knopflerbruce

I agree that the screens look weird, but he has also published cpuz validations which show that he has CPUs that OC well. Those validations can't be questioned, so... the mods should ask themselves if the CPU that validates at frequency x will be capable of benchmarking at frequency y. And the "setup" of his screens is quite simple so it MAY be easier to reproduce the exact same layout time after time than OP thinks. Btw, nice title. Should I call you cheater, too, if I find the smallest error in one or two submissions??

Memory plays a big role, so get a nice pair of sticks and mess around with the timings:p

I did try Server 2008 once (very briefly), and that OS was definately not that much faster more or less "out of the box". if all Vista scores had been that much faster it wouldn't have been a problem, as it would be obvious that the OS itself would be faster. But that's not the case, most Vista results are pretty normal - only like 20 scores are wicked fast. I'm not saying you should flag every single one of them, but it would be nice if someone would confirm that the tweak or whatever is legal.

http://www.hwbot.org/result.do?resultId=849451 http://www.hwbot.org/result.do?resultId=863329 RS690 is an AM2 mobo... can't really be this core. Ask him for a CPUZ validation of his 5000+ Toledo

Ahhhh, I thought the problem was the other way around - that you couldn't figure out where to post because there were too MANy categories:p In this case i agree with you, there are just too many differences, and a split-up is needed.

Do you think I run my benches on a 24/7 OS? I have 700-ish results, it's pretty obvious that I have some drives for benching. By all means, I'm no expert tweaker, but I know "the basic stuff", which should be enough to avoid being ~10% behind in the PP department. If someone posted a 6.00s run of spi1m at 6.6ghz wouldn't you want to find you how someone can get that kind of score at the given clock? I'm not saying anyone CHEATED here, but something is a bit weird. I've seen people freak out for 44k PP 1m runs, which is only ~2.25% off the normal "limit" at 45k. Here you have an irregularity which is 3-4 time greater, and you ask me to install a clean OS:o

It's easy to find out the correct category, all info to figure this out is visible in GPUZ. btw, I find only one 6200A category (but alot of other 6200 ones, but they don't match the list you posted:p)

Ticket ID: 403 Priority: Low The name of the CPU is a bit wrong, compared to the others so it\'s at the wrong place in the listing. Please move \"5400+\" in front of X2

Ticket ID: 402 Priority: Low These CPUs don\'t exist, all the X2 4000+ Toledo cores are mislabeled by wPrime, and the others belong to other categories (at least that was the case the last time I checked). The other two are empty and there is NO chance that AMD managed to create a 2.8GHz stock Manchester part.

Ticket ID: 401 Priority: Low These two are the same, please merge

Rejected doesn't really mean anything these days because of that issue. But it's not a real problem, as you WILL get a link with all the details. A REAL problem would be if the file wouldn't be uploaded at all:p

What's interesting is that only a small part of the vista scores get that kind of PP;) Someone should look into it IMO... I'm not a great fan of asking every bencher to reveal his tweaks, but this could very well be a bug. The difference is just too big. Agree? It's like someone starts posting 1m scores at sub-6.5s:p

http://www.hwbot.org/result.do?resultId=763110 1.55 XP http://www.hwbot.org/result.do?resultId=608872 online submission http://www.hwbot.org/result.do?resultId=652561 1.28 vista http://www.hwbot.org/result.do?resultId=712943 1.55 vista http://www.hwbot.org/result.do?resultId=723934 1.55 vista http://www.hwbot.org/result.do?resultId=752347 1.55 vista http://www.hwbot.org/result.do?resultId=654304 1.28 vista http://www.hwbot.org/result.do?resultId=748928 1.55 xp http://www.hwbot.org/result.do?resultId=807804 1.55 vista (?) http://www.hwbot.org/result.do?resultId=731455 1.55 xp http://www.hwbot.org/result.do?resultId=587019 online submission http://www.hwbot.org/result.do?resultId=816027 vista 1.55 To me it looks like some kind of bug/tweak that happens in vista;) Maybe the xp scores I listed here were done with xp skin in vista, I'm no vista user so I cant spot the small detail differences.

I don't have a clue... http://www.hwbot.org/hardware/processor/Opteron%20165%20.09 http://www.hwbot.org/hardware/processor/Athlon%2064%206400_%20X2 http://www.hwbot.org/hardware/processor/Opteron%20170%20.09 http://www.hwbot.org/hardware/processor/Athlon%2064%206000_%20X2 http://www.hwbot.org/hardware/processor/Athlon%2064%205000_%20X2%20Windsor http://www.hwbot.org/hardware/processor/Athlon%2064%204600_%20X2%20Windsor http://www.hwbot.org/hardware/processor/Athlon%2064%204400_%20X2%20Toledo http://www.hwbot.org/hardware/processor/Athlon%2064%203800_%20X2%20Windsor http://www.hwbot.org/hardware/processor/Athlon%2064%203800_%20X2%20Toledo There are some scores with WAY better performance products than the rest if you look in the top 3 of all these categories. I don't have a clue what that special tweak is, but it certainly makes it close to impossible to break their scores without it.

There is some special tweak that cuts off like 10% of the total score, too. Some use it... no clue what it is, though:rolleyes:

Make a sticky, massman;)

Nothing has been deleted, it's just a bug... it WILL be fixed!

I lost one golden cup, no idea about silver etc - which makes sense if this is a GPu score bug, since nearly all my scores are for K8 CPUs:D Being robbed for 1700 points would be a bit... OMG:D If it was the other way around.

You can still get the wprime scores;) It's impossible to beat them with a single socket, so there is a similar problem - but in the other "direction", too.

I think every global top20 score has to be approved by the mods before it's added to the ranking. Or they check the scores themselves. It's not that they all get reported at least:)

I've had a couple of nasty experiences with CPUZ lately... I use one CPU, and press F7, but when I upload it the link may lead to a different CPU. For instance the Athlon 64 2600 I asked Turrican to add. It WAS OK, but now it's linked to my old LE1660-validation. Since I have no chance to eliminate this bug by using a validation file, I'm asking for an exception to the validation rule, so I can upload the BMP file you get when you press F5, as that one can't be messed with;) Ticket ID for 2600: http://www.hwbot.org/forum/helpcenter.php?do=ticket&tid=257 Result: http://www.hwbot.org/result.do?resultId=866062 Edit: same happened to the 2650e-validation. Same solution as well, I uploaded my F5 screenie

Ticket ID: 380 Priority: Low http://valid.canardpc.com/show_oc.php?id=585791 - 200*11.5 stock... There are two other models available, 1354 and 1352 (11 and 10.5-multi), you may want to add those as well

The rules do look a bit strange to OC'ers, but WE don't decide who should be invited etc etc. However, if there is no qualification etc - just handpicking of benchers - it's not like it's a competition that decides who is the best overclocker out there or something;) As long as no-one claims that that's what it is, then it's fine by me:p Hopefully this "test round" will succeed so the REAL stuff can come to life later on.

This isn't really a CPUZ issue (I think), more a processor problem:)

I don't think it's a good idea. Even if "the gentleman way" would be to tell those who wonder what batch/stepping you've got, it shouldn't be mandatory. It's close to revealing all tweaks in a notepad document:p Plus, it's very time consuming. (make screenshot, take a shot of the CPU, edit it, paste it into the old screenshot).