Canon EOS R5 records for 4 hours in 4K30 HQ - Updated

A new video from nolifedigital has gone a bit further in the heat testing of the EOS R5.

Differing from Gerald Undone's great testing, no life tested without having the battery (using a dummy adapter), and without cards in the camera and recording externally to an Atomos Recorder.

It seems that the EOS R5 in this configuration recorded during his tests over 4 hours and still did not display an overheat warning.  This is with recording externally 4K30HQ - which is 4K oversampled from 8K.

This may be certainly a workaround for the problem.

Update:

After further testing, it seems as if the EOS R5 will record continuously with 4K30HQ with an LP-E6N battery until the battery dies, but without cards in the camera (so around 2 hours of record time).

So it sounds like the limitation is around having cards in the camera.  It's hard to say that this is still a software limitation itself because otherwise there would be no real need for a "cooldown" period, but the camera could be doing something with the cards, even while recording that is simply adding to the problem.

Update #2:

Some people can't replicate this, but a user in dpreview may have narrowed it down to PAL versus NTSC.  

The user, dellfonic, first tried to reproduce the test, and couldn't;

I was hoping that you were right about the it being a software problem but it isn't.

As per the video, I took out all cards, inserted a dummy battery and proceeded to record in 4K HQ ALL-I to the Atomos Ninja V in Prores 422. As suspected, the temperature warning came on 35 mins. I have repeated this test with the same result.

I also tested recording in 'standard' 4K HQ and got over 2 hours before the battery ran out. The question is, whether the HDMI output to the Atomos provides you with oversampled 4K footage or the line skipped mush you get with internal recording.

However, in a eureka moment he decided to check if there was a difference between PAL and NTSC, and there was;

Of course... so, having switched from PAL 25p to NTSC 24p, I've just broken through the 35 min barrier, approaching 1 hour. I'm convinced now that they've limited PAL to pass apparent forthcoming EU laws on temperature limits. It does feel fairly hot on the back and top of the body...

There seems to be some sort of new regulation on the thermal characteristics of, I would imagine, all electronic goods being sold in the EU. If this is the case, then the camera may be overly aggressive with temperature management, getting ahead of the EU regulations.

It would also explain why some people are unable to replicate nolifedigital's results.

As far as why with the card? there seems to be some throttling that occurs on CFE cards if they get too hot. nolifedigital noticed that even when recording to the Atomos controller, the card itself got warm.  Warmer than for instance, the camera itself.

This article goes into CFE heating and throttling that may be occurring with the R5.

There's a lot more that needs to be done, and we certainly need some information from Canon concerning the overheating and what they plan on doing about it.