Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Gaming on M395X: I'm satisfied
- Thread starter MandiMac
- Start date
- Sort by reaction score
Hmm it's curious that your real world usage on your machine doesn't seem to match what the barefeats benchmarks show.
Although barefeats mention high temps, they don't mention that this then throttles preformance. Perhaps we are all reading to much into numbers and missing the point that this is a great machine?
Although barefeats mention high temps, they don't mention that this then throttles preformance. Perhaps we are all reading to much into numbers and missing the point that this is a great machine?
Again, I think that Barefeats did play the games and tested all the usage under OS X, while I'm playing under Windows 10. So that factor is not to cast aside lightly. But yeah, I've got more.
I started StarCraft II with 43/47 degrees Celsius (GPU/CPU).
I stopped a good bit later with 53/60 degrees. Stutterfree and no problems.
Then I hopped onto Diablo III, because who cares: 50/58 degrees at the beginning.
Around 30-40 minutes later, I stopped with 51/56. It actually cooled down while being buttery.
Then I tried Shadow of Mordor on High Settings (which gave me around 45-60 FPS), in the tutorial it actually cooled down to 45/50 degrees. After playing for around 30 minutes, the temperatures were 56/59 degrees.
Last, but not least, I played now some Torchlight II with 50/56. It ended with 50/54, while I got around 75 frames/sec on the 5K resolution. You can imagine that the GPU was bored at 1440p and VSync on.
I can only repeat myself: The stories of the overheating are greatly exaggerated. I will try some BOINC overnight and log that somehow, let's see what a CPU stresstest will do.
Good point. From my limited understanding of processors (both CPU and GPU), experts say that processors have a certain 'personality', for lack of a better term. You can take two identical processors off the assembly line, and they will each perform slightly differently (performance/power consumption). I don't believe it is a huge variance, I am guessing it is in the range of 1-3% (just a guess). This could have something do with slight variations in both benchmarks and real world use. Also consider environmental conditions (temperature, humidity, elevation). Believe it or not, these do affect performance, even if ever so slightly.
EDIT: spelling
Serban, with which app do you measure these temperatures?
I need an app for Windows 10.
It seems not like it.
I give him the benefit of the doubt, however, that either
a) iStat Pro measures directly on the GPU itself and some gives you some sort of GPU temperature or
b) Everest measures not directly on the GPU.
I only have the figures I wrote above, so I'd need ideas what I can do unter Windows 10 to monitor the temperatures.
I am pleased by your numbers, but it's rather odd that Windows could possibly be giving better cooling performance than OS x when this is what this machine was designed to run, poor drivers!??
Actually it is not surprising at all. Games are known to be much more optimized for windows. Henceforth, they will utilize GPU more efficiently to achive same results as under OS X. Hence why they are able to run cooler and perform better.
Or if I can pu it other way: its not the cooling that is better, gfx card can render more frames faster while not running as hot due to better driver support and game optimization.
It is a great machine, the new one is slightly more powerful (5-10%) on CPU and GPU, and like 2-300% as for the SSD performance, so overall really not too bad
For the temperatures, we know since last year that the AMD 5Ks are hot, but does not mean it's a problem, to me it's only a fact, irrelevant to the day to day usage, mine is working 24/7 since December 2014 and it's still perfect.
I won't be surprise if in full load (running games @ 1440p with the VSYNC on is far from it) the temp are similar to the M295X, it's only a higher clocked rebrand, a very nice GPU in my opinion.
I did hours of tests and with the same games OS X and Windows are almost identical, for Blizzard games as an example, the OS X version is very well optimized, no difference to the Windows version.
So the temp and fan speed are exactly the same.
That's some truth right here.
As for me, I got BOINC, I got Afterburner and I let it and Everest both log their sensors. We'll see what this one turns out - but for now I'm off for a while, gonna need some sleep after all
Will report back asap!Any views on what can be expected on Starcraft2, Battlefield 4 or Elite Dangerous? I am more worried about Battlefield 5 (which is expected next year / 2017), also Rainbox six, The Division all come up next year. Will this give me a good enough playable experience. I am not an expert gamer, but someone that works in the creative world that although has a PS4 and a XBOX just hates the controller and wants a mouse and keyboard back.
Cannot make my mind up about going the top of range iMac or going Hackintosh and all the hassles that opens up.
Cannot make my mind up about going the top of range iMac or going Hackintosh and all the hassles that opens up.
What hassles? I took the hard drive out of my macbook pro and put it into a PC assembled with known compatible parts and guess what? It booted up right away into OSX. Clover bootloader is quite amazing because you can make any required fixes for onboard sound/audio over hdmi/ethernet (all very easy to find, thanks internet) without touching the OS files. That means things don't break when you update it like a normal mac via the app store. Also I get 300FPS on steam games in OSX. I don't even boot into windows for games since the only one I play is available on OSX also.
What difference did you notice? I know OSX has mouse acceleration on by default but smooth mouse fixes that. The textures being loaded are also the same. is it the framerate or something else?
Generall gameplay is different. Input lag from the mouse is incredible, that alone renders the game unplayable. Fps difference is also very much noticable, which makes game by far not as smooth. I've been playing cs for more than 15 years and know the game qute well.
I decided I will keep my pre-order with Bhphoto for the i7, M395X... Im tired of looking at at benchmarks some people are happy, others not so much making me question or not to keep every single day... it never ends this nightmare.. this is getting annoying. LoL
Going to stay with it and deal with it.. DONE!!
Hopefully they will ship mine this week..
Going to stay with it and deal with it.. DONE!!
Hopefully they will ship mine this week..
Thanks for mention from gian8989,
I always go for fastest CPU that I can afford within budget.
I never thought i7 and i5 could be an issue of heat difference until know.
I google the skylake profile (http://wccftech.com/intel-6th-gener...med-core-i76700k-core-i56600k-coming-q3-2015/)
The i7-6700K used in late 2015 5K comes with TDP about 91W.
There are 2 i5 CPU (Core i5-6500 and i5-6600), both have TDP about 65W.
It seems i7 could generate 40% more heat than i5 if all CPU work at maximal load.(of course i7 handle more jobs at that time, may have better benchmark, or more fps)
But that's difference occur when both CPU reach maximal workload. So, if I use VSync to lock the fps, I think the CPU workload for i5 and i7 should be the same. Are there still temperature differences between i5 and i7??