The E0 Q9550 SLB9V thread (finally!)

[vws]

Monza, I'm very embarrassed to say it's an ECS G33T-M2 board that I used to bench all these. I got the board virtually free from a Q9550 bundle deal back in Nov. last year, when I luckily scored that L835B078 chip. According to my inquiry with the ECS engineers the board supports CPUs of designed power no greater than ~95W, which means it can not sustain highly OC'd quads through long stress tests. That's why I mostly resort to short benches such as OCCT standard (not Linpack) 1hr tests. Memtest86+ is fine, but with the generic BIOS (with no controls of vCore, vNB, vTT, GTL, clock skews, etc., i.e. mostly by raw settings) I am only able to test FSB up to 453 MHz and memory up to 545 MHz (in 5:6 ratio). But I do try my best to optimize whatever is available to me. The highest bench-able speed for OCCT is FSB 442 / memory 531. Benching at this speed the Q9550's droop ~0.07v off their VID values (the board automatically sets vCore by stock VID), but the three Q9550's that I decided to keep have all passed the OCCT 1hr benches.

[monza1412]

Monza, I'm very embarrassed to say it's an ECS G33T-M2 board that I used to bench all these. I got the board virtually free from a Q9550 bundle deal back in Nov. last year, when I luckily scored that L835B078 chip. According to my inquiry with the ECS engineers the board supports CPUs of designed power no greater than ~95W, which means it can not sustain highly OC'd quads through long stress tests. That's why I mostly resort to short benches such as OCCT standard (not Linpack) 1hr tests. Memtest86+ is fine, but with the generic BIOS (with no controls of vCore, vNB, vTT, GTL, clock skews, etc., i.e. mostly by raw settings) I am only able to test FSB up to 453 MHz and memory up to 545 MHz (in 5:6 ratio). But I do try my best to optimize whatever is available to me. The highest bench-able speed for OCCT is FSB 442 / memory 531. Benching at this speed the Q9550's droop ~0.07v off their VID values (the board automatically sets vCore by stock VID), but the three Q9550's that I decided to keep have all passed the OCCT 1hr benches.

ouch mate, I feel your pain there...but I must say that 442 fsb in a low profile motherboard is an impressive result.

[Marvin]

Hi, just jumped to 1.38 chipset voltage and 1.26 vcore.

[ViViD]

Realy Great Chip You Got M8
1.26 Vcore It's Damn Low

Great Work And Great Batch Number

[Boulard83]

My last 1hr prime on a HOT summer day

P5Q fsb @ 450
WCed Q9550 @ 3.825ghz
4g Pi-Black @ 900mhz 4-4-4-12
BFG GTX285 OC2
PC P&C 750w

[dekruyter]

Boulard83, what are your voltages/bios settings to get that overclock...Your cpu is below the VID on CPUZ/ID. Just curious.

CPUv
CPUt
CPUpll, GTL's, etc.

[Boulard83]

Boulard83, what are your voltages/bios settings to get that overclock...Your cpu is below the VID on CPUZ/ID. Just curious.

CPUv
CPUt
CPUpll, GTL's, etc.

BIOS settings
Vcore : 1.215 for 1.208 under load.
GTL : .65x
FSB term : 1.56
NB : 1.36
SB : 1.10
PCIe : 1.50
Dram : 2.00

[dekruyter]

Boulard83,

So you undervolted the Q9550, to get 3.8.................amazing

But, you upped the CPUv termination (vtt/vfsb) from ~1.2, to 1.5, isn't that a bit high? Very high? Is this how you were able to underclock the cpu and still get 3.8?

what about Pll?

I assume you needed that high mch to overclock the ram?





..

[Boulard83]

BIOS settings
Vcore : 1.215 for 1.208 under load.
GTL : .65x
FSB term : 1.56
NB : 1.36
SB : 1.10
PCIe : 1.50
Dram : 2.00

My error ... SORRY !

Vcore : 1.215 for 1.208 under load.
GTL : .65x
FSB term : 1.36
PLL : 1.56
NB : 1.36
SB : 1.10
PCIe : 1.50
Dram : 2.00

Im still tweaking this OC few volts may be lower... i need TIME ... my goal is 4.08 24/7.

[dekruyter]

ahhhhhhhhh, that's better. I was going to say that a termination of 1.5 for a 45nm CPU is NOT GOOD, and death may be nearby; actually, 1.36 is also still quite high, no? I'm not sure why you're working under your VID.....running @ the VID (or slightly above) is good for the chip, no? Then you might be able to lower the termination? Just curious if my thinking is right on your overclock, and how it might relate to mine on an EP45-UD3R.

[Boulard83]

ahhhhhhhhh, that's better. I was going to say that a termination of 1.5 for a 45nm CPU is NOT GOOD, and death may be nearby; actually, 1.36 is also still quite high, no? I'm not sure why you're working under your VID.....running @ the VID (or slightly above) is good for the chip, no? Then you might be able to lower the termination? Just curious if my thinking is right on your overclock, and how it might relate to mine on an EP45-UD3R.

Yeah i think you can be right. My best try was a 4.08 with 1.275 Vcore but this was with VERY fast bios settings and i ran a 3Dmark06 but thats all.

I think i can do 4.08 ( 480fsb ) with my VID as Vcore. When ill reach my 4.08 ill drop each volts one at the time in order to find the lowest possible for each one. Those arnt the lowest possible.

THX again.

[Marvin]

play with gtl and you will reach it.

[OC Nub]

Put my 9550 in for some more testing today, working on this now. Just over an hour blend.

[OC Nub]

Ran 2hours 0 errors, will shoot for 4400mhz next.

[andressergio]

Ran 2hours 0 errors, will shoot for 4400mhz next.

that's a hella 9550 mate now put back the 9650 and let's keep on playing lol

Cheers Tom
Sergio

[OC Nub]

Thanks Sergio. Just put it back in.

[andressergio]

Thanks Sergio. Just put it back in.

[vws]

Thanks Sergio. Just put it back in.

Hey, some people are going to be disappointed if you stop the show... keep it coming my friend! We Q9550 E0 owners can use some inspiration, too.

So is this the same chip you showcased in this post? I'm very impressed to say the least, a Q9550 so competitive and comparable to the performance level of those golden Q9650's! Wow! And that's what this thread is about!

I wonder if yours is one of the new C batches - don't recall you've posted the chip info in this thread. Care to share more details?

[lordmagnum]

Is It proven that turning on LLC will damage quicker than keeping it turned off ?

[Zucker2k]

Gigabyte EP45C-UD3R Bios F5D
Q9550 SLB8V (E0) Batch: L844B782
Aircooling: Xigmatek SDT 1283 Single Fan

More later...

[Zucker2k]

Gigabyte EP45C-UD3R, Bios F5D
Q9550 SLB8V (E0) Batch: L844B782
Aircooling: Xigmatek SDT 1283 Single Fan

More later...

[dekruyter]

Is It proven that turning on LLC will damage quicker than keeping it turned off ?

For me and my chip, without LLC, I couldn't even overclock to 3.4...at least not easily. So it's LLC for me.

[OC Nub]

Hey, some people are going to be disappointed if you stop the show... keep it coming my friend! We Q9550 E0 owners can use some inspiration, too.

So is this the same chip you showcased in this post? I'm very impressed to say the least, a Q9550 so competitive and comparable to the performance level of those golden Q9650's! Wow! And that's what this thread is about!

I wonder if yours is one of the new C batches - don't recall you've posted the chip info in this thread. Care to share more details?

It is the same chip, it is an older EO batch#L832B701. Next weekend I might try again to get higher. Any more than 4400mhz will be hard to get stable.

Very nice vcore Zucker.

[OC Nub]

I used the same template to run 4350 on this 9550 as I used for 4500mhz on the 9650 with a little less vcore, I didn't tweak to lower vcore or vtt with this run. These are the same chips minus 0.5 multi as the 9650 and have potential to clock very high on the right board. If you want to get high clocks the 9650 thread can help with settings but must have the right board for it.

[lordmagnum]

For me and my chip, without LLC, I couldn't even overclock to 3.4...at least not easily. So it's LLC for me.

What I read about LLC on a XSystems' thread realy scared me.