Is Scotch tape the answer to your Canon EOS R5 overheating issues?

Aug 22, 2020
101
110
I give credit when due even on the interwebs, and all I have seen so far are whiners and armchair engineers making idle speculation or outright controversy out of nothing. Speculation taken too far makes people fools. No offense, but IT doesn't make you an expert in the Canon R5 thermals. Neither do people pointing a FLIR at the R5 suddenly know all about the situation. None of these tests equate to die temperatures for ASICs and other internals, and you have no idea what temperature threshold is the protection limit for these parts. Some parts really don't like to be hot over their lifetime and so for reliability are kept at lower temperatures.

I have specific experience with high speed circuit design and thermal management and will leave it at that, but I would never go around speculating I understand a design unless I did the thermal analysis or pulled out schematics or looked at the code for ASICs, FPGAs or uControllers myself. To pretend I know, even with my experience, is a lie unless I have done adequate testing and reverse engineering. On a product like this, insanely compact and complicated, that isn't going to happen from some youtube test. To sit around doing inane amateur tests and then to make biased assumptions about artificial timers and blah blah blah is people stoking their ego, NOT honest intellectual discourse. Sometimes saying "I don't know" is the best course, but too many people don't like the feeling of admitting it, even when the subject matter is way out of their area of knowledge.

Calling the time limits artificial or fishy in the first place is an assumption and an emotional label. Engineers have to make a choice on how to put in stops around various physical realities of the components and their interaction as a system. Even if the record times WERE arbitrarily chosen, SO WHAT?! Canon told us about it including the part about other camera activity reducing those times. So for the millionth time I suggest people go buy another camera if you don't like it instead of falsely acting like some detective! Since people are not having an honest discussion about how electronics and product engineering actually work, I will continue to down play what these people are saying because the way they are going about this is amateurish, has an unintelligent and non-rigorous tone, and the vast majority of them have an agenda of some sort: to get clicks or prove the camera is flawed. It isn't; Canon told us exactly how it behaves before shipping started.

There are plenty of reasons the camera internals might heat up just in the menu or shooting stills. In my own tests, just using the EVF for long periods of time generates heat. IBIS also seems to generate heat. Try a 1 hour timelapse at 5 second intervals and just disabling those two is the difference between a warm and cold body at the end of the hour. So I could draw conclusions from my own amateur test that something about the heat and cooling is real, not artificial, but why bother? The camera worsk for me as designed. I'm sure everyone has heard that removing the CFE card helps. SERDES for the CFE might always runs idle characters even without data flowing. We could speculate all day. I could do some pretty fancy tests if I really wanted to, or I could buy the right tool for the job (which I did, and it's a pretty great camera with a few flaws like every product must have).

If Canon has some errors to fix or optimizations for the firmware that make this better, great. Not uncommon for new product. If they actually made a hardware mistake and fix it, great, but I find that very unlikely. However, until I hear well researched logic that doesn't fall apart trivially I will keep calling out all these hokey tests and assumptions for what they are: hot air (see what I did there)?

The only testing I've seen that actually benefits anyone is from those actually working with the camera instead of trying to make it fail, or providing workarounds like using external recorders- those people are accepting the camera for what it is at least instead of continuing on this path of labeling things nefarious or fishy or artificial or whatever.

That's all fair and square and my assumption is still that this is due to the first-gen IBIS unit in the camera which can't take the heat. But several people tested this camera at 4C, 15C, or 25C and all of them find that even just some photos or idle time will decrease the record times to a handful of minutes with recovery times of +30min to be able to shoot in the advertised modes again. Thus I do truly wonder whether this device is not very poorly engineered since it does not really deliver consistently what it advertises and needs apparently super special care to not fall apart internally (some hyperbole there, but it is meant to illustrate a point). If these internal heat problems are real I would worry about shooting sports in the summer in hot regions under the sun, too.

Besides, why would it be unlikely for Canon to have made a hardware mistake? This is there first attempt at IBIS. Plenty of space for issues already there. And you only accept the tests by people that accept the camera with its limitations and thus work around them? Isn't that some really strong bias? Literally every review so far has said that 8K and 4K HQ are ruled useless due to recovery time, but this is irrelevant to you because they have not fiddled around with the camera to open up card slots or attach external recorders without batteries inserted to make things work that are on the spec sheet? Some of these measures also render the weather sealing moot. So at that point it's not an EOS R5 anymore which is weather sealed and offer 4K HQ.

Meanwhile, I really hope there is some simple FW fix around the corner that unlocks the true potential of this camera since the 4K HQ is truly beautiful and it deserves to be able to be used by capable hands without the immense restrictions that the current situation puts on them.
 
  • Like
Reactions: 1 users
Upvote 0
i am sorry, i am late, who is on duty today?) i mean, is it been discussed that the method of pulling out the battery brokes the recorded file?
so, the method confirms canon's counter but nothing else, you cannot record 8k continuously

But you don’t have to wait a couple of hours to continue recording in 8K.

The big problem for me is that it seems the camera is reading it’s temp from a timer rather than getting a value from the thermometer.

I hope this is fixed in a firmware update.
 
  • Like
Reactions: 1 user
Upvote 0
Jul 16, 2012
486
298
Its all just overclocking really.

Yes Canon might have done something to sell this at a price point, just like Intel has in the past. If they have thats less a moral catastrophe and more potentially a huge opportunity.

Yes theres the usual groups of 'it willl explooode' vs the 'theres no risk at all'. Depending on the particular CPU, you either saw massive increases or dead chips. My personal view is this is more like trying to overclock a laptop than a desktop. I salute those who will carry out those early tests to see which it is, it wont be me.
 
  • Like
Reactions: 1 user
Upvote 0
What people are assuming here is that it's ONE temperature reading. I wanted to write an article specifically about this one video, but I'm tired of talking about overheating.

you can't do prediction based upon one temperature value, but need a series to be able to tell if the internal temperatures are going up or going down and rate in which they are. only with that, and armed with foreknowledge on how much atypically rate in which temperature increases with recording can you roughly predict how long you can record for.

Canon is probably keeping those in local memory and dumping it all to nvram when the camera shuts down.

not writing that probably just gets the camera to throw up it's arms and go well this is screwed .. let's start with 5 minutes.

There is STILL some wonky code in there. but is it wonky with purpose or just the fact that someone made bad assumptions when writing firmware? That we don't know, and probably never will.
 
  • Like
Reactions: 4 users
Upvote 0
Aug 27, 2019
667
1,414
What people are assuming here is that it's ONE temperature reading. I wanted to write an article specifically about this one video, but I'm tired of talking about overheating.

you can't do prediction based upon one temperature value, but need a series to be able to tell if the internal temperatures are going up or going down and rate in which they are. only with that, and armed with foreknowledge on how much atypically rate in which temperature increases with recording can you roughly predict how long you can record for.

Canon is probably keeping those in local memory and dumping it all to nvram when the camera shuts down.

not writing that probably just gets the camera to throw up it's arms and go well this is screwed .. let's start with 5 minutes.

There is STILL some wonky code in there. but is it wonky with purpose or just the fact that someone made bad assumptions when writing firmware? That we don't know, and probably never will.
Mad respect for not writing another article about this.

I salute you!!!!
 
  • Like
  • Love
Reactions: 3 users
Upvote 0
Sep 29, 2018
325
270
Maybe Magic Lantern will crack this one day and be able to tell us what is really going on. If you look at what they did for the 5DII it was amazing.

In the mean time winter is coming to the North side of the world. If these things still claim to be over heating in a Canadian Winter @ -35C with the wind blowing I will never buy another Canon Camera.
 
  • Like
Reactions: 1 users
Upvote 0
haha. I actually still have the article up on my screen. I don't know, but you know it's too much. the REAL problem isn't the R5 right now, it's Canon's silence.
Between the stock shortages and internet outrage, I do hope Canon remembers what put it on the map. Performance and specification did not alone elevate them to #1.
 
Upvote 0
Aug 27, 2019
667
1,414
haha. I actually still have the article up on my screen. I don't know, but you know it's too much. the REAL problem isn't the R5 right now, it's Canon's silence.
Yeah I think an official "We are working on bug fixes and looking to improve recovery times" that will help.

They have to resist the temptation to say nothing even if they feel this is all much to do about nothing.
 
Upvote 0
Maybe Magic Lantern will crack this one day and be able to tell us what is really going on. If you look at what they did for the 5DII it was amazing.

In the mean time winter is coming to the North side of the world. If these things still claim to be over heating in a Canadian Winter @ -35C with the wind blowing I will never buy another Canon Camera.
has Magic Lantern cracked any recent camera?
 
Upvote 0
the REAL problem isn't the R5 right now, it's Canon's silence.

To be fair to Canon, if they are looking at alternative bitrates I would guess that in itself is not a small piece of work. They will not only have to test it quite exhaustively, but they will also have to provide reliable timings. Add to this, they may well provide that beta fw to various sites at least a week in advance.

Now if they want to adjust timings in the fw to be a little less conservative, well that would take longer too. And no doubt those engineers would have been assigned to other products which they are trying to launch as well as other firmware issues (I think I may have found a small bug, but the workaround is easy).

Bottom line - if Canon states anything before they have additional firmware, then they are admitting indeed there is an issue and that leaves them exposed. If they offer some different bitrate modes without changing anything else then they’re saying we wanted to include these, but couldn’t before - no culpability.

I went out with the R and R5 yesterday just to do some further tests - fairly static birds, and honestly if I could sell both my Rs for a grand each, I think I would buy another r5. After a week with it, I will accept the downsides because for me the benefits outweigh them. And maybe I was a little unrealistic.

Whereas I completely accept that an Atomos may be a solution for some (many?) I honestly cannot see me ever using one (the practicality means for me it would only work for landscapes) - my bag is full enough as it is, so adding another system with different batteries, a cage, a charger doesn’t quite make sense for me (especially if I am flying anywhere). I’m hoping lower bitrate modes may help. If not the 800 I saved by not buying the Atomos with selling two Rs means I have my 2nd R5, and if I could ditch the 5dsr and 5iv, then I could get a 3rd one which my gf could use it on holiday - ha ha. It’s nice to dream.
 
Upvote 0
Yeah I think an official "We are working on bug fixes and looking to improve recovery times" that will help.

They have to resist the temptation to say nothing even if they feel this is all much to do about nothing.
But doesn’t that admit there is a “problem”? I think that may leave them open unless it was a bug and that in itself points to poor testing...

The more I read here (thanks to those who have provided sound reasoning), the more I think it will be just additional modes.

I may be wrong, but is it not the case that Japanese culture frowns on admitting failure? Isn’t that why Olympus are selling their photo division so that someone else can dispose of it quietly, and there is no loss of face. Would Canon admit that they didn’t test their flagship r series and there is an easy to find bug?
 
Upvote 0
How many people REALLY need 8K video? Right now...
It gives you the same advantages as high sensor MP for stills shooting
  • Ability to crop your video - be that for better composition, "zooming" in on your main subject and other fx
  • Less so now they have IBIS, but higher res modes were better when stabilising footage
  • Downsample to 4K - with less noise and better quality (hence why Canon offers downsampling in other modes)
Bottom line, if you could shoot the same in 8K (ignoring processing etc), then you have more flexibility but yes it is 4x the data potentially.

Canon offered it, why would I not want to use it?
 
Upvote 0
Generally, in the USA, a modification, temp or permanent, that defeats a built-in safeguard would violate a warranty--if the manufacturer knew about such a mod.

But this all seems too desperate too soon. I'd think somebody who depends on highest quality 4k video would have dedicated equipment.

Basically, what I'm seeing are people who have expendable funds playing with gear that has been tagged as problematic and/or controversial. Some of the "experiments" have been illuminating! Such as running to an external monitor without the cards in the body, resetting the overheating timer by removing the internal battery...Makes one wonder if Canon's own engineers are at least this clever!

Also makes me wonder when Canon is going to have some breakthrough news about this mess. Can't google for new reviews without seeing the blasted overheating hyperbole popping up.

Spot on. I have a dedicated camera for my 4K needs. I considered the R5 but due to my needs I went and bought a Canon C200B for basically the same money (100 bucks more).
 
Upvote 0