some run benches with classic AIR and normal ambient

[crazydiamond]

ufffff! cant wait! :-)

[FlanK3r]

no, its not Bulldozer sample :-D.....

[El Gappo]

Cant wait to see how it performs flanker. Not seen much of any board that aint the CHIV lately.

Been airbending myself this morning. New target for you





Board needs more voltage than most and this 25C ambient isn't helping.

[FlanK3r]

Gappo-nice 1 man...know, im fearing a bit, Thuban is not Deneb, its bit diferent and want no kill chip :-), did u try validation with AIR? Which your series?

OK, with this lower, il not fear voltage :-D (bad quality photos):
again 1015, but EPMW, SN 00208

[crazydiamond]

nice el gappo one of my cores likes to crap out shortly over 4600mhz so its hard for me to make big valids or get great clocks on these single thread benchies like that when im trying to run them w/ all cores clocked up the same

edit: and AWESOME!!! flank3r!!! make sure you do some good testing on it before you torture it, im gonna guess stability is similar to your dpmw but not as golden for valids,,, we'll see if im close

[FlanK3r]

1.62V is for my "gold" chip dangerous, risk to possible kill. But for this second....hm, if i kill it, kill it :-D. But im hoping for later LN2 sesion with this EMPW chip.

[El Gappo]

@Crazy diamond. disable cores or set affinity or both.

@flanker, tis a 1015GPMW and not the best :/ The board also requires more voltage for the same clocks than other boards... But I did not set 1.62 it just thinks I did It's set to 1.6 I believe, which was overkill. I reakon 1.57 would do it on a cool day and if my heat sink wasn't full of these

The perils of having an awesome techstation?


SUICIDAL MOTHS

Cleaned this heat sink 4 times since last night. Finally decided to put some netting up at the windows.


Oh and btw moths are not conductive :p


Cant really do any validation stuff atm, need to install vista

[BeepBeep2]

@SUICIDAL MOTHS
Oh and btw moths are not conductive :p

Sigged.

[El Gappo]

LOL the worst thing is I had bum lube on the board from my phase because I was planning to map out some read points do a little air testing and stick it back on phase... Well the little delta fan I had on the NB sucked a big one up and splattered it all over my bum lube Didnt look to pretty.

It's this bloody power phase led that attracts them. I'll be sure to mention it in the review haha.

[FlanK3r]

@SUICIDAL MOTHS

:-D, n1. Maybe i can try kill some mosquitos and load Ultra Kaze at maximum

[crazydiamond]

i really dont know how to say how awesome this thread just turned!

[FlanK3r]

Today is a bit hot, but i have time for a lot benching my second Thuban series 1015 EPMW. First looks at screenshot . Now im instaling at second Thuban support DVD and than benchmarking with BOX

[FlanK3r]

not extra chip to time....i testing only with BOX, but definetely is whorse than my first chip...

[crazydiamond]

worse for valid and benchmark speeds?

[FlanK3r]

Ok, done, i have not too much nerve
In my room was hot, 27 C and i was warm too . This second Thuban 1015 EPMW is not so good as my first 1015DPMW. I used 500W PSU Seasonic from my x4 965, 1x DDR3 module A-Data, GA890FXA-UD5 with F3 BIOS, fresh OS and CPU+BOX cooler.
I thinking, AMD box is not so bad, meaning with 1.45V is not problem overclock stability. I tested first wprime burning at 1024M, no tweaking RAMs etc. Tried 4100 MHz, 1.5V and temps was max 67 C :-)


Than i found max overclock for validation. Compared with first chip, i was a bit disappointed. Think, this is about a normal Thuban chip:

4510 MHz with 1.535V, maybe with more time i can got about 4550 MHz

validation

Time to Everest benchmarks between 4300-4350 MHz














superpi 1M was last...no tweaking RAMs timing

[FlanK3r]

my system 1015 EPMW from testing - video

And a lot screens:




Black Editions

[BeepBeep2]

That is NOT normal for a thuban chip.
The chip is too hot. Remember that Thubans temp sensors are broken.
67c is probably 80c.

My 965 does 4494 valid, and SuperPi at 4400+. Not at that voltage, but certainly safe volts for Deneb. (1.61v)

Temps make a lot of difference.

[FlanK3r]

no, at some boards coretemp show corect and here at Gigabyte was it ok, cause idle temps was 34 C (rooms 27) and "gold" Thuban show me at Cros IV Formula 24 C (its 10C diference, yes, here is not correct)

[BeepBeep2]

no, at some boards coretemp show corect and here at Gigabyte was it ok, cause idle temps was 34 C (rooms 27) and "gold" Thuban show me at Cros IV Formula 24 C (its 10C diference, yes, here is not correct)

CoreTemp shows temperatures on reported from the CPU.

34c is most likely still NOT correct.
On the box cooler, my 965 at stock volts idles/clocks at 43c.

[FlanK3r]

remember, stock 965 and 1090t is diferent..and i had not chip, where isrunning at more than 62 C normaly...Denebs with higher temps blackscreen or frozed system first

[FlanK3r]

Today i have for us max low voltage stability. I set some steps in windows, because i was boring going again to BIOS :-)
So, how do u like 1.3V and 3954 MHz in PRIME? PS: Coretemp show me not corectly, think reality is 10 C higher. Still trying go higher than 4300 MHz (exactly 4309 MHz) in LINX or PRIME, but i have not luck :-/

[felix_w]

Nice result...1.3V for almost 4G...

What was your ambient temp ?

[FlanK3r]

21 C now, think, normally i have between 20-22 C, only some hot days i had 26-30 C :-)

[felix_w]

Right now in northern Athens its 31C (11:54PM)...it seems you have 12months/year Bench-Ready-Weather in Czech Republic, right ?

[FlanK3r]

no, we have outside now about 20 C, but in Juni and Juli was hot, 25-37 C outside, i was after work mostly at swimming pool :-) Heh, but 31 at "night", its not good, mainly not for sleeping :-D