Phenom II X6 1090T fun

S

SicKlown42012

2[H]4U
Joined
Jul 6, 2008
Messages
3,319
Just got my new chip in last night, and after a few hours, I've finally gotten to 4Ghz. It's taken 1.48v, but it's rock stable. I also upped my IMC to 2.6Ghz with a small voltage boost. Tried to go higher, but it wouldn't even post at 2.8 no matter the voltage. All in all, I'm happy. Now I don't have to worry about closing background apps when I decide to game.



[URL=http://img269.imageshack.us/i/capturezar.png/]

Uploaded with ImageShack.us
[/URL]
 
nice.. you could probably get more out of the IMC with some more voltage since your running 8 gigs of ram but for the performance different vs temp its not really worth it.. but nice overclock..
 
sirmonkey1985 said:
nice.. you could probably get more out of the IMC with some more voltage since your running 8 gigs of ram but for the performance different vs temp its not really worth it.. but nice overclock..
Click to expand...

The only way i can get the IMC higher is to take out two sticks. It reached 3.2Ghz at the same voltage as the CPU, but I need all 8 gigs.

The only problem I see is that the Core sensors are off. With Cool 'n Quiet enabled, Overdrive, HWMonitor, and CoreTemp show them idling at 22c, but the room temp is 24C. One of the review sites brought this up. My best guess is that it's actually 10c off, which means I'm reaching 54C at full load. Not bad for 6 cores at 4.0.
 
54C sounds about right, my Everest Ultimate readings for 4Ghz on 1090T were around 55C. But I got there with 1.375vcore hehe.
 
That's about what I got also. For the life of me, I couldn't get over 4Ghz without bumping up the voltage a lot. But I can get 3.9Ghz stable in prime95 and OCCT with only 1.375V.
 
god, this makes me want to order mine now :p waiting a few days sucks(guess i should just put my order in tomorrow, saw the Egg dropped the price by 10 today.....too bad i have tests all next week and the week after)
 
Nice. I can't wait to get my hands on mine but I gotta wait another week or so.

What's your CPU-NB voltage?
 
SicKlown42012 said:
1.25v, so +.10v over the auto setting, which for me was 1.15v according to AMD Overdrive.
Click to expand...

that's what I use also...nothing too extreme.

I've been out of the loop for so long since the days of the Athlon64 OC with the HT bus. Just starting to figure it out again. up to this point, I just raised the multi.
 
Since going with Black Edidtion processors with my 940, I haven't touched the bus speed. My motherboard gives me control of the IMC, CPU, RAM, and HTT using multipliers. I don't think I can go without this level of control anymore.
 
yes! BE chips are where it's at! I too went BE with the 940 and never looked back! You know what they say lol "once you go Black, you never go back!"
 
So do these new AMD CPU's require less voltage or what? Thats what I'm getting at.

Can't wait to get one to replace my Phenom II X2 rig along with a 890FX motherboard. Just need to wait for them to be in stock.
 
I have a dumb question I guess, but why is it that in the CPU-Z screenshot it says AMD Phenom II X6 1095T instead of 1090T ?
 
Regarding stock voltage, I've seen some people report undervolting and overclocking with moderate success. Something in the 1.21V-2.5/3.5Ghz (non turbo/turbo)range. Pretty impressive.
 
My 1055T is stable at 3.7 on stock volts. I haven't tried higher yet.
 
John Vuong said:
So do these new AMD CPU's require less voltage or what? Thats what I'm getting at.

Can't wait to get one to replace my Phenom II X2 rig along with a 890FX motherboard. Just need to wait for them to be in stock.
Click to expand...

Yes. AMD and GlobalFoundries added a Low-K Di-electric material in the design. Which means the Thubans and it's derivatives will leak less current, enabling higher speed with lower voltage. They still can't match Intel's process yet, but GF is working on rectifying that for the next node.
 
Hmm I think I have to disagree with that about process tech. IMO AMD's new 45nm SOI process with ultra low k dialectrics is superior to intels 45nm HKMG process and is actually on par with the 32nm node. There isn't a chance in hell intel could manufacture their 6 cores on 45nm and keep it at 125w. They can't even do that on 32nm and the peak power is a helluvalot higher than the documented 130W of that 980x. AMD and GF has worked some magic here.
 
I agree there's something inherently different about AMD and Intel's architecture, considering the same number of cores, at the same clockspeed, on the same process node suck down more power on Intel than they do on AMD, however you've got to remember Intel's IPC is so much higher, so it's probably more fair to compare a lower clocked Intel chip to a higher clocked AMD chip.

At the end of the day, my vote for the "most advanced manufacturing" is a draw.
 
flippin_waffles said:
Hmm I think I have to disagree with that about process tech. IMO AMD's new 45nm SOI process with ultra low k dialectrics is superior to intels 45nm HKMG process and is actually on par with the 32nm node. There isn't a chance in hell intel could manufacture their 6 cores on 45nm and keep it at 125w. They can't even do that on 32nm and the peak power is a helluvalot higher than the documented 130W of that 980x. AMD and GF has worked some magic here.
Click to expand...


arguements mute when you consider the power usage to actual performance.. when you consider that AMD's 6 cores while sure it uses less power then the i7 980x.. it doesnt even come close to matching the performance of the 980x...

its like saying a geo metro is better then a Corvette.. sure the geo metro gets 40 mpg from point A to point B.. but a corvette gets 30mpg from the same point A to point B in half the amount of time..

i prefer AMD over intel any day but honestly trying to pull shit out of your ass to make AMD's architecture seem better then intel's architecture is just getting old..
 
i cant equate the cost of the hardware to the amount of time it takes to simply click the "close" button on a program....i mean, have you added up how much you paid out of your own pocket in order to not close those programs so you can play a game? just curious how thats working out
 
nobody_here said:
i cant equate the cost of the hardware to the amount of time it takes to simply click the "close" button on a program....i mean, have you added up how much you paid out of your own pocket in order to not close those programs so you can play a game? just curious how thats working out
Click to expand...

Because closing programs means that they're no longer doing what I need them for. Being able to convert videos and rendering ray-traced images while also having fun gaming is more than enough incentive. No more having to choose what I want to do at that moment and just doing a bit of everything is quite nice. The fact that it's a better overclocker than my previous 940 is only some icing on the cake, so to say. If you don't multi-task, then these aren't for you. But a few of use to demand a lot from hardware and will pay for it.
 
Lol @ shit out of my ass. Where would you suggest I shit from, if not from my ass? Do you have a special place for yours? :D Anyway, my response was not about architecture, but process tech. But fine if you want to bring up architecture, would it not be remiss of you to ignore that it takes intel 30% more core logic to achieve the extra performance? We can go around in circles all day, but my ORIGINAL point stands: AMD's new 45nm node appears favorable over intels 45nm node. That's it, that's all.and without the HKMG that intel ands its band of surrogates claimed ad nauseum was critical if AMD wanted to compete at 45nm and below. Don't turn it into something its not, my comments are based on widely available data.
 
I haven't done any stability torture tests yet but as far as benchmarking WPrime 1024M, 3DMark06, Everest, and general computer usage I am seeing:
3.8GHz / 2.6GHz NB @ stock voltage
3.9GHz @ 1.35v
4.0GHz @ 1.4v

If I boot low and raise my multiplier with AOD I can do fun stuff like validate 4.0GHz at stock voltage too:
http://valid.canardpc.com/show_oc.php?id=1162086

I don't know how impressed someone should be going from a C3 965 to a 1090T but for me, going from an launch week 940BE to a launch week 1090T was a rather nice improvement.
 
DedEmbryonicCe11 said:
I haven't done any stability torture tests yet but as far as benchmarking WPrime 1024M, 3DMark06, Everest, and general computer usage I am seeing:
3.8GHz / 2.6GHz NB @ stock voltage
3.9GHz @ 1.35v
4.0GHz @ 1.4v

If I boot low and raise my multiplier with AOD I can do fun stuff like validate 4.0GHz at stock voltage too:
http://valid.canardpc.com/show_oc.php?id=1162086

I don't know how impressed someone should be going from a C3 965 to a 1090T but for me, going from an launch week 940BE to a launch week 1090T was a rather nice improvement.
Click to expand...

I can boot at 4Ghz with 1.375V and it will run fine, but it will fail prime95 after about 15 minutes. At 1.4V, it will fail prime in an hour. At 1.425V, it will fail in 7 hours.
 
You must log in or register to reply here.
Back
Top