HWBot: Windows 8 benchmark errors don't show up on AMD PCs

Visso's Windows 8 Tablet PC uses an AMD processor

A few days ago, the HWBot website, which collects PC benchmark scores, decided to ban all Windows 8 scores from its database. The site claims that the decision was due to a change in the real time clock (RTC) in Windows 8 that can cause benchmark programs to show inaccurate results if the CPU base clock (BCLK) frequency is changed in software.

At the time, HWBot tested this issue on a PC with an Intel Haswell processor. Today, the site has posted an update on the matter which says that the RTC issue does not appear when benchmarks are run on a PC with an AMD processor. The site says, "In fact, the clock drifting on Intel based systems seem to resolve itself when switching from an Intel based system, to AMD, and back to Intel."

As a result, the HWBot benchmark database will now accept Windows 8 benchmarks scores that are generated from an AMD PC. The site is still blocking all new Windows 8 benchmark results from Intel systems and existing Windows 8 benchmark scores will be blocked if they appear to be a new world record or top score or if they are "seemingly out of line."

Even though HWBot is accepting Windows 8 benchmark scores from AMD rigs, the site still feels the RTC problem lies with Windows 8 itself and not with Intel. The site states:

Windows 8 seems to be using a different source to fetch “real time” on an Intel based system compared to an AMD based system. It’s still guess why and which at the moment, but it seems to be related to the power saving features integrated in Windows 8.

Finally, the site also claims that Windows 8.1 suffers from the same RTC issue as Windows 8. We have emailed Microsoft for comment but have yet to receive a response. Futuremark, which makes the PCMark and 3DMark benchmarking programs, has already posted a response to the HWBot bans, saying, " ... the steps required to exploit this issue are quite unusual and could not be happened upon by accident."

Source: HWBot | Image via Visio

Report a problem with article
Previous Story

LG unveils insane 5.5-inch screen with Q-HD resolution

Next Story

The researcher who hacked Zuckerberg's page to get rewarded but not from Facebook

25 Comments

Commenting is disabled on this article.

Have they ever thought its an intel chipset driver issue and NOT a problem specific to windows 8 ? They seem soo fast to blame microsft for this.

This issue is discussed in today's update:

We also received a couple of emails debating the topic of which is to blame, Windows 8 or Intel CPUs. Based on the information we gathered from clock drift problem on various operating systems and hardware platforms, we are still of the opinion that Windows 8 is the root cause of the problem. Windows 8 seems to be using a different source to fetch “real time” on an Intel based system compared to an AMD based system. It's still guess why and which at the moment, but it seems to be related to the power saving features integrated in Windows 8. With Windows 7 unaffected, and Windows 8 affected on all Intel platforms (including old generations), the root of the problem lies with Windows 8. Having the Intel system unaffected when coming from and AMD Windows 8 installation indicates that the correct timers are available on Intel too, but unused by Windows 8.

Just because they cant find the cause, doesn't make it Windows 8's flaw.
I rather wait for a more technical word from Microsoft before pointing fingers. If they made any changes to the RTC clock for some reason, it is documented and probably has had a reason to do so.
That it took those guys a year (or more) to even find this, and then before even checking if the issue was there on AMD, they spouted into the world "WINDOWS 8 FAILS AT TIME".....
Not a very professional and technical approach from their end and now they just go like "well it works on AMD, so it's still Windows 8's fault".....

Shadowzz said,
Just because they cant find the cause, doesn't make it Windows 8's flaw.
I rather wait for a more technical word from Microsoft before pointing fingers. If they made any changes to the RTC clock for some reason, it is documented and probably has had a reason to do so.
That it took those guys a year (or more) to even find this, and then before even checking if the issue was there on AMD, they spouted into the world "WINDOWS 8 FAILS AT TIME".....
Not a very professional and technical approach from their end and now they just go like "well it works on AMD, so it's still Windows 8's fault".....
You're completely misrepresenting what they're saying and their arguments.

No he's not. They could have checked AMD hardware before going public with the claim that the problem relies solely on Windows 8.

Asik said,
You're completely misrepresenting what they're saying and their arguments.

I'm not defending 8 before I even know the cause, but I do think they've made themselves look extremely unprofessional here. If I liked their product at this point I'd be looking at their competition.

Randomevent said,
I'm not defending 8 before I even know the cause, but I do think they've made themselves look extremely unprofessional here. If I liked their product at this point I'd be looking at their competition.
Could you guys actually make sense for a minute? Exactly what claim (please provide the exact quote) did they make that was then proven wrong?

Asik said,
Could you guys actually make sense for a minute? Exactly what claim (please provide the exact quote) did they make that was then proven wrong?

I didn't say wrong, I said unprofessional. If I prefer 8 and am using their product they're making their product less functional to me even though I did nothing wrong. I could've cut off my leg to spite my ingrown toenail, but it seemed like a bad idea at the time. Futuremarks response on the other hand is much more what I expect from a software company.

Randomevent said,

I didn't say wrong, I said unprofessional. If I prefer 8 and am using their product they're making their product less functional to me even though I did nothing wrong. I could've cut off my leg to spite my ingrown toenail, but it seemed like a bad idea at the time. Futuremarks response on the other hand is much more what I expect from a software company.
Until there's a way to tell if a user modified his BCLK at run-time, there's no way to tell if a Windows 8 benchmark tells the truth (exception made of AMD). That makes it unreliable. HWBot is in the business of providing a meaningful database of benchmarks. How do you suggest they should act?

Of course Futuremark told its users to keep using its product; they're in the business of selling their benchmark tool.

Asik said,
HWBot is in the business of providing a meaningful database of benchmarks. How do you suggest they should act?

In a manner that doesn't discourage people from using their product. I figured I covered that already.

Randomevent said,

In a manner that doesn't discourage people from using their product. I figured I covered that already.
How? They can't accept unreliable results. They're just being honest with their users.

The site says, "In fact, the clock drifting on Intel based systems seem to resolve itself when switching from an Intel based system, to AMD, and back to Intel."

Does not compute. Someone want to clarify this for me?

adrynalyne said,

Does not compute. Someone want to clarify this for me?


Just a guess, but I think they are talking about moving a hard drive from an Intel box, to an AMD box, then back to the Intel box -- without reinstalling the OS. Maybe there is something that gets set in the registry.....

That is what it sounds like, but Windows usually doesn't play nicely with such things. How are they getting around the reinstall?

It seems to me for it to be an accurate test, they wouldn't be allowed to do a Windows reinstall/repair.

Actually. At least for me, I've seen successful mounts (besides video configuration) on a Gateway netbook and on an Acer 11" laptop. I know that in the end, they're both from the same manufacturer. Both have intel processors: one Atom and one Celeron...

Windows just popped the "Installing Device Driver Software" prompt and the computer ran incredibly well.

That is less surprising with it being the same OEM and CPU brand. I guess maybe they shopped around for motherboards using the same SATA controller and drivers...I dunno.

Unscientific test and faulty reasoning IMO.

adrynalyne said,
That is what it sounds like, but Windows usually doesn't play nicely with such things. How are they getting around the reinstall?

"Usually" the problem is in driver of sata/pata controller: Windows disables unused boot drivers for better start-up performance. So if you want to move system on motherboard with different chipset manufacturer or just change SATA mode from IDE to AHCI you need temporally enable all drivers, see http://support.microsoft.com/kb/922976/en if you want to know how.

Also in Windows 8 Enterprise there is ability to create portable version "Windows To Go" which is created for people who want to have Windows on USB flash drive and load multiple PC's from it. It has all drivers enabled by default.