Jump to content

Featured Replies

Posted

Nice result. Interesting to see that high cpu frequencies can improve the score so much. Congrats!

Is 29ns even possible on Gulftown IMC/DDR3? :/

 

Not really.

But maxxxmem is known for its... inaccuracy, so for maxxxmem 29ns mem latency is possible :P

  • Crew

Almost the same clocks as your 2109.9 score, your bandwidth did not change a lot but latency dropped 7 ns?!

Don't think this is possible, should be another bugged run.

  • Crew

maxmemm is just a bit all over the place, its not a bench you can consistantly push the same scores across different platforms, os differences can also cause this bench to differ greatly... its probably not a bugged run (he can probably duplicate this over and over) but more so a bugged benchmark

my fellow croatian did 2700+ points in maxmemm and he could run it over and over again, but the score was declared as bugged by hwbot moderation team...

microman.jpg

________

DODGE DAYTONA

Edited by timmyCRO

It's not a same settings ;) He played with subtimings. I know that, because i was there and helping with LN2. In first run he used setFSB to make freq higher, in second he boot, and just increase multiplier. Somehow whole system is faster that way and better scalling that way...

 

The real question is: If the maxmemm is so bugged software, how we can know that the first run was not bugged to? Or any other run??!?!

 

For example. If I score 2350 in maxmemm, how we can know that is real score? Maybe is 2300 ;) But that "50" iz enough for victory! :(

 

I realy bealive in hwbot crew, but this time i think that maxmemm is realy baaaad idea for importmant qualifications as "GOOC 2010 Serbia (ex-YU) Online Qualifiers".

 

:wave:

In comparison to the other scores already in the database (so not from the competition), the only reason why this score could've been possible was if the CPU speed would give a huge boost (others have 5.1G, here 5.5GHz). His first score was with the same CPU frequency and was not much faster than the others with same mem/unc frequency.

 

Subtimings is not really an explanation. Everyone plays with subtimings.

 

Problem is that this benchmark's producing bugged runs instead of crashing :(

MaxxxMem is TOTALLY incosistent, even worse than passmark ( that's an amazing achievement :D )

 

Not really.

 

The benchmark is in fact pretty consistent as long as your system is perfectly stable. At the very edge of stability, the benchmark is producing out of the ordinary latency scores (bandwidth figures are mostly ok).

Good news. Cancel maxxmem cause program is bugged. Period.

 

Program isn't so much explored.

How the hell we don't know if there is a lot of bugged runs [ even for 0.3ns ] on here. Do we need to check/verify all 600 [ lol] subbmited scores during a big championship - and explore how program scales...Could somebody tells us tweaks for maxxmem since tight OS, sub latency are not appropriate.

 

Disgusting.

 

@timmy - I am really feel sorry for your "croatian" friend but could you tell us more if he could repeat same runs ? What does that mean ? If someone have tight OS, cool sub latency and not 6+ghz he could not have a high score ?

 

Hwbot what's deal with maxxmem ?

Edited by Oki

In comparison to the other scores already in the database (so not from the competition), the only reason why this score could've been possible was if the CPU speed would give a huge boost (others have 5.1G, here 5.5GHz). His first score was with the same CPU frequency and was not much faster than the others with same mem/unc frequency.

 

Subtimings is not really an explanation. Everyone plays with subtimings.

 

Problem is that this benchmark's producing bugged runs instead of crashing :(

 

I agree Massman, PCAXE crew realise that after competition... But.... Jale00 and old other overclockers around the world will not be happy...

For example, Jale00's Pi Fast is much better than the other guys in competition even on much higher frequency. If he know (that was the first time he was using maxmemm..) that was bug, he will concetrated on other bench software... It was run with others, and time...

 

I am so sad for overclocking as a sport because all of this...

 

Maxmemm has only 553 records...c c c c... Bad call if You ask me...:( :( :(

Edited by animaN

So why does my perfectly 24/7 stable system ( which is not pushed at all, just 4GHz i7 980X & DDR3-2000 7-9-7-24 ) scores differ by 100pts and up to 400pts from run to run ? :D

So why does my perfectly 24/7 stable system ( which is not pushed at all, just 4GHz i7 980X & DDR3-2000 7-9-7-24 ) scores differ by 100pts and up to 400pts from run to run ? :D

 

Because maxmemm is worst bench software ever...

Good news. Cancel maxxmem cause program is bugged. Period.

 

Program isn't so much explored.

How the hell we don't know if there is a lot of bugged runs [ even for 0.3ns ] on here. Do we need to check/verify all 600 [ lol] subbmited scores during a big championship - and explore how program scales...Could somebody tells us tweaks for maxxmem since tight OS, sub latency are not appropriate.

 

Disgusting.

 

Hwbot what's deal with maxxmem ?

 

There aren't that much bugged scores ... no need to become overly paranoid. From the June OC Challenge, I've learned that MaxxMem is pretty consistent at stable settings (within 10p ~ 2% variance). When NB/IMC is pushed to the edge, latency gets bugged ...

 

Our deal with MaxxMem is just that we include the bench if proven stable. Nothing else.

 

So why does my perfectly 24/7 stable system ( which is not pushed at all, just 4GHz i7 980X & DDR3-2000 7-9-7-24 ) scores differ by 100pts and up to 400pts from run to run ? :D

 

That's because MaxxMem cannot work well with altered bios microcode ...

Ok, Massman please explain this difference on 24/7 stable system which I'm using for CAD drawing and rendering.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...