AMD Radeon R9 M295X Core Clock Throttling, Heat, and Performance

[thundersteele]

PS. Steve23094 (or anybody) would you be so kind and make a practical test: set fan manually to run 2300rpm (or whatever the Apple default maximum is) and game about one hour (some modern game, whatever you like to game). Log the maximum GPU temp with GPU-Z. Then repeat the test by setting fan to maximum 2700rpm and check the same GPU temp (of course restarting the GPU-Z first to reset the logged maximum temp value). Is there any difference? I really liked to see couple of celcius lower temps with max 2700rpm.

I didn't do exactly what you requested. But this is what I observe on my 5k iMac (i5, M295X). Note that I don't use any third party fan control or other tools that might mess with the fans, and I do everything in OSX.

Gaming: Max fan speed 2300rpm (GPU up to 104C), no visible sign of throttling
GPU benchmark: Max fan speed 2300rpm (GPU up to 105C), no visible sign of throttling
Prime95 + GPU benchmark: Fan speed up to 2700rpm, CPU throttles (100C max temp), GPU up to 105C, no sign of GPU throttling

The main message from this for me is that the system will ramp up the fan to 2700rpm if necessary. Does this mean the GPU is fine with 2300rpm, not throttling? I don't know, but it looks like that to me.


Overall one should remember that cooling is limited in an all-in-one. Throttling in max load (GPU+CPU) situations is acceptable, though not optimal. Now, I come from a series of MBPs, so even under full load the noise level and performance is much better on the 5k iMac. Those who come from a 2012/2013 iMac might experience this differently, and I understand that.

 

[tmorterlaing]

Well, I was the guy who said that it would be better with 2700rpm than 2300rpm. And yes, I am still thinking the same way.

Also this proves, in opposite what Astelith claims, and the same I on my earlier posts said, that the cooling of this iMac is just insufficient. When on 100% CPU load + 100% GPU load it cannot keep temperature in designed limits even with the max fan speed 2700rpm and ultimatelly CPU plus GPU needs to throttle to lower the temps.

NOTE! My point however is that I am not happy with the insufficiently built cooling system of riMac, but at this time I am so keen on the 5k display so I may be able to live with such way to tinker fan speeds with 3rd party apps.

(Can't believe I'm doing a switcharound here) but form what I can tell, the above tests almost prove it's not insufficient... to an extent. If you can name an app the will run the CPU and GPU at 100 percent, I'd be surprised. I can't think of one, I and I professionally use After Effects, an array of editing software, youtube in the background checking vids, separate screen attached etc. It's quite difficult to push all 4 cores of the CPU to full load. Very difficult in fact. I don't know about he GPU, but the i7 at least is one hell of a CPU for an all in one.

PS. Steve23094 (or anybody) would you be so kind and make a practical test: set fan manually to run 2300rpm (or whatever the Apple default maximum is) and game about one hour (some modern game, whatever you like to game). Log the maximum GPU temp with GPU-Z. Then repeat the test by setting fan to maximum 2700rpm and check the same GPU temp (of course restarting the GPU-Z first to reset the logged maximum temp value). Is there any difference? I really liked to see couple of celcius lower temps with max 2700rpm.

Well, we tried the inverse of that, with the fan running at 2000RPM and that didnt seem to make a difference

The fan at 2300rpm is 38dB, according to the Apple discussions board. Either way, it's pretty darn loud, which is what caused my initial concern and led me here. It's also loud enough for hundreds/thousands of others to question it online too.

I also found the fan spinning up on desktop applications - even just exporting and playing 1080p 60FPS video in iMovie or QuickTime. A few tasks that my 2010 iMac breezes through, caused heat and fan noise on the 5K iMac. Even a few times watching YouTube. Obviously it's a lot faster, but at a heat/noise compromise. There's quite a few posts about this in particular as it's more common than gaming.

Those who "never hear the fan" simply don't do any activities that tax the system. We've proved that here.

I don't think we should compare a fully-loaded iMac to a low-end laptop with no discrete GPU. Instead, compare against previous fully-loaded iMacs. The best would be 2012/2013 models as they are identical in terms of internal/external design. I haven't see any throttling on those, but maybe they do...?

You'll be able to make your own decision when it arrives. My opinion of another current gen iMac would be forever tainted by the findings in this thread.

I agree, the only reason I'm doing a sell and buy of the new one is so I'm up to date with imacs. Then when the next (hopefully quieter!) one comes out I'll be doing another 'exchange'. One thing though, what sort of heat and noise were you looking at with your video example? I just played a 1080P MKV through VLC for half an hour and the fan got to about 1630rpm at max - which I still consider quiet enough (thats about what my old one would do when gaming).

The main message from this for me is that the system will ramp up the fan to 2700rpm if necessary. Does this mean the GPU is fine with 2300rpm, not throttling? I don't know, but it looks like that to me.

Looks that way - it looks as if the 2700rpm is observed for times when everything is under load, not jus the GPU. Which i guess would be a good sign. The cooling isn't quite good enough for full load on everything, but again - I'm not sure what real life implications that has. I for one can't actually get my CPU and GPU to max out at the same time.... well maybe I could, if I was going to try and export a video from Premier, render prores files from after effects, encode with MPEG streamclip and play a game at the same time..... But anyone actually needing to do that should be looking at a Mac pro anyway...IMO.

Overall one should remember that cooling is limited in an all-in-one. Throttling in max load (GPU+CPU) situations is acceptable, though not optimal. Now, I come from a series of MBPs, so even under full load the noise level and performance is much better on the 5k iMac. Those who come from a 2012/2013 iMac might experience this differently, and I understand that.

Still, 2300rpm is too loud for an imac. Apple have unfortunately set themselves a bar of quietness that they should be matching now (subjectively speaking of course). For them to show that level of noise isnt distracting would suggest they didnt test the prototypes under load and actually think about it, which is concerning as far as their quality checking attitude is concerned.

 

[Alesc]

(Can't believe I'm doing a switcharound here) but form what I can tell, the above tests almost prove it's not insufficient... to an extent. If you can name an app the will run the CPU and GPU at 100 percent, I'd be surprised. I can't think of one, I and I professionally use After Effects, an array of editing software, youtube in the background checking vids, separate screen attached etc. It's quite difficult to push all 4 cores of the CPU to full load. Very difficult in fact. I don't know about he GPU, but the i7 at least is one hell of a CPU for an all in one.
[...]

If you wan't to see the 8 cores at 100%, open Terminal and run 8 or 9 "yes" processes: it works
http://osxdaily.com/2012/10/02/stress-test-mac-cpu/

 

[cynics]

I didn't do exactly what you requested. But this is what I observe on my 5k iMac (i5, M295X). Note that I don't use any third party fan control or other tools that might mess with the fans, and I do everything in OSX.



Gaming: Max fan speed 2300rpm (GPU up to 104C), no visible sign of throttling

GPU benchmark: Max fan speed 2300rpm (GPU up to 105C), no visible sign of throttling

Prime95 + GPU benchmark: Fan speed up to 2700rpm, CPU throttles (100C max temp), GPU up to 105C, no sign of GPU throttling



The main message from this for me is that the system will ramp up the fan to 2700rpm if necessary. Does this mean the GPU is fine with 2300rpm, not throttling? I don't know, but it looks like that to me.





Overall one should remember that cooling is limited in an all-in-one. Throttling in max load (GPU+CPU) situations is acceptable, though not optimal. Now, I come from a series of MBPs, so even under full load the noise level and performance is much better on the 5k iMac. Those who come from a 2012/2013 iMac might experience this differently, and I understand that.

The bolded is where opinion will vary. An all-in-one is just that, all the components in one device. Therefore Logic will dictate that it's cooling capacity should be determined by the components with in.

If you built a computer and saw any signs of processor throttling you didn't do a good job building that computer. Same would apply if you were capable of building an all-in-one.

This certainly should be "the nature of the beast" situation.

I'm very confident the next generation will have a different design to better cooling and thus quieter operation. That will leave of questioning the current design as not very good. That's just a guess though.

 

[tmorterlaing]

The bolded is where opinion will vary. An all-in-one is just that, all the components in one device. Therefore Logic will dictate that it's cooling capacity should be determined by the components with in.

If you built a computer and saw any signs of processor throttling you didn't do a good job building that computer. Same would apply if you were capable of building an all-in-one.

This certainly should be "the nature of the beast" situation.

I'm very confident the next generation will have a different design to better cooling and thus quieter operation. That will leave of questioning the current design as not very good. That's just a guess though.

I hate to, but I agree again - this is a $4000 machine, and 'optimal' really shouldn't be all we're hoping for. When do we think the next riMac will be out(ish)? I for one will be selling this one the second the new on is announced!

 

[steve62388]

I hate to, but I agree again

You're very agreeable.

 

[tmorterlaing]

You're very agreeable.

Thanks... I think It's just I can see both sides of the argument. I don't think this is a big deal, because in the real word it doesnt make any difference to anything.

However, I do think it's disappointing Apple have made such a loud product is all. I'm interested to see how this progresses, and also if they stick with AMD for the next batch of rimacs. It's almost as if they use AMD in their trial machines whilst they wait for NVIDEA to create something they actually want to use....

 

[phpmaven]

Thanks... I think It's just I can see both sides of the argument. I don't think this is a big deal, because in the real word it doesnt make any difference to anything.

However, I do think it's disappointing Apple have made such a loud product is all. I'm interested to see how this progresses, and also if they stick with AMD for the next batch of rimacs. It's almost as if they use AMD in their trial machines whilst they wait for NVIDEO to create something they actually want to use....

"Loud product"? That's a good one

This is by far the quietest computer I've ever owned, period. I can't believe we are still flogging this dead horse.

 

[ioxakoyba]

I'm very confident the next generation will have a different design to better cooling and thus quieter operation. That will leave of questioning the current design as not very good. That's just a guess though.

How are you so confident about that? Considering Apple's last "innovations" in the past year or two, I could bet my horses on that Apple just uses the same design, same unefficient cooling system and puts a even more heat producing and even more throttling AMD R9 m395X inside the riMAc late-2015. And everybody's just A m a z e d !

If Apple has made this product so bad, and don't care about it, why on earth they would care about it year later either? After all it's just a percentile who even notices that this thing throttles - the other 99% are just so amazed of this product that has beautiful 5k screen sharp an thin as razor.

But your guess is just as good as mine.

 

[cynics]

"Loud product"? That's a good one



This is by far the quietest computer I've ever owned, period. I can't believe we are still flogging this dead horse.

You've never owned any other iMac?

 

[phpmaven]

You've never owned any other iMac?

I own several. I've owned dozens of computers in my life, and this is the quietest. I can just barely hear the fan if I put my ear right up to the bottom of the screen.

 

[nj-mac-user]

I own several. I've owned dozens of computers in my life, and this is the quietest. I can just barely hear the fan if I put my ear right up to the bottom of the screen.

I agree. Granted I own a 5K iMac i5 model with the lessor of the two GPU's, this iMac is quieter than my previous 27" iMac when the fan kicks into higher RPMs. Then again, all of this "its too loud"... "its not loud at all" is all relative. Some people have more sensitive hearing than others even if they claim they don't.

 

[andy9l]

"Loud product"? That's a good one



This is by far the quietest computer I've ever owned, period. I can't believe we are still flogging this dead horse.

The past few pages prove the horse is very much alive. Whether or not it bothers you or affects your work is another topic. Your GPU throttles under load because the iMac fails to cool it sufficiently, assuming you have the M295X.

I own several. I've owned dozens of computers in my life, and this is the quietest. I can just barely hear the fan if I put my ear right up to the bottom of the screen.

Agreed. But this thread is discussing periods where the GPU is being pushed.

I agree. Granted I own a 5K iMac i5 model with the lessor of the two GPU's, this iMac is quieter than my previous 27" iMac when the fan kicks into higher RPMs. Then again, all of this "its too loud"... "its not loud at all" is all relative. Some people have more sensitive hearing than others even if they claim they don't.

The M290X is not in question, at all.

 

[cynics]

I own several. I've owned dozens of computers in my life, and this is the quietest. I can just barely hear the fan if I put my ear right up to the bottom of the screen.

Well sure, sitting idle its very quiet. Not the quietest computer I've ever used but certainly quiet. Its also quiet if its turned completely off.

But as the topic dictates I'm referring to the required fan RPM relative to usage vs previous Macs.

As we've seen with the Valley benchmark, same settings as certain previous iMacs, with similar results only difference being the RiMacs temps and fan noise. And if its benchmarking the same on the same resolution but fan is running at 2300 RPM vs 1200 RPM, well its not the quietest.

 

[nj-mac-user]

The past few pages prove the horse is very much alive. Whether or not it bothers you or affects your work is another topic. Your GPU throttles under load because the iMac fails to cool it sufficiently, assuming you have the M295X.




Agreed. But this thread is discussing periods where the GPU is being pushed.




The M290X is not in question, at all.

I know. I was referring strictly to fan noise when hot.

 

[tmorterlaing]

I agree. Granted I own a 5K iMac i5 model with the lessor of the two GPU's, this iMac is quieter than my previous 27" iMac when the fan kicks into higher RPMs. Then again, all of this "its too loud"... "its not loud at all" is all relative. Some people have more sensitive hearing than others even if they claim they don't.

It's not relative actually. If you put measure the sound in decibels in the same environment its not a subjective thing on any level. The rimac with the M295X is significantly louder under GPU load. The only subjective bit is whether that bothers you or not.

 

[Astelith]

It's not relative actually. If you put measure the sound in decibels in the same environment its not a subjective thing on any level. The rimac with the M295X is significantly louder under GPU load. The only subjective bit is whether that bothers you or not.

The worst thing is that some iMac are cooler than other (and I'm a lucky one), this is a fact, and probably it's a poor assembly issue.
For such an expensive machine we are expectng more quality, shame on Apple.
So far we all have more or less the same performance but with very different noise level due to the fan with 400 rpm of difference.
In my opinion, who have "the noisy version" should give it back to Apple and get a proper one

 

[tmorterlaing]

The worst thing is that some iMac are cooler than other (and I'm a lucky one), this is a fact, and probably it's a poor assembly issue.
For such an expensive machine we are expectng more quality, shame on Apple.
So far we all have more or less the same performance but with very different noise level due to the fan with 400 rpm of difference.
In my opinion, who have "the noisy version" should give it back to Apple and get a proper one

My replacement *should* arrive tomorrow - I'll let you now how I get one with it Interestingly perhaps it was delayed actually - it had an original shipping date of the 9th being the latest it would arrive, which got changed....wondering if this has anything to do with it... probably not. Wishful thinking

 

[Astelith]

My replacement *should* arrive tomorrow - I'll let you now how I get one with it Interestingly perhaps it was delayed actually - it had an original shipping date of the 9th being the latest it would arrive, which got changed....wondering if this has anything to do with it... probably not. Wishful thinking

So a big Good Luck ! Hope you'll get a lucky one, let us know

 

[steve62388]

If you put measure the sound in decibels in the same environment its not a subjective thing on any level.

I have only seen one post here (and no reviews) from someone that has actually used a decibel meter to measure it's volume, and that was just on the riMac so there was no comparison to previous models.

The only subjective bit is whether that bothers you or not.

This is true.

 

[ha1o2surfer]

So, what happened in my video where the clock speed dropped to the 700s?

http://vimeo.com/116208647

I mean, the correlation between FPS drop, clock speed throttle and GPU heat seems to be 1:1...?

WOW that temp rises fast!!! In order to get my 970 to rise that fast, I stopped the fan with my finger. and even then it won't hit 100c!

 

[cynics]

I have only seen one post here (and no reviews) from someone that has actually used a decibel meter to measure it's volume, and that was just on the riMac so there was no comparison to previous models.







This is true.

I feel there is a bit of confusion (maybe it's me).

But when we are talking about noise it's in reference to the RiMac needing to utilize the fan more and at higher speeds in comparison to other iMacs.

At any given RPM the RiMac and late 2013 sound the same. Or at least very similar, any differences you can probably find just comparing two 2013s or 2 RiMacs to each other.

So they sound the same/similar. However performing the same task the RiMac can be much louder. Simcity for example, on its maximum settings, my iMac never increases off 1200 RPM, my buddies RiMac runs at at least 1800 - 2000 RPM (full screen hard to tell exactly). It's much louder with no visual or other noticeable differences in game.

So it ends up being a much louder computer.

----------

As far as decibel meters go there are plenty of apps. However accuracy is questionable so one would need to use the same device to measure on both machines for just a comparison.

 

[andy9l]

@cynics, no confusion - what you've written is spot on.

 

[WilliamG]

The worst thing is that some iMac are cooler than other (and I'm a lucky one), this is a fact, and probably it's a poor assembly issue.
For such an expensive machine we are expectng more quality, shame on Apple.
So far we all have more or less the same performance but with very different noise level due to the fan with 400 rpm of difference.
In my opinion, who have "the noisy version" should give it back to Apple and get a proper one

And again, NONSENSE. I'm sorry, Astelith, but all you're doing is making poor people go through hassle of exchanging their iMacs because YOUR iMac is not working according to how Apple specified.

ALL RiMacs should be running at 2300rpm under "normal" CPU/GPU load. I think I read earlier that it's possible to REALLY stress (artificially) an RiMac so it hit 2700rpm, but it should be 2300rpm.

I've confirmed this on THREE different RiMacs - all 3 have hit between 105-106C GPU under gaming load, and all 3 hit 2300rpm and stay there. Normally I'd agree that my evidence (like yours) is anecdotal, but that everyone else EXCEPT you is experiencing 2300rpm fan speeds should be telling you something. You didn't get some magical unicorn iMac!

----------

My replacement *should* arrive tomorrow - I'll let you now how I get one with it Interestingly perhaps it was delayed actually - it had an original shipping date of the 9th being the latest it would arrive, which got changed....wondering if this has anything to do with it... probably not. Wishful thinking

I'm not sure what you're replacing your iMac for, but if it's because your iMac is perceived as "loud," your replacement will be exactly the same. 2300rpm under GPU/CPU load. I don't intend to sugar-coat these responses anymore, since I firmly believe Astelith is putting false hope into a LOT of people's minds lately.

There's something wrong with his system, anyway, given how long it takes him to alt-tab/command-tab between games and desktop (not to mention his GPU temps are the same as ours, period, which he continues to deny).

----------

I feel there is a bit of confusion (maybe it's me).

But when we are talking about noise it's in reference to the RiMac needing to utilize the fan more and at higher speeds in comparison to other iMacs.

At any given RPM the RiMac and late 2013 sound the same. Or at least very similar, any differences you can probably find just comparing two 2013s or 2 RiMacs to each other.

So they sound the same/similar. However performing the same task the RiMac can be much louder. Simcity for example, on its maximum settings, my iMac never increases off 1200 RPM, my buddies RiMac runs at at least 1800 - 2000 RPM (full screen hard to tell exactly). It's much louder with no visual or other noticeable differences in game.

So it ends up being a much louder computer.

----------

As far as decibel meters go there are plenty of apps. However accuracy is questionable so one would need to use the same device to measure on both machines for just a comparison.

This is EXACTLY true. I had a top-end 27" 2012 iMac, and now a top-end 2014 RiMac. They both sound EXACTLY the same. The issue is that the 2014 RiMac hits the higher fan speeds much, much more easily than the 2012 model did under the same circumstances.

For example, Diablo 3 in OS X. I know my 2012 iMac was very, very quiet playing this game in OS X. As I recall, it never went above 1200rpm (default speed). With the RiMac? Fans ramp up to 2300rpm in not very long.

 

[Astelith]

And again, NONSENSE. I'm sorry, Astelith, but all you're doing is making poor people go through hassle of exchanging their iMacs because YOUR iMac is not working according to how Apple specified.

ALL RiMacs should be running at 2300rpm under "normal" CPU/GPU load. I think I read earlier that it's possible to REALLY stress (artificially) an RiMac so it hit 2700rpm, but it should be 2300rpm.

I've confirmed this on THREE different RiMacs - all 3 have hit between 105-106C GPU under gaming load, and all 3 hit 2300rpm and stay there. Normally I'd agree that my evidence (like yours) is anecdotal, but that everyone else EXCEPT you is experiencing 2300rpm fan speeds should be telling you something. You didn't get some magical unicorn iMac!

----------



I'm not sure what you're replacing your iMac for, but if it's because your iMac is perceived as "loud," your replacement will be exactly the same. 2300rpm under GPU/CPU load. I don't intend to sugar-coat these responses anymore, since I firmly believe Astelith is putting false hope into a LOT of people's minds lately.

There's something wrong with his system, anyway, given how long it takes him to alt-tab/command-tab between games and desktop (not to mention his GPU temps are the same as ours, period, which he continues to deny).

----------



This is EXACTLY true. I had a top-end 27" 2012 iMac, and now a top-end 2014 RiMac. They both sound EXACTLY the same. The issue is that the 2014 RiMac hits the higher fan speeds much, much more easily than the 2012 model did under the same circumstances.

For example, Diablo 3 in OS X. I know my 2012 iMac was very, very quiet playing this game in OS X. As I recall, it never went above 1200rpm (default speed). With the RiMac? Fans ramp up to 2300rpm in not very long.

So my iMac is broken because it's cooler, mhh... ok... like the other who voted <100 degrees in the valley benchmark's thread.

Happy to have a faulty machine then