Live MB OC Report :: ASUS Rampage Extreme

[sup3rfly]

yo bro...

its a gtl issue afaik.

a lot of stability on this board in the end points to gtl being the cause.
using a few orthos custom and silly tricks and ibt...

so now trying to understand whether skews can fix this or not.
setting the cpu delay to around 200 shows there is other ways to fix this.
but still inconculsive

weird other ddr3 boards was not at fussy but other ddr3 boards cant do 10500 on everest read for 400fsb at 1.25vnb.

but for 4x1gb i think will just give up

this board at 500fsb 1600 CL7 is already damn fast at only clock skew moderate vnb 1.25v is just a bit slower than 2000mhz CL8 trd 7 clock skew
strong vnb 1.6. still cant understand this part.. cause in normally u expect the difference to be big .

yo :P
just got my RE... well i find a way to prevent that no display to happen, press DEL when u see its detecting hdd...confirm wont get no display... well i see in the bios there is 4 cpu gtls... for dual core i will need to adjust the 1st 2 rite??? the other 2 should be for quadcores i suppose?

[cstkl1]

read the previous post bro..
mentioned about the dram vtt as one of the causes of the hanging issue

one thing
i definately dont think u need that much vnb to run 2000mhz CL8

reason is this
at 500 fsb
strap 333 /moderate 1600 CL7 1T is giving performance around read 11200 for everest at 1.25vnb vdimm 1.77 tRD 8

strap 400/strong 2000mhz CL8 1T is giving performance around read 11600 for
everest at 1.6vnb vdimm 1.9 tRD 7

just doesnt make sense.. so something else must be the problem that i am
upping the vnb to compensate since write and copy performance is the same.

also noted was the gain i get from 1866 1.8v going up to 2000mhz 1.9v both at CL8 tRD 7..
the gain is disappointing for read .

[vega22]

PLL definitely did not help me.
So far, only i reach solid stability @425 Fsb/1701mhz by the +20 NBGtlref trick and slowing performance to PL7, AI C. T. Strong. It´s important too the trick SB +1.5V = PLL, both at 1.52 in my case.

EDIT: strap 333

[vega22]

I want to ask something here to. I'm currently using on Rampage Formula (flashed P5E X48) vNB 1.51v and 0.63 NB GTLREF. I know, that by the CPU the VTTxGTLREF must be less than 1.0 (so, the Vref must be less than 1.0). So, what is it about the NB Vref. I'm asking, because my stability depends on NB Vref. So, even for benching i have to use vNBxNBGTLREF>1.0 Is there any problem NB Vref to be more than 1.0?

As you just read, i´m too in q9550/G.skill and my stability too depends on NBGtlref. Very interesting.

[Solarfall]

damn so close to 6.2 ghz.. i defineatly need some more bruter power than my cascade

[ToyTen]

Hi Solarfall, GG there for cascade cooling

I own a RE and the same E8600 cpu batch as you and I was wondering of your max fsb?

Also I'm actually testing the 0801 BIOS version and previously the 0701. Do you think I should test the one you played with here ?
I tested the 0401 and dispite the good feed it has, I can't boot OCing with it Setup is straight back to default at reboot

So far I think I am cpu limited for the fsb

My goal is to get rocking setup for Super Pi 32mo btw.

Any help would be great

'wish you get some ln² asap
ToyTen

[Solarfall]

Hi Solarfall, GG there for cascade cooling

I own a RE and the same E8600 cpu batch as you and I was wondering of your max fsb?

Also I'm actually testing the 0801 BIOS version and previously the 0701. Do you think I should test the one you played with here ?
I tested the 0401 and dispite the good feed it has, I can't boot OCing with it Setup is straight back to default at reboot

So far I think I am cpu limited for the fsb

My goal is to get rocking setup for Super Pi 32mo btw.

Any help would be great

'wish you get some ln² asap
ToyTen

well i've tested all the publically available bioses and found that the one im using now works best for me.. about the max fsb im not too sure cos i just got it but i hear that the previous owner was able to bench it with +6.4ghz clocks soo that would be 640+ fsb.. this particular patch of E8600 scales well with high volts and cold temps (so i've heard and is seems to be true too)..
btw what sort of volts are you using?? it would help if could post them

[spoof]

It´s important too the trick SB +1.5V = PLL, both at 1.52 in my case.

I didn't get this? So far, looks like 465 is absolute max on my MB, for 467 higher voltages doesn't help. I'm wondering what should we do ...

[ToyTen]

well i've tested all the publically available bioses and found that the one im using now works best for me.. about the max fsb im not too sure cos i just got it but i hear that the previous owner was able to bench it with +6.4ghz clocks soo that would be 640+ fsb.. this particular patch of E8600 scales well with high volts and cold temps (so i've heard and is seems to be true too)..
btw what sort of volts are you using?? it would help if could post them

My max fsb under auto cascade(propane/CO² @ -60°C approximatively) is 651 with this cpu, using 1.5vFSB and NB on air, (1.85vNB... and on a x48 Premium board) don't know if i should push the vFSB furtherer since this RE allows me to do so?

It completes a spi32mo at 620 (PL9)cpu under D-Tek Fusion & nb under Swiftech mcw30. But couldn't do better atm with following settings.

Generally I use this setup(cpu/nb watercooled):
- ai OC tuner: manual,
- OC from cpu/memory: auto,
- LLC enable(should I turn it off ??),
- 1.8vNB+ and DDR is at divider 5:8, strap333 cas7. PL9~11,
- minimum/max vFSB: 1,22/1.5v,
- vPLL/vSB 1,5v/vSB1,05: lowest,
- CPU/NB GTLRef's: +10~40mv(0/2/3)and auto for (1),
- no skew at all or all auto excepted the CPU skew: 300

thx

[Solarfall]

My max fsb under auto cascade(propane/CO² @ -60°C approximatively) is 651 with this cpu, using 1.5vFSB and NB on air, (1.85vNB... and on a x48 Premium board) don't know if i should push the vFSB furtherer since this RE allows me to do so?

It completes a spi32mo at 620 (PL9)cpu under D-Tek Fusion & nb under Swiftech mcw30. But couldn't do better atm with following settings.

Generally I use this setup(cpu/nb watercooled):
- ai OC tuner: manual,
- OC from cpu/memory: auto,
- LLC enable(should I turn it off ??),
- 1.8vNB+ and DDR is at divider 5:8, strap333 cas7. PL9~11,
- minimum/max vFSB: 1,22/1.5v,
- vPLL/vSB 1,5v/vSB1,05: lowest,
- CPU/NB GTLRef's: +10~40mv(0/2/3)and auto for (1)/auto,
- no skew at all or all auto excepted the CPU skew: 300

thx

maybe you could drop the nb volts to 1.7v and bring up the fsb voltage to around 1.6v thats what i usually have those 2 voltage settings at.. i also have the load line calibration option enabled it helps to keep the vdroop in check when your stressing your cpu

[ToyTen]

Oki, I asked about the LLC coz some of us had probs with it not me but still I wasn't so sure of it to bottleneck me.

I start testing on your advice for vNB/vFSB right away.

Yet i passed 16 loops of the 32mo at 630x7 PL9 strap400 twister moderate but not more. Thank you

[cstkl1]

Oki, I asked about the LLC coz some of us had probs with it not me but still I wasn't so sure of it to bottleneck me.

I start testing on your advice for vNB/vFSB right away.

Yet i passed 16 loops of the 32mo at 630x7 PL9 strap400 twister moderate but not more. Thank you

hmm hope this might help u or may not

this what happens when i am upping my multi

fsb 466 strap 400 strong 1866 1t vdimm at 1.8v multi 8.5

memtest 5 pass stable to confirm vnb@ 1.33vnb and vdimm @ 1.8vnb
ran orthos/memtest hci/memtest86 5 passes/ IBt perfect

multi 9.5 vcore found.
memtest failed
up sb 1.05 - 1.1 , sb 1.5 - 1.55 , pll to 1.55 dram vdimm to 1.83
pass everything...

so i guess u would have tried the high fsb clocks at low multi first and then upped the multiplier. so this might help

[ToyTen]

Thank you cstkl1, I'm not trying i get it more stable than the time of a Super pi 32mo here, stability absolutely can't be found at such voltages for every day use, it would damage it too fast i guess.

the point was to complete it at the moment:


fsb630 PL9 was the goal. Now I can prepare the rig for ln2 benching. It might probably go higher in fsb, I hope so.

I actually need 1.501vFSB / 1.79 vNB / 2.26vDDR to acheive this, there's no way I can keep it for everyday use.

[Solarfall]

hey ToyTen .. what ram sticks do you have??

[ToyTen]

hey ToyTen .. what ram sticks do you have??

Cell shock 12800 7-6-6-18 @ 1.9v

[Solarfall]

Cell shock 12800 7-6-6-18 @ 1.9v

cellshocks FTW.. btw im pretty sure that when you find a good voltage settings 640fsb is doable with that chip.. i was able to reach 642fsb with my previous e8600 and it was not as good clocker than what i have now, i still have to test the max of this one when i find the time

[ToyTen]

I got those results under water(NB and cpu):




Like 'hem Cell Shock Just too vDDR needed.

[Solarfall]

KICK ARSE !!! ToyTen.. i knew that cpu had few more mhz in it

[ToyTen]

Thx mate

Can't wait to test it deeper under ln2 in early January. On this mobo i might improve it

:wOOOt:

[rob2k]

You need to file between both holes to create a larger hole? I was going to say because as is the peg moved to the inner positions will only go in at an angle, the pushpin won't go into the board vertically straight. Do you have a picture of your CNB-S1L as filed? I'd just be interested to see the result. Thanks again.

Hi Aln688, how's it going ?. I had to file between the holes first to get my round file in and then at the inner most hole by about a 1mm at each end to reduce the hole spacing ......

And fitted .......

Add small washers to prevent the spring going in the hole making the clamping forces uneven.

CNB-S1L = 53 degrees : SLF-1 = 44 degrees (on mid fan speed setting CHA FAN2)

[KanGaXx]

some one know vmem read for this?

[ToyTen]

some one know vmem read for this?

you might find the reading point searching all arround the condo's and more generally arrount the DIMM's. If I do find it I'll come back and tell

[aln688]

Hi Aln688, how's it going ?. I had to file between the holes first to get my round file in and then at the inner most hole by about a 1mm at each end to reduce the hole spacing ......

And fitted .......

Add small washers to prevent the spring going in the hole making the clamping forces uneven.

CNB-S1L = 53 degrees : SLF-1 = 44 degrees (on mid fan speed setting CHA FAN2)

Hello there Rob2K, not bad, you? Thanks for the picture, I get the point now. I measured the distance between the holes on the southbridge, it's 53mm. With the S1L at its inner most points being 54.6mm, obviously a light filing on that heatsink will do the trick. I've looked around online and there really aren't any SB heatsinks that will fit this board, it's no surprise people are turning to this option. I have an HR-05 SLI/IFX in my computer parts box, I'm just reluctant to use it as it seems OTT. One item I do wonder about is the RPM on that SLF, does it sound loud? I have a slow speed 80x15mm 2000 RPM fan I could use on my HR-05 SLI/IFX, just toying with the different options, but your board with your SLF and modification looks great. Cheers.

[rob2k]

Hello there Rob2K, not bad, you? Thanks for the picture, I get the point now. I measured the distance between the holes on the southbridge, it's 53mm. With the S1L at its inner most points being 54.6mm, obviously a light filing on that heatsink will do the trick. I've looked around online and there really aren't any SB chips that will fit this board, it's no surprise people are turning to this option. I have an HR-05 SLI/IFX in my computer parts box, I'm just reluctant to use it as it seems OTT. One item I do wonder about is the RPM on that SLF, does it sound loud? I have a slow speed 80x15mm 2000 RPM fan I could use on my HR-05 SLI/IFX, just toying with the different options, but your board with your SLF and modification looks great. Cheers.

Yes, good here too thanks. Just changed my Thermalright sinks to MST-81's - not an easy thing to do with the board in situ !.

The SLF fan is rated at 28dba at its max speed of 8000rpm, compared to the 6x120mm fans in the case and the 3x40mm fans on the memory mine is inaudible spinning at 4300RPM on the chassis Q control silent setting. You could always set it to between 50-100% speed using the duty setting on one of the 3 optional fan headers for more control. I would recommend this without hesitation or the CNB-S1L if not trying to max out your system - I have to run my S/B at 1.6v.

[cheapseats]

Another strap-performance update guys - minor but useful maybe:


Read and Latency performance at 500FSB/CL8/1:2 (CT Strong)


As posted before, at 475FSB, 200strap/tRD6 and 333strap/tRD7 give equal Read and Latency with CL8, but 400/tRD7 Read and Copy are lower / Latency higher.

Unlike 200 strap at certain CL/FSB, the 400 strap does not allow lower tRD than 333 strap.

200 strap still allows CL8/tRD6 at 500FSB/1:2

Unlike results at 475FSB, CL8/tRD6 at 500FSB gives higher Read than 333 strap/CL8/tRD7. 400strap/tRD7 still has lowest Read and Copy / highest Latency.

Allowing CL8/tRD6, the 200 strap is able to give Read performance more comparable with 475FSB/CL7/tRD6 on 266 or 333 straps (EDIT: for same CPU clock).

The 500FSB/200 strap advantage is not available with CL7, as 200strap/1:2/CL7/tRD5 is disallowed before 450FSB

Consistent with a load increase, 500fsb/CL8 using 200 strap/tRD6, appears to need higher min vNB for same Large FFT (in)stability, compared to 333strap/tRD7.



500x8 / 1000 / 1:2 / CL8 / CT Strong


tRD7 / 400 strap




tRD7 / 333 strap




tRD6 / 200 strap