Testing methods and results

Even in direct sunshine, the EOS R5 was able to record line-skipped 30p footage for nearly two hours (in 29:59 chunks) without overheating. The thermometer was reading 40°C (104°F) by the end of the test.

We tested the cameras in the 4K/30p modes that we believe most videographers would use for the bulk of their primary ‘A-roll’ footage. We set the cameras outside in direct sunshine and set them recording. The ambient temperature was around 25°C (77°F) but the constant unshaded exposure to sunshine meant the cameras will have got a lot hotter than this. Our thermometer, placed alongside the cameras, registered 40°C (104°F) by the end of the test.

4K/30 HQ
4K/30 Std
Overheat warning


26:30 (Est 25:00) N/A
Clip length 39:16* 29:36 1:56:12*
Rest (30 min)
Overheat warning 19:44 21:40 (Est: 20:00)
Clip length 22:03 24:34
Rest (30 min)
Overheat warning N/A N/A (Est: 25:00)
Clip length 22:22 25:17
Failure reason Battery exhausted Battery exhausted Battery exhausted

*Clips with reported length exceeding 29:59 were re-started when the time limit was reached

Upon overheating, we placed the cameras in the shade and gave them 30 minutes to recover. This is about as long as it’s practical to leave people waiting or setting-up a new shot, while shooting a video.

You can see that the EOS R5 promises reduced recording capability even after 30 minutes of cooling but actually shoots close to its full capability again. However, the R6 only regains around 1/2 its capability after a 30 minute cool-down period, presumably because it doesn't have the R5's metal rear plate to dissipate heat. The batteries died before we were able to complete the third test.

The first test was conducted on tripods, shooting a static scene, so for the next test, we walked around in 25°C (77°F) sun with IS On, plenty of scene movement to compress and AF set to continuous, to see whether this shortened the shooting duration.

4K/30 HQ
Overheat warning 37:02 27:00 (Est: 25:00)
Clip length 40:00


Rest (30 min)
Overheat warning 19:04 19:45 (Est: 20:00)
Clip length 20:42 22:15

As you can see, the recording duration for the cameras was not affected by camera movement, scene movement or increased AF activity.

Taking the battery and card out, leaving both door open and the screen extended away from the rear plate of the camera didn't appear to improve recovery times.

We also conducted a test to see whether the compression type significantly changed the amount of work the processor was having to do.

4K/30 HQ
4K/30 HQ
Overheat warning 27:15 27:01
Clip length 29:59 29:54
Rest (30 min)
Overheat warning 21:52 (Est: 20:00)* 20:52 (Est: 20:00)
Clip length 24:53 23:38

*Camera left with card and battery in, and screen folded inwards during cool-down period

This test strongly suggests compression type doesn't have a significant impact on heat build-up. For the cooling period, we took the battery and card out of the camera that overheated first, but left the other camera with all its doors closed and the screen folded in towards the back of the camera. Surprisingly, this doesn't seem to have much impact on the rate at which the camera cooled. Which is a positive in the respect the respect that there aren't any special tricks you need to remember but also suggests there's not much you can do to accelerate cooling: you just have to wait.

Richard's EOS R5 photo experience

Days after conducting these video tests, I went shooting at a local nature reserve using the Canon 800mm F11 USM IS lens. It was another hot day, for Seattle (27°C / 81°F).

In just under two hours I shot 164 images, all in CRAW, some as parts of short bursts. I then switched to video mode to capture some 4K HQ footage and was confronted by 04:00 minute limit, despite plenty of card capacity.

The overheat warning displayed immediately and, after shooting three sub-10-second clips the camera said it would only shoot for another 02:00 minutes.

We also checked whether moving to the camera's 10-bit modes made any difference. An EOS R5 set to record 4K/30p HQ in 10-bit Canon Log recorded for 29:59 while a second camera capturing 10-bit HDR PQ footage stopped after 28:34, both cameras having estimated being able to shoot for 25:00 minutes.

Other usage and cooldown periods

Our initial tests repeatedly showed that both cameras can deliver the recording durations Canon promises, even when tested in warmer conditions (though we'd expect more extreme conditions to have an even greater impact). The only exception being a test of the R6's 4K 60p, which recorded only 26:42 of footage, rather than the 30 minutes Canon promises (though the camera only estimated 25:00).

We then tried testing the cameras with various amounts of time to recover and having spent time shooting stills and being used prior to the start of a video capture. During this test we switched one camera to using a UHS-II SD card, rather than the CFexpress used for other tests, to see if this made any difference.

4K/30 HQ (SD Card)
4K/30 HQ
Initial setup From room temp Half an hour in the sun including shooting 300 Raw+JPEG images
Overheat warning 27:25 (Est: 25:00) 15:01 (Est: 15:00)
Clip length 29:59 18:01
Rest 20 min 10 min
Overheat warning 24:39 (Est: 20:00) 07:29 (Est: 10:00)
Clip length 27:53 10:30
Rest 5 min on USB charge 5 min
Overheat warning 03:31 (Est 4:00) 04:55 (Est: 5:00)
Clip length 06:33 07:10

It's worth noting that half an hour of camera usage in the sun had a significant impact on the camera's internal temperature (and hence estimate of recording duration). Similarly, the less time you can give the camera to cool, the lower the amount of capability is recovered. As you might expect: trying to charge the battery internally (which generates heat), undermines attempts to let the camera cool.