*HOT* | Corsair DOMINATOR GT | 6GB (3x2GB) | DDR3-2000 7-8-7-20 @ 1.65V | *HOT*

[@@@@]

this modules are still not available in my country for those who where able to buy this kits can you test it if it can go a little better with the timings 7-7-7-20-1T 2000Mhz 1.65V instead of the 7-8-7-20-1T 2000Mhz 1.65V at least 4 hrs prime stable? thanks

[jcool]

Or just do 10 minutes linpack instead

[IronWarrior]

Thats only your opinion though. 1600 C7 vs. 2000 C7 is a HUGE difference to alot of people here. And many are willing and able to pay those high prices for the performance.

Indeed, I wish to aquire two 6GB set's for my P6T.

[TheKarmakazi]

this modules are still not available in my country for those who where able to buy this kits can you test it if it can go a little better with the timings 7-7-7-20-1T 2000Mhz 1.65V instead of the 7-8-7-20-1T 2000Mhz 1.65V at least 4 hrs prime stable? thanks

I believe you achieve much better clocks using 7-8-7-x vs 7-7-7-x. Theres virtually no bandwidth/latency difference between those two settings either.

[systemviper]

How are the Dominator 1866 kits, they are 400.00 for 3 x 2G

or should i just get the 1600's.... naaa

[cstkl1]

this modules are still not available in my country for those who where able to buy this kits can you test it if it can go a little better with the timings 7-7-7-20-1T 2000Mhz 1.65V instead of the 7-8-7-20-1T 2000Mhz 1.65V at least 4 hrs prime stable? thanks

Or just do 10 minutes linpack instead

u guys do realise that this is x58 right
and ts posted a 63 hours memtest86

10 minutes linpack is a breeze
and for prime???.. errr this is ram stabiliy test not proc...
for x58 only have to make sure the qpi is stable via memtest
its not as complicated as 775 testing for failure of gtl etc

[cstkl1]

Thats only your opinion though. 1600 C7 vs. 2000 C7 is a HUGE difference to alot of people here. And many are willing and able to pay those high prices for the performance.

not that much bro
http://www.xtremesystems.org/forums/...&postcount=584

unless its 200bclk like the coolaler review
i think this is what he meant

[dinos22]

if u guys think this is bad
here is my conversation with convergent ( local distro for ocz) to the sales guy name melvin

cstkl1: hi i would like to know the pricing of the ocz blade and eta
melvin: i am sorry its not in our pricelist
cstkl1: i fully anticipated that and would like you to order it for me
melvin: ok so this is the 2000mhz CL7??
cstkl1: yup
melvin: y not go for our other cl7's?? platinum etc
cstkl1wtf)... what cl7's are u talking about
melvin: we have some 1333's cl7's etc
cstkl1: eh i would like to run 2000mhz Cl7 at 1.65v 1N
melvin: u cannot run 1N on x58..

here is where i put down the phone

back to the topic..
does anybody know when the gt's will be on sale officially??
dont mind paying for these
somebody in singapore managed to get one magically.

ROFL looks like this sales rep knows his

[cstkl1]

http://www.xtremesystems.org/forums/...&postcount=594

almost there with 2000mhz CL7 but bclk 200 1600 7-8-7-1t

[jcool]

u guys do realise that this is x58 right
and ts posted a 63 hours memtest86

10 minutes linpack is a breeze
and for prime???.. errr this is ram stabiliy test not proc...
for x58 only have to make sure the qpi is stable via memtest
its not as complicated as 775 testing for failure of gtl etc

Lol.. memtest is nice, and I always use it, but linpack will freeze after 10 seconds if CPU VTT voltage is too low (for example) while memtest will happily pass.
Do you even have a x58 setup?

[Yellowbeard]

I don't know if seeing Dinos in a G-string is desirable..............

That would probably depend on how long you had been out to sea

[Oliver]

They are using the new 50nm Elpida ICs.

sry for late reply , but thread is still going he he.

ya they do, now there just need to be good yields with these ICs....

would be interesting to see how far sec hcf0 which is really crazy binned will go cl7...

keep pushing it

[cstkl1]

Lol.. memtest is nice, and I always use it, but linpack will freeze after 10 seconds if CPU VTT voltage is too low (for example) while memtest will happily pass.
Do you even have a x58 setup?

a few actually
do u??.. cause u obviously dont realize that linpack doesnt stress i7 much around 50 percent on average on all cores
try multiple orthos on each cores for small fft priority 10
try it..or in ure case if u dont have it.. get it.. kekeke

[jcool]

Lol - try LinX (latest version) or run IBT 2x, there you go, 100% with HT enabled, and 5-8C higher temps than prime blend (on water).
ALso you obviously have no clue that SmallFFT isn't the most stressful test for i7 - due to the IMC, blend actually makes the CPU run a few degrees hotter than SmallFFT. Observed by me and a few other people as well, somewehre in the middle of the Gigabyte-EX58 thread.
Next time try to fool with someone who doesn't run 3 different i7 rigs

[cstkl1]

Lol - try LinX (latest version) or run IBT 2x, there you go, 100% with HT enabled, and 5-8C higher temps than prime blend (on water).
ALso you obviously have no clue that SmallFFT isn't the most stressful test for i7 - due to the IMC, blend actually makes the CPU run a few degrees hotter than SmallFFT. Observed by me and a few other people as well, somewehre in the middle of the Gigabyte-EX58 thread.
Next time try to fool with someone who doesn't run 3 different i7 rigs

fool u.. ??
this is not a pissing contest
didnt know u can do that i guess run two instances of linx and split the rams to half??

ok read the thread
noticed how u found out a few days ago..
is there any more reason enlightenment that u want to share in detail??

[jcool]

fool u.. ??
this is not a pissing contest
didnt know u can do that i guess run two instances of linx and split the rams to half??

ok read the thread
noticed how u found out a few days ago..
is there any more reason enlightenment that u want to share in detail??

There is a lot you didn't know, and until a few days ago I shared many of your beliefs (SmallFFT being the most stressful etc) but they just don't hold true with i7.

What I didn't like is you claiming that a few passes of memtest would be enough to determine i7 QPI/memspeed stability, therefore spreading false information.

Like I said, from my own testing experience with i7 and 6/12GB of ram, linpack will freeze after 30s if your IMC can't handle the speed/timings/whatever while memtest will still run fine (even if you only run on 50% CPU).

Attached is a LinX run using 64bit mode and 11GB of ram - you can't stress i7 more than that. CPU is watercooled and ambients are low, on a stock cooler this would be >90C. Prime blend reached 61-62C after 53 hours with the same settings, if I'd have run linpack longer it would peak at around 67-68C.

Also the rams get blistering hot after only 20 minutes of linpack, while they are still cool to the touch after 53h of prime blend (ran custom inplace tho)

So linpack is the way to go with i7 - prime used to be better at detecing mobo instabilities, but since the NB doesn't have to do a lot on x58 it hardly matters any more, it all comes down to the CPUs really.

[cstkl1]

post a screenshot of task manager on loading of all the cores
how u did u manage to get it to 100%

just tested
linx.. on a 3gb kit
its way less and doesnt load all cores to 100%
are u using 0.5.4??
and is still cooler than small fft
will post u the linx in abit running on diagnostic mode to get the most mem i can

ok got it..
testing now..
although my aircon is not at full blast will post comparison

[jcool]

Well, running linpack on 2Gb (Vista eats 1GB of your 3) is of course way less efficient than running it on 5 or 11GB... but it should still be at least on par with prime if run at 100%.
Dualist (author of linx) said he discovered a way to force linpack running 8 threads in 1 instance even when there are only 4 physical cores, so we'll be rid of the dual instance run soon enough.

[cstkl1]

Well, running linpack on 2Gb (Vista eats 1GB of your 3) is of course way less efficient than running it on 5 or 11GB... but it should still be at least on par with prime if run at 100%.
Dualist (author of linx) said he discovered a way to force linpack running 8 threads in 1 instance even when there are only 4 physical cores, so we'll be rid of the dual instance run soon enough.

ok here is just 10 mins of linx at room ambient of 25C
the orthos priority 10 was run at 21C

orthos + memtest hci + 4 passest of memtest86


vs linx


do u want me to run an hour.
when i was testing intel 775 e8600 if i can pass orthos priority 10 stable for 1-3 hours it will guarantee my vcore voltage for ibt
for mem for 775 is the 4096 test .. go read the rampage extreme which i posted about
for 775 gtl try the blend and custom priority 10 1min... in rampage extreme
this ensures ibt run on 4gb on 775 up to 100 passes tested.. thats roughly 8 hours
also the difference is temp would be something in the region of e8600 4.25ghz at 1.2625 IBT 68C /e8600 4.5 at 1.375 orthos at 69C

for this i7 i will stick by the orthos priority 10 small fft and memtest86.
single linx would allow me to run 3.8ghz at 89C at 1.224
orthos priority 10 windows diagnostic mode is 94C at 1.24v (1.24 bsod) until the computer will reset with message cpu too hot

anyway how long more u want me to run linx??
what i dont like about ure statement is ure so sure that u have tested every possible test
take note this is with intel stock cooler

[StimulatedBoy]

That would probably depend on how long you had been out to sea

Dear Mr. Yellowbeard,

Any news on when these will be available? Eagerly awaiting or just plain scared of seeing Dinos22 in a G-string? Still haven’t figured out my preferences 100% in this case but seems to be leaning towards first alternative. Then again, you’ll never know and we might miss out on something here...

[dinos22]

damn it ppl stop picking on me
my arse is not THAT hairy

[Tech10]

Anyone hear of these? ( Geil ) Website here http://www.geil.com.tw/products/showSpec/id/149
I was searching the net and stumbled on this page. Looks like nice ram and the heat shield appear to be able to be removed for water cooling. To bad ram couldnt be sold with or without heat spreaders.


GU33GB2000C8TC | 2000MHz 8-8-8-28 3GB (1GBx3) Triple Channel Kit
GU36GB2000C8TC | 2000MHz 8-8-8-28 6GB (2GBx3) Triple Channel Kit
GU33GB2000C9TC | 2000MHz 9-9-9-28 3GB (1GBx3) Triple Channel Kit
GU36GB2000C9TC | 2000MHz 9-9-9-28 6GB (2GBx3) Triple Channel Kit

[DonFaygo]

When are the corsairs being released?

[TRANCEFORMER]

well HEXUS says that it will be released by the end of the month and likely for $500( i really hope that is true )

also there will be 1,866 CL7 edition launched soon after


LINK: http://www.hexus.net/content/item.php?item=16781

[DonFaygo]

just have to decide whether to get the bloodrage or wait for the evga ftw board now, lol