Massman

Girl of your life?

Even ancient scores (pre-2004) I wouldn't allow in the database if the cpu info is missing. It's so darn vital ;D

Sorry, but CPU CPU-Z is like the most essential part of verifying a score.

The problem is that there's no CPU CPU-Z tab in the score. Although a verification link indicates something, it's most definitly not 'enough' to serve as verification. A CPU CPU-Z tab is an absolute necessity.

Haha, I'm imagining you both yelling through the phone without understanding each other "Hey man, I just cracked 29k in PCMark05!" "Sure, I do have a lovely wife. How did the benching go?"

Fixed.

Nice one. I like your bench environment as well

Actually, it's how most new bench tables work. It's much easier than using 4 mounting holes

Correct.

Setup looks really clean, nice!!!

This time, I'll follow the lead of XS for once. Thread closed. This is what happens next month: - Submissions are sent in by mail - Scores will be unveiled at the end of the month on the F1-OC.org microsite - Discussion about the scores will follow afterwards

Perfectly valid

I offered to close it when the bugs were fixed, but Chris was okay with keeping it open.

That's weird. Looking into it, thanks for posting.

Okay, working on this. Already got the results for global/hardware split up ... interesting stuff. Now working on averages.

I'll try to find a matching query .

That's quite high, I think. What are the subtest scores?

We have a hwbot staff meet next thursday, I'll put the sponsor logo on the agenda. Untill then, use a customised desktop or waterlogo to be on the safe side :-)

Their = a manufacturer which is currently in the database?

There seems to be a bit confusion regarding the world record page. In short, the world record page is a list of the highest scores in each benchmark containing both hwboint and non-hwboint scores. It's what we consider the official WR page on hwbot. The hall of fame you're refering to is, I guess, the ranking of the hwboint scores. It's still visible if you go via the hwbot application info page or the summary page (http://hwbot.org/benchmark.do). Also, you can go straight to a BM's ranking by clicking on the world record in the list on the right hand side:

Okay, I see what is happening here. Problem is that the results that were tagged in the past to a certain chipset are not updated with the new information after cleaing up the database. So, it seems like scores were somehow tagged as 'P55' although an X58 motherboard was selected. It looks like most of the scores have the right motherboard indicated, which is quite bizar as we should have a security implented to prohibits non-matching hardware to be selected. To RichBa5tard: I presume that the 'chipset' field of the result is the ID stored in result.mb_chipset_id? Would it be possible to implent that IF a motherboard type has been indicated, that field is overriden with the id found in mb_model.chipset_id?

Must be something wrong; all X58 and P55 motherboards should be completely cleaned up. It doesn't make sense that X58 results (result indicated X58!) show up in the P55 search.

Verified .

Last update before the break. - Tested UD6 more than a couple of times this week; same story: BCLK capabilities for 32M is inferior compared to the GD80. Efficiency at high clocks is inferior as well. It might be an issue with my configuration/OS/skill as other people are in fact producing quite good scores. - GD80 is back to HQ to get a couple of updates, for instance PCI-e clock improvements.