E5-1650 v4 throttle on Prime-95

M

MysteryGuy

n00b
Joined
Jan 31, 2017
Messages
3
Hi;

I'm trying to determine if (what appears to be) unexpected (below base frequency) throttling on my new system is expected or not. When I run various stress test programs like Prime-95 and the Intel Processor Diagnostic Tools Floating Point and Prime number tests, I see the multiplier fall below the 'Base Frequency'.

The symptom I'm seeing on my new system is that some stress tests (like Intel XTU Stress CPU, and the Intel Processor Diagnostic tool CPU Load test) will run up to all available cores at 38x. (The E5-1650 v4 says its base frequency is 3600 MHz, and the max Turbo boost is 4000 MHz).

But, running one or more threads of Prime-95 will cause the cores running these threads to down-clock to 35X, which is below the base of 3600 MHz.

When this is happening, cores not running Prime-95 still report 38X (under the Windows 7 High performance profile). Temps. all appear to be fine.

Similarly, I see slower 35x multiplier use when running the Intel XTU benchmark, and the Intel Processor diagnostic tools Floating point and Prime number tests.

I don't know if it's related, but I came across an Intel document [Intel Xeon Processor E5 v3 Product Family Processor Specification Update] that indicates reduced processor speeds may be encountered when AVX instructions are run on an E5-1650 v3.

I have an E5-1650 v4, but can not immediately find similar specifications for it (as to if and how much throttling would be expected for AVX usage). The version of Prime-95 I'm running (28.10) does say it's using AVX or AVX-2 instructions.

And , if I turn off 'AVX' for Prime-95 in 'local.txt', then I see up to 12 Prime-95 threads all running at 38x/ 3800 Mhz.

I'm trying to figure out if what I see is expected for this CPU. Anyone have an E5-1650 v3 or v4 and can tell me if you see similar throttling when running the current Prime-95 (v 28.10), etc. ?

Any help would be appreciated.

Thanks;

P.S.: My system is an Asus X99-E WS/USB 3.1 with BIOS 3402 running Windows 7 64-bit.
 
Change your CPU Power setting within windows to min CPU Usage XX% and set your power profile to high performance. Tell me what it does after that.
 
This is possibly a TDP issue, or possibly a violation of Icc Max. Download and run CoreTemp and watch the power consumption when under Prime95, if the current TDP under load exceeds the TDP limits (short, long and max) as listed in HWinfo64 5.38, it'll be your problem. Unfortunately, along with locking the Turbo multiplier, Intel also locked the max TDP on the E5-1600V4s, so you can't override it.

My suggestion....sell the V4 and buy a unlocked E5-1600V3 that has limits that you can override...;)
 
Some Xeons got different AVX2 clocks unlike desktop parts.

What is the powerdraw and temperature under the load?
 
Shintai said:
Some Xeons got different AVX2 clocks unlike desktop parts.

What is the powerdraw and temperature under the load?
Click to expand...

I'm beginning to think that this might just be how these are :(. The throttling doesn't seem related to total power as much as running something (even on one core) that triggers it for that core.

I don't see the issue when running Prime-95 without AVX enabled.

The peek powers I see (as reported by AIDA64) are:

Idle : 12X, 8.47 W, 11 A

Prime-95, one thread, AVXs enabled, Throttling 35X (one core) = 24.46 W, 22 A, CPU 1.112 V

Prime-95, 12 threads, AVXs enabled, Throttling 35X (all cores) = 59.60 W, 59 A, 1.010 V

Prime-95, one thread, AVXs disabled, No Throttling 38X (one core) = 28.58 W, 24 A, 1.191 V

Prime-95, 12 threads, AVXs disabled, No Throttling 38X (all cores) = 71.44 W, 60 A, 1.191 V

Intel Processor diagnostic Tool, 'AVX test', No throttling 38X (one core) = 26.19W, 22 A, 1.191 V

Intel P.D.T. 'Floating Point test', Throttling 35x (all cores) = 37.80 W, 34 A, 1.112V

If that's to be believed (and I'm guessing the real total power should be a bit higher), there's a throttling case of one-core at only 24.46 W total, yet a Non-throttling all-cores case with higher total power of 71.44 W. The total power rating is 140 W for this CPU.

I found a program called 'Limit reason' that says the throttling (when it's happening) is caused by 'CORE P1', although it doesn't explain what that means.

As far as the Intel AVX test not showing throttling goes, it sounds like this type of Xeon throttling happens when using AVX and a larger '256-bit' register. So, maybe their AVX test doesn't use a 256-bit register?

If this is how they are, it's very disappointing to see that their Xeon parts apparently run slower than the equivalent I7 parts (I7-6850 in this case?)

I'll try HWINFO next to see what it says.

Thanks;

P.S.: CPU 'Package temp.' never gets above 62 C, and the cores all report less than that.
 
I' sure I've read somewhere that AVX clocks are lower than none AVX clocks due to the extra heat generated
 
MysteryGuy said:
If this is how they are, it's very disappointing to see that their Xeon parts apparently run slower than the equivalent I7 parts (I7-6850 in this case?)
Click to expand...

The short explanation is consumers are hysterical about it while the enterprise understands. Its very silly in reality for the consumer versions to run full speed but people would whine if say a 6700K ran 3Ghz with AVX2 instead of 4Ghz. Despite giving 50% higher output at 3Ghz. Then a 6700K could have been close to, if not a 65W chip.
 
Has the original OP made any more progress with this, just out of interest? I have almost the same system and while this doesn't bother me, you might find that the latest BIOS update does refer to the Turbo feature, also (and you probably know this), but so far as the X99-E WS is concerned, there is a Turbo Ver 3.0 download on the Asus website. Not sure if that helps, but I thought some positive feedback might be more functional.
 
Bikerchris said:
Has the original OP made any more progress with this, just out of interest? I have almost the same system and while this doesn't bother me, you might find that the latest BIOS update does refer to the Turbo feature, also (and you probably know this), but so far as the X99-E WS is concerned, there is a Turbo Ver 3.0 download on the Asus website. Not sure if that helps, but I thought some positive feedback might be more functional.
Click to expand...

I eventually decided this is probably just how they are. Thanks for the heads-up on the new BIOS. I'll give it a whirl, but I'm not expecting it to change much (as far as the throttling I see with some AVX programs).

I've been using the Intel 'Turbo Boost Max Technology 3.0 application' 1.0.1.9.

I don't know if it means anything, but I noticed that the 3502 BIOS description at the ASUS website for the plain X99-E WS mentions

"Support Intel Turbo Boost Max Technology 3.0"

while the 3502 description for the USB 3.1 version leaves that line out.

I'm not sure what

'update windows update HWID to 9'

means though...
 
MysteryGuy said:
I eventually decided this is probably just how they are. Thanks for the heads-up on the new BIOS. I'll give it a whirl, but I'm not expecting it to change much (as far as the throttling I see with some AVX programs).

I've been using the Intel 'Turbo Boost Max Technology 3.0 application' 1.0.1.9.

I don't know if it means anything, but I noticed that the 3502 BIOS description at the ASUS website for the plain X99-E WS mentions

"Support Intel Turbo Boost Max Technology 3.0"

while the 3502 description for the USB 3.1 version leaves that line out.

I'm not sure what

'update windows update HWID to 9'

means though...
Click to expand...

Thanks for getting back and you're very welcome for the BIOS info. That's a real shame about the BIOS update 3502 (which I'm certain I've applied) not mentioning Turbo related improvements for your USB 3.1. version. I have got Prime-95 so I might see how mine performs and get back to you.

What is a shame is the lack of reviews for this CPU, although I did see THIS ONE which (when translated) mentioned that the 4GHz turbo is rarely achieved. Interesting and for some safe fun, I used the automatic overclocking features of the motherboard and the BIOS claimed a Turbo of 4.2GHz (despite this CPU being locked). But when booted into an OS Win 10 environment, I don't recall ever seeing an improvement, using CPU-Z for monitoring.
 
You must log in or register to reply here.
Back
Top