Clock Bug Leads to False Ryzen Benchmark Scores

[Pachyderm]

RTC bug still prevalent for AMD Zen 2, which can alter the system's perception of time.

 

Tom's Hardware Article

 

Quote

Benchmark results for any AMD platform with an alterable base clock have been banned from HWBot, the organization that maintains the world’s largest competitive overclocking database.

Quote

RTC bug is a timer flaw that alters the time perceived by a benchmark as a result of over- or under-clocking the base/reference clock of the system. 

 

[cj09beira]

from what i remember this is because Microsoft wanted to be able to use hardware that didn't have an real time hardware circuit, they should have returned to the proper way with win 10

[porina]

From the linked article:

Quote

HWBOT had no other choice but to disable points for all Ryzen 3000 CPUs for all ‘unsafe’ benchmarks. like SuperPi, wPrime, Cinebench, Geekbench, and so on.

I still have my points... but note I did comply with hwbot requirements and used Windows 7 with Ryzen.

https://hwbot.org/submission/4195019_mackerel_cinebench___r15_ryzen_7_3700x_2238_cb

 

The limit of Win7 with Zen 2 CPUs is lack of CPU USB driver, but people have hacked the installer files to allow the old ones to install and it still seems to work.

 

More interesting is this:

Quote

Zronek explains that there are even more pressing issues with HPET: “The biggest problem of forcing HPET are the system-wide implications. I’ve talked about the System Clock Interrupt above, the periodical wakeups for the OS. Well, enabling HPET will switch the time source for the interrupt to HPET. So the kernel tick and every tick-dependent API function for time keeping will rely on HPET now. That’s a good thing, because it solves the problem, right?”

 

“Sadly, that’s only true for older CPU generations. Modern CPUs with high core counts take a serious performance hit when enabling HPET. Although it’s a very precise timer, it resides on the chipset and has slow access times. If it’s used for every timing operation in the system, it bottlenecks the CPU and will lower your FPS or even bring stuttering to the Windows UI. The situation is especially bad on Kaby Lake X and Skylake X, where I’ve detected this anomaly first. But Ryzen and Threadripper are negatively impacted as well. I’m calling it the ‘HPET bug’ (not very unique, I know) and I’ve written a full article that features a small benchmark app called “TimerBench” to test the impact of the timer configuration on your own system.”

Since I do participate on hwbot with my various systems, I tend to have HPET on. Actually, that's not entirely correct. In discussion elsewhere with the writer of y-cruncher benchmark, the requirement for hwbot is that the system platform clock is enabled. This could be one of many timers, of which HEPT is one of them.

[RejZoR]

Isn't even HPET considered buggy and not particularly useful? Wasn't TSC suppose to replace it?

[Arika]

Oh would you look at that, AMD isn't perfect. 

[Flying Sausages]

2 minutes ago, Arika S said:

Oh would you look at that, AMD isn't perfect. 

My wallet is still saying go with Ryzen over Intel.

[mr moose]

22 minutes ago, Arika S said:

Oh would you look at that, AMD isn't perfect. 

You you clearly know nothing about anything, AMD are omnipotent and if it wasn't for ryzen my cat would be dead and I wouldn't have a job.  All hail AMD,  from hear on AMD shall be known as the Almighty Man-sized Deity.  On your knees Intel heathen...

[Animal901]

On a tech elitist forum and you typed "you" twice straight off the bat.  How did you almost kill your cat over a chip? Those goofballs with the old apple towers still have jobs and it takes them 2 hours to transfer 40 gigs over usb 2. Lol...... 

[mr moose]

3 minutes ago, Animal901 said:

On a tech elitist forum and you typed "you" twice straight off the bat.

you you don't understand modern old time aussie lingo then.  

 

3 minutes ago, Animal901 said:

  How did you almost kill your cat over a chip?

I am pretty sure that had I not taken my case over to the work bench to replace the CPU (with the Ryzen) that it would have fallen over and squashed my cat.  I mean, that is a very highly likely outcome if i owned a cat and it happened to be under the case at the time.  therefore the Ryzen saved my cat. 

 

3 minutes ago, Animal901 said:

Those goofballs with the old apple towers still have jobs and it takes them 2 hours to transfer 40 gigs over usb 2. Lol...... 

I still own a G4 case which was used right up until a few months ago,  does that count?

[Animal901]

And there is the sole reason we cannot make it to 128bit. 

[williamcll]

5 hours ago, Animal901 said:

And there is the sole reason we cannot make it to 128bit. 

But the console did it and look where it lead them.

[Fnige]

5 hours ago, williamcll said:

But the console did it and look where it lead them.

to 128bit 

 

12 hours ago, mr moose said:

You you clearly know nothing about anything, AMD are omnipotent and if it wasn't for ryzen I would be dead and my cat wouldn't have a job.  All hail AMD,  from hear on AMD shall be known as the Almighty Man-sized Deity.  On your knees Intel heathen...

fify

Edited September 5, 2019 by SlimyPython
and myself

[ravenshrike]

So does the clock bug interfere with any actual programs that people use, and not merely benchmarking tools?

[RejZoR]

5 hours ago, ravenshrike said:

So does the clock bug interfere with any actual programs that people use, and not merely benchmarking tools?

Just benchmarks. Because benchmarks just rack up the work and compare it to time taken and churn out the score based on that ratio. This doesn't in any way change how long it takes for lets say Ryzen 3900X to compute a workload in actual time counted by your iPhone or a dedicated stopwatch. Bottom line, it doesn't affect actual, real performance and work done. It just affects what benchmarks display as end score. We already know Ryzen CPU's are great, this doesn't really change much for end users.