DrMrLordX
Lifer
- Apr 27, 2000
- 22,493
- 12,367
- 136
If I mounted a top fan as intake to point it straight at the VRM heatsinks, would that work? Would it mess up airflow for the rest of the case too much?
If you have one as exhaust, that would probably work better. Usually, ATX cases have intakes in the front and (maybe) on the side, and then exhaust in the back and on the top. Beyond that it's up to you whether you are trying to maintain positive pressure in the case.
I will only say this to you - I ran Cinebench R15
Did you run Prime95 smallFFTs? Cinebench is actually pretty relaxed. It doesn't even use 128-bit SIMD. It's SSE4.1a or something. The most current I can get my 1800x @ 4.0 GHz to draw is 135a (contrast this with y-cruncher, where I saw 152a at one point). I have Prime95 running right now and it's pulling 145a . . .
on this board with my 2700X at 4.25GHz and 1.45V with power draw for CPU well above 250W (390W system power consumption).
No active cooling of VRMs and board handled it fine.
1.45V? Ouch. How much current did it draw? HWiNFO64 should tell you. Honestly, I'd like to know. In something like y-cruncher or Prime95.
BTW if you think you need 160A for 3.5GHz on all cores on even 14nm ... my 12C Threadripper takes way less than that!
My 1800x easily pulls over 90a average @ 3.7 GHz with 1.2v vcore. You might have a better bin that lets you run lower voltage, but I would be fascinated to know what your Threadripper pulls @ 3.5-3.7 GHz with the lowest voltage you can maintain. In fact, testing that threadripper at different clocks and voltages might give us a pretty good idea of what current Matisse will want.
We still have to remember also that Matisse is a wider core. That will up current draw as well, particularly in workloads that can use the extra performance - notably, AVX2 workloads. I can easily see 16c Matisse pulling in the 140-180a range @ 3.7 GHz depending on what voltage it takes to get there and what workload you run. Believe me that when I get my 12x 3900x, I will do some benchmarking to see how current draw is affected by voltage and clockspeed. I already did this with my 1800x in the narrow clockspeed range that it will run via static overclocks (3.6 GHz and lower makes the machine lock up).
You can see it here:
http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=bookmarks/confirm?content_type=post&content_id=39840575
Peak current draw for 3.7 GHz @ 1.2v was 112a. If I were able to boot and run @ 1.0v then peak current draw would probably be in the 80-90a range. Double the core count, and you're likely looking at 160-180a. Add AVX2 capabilities . . . do you see where I am going here?
I have electronic engineering education and some marketing will not sway me from the facts.
Then deal in facts. I do not care for marketing. But I am aware that current draw is a real concern.
3.64GHz if we're splitting nickels, 3.64, 3.7, is 60Hz going to make a difference?
Like I said, blame the 1800x for having a crappy boost algorithm. If it had PBO2 or the equivalent, I'm sure it would hit 3.8-3.9 GHz with proper cooling and VRMs, and more if 14LPP weren't holding it back so much. Something it would never do on a 3-phase motherboard. Put a 2700x on there and see how far it goes . . .
I don't expect to see any but I'll report either way.
It won't unless you try hand overclocking, which is the only way to get an 1800x to run higher than 3.7GHz all-core. XFR was pretty lame on 1st-gen Ryzen chips. At best you'll get 3.7 GHz all-core and 4.1 GHz single-core. And that's if you cool the heck out of it.