(UPDATE 03/09) Windows 8 Results (Temporarily) Not Accepted At HWBOT - Benchmark Result Veracity Compromised

  • News, benchmarks, Editorials, Articles
  • 124
  • HWBOT

(UPDATE 03/09) Windows 8 Results (Temporarily) Not Accepted At HWBOT – Benchmark Result Veracity Compromised

Author: Pieter-Jan Plaisier

As the result of weekend-time research, the HWBOT staff has decided to invalidate all benchmark records established with the Windows8 operating system. Due to severe validity problems with the Windows8 real time clock (“RTC”), benchmarks results achieved with Windows8 cannot be trusted. The main problem lies with the RTC being affected when over- or underclocking under the operating system. The operating system uses the RTC as reference clock, and benchmarks use it to reference (benchmark) time.

Update 03/09 – Revised Moderation Guidelines Concerning Windows 8, Futuremark Benchmarks Allowed (with new SI)

Following our recent news item regarding Futuremark’s solution for the Windows 8 RTC bug, we have updated our HWBOT moderation guidelines. They now include exceptions for Futuremark benchmarks with ORB link and System Info 4.20. The updated rules:

  • All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
  • Existing Windows 8 based benchmark submissions will be blocked if seemingly out of line or world record or top score.
  • Existing Windows 8 based benchmark submissions will not be blocked or removed if they don’t meet the requirements above
  • Existing Windows 8 based benchmark submissions will not be blocked or removed if they were made on an AMD system (New ones will be blocked)

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

  • non-overclockable notebooks may be used in combination with Windows 8
  • 3DMark and PCMark runs on Windows 8 if an ORB Link is included, and SystemInfo 4.20 (unofficial download) is used.

Update 24/08 – Revised Moderation Guidelines Concerning Windows 8

Following the findings of Ocaholic.ch (reference), the HWBOT moderating team has revised its rules regarding the usage of Windows 8 for benchmark submission at HWBOT. The major change is that the combination of AMD and Windows 8 is also not allowed, as in contrary to previous findings the platform is affected too. The revised rules can be found below.

  • 1) All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
  • 2) Existing Windows 8 based benchmark submissions will be blocked if seemingly out of line, world record, or top score.
  • 3) Existing Windows 8 based benchmark submissions will not be blocked or removed if they don’t meet the requirements above

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

  • 1) non-overclockable notebooks may be used in combination with Windows 8

Update 21/08 – Updated Moderation Guidelines Concerning Windows 8

Following up on the recent developments on the topic of Windows 8 RTC, read the updated news item here, Head of Moderating Christian Ney has updated the moderation rules regarding the usage of Windows8. Here is a list of the updated rules:

  • 1) All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
  • 2) Existing Windows 8 based benchmark submissions will be blocked if
    • seemingly out of line
    • world record, or top score
  • 3) Existing Windows 8 based benchmark submissions will not be blocked or removed if they don’t meet the requirements above

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove the result.

Exceptions to the Windows 8 rule:

  • 1) AMD-based systems may be used in combination with Windows 8
  • 2) We are still debating on how to address non-overclockable notebooks

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.

Thank you for your understanding.

The HWBOT Staff.

Update 20/08 – Windows 8.1 Affected, AMD Problem-Free(?)

We have continued our testing on more platforms. Windows 8.1 is, just like its older brother, affected by at-runtime base clock frequency adjustments. Intel platforms back to LGA775 are affected. From the looks of it, AMD systems are unaffected. 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. Later today Christian Ney, Head Moderator, will publish an update to the rules at HWBOT concerning the usage of Windows 8. Stay tuned!

The following combinations of platforms have been tested

Operating SystemPlatformRTC Bias / Clock Drifting
Windows 8.1HaswellYES
Windows 8HaswellYES
Windows 7Haswellno
Windows 8Ivy BridgeYES
Windows 8Sandy BridgeYES
Windows 8GulftownYES
Windows 8WolfdaleYES
Windows 8Trinityno
Windows 8Piledriverno

How does it work and how to spot it?

  • Boot in OS at 130MHz BCLK and 32x CPU Ratio
  • Reduce BCLK frequency to 122MHz at runtime (in OS). Set CPU ratio at 34x to maintain the same frequency of 4160 MHz
  • Check benchmark before and after the downclock

You can also easily spot the RTC problem using Windows Clock (see video below)

More information

Background Information – Remember Heaven?

At the moment of writing, we do not have the full technical what’s and how’s figured out. Since this problem affects everyone who is passionate about overclocking, it is important to provide an explanation. It is far from the complete story, but it should be enough for you to understand why we have decided to ban Windows 8 from HWBOT.

Do you remember the history of Unigine Heaven at HWBOT? About three years ago, we launched a wrapper for the back-then brand new DX11 benchmark software Unigine Heaven. The wrapper featured an easy and secure way to submit benchmark scores to HWBOT via data files. On November 2, 2010 we posted a response to an on-going discussion about downclocking in Windows affecting the benchmark score. To make a very long story short, by downclocking in Windows the Heaven benchmark time runs slower than it really is. One second in the Heaven benchmark is equal to 1,x second of real time. Because there is more time within a second, the system can render more frames. The benchmark itself is unaware of all this – for the benchmark, one second is still exactly one second. In the end, the system renders a higher amount of frames in a longer timeframe. In the result calculation for frames per second (“FPS”), the “frames” have increased but the “second” remains the same. You get a higher score.

Later that week, we published a V1.03b version of the wrapper, which fixed the downclock issue. Without going too much into detail, the wrapper uses a second source to verify the benchmark duration and takes the relative measured time difference into account when calculating the final score. Problem solved.

Windows8 – “Support for all devices”

As you know, Microsoft is trying to come up with a unified operating system and user interface for a wide range of devices, including tablets, smartphones, Xbox One, and the desktop PC. Building this unified platform is not easy. It is not just a matter of creating an interface that can be used with a multitude of input devices (finger, mouse, controller), but it also needs to support as many devices as possible. Getting the software to run out of the box on as much hardware as possible is the challenge they are facing.

Of the many aspects to fine-tune, one feature in particular is causing Windows8 to be practically useless for (competitive) overclocking: the RTC. Quoting Wikipedia, “A real-time clock (RTC) is a computer clock (most often in the form of an integrated circuit) that keeps track of the current time. Although the term often refers to the devices in personal computers, servers and embedded systems, RTCs are present in almost any electronic device which needs to keep accurate time.” Sparing you the details of development process, compared to Windows7 and previous versions, Microsoft made changes to how it measures time to be compatible with embedded or low cost PCs that do not have a fixed RTC clock. After all, having a fixed RTC clock adds cost to a platform.

Your PC system uses the RTC for many things. For example, it ensures the Windows Time on your machine is accurate. For most benchmark applications, the RTC is used as reference clock when executing the benchmark code. By synchronizing with the RTC, the benchmark knows exactly how much time has passed, and takes that value into account when calculating the performance of your system.

It’s all relative (part2)

In a previous editorial (“Critical thinking – Should Maximus V Extreme be …”), we already touched the topic of the relativity of time in our overclocking universe. “The concept of ‘time’ on a PC configuration is, if not synced via network or internet, an arbitrarily defined constant designed to ensure that the configuration is running in sync with the real world. In other words: hardware and software engineers ensure that ‘one second’ on your PC equals ‘one second’ in real time. One of the reasons why it’s so important to have the PC’s timer line up with the real world time is to ensure that your PC can produce accurate measurements and predictions.” The points we brought up in that editorial are relevant again. To ensure that the arbitrarily defined constant of ‘time’ is the same on everyone’s benchmark system, we rely on the OS and hardware. This worked quite well, until Windows8 came around.

The problem builds on the problems we faced with Heaven. When downclocking the system under Windows8, the Windows RTC is affected as well. The biggest difference between Windows7 and Windows8 is that now all benchmarks (no exception) are affected.

Examples: Benchmarks and Windows Time.

Let us make this more practical. On our Haswell test system we downclocked the BCLK frequency by about 6% from 130 MHz to 122MHz. Using a CPU ratio of respectively 32x and 34x, the resulting CPU frequency remains 4160MHz. Then we ran comparison benchmarks. Here are a couple examples:

BenchmarkResult at defaultResult after underclockDifference
Wprime32M5.6055.235+ 7.07%
Prime9513.62213.423+ 1.48%
Heaven DX111417.9821522.567+ 7.38%
SuperPI 32M7min 337min 10.799+ 5.10%
PiFast15.7814.88+ 6.05%
Aquamark3351,160371,902+ 5.91%
3DMark Ice Storm141,977150,526+ 6.02%
3DMark Cloud Gate2046221676+ 5.93%
3DMark Fire Strike47275012+ 6.03%

As you can see, even the latest 3DMark is affected by this. In Windows7, most benchmarks are unaffected by this kind of “overclocking”.

We can also demonstrate the Windows8 RTC problem with the Windows Time. Have a look at the videos below. In one of the videos, we underclocked the base clock frequency by 6%. After 5 minutes, Windows Time was already 18(!) seconds behind real time. When overclocking the base clock frequency, we can see the opposite effect. We overclock by roughly 4%, and after two minutes, Windows Time is 3 seconds ahead of real time.

Conclusive lines.

First, we would like to apologise if this editorial seems put together in a hurry. If it seems so, it is because it is. As we hope you gather from the data, we provided in the article, it is not possible for HWBOT to accept any benchmark results or records achieved using Windows8. Simply no benchmark – not even 3DMark – is unaffected by Microsoft’s RTC design decisions. As a result, it is impossible to verify the veracity of a system performance indicative in Windows8. The resulting score of any benchmark is relative to the RTC bias of that system.

In order to compare we need the RTC bias to be equal for all systems. With Windows8, we can only use question marks.

Practically speaking, it is not allowed to:

  • exploit the Windows8 RTC for benchmark purposes
  • submit any benchmark results with RTC bias

The HWBOT staff will:

  • Block any seemingly out-of-line Windows8-based benchmark results
  • Block any Windows8-based benchmark record, even if the score seems in line with the expectations

We hope that everyone is willing to adhere to these new guidelines. The HWBOT Staff is as passionate as you are about overclocking, and this kind of external problem is affecting our hobby in a way that makes we cannot easily resolve. We are friends amongst each other, and respect the effort your fellow overclockers put into overclocking results.

Thank you for your understanding.

The HWBOT Staff.


124

Belgium Massman says:

Since this is all very new to the staff, we are still looking into the consequences and possible solutions for this problem.

//edit: seems like Wordpress is broken due to traffic. I'll cross-post the article below.

How to recreate this issue

- Boot in OS at 130MHz BCLK
- Reduce BCLK frequency to 122MHz at runtime (in OS)
- Check benchmark before and after the downclock

Affected

- Windows 8 + Haswell = YES
- Windows 7 + Haswell = NO
- Windows 8 + Trinity = NO
- Windows 8 + LGA775 = YES
- Windows 8 + AM3 = NO
- Windows 8 + Ivy Bridge = YES
- Windows 8 + Sandy Bridge = YES





As the result of weekend-time research, the HWBOT staff has decided to invalidate all benchmark records established with the Windows8 operating system. Due to severe validity problems with the Windows8 real time clock ("RTC"), benchmarks results achieved with Windows8 cannot be trusted. The main problem lies with the RTC being affected when over- or underclocking under the operating system. The operating system uses the RTC as reference clock, and benchmarks use it to reference (benchmark) time.

Background Information - Remember Heaven?

At the moment of writing, we do not have the full technical what's and how's figured out. Since this problem affects everyone who is passionate about overclocking, it is important to provide an explanation. It is far from the complete story, but it should be enough for you to understand why we have decided to ban Windows 8 from HWBOT.

Do you remember the history of Unigine Heaven at HWBOT? About three years ago, we launched a wrapper for the back-then brand new DX11 benchmark software Unigine Heaven. The wrapper featured an easy and secure way to submit benchmark scores to HWBOT via data files. On November 2, 2010 we posted a response to an on-going discussion about downclocking in Windows affecting the benchmark score. To make a very long story short, by downclocking in Windows the Heaven benchmark time runs slower than it really is. One second in the Heaven benchmark is equal to 1,x second of real time. Because there is more time within a second, the system can render more frames. The benchmark itself is unaware of all this - for the benchmark, one second is still exactly one second. In the end, the system renders a higher amount of frames in a longer timeframe. In the result calculation for frames per second ("FPS"), the "frames" have increased but the "second" remains the same. You get a higher score.

Later that week, we published a V1.03b version of the wrapper, which fixed the downclock issue. Without going too much into detail, the wrapper uses a second source to verify the benchmark duration and takes the relative measured time difference into account when calculating the final score. Problem solved.

Windows8 - "Support for all devices"

As you know, Microsoft is trying to come up with a unified operating system and user interface for a wide range of devices, including tablets, smartphones, Xbox One, and the desktop PC. Building this unified platform is not easy. It is not just a matter of creating an interface that can be used with a multitude of input devices (finger, mouse, controller), but it also needs to support as many devices as possible. Getting the software to run out of the box on as much hardware as possible is the challenge they are facing.

Of the many aspects to fine-tune, one feature in particular is causing Windows8 to be practically useless for (competitive) overclocking: the RTC. Quoting Wikipedia, "A real-time clock (RTC) is a computer clock (most often in the form of an integrated circuit) that keeps track of the current time. Although the term often refers to the devices in personal computers, servers and embedded systems, RTCs are present in almost any electronic device which needs to keep accurate time." Sparing you the details of development process, compared to Windows7 and previous versions, Microsoft made changes to how it measures time to be compatible with embedded or low cost PCs that do not have a fixed RTC clock. After all, having a fixed RTC clock adds cost to a platform.

Your PC system uses the RTC for many things. For example, it ensures the Windows Time on your machine is accurate. For most benchmark applications, the RTC is used as reference clock when executing the benchmark code. By synchronizing with the RTC, the benchmark knows exactly how much time has passed, and takes that value into account when calculating the performance of your system.

It's all relative (part2)

In a previous editorial ("Critical thinking - Should Maximus V Extreme be ..."), we already touched the topic of the relativity of time in our overclocking universe. "The concept of ‘time’ on a PC configuration is, if not synced via network or internet, an arbitrarily defined constant designed to ensure that the configuration is running in sync with the real world. In other words: hardware and software engineers ensure that ‘one second’ on your PC equals ‘one second’ in real time. One of the reasons why it’s so important to have the PC’s timer line up with the real world time is to ensure that your PC can produce accurate measurements and predictions." The points we brought up in that editorial are relevant again. To ensure that the arbitrarily defined constant of 'time' is the same on everyone's benchmark system, we rely on the OS and hardware. This worked quite well, until Windows8 came around.

The problem builds on the problems we faced with Heaven. When downclocking the system under Windows8, the Windows RTC is affected as well. The biggest difference between Windows7 and Windows8 is that now all benchmarks (no exception) are affected.

Examples: Benchmarks and Windows Time.

Let us make this more practical. On our Haswell test system we downclocked the BCLK frequency by about 6% from 130 MHz to 122MHz. Using a CPU ratio of respectively 32x and 34x, the resulting CPU frequency remains 4160MHz. Then we ran comparison benchmarks. Here are a couple examples:




http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/7.14.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/122.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/130.jpg

http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/cg1.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/cg2.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/fs1.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/fs2.jpg

http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/141.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/141.jpg
http://news.hwbot.org/Articles/9824_BreakingWindows8Disallowed/150.jpg

As you can see, even the latest 3DMark is affected by this. In Windows7, most benchmarks are unaffected by this kind of "overclocking".

We can also demonstrate the Windows8 RTC problem with the Windows Time. Have a look at the videos below. In one of the videos, we underclocked the base clock frequency by 6%. After 5 minutes, Windows Time was already 18(!) seconds behind real time. When overclocking the base clock frequency, we can see the opposite effect. We overclock by roughly 4%, and after two minutes, Windows Time is 3 seconds ahead of real time.

http://www.youtube.com/embed/brkDKgvXKko
http://www.youtube.com/embed/XUi0zpRrtMM

Conclusive lines.

First, we would like to apologise if this editorial seems put together in a hurry. If it seems so, it is because it is. As we hope you gather from the data, we provided in the article, it is not possible for HWBOT to accept any benchmark results or records achieved using Windows8. Simply no benchmark - not even 3DMark - is unaffected by Microsoft's RTC design decisions. As a result, it is impossible to verify the veracity of a system performance indicative in Windows8. The resulting score of any benchmark is relative to the RTC bias of that system.

In order to compare we need the RTC bias to be equal for all systems. With Windows8, we can only use question marks.

Practically speaking, it is not allowed to:

- exploit the Windows8 RTC for benchmark purposes
- submit any benchmark results with RTC bias

The HWBOT staff will:

- Block any seemingly out-of-line Windows8-based benchmark results
- Block any Windows8-based benchmark record, even if the score seems in line with the expectations

We hope that everyone is willing to adhere to these new guidelines. The HWBOT Staff is as passionate as you are about overclocking, and this kind of external problem is affecting our hobby in a way that makes we cannot easily resolve. We are friends amongst each other, and respect the effort your fellow overclockers put into overclocking results.

Thank you for your understanding.

The HWBOT Staff.

Italy Atlas Rush says:

What if my Windows 8 is always connected to the Internet? What if I just overclock using multipliers, not modifying BCLK neither straps? Is there a way to check the RTC without the need of blocking all of the Win8 results?
I think it's a too drastic decision to block all of the results made with W8, it means that Windows 7 will be the last supported OS for Overclock, since Microsoft's path will keep working on modified versions of Windows 8, not on 7's ones. There should be a way to check if a score is "tricked" or not, maybe with a video, or something like that, like "If you wanna use Windows 8, don't use BCLK clocking/downclocking and post a video proving that you didn't modified clocks before taking the screenshot", it's a viable compromise, I think. Besides, top league clockers don't use W8 for WRs, they use tweaked versions of XP, 7 and S2003, since 8 is not really a bench system...

Belgium Massman says:

To clarify, we don't want to take the position of "everyone is guilty, until proven otherwise". The problems with the Windows8 RTC are of such nature that RTC bias (incorrect clock) can happen without the user knowing. The matter of the fact remains that any score with a biased timer is an false result. We are looking into ways to get Win8 back on the approved list. But given even Futuremark's 3DMark is affected, I think this will be quite a complicated issue. Thank you for your suggestions Ciruzzo!

United States Mr.Scott says:

Nice work.

Thanks for the info. :)

United Kingdom borandi says:

This obviously makes an issue for regular reviewers benchmarking in Win8 + pre-overclocked systems.

Italy Atlas Rush says:

Massman said: To clarify, we don't want to take the position of "everyone is guilty, until proven otherwise". The problems with the Windows8 RTC are of such nature that RTC bias (incorrect clock) can happen without the user knowing. The matter of the fact remains that any score with a biased timer is an false result. We are looking into ways to get Win8 back on the approved list. But given even Futuremark's 3DMark is affected, I think this will be quite a complicated issue. Thank you for your suggestions Ciruzzo!
You're welcome Pieter!
borandi said: This obviously makes an issue for regular reviewers benchmarking in Win8 + pre-overclocked systems.
That's exactly what's my point, since I'm one of them :\

United States Redmax says:

good work! Pretty sure this will cause issues throughout the industry, hardware review and otherwise.

Christian Ney says:

Saito said: Can't we patch windows 8 maybe? and remove RTC from win 8?


Tell Microsoft :D

South Africa Saito says:

Can't we patch windows 8 maybe? and remove RTC from win 8?

United States Fastvedub1 says:

Wow this is a bummer,I just started doing some benches with windows 8,As i have switched over to windows 8 on my machines....

United Kingdom borandi says:

Has this been tested across platforms/motherboards/BIOS revisions? It would be good to see a distribution of system setups that show these problems. I know you've said 'weekend research', but at this time all we are seeing are the results of one setup?

Netherlands neliz says:

Good riddance! :)

Now if only Microsoft just listened to the OC community and cancelled Win8, that'd be great.

South Africa Vivi says:

Damn now got to clock extra hard on win7 to get 100p back on firestrike :p. Ok so my win8 firestrike score will be removed right? Ill go disable the points

South Africa Vivi says:

Woops i cant edit my score because it was for a competiton, admins plz disable the points for it

Brazil ExPl00iT says:

Win7 Wins

Italy Atlas Rush says:

Neliz, a society like Microsoft really don't care about Overclocking. If you had a hundred millions copies-selling company, would you care for 100k people in the world? C'mon man, be real. It's not something SPECIFICALLY HURTING CLOCKERS, it's just something used to unify a system among different kind of devices, similarly to what Google did with Android for Smartphones, Tablet, Phablets and also Chromebooks (that can run Android OS on it).

United States Mr.Scott says:

Any idea on the time frame for the W8 sub removals?

CL3P20 says:

Vivi said: Damn now got to clock extra hard on win7 to get 100p back on firestrike :p.


**beat me to it :D

Germany Dancop says:

That's absolutely a mess...ban win8 means minus 400 points to me!
I won't buy my 3 titans back to achieve same quad and tri sli result!
Really...it would be ok, to ban future results, but not to ban all of them!
Especially the old ones...how shall I get back my scores? Is it my problem, that not everyone benches win8? No!
Even so, it's not the problem of others, that I sold my titans...BUT! it's impossible to some of us, to get the points back!!!

Poland ivanov says:

What about PRO OC Fire Strike Extreme stage? I already submitted one fine result and wanted to push it further :( As far as I know other overclockers also made submission on Win8 for this stage.

United Kingdom borandi says:

--> What about all the MOA results.

South Africa Vivi says:

borandi said: --> What about all the MOA results.


none of them look dodge so i dont think its a big issue, this is more about moving forward than looking back, hwbot only saw the issue now so they have to act on it.

i wondered why i wasnt getting an increase when i upped my BCLK lol

Russian Federation Antinomy says:

Massman, I'd like to ask about an exclusion for this rule - non-overclocked notebooks. There are users who benchmark notebooks obviously on stock settings (coz there's no overclocking given by the vendors). Since it's an overclocked issue, could stock settings that run on mobile devices only, be allowed? You know, Win8 is kind of preinstalled on these ones.

Greece varachio says:

I feel really lucky no having win8 results submitted so far...
Thanks for the info Hwbot staff...
I will keep a safe overclocking distance from Win8 until further notice...
Is there any hope to resolve easy this issue?
A patch for every single benchmark used within Hwbot maybe seems more feasible to me... huh?

Australia macsbeach98 says:

I feel sorry for the people that have submitted with Win8.
But it has to be a level playing field
The only real solution I could see would be a wrapper for every benchmark that references the real RTC clock at firmware level. Software can only divide or multiply it not actually change it.
Or just ban Win8 completely

Italy Atlas Rush says:

Wait, wait, maybe I got it wrong but...is it retroactive too? The post clearly says: The HWBOT staff will: Block any seemingly out-of-line Windows8-based benchmark results Block any Windows8-based benchmark record, even if the score seems in line with the expectations So only Records and out-of-line results will be blocked, right? Submitted scores that are not a record won't be blocked? Else I'll lose half of my points since I clocked under Win 8 with my old Ivy Bridge and with my Haswell, I'm a reviewer and the HW I test usually goes back to the brand/Press office, so I can't rebench everything I had on Win 7, I don't have anymore the 95% of the HW i benched, it will be a pain in the ass, and I don't think I'll be the only one to suffer from the rule being retroactive... So? Could I have better explanations about which scores will be removed? Also because I smell like people will hunt for clockers with same hardware but better scores and Windows 8, just to have an excuse to report and gain positions... Goddamn, that's ridiculous.

United States Mr.Scott says:

IMO, better to not pick and choose. It'll create animosity. All W8 subs should be blocked until further notice.
It is not any of the benchers faults, but it is unfortunate for them.

Italy Atlas Rush says:

Well, then also MOA scores and competitions scores MUST be removed, moreover they have a real reason to fake results, they got sponsored so being in the upper rankings means holding the sponsorship. Right? It's not any benchers' faults, but we are unfortunate.

Australia JJJC says:

The unfortunate thing here is previous subs on W8 will have a small efficiency advantage over 7 for certain benches, but that is something I could wear instead of calling for all 8 subs to be banned. I would be willing to bet the absolute vast majority of subs are inline with where they should be. For MOA I don't see any issue, none of the subs showed any signs of unusual scoring as far as I can tell. Just to confirm I can see why w8 has now been banned but I don't see the need to remove all previous subs.

Australia JJJC says:

The unfortunate thing here is previous subs on W8 will have a small efficiency advantage over 7 for certain benches, but that is something I could wear instead of calling for all 8 subs to be banned. I would be willing to bet the absolute vast majority of subs are inline with where they should be.

For MOA I don't see any issue, none of the subs showed any signs of unusual scoring as far as I can tell.

Just to confirm I can see why w8 has now been banned but I don't see the need to remove all previous subs.

Belgium Massman says:

We are still discussing the moderating consequences in the staff forum. This is what my suggestion was, I'm posting it here just so you have an idea of what I think about it. Ultimately, it's for the moderating staff to decide.

The main problem of this story for us, is that the benchmark veracity simply cannot be confirmed. It's not a story of black or white; it's not 0 or 1. All we can do with Windows8-based benchmark results is put a question mark next to it. Which puts us in a difficult position moderating-wise. Since the exploit is predominantly related to the uncommon practice of underclocking at runtine, here's my suggestion:

1) block or "no point" all Windows8 submissions that seem out of line
2) leave all old "correct looking" Windows8-based submissions as they are
3) we do not disqualify competition results (eg: moa, pro oc)
4) put out a request to the community to take down scores with Windows8 and underclocking
5) block / "no point" all new Win8 based results for now - especially in future competitions, no Windows8 will be allowed

Then we can gradually start adding exceptions to the Windows8 ban, although I don't know for sure if there will be any. The nature of the exploit is related to the RTC clock, not the underclocking itself. To put it differently, underclocking is a method we use to affect the RTC. There may be more methods like that out there. I'm hoping someone with better knowledge of the Window OS can uncover the technical problem.

I just want to repeat again that the issue we found on Sunday is not the fault of HWBOT or our userbase. There is no reason to take exhuberant measures against people who have run Windows8 in the past - it's very likely that this exploit has not affected too many results so far. We have trust in the community and know pretty much everyone here is on HWBOT because they love overclocking. We're all in this together, so we might as well be honest to each other.

What do you guys think?


By the way, I think it is very important to spread the word on this issue to as much people as possible. As a community we stand strong, and we should not allow people/companies claiming records or top scores on Windows8 systems.

Belgium Massman says:

borandi said: This obviously makes an issue for regular reviewers benchmarking in Win8 + pre-overclocked systems.


Antinomy said: Massman, I'd like to ask about an exclusion for this rule - non-overclocked notebooks. There are users who benchmark notebooks obviously on stock settings (coz there's no overclocking given by the vendors). Since it's an overclocked issue, could stock settings that run on mobile devices only, be allowed?
You know, Win8 is kind of preinstalled on these ones.


The Windows8 RTC bias only occurs when the base clock frequency is downclocked at runtime. That means pre-overclocked, or not overclocked, systems are not affected. As long as you run the benchmarks at boot frequency, the benchmark will run completely normal.

Allowing an exception for notebooks is a idea we can look into. A problem might be that nowadays high-end gaming notebooks can actually be overclocked as well. Even via BCLK frequency. As far as we know, all systems that all at runtime base clock frequency tuning are affected by this issue.

borandi said: Has this been tested across platforms/motherboards/BIOS revisions? It would be good to see a distribution of system setups that show these problems. I know you've said 'weekend research', but at this time all we are seeing are the results of one setup?


The test results in the article are from one system, but the problem description (RTC issue) is based on several sources and thus from several systems. We didn't test all different platforms, so compiling a list is not a bad idea.

We can start doing that right now. All help is appreciated! :)

Italy Atlas Rush says:

Massman said: We are still discussing the moderating consequences in the staff forum. This is what my suggestion was, I'm posting it here just so you have an idea of what I think about it. Ultimately, it's for the moderating staff to decide.
[Quote] etc etc



By the way, I think it is very important to spread the word on this issue to as much people as possible. As a community we stand strong, and we should not allow people/companies claiming records or top scores on Windows8 systems.[/QUOTE]

That seems the less "toxic" decision to get, it's the right compromise between avoiding future hackers and preserving previously submitted scores... And that follows the guidelines set by the evolution of rules added: it's valid since its introduction, not retroactive, it prevents cheating, falsification and things like that, probably the best decision, I hope this will be the picked decision.
Sorry for my shiiiiiiiiitty english but it's 6:10 A.M. here and I'm SOK'ing LOL

Australia Dinos22 says:

I would stop Win8 sub intake until there is a fix to be honest because you will have a tough time regulating this

Russian Federation Antinomy says:

Massman, I'm looking forward for a consensus on mobile devices. I'm aware of bus overclocking via XTU in notebooks. Just as I said, we can allow Win8 only for stock mobile devices. Overclocking -> Win7. Overclocking usually is about the pricey toys with i7 xtreme mobile and notebooks like Alienware (heard of MSI trying to do similar too).

United States remixedcat says:

Is Server 2012 still affected??? Does it do anything differently???

Argentina sebaz_ri says:

I wanted to upgrade from Vista to 8.1 but i think that i'll stay with 100% legal scores

United States cowgut says:

Wow just wow....Ok boys you know what you have to do.
I feel bad for the guys who will have to block there scores,the guys who were benching during rules change and for the mods who have to deal with the issue
Good luc to all

Russian Federation Antinomy says:

It's funny how history comes back from where it started. The NVIDIA NF2 jumping timer. Then Vista with it's new HPET. Then Maximus V Extreme. Now Windows issue again. Will clock games ever end? :)

Belgium Massman says:

Antinomy said: It's funny how history comes back from where it started. The NVIDIA NF2 jumping timer. Then Vista with it's new HPET. Then Maximus V Extreme. Now Windows issue again.
Will clock games ever end? :)


AMD HTT Clock Fluctuation ;)

United Kingdom borandi says:

There should be a big banner, as the article is a little unclear how the problem is produced.

This issue is caused by changing BCLK while in the OS not by BCLK changes in the BIOS

This is because the system makes an internal calculation based on the BCLK at boot time, which the OS uses. This value is kept even if the BCLK changes in the OS, causing this discrepancy.


---

I had to reread the main post several times to realise the extent of this issue. Making it clear to begin with (rather than adding tons of words describing the preface or the lead up) would help a lot. Start with the conclusion written in a very obvious way (like newspapers, scientific research journals) then describe the history and steps you used to reach the conclusion.

Belgium Massman says:

You're right, Ian. Added the following to the starting post. How to recreate this issue [list] Reduce BCLK frequency to 122MHz at runtime (in OS) Windows 8 + Haswell = YES Windows 8 + Trinity = NO [/list]

Belgium Massman says:

Trinity setup seems unaffected: SuperPI result is correct, and Windows Clock doesn't run slower. Anyone who can help out testing different platforms?

Iran stormycaspian says:

Is there any THING windows 8 can do correctly???

United States MetalRacer says:

borandi said: There should be a big banner, as the article is a little unclear how the problem is produced.

This issue is caused by changing BCLK while in the OS not by BCLK changes in the BIOS

This is because the system makes an internal calculation based on the BCLK at boot time, which the OS uses. This value is kept even if the BCLK changes in the OS, causing this discrepancy.


---

I had to reread the main post several times to realise the extent of this issue. Making it clear to begin with (rather than adding tons of words describing the preface or the lead up) would help a lot. Start with the conclusion written in a very obvious way (like newspapers, scientific research journals) then describe the history and steps you used to reach the conclusion.


Thank you, I was struggling to understand how this was happening.

I'll run some tests to see if this is happening with the X79 platform.

Canada Trouffman says:

Need to promote that widely, I feel sad fo rthe overclcoker that submitted scores on Win8... and i feel even more sad for the moderatrs that will have to deal with it :/

Anyway Win8 benchmarks submissions should be suspended until further notice :)

Not sure MS will fix it tho... but it's stupid how the Windows tim is even influenced xD

Netherlands rsnubje says:

Would be better if the benchmarks would have a file to refer to instead of windows's time, especially the ones from futuremark.

Germany deamon says:

But but but...Mocrosoft would NEVER fudge numbers to make their newest OS look better than older OSes so they could gin up more sales... microsoft is way too honest to do such a thing. Next, we'll be hearing some malarky that the NSA is spying on the citizens its suposed to protect.

Italy Atlas Rush says:

deamon said: But but but...Mocrosoft would NEVER fudge numbers to make their newest OS look better than older OSes so they could gin up more sales... microsoft is way too honest to do such a thing. Next, we'll be hearing some malarky that the NSA is spying on the citizens its suposed to protect.


Man, it's not funny, just because it doesn't make any sense at all. An unified architecture (and a RTC defined as Windows 8) is not made to pump up the sales, it's totally unrelated -.- that's for optimizing costs for an OS that has to work on several devices. And regarding the NSA fact, I wouldn't shout it that loud, you know, they're already here, too, in disguise.
lol.

United States Bobnova says:

Very odd Trinity doesn't do it. What about older stuff? Llano, IB, SB, 775, etc? I don't have win8, or I'd help check :D

India cyberwarfare says:

What if we dont use the RTC from win8 itself and make a bench sync with a server OTA. this way the the time used will be governed by the centralized server and not the bugged system

Christian Ney says:

Installed Windows 8 on a new drive and testing more platforms atm, so far:


Affected:
- Windows 8 + Haswell = YES
- Windows 7 + Haswell = NO
- Windows 8 + Trinity = NO
- Windows 8 + LGA775 = YES
- Windows 8 + AM3 = No
- Windows 8 + Ivy Bridge = YES
- Windows 8 + Sandy Bridge = YES

Norway knopflerbruce says:

So, for now it looks like AMD chips are not affected :p Intredasting.

Spain [DRACO] says:

Christian Ney said: Installed Windows 8 on a new drive and testing more platforms atm, so far:


Affected:
- Windows 8 + Haswell = YES
- Windows 7 + Haswell = NO
- Windows 8 + Trinity = NO
- Windows 8 + LGA775 = YES
- Windows 8 + AM3 = No
- Windows 8 + Ivy Bridge = YES
- Windows 8 + Sandy Bridge = YES


interesting

Then who use window 8 on the moa, the tube easier?

records with win 8 are not valid ... not?

Australia Dinos22 says:

so it's basically more of a Windows 8 + INTEL problem. AMD does not seem to have the same issue. Time for Intel to patch this up!

United Kingdom borandi says:

In case anyone is interested, here is Intel's RTC in Z87:

http://t.co/xbocjOYNqj

Pages 50 and 77

United States dreamss says:

this is why people should use the hardware hpet not the software dynamic clock.
to disable it: BCDEDIT /SET DISABLEDYNAMICTICK YES

dynamic clock is known to cause other issues like the incorrect DPC readings in latencymon and incorrect ratios in wintimertester
[img] http://img401.imageshack.us/img401/2678/dpctesting.png [/img]

Belgium Massman says:

I tested

BCDEDIT /SET DISABLEDYNAMICTICK YES
this weekend, but it didn't make any difference.

Singapore janwas says:

Benchmarks should access the HPET directly: http://wassenberg.dreamhosters.com/articles/timing_pitfalls.pdf Microsoft have consistently managed to screw up their timing APIs for YEARS. Perhaps the sleep-deprived microserfs are indeed incapable of something simple that works. I have recently updated this driver and signed it so it will work on current Windows.

Christian Ney says:

After intrnal discussion:

1) All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
2) Existing Windows 8 based benchmark submissions will be blocked if
- seemingly out of line
- world record, or top score
3) Existing Windows 8 based benchmark submissions will not be blocked or removed if they don't meet the requirements above

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

1) AMD-based systems may be used in combination with Windows 8
2) we are still debating on how to address non-overclockable notebooks

India thebanik says:

[QUOTE=Massman;260933]You're right, Ian. Added the following to the starting post.

How to recreate this issue
[list]
Reduce BCLK frequency to 122MHz at runtime (in OS)
Boot in OS at 130MHz BCLK x 30 multiplier
Check benchmark before and after the downclock

or do you want to say 122 x 30 = 3660 Mhz would still be better by 4-7% ?

My understanding is that 122 x 32 = 3904 Mhz (virtually same clocks) would be better by 4-7% and not clocks of 3660Mhz.

United States xxbassplayerxx says:

[QUOTE=thebanik;261236]Ok, this line is causing confusion among many non-overclockers, since this news seems to have gone big. [*]Boot in OS at 130MHz BCLK [B]x 30 multiplier[/B] [*]Reduce BCLK frequency to 122MHz at runtime (in OS) [B]x 32 multiplier[/B] [*]Check benchmark before and after the downclock or do you want to say 122 x 30 = 3660 Mhz would still be better by 4-7% ? My understanding is that 122 x 32 = 3904 Mhz (virtually same clocks) would be better by 4-7% and not clocks of 3660Mhz.[/QUOTE] 130 * 30 = 3900 122 * 32 = 3904 This was done so that the tests were run at the same speeds. The one with the lower BCLK scored significantly higher in all benchmark tests since the calculated vs actual time elapse was skewed to trick the OS and benchmark into thinking that more work had been done in the same amount of time.

India thebanik says:

xxbassplayerxx said: 130 * 30 = 3900
122 * 32 = 3904

This was done so that the tests were run at the same speeds. The one with the lower BCLK scored significantly higher in all benchmark tests since the calculated vs actual time elapse was skewed to trick the OS and benchmark into thinking that more work had been done in the same amount of time.


I was just giving an eg without finding the exact multiplier required to match the speed.

If you see main thread has been edited out after my post to correctly imply what they wanted to convey.

United States xxbassplayerxx says:

Ah, I didn't quite understand what you were saying at first. Apparently he didn't mention raising the multiplier at first and you were trying to get him to clarify that.

Brazil Rbuass says:

So... what about the scores approved in MOA qualifiers and other competitions? Some people used W8, and it can change the final results; If it will be removed, also, it can be a new issue.

South Africa Vivi says:

rbuass said: So... what about the scores approved in MOA qualifiers and other competitions?
Some people used W8, and it can change the final results;
If it will be removed, also, it can be a new issue.


MOA isn't effected by this i think, everyone already has their plane tickets being issued. results will probly get removed but still valid for comp :P

Brazil Rbuass says:

Vivi said: MOA isn't effected by this, everyone already has their plane tickets being issued. No need to make a scene about this now for no reason at all. No results are out of order

I asked because this:
"the HWBOT staff has decided to invalidate all benchmark records established with the Windows8 operating system"
That's nice to know nothing will change in the competition scores before :)

South Africa Vivi says:

rbuass said: I asked because this:
"the HWBOT staff has decided to invalidate [B]all{/B] benchmark records established with the Windows8 operating system"
That's nice to know nothing will change in the competition scores before :)


i want to go to moa so badly this year i will even rebench my titan in XP if they want hahah, see you there bro

Brazil Rbuass says:

Same here Vivi. See you there. ;)

Canada Vinster says:

Has anyone seen this information; http://www.ocaholic.ch/modules/smartsection/item.php?itemid=1071

Christian Ney says:

After further testing (Update 24/08 – Updated Moderation Guidelines Concerning Windows 8):

1) All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
2) Existing Windows 8 based benchmark submissions will be blocked if
- seemingly out of line
- world record, or top score
3) Existing Windows 8 based benchmark submissions will not be blocked or removed if they don't meet the requirements above
4) Existing Windows 8 based benchmark submissions will not be blocked or removed if they were made on an AMD system (New ones will be clocked)

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

1) non-overclockable notebooks may be used in combination with Windows 8

Russian Federation Antinomy says:

Thank you, CN and the rest of involved crew!

Exceptions to the Windows 8 rule: 1) non-overclockable notebooks may be used in combination with Windows 8
This was a definitely right decision! For all those crunchers who benchmark notebooks at stock.

South Africa Saito says:

will we be able to use AMD on WIN 8 since it is unafected

Christian Ney says:

No, AMD can be affected.

Christian Ney says:

After further testing (Update 03/09 – Updated Moderation Guidelines Concerning Windows 8):

1) All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
2) Existing Windows 8 based benchmark submissions will be blocked if
- seemingly out of line
- world record, or top score
3) Existing Windows 8 based benchmark submissions will not be blocked or removed if they don't meet the requirements above
4) Existing Windows 8 based benchmark submissions will not be blocked or removed if they were made on an AMD system (New ones will be clocked)


Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

1) non-overclockable notebooks may be used in combination with Windows 8
2) 3DMark and PCMark runs on Windows 8 if they meet the two following requirements:
- ORB Link
- Use of SystemInfo 4.20 or newer. download it here (not yet on FM's website)

Poland Xtreme Addict says:

okay, so FSE windows 8 here I come!

Belgium Massman says:

Updated article with your info, thanks for the major effort Christian!

Christian Ney says:

Following XTU Hotfix (Update 08/09 – Updated Moderation Guidelines Concerning Windows 8):

- All new Windows 8 Benchmark Results will be blocked, including for on-going competitions
- Existing Windows 8 based benchmark submissions will be blocked if seemingly out of line or world record or top score.
- Existing Windows 8 based benchmark submissions will not be blocked or removed if they don’t meet the requirements above
- Existing Windows 8 based benchmark submissions will not be blocked or removed if they were made on an AMD system (New ones will be blocked)

Note: if you recall downclocking when using an Intel based Windows 8 system, please remove that result.

Exceptions to the Windows 8 rule:

- non-overclockable notebooks may be used in combination with Windows 8
- DMark and PCMark runs on Windows 8 if an ORB Link is included, and SystemInfo 4.20 is used.
- XTU version v4.2.0.8 runs in Windows 8

United States PROBN4LYFE says:

Really?!... So you basically want to leave the top "clique" on the same OS that they basically overbear us all on....nice. And don't give me that BS about we can all use a stripped version of the same. I never thought as a "pure" overclocker to alter what I was using in front of me needed to be altered to give me a better score. It's fine with me guys really, I guess you need it...I know what I do real time not on a bench overnight...your world. I have beta tested Win 8 for over a few years...now it's no good for benches? Whatever. At subzero it does have false readings just like a Radeon does when it's clocked too far on air in Aquamark. Black screen...bench finishes with your "Oh Yeah!" face. Same with 3dmark03...so you pull the win8 scores because somebody is QQin again...get nuts fellas.
oh my bad the "King" spoke before me...now I'll earn my right.
Probn4lyfe(OCN)

Belgium Massman says:

Have you actually read the explanation?

United States PROBN4LYFE says:

Mass...reallly? Install OS>drivers>bench.
Plz tell me what's wrong with that?
I don't overdo overclocking Mass...no frills no thrills...I set the cooling in and see where I fold bro. I dont "alter" files or use software. I thought that was what defined me as a "real" Overclocker...maybe the game doesn't include us now...seems so.

United States PROBN4LYFE says:

Oh and send me cheap potatoes plz homie! lol Nah for real these guys get paid for it and I do it for fun. But you have to protect your rep.... I won't ever submit sub zero....and I have done it at home...but if you guys keep letting this stuff fly...why should I compete if I can't even have a field?

Belgium Massman says:

There is nothing wrong with you, your overclocking, or anyone else's overclocking. Windows 8 has a fundamental problem with the choice of timer, causing benchmarks to be invalid when the BCLK is downclocked at runtime.

United States PROBN4LYFE says:

borandi linked me some info...I see what you were talking about...I didn't take the time to get in depth with it :(

South Africa Vivi says:

lol what

Australia JJJC says:

Vivi said: lol what


Haha, this

United States Splave says:

8.1 allowed with xtu?

GENiEBEN says:

Splave said: 8.1 allowed with xtu?


If it's the latest XTU with the timer fix.

United States Splave says:

thanks benzino

Belgium Massman says:

V4.2.0.8 has the timer fixed, and is also the only allowed version.

Greece phil says:

Is systeminfo 4.22 allowed?

Belgium Massman says:

Lol, Microsoft never fixed it ^^

Puerto Rico chispy says:

I am sorry to bring back this topic up for discussion , but it is needed.

We need and updated rules on what benchmarks can be run on W8.0 or W8.1 and I mean 100% cristal clear , also we need hwbot new rules to be added to each benchmark rules page and update each benchmark rules page individually regarding W8 usage , I am still not sure what is allowed and what is not , it is not clear enough as it is , it is a big Grey area. Please we need the hwbot staff to make a new and updated rules + complete explanation of exceptions + a new statement regarding W8.0 and W8.1 situation. Please Pieter ,Genieben , Christian and all hwbot staff discuss here and lets continue to find a real solution for the problems with w8.0 and w8.1.

Norway knopflerbruce says:

Well, win 8.0 is not allowed in most cases. One exception was made for non-overclockable notebooks, but for people like you it's not allowed for any hardware you'll be running under normal circumstances. As for win 8.1 I don't really know. It all depends on RTC.

Puerto Rico chispy says:

knopflerbruce said: Well, win 8.0 is not allowed in most cases. One exception was made for non-overclockable notebooks, but for people like you it's not allowed for any hardware you'll be running under normal circumstances. As for win 8.1 I don't really know. It all depends on RTC.


Define people like me Knut ?

Norway knopflerbruce says:

People overclocking REAL hardware :D

Belgium Massman says:

[QUOTE=chispy;289945]I am sorry to bring back this topic up for discussion , but it is needed. We need and updated rules on what benchmarks can be run on W8.0 or W8.1 and I mean 100% cristal clear , also we need hwbot new rules to be added to each benchmark rules page and update each benchmark rules page individually regarding W8 usage , I am still not sure what is allowed and what is not , it is not clear enough as it is , it is a big Grey area. Please we need the hwbot staff to make a new and updated rules + complete explanation of exceptions + a new statement regarding W8.0 and W8.1 situation. Please Pieter ,Genieben , Christian and all hwbot staff discuss here and lets continue to find a real solution for the problems with w8.0 and w8.1.[/QUOTE] From the rules page: Exceptions to the Windows 8 rule: [list] [*]non-overclockable notebooks may be used in combination with Windows 8 [*]3DMark and PCMark runs on Windows 8 if an ORB Link is included, and SystemInfo 4.20 (unofficial download) is used. [*]Intel XTU from v4.2.0.8 and up[/list]

Puerto Rico chispy says:

[QUOTE=Massman;289998]From the rules page: Exceptions to the Windows 8 rule: [list] [*]non-overclockable notebooks may be used in combination with Windows 8 [*]3DMark and PCMark runs on Windows 8 if an ORB Link is included, and SystemInfo 4.20 (unofficial download) is used. [*]Intel XTU from v4.2.0.8 and up[/list][/QUOTE] what about catzilla , heaven and 2Ds like hwbot prime and cinebench R15 ? ids it allowed to use w8.1 ?

Belgium Massman says:

No, they are not on the exceptions list.

Puerto Rico chispy says:

Massman said: No, they are not on the exceptions list.


Thank you Pieter that's all i want it to know. People that ran recently benchmarks on w8.1 take notice and please follow the rules , rules apply for all and everybody , no exception on any person , man up and do the right thing with your non legit scores.

Australia SniperOZ says:

Could I please have some clarification on windows server 2012 ?

also is systeminfo 4.20 and above i.e 4.23 ok???... or not?

Belgium Massman says:

It applies to all Windows 8 based operating systems. That includes Server 2012. Systeminfo 4.20 and above are okay IF you provide a verification link and the verification link doesn't say the timer is off.

Australia SniperOZ says:

Thanks mate.... I was sure that was the case..... but helps to ask, cheers

United States Mr.Scott says:

chispy said: Thank you Pieter that's all i want it to know. People that ran recently benchmarks on w8.1 take notice and please follow the rules , rules apply for all and everybody , no exception on any person , man up and do the right thing with your non legit scores.

Seen quite a few on 8 and 8.1 lately. The list will be big.

Bulgaria liubutu16V says:

I just finished testing and I found out that this "bug" isn't here for me. AMD CPU, tried default settings, OC, UC, not even a second behind or ahead of my smartphone stopwatch. Win 8.1.
Edit: I can make videos and upload them on Youtube for you to see for youself.

United States [GF]Duke says:

Is it up to us the users to report the windows 8 and 8.1 scores that have the wrong sysinfo? I have seen so many lately. I don't want to seem like d..k for reporting everyone that I see. I just seems unfair to the rest of the community that people are putting scores in and getting points and cups for something not allowed.

Belgium Massman says:

Yes, please report them. The moderators can't follow all categories at the same time :).

GENiEBEN says:

'[GF]Duke said: Is it up to us the users to report the windows 8 and 8.1 scores that have the wrong sysinfo? I have seen so many lately. I don't want to seem like d..k for reporting everyone that I see. I just seems unfair to the rest of the community that people are putting scores in and getting points and cups for something not allowed.


If you don't like reporting just collect the links and PM me.

websmile says:

What I see lately is more and more 2d screenshots which do not give any hint on os used, because this is cut and you only see part of the desktop... - what about these? I mean it would be easy to use win8 there, programs like UC don´t give info on used OS iirc, shouldn´t the screenshot rules be updated for this reason or is this already "illegal" ?

GENiEBEN says:

websmile said: What I see lately is more and more 2d screenshots which do not give any hint on os used, because this is cut and you only see part of the desktop... - what about these? I mean it would be easy to use win8 there, programs like UC don´t give info on used OS iirc, shouldn´t the screenshot rules be updated for this reason or is this already "illegal" ?


Easy to spot Win8 from theme design.

Portugal aGeoM says:

Hello to all

[QUOTE=]Existing Windows 8 based benchmark submissions will not be blocked or removed if they were made on an AMD system (New ones will be blocked)[/QUOTE]

[QUOTE=]We have continued our testing on more platforms. Windows 8.1 is, just like its older brother, affected by at-runtime base clock frequency adjustments. Intel platforms back to LGA775 are affected. From the looks of it, AMD systems are unaffected. 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.[/QUOTE]
... not itself, but only when the OS resets on AMD platform...

I'm trying to understand why Windows 8/8.1 with AMD systems are blocked, if AMD is the cure ;) ...and I can't!
There is no need to point the blame to Windows 8, INTEL or perhaps the OC tool used, a fact its a fact, and in this case the fact is the AMD systems are not affected.

That's all, and be well...

Belgium Massman says:

The major change is that the combination of AMD and Windows 8 is also not allowed, as in contrary to previous findings the platform is affected too
The RTC can be enabled on AMD systems too

Portugal aGeoM says:

Massman said: The RTC can be enabled on AMD systems too


How, may I ask?

GENiEBEN says:

aGeoM said: How, may I ask?
bcdedit /set {current} useplatformclock No

Portugal aGeoM says:

GENiEBEN said: bcdedit /set {current} useplatformclock No


Thanks, I have one machine with win 8.1 SE, will try on that one.

Be well...

Kazakhstan TerraRaptor says:

Windows 2012 server allowed?

GENiEBEN says:

TerraRaptor said: Windows 2012 server allowed?


No.

United States boobteg4642 says:

What about windows 8.1? Because if this is the case, I don't think I'll be changing my entire operating system, again, to participate in your websites submissions anymore. Windows 8 is not new, in fact windows 9 is amongst talk. Either figure out how to run your stoneage benchmarks on newer computers or your going to lose a lot of members real soon. Unless your a die-hard Windows XP fan, then your just an idiot.

GENiEBEN says:

boobteg4642 said: What about windows 8.1? Because if this is the case, I don't think I'll be changing my entire operating system, again, to participate in your websites submissions anymore. Windows 8 is not new, in fact windows 9 is amongst talk. Either figure out how to run your stoneage benchmarks on newer computers or your going to lose a lot of members real soon. Unless your a die-hard Windows XP fan, then your just an idiot.


Any 8 based OS is ruled out ATM, that includes 8.0/8.1 and Server 2012/R2. I wouldn't hope for a fix coming to 9 either.
The problem is not related to 'stoneage' applications. Real members wouldn't bench on latest OS anyway.

EDIT: you're*

United Kingdom borandi says:

boobteg4642 said: What about windows 8.1? Because if this is the case, I don't think I'll be changing my entire operating system, again, to participate in your websites submissions anymore. Windows 8 is not new, in fact windows 9 is amongst talk. Either figure out how to run your stoneage benchmarks on newer computers or your going to lose a lot of members real soon. Unless your a die-hard Windows XP fan, then your just an idiot.


The problem is the operating system invalidating timings, so the benchmarks that rely on timing in the operating system are being skewed. Ask Microsoft to fix it, it's a problem on their end.

Also add to the fact that most overclockers here use fresh systems and change components on an open test bed, not their systems at home. That's how we get started. One disk drive for each OS perhaps, or different driver versions as required to get the best score.

It's kind of like asking Man United why their VIP boxes don't contain TVs watching the match. Everyone watches the match at home to get interested, but in a VIP box you can look out the window and see it live.

United States kjjweber says:

The Fire Strike Extreme world record says Operating system 64-bit Windows 8 (6.2.9200)
http://hwbot.org/submission/2587820_kingpin_3dmark___fire_strike_extreme_geforce_gtx_780_ti_8950_marks
http://www.3dmark.com/fs/2442579 7/21/2014, Am I reading this correctly, we can use win8 now ?

United Kingdom ObscureParadox says:

kjjweber said: The Fire Strike Extreme world record says Operating system 64-bit Windows 8 (6.2.9200)
http://hwbot.org/submission/2587820_kingpin_3dmark___fire_strike_extreme_geforce_gtx_780_ti_8950_marks
http://www.3dmark.com/fs/2442579 7/21/2014, Am I reading this correctly, we can use win8 now ?


If you use the latest system info for 3DMark then it's fine I believe.

Canada Trouffman says:

http://hwbot.org/news/9824_breaking_windows_8_benchmark_results_no_longer_accepted_at_hwbot/ With the latest system info its fine.

United States Redwoodz says:

We can see the difference when changing blck,but how do you know even at stock blck Intel is reading correctly? How would you know if even old OS with stock blck is reading correctly?Just realized the Northwood P4 on Vista32 I've been using seems to get ahead of time,even connected to the net constantly and time updating once a week.Just checked it and was 1.5 min fast. Ponderings...

Please log in or register to comment.

Leave a Reply: (BBCODE allowed: [B], [QUOTE], [I], [URL], [IMG],...)