knopflerbruce

The reason why the limit was created was to eliminate quantity benchers to get into approx. the top 10's, maybe 20's (that's what it sounded like when I read the old threads from when v2 was released at least). Even if you double the limit we're no threat to the top 20:o Now we struggle to get into the top 150:o In some way I understand your 79 of 12000 argument, but I'm sure that no more than 79 people had serious issues before the limit was created either, so if you need the majority to actually have a problem with something before you change it, I'd say the limit shouldn't be raised - it should be removed, as it was obviously a mistake to create it in the first place. (longest sentence ever... sry) I have another suggestion, too. Could you make an "overall league", which would be as if there were no limits? For some reason I just don't feel that it's alot of prestige to have a high rank on the HW masters list. No-one ever mentions it anywhere, like it doesn't exist... It's like 1 Bundesliga vs. Oberliga:p Maybe such an overall rank would make it a bit more interesting, as you fight the whole bot, not just the ~100 who try to get hardware points. As a result, you may get alot more results for older HW which have maybe 1 or 2 entries per bench today, which wouldd be great, right?

In this case, in the 4850-category, as 4850>3850.

As long as a LOT of the HW listed doesn't even have all "point slots" filled, the limit has to be there, or it would be VERY easy to get a top 5 global score. Just bench a whole lot of old-school HW @ stock - till you've reached your goal. However, the limit is awfully low, it needs to be raised a bit... You're no threat to anything when you're at 160th place (300 points).

I wrote that I used a DFI NF4 SLI-DR Venus for alot of results in the past. It never got added, and the same with my ASUS A8N-SLI Deluxe. I'm just a bit too lazy to add mobo info when it's not listed:o

I also received a PM that explained this:D Glad to see that DDR2 2GB modules can do that well, too - and of course that it looks to be a real score:)

Maybe he got some suspicious scores before, but this one looks real to me. One small "problem" is that I never heard of 2x2GB DDR2 that can do that high 1T benching, though. But I'm so absolutely not an expert on DDR2 so that could very well be lack of knowledge on my side:p I'd say we wait and see if more CPUs are capable of benching at these insane speeds:)

I notice that alot of old scores have missing screenshots, not just the one you just mentioned. Why is that? Something is a bit weird here;) Are all images deleted after two years or so+ I see that even when I search XS, images in oooold threads have vanished, which is annoying when I look for mods etc... but to have a score deleted that way is much, much worse. I'm planning to buy a couple of memory sticks for my screenshots, just in case they start disappearing from HWbot, too...

LOL! At least they use the same OS installation, mobo and memory. I'm willing to bet they use the same chip as well, 3 different ES's that have almost indentical potential:D Maybe they didn't know about the HW sharing policy:confused: If they claim all scores are real I'd be interested in knowing how they can explain why everything seems almost identical:p

If they have a tweak they don't want to share, it's OK. If everyone shared every tweak they know, the guys with the fastest CPUs will always win. Which is kinda boring, and not fair to the guys who really know how to tweak. There are at least two ways to get on top of the charts: get the best HW, or have the best tweaks. If your tweaks aren't good enough, you just have to go look for some better memory sticks or CPUs:) I wouldn't mind knowing Coolaler's tweaks, though. Would make it alot easier for me to grab some more 1m 939 scores:D

Just a quick HW sharing question... my opteron 165 WR chip was bought from a team member who submitted a score here. That's legal, right? No HW sharing issues?

Ticket ID: 44 Priority: High I noticed that one of my scores had been moved from Opteron 165 \"Denmark\" to Opteron 165 \"Toledo\". \r\n\r\nFirst, these cores are identical - no performance difference whatsoever, so there should be no need to separate them in the first place.\r\n\r\nSecond, these chips are all labeled as Denmark cores afaik. At some point CPUZ stopped reporting the core name to be Toledo, and then reported Denmark instead (which is correct). This can be easily proven by using an old and a new CPUZ version at the same time. I can\'t, as I don\'t have the necessary equipment to use a de-lidded CPU, but I noticed that Untouchable\'s score shows a Toledo core - and my chip is the same as he used. He didn\'t have the time to really max it out, so I bought it from him.

Not very many people do this, and IMO phase change cooled air = phase change cooling, Same with a water chiller. The last example is IMO either dice cooling, or perhaps phase change (the temps you get will be comparable to a SS, even if it's not using a compressor)

You speak Polish, that post should be piece of cake for you:) :D:D

If you look at the black field on the CPU you'll see if it says mobile AMD Athlon or just AMD Athlon. If it's the last one, then you don't have a mobile chip - even if it's lidless in the first place. I know that, as I have like 4 or 5 myself, and I was a bit surprised when I saw that they weren't mobiles - just non-lidded desktop CPUs. "Real" mobiles DO exist, they're just a bit more rare:)

...with a valid checksum... right:D What bothers me is that you don't have to upload a screenshot , as it's quite hard to find out the configuration of the bench PC without it. If my score is beaten, I'd like to see what it took to beat it, not just see the time, and that's it. Mem speed/timings, CPU speed, mobo etc should always be easy to find, just in case someone wants that (vital) information. That's not the case with the online submission system, I usually see pretty much nothing except the score!

+1 (I'm expecting to get one of these cards within a couple of week)

I like wPrime, but how can it be so hard to keep a checksum validation mechanism bug-free? I hope someone would spend a little time, trying to make a version that works 100% so there will be no more discussion;)

I think most (if not all) scores are 754. The 939 version is quite rare. ALso note that 939 supports dual channel, so the superpi/pifast/wprime-scores of s939 3000+'s should be quite easy to spot - dual channel helps quite a bit.

Ticket ID: 19 Priority: Low http://valid.x86-secret.com/show_oc.php?id=260824\r\n\r\n(200x10, 1.50v stock)

Ticket ID: 18 Priority: Low http://valid.x86-secret.com/show_oc?id=204045\r\n\r\n(200x9 stock speed, 1.50v)

I've seen over 300mhz on 2x1GB, but I have NO clue about 4x1GB. Highest I ever saw was 275mhz stable, but I don't know if that was some kind of record.

Check your spam folder. If that doesn't work, reupload and write down the ID somewhere:p

There's no need to block scores that are normal for the hardware. If something is a bit weird, then perhaps it should be blocked because of missing CPUZ screens, but as long as it's out of top 5, I'd say it's not a problem.

A minimum and a maximum PP is OK, as long as you're not too strict (it shouldn't block scores that are just below the best PP, or block a score made with alot of apps running, and slow memfreq + bad timings.). Just remember that PP is different for every CPU model;) Show PP: yes! The mod stuff is yes as well IMO.

CAS is the best way to sort the scores. IMO.