Broke Quad + 8GB + 500FSB+ Barrier with Max Formula

[Zucker2k]

I have been scouring the threads to see if this has been done by someone already but came up short. I've seen many posts where folks thought this was an impossible feat so I decided to share this. A picture is worth a thousand words...

PS: I'm currently running 7x500 for 24/7.
PPS: Second screenie is E8400 @ 8x533 with 8GBs.


Setup:
Maximus Formula Rev. 102G (with Rampage Beta Bios 0403)
Core 2 Quad (L741A941, Vid. 1.2625)
2 kits of G.Skills: F2-8000CL5D-4GBPQ for 4x2GB @ 1.80v (Bios set)
Cooling: Water ( Check sig.)

[Darkstormz]

care to share your settings?

[Zucker2k]

care to share your settings?

These are my settings for 3.5Ghz (7x500) and 8GB on Max Formula running with RF bios 0403:

Extreme Tweaker
Ai Overclock Tuner : Manual
OC From CPU Level Up : AUTO
CPU Ratio Control : Manual
- Ratio CMOS Setting : 7
FSB Frequency : 500
FSB Strap to North Bridge : 333
PCI-E Frequency: 110
DRAM Frequency: DDR2-1000 (1:1)
DRAM Command Rate : 2T
DRAM Timing Control: Manual
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay : 3
Row Refresh Cycle Time : Auto (for 52, can't be set manually in bios)
Write Recovery Time : 6
Read to Precharge Time : 3

Read to Write Delay (S/D) : 8
Write to Read Delay (S) : 3
Write to Read Delay (D) : 5
Read to Read Delay (S) : 4
Read to Read Delay (D) : 6
Write to Write Delay (S) : 4
Write to Write Delay (D) : 6

Write to PRE Delay : 14
Read to PRE Delay : 5
PRE to PRE Delay : 1
ALL PRE to ACT Delay : 6
ALL PRE to REF Delay : 6

DRAM Static Read Control: Auto
Ai Clock Twister : Auto
Transaction Booster : Manual

Common Performance Level [9]

Pull-In of CHA PH1 Disabled
Pull-In of CHA PH2 Disabled
Pull-In of CHA PH3 Disabled
Pull-In of CHA PH4 Disabled
Pull-In of CHA PH5 Disabled
Pull-In of CHB PH1 Disabled
Pull-In of CHB PH2 Disabled
Pull-In of CHB PH3 Disabled
Pull-In of CHB PH4 Disabled
Pull-In of CHB PH5 Disabled

CPU Voltage : 1.320 (CPU-Z)
CPU PLL Voltage : 1.50v
North Bridge Voltage : 1.65v
DRAM Voltage : 1.80v
FSB Termination Voltage : 1.50v
South Bridge Voltage : 1.050v
SB 1.5V Voltage : 1.50v
Loadline Calibration : Enabled
CPU GTL Reference : 0.67X
North Bridge GTL Reference : 0.67X
DDR2 Channel A REF Voltage : Auto
DDR2 Channel B REF Voltage : Auto
DDR2 Controller REF Voltage : Auto

Advanced CPU Configuration:
Everything Disabled

[screwtech02]

Sweet man!! I'll have to give that a go on my rig.. Any other bench's/gaming issues???

[Zucker2k]

Sweet man!! I'll have to give that a go on my rig.. Any other bench's/gaming issues???

No issues here.

[Grnfinger]

Very good!
That is some very nice bandwidth score's on the wolfdale
I might try out your setting on my quad, my vid is alittle lower than yours, how are the bandwidth score on the quad?

[Viss]

Very very nice Q6600 you got yourself Zucker2k m8

[Grnfinger]

I dont have 8GB's like you but here is my 500FSB quad



PL9 is not to cool I will try the pull-ins and see if that helps any for improved speed

[ZenEffect]

I dont have 8GB's like you but here is my 500FSB quad



PL9 is not to cool I will try the pull-ins and see if that helps any for improved speed

damn i should have kept that cpu... 500fsb drops 2 cores on mine right now

oh well, tomorrow my e7200 should arrive. it will be abused beaten and battered.

i think 333 strap gives 1200 for ram so you can do pl7

[Grnfinger]

it does and trying it now, then I will run 3dmark for some scores

[ZenEffect]

it does and trying it now, then I will run 3dmark for some scores

whats your bandwidth like? somewhere around 47.7ns latency and 15000 read?

[kup]

I dont have 8GB's like you but here is my 500FSB quad

http://i237.photobucket.com/albums/f...500fsbquad.jpg

PL9 is not to cool I will try the pull-ins and see if that helps any for improved speed

Dude, you've lost 2 cores...

[Grnfinger]

Dude, you've lost 2 cores...

Doh!
I didnt even see that, I'm running the NB at 1.53 I will up it and see if the cores come back

[ZenEffect]

Doh!
I didnt even see that, I'm running the NB at 1.53 I will up it and see if the cores come back

this is why i got the dfi... its all in the gtl control my friend

[Zucker2k]

Thanks guys....

Very good!
That is some very nice bandwidth score's on the wolfdale
I might try out your setting on my quad, my vid is alittle lower than yours, how are the bandwidth score on the quad?

These rams don' like tight timings but they're quite decent.... see below

Very very nice Q6600 you got yourself Zucker2k m8

Thanks loc.o m8. Wish I was as extreme as you... haven't messed with LN2 or dry ice yet.

[eva2000]

congrats man best i did was 4x1GB Q6600 @490FSB for Nautilus 500 H20 cooled

[ZenEffect]

zucker, what strap are you using?

[Zucker2k]

congrats man best i did was 4x1GB Q6600 @490FSB for Nautilus 500 H20 cooled

Thanks Eva m8, that vcore for 3.4Ghz+ looks sick for a 65nm quad

zucker, what strap are you using?

Zen, I'm using the 333 strap. I'm booting straight from bios, no clock-gening.

[jonny_ftm]

These are my settings for 3.5Ghz (7x500) and 8GB on Max Formula running with RF bios 0403:

Extreme Tweaker
North Bridge Voltage : 1.65v
DRAM Voltage : 1.80v
FSB Termination Voltage : 1.50v

Advanced CPU Configuration:
Everything Disabled

I won't comment on NB voltage as high since you use high FSB on 8Gb RAM, and I suppose you tried to lower a notch, right? On the other side, VTT at 1.5v is about 25% out of intel specs, so as if you were giving your vcore some 1.875v

Your vDDR voltage is impressingly low at 1000MHz, you got there great sticks. Giving them some 2.0 to 2.1v would be safe (unless the maximus overshoot them as it is known for...) and could give you some better clocks

Overall, I can only say you got a great benchmarking performance there with 500 FSB on a quad + 8Gb RAM on the X38.

But, I doubt it is the sweet spot for your setup. You have a good CPU and memory combo running high specs on low voltages. Your Q6600 will be safe with up to 1.45v real vcore and your memory could kick the 1200MHz at 2.10 safe voltage. Your CPU should easily be stable at 3.9GHz at an acceptable vcore (1.45-1.47v) which on 8x multi will give you 488FSB and 976MHz on the RAM at strap 333 or if you manage strap 266 stable you'd outpass 1200MHz on the RAM

In the worst scenario (3.9GHz CPU, strap 333, memory at 976MHz), running the memory lower than 1000 MHz, you could tighten the timings, the TRFC or at least enable static read control and some more relaxed booster settings to get better bandwidth. Even at same memory settings than your actual 500FSB performance, the +400 MHz on the CPU will compensate the small decrease in memory bandwidth.

Maybe you tried it all, but I'm certain, 8x480 to 8x488MHz is a better 24/7 setup in your case and should give you lower vNB and lower VTT

[Zucker2k]

Thanks jonny I'm trying to go slow on the setup. I am goint to try to hold it at 8x500. I know the fsbt is a little high too, and it runs fine at 1.38 (which is where I'm going to try to hold. I'm still trying to find that sweet spot. I P95 primed (blend mode) overnight and woke up this morning find that thread one crapped out after 7 hours and 41 minutes. So I'm close; I'll play with fsbt and ich&ich 1.50v which are both at stock. Oh, these rams won't do 1200, highest any of them run individually was 1120mhz.

[Zucker2k]

I went back to the store where I got my quad and picked two more:

Quad No. 1 (above): L741A941 (Vid. 1.2625) last 4 wafer digits: 0238 = Broke 500+ fsb
Quad No. 2: L741A941 (Vid. 1.2500) last 4 wafer digits: 0475 = Broke 500+ fsb
Quad No. 3: L746B022 (Vid. 1.3125) last 4 wafer digits: 0870 = Broke 500+ fsb
Quad No. 4: L746B022 (Vid. 1.2875) last 4 wafer digits: 2233 = Could hardly do 450 fsb

[Sparda]

Your Q6600 will be safe with up to 1.45v real vcore and your memory could kick the 1200MHz at 2.10 safe voltage. Your CPU should easily be stable at 3.9GHz at an acceptable vcore (1.45-1.47v) which on 8x multi will give you 488FSB and 976MHz on the RAM at strap 333 or if you manage strap 266 stable you'd outpass 1200MHz on the RAM

Mehh this is 65nm Quad Q6600 not a 45nm architecture. So usually you can pump the vcore up to 1.55V with a good cooling. Though very high vtt is not recommended. Pump up the voltage until it dies...lololol rolftomato

Still it an impressive result with a retail kits. Should be better if there a long prime or occt shots.

[jonny_ftm]

Mehh this is 65nm Quad Q6600 not a 45nm architecture. So usually you can pump the vcore up to 1.55V with a good cooling. Though very high vtt is not recommended. Pump up the voltage until it dies...lololol rolftomato

Still it an impressive result with a retail kits. Should be better if there a long prime or occt shots.

If you read well around, you'll understand that 1.45v with some vdroop bypass would overshoot for 1.55v
Anyway, 1.45v is absolutely safe, 1.55v with overshoot will give you 1.65v so still within a 10% of intel specs and should be ok for some years with proper cooling

[Speederlander]

How stable is this with prime, orthos or looped 3DMark?

[Zucker2k]

How stable is this with prime, orthos or looped 3DMark?

Run P95 blend overnight for 7hrs 41 minutes before thread one crapped out.

Edit: Will try again after some fine-tuning, got two other quads that do 500fsb easy so I'm playing with those now. 7:41 is pretty stable for me tho. Small ffts is cake walk; blend is the main thing with 8GBs as the NB is stressed more @ 1.65v+