knopflerbruce

http://www.hwbot.org/result.do?resultId=685155 X2 4400+ Brisbane or Windsor in Toledo rank http://www.hwbot.org/result.do?resultId=686907 Athlon 3500+ in Sempron 3500+ rank

I know Bwanasoft makes some "test scores" sometimes, to try out new rankings or something. Nothing to worry about:)

Hmm, maybe your PCI lock is off, I guess that kind of problem could be because of that.

There will (probably) be a new category, and then the wrong scores will be moved to the right one.

I don't know it's cheating. All I know is thatg you think it is. Please, show me where it says it's not legal? I'm sure the GOZ users will stop using the driver if you can give us a link or something. What is so special with these drivers, btw? How can you get 15% extra by using those drivers? Uber tweaked??

Maybe I missed something, but where does it say that GOZ drivers are illegal? And you shouldn't call people cheaters all the time, let the mods be the judges. Would you like it if I write a post here like "gradus cheater [link]"? You post the links here, that's great. Then a mod will look at it and if necessary fix it. Making accusations like that is just plain stupid.

How long does it run, btw? Before we have a winner I mean:) (seems obvious that it's going to be a single card, but who knows)

Number of sockets is the same as number of pcie-slots for GPUs. 2 cores glued together to one CPU-pcb is the same as 2 GPUs "glued together" on one gpu-pcb IMO. I voted for 3870 x2 as a single card:)

This bug made a mess a while ago, but I think new results are not rounded like that anymore:)

http://www.hwbot.org/result.do?resultId=690195 This score was reported by a user. He thinks it's bugged or something, but I think it's ok. When I search the orb I find plenty of amd x2 / 7900gt setups that do about 30-36k in 3dmark03. Most of them have higher GPU clocks than me, but very few of those scores are with a CPU @ 2.8 and above, my chip was 2.9ghz. And I have 512mb video memory, the others have have 256mb. (the 512mb version is quite rare). Are the mods with me on this one? Edit: this COULD be an SLI run, I remember I had 2 of these cards once, but I'm sure I benched a little bit before I got the 2nd as well. If the score really IS way too high for a single 7900gt 512mb, then move it to the sli rank instead. This was just a 3dr file I found "by accident", and I really don't know much about how these cards compare to each other.

Ah, I felt a bit special when someone with a name like that sent me a PM, but now... But who dares to check out the link?

I don't OC high-end video cards, but this sounds a great idea to my ears anyway:D But are you talking about CORE frequency, MEMORY frequency, or the total sum of those?

Well, it's not 100% true, I have 300 points flat... Guess why? And I'm like... what? 130th? So atm 130th place is as high as you can get without a "big" wallet;)

I think this discussion has missed something important: the problem isn't that someone has like 4 spare qx9650 without any hwbot scores that he decides to let members of his team use for a session. What's importnant to avoid is that the SAME CPU is used several times within the same team. I'm sorry, but I don't see ANY reason why someone can't share a piece of HW that hasn't been used to make points here in the past. The ownership rule was made to avoid people sharing GOLDEN chips, not spare chips they won't use for hwbot anyway.

I think he means that his friend is not a HWBot member, or at least doesn't use that particular CPU to get points in 2d benchmarks. IMO that's OK, as that particular CPU is not getting points for, say, 2 spi1m runs by 2 people.

This mess could be sorted pretty easily simply by showing the mods pictures of the HW so they know that the HOT members in this case owns the hardware. Sure, it's not EVIDENCE, but how can anyone PROVE anything in a case like this? The "suspiciouness" of the scores would be reduced if people can see what was used during the session. Of course no-one can know if one user's cards were used for all scores, and some of the pictured items were not used at all. As long as there is no such thing as a gpu/cpu-S/N-reading feature I don't think anyone can ask for more evidence then pics of the parts. One can't video tape the whole event, that's crazy:) It's pretty obvious that during a bench session where many people post great scores, some might ask some questions about HW sharing. Taking a few photos during the session would solve everything. I think alot of people who bench together take pictures anyway, so it shouldn't be a big problem. I hope that's the case in the future. At least I will prove that I have access to 2 DFI Venus'es (of 1000 made) and a truckload of 939 CPUs when I go to the next bench party, as some people might want to try some of my things for different purposes.

I imagined this would be a problem, but you did something similar before: I can't get more points for my team unless I get an Intel rig, as my team only benefits from 300 of my 430-440 points already. Time to retire from HWBot?? I mean... I can't get more team points nor "personal points", only more cups and a higher hardware master rank. The member you talk about will be able to fight for more personal points, which I can't, but apart from that it's the same situation. Which means that I should have less motivation to continue benching here than a fresh dude, since my points are useless, except for a tiny ranking list. I know it's not a 100% bullet-proof solution, but this may be about finding the best solution, not the perfect one. And I find it a bit hard to believe that most people bench ONLY for their teams and not for themselves;)

My solution to the shared HW/team score problem: Only the best score for each CPU (or GPU)/benchmark combo will count as team points. Which means that if 20 people from OCTB post an e2180 score, only #1 will help the team.

I always get a little worried when I see that someone beats one of my scores, but knowing which score that was beaten isn't so easy. Is it possible to be notified when someone "steals" a cup?

That would be hard to do, because it's impossible to know exactly how much lower. I don't mind that CPUZ photos are "illegal", my problem was that it wasn't written anywhere. That problem will be taken care of thanks to 1Day I think:) As long as the rules are clear, and fair, then it's no problem. When it comes to superpi scores etc i hope a frozen screen is enough, if all CPUZ windows etc are visible.

Hehe, didn't think about that way of cheating as you just described. But, really, you should write it somewhere so people knows about this. (that photo's can't be used to validate CPUZ scores). If I had known about this "rule" I would've upped the voltage a little bit more just to get that damn validation. Thinking it didn't matter, I moved on to the next benchmark. How could I possibly know that this way of proving my score wasn't allowed? Can't see it mentioned anywhere...

Errrrr, HWBot is a place where people compete. No need to say sorry for breaking a few (or MANY) records;) It's not your problem that others can't match your scores.

I had a score blocked because a mod wanted a proper validation, and not a picture of my frozen desktop. Where does it say that a photo isn't good enough? If that rule doesn't exist, please unblock my score. Score: http://www.hwbot.org/result.do?resultId=671222 The CPUZ benchmark is (for us HWBot users at least) only a nice way of proving insanely high clocks. Which I did by using a camera. The score is there, it's no doubt about the fact that my Opty 185 ran at 3817mhz for a second or 2. Sure, it froze, but... like I said, it's no stability benchmark:) I've seen alot of people on forums (like xs) use photos to prove scores, why shouldn't we be allowed the same here? It's MUCh harder to fake a photo than a normal screenshot, you know.

What's so wrong here? Different clocks, different scores... There are quite a few members in that team, it's not so strange that 2 or more of these guys own a 7800gtx and a decent CPU each. It's not good news for anyone if 2 scores in the same hardware category by the same team is all it takes before people wonder if they cheat.

Something is a bit strange here: Opty 165 http://www.hwbot.org/result.do?resultId=624465 http://www.hwbot.org/result.do?resultId=618550 This is WAY faster than normal scores, wrong pifast version perhaps? 3200+ Manchester http://www.hwbot.org/result.do?resultId=616252 Looks like a wrong pifast version, or at least the REAL time is missing from the screenshot. 3500+ Manchester http://www.hwbot.org/result.do?resultId=593187 http://www.hwbot.org/result.do?resultId=654159 Same score, please delete one of them. FX55 San Diego http://www.hwbot.org/result.do?resultId=574370 Can't see the screenshot (not registered with the forum where the screenshot can be found), but the result seems a bit too good. 5 sec less, at 500mhz less than 2nd place? Errr.... FX57 http://www.hwbot.org/result.do?resultId=618538 http://www.hwbot.org/result.do?resultId=618545 Both results are just too good (again). X2 3800+ Manchester http://www.hwbot.org/result.do?resultId=635655 Same as above. X2 4200+ Toledo http://www.hwbot.org/result.do?resultId=562499 Wrong version? It was also posted in the X2 4200+ Manchester section, even if it's the wrong core. Most of these results have this difference with the normal scores: Total Physical memory Allocated: 61757 (normal scores have all 61372).