What’s next for the Canon EOS R5 and Canon EOS R6?

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
It there was a manufacturing issue they wouldn’t need to issue the Overheat timings advisory and have a menu feature which you enable to protect against overheating, no?

second there was a YT video showing the Canon and Sony overheating in Florida - the guy had a temp sensor which could read the heat from the bodies as they started to record. Iirc, around 120 to 130 degrees constant, with the Panasonic being lower by about 10 degrees. It may simply be the Sony sensor can handle the heat and the Canon one can’t for the same period. Whether 12MP has a bearing on that I don’t know. Presume the same with the Panasonic (to the Canon) and hence why they designed a fan in theirs. Maybe Sony needed 5 years to design a solution which could dissipate the heat / take the heat better / didn’t generate as much heat. Even with IBIS which apparently causes challenges.

My best guess is either they’re taking design from a 1Dx / R1 body and modifying it for the R5/R6 along with firmware changes, or they’ve had a lightbulb moment. Combined they think they can mitigate it to a point, but as you suggest, a complete fix? Not in this generation. I think 8K isn’t really the issue for them - as in, maybe a 1.5x the time would be good enough. I think most concern comes from the 4K modes shutting off. If they can prolong that in the HQ and 120fps without lowering quality to over an hour and offer a gizmo which can cool it quicker (maybe the lens mount cooler but designed just to cool it, not always active), then that might be enough.

Alas it does mean that stills shooters have to wait...
d
We'll see. We know already that an external fan will work. Canon ex[

Astrologist have used external DSLR cooling for over a decade.
We already know that an external fan will work. It most often does, Regardless of the interior design of the camera. It's done all the time in the advvent of loss of air conditioning in electronic equipment rooms. The first thing we do is bring in large fans. Most often we can keep entire buildings with racks of equipment cooled for days. Cool that camera body with a fan/blower or the like an one can keep the temps down. I will record my device temps and setting when it arrives today, and proceed with designing my own supplemental cooling until the Tilta, SmallRig, or Canon devices are available. Not are difficult task.
have you seen a photo where videographer was trying to bring R5 temperature down with a massive leaf blower? it didn’t help Much apparently. :)
 
Upvote 0

Starting out EOS R

EOS R5 - RF24-105mm F4L, RF70-200mm f2.8L
Feb 13, 2020
295
315
Super cool. Mine's coming some point today. Disheartening to hear recalls and stuff when first waking up, but if it perfects the body, why not? My atomos is a nice work around for some of the issues for me, but I'm still super stoked. I don't usually pre-order or jump in the first batch with big purchases like this, but I've been waiting for the R5 for a while now. My 5DII and 5DIV were great when I got them a few months after launch, but the R5, hoping it's better.
Always fun unboxing and setting up everything :ROFLMAO:

Haha, the 1st thing I did was complete all the registration details for the free EF adapter and lens reward and had a couple of hours playing, just like a little kid at xmas lol.

Not sure if the Canon website is having issues though as for some reason it won't let me register it on the Canon CPS membership page. It's stupid things like this that although not massively important, really annoy me. Maybe I'm Turing into a grumpy old man!

Tomorrow looks like it will be a sunny day here in the UK so I'm hoping to be out and try the animal AF on our trainee guide dog when she's free running.
 
  • Love
Reactions: 1 user
Upvote 0

cornieleous

5D4 + R5
Jul 13, 2020
208
737
We already know that an external fan will work. It most often does, Regardless of the interior design of the camera. It's done all the time in the advvent of loss of air conditioning in electronic equipment rooms. The first thing we do is bring in large fans. Most often we can keep entire buildings with racks of equipment cooled for days. Cool that camera body with a fan/blower or the like an one can keep the temps down.

Are you assuming everyone is shooting indoors? How would event shooters, people using on camera mic, or people moving the camera use a fan?

Beyond the impracticality, an internal OR external fan only works if ambient air is cool enough to reduce heat fast enough. Canon already performed tests at room temperature, so blowing air across the body is not likely to improve things without more heat dissipation surface. Expelling heat from the body with an internal fan might help if they wanted to redesign the camera. In the case of sunlight or certain studio lights there is then an external heat source adding to the problem. In air that is too hot, blow all the air you want inside or out, it will not cool well. In extreme temps like the desert, almost no cooling would be achieved with any fan.

The real question is, how much cooling is enough to make the camera shoot long enough for people to stop complaining. In real world use, I think it already shoots long enough for reasonable use, but many people are not reasonable.
 
Upvote 0

SecureGSM

2 x 5D IV
Feb 26, 2017
2,360
1,231
It's totally possible the thermal couple between the sensor on the IBIS sled and the mag alloy is the problem for exactly that reason. But it doesn't seem to be given that you can run the system at full bore for 20 minutes before issues crop up and then you need a really long cool down. That seems more like hitting the sink capacity.

I probably could have condensed that entire post down to "Those people are idiots. There's a huge heat sink on the R5—it's called the body"
Two more solutions that twill take care of the issue: relaxed codecs and.. auto focus.
 
Upvote 0
There doesn't need to be a recall. That's stupid. It doesn't have a fan. There are a few options. And not do what Sony did and raise the threshold so that it operates while overheated. Nor use a Auto Pwr OFF Temp (HIGH). In a few weeks there will be third party devices and Canon's own adapter.. I believe this is what the slow down is. I have already called Tilta, and communicated with SmallRig. Tiltas unit will be available mid - late August. I have yet to hear from SmallRig. Myself will order the R cage and design my on small fan attachment and or peltier cooling with USB control similar to that of Tilta unit.
If they could provide an addon then indeed they could continue to sell and stills shooters would be ecstatic....
 
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
It's totally possible the thermal couple between the sensor on the IBIS sled and the mag alloy is the problem for exactly that reason. But it doesn't seem to be given that you can run the system at full bore for 20 minutes before issues crop up and then you need a really long cool down. That seems more like hitting the sink capacity.
There is another thread here where results with an external recorder show that overheating in 4K HQ occurs only after over 70 minutes, despite the quality being identical to internal recording. So while heat is generated just by the read out, IBIS, downsampling and application of image profiles, it seems to me like a major source of heat is actually writing to the fast cards. The 8K overheats roughly the after the same time in RAW or compressed mode. So I don't think the encoding is the main source of heat.
 
Upvote 0

davidhfe

CR Pro
Sep 9, 2015
346
518
have you seen a photo where videographer was trying to bring R5 temperature down with a massive leaf blower? it didn’t help Much apparently. :)

All this stuff is way over our pay grade, but air's actually a pretty good insulator. I'm not surprised that blowing on a 2x3 inch piece of plastic (the housing behind the tilty flippy) doesn't do much. The little cooling gimmick that's going around at least had a set of radiator fins the fan was set in to. That alone would be like 8x the cooling surface for that fan to work against.
 
  • Like
Reactions: 1 user
Upvote 0
There is another thread here where results with an external recorder show that overheating in 4K HQ occurs only after over 70 minutes, despite the quality being identical to internal recording. So while heat is generated just by the read out, IBIS, downsampling and application of image profiles, it seems to me like a major source of heat is actually writing to the fast cards. The 8K overheats roughly the after the same time in RAW or compressed mode. So I don't think the encoding is the main source of heat.
I thought HDMI had uncompressed video and the external recorder does the encoding?
 
  • Like
Reactions: 1 user
Upvote 0
Of course, and yeah I'm being quite glib here. Regardless of what design goals were set out 4 years ago (45 mp was most likely chosen because they wanted 8K to be a feature) you need to make a call based on the engineering that's happening on the ground. Either:

- The camera was engineered to handle 8K, went to market with that headline feature assuming it would work, and there's a defect to be addressed.
- The camera couldn't hit the 8K goal, but it was marketed as such anyways.

If it's the latter, it was known the first time engineering samples came off the line a year ago and it was decision time back then. We don't really know what happened inside of canon here—maybe engineering thought they could fix it. Maybe marketing got flustered by a leak. Maybe I'm wrong and tomorrow canon says firmware 1.01 will address these problems and we can all go skipping off into the sunset.
option 2 - they thought the record times were enough, and the decision was made to go to market with limits. I don’t think any 1 department made the decision. Head of their camera division would have bought in to it.
 
  • Like
Reactions: 1 users
Upvote 0

cornieleous

5D4 + R5
Jul 13, 2020
208
737
Also, wow, that EOS thread about a heat sink. A heat sink is a device that you throw (sink) waste heat into so that it can be dissipated. The R5 has a huge heat sink—the mag alloy body. It doesn't matter how big the spreader is on the back of the PCB as long as it can can transfer that heat from the processor and sensor to the body as fast as it's generated. Given the 20-30 minute recording time, I'd say that's the sink (body) reaching it's capacity. This sounds very different from, say, a PC that's got inadequate cooling. Not sure about the lack of thermal compound—that picture was tiny—I suppose it could be an issue but only if it's heat transfer that's the problem not heat dissipation. But you don't put cooling fins (like you'd see on a CPU) on a thermal setup like this because fins are used to dissipate not transfer heat.

The problem* here seems to be that the sink (mag alloy) can't effectively dissipate the heat generated once it's full. This is why the R6 overheats faster despite processing half the information as the R5. No mag allow to sink heat to! I feel like folks are looking at this and saying "gosh it's not designed like my PC canon is dumb" without a pinch of thought about what's going on:

- Rate the sensor and processor generates heat. Can be mitigated by firmware (more efficient or less use of hardware) to some degree but there are also just the realities of the TDP of these components.
- Is the heat from the sensor and processor being pulled away quickly enough to prevent them from overheating (stability, eventually hardware damage).
- What do you do with that heat—where does it sink to?
- Once it's been sinked, how do you dissapate heat from the sink. PCs have big ass sinks with hundreds of fins because they're paired with big ass fans to remove the heat from the sink.

Where the problem in that stack lies greatly affects what Canon can do. If the entire thing is working as anticipated, I don't see how you fix* it w/o re-engineering the entire body.

In a perfect, perfect world there's a defect in a batch of Digics or sensors that's causing them to create more heat than their spec'd TDP. The delay is to fix that mfg problem but wow that's idle speculation for the ages. This is starting to feel an awful lot like "Canon is re-working the 5D4 to add better 4K codecs!" and folks need to get their own expectations in order.

*I am still not 100% convinced there is a "problem" here except for Canon's marketing group which headlined 8K and then said "uhhh well sorta" in conjunction with not putting the bodies in the hands of anyone except explorer.


Well reasoned. I agree there will not be a problem for most users and use cases. The marketing should perhaps have been more cautious.

Not sure Canon, busy making great cameras, was prepared for the negative whiny world of youtube parasite do nothing reviewers and their dim-witted followers who eat up their conflict laden drivel. I am continually surprised at how low some of these reviewers will stoop, how large their egos, how gullible their audience is, and how they and their followers do almost everything on irrational emotion instead of thinking. It is truly sad that some of the most careless, abrasive, immature and just gross personalities are listened to by so many.

Canon probably thought the use case for high frame rate and high bit rate 4K modes and 8K would never be thought by anyone to be used for long form shooting. Enter a bunch of griping brand elitists and amateurs with unrealistic expectations and now there is an alleged fatal flaw in the product. Consumers have become dumb and spoiled, and the voice of un-reason is pandered to in the modern world, rather than ignored or refuted. It really is the most complex problem businesses face- how to please an increasingly whiny user base.
 
Last edited:
  • Like
Reactions: 1 users
Upvote 0

AlP

EOS R5
CR Pro
Sep 5, 2018
94
188
There is another thread here where results with an external recorder show that overheating in 4K HQ occurs only after over 70 minutes, despite the quality being identical to internal recording. So while heat is generated just by the read out, IBIS, downsampling and application of image profiles, it seems to me like a major source of heat is actually writing to the fast cards. The 8K overheats roughly the after the same time in RAW or compressed mode. So I don't think the encoding is the main source of heat.

So maybe Canon will offer a firmware upgrade allowing to operate the camera with the card door open??
Jokes aside, is there a major difference in terms of internal memory requirements between external recording and writing to the card(s)? There are 4 memory ICs around the processor and close to the card slot, so that area could quickly become a hot spot.
 
Upvote 0

Bert63

What’s in da box?
CR Pro
Dec 3, 2017
1,072
2,335
60
H265 takes more time to encode than H264. Do a test and see if your PC can encode 4K 60fps at near real-time with good quality to see the challenges faced and hence why you need hardware support. H265 was designed for 4K and 8k which is probably why it’s being used and not h264...


I have nothing in my house that will record anything at 4k60p or I would do this. My PC is a beast and it would be an interesting test.
 
Last edited:
Upvote 0
So maybe Canon will offer a firmware upgrade allowing to operate the camera with the card door open??
Jokes aside, is there a major difference in terms of internal memory requirements between external recording and writing to the card(s)? There are 4 memory ICs around the processor and close to the card slot, so that area could quickly become a hot spot.
Actually that would be a useful feature and apparently the a7s can do it - allows them to swap out a card in one memory slot while the other is being used!
 
  • Like
Reactions: 1 user
Upvote 0

Joules

doom
CR Pro
Jul 16, 2017
1,801
2,247
Hamburg, Germany
I thought HDMI had uncompressed video and the external recorder does the encoding?
I didn't claim anything else, did I? But I may have made some poor assumptions

My argument: Internal 8K H.265 overheats after ~25 minutes. Internal 8K RAW overheats after ~ 25 minutes. I was assuming that RAW does not involve any compression and therefore concluded that encoding is not the reason for the overheating. But, looking at the supposed bitrate of 2600 Mbps, there is some compression present, or my math is wrong: (8192 * 4320) pixels * 12 bit per pixel * 30 per second = 12,740 Mbps.

So yeah. We don't know the cause of the heat too well and therefore all this speculation about a possible fix seems pretty of the rails to me.
 
Upvote 0