Very low floating point math score.

[David (PassMark)]

tho i did try ryzen masteronce

One of the suggested changes when using the Ryzen Master application was to permanently change the clock source in Windows to be HPET. So even if you uninstall Ryzen Master, then it won't effect the clock source. I don't know if you did this or not.

[kerberos_20]

well the thing is that i didnt overclock it, cpu jumps up on its own to 4.1ghz, tho i did try ryzen masteronce , but since it was messing up with vddr voltages a bit than it got deleted

[David (PassMark)]

AMD is giving out conflicting information. They are saying HPET should be off for performance reasons, but they are also saying it should be on when using the AMD Ryzen Master application which is used for overclocking to increase the performance. Which of course makes no sense.

The way of measuring time on a PC has changed over the years. From oldest to newest the various hardware is,
RTC, Real Time Clock
PIT, Programmable Interval Time
PMT, Power Management Timer
HPET, High Precision Event Timer
TSC, Time Stamp Counter

You can tell the frequency of the clock (and this which one is in use) by using the Windows QueryPerformanceFrequency() call,

Anyway all new systems should be using the TSC. So even if the HPET is enabled in BIOS, it shouldn't be used. So from that point of view it makes sense to turn it off.
BUT the instructions from AMD for using the AMD Ryzen Master application tell you to turn it on in BIOS, then reconfigure Windows to use it. Why I don't know as this make little sense.

Here is what I suspect is happening. When you overclock Ryzen inside of the O/S then the TSC clock also get effected (sounds like a bug in the CPU to me, but anyway). Meaning the clock is no longer accurate and benchmark frame rates won't be accurate either. So to avoid this problem AMD suggests you force the use of HPET when overclocking inside of the O/S. The HPET timer remains accurate regardless of CPU clock speeds. This isn't a problem I suspect if you overclock from BIOS.

An inaccurate clock might also cause the video stuttering you saw.

So by forcing HPET to be on you get an accurate clock back, but there is extra interrupt processing overhead required when you are using HPET.
But if you aren't overclocking then you can leave HPET off I would think.

[kerberos_20]

hi, ive recently purchased fatality mobo with ryzen cpu...
first few days was everything okay....but one day my pc got sluggish....playing youtube producing sound stutterin...fps drops in games...etc
syntetic benchmarks were lookin all fine...except pass mark cpu floating point test...that one got crappy low...from 11.5k point (when it was new) to ~4.5k
after a bit googlin i found nothing usefulll...so ive tryied reinstall OS...no success..
then i reseted CMOS and everything got back to normal...
so i did change bios settings one by one until ive got to ACPI HPET table...
amd kinda says:
Make sure the system has Windows High Precision Event Timer (HPET) disabled. HPET increases the polling resolution of the system’s timer for certain performance monitoring utilities and the increased poll rate can compromise everyday application performance. HPET can often be disabled directly in the BIOS.
...
but in my case disabled HPET hurts my PC..
so im wondering...amd failxD

[dododa]

In my case yes. New win install, everything works as it should. Still trying to figure what was the problem in the first place. I've access to the troubled 8.1 installation, So I've just been fooling around trying to find the cause. Went through the 7700 thread looking for examples, and I can't find anything.

MB is Asrock, and in the 8.1 installation is had some asrock autotuning thing installed. Uninstalled the thing, Speed step was on, also tried with it off, In both cases I'm still getting low Floating Point Math. It came at 2580 in the last test, it's going up/down by some 50 points. The other win installation is at 9000 (+-100). I'm thinking there's also an option that it's got something to do with some autotuning residue from the old MB/CPU combo that's still embedded in the system. But again not certain. It's curious that in 7700 case user also had similar increase when he fixed it. Roughly 3x.

[LocalGamer]

Was this problem solved ?

[David (PassMark)]

Cross reference to another post that appears to have a similar issue, but with the i7-7700 CPU.


So as a summary, this seemed to help:
- Turning on Intel Speed step in BIOS
- Uninstalling Gigabyte bloatware (e.g EasyTuningService)
- Full reinstall of Windows if above don't work.

It still doesn't fully get to the bottom of the problem, but at least there are some things to try.

Update: It was also determined that Dual Intelligent Processors 5 software (under ASUS AI suite) causes a problem for AMD ThreadRipper CPUs.

[David (PassMark)]

So maybe a device driver problem, or some 3rd party piece of software messing things up.

[dododa]

Stumbled on this page while troubleshooting a gaming comp. I had similar case, and it helped me somewhat. Here's what I did in case it helps someone.

Win 8.1. 6600k, overclocked to 4.2ghz, no problems in most cases, heavy stuttering in few games. It would go from 40-60fps to single digits in cases where physics was concerned. I had some suspictions it was cpu's fault. Ran passmark, in most cases it was on par with similar systems, but in floating point math, it was very low. Seme results no matter what clock speed I tried. Can't remmember correctly, but it was in 2000 range. Anway, tried it on completely new Win install to rid of potential driver and registry problems. Win 10 on spare SSD (that already had some problems of it's own), and problem was immidately gone. Only tried with stock 3.9ghz. Games without stuttering, passmark score on floating point math: 8837. No idea what exactly was the fault, malware, no win reinstall on new cpu, overclock, drivers... But the problem was gone. Mind also that I'm not completely certain that it was cpu, it may also had something to do with ssd disk that 8.1 were installed on.

[David (PassMark)]

I didn't know what the PCU was. PCU = Power Control Unit or Package Control Unit, depending on who you believe. It is a new module inside Intel Skylake CPUs.

The best explanation I found was from anandtech.com.
"The PCU is essentially a microcontroller that monitors and computes the power requests and consumption portfolio of the separate silicon areas, providing information that can allow parts of the CPU to be power gated, duty cycled, or adjust in frequency and voltage."

I also read some that on some motherboards you can disable the PCU, but the side effects are pretty bad. Like the loss of the internal iGPU, the loss of all power control (the CPU never idles as it should) and lower speed from certain instruction mixes. In particular AVX instructions running slower.

The bit about duty cycled and frequency adjustment is interesting. So it is plausible this PCU module is now damaged and doing some frequency adjustments that it probably shouldn't be. Going to had to confirm however without swapping out the CPU for a new one.

[iSilver]

I have heard few heavily overclocked skylake cpus having problem with AVX and floating point calculation performance coz of its PCU going bad. I'm not sure about the details though. Those were after overclocking competition.

[David (PassMark)]

So even running 1 core is slow.
Still looks like throttling, but that doesn't explain why only the floating point result is effected.
So a bit of a puzzle.

Other floating point benchmarks are,
The Classic Whetstone
http://www.roylongbottom.org.uk/whetstone.htm

Spec CPU2006
https://www.spec.org/cpu2006/CFP2006/

But they aren't friendly bits of software to download & run.

SANDRA might also have floating point tests. Beware however some of these might use different CPU instructions (SIMD SSE in-particular) so the same circuits in the CPU might not be used as in our test.

[Falcy]

.
This chart was from a 4 core i7-4770. So you see near linear scaling to 4 processes. Then a levelling off and even a slight drop in throughput as more load is added. Which is slightly surprising as hyperthreading should have kicked in and gave some slight improvement from 4 to 8 processes. Hyperthreading always was a bit hit and miss however.

Maybe looking at the scaling on your machine would give some insight into the issue.

1 core: ~1330
2 core: ~2280
3 core: ~2430
4 core: ~2480

[David (PassMark)]

Posting that code snippet above made me thing of something else to try as an experiment.

In the Preferences window change the number of processes to 1, then 2, then 3... etc... and record the result each time.

You should see a progression like this.
.

Floating point test, number of processes	Benchmark result, Operations/Sec
1	2276
2	4342
3	6354
4	7545
5	7847
6	7761
7	7728
8	7382
9	7417

.
This chart was from a 4 core i7-4770. So you see near linear scaling to 4 processes. Then a levelling off and even a slight drop in throughput as more load is added. Which is slightly surprising as hyperthreading should have kicked in and gave some slight improvement from 4 to 8 processes. Hyperthreading always was a bit hit and miss however.

Maybe looking at the scaling on your machine would give some insight into the issue.

[David (PassMark)]

As the benchmark name of the test implies, floating point maths is being done.

The test is really simple & the code hasn't changed for years.

Simplified, it looks like this,

float *testdata;
double *pdtestdata;

while (1) {
val1_f = testdata[ind];
val1_d = pdtestdata[ind++];
val2_f = testdata[ind];
val2_d = pdtestdata[ind++];

j_f=val1_f + val2_f; // Line 1
j_f=val1_f - val2_f;
j_f=val1_f * val2_f;
j_f=val1_f + val2_f;
j_f=val1_f - val2_f;
j_f=val1_f * val2_f;
j_f=val1_f + val2_f;
j_f=val1_f - val2_f;
j_f=val1_f * val2_f;
j_f=val1_f / val2_f; // Line 10
<snip>

//64-bit Double math added in PT6.0
j_d=val1_d + val2_d; // Line 1
j_d=val1_d - val2_d;
j_d=val1_d * val2_d;
j_d=val1_d + val2_d;
j_d=val1_d - val2_d;
j_d=val1_d * val2_d;
j_d=val1_d + val2_d;
j_d=val1_d - val2_d;
j_d=val1_d * val2_d;
j_d=val1_d / val2_d; // Line 10
<snip>

//If time expired quit loop and calculate operations per second
<snip>

} //loop end

This loop is threaded and runs on all available cores (or the number specified by the user in the preferences window).