Andrew Reid

Hurray at least. This is the important thing, that it gets out there and prompts Canon to rescue the situation as they should And that it gets out there credited to the right people, unlike with TONEH, even better. Never under estimate the power of embarrassment! Firmware update in T minus 10,9,8...

I have been a long-time follower of Roger's blog at Lens Rentals, which gives us not just a rare glimpse inside camera gear, but also an expert eye on construction quality, weather sealing and the repairability of it all when you drop it! To celebrate Roger's latest article I had the chance to put some questions his way, to dive into his thoughts on the 600mm, the potential for a 'universal' mirrorless mount and of course the EOS R5 overheating controversy. Read the full blog post and interview here: https://www.eoshd.com/news/an-interview-with-roger-cicala-of-lens-rentals-on-canon-rf-600mm-teardown-and-more/

@horshack Are you working with the 8K MP4 DAT files, or 4K HQ truncated recordings too? I can send you a few different flavours of file if this helps, might be interesting to compare the different codecs. I also need to test 4K HQ with the battery pull after starting and stopping recordings prior to one last sacrificial short clip to facilitate the battery pull during recording, and seeing what the camera remembers from the previous 4K HQ clip record times in NVRAM and what it doesn't. Card swap is also key to do.

😂😂😂😂😂😂😂😂😂 A CLASSIC.

Thanks Kaylee!! I think we should get a new guinea pig meme going.

Very good point! I am pretty sure the REAL temperature is checked by the ARM CPU and other components though, if not by Canon's firmware... And if the camera gets too hot it will likely hang or crash, so we'll have prior warning if there's trouble. I had no issues whatsoever doing the long 8K stints. Back did not get hotter after the first 10 minutes. By 50 minutes it was the same.

ARM based processors will likely have their own thermal hardware shut down algorithms and protections. I am 99% sure the timer is controlled by firmware, with timing metadata saved to the internal battery backed RAM on the PCB and possibly also to the clip metadata on the card.

Daniel I agree 100% We are customers at the end of the day and deserve some respect and to be levelled with in an honest way!! It really drives me up the wall, the fact that I can break all this news on the blog and in Canon's eyes - I AM the villain right?!!? ME! The guy who said Canon had finally turned a corner, rushed out to give them my hard earned £4200 only to find all my worst suspicious were true. The fact they have not apologised to me or to customers like you after all this investment is a DISGRACE. If a class action happens I would support it 1000% I cannot believe the depth of stupidity at the senior management level in Japan. Why do they feel it is ok to release a product with an "official overheating problem" let alone an artificial one?! It would be industrial level sabotage if it wasn't so self inflicted. I cannot believe how slow they have been to respond with an apology and promise to investigate and fix the firmware. They are actually standing their ground on this?? WHY? Out of fear they might lose a few sales in a niche business division that deals with expensive camcorders? The EOS business is surely bigger and should be pulling the strings, not the other way round.

I don't have one and won't be buying one that's for sure. Soon the R6 will be announced and I am sure people will put it through the same stringent tests as the R5. The field tests I've seen of the EOS R6 so far don't bode well, with cut off times as bad as, if not even worse. It has no fall-back "not limited by cripple hammer" 4K line skipped mode either so it is a 1080p camera basically

Did you miss the Sigma L mount 85mm F1.4? Looks great.

Hold on a minute and listen. We do not have a monster at all. It remains crippled. And in dire need of a firmware update. The actually reliable video modes are line skipped mush. But most of all think about your principals. It pisses me off to be honest when a company has been shown to cheat their customers, but then the customer just goes "ah what the heck" and rewards them anyway to the tune of 4 grand. If you really feel strongly about the fact that Canon have lied, or have even been "tricky bastards" - why would you then rush out to reward them for that with $4000 on a new camera?

Just in case anyone missed it here is the hack methodology in full: https://www.eoshd.com/news/this-is-my-canon-eos-r5-recording-8k-video-50-minutes-straight-until-empty-battery-video-evidence/ Here’s what I did during the 50 minutes 8K test: 00:00 – Power on the EOS R5 normal from cold 01:00 – Remove battery card door, camera powers down but power switch remains in “on” position 01:30 – Insert small screw in the battery door sensor as pictured below. Camera turns back on. 02:00 – Set video mode to DCI 8K resolution (IPB codec) to SD card. 02:10 – Hit record 17:00 – Roll to approximately 17 minutes (“overheat” warning icon will be flashing) 17:05 – At this point pull battery during recording 17:15 – With battery disconnected remove SD card and put freshly formatted SD card in the slot instead 17:20 – Push battery back in and power will return, overheat warning is gone and timer is reset 35:00 – Record for another 17 minutes! 50:00 – End of test I could go until the battery died, basically.

It has yet to be proven that 8K RAW goes indefinitely, but yes it is very impressive hardware. There may be thermal throttling or lock-ups somewhere during long sessions, we don't know - but what is now proven without a shadow of a doubt is how artificial the 15 minute limit is and how it counts down on a timer rather than a high maximum temperature cut off point. My opinion gets stronger every day that all this is about segmenting it from the Cinema EOS cameras. But we all know this is a stupid business decision, because most Cinema EOS camera owners wanted to buy an EOS R5 as a b-cam, second body, third body, handheld rig, you name it. In my opinion Canon seems to want to sell every professional shooter a minimum of 2 very expensive cameras for the same shoot - one single hybrid camera at $4000 plus very expensive high margin optics is not enough it seems. Shoot stills? Buy EOS R5. Shoot video? Buy C500 Mark II. Shoot both. Buy both. Yup it is terrible for us who just want to buy the EOS R5 for it's amazing hardware and cinematic images. We're always the one to get screwed it seems. This is why Canon, in my opinion, has been losing video users for years. It is faulty thinking, throughout the company. I don't know which part of Canon is most to blame - USA or Japan? Both? EOS or Cinema? Maybe both? Or do the EOS guys secretly really dislike the politics that the Cinema business has brought to the company and the restrictions it has placed on their products and all the stupid segmentation? Either way, they must change. I am up to 1 hour of 8K H.265 which is very special from a 45MP full frame sensor. It makes it even more frustrating that Canon has turned it into a lemon then! If Canon chose to lie to their customers about limitations on a $700 consumer camera that was boring and unexciting, the passion wouldn't quite be as strong to uncover the truth and fix it... That it is a very expensive flagship camera and the first of its kind in the world, makes it a hugely emotional topic for me & you, and rightly so.

Exactly - if nothing to hide, come out and say it... I honestly think they fear the legal consequences of this. And are uncertain on how to proceed policy wise. Canon are a big committee and do not move quickly on important decisions. It seems obvious to us - Just apologise and fix the problem, give us the impressive camera hardware as the engineers intended it and not crippled version with a firmware timer. For $4000 we are within our rights to request it. But to Canon it is probably all out political nuclear war and all sorts of manager arguments, in multiple departments and even multiple regions. I am sure some in the company are absolutely ashamed of what has happened. But they are not going to go above the level of the CEO and come out and say it publicly. Hierarchies are VERY strict in corporations, especially Japanese ones. We should give them yet more time to respond. It may be that Canon choose to ignore all their customers for 1-2 months completely and then just spring a firmware update on us. Honestly, it wouldn't surprise me if we never get an apology or really any honest answers.

It takes time for camera body to absorb the cold temperatures, and the inside is somewhat insulated. What the fridge does, is to keep the camera a nice 10-15C below the temperature it reports in a normal room. So it has no ambient temperature 'excuse' for overheating, and with no video recorded, no high CPU load excuse either, thus more proof of the timer.

Like I said in the article, I have approached everybody I know who works for Canon, and one of them is Michael Bravin, ex-ARRI, so we are not talking about the regional chinless wonder junior sales reps but people who have clout, who could make a difference and who I assume to be ethical people willing to help. Let me tell you... I have got nowhere fast. The same with Canon UK. I believe they are probably just sticking their heads in the sand or probably don't even understand the depth of customer anger there is about this. Or maybe we are a niche who don't make them the extra billions they are targeting so fuck us. Just one of the reason's we've a right to be very angry. Everybody who has spent $4000 on this camera (or even more in many regions) has a right to an honest explanation, and an apology, and most certainly a fix. I do agree that maybe a less harder, pro-corporate position would make for a more harmonious relationship between EOSHD and Canon. But I cannot sacrifice my principals - The right strategy at the moment is to simply carry on speaking the truth to bullshit, with factual evidence to back it up and as many deep technical discoveries as possible ideally with the help of Magic Lantern. We must pool our knowledge. Then Japan can do what the like to be quite honest. The ball has been in their court since day one and will continue to be. If we don't like what they have done, the onus is on us to stop being nice little easily-led consumers and to spend our money elsewhere! Perhaps even in the high end smartphone camera market instead.

Magically Screwed! Thanks for your good work Horshack! I'll be on hand again tomorrow if you need me try anything at this end.

It is because the camera may write the last clip length and 'overheat' timer status to the cripple clock memory when you hit stop on the recording. Going to test a few things around this aspect soon.

Ah Sony are just another bunch of opportunists really, using the suffering of a rival to drum up some more sales - I'm no great fan of the A7S III. I think this is a watershed moment, and I hope it encourages people to spend their money differently. Maybe buy more from Sigma, who are family owned. Let's put business ethics in the spotlight. Let's see what really goes on behind closed doors and what they try to hide from their customers.

The other ugly thing is that I have taken an absolute beating online with people questioning my reputation left right and centre for revealing this timer and behaviour from Canon. Canon meanwhile, brand reputation virtually untouched, not two hoots from them, no statement, no recall and the EOS R5 continues to fly off shelves!! BBC and CNN tech reporters should be all over it. Where are they?

Yes certainly possible but it is one heck of a reverse engineering project with many complexities and arduous effort, and would likely take a long time even for somebody as skilful and experienced as a1ex. I think after being mislead we deserve Canon to come out with an official fix, and a public apology. It would also be smart of Canon to make a development kit so that Magic Lantern didn't have to reverse engineer in order to add modules and code. Can you imagine if developers had to do this to make Android apps?!

Interesting point. I'll see if I can give it a go. First I need to work on a YouTube video showing it going for 1 hour in 8K. That should shut some people up.

On topic please people.

Updated article with more clarifications. It is battery backed RAM not NVRAM EEPROM (that's why the Baidu internal battery pull worked) but the basic workings are the same https://www.eoshd.com/news/canon-eos-r5-so-called-overheat-timer-defeated-by-a-single-screw-in-battery-door/ A lot to get our heads around, so please if you are commenting on this topic, make sure to read the full article first.