upgrade" for x amount of money so that such limitations may be lifted. It would be similar to the updates we have seen in other cameras (or was it Canon itself as well?) that you paid an amount of money to have a feature upgrade such as support for an specific codec. So when Canon eventually drops the price of the camera, voila! it will be able to charge the same amount!

To be honest though, in the electronics industry things like this happen ALL the time. Firmware is the closest possible piece of software, so things do get tricky there and a lot of products with practically the same hardware have only changes in firmware. If Canon has done this thing here, I guess it got a little bit more clumpsy than the majority of the industry.

On the other hand, I think that Canon could be playing a very caution game here. Initial production of sensors and chipsets could very well be prone to overheating issues and maybe production is not stable enough to ensure a proper supply of chips of specific capabilities, meaning that Canon for safety went for the lowest denominator. After all, with the kind of clientelle that Canon has, dependability and reliability are top priorities. As such, Canon went with the safest solution and maybe eventual firmware updates will provide the time required to iron out more proper ways to measure temps and apply safety algorithms. Just saying.

[Ilkka Nissila]

20 minutes ago, Andrew Reid said:

It's not a plastic box though. It's alloy

From the CPU, there are layers of static air, circuit board, static air, metal and plastic at least. There is no effective thermal conduction path to the metal chassis either from the outside or from the processor. Thus the processor cannot effectively disseminate heat and outside cooling has only a delayed effect on the internal temperature of the components.

[horshack]

2 hours ago, Andrew Reid said:

Even if it turns out that the EXIF "camera temperature" has no relation to the temperature of the DIGIC X chip, the test results still confirm fake overheat timers. Why - let me explain:

1. CDA TEK's (BTM_PIX)'s app reads the temperature inhibitors. If the EXIF temp sensor is on a separate board and is not reading the CPU core temps, it would still increase if the CPU got hotter. The CPU is the primary heat source in the camera body and heats up the entire unit. We show that the temp inhibitors toggle on over time, even as the EXIF reported temp stays level at 46C or even decreases (as well as camera body external temps to the touch).

2. After powering off an electronic device or CPU, the part dissipates most of the 90C+ heat within seconds. If opening the card door, the chipset even then has a venting hole. Hot air escapes and dissipates quickly. When I switched off the camera for 1 hour and opened the card door, removed battery and lens, the temp inhibitors were left on even after 1 hour. It is impossible for the CPU to maintain 60C+ temps (33C above ambient temperature!) in this situation with no power.

3. The temp inhibitors flag ON even when the camera is not recording. They flag ON in the Wifi menu. ON in live-view. ON during stills shooting. If these processes are causing problematic temperatures for the CPU, it should shut the camera down in stills mode, but it doesn't... it can be left on.

Sure, but see my point 1 & 2 above.

A separate temp sensor would not stay at 46C if a main CPU at 90C started heating the body and internal air.

And heat distribution via air does not take 1 hour between a hot silicon chip and the ambient air right next to it and in/around the card slot.

The Chinese heat gun shows this not to be the case, but anyway, if the sensor is the limiting part that gets too hot, it would have to shut down during an 8K (45 megapixel) photo shoot at the same time it shuts down for 8K video.

1. There will be a point where both DIDIG/SDRAM and internal temps reach apparent homeostasis. Under constant processing load the hottest chips will heat up immediately and reach a plateau quickly, which will be based upon their thermal profile relative to ambient. The ambient temperature around the chips and inside the camera chamber will rise more slowly (hysteresis). As ambient continues to slowly rise the heat-generating chips will get hotter as well but on a slower ramp. This thermal feedback loop will slow over time as the camera's dissipation reaches homeostasis with the heat generation.

2. After powering off the hottest chips will cool immediately, however the ambient temperature around the chips will not. When power and processing load is reapplied those chips will quickly heat back to their pre-shutdown temps since their temp is a function of heat generation relative to ambient temp.

3. What you call "temp inhibitors" is a combination of the latent ambient temperature and logic that is projecting the ongoing heat ramp, which factors in the thermal lag affects described above (hysteresis).

The fact the freezer experiment posted on Fred Miranda returned the camera to full video operation times in only 25 minutes serves to both disprove the "timer-based cripple" theory while also demonstrating how the thermal management algorithm is actually working. The logic is serving two purposes - 1) Keep all chips on the PCB within their thermal design limits and 2) Project how much time remains before the chips will reach their thermal limits. The logic for #2 must factor in thermal hysteresis, which to the untrained will may look like intentional crippling and temperature-indifference since the algorithm wont respond as immediately or definitively as one would expect based on instantaneous temps but in reality is actually projecting out a longer-duration thermal ramp. You can see the result of this calculation in the available video-time value reported to the user in the UI.

 

[MurtlandPhoto]

This is truly amazing testing @Andrew Reid! I was honestly skeptical in the very, very beginning that Canon could possibly fake the camera overheating as a means to cripple it and protect the cinema line. But, wow. Using their own tools and measurements shows the damning truth. Your test results should be used as exhibits A-Z in a class action lawsuit. Truly sad that Canon thought they could pull this $hit.

[Emanuel]

Nothing new to see here. Pure story already told and seen. They all tend to do it or the majority at least. So, we need a new order telling them they will end not nice in the picture in this age of digital revolution and internet... That's it.

I still hope for popular demand they will be back. They did it with 5DII, even though there was no C line yet and that was actually the beginning of theirs.

[ajay]

How about opening the back of the unit w/o disconnecting anything. Blow air across the inside of the camera using a small muffin fan and record as you normally would. The temperature should definitely stay cooler. Check run times and recovery times.

If it's crippled by firmware, times shouldn't change. If recovery times are drastically reduced, it is strictly their way of managing the temperature of the camera.

My guess is, it's probably both. But it would be interesting to see how a fan blowing directly on the components changes things.

[Video Hummus]

30 minutes ago, ajay said:

How about opening the back of the unit w/o disconnecting anything. Blow air across the inside of the camera using a small muffin fan and record as you normally would. The temperature should definitely stay cooler. Check run times and recovery times.

If it's crippled by firmware, times shouldn't change. If recovery times are drastically reduced, it is strictly their way of managing the temperature of the camera.

My guess is, it's probably both. But it would be interesting to see how a fan blowing directly on the components changes things.

Yes.

Someone needs to take the bloody back panel off without disconnecting the screen and be able to power the camera on and record 8K. Take that shit into a walk-in freezer at -4C or 0C with a fan and blow freezing temps over the back circuit board...if it overheats or doesn’t slow down any of the timers than the timers are a closed loop and temperatures do not effect them! Meaning, it’s all software regulated.

If the internals being exposed to freezing temperatures doesn’t improve the “overheating” then Tilta’s stupid cooling kit is beyond ridiculous.

The fact that there has been multiple independent test of the camera overheating roughly the same time in cold, hot, and extremely hot conditions tells me that the outside ambient temperature has nothing to do with the cool down timers at all. Period.

4 hours ago, Andrew Reid said:

The cool down times are unreasonable. After a shut down at 62C, I can leave my camera on a desk for 1 hour turned off and come back to find only 5 mins 8K. There is no chance in hell any silicon or metal stays above 60C for that period of time inside a device that has no power. I even removed the battery, took the lens off and opened the card door to vent the internals. No 'weather sealing' at play here.

Leave it in a fridge for 1 hour...I bet you it will tell you 5 mins of 8K left as well. Make sure to put a beer in there as well so you can drink it after your realize what your 4400 euros got you.