R5 firmware 1.6 -- Overheating Fixed?

AtticusLake

AtticusLake

Well-known member
Canon just released firmware version 1.6 for the EOS R5; the big add in 1.6 is the new "Auto power off temp" option, with settings "Standard" and "High". Since the release, I've been testing how the new option affects overheating. It seems to be a big improvement. So I thought I'd share my results.

All tests were shooting the TV, so the camera had some motion to process. The room was about 19-20 °C (66-68 °F). Obviously the 30 minute record time limit (7.5 minutes in 120 FPS) kicked in, so I just pressed record again each time the camera stopped, until it actually overheated and shut down -- or I got fed up and went to bed.

In all cases, once the camera shut down, I stopped the test. Of course you can let it rest a bit and get more time out of it, but how much time depends on how long you let it cool. With infinitely many options, I couldn't be bothered testing this. The camera was allowed to get stone cold between these tests.

Auto power off temp = Standard:
  • 4k HQ 23.98 warning @ 31m 02s, shutdown @ 40m 18s
  • 8k IPB 23.98 warning @ 25m 50s, shutdown @ 33m 57s
  • 4k LQ 59.94 warning @ 1h 11m, shutdown @ 1h 41m
  • 4k LQ 120 warning @ 14m 45s, shutdown @ 19m 53s
Auto power off temp = High:
  • 4k HQ 23.98 ran for 3 hours, no warning or shutdown, gave up
  • 8k IPB 23.98 ran for 4 hours plus, white warning after an hour or so, red warning intermittent, no shutdown
  • 4k LQ 59.94 ran for 3 hours, no warning or shutdown, gave up
  • 4k LQ 120 white warning @ 22m 50s, red warning @ 29m 45s, shutdown @ 35m 54s
In the cases where the camera ran for 3 hours or more, it just didn't seem to be getting hotter, so I couldn't see much point in continuing. The camera was pretty warm after these tests, but not crazy; though the memory card was too hot to hold. It cooled down pretty fast, though.

For a mirrorless camera without active cooling, this looks about as good I would expect.

Obviously I can't guarantee that you'll get these run times. Individual variations between cameras, different environments etc., will all have an effect. In particular in the 8k test, the warnings were showing intermittently after the first hour or so, so the camera was clearly close to the edge.

Please bear in mind that in "High", the camera can get hot, so watch out for low-temperature burns if you're holding the camera for a while.

I've uploaded a more extensive write-up to my web site: https://moonblink.info/MudLake/gear/r516
 
If you're feeling adventurous, you can take it apart, add a custom heatsink and thermal pad, and throw on an active cooling fav for good measure (see this DIY Perks video).
Canon did the same thing that Sony did with the A6300, they engineered it to fail so that you need to buy one of their cinema cameras.
 
I'm glad it's working for you, but it's really a Band-Aid solution to poor engineering.

You might be okay indoors, but not necessarily outdoors in direct sunlight, plus excessive heat elevates the sensor noise.
 
I'm sure you're right about sunlight. But I really don't know what you're comparing the R5 to. The Sony A1? It also overheats, eventually. Seems to me the R5 has easily reached par with all the other small, mirrorles cameras -- i.e. no active cooling. Of course, if you're willing to get a bigger, noisier camera, then get one with active cooling. The R5C is excellent, and so is the C70.

As for "poor engineering", I don't know what you're basing that on. The DIY perks video is cool, and cute, but totally irrelevant. DIY perks can duct the heat out to the back, but Canon absolutely CAN NOT. As a real product company, you ABSOLUTELY can not do anything which might harm your customers, and low-temp burns are on the minds of the electronics companies. DIY perks doesn't have to care because he's doing it for himself. So I'm quite sure Canon know how to do heat pipes etc., but they didn't on the R5 because there's nowhere to put the heat safely. Of course the R5C did exactly that and put the heat behind the fan.
 
AtticusLake said:
I'm sure you're right about sunlight. But I really don't know what you're comparing the R5 to. The Sony A1? It also overheats, eventually. Seems to me the R5 has easily reached par with all the other small, mirrorles cameras -- i.e. no active cooling. Of course, if you're willing to get a bigger, noisier camera, then get one with active cooling. The R5C is excellent, and so is the C70.

As for "poor engineering", I don't know what you're basing that on. The DIY perks video is cool, and cute, but totally irrelevant. DIY perks can duct the heat out to the back, but Canon absolutely CAN NOT. As a real product company, you ABSOLUTELY can not do anything which might harm your customers, and low-temp burns are on the minds of the electronics companies. DIY perks doesn't have to care because he's doing it for himself. So I'm quite sure Canon know how to do heat pipes etc., but they didn't on the R5 because there's nowhere to put the heat safely. Of course the R5C did exactly that and put the heat behind the fan.
Click to expand...
I know it makes me sound like a conspiracy theorist but the camera companies especially Canon are famous for their "cripple hammer". I watched a video that took apart the R5 and there were very basic things they could have done like using thin copper heat plate over the cpu.

The motivation is simple they don't want hybrid cameras to compete against their cinema line. There is no legal contract that prevents them from crippling a camera. It's probably one of the first thing they consider when developing a new camera how do they design it not to compete against their existing cameras. For a long time they imposed an artificial 29 minute rule. Many speculated that it was due to higher tax rate for European cameras. But it brought up all sorts of questions but why are American cameras limited? or why don't they give an option to pay more for one without the limit? Again this was the ole cripple hammer at work. For the record the over heating was initially an artificial limit they imposed. Evidence to this is a few months after the bad press of the over heating they quietly released a firmware "fix".

So its good its less prone to over heating but I wouldn't go so far as giving Canon any credit for fixing a problem they knew of before the release.
 
From a business perspective, what could of happened (and does happen all of the time but not saying it did in this case) is a contract expired.

Company A and company B legally agreed to do something or not do something for 12-18-24 months and then once the time has passed life changes. (That FW in 2020 barely did anything...it was just an attempt at damage-control.)

In my opinion, initially releasing a R5 which did not overheat (it has nothing to do with heat, lol) would have been very, very bad for Sony.

And that cannot happen in Japan.
 
I don't think they did it for Sony's benefit. They wanted two version of the same camera. One Photocentric R5 and the other Videocentric the R5c. At the time we didn't know they had plans to release the R5c, which by the way has it's own cripple to discourage people from using it for photos. They took away image stabilization. Sony has the same approach of having parallel lines of mirrorless cameras for video and photo.
 
They've both been doing the same thing for 10 years, so although we didn't know what it was going to be called, we knew - at least I knew - something was coming (and Canon Rumors and Canon Watch also said a more advanced video version of the R5 was coming a few months after the R5 was released).
 
Peter C. said:
I know it makes me sound like a conspiracy theorist
Click to expand...

Yup, 'fraid so. ;-)

Peter C. said:
... I watched a video that took apart the R5 and there were very basic things they could have done like using thin copper heat plate over the cpu.
Click to expand...

And what would that do? Does copper have some magic power to destroy heat?

No, it doesn't. The heat HAS to go SOMEWHERE -- that's basic physics. As I explained already, if you duct the heat out to a hot spot on the back, you get sued for low temp burns. For Canon, that is absolutely not on. DIY Perks can do it because DIY Perks isn't going to sue himself. So instead the R5 dissipates the heat throughout the camera, which leads to heat buildup.

The consipracy theorists have gone mental over the R5, and it's easy to see why -- Canon poured fuel on their own bonfire by releasing the R5 in a half-ready state. But this wasn't the first product to be released too early, and it won't be the last, and none of that makes the consipracy theorists right.

The reason that the consipracy theorists exist is because they have no clue how hard it is to actually develop a complex high-tech product. I've spent close to 40 years developing complex high-tech products -- not cameras, but I think I know a little about how this works. And how it works is WAY more complex than most people imagine. There are technical challeneges you wouldn't have a clue about if you hadn't done it, and pressure to get the product out on time is unbelievable. I was eating three meals a day at my desk, often 6 or 7 days a week, to get products out the door.

Not having to deal with this stuff in real life makes armchair punditry very easy. But that doesn't make it right.

Do Canon play the cripple hammer game? Yes. Everyone does. Ever heard of a company called Apple? It's hard not to see the 30-minute limit in that light, at this point. As far as that one goes, put on yer tinfoil hats and rant.

But if you're going to criticise the R5 for lousy engineering, then please say what you're comparing it to. In other words, someone please post a list of fanless cameras that can record 8k for 4 hours.
 
Did you ever walk by the coding department during those 40 years?

Just reset the R5's timer that appears to be programmed into the camera and then you can record all day, so we don't have to list any cameras because we have one right here! :)

https://www.youtube.com/watch?v=taUj9UCKG50

If you're deep in product development, I would think you understand conspiracy theorists more than most because you have firsthand knowledge of business and the way the world works. People like engineers are always skeptical and open-minded (at least the ones I've met who are smart and kooky and love to control the things they make/work on).

There are other videos of people measuring the heat inside the camera or barely recording with it and it was still overheating based on what appeared to be artificial countdowns (back in 2020).

And then two years later firmware fixes it? How do you explain that? They didn't re-design the camera, you know? (We're talking going from minutes to hours...)
 
But with all of that said, we'll always expect heat from electronics and whatever they did now is amazing - but just wish it would have been like that on day 1 when I had the camera (July 31st, 2020).
 
Atticus, you're probably one of the last people on this forum who doesn't believe the conspiracy theory
 
AtticusLake said:
And what would that do? Does copper have some magic power to destroy heat?

No, it doesn't. The heat HAS to go SOMEWHERE -- that's basic physics. As I explained already, if you duct the heat out to a hot spot on the back, you get sued for low temp burns. For Canon, that is absolutely not on. DIY Perks can do it because DIY Perks isn't going to sue himself. So instead the R5 dissipates the heat throughout the camera, which leads to heat buildup.
Click to expand...
Here is visual proof of how poorly they built the heat dissipation. You can jump to 3:51

The point is that purely from an engineering point of view they knew there would be a thermal issue and did nothing to address it because they don't want people using it primarily for video.


I do find it funny how defensive people get about their gear. That the idea a manufacture would do things intentionally to limit a camera for financial reasons is inconceivable (Princess Bride).
 
Last edited:
NorBro said:
Did you ever walk by the coding department during those 40 years?
Click to expand...

No, I didn't walk by it. I was running it. And we shipped a whole bunch of products (tablets & phones) on time, to budget, and ahead of our competitors.

If you're going to criticise my industry experience, then maybe you should tell us what yours looks like.

NorBro said:
Just reset the R5's timer that appears to be programmed into the camera and then you can record all day, so we don't have to list any cameras because we have one right here! :)
Click to expand...

Take a look at that video; it's 2 years out of date. Yes, they did originally use a timer, but they don't any more -- not for a long time. So there's nothing to reset. Yes, using a timer was a bad idea, and a lazy way to control heat, but they fixed that a couple of years ago. The whole point of my post is THAT YOU CAN IN FACT RECORD ALL DAY ON THE R5 RIGHT NOW. You don't need to reset any mythical timer. So I just can't see what point you're making.

NorBro said:
If you're deep in product development, I would think you understand conspiracy theorists more than most because you have firsthand knowledge of business and the way the world works.
Click to expand...

Exactly.

NorBro said:
And then two years later firmware fixes it? How do you explain that? They didn't re-design the camera, you know? (We're talking going from minutes to hours...)
Click to expand...

I think what happened is that Canon started with a really conservative idea of how hot a camera could safely get. Which is a good place to start. I'm very sure that a large part of that was avoiding low-temp burns -- a point I keep raising, because it's important.

Over time, they've improved the efficiency of their data processing -- the usual thing that happens is that as time goes by, they write the software to offload more and more work onto hardware assist / DSPs. And, they've clearly reset their expectations of how hot a camera can safely get. Run it on 4kHQ for 4 hours, and see how hot the CF card is. IMHO, they're pushing it pretty hard now, but hey, it seems to work.

I wonder if they did temp measurements of the Sony A1 and thought, whaddaya know, we can go hotter. But that's just speculation.

And still, no-one has posted the name of any other camera which can (thermally) do what the R5 does. I'm not claiming that the R5 is perfect, and as I've already said, it was clearly released before it was really ready (in terms of what the marketing was claiming). And it's still far from perfect. But if you're going to criticise its engineering, then please at least tell us what you're comparing it to.

NorBro said:
But with all of that said, we'll always expect heat from electronics and whatever they did now is amazing - but just wish it would have been like that on day 1 when I had the camera (July 31st, 2020).
Click to expand...

Absolutely agree, on all points. It was a big letdown on launch, and good Lord it's taken them time to fix it. But they seem to have pretty much got there -- and if you have worked in product development, you should have a healthy respect for how long these things can take.
 
You're either really naive or are providing humanity the benefit of the doubt. lol

It doesn't matter though and it's all good. If this was 2 years ago, I would be all about this convo but the camera is old and no one cares anymore.
 
ahalpert said:
Atticus, you're probably one of the last people on this forum who doesn't believe the conspiracy theory
Click to expand...

I wonder how many people on this forum have spent as much time in high-tech product development.

Conspiracy theorists exist because sometimes they're right. Governments and corporations do in fact do shady, dubious, unethical, illegal, immoral, and downright fattening things. I already mentioned Apple. I also mentioned the R5's 30-minute record limit. There are plenty of other examples.

The problem is that conspiracy theorists get carried away with their occasional successes, and how much fun it is to speculate like that, and they don't have the inside knowledge and experience to judge how probable their theories are -- so they get completely out of hand.

And just when you want to want to slap them all down and tell them they're crazy, the corporations come along and do another insane immoral thing, adding more fuel to the fire.

So yes, it can be confusing. But jumping on the conspiracy theory bandwagon every time a product has a flaw is just lazy. Believe me, there are plenty, plenty plenty of reasons why products get launched with flaws. In terms of software, if you're using a phone right now, bear in mind that the manufacturers probably have thousands, if not tens of thousands, of known outstanding bugs logged against it. Those bugs don't exist because the company wants their product to look bad -- they exist because that's how complex software is. Android, for example, was 35 million lines of code when I worked on it. By now I think it's over 60 million. Check me if you like, the source is online.

The idea that Canon would deliberately sabotage their products to make them look bad is pretty wierd. The idea that they have some kind of secret deal with Sony is just silly. There are plenty of technical reasons why the R5 would suffer from heat issues -- because it's doing an insane amount of data processing in a tiny, compact package.

Does deliberate product crippling play into it partly? Maybe a bit, but I'm guessing mostly not.

On the other hand, is the 30-minute limit a deliberate bit of product crippling, to protect their cinema camera line? I can't think of a more likely reason. So the conspiracy theorists would have my support there.
 
You must log in or register to reply here.
Back
Top