Massman

Did you either use Mipmap or disable post-processing on that run?

You should have recieved an email, no?

Moderator tool says: "371 submissions left to moderate!" Not that bad, knowing that the number has been over 2000 in the past . Please understand that in certain periods every single crew member has a lot going on besides hwbot. This still is a hobby site, which doesn't mean we shouldn't do our work thorough, but which does mean that it can take a while before we can solve problems.

There are no points given to the Sisoft benchmark and, by general consensus, everyone can choose the version he or she wants for the best result.

It's just not possible to compare results of two completely different platforms. It's purely a matter of compatibility of scores; that's why, for instance, speedhacks are illegal as well. Surely, one could get a better score, but the test environments are nowhere near similar.

I guess so

http://www.hwbot.org/hwbot.post.do?postId=386 There might've been some changes as the more results, the more cpu power is needed, but only RB can confirm that

Take this to PM/mail, thanks.

If I remember correctly, you were one of the people who actually worked in the CDT-thread over at XS, no? Anyway, the fact that you're quitting the overclocking scene partially because multi-dollar Vantage rigs rule the world has nothing to do with the actual money you have to spend on your rig, but has everything to do with what goals you set for yourself in this overclocking community. If you're aiming for top spots in every benchmark, then yes, the multi-doller Vantage rigs will mess things up for you pretty badly, assuming you will not find any source of hardware besides your own pocket. If you, on the contrary, aim for having fun in overclocking, it's pretty easy to have a lot of fun overclocking ... very easy in fact. I have at the moment the gear to aim for top spots in some global rankings: E8600 with excellent stepping and a 4870X2. But instead, I rather spend my time hunting for sub30s 1M with my Athlon XP ... weird no? And I promise you, when I've finally broke 30s, I'll probably start benching my Asrock AGP motherboard, for which I bought an X6800. It's all a matter of personal objectives. Now, I do know what comment will follow: Hwbot doesn't allow people to bench in old hardware categories, because the big points are only to find in the global rankings. Well, we are actually working on a new hwboints system that would hand out more points to excellent old hardware overclocks. From the moment RB gives me approval to post the suggestion in public, you'll know what it's about. Next there's the rather incorrect analogy of the KGB-alike instances. I think it's more than normal, at least these days, that scores can be questioned and discussed, which is in fact completely the opposite of what you like us to believe. There's room for discussion and doubt, which is not the fact in what you say. There's a giant difference between being accused and being convicted here, please understand that. If you have indeed no place in such world, namely the overclocking scene, than you will have a hard time anywhere else as well. Everywhere in this world, when someone is better than the rest, people will question that person and his results. Jezus supposed to do miracles, but I know many people, including me, have a hard time believing he actually walked over water. Bolt broke two world records on the latest Olympics, but even the head of the Olympic comitee has doubts and claims that "we'll only know the true champions of these Olympics in 2016". Do you remember what happened when Kinc and Shamino broke the 3DM06 world record a few years ago? I'll spare you the details, but they eventually pulled that score, although it was proved more than legit a few months later. Which is a comment I agree totally on. If you want to give comments on other people, accept that you will be given similar comments back. There's a very nice thread in here that illustrates what is being said here. Actually, he never said you didn't bench in public or don't have any experience in overclocking whatsoever. All he tried to explain was that overclocking is not predictable, which is actually incorrect as I explained underneath. That's not the point, John. The point is that this kind of gem hardware can always be pushed, in front of an audience, to certain frequencies which make it special overclocking gear. If he hits 650Mhz instead of 660Mhz, it's fine as well, it's just a matter of showing that with a little bit of luck and patience it's possible to achieve that frequency. Same goes for the videocards. If he hits 1210Mhz instead of 1250Mhz, everyone will believe that 1250MHz is doable in an enclosed environment where you have all the time to work up to the highest frequencies. Overclocking is actually quite predictable in this way that equal hardware at equal frequencies should perform pretty much equal, variance allowed of course. It's not like your CPU will do 6500MHz one day and only 5700MHz the next day in the same conditions. And if it does, there's definitly something wrong with that CPU. That was my initial thought as well

Okay, I think we've seen and said enough about this. From what I understand of this discussion, the main problem is the fact that Team Latvia (Tapakah, MrLobber and Kiwi) always seems to have very, very good hardware to bench with, be it either videocards, processors, memory kits or motherboards. If I understand correctly, the actual scores are not being questioned, nor are the actual frequencies that have been achieved on that particular hardware; it's a matter of 'too much good hardware for one person', especially since those persons claim not to have handpicked hardware. Now, let me clear up the term 'handpicked' as I think we're using different definitions here. If Tapakah refers to handpicking, he means hardware that has been selected by the manufacturer and send to a bencher in order to achieve outstanding frequencies. Pro and Gautam probably use the definition to handpick equals to select one out of many samples, which is very different from the definition Tapakah uses. In the first one, you only have one piece of hardware that you work with: the very good overclocking cpu/vga/memory/motherboard; in the second definition you have to pretest very much hardware in order to FIND the best overclocking one. As far as I know (I didn't check it myself), Tapakah has worked in the warehouse of an hardware distributor at the RMA department. If that's true, he would have acces to A LOT OF hardware, which he could pretest anytime he wanted to. Considering the amount of time he spends on overclocking, from what I've heard from MrLobber, it wouldn't come to a surprise to me if he indeed has spent that much time on pretesting the hardware. Now, this process can be called handpicking hardware, if you use the second definition, no? Furthermore, chances of finding the golden hardware would raise quite a bit. In addition, it's a known fact that not all the hardware is Tapakah's, which means that lightning didn't struck at the same place every single time. BUT, even if we rule out the hypothesis of the incorrect definition, the scores still look valid, as there are no scores that are really out of line, as Gautam already said. Therefor, it would seem a little drastic and in fact way over the top to consider the results as cheated and invalid, as we only have a limited amount of tools which we can use to identify a cheat. Statistics can be used to raise questions, but cannot be used to convict a person. As already said, there is a way to decrease the questions raised about the legitimacy of Team Latvia's scores, namely bench in public. The longer this is postponed, the longer people will have questions, obviously, so I would ask Sam to think about HIT again. As far as I know, the people who organise the bench session at HIT can help you to come and university schedule ... well, skipping a few lessons shouldn't be that bad (). I see you have plans on visiting Cebit09, so hopefully we will actually meet there. Regarding the other comments: - Unlike what some persons apparenly think, being a member of the OCX Team does not make to a cheater and the membership can not be used as an argument in this kind of discussion. - Although everyone has the right to question a score, I would still very much stress the fact that we are on a public forum and therefor should be polite. Behave, behave, behave!! - Yes, we all know that benchmarks can be cheated ... I think we also know that people do actually cheat, just like you'll find people cheating in every sport. I believe that the correct question is not "How can we cheat", but "how can we reduce the number of cheaters"

To whine = moaning, bitching around Whine = a alcohol-containing drink which (in france) is often accompanied by cheese as some sort of snack. I'll be in Russia as well, by the way .

Sorry, can't solve the problem, only RB can. Only thing I can do is force the server to recalculate the points, but that asks quite a lot from the server.

I forced the bot to recalculate the boints for now, previous toplist was about 8 scores, the lowest one having 1,5p. Problem is that the boints are not recalculating at all, so I think the boints schedule is messed up.

Most definitly a server issue, the top scores of today are not that great ....

May be a slow server issue

The superpi 1M score. Hardware sharing: two members did a joined bench session and uploaded results of each other's bench system. Members A and B A owns System C B owns System D => A submits scores with D, B submits scores with C. This is violating the hardware sharing rules, so I blocked them.

You are allowed to lend your CPU to someone else for 3D benchmarking and you are allowed to lend someone else's GPU for 2D benching. No cpu for 2D and no gpu for 3D. Your scores will not be deleted if you actually own the hardware. We are talking about SHARING, which means that you have ONE setup and TWO people benching it. If you buy yourself an 4870X2, your scores will not be deleted because I have uploaded results with an 4870X2 . The blocked scores will not come online, because they violated the hardware sharing rules.

Have you checked the score since you posted it here?

In many cases, it's difficult to know, because we don't attend every bench session. We have had some issues regarding this in the past and the best solution to prevent scores is to take pictures of the hardware you'll be using during that bench session (for instance, if you have 1 GF8800u and the other has 1 GF8800u, make one pictures of the two cards). In this case, it was quite easy, because you are on my team.

Thanks for pointing out that we still don't spend enough time on HWbot; from now on I'll try to ditch all my other social activities to be more active

I am. The problem is hardware sharing: http://www.hwbot.org/hwbot.post.do?postId=689 Actually, it is. If we don't apply these rules, it's so easy to get points: one bencher has a golden setup, he invites 100 members to his house and lets them bench his setup. 100 top scores of one team => gigantic boost. You should, actually. They should've been sent to crochr@hotmail.com and if not, please let us know, because that might be a bug.

Have fun doing so, but please don't complain when we are not able to keep up with your reports. It's because if these kind of actions that the result moderators are behind on schedule.

http://www.hwbot.org/hwbot.post.do?postId=717

Gpu-Z is absent

But completely useless when you want to compare hardware setups ;-).