flanker
Members
-
Joined
-
Last visited
Everything posted by flanker
-
-
-
where is Stilts magic 8 GHz? I saw some screen few days ago
-
depends at cooling...high air or allinone watter, you can set 1.55V. But there is a problem, you can have CPU part stable OK, but iGPU part not (if overclocked both parts in APU). With higher GPU clocks is harder to get more MHz at DRAM. Big stress for IMC.
-
-
I have one 8350 with core 3/4 the best and second chip has great core 0/1. I tested first one, about 8040 MHz with reservation (only 1.95V). Second chip will be the beast, 1.56V and 5730 MHz with "air" and normal ambients in room
-
nice and thx man!, do you have some validation results?
-
but still very nice!
-
-
awesome for watter cooling
-
-
what about driving settings? Voltage of DRAM, IMC? Did you tried different dividers?
-
-
looks, my 1.60 x64 and win7 working good and I want not change it until will be fixed
-
Hm...Im stay at 1.60x64....
-
-
-
only one RAM can u help a bit...I think, there is only big diference between batchs and pieces.
-
-
-
-
:-)
-
7100? Nice!
-
the banik: the main problem is, no much people testing FX under LN2, this is real problem. If u seen Chews* FXs, most of these scaling good over 8 GHz in quick test. And it was not binned chip, after this he took one for LHe with "lucky" hand. Example from my experience: I have 4 FX chip home. And there are differences. One is good for AIR with very good power consumption. But poor with LN2 (7300 MHz max). Second is 7700-7800 MHz chip. Died under LN2, at air was not special, average chip. Others two are 8150. One is average at AIR with "good" power consumption. Second is great AIR chip, but very hot and hungry. This chips still waiting for LN2, my estimate is second chip will be 8 GHz chip with LN2. If I got support as some guys here, I can testing more FX chips with LN2. Ln2 is cheap here and is about 5km from my home Did u read all my text? If he show good run superpi at 8500 MHz, after is at least 8800 MHz validation possible. 8500 MHz superpi must be at least 10-10.5s (without much tweaking uncore, operating systems and bad skill). Tuned FX at 8 GHz has superpi1M about 10.6s
-
we will see for next proof, maybe he will try again in superpi, if superpi run will be about 10-10.5s at 8500 MHz, I have no problem believe 9 GHz validation. Between FX are very big diferences, at air some can hit only 4500-4550 MHz stable, but there are few over 5000 MHz stable. The same with LN2, worst of these hit only 7000-7200 MHz, the best over 8500 MHz.