Sony a6300 announced

For what it's worth, the idea that adding 4K recording to the a6300 will mean it has to overheat seems like fairly unsupported speculation. I owned the A7S, A7RII, and A7SII and have never once had any of them overheat on me (and I used them as static cameras rolling 4k continuously for hours). And even if they did overheat on some people, that may have been more connected with IBIS which may restrict the cooling allowable on the sensor given that it has to be physically isolated from the other internals. My 2 year old $250 cellphone can record 4k video without overheating... Not to mention Sony released a firmware that reportedly helped reduce the incidence of the issue. Again I never had any of my cameras overheat before or after the newer firmware.

I'm also not sure any of your conclusions about the effect of sensor size on thermal dissipation are necessarily founded either. So much depends on the electrical design of the sensor array and design thresholds. People do like to theorize a lot on these forums and at some point enough repetition of unsupported theories seem to become prevailing wisdom, but that is a dangerous cycle that can make everyone less informed.

My S35 4k RED Scarlet had a huge heatsink and active cooling fan(s) and overheated far more easily than my passively cooled A7RII with isolated, stabilized sensor shooting 4k in FF35 or 5k (downscaled to 4k) in S35...
 
nyvz said:
For what it's worth, the idea that adding 4K recording to the a6300 will mean it has to overheat seems like fairly unsupported speculation. I owned the A7S, A7RII, and A7SII and have never once had any of them overheat on me (and I used them as static cameras rolling 4k continuously for hours). And even if they did overheat on some people, that may have been more connected with IBIS which may restrict the cooling allowable on the sensor given that it has to be physically isolated from the other internals. My 2 year old $250 cellphone can record 4k video without overheating... Not to mention Sony released a firmware that reportedly helped reduce the incidence of the issue. Again I never had any of my cameras overheat before or after the newer firmware.

I'm also not sure any of your conclusions about the effect of sensor size on thermal dissipation are necessarily founded either. So much depends on the electrical design of the sensor array and design thresholds. People do like to theorize a lot on these forums and at some point enough repetition of unsupported theories seem to become prevailing wisdom, but that is a dangerous cycle that can make everyone less informed.

My S35 4k RED Scarlet had a huge heatsink and active cooling fan(s) and overheated far more easily than my passively cooled A7RII with isolated, stabilized sensor shooting 4k in FF35 or 5k (downscaled to 4k) in S35...
Click to expand...

If you say you have been recording 4k for hours non stop, is it safe to assume you mean recorded externally, since internally it would clip at 30 min (and the a7s can't do it internally anyway)?

Because we already know that circumvents the heat issue which makes your argument a non starter.

And I agree that there is an awful lot of theorizing on forums. However, I base my personal thoughts on my personal experiences, occasionally discussing them with engineers when I get the opportunity to (and the bit of electrical engineering in my own blood and humble background comes in handy, although I don't claim to be any expert). So while I cant vouch for anyone else, I can say that I've noticed all small bodied Sony cameras that I personally have had the pleasure to use, from the nex models to the axxxx to the a7, have overheated in some capacity while recording video INTERNALLY. Externally was never an issue. Heck, my A5100 is only useful as a video recorder externally since it often has to cool off within 10 minutes of use outside when recording internally. This is not a theory, this is my experience personally.
The A7S was the first not to have this problem rear its head at me. Now why that happens I can only theorize based on what I think I understand about these cameras. It seems that the processing/encoding is what generates the most heat. When the encoder is idle and the camera is just acting as a dumb pipe, feeding the hdmi out on the A5100, no problem. Hit record, problem. Never heard of an a7s overheating until the ii made internal 4k an option. You could argue that's heresay as I didn't witness it myself (don't own the ii), but I can tell you that running externally also brings my battery life from 45 min of recording time on the a7s, to upwards of an hour and a half! If all that juice is being consumed by the internal encoding process, you don't need an electronic engineering degree to know some of the juice is being radiated out as heat somewhere in the pipeline. Looking at these observations, it isn't hard to draw a logical conclusion that the more difficult computationally 4k encoder would draw more juice, create more heat, and be the reason we hear about a7sii overheat cases in 4k, whereas 1080p or original a7s stories of that are nearly non existent. Again, without actually being a Sony engineer, these are just claims make. However they are rational explanations to observable behaviors.

Now, considering the A6000 like all compact apsc ILCs from Sony, used to overheat until the most recent firmware update (which made it less frequent, but didn't eliminate it as I'm told), all the more so adding a more computationally expensive process to it all-else-being-equal should make it worse.

Hey, I could be wrong. Heck I'd love to be. But I have good reason to believe this is the case and no reason to believe otherwise.
 
Last edited:
Oh, and to be clear, your cell phone is not doing the amount of work your camera is. I'd guess it is pixel binning and not sampling the whole sensor- compare it side by side with your RED and tell me if it is even in the same ballpark of 4K image quality!
And whether or not you think there is reason to believe me, the size of the sensor and amount of embedded copper wire length it would need to travel, I'm nearly certain has an effect on heat as well. Back in my college days, we studied CPU architecture and one of the concepts was that cpus are able to get faster and run cooler in newer generations by reducing the nanometer die size in manufacturing. Smaller circuits means shorter travel means less energy required and faster responses. They can be clocked higher and not require as much heat to be dissipated. So, I don't think it crazy to apply the same dynamic to sensor size the the copper wiring embedded within it.

But anyway, you get my point.
 
Last edited:
From what I've read on other threads, it does some punch in but not much (something like going from 1.53x crop to 1.60x crop, though we're still not sure).
 
It has a 3:2 24 MPX (6,000 x 4,000) sensor. It does 4K in 16:9 (6,000 x 3,376, oversampled). So, off specs, it uses the entire available width. No crop.

PS. I would think that 14 nM processors - which is the current standard - would produce a lot less heat than the previous generation's 28 nM.
 
Don't forget - cameras aren't like cell phones or desktop computers, they don't necessarily use whatever is "current generation" from a microprocessor perspective. Heck, I remember waiting for Canon to bring out the digic 5, but they kept rolling out the same digic 4 chips for years, putting the same ones off the same production line each year until they design a new one. Sony isn't much different. The a6300 is still using the Bionz X chip, and the sensor is new but not using their newest tech either.
 
It's true that cameras, especially professional cameras, seem to lag a few years in internal technology. Ironically cell phones seem higher-tech, quite probably because of two things: (1) far more units sold and (2) fiercer competition. Take the latest chip, the size of your fingernail and about $30 each:

Snapdragon 820

- debayer and process from up to a 28 MP sensor at 30 fps (higher fps from lesser MP, I'm sure)
- record up to 4K to H.264 or H.265
- all this while doing other cell-phony things like WiFi, driving a 4K 60fps display, etc.

and probably not overheating.
 
combatentropy said:
It's true that cameras, especially professional cameras, seem to lag a few years in internal technology. Ironically cell phones seem higher-tech, quite probably because of two things: (1) far more units sold and (2) fiercer competition. Take the latest chip, the size of your fingernail and about $30 each:

Snapdragon 820

- debayer and process from up to a 28 MP sensor at 30 fps (higher fps from lesser MP, I'm sure)
- record up to 4K to H.264 or H.265
- all this while doing other cell-phony things like WiFi, driving a 4K 60fps display, etc.

and probably not overheating.
Click to expand...

Cellphones have been increasingly having more and more overheating problems, and have built in throttling to prevent it. The technology is well ahead of camera processors though, for sure.
 
combatentropy said:
It's true that cameras, especially professional cameras, seem to lag a few years in internal technology. Ironically cell phones seem higher-tech, quite probably because of two things: (1) far more units sold and (2) fiercer competition. Take the latest chip, the size of your fingernail and about $30 each:

Snapdragon 820

- debayer and process from up to a 28 MP sensor at 30 fps (higher fps from lesser MP, I'm sure)
- record up to 4K to H.264 or H.265
- all this while doing other cell-phony things like WiFi, driving a 4K 60fps display, etc.

and probably not overheating.
Click to expand...

Yes, well, you have to understand that the cell phone market has artificially accelerated the rate in which "generations" of tech come out- mostly for the reasons you stated above, I'd guess. Its not that the camera market is slow, its that the mobile market is crazy-stupid-fast in ways that aren't cost effective in most other markets.
I mean, I remember seeing articles in the EE Times and seeing proof of concept for certain SOC designs or RFID tech like NFC with patents filed the better half of a decade before we saw them in the first consumer device. Just because it exists doesn't mean it is ready to be used (cost effective or otherwise) in all applications. The mobile market, however, sells in such quantity and in such fierce competition that I've never seen concepts rushed to the hands of consumers quite like this before.

The BIONZ X chip has been around for quite a few models of camera already by Sony. The technology is out there already to make a newer and more efficient one, and probably was already out there before the first BIONZX camera shipped. But the current model still does an admirable job and probably has a bit more life left in the eyes of Sony before they decide it needs an overhaul. And that overhaul will probably skip a generation or two of "mobile generation technology", and then be on the bleeding edge of state-of-the-art until it eventually falls behind and demand is finally there for the next one.

TL;DR: Cameras and Cell Phones can't be compared in terms of technological generations. And there's nothing ironic about this at all.
 
M. Gilden said:
... TL;DR: Cameras and Cell Phones can't be compared in terms of technological generations. And there's nothing ironic about this at all.
Click to expand...
That may be true about Canon and Nikon but it certainly does not seem to be the case with Sony over the last couple of years. A7S/II, A7RII, RX100 IV, RX 10 II and now A6300 certainly represent far more than marginal improvement over their predecessors. On their pro tier cameras, FS7 has been a grand slam (and I am not talking about Denny's here) as well. The ILC market is still pretty strong but companies not involved in the smartphone industry find it much more difficult to spread their R&D costs over a wide range of products. Plus, Sony more than dabbles in the high end pro market with their F-series cinema products. IMO, Canon takes a couple of more naps and it'll find itself competing with Kodak for the title of the company that once was.
 
The A6300's sensor is a new design, now using copper wire instead of aluminum. The wire is also smaller than the previous generation, possibly by a significant amount (180nm -> 45nm rumored). It's small enough to give an IQ bump similar to BSI by getting the wire out of the way of the sensor. This could also reduce/eliminate the occurrence of overheating completely. Either way, there's less than 3 weeks left till the camera ships, so we'll know soon enough. ;)
 
M. Gilden said:
If you say you have been recording 4k for hours non stop, is it safe to assume you mean recorded externally, since internally it would clip at 30 min (and the a7s can't do it internally anyway)?
Click to expand...

No, I'm talking about recording 4k internally and re-rerolling immediately every 30min.
 
M. Gilden said:
Oh, and to be clear, your cell phone is not doing the amount of work your camera is. I'd guess it is pixel binning and not sampling the whole sensor- compare it side by side with your RED and tell me if it is even in the same ballpark of 4K image quality!
Click to expand...

It's not pixel binning, it's a 13MP 4:3 sensor so the 4K is an unbinned 1.89:1 crop with nearly the same horizontal field of view... If I was talking about binned 1080p or low-bitrates or external recording I would have noted that, you seem to be looking for some very obvious way around my argument so you don't have to justify your lack of evidence for yours. Why would my phone recording 100Mbps unbinned 4k mp4 files be doing a lot less work than another camera also recording 100Mbps 4k mp4 files? I think you are trying to draw a connection that while sometimes true, is hardly a rule especially given the many different pixel, array, and processing architecture designs...
 
Last edited:
M. Gilden said:
Oh, and to be clear, your cell phone is not doing the amount of work your camera is.
Click to expand...

I thought you wanted something like a RED camera to do less work to the image, not more.

When Andromeda came out, allowing raw output from the DVX100, it was like a revelation, all because it output uncompressed video, leaving the grading up to us, not the camera's presets.
When Ikonoskop came out, its images were beautiful too, again because it was unprocessed. For this reason it also had no fan.

We all should know by now that compressing a raw signal to MPEG-4 is far more processor-intensive than any magic sauce a high-end camera is doing to an image before encoding it to a far more lightly compressed image. So I would say the iPhone is doing far more work than any professional camera. Cell phones even have those built-in special effects, on top of all the other stuff.

Professional cameras have better images mainly because: (1) you can use a better lens, and (2) the sensor is bigger, not because it's doing more processing. They're not even processing more pixels or higher frame rates than cell phones anymore.

Professional cameras cost more mainly because of demand. If you lined up every person in the world and offered them an iPhone 6S or a RED Epic for free, 99% of them would choose the iPhone. The RED Epic is this huge beast that they can't put in their pocket, much less fit in their suitcase, and they will look like a total dork carrying it around on their next night on the town. However, if for some reason, RED could sell 10 million RED Epics instead of 10,000, you can bet that a RED Epic could sell for $1,000 and still make a profit.
 
I think this camera may find its way into a lot of camera bags. Its so tiny and the quality to pretty ridiculous for what it is. Man, I remember shooting on a tricked out HVX200 with REDrock system and firestore drive....ect....weighed about a million pounds....that sucked.

This little feather light camera is really a heavy lifter.
 
I was almost sold on this camera and that footage just sealed the deal. I will wait for the first wave of reviews though, to make the final decision.
 
I'm also looking at the 6300 to use as sidekick to my FS7. But as I have no intention buying any other lens than the set of ten Canon L lenses I already have question is how it'll work with these and speedbooster?
 
You must log in or register to reply here.
Back
Top