SIX is Sexy! (Unboxed)

Quote:

---

Originally Posted by FlanK3r

RAMs...AM3 score is better than AM2+ in most aplliacations, example here:

---

HMmm I dont see a big difference. Only 300 points in the X-CPU bench. Thus it is the multicore bench, but we are talking here about the single-thread performance.

Furthermore, the used RAMs for the X6 system are quite good:

Quote:

---

System specs:
CPU: AMD Phenom II X6 1055T
MB: MSI-790FX-GD70 Bios v1.10b8
RAM: Corsair CMD8GX3M4A 1600 C8 at 1333-9-9-9-24 (all auto)
GPU: 4770 CF stock clocks
PSU: Corsair HX850

---

@informal:
Yes, saw that problem with the NB clocks, but as stated above, the andandtech results are still better, and these are not oc'ed.

@jcmarfilph:
Hmm ok C&Q ... could be a problem, indeed.

@informal (again ;-)
Was C&Q enabled at your run @3.2 GHz ? If yes could you try once with disabled C&Q ?

thx

Opteron146

Quote:

---

Originally Posted by jcmarfilph

CnQ was enabled too so that could affect the result.

And correct me if i am wrong, Turbo Core is just spiking on when needed so a static 3.3Ghz > dynamic 2.8Ghz(Turbo Core ON and CnQ ON)

:)

---

or the fact hes taking screenshots while its running. and yes CnQ its a bit odd to work with, ill test that when i get home, but i know for a fact in gaming, if i have my windows power profile say minimum 5% max 100%, some games only make it go up to 50-60% if i have it on balanced or power saver, only high performance lets it hit 100% all the time. and yes, there is a fps difference. WoW is so badly coded that i only used 25-30% of my cpu power, but is 100% relevant to cpu speed, so if 25% of 3ghz, does not mean its 50% at 1.5ghz, its still 25%, :down:, so i only use high performance with CnQ, and mod the profile to let it reach 5% for minimum

No CnQ in my case :). I disabled it and i use K10stat only ;). But the profile was set to 3.2Ghz for both perf. profiles so you can say my run was @ 3.2Ghz. There is still 4% IPC difference between my run and the X3 you linked(could be 2.5Ghz NB effect).

Quote:

---

Originally Posted by Opteron146

Furthermore, the used RAMs for the X6 system are quite good:

Opteron146

---

What do you mean 1333@CL9 is quite good? Good 1333 RAM should be running at CL6.

here you can see bus speed 312mhz : http://www.xfastest.com/viewthread.php?tid=41074

ridney
When you run the x264 bench you'll do it also at stock speeds right?

Quote:

---

Originally Posted by richierich

What do you mean 1333@CL9 is quite good? Good 1333 RAM should be running at CL6.

---

It's better than DDR2 ;-)
I dont expect great gains of CL6 memory. The prefetcher should do the main work for cinebench. Low latency RAM is mainly useful for games, only, because their acces patterns are not as regular.

@informal:
Thanks ... hmm maybe it could be the +4MB L3 cache ?

Quote:

---

Originally Posted by Opteron146

It's better than DDR2 ;-)
I dont expect great gains of CL6 memory. The prefetcher should do the main work for cinebench. Low latency RAM is mainly useful for games, only, because their acces patterns are not as regular.

@informal:
Thanks ... hmm maybe it could be the +4MB L3 cache ?

---

Yep,it could be the cache or something related to minor core changes AMD did with Deneb(latency of some instructions is decreased,L1 and L2 caches have slightly lower latencies overall).

Nice results!

And lovely pics. :D

Quote:

---

Originally Posted by FlanK3r

ridney: nice result in Cinebench, can u go a bit higher, exmaple with 3700-3800 MHz?

---

i can go much higher than that :up: i think at 1.375v default, i can go as high as 3.7 cinebench stable

Quote:

---

Originally Posted by radaja

ridney,nice results so far.i cant wait to get one.six cores is going to be fun and affordable:up:

thanks for showing me how to do this.this is a great feature that i had no clue about

---

thanks radaja! :)

Quote:

---

Originally Posted by Nintendork

ridney
When you run the x264 bench you'll do it also at stock speeds right?

---

yep, stock and to whatever maximum oc stable i can get :yepp:

Quote:

---

Originally Posted by -n7-

Nice results!

And lovely pics. :D

---

thanks n7 :) six is definitely sexy!

http://img109.imageshack.us/img109/5...1m01270bus.png

WOW nice 6 cores at 4.5Ghz:up: that is just awesome

That OC scream cinebench and x264 xDDD

don't be too excited gentlemen, that's on turbo :D i can't find a way to turn it off, damn!

Turbo can only be disabled on 800 mobos?

Still, at least your system doesn't crash right away, clocking that high in turbo mode.

i think these early bios doesn't have an option to disable it yet :shakes:

if turbo picks up 3 cores then i crap out

Enlighten me on this turbo thing please.

You overclocked it to what, 4.1GHz? And than turbo kicks in and raises one core to 4.5GHz?

So that makes 1M ran at 4.5GHz?

Quote:

---

Originally Posted by ridney

i think these early bios doesn't have an option to disable it yet :shakes:

if turbo picks up 3 cores then i crap out

---

EDIT: Mistaken math.

Awesome results!

This is providing some incentive for going with an 1090T as the multiplier control will allow the same hefty OCs without the Turbo getting out of control due to FSB scaling.

Quote:

---

Originally Posted by Zeus

Enlighten me on this turbo thing please.

You overclocked it to what, 4.1GHz? And than turbo kicks in and raises one core to 4.5GHz?

So that makes 1M ran at 4.5GHz?

---

Turbo adds 2.5 to the multiplier, at stock that's 500, at 275 FSB it's 687.5.

zeus, on that 1m pi run i set affinity to core#1 so that it wouldn't be bouncing off from core to core and kicking all three cores to turbo mode. normal clocks at 275HTT x14 multi is 3.85GHz but when turbo kicks in it would be 4.538GHz on 1 to 3 cores. If all 3 cores are on turbo, 1.475vcore is not enough and craps out :shrug:

sorry to disappoint you guys but i don't think i can go any higher than this unless i can find a way to disable this turbo, if i up HTT to 280 i would be getting 4.62GHz on turbo and i don't have much volts for that clock

http://img694.imageshack.us/img694/310/cb001.png

nice, 22k on CB10, thats a number worth oggling over.

Thanks for the explanation Ridney and AMD-C.

That cinebench is awesome, almost 22K!

So, allow me to do a little math: your're runnning 14x270= 3780 and once turbo kicks in 3 cores go to 2.5x270=675+3780=4455MHz?

Basically you ran that bench with 3 cores at 3780 and 3 cores randomly switching to 4455MHz, right?

(A part from the cpu) these are what i call great pics!

Quote:

---

Originally Posted by Zeus

Thanks for the explanation Ridney and AMD-C.

That cinebench is awesome, almost 22K!

So, allow me to do a little math: your're runnning 14x270= 3780 and once turbo kicks in 3 cores go to 2.5x270=675+3780=4455MHz?

Basically you ran that bench with 3 cores at 3780 and 3 cores randomly switching to 4455MHz, right?

---

nope, its 6 cores at 3870, it ONLY uses turbo if 1-3 cores are loaded, 4-6 means no turbo, which is why if exactly 4 cores are used, these chips can seem slower than older PIIs

Ah, ok now it's clear, tnx.

So 22K at just 3870? Can anyone say 24K on water? :)