-
Posts
3656 -
Joined
-
Last visited
-
Days Won
29
Content Type
Profiles
Forums
Events
Blogs
Posts posted by TaPaKaH
-
-
dinos, CW without maxmem actually helps and quite a bit
don't think Hicookie's eff is possible without CW, this is BY FAR (sub2150k vs 2200k+) the best 6.1ghz+ GT efficency ever seen
-
looks like you've messed up your waza a bit (started pi far too soon), but AMAZING efficency there!
-
also, it's 6000MHz, not 6200
congrats on finding that CPU, getting/running one of those top Prescotts is definitely not easy
-
"magic" button back here too, thanks!
-
so basically, you're looking for a man in pink strings who will run some 3D benchmarks with his handpicked 980X and GTX480, right ?
-
more like δT(°)/δt(s)->0 , as pot doesn't move along some imaginary Ox/Oy/Oz axis anyway
-
who said this is w/o copywaza? i only see a lack of maxmem
-
...don't have a backup against this
-
am not yet into AM3 - is mem controller on low end CPUs so crippled that 800MHz is an "achievement" ?
-
...hope he doesn't have backups
-
George, you could have bent one of memory slots (trying to keep it 2mm further away from the cpu) like I did on mine - solder quality on these boards is crappy in some places
5mins in the oven should fix your issue
-
what's the PCB code ?
-
with a very good 870 and 1Gb module it is easily possible, I did 274 with 2:12 (link) with 2Gb module
-
congrats on the lead in the "best CPU" race
-
can't see the screen ... but 9:19 with Clarkdale? wow!
-
whew ... I thought this was on H55N so my 32M is in trouble
-
tighter tRFC (~60) / tRTP (~3) don't help ?
-
I can confirm that "000" string is because of BIOS microcode option, not because of someone trying to hide it's an ES
-
well, that's actually nothing new .. even in P6T times you needed to boot around 4.8G with low multi but high Vcore, adjust multi from OS and bench
-
can you tell me the batch of that 631 ?
-
on hoes, Massman enters ... will my backups hold up ?
-
can you please tell batch and PCB code?
-
how the hell did you get that eff out of a Clarkdale with that low QPI and that loose mem ?
-
1175 7-8-7 (tRRD/tRTP/tWTP/tWTR(same rank) = 4/4/20/18) and 940 6-6-5 (2/2/18/16) gets almost exactly the same score for me (around 8:49.1), changing tRP from 7 to 6 gained me good 1.5 seconds
The official GIGABYTE H55 miniITX contest thread.
in HWBOT Competitions
Posted
As I already told on OCX, there is a valid reason why this cannot be done now (specifically, not having any Lynnfield at the moment).
If this is a case of double standard, fair enough, use the "gigabyte employee rule" to remove my (not my own, obviously) super-cheat-photoshopped scores, it's not like I lose anything in a "fun" contest.