Printable View
Finally we can see some review great mobo Corsshair IV Extreme in Czech Republic. Is more than good, he has been achieved WR AMD at superpi benchmarks:eek::up:
Thanks WR user and writer of reviews OBR from PCtuning.cz
http://pctuning.tyden.cz/ilustrace3/...rd_2_small.png
http://pctuning.tyden.cz/ilustrace3/...cord_small.png
review ASUS Cross IV Extreme review google translate
Is it just me or is that 1m a little too fast?
OBR has a track record of awesome pi efficiency :)
Solid benchmarks and clock speeds :)
Better RAM = sub 10m 32m? :D
Very well done to him :toast:
Are the 4 all PCI-E slots x16 wired ?
congratulations!!!! cant wait for sub 10 min :up:
it did strike me that mem timings are so high? there is big chance for sub 10min
how did he get low 9s 1M when most people at clocks like that were getting low 10s?
I'd like to see some matx Crosshair boards :D
Booj and others doubters:
yes, because with superpi and wazza is not cheating, so u can got under10min think. This point is good IMC and high CPU freq. 6909 MHz, NB 4605 MHZ, RAM 2046 MHz 9-8-8-24-40-1T. For 1M at AMD is better higher RAM clock than timings and next +80 MHz CPU clock diference. If u can, try it at home (setings with some RAM clock and timings in 1M, lower timings and lower RAM clock with lower CPU clock, higher RAM clocks with higher CPU clock etc etc. can combined to diferent HT clock, its playing at much much hours :-) )
Look at MADs result, its 2 AMD WR for 1M superpi. I Know from MAD222, he not use tweak OS and not wazza.
http://www.hwbot.org/signature.img?i...se&iehack=.jpg
CPU clock: 6831 MHz
NB clock: 4807 MHz
RAM 1686MHz 6-6-5-15-11-1T
For me, 1M result is really corect with rules
Yeah, that 1M is incredibly fast. The 32M is also a bit funky compared to the top 3 X4 965s. Either CPU-Z is reading the timings wrong or we've all been approaching SPi 32M the wrong way.
very nice clocks...:up:
that was done using Windows 7?
the 1m is bugged, got removed from bot just some FYI sub 65K effeciency is not even remotely possible when sub 67K is the goal.
Not to take away from fact that if he runs it again......it will be a 1m wr at those speeds.
gomeler: yes, its quick, but not unreal with this setings, maybe later OBR will test this same chip and tell him record some video...OBR is big talker, but not cheater, believe me.
and with wazza tweak it is impossible do u think?:shrug: MAD dont used tweak OS etc...think, with some legals tweak it is possible...Quote:
Originally Posted by chew*
We are talking 1m here..............waza is not shaving off 3k worth of effeciency on a 9 sec bench. The diff from a well tweaked run to a non tweaked run would be lets say 67,900 effeciency versus like 66,900 effeciency.Quote:
Originally Posted by FlanK3r
Don't take my word for it take hwbots ;) Results get removed for reasons, not just for kicks.
In all seriousness the best I have ever seen is sub 67K. This is 64k effec......thats a huge huge jump.......
No one saying he cheated, he just got a bugged run, it happens. No one doubts at those speeds without a bugged run it will surely be 1m record either ;)
seems legit to me, i sense a lot of, can i say envy here? just because he does something over and above anything that has been seen, doesnt mean it isnt possible.
new tweak maybe?
and heres a random pic of my new invention, going into production 2011,
http://www.jonco48.com/blog/shoppingcart.jpg
chew dosnt get jealous, he just gets more LN2
Sure ok i'm jealous and its a new tweak.............:rolleyes:Quote:
Originally Posted by DosDuoNo
The fact remains it was removed from hwbot.....the question you then need to ask yourself is why.........then you need to follow up that question and ask yourself why OBR did not challenge the fact it was removed.
BTW the 32m result did not get removed. Just some fyi.
I don't want to call anyone a cheater unless it's blatant. More likely a bugged run. I'll be the first to congratulate him if such efficiency can be replicated.Quote:
Originally Posted by FlanK3r
i neither said 'Chew*' or 'Jealous' in my post, so exactly where are you getting your info?Quote:
Originally Posted by Manicdan
as for asking questions, i am pretty good at making my own investigations as to who, when, why and how, as have been not bad at this since i was say............. 4 years old.
its legit, as far as i am concerned, hwbot is a farce, simply in place to massage the ego's of those who pusue its 'brilliance'
have you seen my new trolley bike? wht you think?
Ahhh ok so i'm gonna use OPB tweak and beat that "record" on air then......rules don't seem so bad now.Quote:
Originally Posted by DosDuoNo
This is nothing about ego's here this is about software hardware limits and the limit is like sub 67K effeciency until amd does something about CPC performance.
its very simple, clock speed multiplied by result = effec.
try and break 67K, when you realize that task is hard enough sub 65K seems unrealistic.
go right ahead then, but i think your confusing me with someone else :rolleyes:Quote:
Originally Posted by chew*
thats it, you confusing me with someone that actualy cares what you do. :rofl:
which is probably why you thought i was replying to you, instead of giving a sarcastic perspective of chew's laid back personalityQuote:
Originally Posted by DosDuoNo
visit the AMD section, half his posts are him trying to help others get better results. his knowledge is one of the best that section sees.
need more coffee..........
me too this brown nosing is killing meQuote:
Originally Posted by OC Maximus
It's not a big deal if he runs it again he will hit around 9.700 get record and call it a day.
Like i said great result regardless he will have record at those speeds :)
pretty amazing runs there :) maybe even be able to squeeze out a few extra mhz if he disables 2 cores like i have seen others do :shrug:
next will validation too (7GHz+ think easy)
Hey, Flankerp...he must be a great AMD Overclocker:up:
Nice to see new WRs
Frankly, I would say 6.9GHz @ 9.3s is an unrealistic no. even with tweaks which may not make a very big diff. :rolleyes:
Anyway, congrats to OBR & hope to see more AMD WRs from him soon !:)
Be carefull madd, stating the obvious apparently nowadays makes you a troll or jealous :(Quote:
Originally Posted by Mad222
yeah watch out guys DosDuoNo is hungry, dont feed the troll!
And as has already been stated that 1M is def bugged. Cant wait for a rerun and a valid WR! :up: Great clockin silicon u got there
very nice :up:
Solid performance there :D
Keep pushing it :up:
:yepp::yepp::yepp: Brian-Y it's 7 64bitQuote:
Originally Posted by Brian y.
lol mr trackball u got it ;)
I don't think anyone has ever run SuperPi at that frequency... When at that speed I think it's safe to say anywhere remotely near that frequency, as once you get to that speed every MHz is more difficult that the last. I learned that when I hit 3.916GHz on my FX-55 and thought 4GHz would be a breeze :D
I'm working on my 8th GHz now, the first seven failed :rolleyes:
:rotf:Quote:
Originally Posted by chew*
chew* is right, this efficiency is simply impossible. Copy waza does very little for 1M, especially with the cache current cpu's have. Take it from a guy who has a fair share of experience with SuperPi 1M. ;)
MAD+Chew: ok, hope, we will se "true" in next LN2 station...Maybe really it was some bug, god know it.