My Canon EOS R5 recording 8K video 50 minutes straight

[horshack]

1 hour ago, Electroholic Anonymous said:

The ambient temperature is perhaps only 1 or 2C lower, that should not make such a difference. I think it's mainly convection, and that I held it in my hands perhaps a minute at the start and end of each run earlier, and now on the tripod I did not. Ah, and that's it, I used 8K 24p now, and 8K 30 earlier, that's of course it. I specifically chose 24 p for the long test, because that is the mode I am personally most interested in. I could run the 8K30p also, we'll see. The inordinate number of RAW+jpeg's that I shot at the end produced 64C exifs after only 257 photos. So that's in line with my personal experience of the R5 feeling hotter hammering lots of stills than recording 8K.

Rereading this I'm thinking maybe the 24P made the difference for the newer 62C plateau vs the earlier's 30P at 72C. That's a lot more data being read off the sensor and pushed through DIGIC.

[Electroholic Anonymous]

4 minutes ago, horshack said:

Rereading this I'm thinking maybe the 24P made the difference for the newer 62C plateau vs the earlier's 30P at 72C. That's a lot more data being read off the sensor and pushed through DIGIC.

Haha, as we say here: "Great minds stink alike." 😉

I might do an 8K30p next, or perhaps the even more demanding 4K120, ah but 4K120 does not write to SD, for that we would have to wait for a DAT file recovery option.

[horshack]

1 minute ago, Electroholic Anonymous said:

Haha, as we say here: "Great minds stink alike." 😉

I might do an 8K30p next, or perhaps the even more demanding 4K120.

Joking aside it's nice to have a new tech-savvy contributor working on this!

Not sure the camera will let you do 4K120 on the SD card - the R5 manual states 4K120 only supports ALL-I, at a data rate of13447 MB/minute, which works out to ~224 MB/s.

[hoodlum]

1 hour ago, wolf33d said:

FW update to fix thermal issue coming tonight
https://www.canonrumors.com/canon-eos-r5-firmware-update-coming-in-the-next-24-hours/

Obviously it's not like they will release many of them so the moment of truth is tonight. Tomorrow we will know if the camera is usable or not. 
Either they will go deeper in the same and shit on us with allowing something like 5min more recording and little shorter recovery (but overall still timer based), or they will make it right and make it non timer based but temperature based making this camera perfect or if they can't (because of poor thermal data access) then at least push the timer as much as they can according to extensive testing. From what we see in the test from Andrew and others, that camera should have no problem going to the recording limit of 30min and then recover in that time or less. 

Fingers crossed. 

Better to keep our expectations very low.

"Update: A source that is testing the new firmware has warned me not to expect too much of a difference in thermal performance with the new firmware. I guess we’ll know for sure quite soon."

[Electroholic Anonymous]

15 minutes ago, horshack said:

Joking aside it's nice to have a new tech-savvy contributor working on this!

Not sure the camera will let you do 4K120 on the SD card - the R5 manual states 4K120 only supports ALL-I, at a data rate of13447 MB/minute, which works out to ~224 MB/s.

Thanks! I had a lot of fun doing the test and preparing it while discussing with you. By the way the Fro knows you:

 

[Electroholic Anonymous]

Perhaps it is even a binning issue. Binning as in CPUs of Intel for example. Some can be overclocked substantially higher than others. Especially on a fresh new range when the production process has not stabilized fully yet. People like der 8auer make a whole business out if it, selecting and selling the fastest bins for a premium. Let's say your friend got a champion R5 and you got a lemon R5, you would be happy that they behave the same, and neatly overheat at 15 or 20 minutes. It's still all conjecture at this point. Some truth from Canon would be very welcome in this regard.

[Andrew Reid]

1 hour ago, horshack said:

Thanks for the .CSV data. Here's your data plotted. The running elapsed time was calculated from the EXIF timestamps:

Thanks for the chart. It's good to have it in black and white that there's only a 6-8C increase in temps after 20 minutes.

Certainly not enough to justify the shutdown (let alone a long lockout).

And also interesting to see the temps are basically flat for the last 50 minutes after the first 50 minutes.

I think if we are going to put credits on stuff by the way, make sure to include @BTM_Pix

His app has been key to understanding the temperature status and EXIF temp.

[horshack]

1 hour ago, Electroholic Anonymous said:

Haha, as we say here: "Great minds stink alike." 😉

I might do an 8K30p next, or perhaps the even more demanding 4K120, ah but 4K120 does not write to SD, for that we would have to wait for a DAT file recovery option.

FYI, I started a new dpreview thread about all the experiments here, which is a continuation of my original thread here.

[ajay]

According to TechRadar, the firmware doesn't address the overheating and actually engages it when using an external monitor:

"Canon has today released the EOS R5’s first firmware update and it’s available to download from the camera maker’s regional websites around the world. However, if you were expecting the firmware to improve on the current video limitations the R5 suffers from (as previous reports indicated it would), you’re going to be sorely disappointed.

The EOS R5 firmware version 1.1.0 only brings a couple of minor upgrades – it improves image stabilization for video recording, while also improving the combined stability of the camera’s in-body image stabilization system when paired with a stabilized lens, specifically the RF 100-500mm f/4.5-7.1 L IS USM telezoom.

According to Canon, the update also ensures that the overheat control on the EOS R5 doesn’t get disabled when using an external monitor or recorder. It will also improve the accuracy of the video time display when recording several short clips back to back.

None of these are quite what we were expecting but Canon has said there’s more to come."

[Video Hummus]

3 minutes ago, ajay said:

According to TechRadar, the firmware doesn't address the overheating and actually engages it when using an external monitor:

"Canon has today released the EOS R5’s first firmware update and it’s available to download from the camera maker’s regional websites around the world. However, if you were expecting the firmware to improve on the current video limitations the R5 suffers from (as previous reports indicated it would), you’re going to be sorely disappointed.

The EOS R5 firmware version 1.1.0 only brings a couple of minor upgrades – it improves image stabilization for video recording, while also improving the combined stability of the camera’s in-body image stabilization system when paired with a stabilized lens, specifically the RF 100-500mm f/4.5-7.1 L IS USM telezoom.

According to Canon, the update also ensures that the overheat control on the EOS R5 doesn’t get disabled when using an external monitor or recorder. It will also improve the accuracy of the video time display when recording several short clips back to back.

None of these are quite what we were expecting but Canon has said there’s more to come."

The pants shitting continues.

[kaylee]

u guys r so smart @BTM_Pix et al

tysm for all your hard work. yall have really come through for the people

[mechanicalEYE]

https://translate.googleusercontent.com/translate_c?depth=1&pto=aue&rurl=translate.google.com&sl=auto&sp=nmt4&tl=en&u=https://cweb.canon.jp/drv-upd/eosd/eosr5-firm.html&usg=ALkJrhhjgCoNGbTHOjJ5KBPzgPnWPWwAbg

[Emanuel]

Seems not a single word on topic... or I am lost over here? I am not as smart as @BTM_Pix ; ) so please can anyone explain what does this shit exactly mean? Thank you!

I only wanna know if I will buy this thing or not... Firmware fix or mod/hack, anything serves if it fits the bill and does the job just fine, I am not so much picky :- D

[mechanicalEYE]

 

[Emanuel]

Why doesn't Canon UK support EOSHD as they do with these dudes? (and I do appreciate Gordon Laing's objectivity anyway)

Because they serve the brand's goals instead.

Here's why we should ALL support the spirit and good efforts of this mod team here (I'd rather call it mod than hack as matter of fact!) It is here where filmmaking tech vanguard is (Andrew's credit/credentials), not where the commercial target rises up.

E :- )

[horshack]

34 minutes ago, ajay said:

According to TechRadar, the firmware doesn't address the overheating and actually engages it when using an external monitor:

"Canon has today released the EOS R5’s first firmware update and it’s available to download from the camera maker’s regional websites around the world. However, if you were expecting the firmware to improve on the current video limitations the R5 suffers from (as previous reports indicated it would), you’re going to be sorely disappointed.

The EOS R5 firmware version 1.1.0 only brings a couple of minor upgrades – it improves image stabilization for video recording, while also improving the combined stability of the camera’s in-body image stabilization system when paired with a stabilized lens, specifically the RF 100-500mm f/4.5-7.1 L IS USM telezoom.

According to Canon, the update also ensures that the overheat control on the EOS R5 doesn’t get disabled when using an external monitor or recorder. It will also improve the accuracy of the video time display when recording several short clips back to back.

None of these are quite what we were expecting but Canon has said there’s more to come."

Some said the HDMI w/o card long-recording workaround was a bug. I didn't believe it. Looks like they were right. Shaking my head. This also implies the thermal management isn't to protect the cards, otherwise why would they still throttle the camera without cards installed (assuming this update throttles both with and without cards).

[ajay]

So based on Gordon's tests, the new firmware takes into account external temperatures and recovers quicker when cooling is applied. I guess we're back on for the Tilta Cooling Kit. 😀

The good news is that now the firmware can be downloaded and hopefully Alex and others @ ML can start working on additional firmware to add functions to the crippled R5.

[ajay]

According to Gordon's communication with Canon, the R5 has three temperature sensors: (2) internal...near sensor/autofocus system? and (1) external ambient temperature.

That still doesn't address the timer that's being employed. Doesn't make sense really. Could just be a bunch of BS from Canon.

[Emanuel]

This is just a mere business. This put them on shame. Embarrassing to say the least. Hackers/modders are more dignified (and reputable parties) than established manufacturers as Cannot. What a strange world we all live in.

[ntblowz]

7 minutes ago, horshack said:

Some said the HDMI w/o card long-recording workaround was a bug. I didn't believe it. Looks like they were right. Shaking my head. This also implies the thermal management isn't to protect the cards, otherwise why would they still throttle the camera without cards installed (assuming this update throttles both with and without cards).

so no more 3-4hr 4KHQ external recording when using new firmware?!