DPReview: Review of the Canon EOS R5

[dtaylor]

Can you please calculate how quickly that would happen?

It would vary with ambient temperatures and airflow rate but we're talking minutes.

(I wonder why people use radiators when they want to dissipate heat by airflow. Should be totally unnecessary, right?)

Internal combustion engines are big three dimensional pieces with internal heat build up. An exposed sensor is effectively a 2D piece with its flat surface exposed to air. The sensor is like a radiator in this respect and would respond very quickly to airflow.

Have you tried to swap the magnesium frame?

That's not a source of heat.

Not if you touch it in the area where it's covered by a low thermal conductivity plastic.

That would be a thermal design issue in a device like this, wouldn't it?

 

[Mt Spokane Photography]

Still waiting to hear if Canon will duplicate the R5 into 5Dmark5 (with OVF).

You did hear that there will be no 5D MK IV. Thats the current situation. It might change in a year, but if sales of R5 and R6 are good, it won't happen.

There will not be an EOS 5D Mark V [CR2] - Canon Rumors

I have been told that the EOS 5D series is going the way of the EOS 7D series. There will be no direct DSLR successor for the EOS 5D Mark IV. The source claims that the EOS 5D Mark V development was stopped “some time ago”, and that there are no plans to make an […]

www.canonrumors.com

 

[rbielefeld]

I downloaded to look at the files. The STK head on was shot with a Sony 9 II + 200-600mm according to the EXIF. How does the Sony compare with the R5?

Yes, that was meant to be for comparison. The R5 AF is as good or better. The head on shots were always tough to canon cameras in the past given my experience, but it seems to be much better. The animal eye AF is much better than a9II IMO thus far.

 

[rbielefeld]

Nailed the eye focus on these shots. I assume you were using the eye focus option?

Yes, and no. Sometimes Zone AF. The Osprey was animal eye if I remember correctly.

 

[mppix]

It would vary with ambient temperatures and airflow rate but we're talking minutes.

This depends on thermal capacity of the sensor (low), airflow, and local air temperature. Sensors have an filter array in front and there is likely a stagnant air layer between that and the sensor; so there may not be a direct airflow on the sensor. This acts as an isolator.
I don't think this can be determined without CFD analysis.
However, I agree that the sensor is unlikely the problem area.

[Have you tried to swap the magnesium frame?] That would be a thermal design issue in a device like this, wouldn't?

The frame is likely less thermally conductive than what would be ideal so it traps heat. This is likely part of the problem (together with conducting heat away from the problem spots).

 

[Random Orbits]

Before the 1DX3 came out, I remember rumors of the MP count going up, and when it was released, people were disappointed that it stayed at 20MP. If Canon had been fielding prototypes with different resolutions, I wish Canon had picked something closer to 26-28MP. Perhaps user feedback had pushed Canon to choose 20MP, but since it ended up being used by the R6, something higher would have been nice. Going from 30 in the R to 20 in the r6 is hard to sell. Canon should have tailored mRAW options for photogs that wanted smaller file sizes... just a thought.

Personally, I've been a 5 series guy after using the 5D2 years ago, so I'm opting for the R5. Both the R5 and R6 do 4K video better than the R (which isn't hard). For now, I'll trade in the 5D4 and a few EF lenses for the R5 and retain the R as backup. But I can see looking at the R6 to replace the R once the price drops and its available on sale in the refurb store.

Sony needed the A7S3. Without it, The R5/R6 would have won against the A7R4/A7S2 pairing. Good on Sony's part for developing the A7S3. However, with the R5 out, what is the appeal of the more expensive A9II?

 

[davidhfe]

It would vary with ambient temperatures and airflow rate but we're talking minutes.

Internal combustion engines are big three dimensional pieces with internal heat build up. An exposed sensor is effectively a 2D piece with its flat surface exposed to air. The sensor is like a radiator in this respect and would respond very quickly to airflow.

That's not a source of heat.

That would be a thermal design issue in a device like this, wouldn't?

I've sort of lost the thread here but I'm not sure these assumptions hold. A CPU is the closest to what we're talking about here and a poorly cooled CPU doesn't behave like these cameras are behaving. ICs seem to come up to temperature very quickly under load, and that heat needs to be removed very quickly to prevent damage. These cameras are building up heat somewhere that's very hard to remove it from.

And CPUs very much do utilize radiators when you're talking about high TDP chips. You have a 2D surface and couple a spreader to it. The spreader makes sure that there's a nice even contact surface for your cooling solution to attach to. That spreader then needs to be attached to a sink of some kind; you can couple it directly, with a heat pipe, or with some pump that moves a cooling fluid around. Usually that sink is going to be a big radiator with a fan.

Apologies in advance if I've mis-represented your point.

 

[davidhfe]

The frame is likely less thermally conductive than what would be ideal so it traps heat. This is likely part of the problem (together with conducting heat away from the problem spots).

Yeah, it certainly seems like the body is acting as a sink, but then the exterior rubberized grips etc are acting as too good of an insulator, making it very difficult to cool that sink down.

 

[Quarkcharmed]

Armando Ferreira did a real 8k production with a real team and real actors. He could not shoot everything in 8k and ended up titling a follow up video "The Canon EOS R5 is NOT AN 8K CAMERA!" because of the thermal issues. He is a professional, not a mere "self-claimed influencer."

There is a thermal issue.

There's certainly an issue with click-baiting on YouTube. Professionals also need likes, views and subscribers and there's a lot of professionals who are at the same time the 'self-claimed influencers'. Especially when they lie right to our faces. There's flawed Canon's PR, too.

All this internet hype around the R5 overheating accelerated after the A7SIII announcement which brings some interesting questions about dark PR on the table. From all this buzz it may look like everyone's concerned about video only and is only going to buy the R5 for its video features. So, because the R5 cannot shoot 8K video for 1hr, those people will switch to Sony A7SIII. Which, in turn, doesn't shoot 8K at all and only has 12Mp sensor, but it's all cool and great just because.

But again, I think there's a high chance Canon will do some 'restructuring' in their PR/marketing department after this launch.

 

[reefroamer]

Expecting? I didn't market it, Canon did.

To my knowledge, Canon has never promised or marketed unlimited 8k on the R5. Or 4K120, etc. Conversely, they have documented the limitations.

 

[Deleted member 384473]

here is a good news for you.
RED release RED KOMODO with RF mount. it also have AF. and its priced around 5000-6000$.
.
if R5 become perfect as you imagine. RED KOMODO is just DOA.

I've followed the Komodo but I don't care to waste time on beta. I also doubt the AF is anywhere near Sony/Canon. Also, the RF support is not supported at launch so they're including EF adapters from Canon. I am excited that another camera provider has utilized the RF mount though! Its just too early on in the game.

 

[Deleted member 384473]

Did they? Or did you just read too much into pre-release hype?

Nope. Just entertained what Canon marketed on their website.

 

[Jack Douglas]

Just give the R5 some time and we'll see it's not really too bad, not too bad! However, this thread by comparison is really bad - a waste of time and I don't know why I persevered.

Jack

 

[Deleted member 384473]

The way you worded your original comment made it sound like you bought RF lenses in the hope that the next R body would meet your needs precisely, my mistake I guess?

No, I get that. I just emphasized I invested in RF because I can only use it with the R system unlike EF where I could use it with other brands.

 

[Deleted member 384473]

There's certainly an issue with click-baiting on YouTube. Professionals also need likes, views and subscribers and there's a lot of professionals who are at the same time the 'self-claimed influencers'. Especially when they lie right to our faces. There's flawed Canon's PR, too.

All this internet hype around the R5 overheating accelerated after the A7SIII announcement which brings some interesting questions about dark PR on the table. From all this buzz it may look like everyone's concerned about video only and is only going to buy the R5 for its video features. So, because the R5 cannot shoot 8K video for 1hr, those people will switch to Sony A7SIII. Which, in turn, doesn't shoot 8K at all and only has 12Mp sensor, but it's all cool and great just because.

But again, I think there's a high chance Canon will do some 'restructuring' in their PR/marketing department after this launch.

Yes. Lots of dark PR that was definitely fueled further by the launch of the A7SIII, which shouldn't at all be compared to the R5. However, theres no shooting 8K past 20 minutes and thats if you're efficient with your shooting AND shoot no stills. Youtube is mainly going to focus on video & overheating for those clicks, for sure. I've only seen two videos focused on stills but my algorithm is all set for video. Now, I wonder what the rumor RF video camera is all about... I wonder if it'll be at all comparable?

Also... Who was asking for 8K? I believe the majority of video shooters just wanted reliable 4K FF 10 Bit... lol

 

[dtaylor]

I've sort of lost the thread here but I'm not sure these assumptions hold. A CPU is the closest to what we're talking about here and a poorly cooled CPU doesn't behave like these cameras are behaving. ICs seem to come up to temperature very quickly under load, and that heat needs to be removed very quickly to prevent damage. These cameras are building up heat somewhere that's very hard to remove it from.

That depends entirely on the chip. Most people who have an impression of such things have it from personal computers with wildly varying loads and fan speeds, and chips that can do a lot more work and even 'turbo boost' compared to an embedded ARM chip.

That said, I do not discount the possibility the problem is downstream. My "Canon forgot the heatsink" comment was somewhat tongue-n-cheek, and later I listed three possibilities that occur to me (no heat sink/spreader/pipe; heatsink is there but poor materials/mating; that's all good but heat can't escape shell).

Edit: and the more I think about it, while a chip can 'slowly' reach it's thermal limits, I wouldn't expect it to take 2 hours to cool. So good chance the problem is downstream.

And CPUs very much do utilize radiators when you're talking about high TDP chips.

I actually took the comment I was replying to as an analogy to cars.

Apologies in advance if I've mis-represented your point.

No apologies necessary. We don't know where the problem is and absent a Lens Rentals teardown and/or IR imaging of a camera in action, we can only speculate. But based on the information we do have I think we can rule out sensor, card, or battery overheating.

 

[dtaylor]

There's certainly an issue with click-baiting on YouTube. Professionals also need likes, views and subscribers and there's a lot of professionals who are at the same time the 'self-claimed influencers'. Especially when they lie right to our faces. There's flawed Canon's PR, too.

Armando Ferreira is not a click-baiter and has never lied in a review that I've seen. He was quite generous to the EOS R when everyone was ripping on it for its 4k crop, and he was one of the people to point out the flexibility of the 480 Mbps output.

From all this buzz it may look like everyone's concerned about video only and is only going to buy the R5 for its video features. So, because the R5 cannot shoot 8K video for 1hr, those people will switch to Sony A7SIII. Which, in turn, doesn't shoot 8K at all and only has 12Mp sensor, but it's all cool and great just because.

That's nothing but a straw man. Nobody expected a fan-less, continuous 8k camera. But when you can't expect to grab a HQ 4k clip at a wedding because you've been shooting stills, there's a problem. And it's a problem which will send some people to other brands.

 

[Quarkcharmed]

To my knowledge, Canon has never promised or marketed unlimited 8k on the R5. Or 4K120, etc. Conversely, they have documented the limitations.

They weren't promising an unlimited 8K but they didn't tell us about overheating until recently. The R5 was in fact advertised pre-launch as an 8K camera, and even now the official description says (from canon.com.au)

EOS R5
EOS R5 full frame mirrorless camera features a 45MP full-frame CMOS sensor, 8K 30p and 4K 120p videos, In-body IS up to 8-stops and can shoot up to a blistering 20fps.

So before the launch, people hoped for the best but didn't prepare for the worst. The 8K is still being advertised as a main innovation on top but there's a fine-print disclaimer now

1 Among all interchangeable lens digital cameras. Based on Canon research as of 9th July 2020. 1 Only when setting crop [Off], up to 29.97 fps / 25.00 fps. The video recording time of the Canon EOS R5 is limited by heat.

All Canon had to do was to put a similar disclaimer in the pre-launch advertisements. It wouldn't hurt the sales and preorders.

Personally I don't care at all - I'm looking forward to finally getting my preorder which is delayed till Oct.

 

[reefroamer]

It would vary with ambient temperatures and airflow rate but we're talking minutes.



Internal combustion engines are big three dimensional pieces with internal heat build up. An exposed sensor is effectively a 2D piece with its flat surface exposed to air. The sensor is like a radiator in this respect and would respond very quickly to airflow.



That's not a source of heat.



That would be a thermal design issue in a device like this, wouldn't it?

I disagree. It’s a thermal LIMITATION of the design, which prioritized a small, weather-sealed body with a high Resolution, fast-readout sensor requiring a very high performance DIGIC processor. Meanwhile, where is the outrage over a 12MP sensor in the A7s3 that most stills shooters would consider a joke in the year 2020. Obviously, that was the compromise Sony made. I wouldn’t even consider that a hybrid camera. It’s primarily video. The R5 makes it obvious to me that Canon accepted the thermal limitations on video as the price paid to achieve other product objectives.

 

[Quarkcharmed]

Armando Ferreira is not a click-baiter and has never lied in a review that I've seen.

He lies right in the title. " The Canon EOS R5 is NOT AN 8K CAMERA! " If it's not a click-bait and a lie, then what is? We know it is an 8K camera!

That's nothing but a straw man. Nobody expected a fan-less, continuous 8k camera. But when you can't expect to grab a HQ 4k clip at a wedding because you've been shooting stills, there's a problem. And it's a problem which will send some people to other brands.

Are you personally shooting weddings and considering A7SIII instead of R5? I don't know about 'that guy'. I know about myself and my own requirements. I don't have the 8K video on my requirement list. It's nice to have, but even if I shoot it, I can't imagine shooting 8K for longer than a few minutes.

So personally I don't care at all, and I don't know why I should care about wedding photographers requirements. If they're happy with the R5 - fine. If they drop Canon and switch to A7RIII and shoot weddings at 12Mp - it's their problem.
If I ever shoot a wedding, I'll consider all requirements and choose the gear accordingly.