Antinomy

You're not messing with the benchmark - only the driver's optimizations

Hm, XP didn't give me to change Aquamark's priority - "access denied" that's what taskmanager was talking about. This tweak is about application-specific driver optimizations like it was with 3DMark 2001 and all the rest of em, right? Does the score change only on NV or cards of both manufacturers?

Looked at the bottom of the forum's main page and saw that today is TaPaKaH's birthday. HB, Sam! Keep pushing and shocking us with your records 20 years is quite a nice age - not too young to be defined as a kid and so many years ahead. Enjoy this day and do things that you like to (not only today).

It was cooled by LN2.

Outstanding! =)

I even got some backups for qwerty84

I like the number =)

http://www.overclockers.ru/images/lab/2008/12/22/SPI/Rabstol4_FULL.JPG - I hope you get what this screen is despite the language It's the control panel's display properties. Then you should add an OS check - it definitely shouldn't add any reg-key to my XP Alexander

Mutant_Tractor, what does the "transparency windows" tweak do? Checked it, restarted, the checkbox dissapeared in BTTB but subtest score didn't change. Is it OS specific? You might disable video hardware acceleration is screen properties - help for superpi. If it doesn't do this already

Massman, I can say that I like the direction were the HWBot staff is going. There are many doubtful things. But the idea about one best result per team is nice, I've rechecked this a week ago. For those who say that it kills small teams - not it's not. It hurts big team much more - the ones having dozens of same videocards and one not having them The new rules encourage to use not only the popular hardware but the rare one too. This will discourage newcomers - as we've all started benching from our own computer it's likely that a team won't gain much from a newcomer - if the hardware is popular it's very likely already been benched and no improvement can be done (because of the low skill of a newbie). As for the temperature and frequency limits - I'd start to consider them seriously when there are tools to control both. Anyway - I love you guys doing hard work on burning your brains thinking how to make things better but not screwing them. One tip - don't go too far in making algorithms more complicated. If it takes a cup or two of coffee to understand them - it's one thing. If, as we say, "you can't get it without half a litre" (meaning vodka) - it's not attractive.

Nothing bad - everyone of us was in school. Everyone was a newbie. Not moving forward is the only bad that can be done

I asked for a photo oh his CPU and there's a reason. As for the second datasheet: Pentium processor with MMX technology: –Pentium processor with MMX technology at 233 MHz, iCOMP® Index 2.0 rating = 203 –Pentium processor with MMX technology at 200 MHz, iCOMP Index 2.0 rating = 182 –Pentium processor with MMX technology at 166 MHz, iCOMP Index 2.0 rating = 160 As for the 133MHz, the datasheet says: "Currently, there are no products that support these bus fractions" The first datasheet is referring to mobile Pentium MMX. Or we're going to add "some Pentium CPU with MMX and 133MHz" - not knowing what we want?

Wrong. Can you make a photo of your CPU?

As for the Pentium MMX133 - can you provide proof that it exists? SSpec and links to trustful sites? The ark.intel.com is wicked, very messed up and a crappy tool (bad decision to close processorfinder). I don't remember to see this processor in the datasheets. And link to Intel's datasheet would be nice in proving the existence. Or make a photo of yours.

BenchZowner, how about the cache count?

Adaptec's controllers were always junk What did you run on I-ram's and what on Acard? How do they perform comparing to each other?

Very strange. wPrime 1.55 and additional libraries installed?

What OS and wPrime version do you use?

excellent!

bump :

Usually the user doesn't see reports on his submission - you can check the informer log when reporting. It says smth like "Sent to user - false Team moderator - false Crew - true" So user involvement isn't a big problem.

Except the case when a moderator deletes it, isn't it?

This happened only with the latest Core i7/i5/i3 CPUs. All the rest are on the contrary better using retail samples.

Please, add this stuff to rules That's the only thing I always ask for - the judgement should correspond to the judgement list which are the Rules

Or, you can show a "not exact in round" after the final loop to show me I'm wrong