View Full Version : Rampage Formula and a 45nm CPU? READ THIS!
Do you use a Rampage Formula (a real one, not a flashed Maximus) with an Intel 45nm CPU? If so, this thread applies to you! Do not vote in the poll if you do not have a Rampage Formula and a 45nm CPU!
Ok, I'm sure you all know what Vdroop is. You know, that nasty little thing that causes your CPU's voltage to significantly drop when under load? I'm sure you all also know of the various motherboard modifications that people have devised to counter the effects of Vdroop, most notably the pencil mod. Some motherboards even have an option in their BIOS to disable Vdroop, the Rampage Formula is one such motherboard. However, it has a little known problem with its method of disabling Vdroop:
If you use it with a 45nm CPU, your system will be fully prime stable, but it will crash and BSOD under absolutely no load.
AnandTech was the first source to draw somewhat widespread attention to this issue. A quote from their writeup on the subject:
Finally, let's take one last real-world look at the consequences of removing Vdroop. ASUS' implementation of this feature, labeled as Load Line Calibration and included with their latest line of motherboards, is particularly worthy of our attention for a number of reasons. The first is that setting lower voltages with this option enabled actually results in a condition in which the CPU voltage under load is higher than the idle voltage. Imagine our confusion as we desperately struggle to understand why our system is Prime95 stable for days yet continues to crash under absolutely no load.
Here is a link to the whole writeup:
http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3184&p=5
When I first read the article, I disregarded it, and used LLC on my Rampage Formula with a Q9450. Things were great, fully prime stable, everything was going well. That is of course, everything but the random crashes and BSODs while the system was under absolutely zero load. It took me a while to figure out what was causing the crashes, as I was hoping it was not LLC, because it let me use a significantly lower voltage.
If you read deeply into that AnandTech writeup, they make it sound like Vdroop control is not possible to utilize on 45nm CPUs and remain stable due to technical reasons (even though their justifications are way off base). I was starting to believe that it actually might be true, until I found this:
http://www.xtremesystems.org/forums/showthread.php?p=3021605#post3021605
Apparently, the LLC feature on the Asus P5E works perfectly with 45nm CPUs, but not on the Rampage Formula. Which is just as I thought: It is possible. I did some closer reading of the AnandTech article, and it says that the instability is caused by a sort of "reverse Vdroop", or in other words, the idle voltage is lower than the load voltage.
However, I have done some testing and it appears that the delta between idle and load is much smaller with LLC enabled than without, going by the CPU-z voltage readings (inaccurate, I know). With LLC enabled, I set voltage to ~1.36750 in the BIOS which resulted in an idle reading in CPU-z of 1.352v, and a load reading of 1.360v. With LLC disabled, I had to increase the BIOS voltage to 1.40625v to remain stable. The idle voltage reading in CPU-z was 1.400v, and the load is 1.352v, the same as the idle voltage with LLC. So, with apparently the same stable load voltage without LLC, the CPU is unstable with LLC at idle. I don't know why this is, but just to make sure, I set the BIOS voltage to 1.38v with LLC and the crashing/BSODs remained at idle. But to illustrate, the voltage delta with LLC is .008v (hardly anything), and .048v without (a lot), as well as the deltas being reversed.
Clearly, there is something flawed with the implementation of LLC on the Rampage Formula, and AnandTech is wrong about it not being compatible with 45nm CPU, regardless of their fancy looking graphs. Because now the fact stands that LLC works perfectly on 45nm CPUs with the Asus P5E, which is supposed to be an inferior motherboard.
We need to find a way to jointly contact Asus regarding this issue and have them work on it in a BIOS release. I have already sent two tickets regarding this issue and have not received any response. If many people do so, maybe they will do something.
Anyway, those with the Rampage Formula and a 45nm CPU, please test with LLC on your overclock; does the system crash when idle and remain stable under load? Please vote in the poll.
Helfarch
05-29-2008, 08:42 PM
As I posted in the other thread I get that exact behaviour.
THe pencil mod worked wonders but LLC just made it crash.
Eurasianman
05-29-2008, 08:50 PM
I have LLC enabled, but my computer is never idle (F@H 24/7). From the OP's post, having LLC enabled is only bad if you let your computer idle. Did I read this correctly?
WizardofOz
05-29-2008, 09:08 PM
Just do a pencil Mod and disable the Loadline Calibration. I was watching CPUz go crazy with fluctuation in Voltages prior to making this change. Prime would system reboot constantly. With the Pencil Mod in place, the Board is now Priming without wild fluctuations, in fact the voltages don't even move when under load. Now for example I set 1.200v in the Bios and CPUz reports 1.184v with no fluctuations.
I have LLC enabled, but my computer is never idle (F@H 24/7). From the OP's post, having LLC enabled is only bad if you let your computer idle. Did I read this correctly?
Yeah but thats not the point... we want/need this fixed! And it worked properly on an inferior motherboard, the P5E!
Helfarch
05-29-2008, 09:40 PM
I have LLC enabled, but my computer is never idle (F@H 24/7). From the OP's post, having LLC enabled is only bad if you let your computer idle. Did I read this correctly?
Not just idle, low usage (web + media player etc)
Just do a pencil Mod and disable the Loadline Calibration. I was watching CPUz go crazy with fluctuation in Voltages prior to making this change. Prime would system reboot constantly. With the Pencil Mod in place, the Board is now Priming without wild fluctuations, in fact the voltages don't even move when under load. Now for example I set 1.200v in the Bios and CPUz reports 1.184v with no fluctuations.
In other words, you have exactly the same problem.
Same here I even posted a similar question:
http://www.xtremesystems.org/forums/showthread.php?t=185243
I have almost the same setup here and the same problem so I voted YES :p:
I really hope they fix this issue with a new BIOS or something. :shrug:
003
Anandtech are not wrong, the theory of droop and why you need it is sound and is exactly as described by Intel to XS reps as well as Kris and myself at the XUG event we all attended.
The basics are really simple, if you remove the droop the overshoot moves higher and as such could damage the CPU, plus the circuitry is under much more stain if you remove the droop.
Now LLC or enhanced slope can cure some of these issues if implemented correctly, in this case it sounds like Asus have an issue with the rampage which could be more hardware based over software. If you apply a hard mod keep in mind its best to retain a small droop, thus overshoot will still happen but it will not be as high as with no droop at all.
One last point here is MM measuring of droop is not good, the meter will not respond fast enough, a scope is the best way of telling exactly what is happening.
Helfarch
05-30-2008, 02:52 AM
this would make sense if the droop wasn't so severe. Is there any information about how high it could possibly spike back up?
JohnMike
05-30-2008, 03:12 AM
My Rampage is working fine and stable (Prime95 Small/Blend proof) with my E8400 (4Ghz, LLC enable), but last weekend i was playing COD4 on-line for 2 hours and on a full action scene, sound disapear/black screen and game off to desktop with a #.exe error.
Fire up the game again and no problem at all. This type of issues (multimedia/web applications also) happened to me alot when i had Asus Formula SE (random BSOD). With my ex-Q6600 GO had no problem at all, run for hours, weeks, months total stable without any issue.
I thought that with Rampage these types of issues where gone, because i find Rampage way better board than Formula SE, but seems that this weekend issue was/is an alert. :confused:
Just do a pencil Mod and disable the Loadline Calibration. I was watching CPUz go crazy with fluctuation in Voltages prior to making this change. Prime would system reboot constantly. With the Pencil Mod in place, the Board is now Priming without wild fluctuations, in fact the voltages don't even move when under load. Now for example I set 1.200v in the Bios and CPUz reports 1.184v with no fluctuations.
Is there a link with pencil Mod steps?!
I think this issue is mainly with ASUS Rampage Formula with 45nm Quads actually if someone has this board and a Core 2 Duo on 45nm and the same problem please post so far I think only NEW Quad's suffer from this :confused:
the issue is the "spike" from high load to low load and the dip when going low load to high load, there will always be a droop between set voltage and load voltage though.
When you set VID in bios, you are actually setting the max the voltage should spike to from high load to low load, not the actual idle or load voltage, this is where many end users get confused.
The main issue is multimeters do not respond fast enough to see these spikes and dips, so end users think they have removed the droop or improved the system when infact they could be causing a lot of damage.
I am off to Taiwan tomorrow, if i get a chance i will talk to a few engineers about this and see if something can be done to reduce the spikes and dips but leave the droop as intended. Remember though this is going beyond what intel designed for and as such is going to cost more to implement on a board...so don't moan if something comes out but costs quite a bit more
SureShoot
05-30-2008, 05:11 AM
Hi ,
I am totally confused here since i do not understand Vdropp and LLC very well.
I recently purchased a rampage formula and an intel E8400.
At stock speeds should i disable LLC , if so what is the correct name in the bios , so i know wich one to disable ?=)
It seems every time i buy a piece of hardware something goes wrong =/
I always do a bit of research before , i saw lot of positive reviews on this motherboard and now this hmmm
the issue is the "spike" from high load to low load and the dip when going low load to high load, there will always be a droop between set voltage and load voltage though.
When you set VID in bios, you are actually setting the max the voltage should spike to from high load to low load, not the actual idle or load voltage, this is where many end users get confused.
The main issue is multimeters do not respond fast enough to see these spikes and dips, so end users think they have removed the droop or improved the system when infact they could be causing a lot of damage.
I am off to Taiwan tomorrow, if i get a chance i will talk to a few engineers about this and see if something can be done to reduce the spikes and dips but leave the droop as intended. Remember though this is going beyond what intel designed for and as such is going to cost more to implement on a board...so don't moan if something comes out but costs quite a bit more
Tony, it never happens when at full load or when coming down from full load (so it seems). All I have to be doing is browsing the internet, or going to the bathroom, and after a few minutes, the system crashes, while it is sitting idle.
And if AnandTech is not wrong, I still don't understand why it works on the P5E. Obviously, since hard mods work, there is a way to implement it correctly. And finally, someone said this may be 45-nm quad only, or that quads are more susceptible. This may well be the case.
I still think Asus needs to address this issue. If they got it right on an inferior board, they can get it right on the Rampage Formula.
Leithfield
05-30-2008, 06:19 AM
QX9650 + LLC ENABLED = random BSOD + lockups @ idle (P95 + OCCT stable 8 hours)
QX9650 + LLC DISABLED = nil BSOD's, nil lockups @ idle (P95 + OCCT stable 8 hours)
It is the consumers choice to either enable of disable LLC; ASUS must take ownership of responsibility to address that which is none other than faulty operation.
To: SureShoot
Hi ,
I am totally confused here since i do not understand Vdropp and LLC very well.
I recently purchased a rampage formula and an intel E8400.
At stock speeds should i disable LLC , if so what is the correct name in the bios , so i know wich one to disable ?=)
It seems every time i buy a piece of hardware something goes wrong =/
I always do a bit of research before , i saw lot of positive reviews on this motherboard and now this hmmm
Don't worry on stock in BIOS find LLC which is (Load Line Calibration) it should be on AUTO so if you run stock live it or DISABLE it.
WizardofOz
05-30-2008, 06:51 AM
Is there a link with pencil Mod steps?!
Yes there is a nice photo to show you exactly what needs to be done
http://www.xtremesystems.org/forums/showthread.php?t=181921&page=16
Trying to monitor the effects of no VDROOP with a multimeter is like attempting gene splicing with a magnifying glass and a couple of nails. Without a fast scope the overshoot Tony mentions will never be seen. The VDROOP circuit is an elegant design because of its simplicity and cost effectiveness of eliminating what can be a catastrophic failure.
Properly implemented, VDROOP in no way hinders a 24/7 overclock. All one has to do is look through the DFI P35, X38/X48 threads to see the number of users successfully implementing VDROOP and C1E with their overclocks.
Brama
05-30-2008, 08:55 AM
I have LLC , speedstep and C1E enabled and I never had an instability with this BIOS configuration. I tried with an E8400 and a Xeon 3110 and absolutely I haven't any trouble at idle.
Maybe because my system changes multiplier at low loads, having C1E and speedstep enabled?
I am 450x9 at full load and 450x6 at idle.
mngdew
05-30-2008, 11:26 AM
I guess I'm another person with the LLC issue.
On my case, my QX9650 needs 1.35V(bios)/1.33V(Windows)/1.34V(full load) to be stable at 3.8Ghz when LLC is enabled. Without LLC, I have to set the Vcore in bios to 1.375V which reads 1.35v in Windows, but the voltages drops to ~1.30V under full load.
There have been several occations I got BSOD during idle, but the issue got taken care of by increasing NB voltage to ~1.50V
Properly implemented, VDROOP in no way hinders a 24/7 overclock. All one has to do is look through the DFI P35, X38/X48 threads to see the number of users successfully implementing VDROOP and C1E with their overclocks.
Trouble is, C1E does not lower the voltages on the Rampage Formula. It lowers the multiplier. And EIST dosen't even exist in the BIOS (when overclocked, at least).
Without an overclock, C1E may well lower voltage, or when voltage is at AUTO, but otherwise, it does not touch it on the RF. Asus needs to address the LLC issue. If they did it right on the P5E, then can sure as hell do it on the RF.
Zucker2k
05-30-2008, 01:39 PM
Trying to monitor the effects of no VDROOP with a multimeter is like attempting gene splicing with a magnifying glass and a couple of nails. Without a fast scope the overshoot Tony mentions will never be seen. The VDROOP circuit is an elegant design because of its simplicity and cost effectiveness of eliminating what can be a catastrophic failure.
Properly implemented, VDROOP in no way hinders a 24/7 overclock. All one has to do is look through the DFI P35, X38/X48 threads to see the number of users successfully implementing VDROOP and C1E with their overclocks.
Do you work for DFI? You know why I'm asking this question right? Needless to say, this is a model specific problem since other Asus boards with similar chipsets are not displaying this anomally.
Do you work for DFI? You know why I'm asking this question right?
I don't have a clue why you are asking that. The only reason I mentioned DFI is as far as I know they have the only boards that allow VCORE to change when the voltage is manually set and C1E is enaabled. This in conjunction with VDROOP allows for a completely stable overclock without excessive voltages at idle or under a load.
Everybody wants to moan that VDROOP hinders an overclock. When the truth is it's the method of implementation that is to blame. My post was not meant to insinuate that this particular board did not have an issue. But rather that it requires more then the ability to be able to read the instructions of a $12.00 multimeter to call into question the design of the power circuit Intel has specified.
I don't have a clue why you are asking that. The only reason I mentioned DFI is as far as I know they have the only boards that allow VCORE to change when the voltage is manually set and C1E is enaabled. This in conjunction with VDROOP allows for a completely stable overclock without excessive voltages at idle or under a load.
Everybody wants to moan that VDROOP hinders an overclock. When the truth is it's the method of implementation that is to blame. My post was not meant to insinuate that this particular board did not have an issue. But rather that it requires more then the ability to be able to read the instructions of a $12.00 multimeter to call into question the design of the power circuit Intel has specified.
Exactly -- C1E is not an option here. FYI, I have a very good Fluke DMM :)
But that is beside the point, as a DMM has nothing to do with the problem I have discovered. The fact of the matter is that LLC on the Rampage Formula is flawed, and that LLC on the P5E, an inferior board, works just fine.
Clearly Asus has the ability to make it work. Either by a hardware change, or a BIOS update. If by a hardware change, they need to replace all current Rampage Formulas free of charge, as a feature that is advertised on it as a key selling point is broken.
Another way to address the problem would be by making it possible to use C1E voltage control while overclocked, although LLC would technically still be broken.
Bottom line is that the current state of the board is unacceptable and something needs to be done about it.
TehPwn
05-30-2008, 08:23 PM
Im about to set up my Rampage Formula with my eon X3350 which I am planning on getting to 3.8GHz, I am really worried about this! You think a BIOS flash would fix it? Anyone tried? Thanks, TehPwn!
STaRGaZeR
05-30-2008, 08:34 PM
The only reason I mentioned DFI is as far as I know they have the only boards that allow VCORE to change when the voltage is manually set and C1E is enaabled. This in conjunction with VDROOP allows for a completely stable overclock without excessive voltages at idle or under a load.
Can you give me some links to see the results? Changing vcore when manually selected in BIOS is something I lost in the P4 era. I miss it so much...
Warpcore
05-30-2008, 08:55 PM
Im about to set up my Rampage Formula with my eon X3350 which I am planning on getting to 3.8GHz, I am really worried about this! You think a BIOS flash would fix it? Anyone tried? Thanks, TehPwn!
Well I wouldn't let it scare you to much.For right now just turn off LLC and you'll be fine.Also 3.6 to 3.8 is a good overclock.4 gigs might be pushing the extreme on some cpu's.Keep following this thread and others to see any updates on the information.
Yes I think a Bios flash (could) fix the problem.There are some very smart Guys here that know what's going on.So you might not want to jump right in on the very first day and try that 3.8 Ghz.
With LLC off I don't have to many problems.I've made it to 475 fsb and 4 Ghz on the cpu.But LLC is off with the Vcore at 1.42v's.The vcore dropped only .02 volts.Then I backed off for now to stock speed till I learn a bit more from the Best of the Best.
Can you give me some links to see the results? Changing vcore when manually selected in BIOS is something I lost in the P4 era. I miss it so much...
Sent you a pm.
Lestat
05-30-2008, 10:59 PM
looks to me like you boys need to turn off C1/E1 and EIST
the EPU is not active unless C1/E1 EIST or other throttling is active or you have that stupid software enabled.
Or its active whether you want it to or not because asus screwed it up (which i would never believe... nnnooo.. )
how many of you have C1/E1 EISt enabled ???
RealTelstar
05-31-2008, 04:51 AM
I haven't got my board yet and this topic is worrying me.
Could a p5e dlx be flashed to the rampage bios and in that case, would LLC still be working?
The reason I want this board is to play with the tRD.
PS: Brama which revision says on your pcb? 1.00G?
Edit: typo, tRC for tRD)
looks to me like you boys need to turn off C1/E1 and EIST
the EPU is not active unless C1/E1 EIST or other throttling is active or you have that stupid software enabled.
Or its active whether you want it to or not because asus screwed it up (which i would never believe... nnnooo.. )
how many of you have C1/E1 EISt enabled ??? I have C1E and ThermalManagement (or whatever its called) both disabled; the option for EIST disappears when you are overclocking.
I haven't got my board yet and this topic is worrying me.
Could a p5e dlx be flashed to the rampage bios and in that case, would LLC still be working?
The reason I want this board is to play with the tRC.
PS: Brama which revision says on your pcb? 1.00G? I was also wondering if it would be possible to flash with P5E Deluxe bios. Would somebody be a guinea pig? AFAIK the Rampage Formula is supposed to have some BIOS restore features in case it gets corrupted, so if it fails I think the board would not be lost. 1.00G is the newest revision as far as I've seen. It's what I have.
RealTelstar
05-31-2008, 06:08 AM
I have C1E and ThermalManagement (or whatever its called) both disabled; the option for EIST disappears when you are overclocking.
I was also wondering if it would be possible to flash with P5E Deluxe bios. Would somebody be a guinea pig? AFAIK the Rampage Formula is supposed to have some BIOS restore features in case it gets corrupted, so if it fails I think the board would not be lost. 1.00G is the newest revision as far as I've seen. It's what I have.
I'm thinking the other way around: get a p5e dlx and flash it with the rampage bios. That if it is NOT a bios problem but a hardware one.
People flashed the p5e x38 board with the rampage formula with great success.
http://forums.bf2s.com/viewtopic.php?pid=2092923
Edit: provided link
The Asgard
05-31-2008, 07:02 AM
I had the same problem with my Formula. Fine under prime but would randomly BSOD while loading windows or shutting down. This occured with both my E8400 and Q9450.
Funny thing is I have also had a BSOD while shutting down with my P5K Delux and Q9450 again with LLC enabled. Only happened once so far but might suggest an issue with LLC and 45nm Tech and not a specific board.
The formula was fine with my Q6600
I had the same problem with my Formula. Fine under prime but would randomly BSOD while loading windows or shutting down. This occured with both my E8400 and Q9450.
Funny thing is I have also had a BSOD while shutting down with my P5K Delux and Q9450 again with LLC enabled. Only happened once so far but might suggest an issue with LLC and 45nm Tech and not a specific board.
The formula was fine with my Q6600
So the P5E Deluxe suffers from the same issue, and it is not isolated to 45nm quads. How do we jointly contact Asus about this and alert them to the problem?
RealTelstar
05-31-2008, 02:42 PM
So the P5E Deluxe suffers from the same issue, and it is not isolated to 45nm quads. How do we jointly contact Asus about this and alert them to the problem?
He didnt say he had a p5e deluxe.
Oh, he said P5K. My bad. This is really aggravating because Asus is not responding to any of my support tickets about this.
motopen1s
05-31-2008, 03:20 PM
same here... waiting for a new bios version, for now - just lowered an overclock a bit
doodah10
05-31-2008, 06:23 PM
I am running a Rampage Formula with a Xeon X3350 @ 3.8GHz on air. I've had mine to 3.92GHz long enough to get a CPUZ capture of it. I've had the blue screens at idle but they are so far apart I didn't realize the board had a problem until reading this thread.
I have had these bouts where the Asus Probe alarm goes off for 3 to 7 "beeps" then stops. The culprit is the Vcore has dropped to 0.3v according to probe II but it has never bluescreened while the alarm was sounding and I have LLC disabled. I'm still using the 219 BIOS and I have the 401 but I haven't really found a thread reporting 401 changed or fixed much of anything so I've been hesitant to try it.
TehPwn
05-31-2008, 06:47 PM
Hey, doodah10, thanks for your input man, Please could you tell me the BIOS voltage and CPU settings etc, as I have the same board and chip and Cooler and even the same PSU! WOW! LOL! Anyway if you could do that for me id appreciate it man :), thanks, TehPwn!
I have created a thread on the official Asus forums regarding this issue. Here is the link:
http://vip.asus.com/forum/view.aspx?id=20080601093315468
Please anybody feel free to contribute or help alert Asus to this problem and get some kind of official response.
doodah10
06-01-2008, 06:00 AM
TehPwn,
I got to 546FSB on Auto everything except CPU ratio and FSB freq. Haven't been close since. Have done the R&R of NB cooler tube and thermal paste. Now pretty stable at 3.8GHz. Here are the only settings not auto or default on extreme tweaker page and advanced CPU page. Volts settings in ( ) are Asus Probe values with Prime95 running.
Ai OC Tuner.............Man
CPU ratio...............8.0
FSB freq................475
PCIe....................108
DRAM freq...............950
DRAM timing ctrl........Man
5
5
5
15
DRAM static read ctrl...Dis
CPU volt................1.375(1.31)
NB volt.................Auto(1.62)
DRAM volts..............Auto(2.06)
FSB termination volts...Auto(1.4)
SB volts................Auto(1.075)
LLC.....................Dis
CPU spread spec.........Dis
PCIe spread spec........Dis
____end of Xtreme teaker____
advanced CPU page
C1E support.............Enabled
All else ...............Dis
If anyone see's something wrong or risky with these settings speak up.
TehPwn
06-01-2008, 07:22 AM
Thanks doodah! So you left North Bridge Auto it seems? LOL, was that a typo with 1.8GHz meaning 3.8 there lol, thanks, TehPwn! My X3350 comes next week, so im egar to try it out!
Please do NOT change the subject of this thread.
The Asgard
06-01-2008, 03:39 PM
Oh, he said P5K. My bad. This is really aggravating because Asus is not responding to any of my support tickets about this.
Definately an issue with the P5K Delux ( P35 ). I have disabled LLC and dropped the FSB by 5 to achieve the same prime stability. Will see how it goes.
bigslappy
06-01-2008, 04:44 PM
:p >
GbrilliantQ
06-01-2008, 05:13 PM
I voted no.
I haven't had a problem running LLC, my QX9650 has been running perferctly
http://valid.x86-secret.com/show_oc.php?id=369192
4.4ghz
Don't let this thread die, perhaps it should be a sticky, if only in the intel support section.
DEFLORATOR
06-03-2008, 04:27 AM
Yesterday received my Rampage Formula. Board made in April 2008 (serial starts with 84xxxxxxx), made in China, rev 1.00G, came with bios 0219.
Installed my Q9450 C1 in it, flashed the latest 0403 bios, so far works perfect. LLC is on. For 24 hours haven't noticed any crashes or BSODs under no load (and under load either, very stable board).
My previous P5K Premium also was flawless in this respect, were no problems with LLC enabled.
I also have a Q9450 C1, and the 0403 BIOS, and a revision 1.00G board, and a s/n beginning with 84, and I have the problem badly. I sent a third ticket to Asus about the issue, and they have not responded to any of them.
DEFLORATOR
06-03-2008, 09:31 PM
So, my system is up for 48 hours already since I assembled it, must state that so far I haven't encountered any BSOD promlem under no load with LLC enabled. Using both XP x64 SP2 and Vista x64 SP1 with latest drivers and updates. I was overclocking, playing games, testing, benchmarking and leaving machine idle for many hours, no BSOD problem. In fact I haven't ever seen any BSOD on RF, maybe because I used stable settings during overclocking from my previous P5K Premium, so I knew limits of my hardware.
My config:
"Native" Asus RF rev 1.00G, bios 0403
CPU Core 2 Duo Q9450 C1 cache 12mb @ 3520mHz (440*8) @ 1.35v
Cooler Scythe Mugen
DDR2 2*1024mb PC2-8500 OCZ Platinum SLI Ready OCZ2N10661G in white slots @ 1050mhz (5-5-4-15-54) @ 2.04v
DDR2 2*2048mb PC2-8500 OCZ Platinum OCZ2P10662G in blue slots @ 1050mhz (5-5-4-15-54) @ 2.04v
MSI nv8800Ultra 768mb @ 675/1728/2160mHz
Raid-controller Adaptec ASR-3405 PCI-Ex4 128mb
HDD 2*147gb SAS Fujitsu MBA3147RC 15000rpm 16mb raid0
HDD 2*80gb Samsung HD080HJ SATAII 8mb raid0
Sound Creative X-Fi ElitePro 64mb
TV-Tuner Beholder M6 Extra MPEG-2 Encoder
DVD-RW SATA Optiarc AD-7203S
DVD-RW IDE LG H22L
Case Thermaltake Eureka Aluminum
Powesupply FSP Epsilon 700wt
Brama
06-05-2008, 06:44 AM
PS: Brama which revision says on your pcb? 1.00G?
I don't remember clearly but I think it is a 1.00G. It came with the 219 BIOS mounted.
I confirm again, never had any trouble of BSODs with LLC on.
Ghostleader
06-06-2008, 09:13 AM
I don't remember clearly but I think it is a 1.00G. It came with the 219 BIOS mounted.
I confirm again, never had any trouble of BSODs with LLC on.
Do you have a quad or a dual?
What FSB and CPU speed do you run?
What CPU voltage?
RealTelstar
06-06-2008, 09:48 AM
I don't remember clearly but I think it is a 1.00G. It came with the 219 BIOS mounted.
Mine too. Will assemble this weekend and see :)
Brama
06-09-2008, 09:28 AM
Do you have a quad or a dual?
What FSB and CPU speed do you run?
What CPU voltage?
Xeon E3110 (dual) at 460x9 Vcore 1.32 V
Guys, I am in indirect contact with an Asus rep regarding this issue.
I need a favor of everybody who has a Rampage Formula and a 45nm CPU, from both people with LLC working, and not working. The favor I need is detailed system information in the following format:
Board rev:........1.00G
BIOS:...............0403
CPU:.................Intel Q9450
Memory:...........2x G.Skill F2-8500CL5D-4GBPK
Video card:.......ATI Radeon HD2900Pro
HardDrive:........1x 250GB Seagate 7200.8, 2x 250GB Seagate ES.2 RAID 0
Power Supply:...PC Power and Cooling Turbo-Cool 860
OS:...................Windows XP Pro x64 SP2
LLC Working:.....No
Add in cards:.....BFG PhysX, ESI Juli@
mrab54
06-16-2008, 05:39 PM
Board rev:........1.00G
BIOS:...............0403
CPU:.................Intel Q9450
Memory:...........2x G.Skill F2-8500CL5D-4GBPK
Video card:.......NVIDIA 8800 GTS 512 MB
HardDrive:........Seagate Barracuda 7200.11 ST31000340AS 1TB 7200 RPM SATA 3.0Gb/s
Power Supply:...CORSAIR CMPSU-620HX 620W
OS:...................Windows XP Pro x64 SP2
LLC Working:.....No
Add in cards:.....
Shadowthor
06-16-2008, 07:18 PM
Board rev:........1.00G
BIOS:...............0403
CPU:.................Intel Q9450
Memory:...........4GB G.Skill DDR2 8500
Video card:.......NVIDIA 9800 GTX
HardDrive:........WD 640 GB
Power Supply:...Antec Blue 650W
OS:...................Windows XP Pro x32 SP3
LLC Working:.....No
Add in cards:.....
Wow... we all have the same memory!!! Could that be the problem? I will give this info to the asus rep, they've been testing with corsair memory.
bigslappy
06-16-2008, 11:41 PM
Ya'all got XP also :p
It happens to me in Server 2008 as well, so that is not the issue.
bigslappy
06-17-2008, 12:44 AM
It happens to me in Server 2008 as well, so that is not the issue.
Hmm ! that is a head scracher ? i'll see if it happening on my Vista machine
Leithfield
06-17-2008, 01:07 AM
Board rev: 1.00G
BIOS: 0403
CPU: QX9770
Memory: Corsair Dominator PC8500 2x2GB DDR2
Video card: Gigabyte 9800GX2
HardDrive: 2x WD7500AAKA (AHCI) + 2 X Samsung HD753LJ (AHCI)
Power Supply: X3 ULTRA 1000W
OS: Vista Ultimate64 SP1
LLC Working: NO
Add in cards: Auzentech XiFi Prelude 7.1
:down:LLC enabled = system instability @ idle on all current BIOS releases; also occurred with my former QX9650, and change of ram makes no difference (tested with Corsair Dominator PC8500, Corsair Dominator PC10000, OCZ9200 FlexII, and Crucial Tracer DDR2-8500).
you know I actually told you the best way to use this board without LLC but it seems my words were missed, so here goes again.
Without LLC when you set the VID in bios in theory you are setting the max voltage the board will spike to when coming from load to idle. Now i know you are saying the board is crashing while idling for a while due to LLC setting above chosen VID, this is why I am saying do not use LLC but instead properly tune the CPU OC and you will not need it.
Opening up the CPU's ability to scale FSB will help a geat deal, reducing the need for Vcore is what this is all about, hence why I suggest you test in the following way.
Pick an OC a little lower than where you want to land finally and lower vcore to the CPU till it starts to error in prime after 10 mins or so. Disable LLC and dial in the CPU with GTL then VTT if it needs any. Watch the cores that are failing and dial GTL in to these cores. The goal here is to tune the CPU with as little VCore as possible.
Once stable increase FSB a little and retest and retune if needed, keep doing this until you hit the final OC you want.
Remember with LLC disabled you are setting the VID that really is the max the voltage should spike to coming from load to an idle state, voltage to the processor under load will be lower ;).
LLC is a bull in a china shop approach IMO, it does have issues I agree, it could be engineered a lot better..I agree, but do we need it..Naah not if you spend time tuning the board.
Abducted
06-17-2008, 01:46 AM
Board rev:........1.00G
BIOS:...............0403
CPU:.................X3350
Memory:...........2x 2GB OCZ Platinum PC8000
Video card:.......ASUS 8800GTX 768 MB
HardDrive:........Seagate Barracuda 7200.11 ST3320620AS 320GB 7200 RPM SATA 3.0Gb/s
Power Supply:...Enermax NoiseTaker 600w
OS:...................Windows XP Pro x64 SP2 and Vista Ultimate x86 SP1
LLC Working:.....No
Add in cards:.....
RealTelstar
06-17-2008, 06:41 AM
Board rev:........1.00G
BIOS:...............0403 (thinking to put the 0401 first)
CPU:.................Xeon E3110
Memory:...........2x 2GB G.Skill PC8500 (white slots)
Video card:.......ASUS 8800GTX 768 MB
HardDrive:........Samsung F1 32mb (boot drive), others for storage and backup
Power Supply:...Silverstone Zeus 750w
OS:...................Vista x64 Ultimate SP1
LLC Working:.....Will tell ASAP
Add in cards:.....Dell PERC 4i (pci-e), Creative X-fi fatality
OO3 I think it is more of a quad issue than memory. Will post my findings ASAP, hopefully tomorrow.
RealTelstar
06-17-2008, 06:44 AM
I confirm again, never had any trouble of BSODs with LLC on.
Please tell me which memory do you have. If you fill the complete form would be even better.
Brama
06-17-2008, 08:29 AM
Board rev:........1.00G
BIOS:...............0403
CPU:.................Xeon E3110
Memory:...........2 x 1GB OCZ reaper 9200 HPC edition (300 ns advance skew)
Video card:.......Leadtek 8800GTS 512 MB
HardDrive:........2xSeagate Chetah 15,000 rpm SAS in raid0 with LSI controller PCIe 4x on second PCIe 16x slot
Power Supply:...570 W Tagan
OS:...................Vista 32 business SP1
LLC Working:.....YES
Add in cards:.....Creative X-fi gamers PCI, 1 x RS-232 PCI
I have enabled speedstep and C1E in BIOS
OO3 I think it is more of a quad issue than memory. Will post my findings ASAP, hopefully tomorrow.
I don't think it is limited to quads, as several people with 45nm duals also have this problem, if you look earlier in the thread. Also, my theory about g.skill memory is out the window, as now people with ocz and corsair memory also have the problem.
Tony:
I understand what you are saying. However, the issue here is that LLC works for some people and not for others, indicating there is a problem somewhere. Also, I am able to set a much lower voltage with LLC enabled and remain stable under load. With LLC disabled, I need to set a MUCH higher voltage to remain stable, and the resulting idle voltage is very high. C1E does not lower voltages when overclocking on the Rampage Formula, so that is not a solution either.
mrab54
06-17-2008, 05:07 PM
Just another snippet of info: I was not even getting a blue screen when I crashed sitting idle after coming off of either playing UT3 or prime95 etc. This lead me to believe the problem was definately something to do w/ voltages and not bad RAM etc. Now that I see others have this issue it makes a whole lot of sense.
Leeghoofd
06-18-2008, 12:20 AM
Thing is people test with prime and co and try to get the lowest value (pretty logic)... try upping ya Vcore a bit more then when it was prime stable with LLC and most will have no issues anymore (at least that's what I learned from from my X38/48 experience) but ofcourse that totally ruins the point for most of LLC... GTL tweaking can also help stability... Second thing like 003 mentioned is that on Asus boards is that the Voltage never drops with EIST enabled which also means ya CPU idles at 6x but at a too high voltage...
DEFLORATOR
06-18-2008, 01:51 PM
Board rev:........1.00G
BIOS:...............0403
CPU:................Intel Q9450 C1
Memory:...........DDR2 2*1024mb PC2-8500 OCZ Platinum SLI Ready OCZ2N10661G (Micron D9 BH-3) + DDR2 2*2048mb PC2-8500 OCZ Platinum OCZ2P10662G (ProMOS?)
Video card:.......MSI 8800Ultra 768mb
HardDrive:........ 2*147gb SAS Fujitsu MBA3147RC 15000rpm 16mb in Raid0
Power Supply:... FSP Epsilon 700wt
OS:..................Windows XP Pro x64 SP2, Vista Ultimate x64 SP1
LLC Working:..... Yes
Add in cards:..... Raid controller Adaptec ASR-3405 PCI-Ex4 128mb, Creative X-Fi Elite Pro 64mb, TV-Tuner Beholder M6 Extra MPEG-2 Encoder
RealTelstar
06-18-2008, 02:27 PM
Hmm i have got another idea. Could be the pcb. I know that there are two "versions" one made in china and one made in taiwan. Brama, I'm assuming has taiwan board, having bought it locally in my same town.
It would be perhaps useful to know it. its written in white on a side of the board, "pcb made in...".
nunor
06-18-2008, 05:32 PM
I have a P5E3 Premium and it's prime stable but when i open some sites or playing sometimes i get a BSOD/Reboot. I had to increase the vcore to avoid this. It seems that it's not only the rampage.
EDIT: I tried to open tvtuga.com with firefox(dunno why but that site reboots when the system is unstable):
LCC=Performance
vcore(bios): 1.23125v
vcore(cpuz): 1.208v
Result: Reboot in few sconds
LCC=Auto
vcore(bios): 1.23125v
vcore(cpuz): 1.192v(drops to 1.176)
Result: Browsed that site for some minutes and no reboots.
Hmm i have got another idea. Could be the pcb. I know that there are two "versions" one made in china and one made in taiwan. Brama, I'm assuming has taiwan board, having bought it locally in my same town.
It would be perhaps useful to know it. its written in white on a side of the board, "pcb made in...".
They're all made in china. Designed in Taiwan is what they do say, though.
bigslappy
06-19-2008, 12:04 AM
Hmm i have got another idea. Could be the pcb. I know that there are two "versions" one made in china and one made in taiwan.
It would be perhaps useful to know it. its written in white on a side of the board, "pcb made in...".
RealTelstar :
what does your board say the pcb was made at ?
as i have 2 of these boards & they are different , even the box they were packaged in are different
RealTelstar
06-19-2008, 03:06 AM
RealTelstar :
what does your board say the pcb was made at ?
as i have 2 of these boards & they are different , even the box they were packaged in are different
My board says pcb made in taiwan.
Now is all mounted and is impossible for me to make a picture.
The white sticker says taiwan too, but that means assembly and does not matter.
I think most of the china-made pcb went to the american market. I'm curious about the russian guy above.
I just received my new Rampage Formula from Asus that they tested with a Q9450 and my BIOS settings and confirmed does not have the issue. It says PCB made in china. I am going to send my board back to them so they can perform testing on it. Right now I am swapping the boards and will report back how things go.
bigslappy
06-20-2008, 12:00 AM
just got the email from asus that they shipped my 1st board that i RMA'ed fast turn-around they got it @ 9;00am wed & shipped today at 10;45am hopefully they sent a new board >> as the # 1 = PCI16 x slot was shorted
I will report on what i get back
;p
So far, LLC seems to be working on the new board!!! Unless I post back stating otherwise, you can assume it has continued to work perfectly.
bigslappy
06-20-2008, 12:25 AM
So far, LLC seems to be working on the new board!!! Unless I post back stating otherwise, you can assume it has continued to work perfectly.
COOL ! let us know how it goes !
DEFLORATOR
06-20-2008, 05:00 AM
So far, LLC seems to be working on the new board!!! Unless I post back stating otherwise, you can assume it has continued to work perfectly.
What bios did it have and what's the revision of the new board?
Bios is 0403, revision 1.00G, PCB made in china. Same as the old one. Still working so far, PC was on overnight. I am guessing that there must have been a bad batch of boards produced.
bigslappy
06-20-2008, 09:19 AM
Bios is 0403, revision 1.00G, PCB made in china. Same as the old one. Still working so far, PC was on overnight. I am guessing that there must have been a bad batch of boards produced.
Tell me about it !
Shadowthor
06-20-2008, 08:11 PM
let us know how it goes, if it is good and stable, might need to rma mine as well.
Yes, still going strong. I am convinced there must have been a bad batch of boards.
Shadowthor
06-20-2008, 08:23 PM
i was hoping i didn't have to rma, lol, so what did you mention in the ticket? I don't want to be out of a comp for a week, waiting for the new mobo to arrive.
I got an advance RMA, they sent me the new board and I am sending mine to them now that I received their pretested board. Waiting to hear back now on where I should send mine. In the ticket, I would just mention that you have the LLC issue being experienced by a handful of users, maybe elaborate more.
bigslappy
06-20-2008, 11:36 PM
turn around was Quick on my RMA (different reason but it might have been a LLc issue also but never got that far as the #1 PCI16 was shorted ) they shipped a board the next day after getting it .. will not know if it's a new board till next week >> FedX ground
I didn't open a ticket i just called the RMA dept explained the problem .. if need the # just ask ....
;-p
bigslappy
06-21-2008, 12:14 AM
Sheet ! turn around was SOO quick i didn't notice the board on my doorstep tonite ! .. Damn the wife didn't even tell me !
NEW BOARD in the Retail box also ! WOOT and it's the newer version as it's the newer style box like the one i got from NewEgg last week but the serial # series is the 83 like the RMA board & bios .. the newegg board is a 84 series & the newer bios
I guess i get to keep the cables,software,manuel,fan,poster from the RMA board also
..shipped direct from Fremont CA,The RMA center is in the East !
Testing tomorrow & i also got my 8 gb's of G.Skill 8500 today also looks like a BUZI weekend ! will report later
Shadowthor
06-21-2008, 07:34 PM
how do you tell its the newer version?
As far as I know, you're supposed to send back the old board, and not keep the new software and such. Basically, you swap boards, but keep your old accessories. I had to sign an advance RMA form to get the board before sending mine, and I am getting nervous because the RMA person nor the tech support people are not responding to my e-mail on where to send my old board. It was never made clear where to send it.
bigslappy
06-21-2008, 08:48 PM
As far as I know, you're supposed to send back the old board, and not keep the new software and such. Basically, you swap boards, but keep your old accessories. I had to sign an advance RMA form to get the board before sending mine, and I am getting nervous because the RMA person nor the tech support people are not responding to my e-mail on where to send my old board. It was never made clear where to send it.
NO >
the email with my RMA# said explicatly to send the board & nothing else to the RMA center as they said any else sent with the board (software,cables,& manuels & the like )would not be returned when they shipped the board back from the RMA center. which was the case as they send the new board from Fremont, Ca (the US hdqts of Asus ) which is here in the Bay Area & I got it the next day
NO >
the email with my RMA# said explicatly to send the board & nothing else to the RMA center as they said any else sent with the board (software,cables,& manuels & the like )would not be returned when they shipped the board back from the RMA center. which was the case as they send the new board from Fremont, Ca (the US hdqts of Asus ) which is here in the Bay Area & I got it the next day
I think I worded what I said in a confusing way.
I was under the impression that when you get the new board (and it comes with all the same as new retail), you take your old one, and put it in the box of the new one, and put the new one in the box of your old one so to speak, and then send back the new one, with the new accessories, because you still have everything from the old one. :shrug:
And can you possibly PM me the address to ship the board?
I really need to clarify what I need to send back, because if I can keep all the new accessories and stuff that'd be great!
bigslappy
06-22-2008, 07:54 PM
003 PM sent
Razmatazz
06-24-2008, 08:57 PM
I have a P5E3 Premium and it's prime stable but when i open some sites or playing sometimes i get a BSOD/Reboot. I had to increase the vcore to avoid this. It seems that it's not only the rampage.
I have also expierenced the very same issues as the Rampage Formula owners, but on a P5E3 Premium. I noticed it more when I was running 4x1GB memory. The description as to what people have been expierencing is exactly what other P5E3 Premium owners have been dealing with. The information about LLC has been very informative and helpful to others as well. Thanks
nunor
06-25-2008, 04:20 PM
I have also expierenced the very same issues as the Rampage Formula owners, but on a P5E3 Premium. I noticed it more when I was running 4x1GB memory. The description as to what people have been expierencing is exactly what other P5E3 Premium owners have been dealing with. The information about LLC has been very informative and helpful to others as well. Thanks
LCC=Auto seems to fix the problem here. Now i'm runing my Q9450 @ 3.2GHz 1.13v(load) and it doesn't reboot but if i set LCC to performance even with 1.22v it reboots immediately.
Shadowthor
06-25-2008, 09:38 PM
LLC = Auto usually means disabled.
SoLoR
07-11-2008, 12:44 AM
This seems to be fixed with 0406, 1.3065 in bios, 1.288V real, no matter if under load or idle on my Q9450. To bad this bios have some nasty bug with 3rd party sata controlers, so its not for everyone...
dhc0329
07-17-2008, 09:10 PM
Here's the moral of the story at the end...Provided that you have a good MB allowing you LLC with no BSOD and crash, go ahead and enjoy the LLC with the following conditions:
1. Disable LLC - if you are a borderline overclocker with high CPU VID (having high vcore as maximum voltage under bios) because LLC tends to violate CPU VID going way over
what you have set as a maximum vcore under bios without even you noticing it as it wouldn't show under CPU-Z or Everest. Intel provides about 1.36v as
a maximum threshold, and if you set 1.4v (real voltage around 1.36) under bios, LLC will make it go higher feeding exccessive voltage to your CPU violating this
Intel threshold. This is really stressing your cpu.
2. Enable LLC - if your CPU VID is lower than (let's say) 1.3v. Even at its peak it's highly unlikely that excessive voltage would go higher than 1.36v. I would say the advantage of
having this feature on is to supply stable power at load and bring down your idle voltage substantially.
bigslappy
07-18-2008, 03:45 AM
bios 0406 seems good real good ( BUT not as many choices in setting ddr2 as ealier bios's ) everything else seems OK BUT the external raid issue (0406)it's bad let's hope they come up with a better one soon as i can only set ddr2 in 2 choices & niether is ddr2-1002 which is what i want for my GSkill's ddr2-1000's & i'm running CPU Voltage at lower then 1.3 soo i've never had the LLC issue
ThugsRook
07-18-2008, 06:15 AM
you know I actually told you the best way to use this board without LLC but it seems my words were missed, so here goes again...
i have notice this same behavior on my P5Q... high load fully stable but random idle and low load instability with LLC enabled. with LLC disabled (and a slight vcore bump) the problem goes away.
:yepp:
vBulletin® v3.7.0, Copyright ©2000-2008, Jelsoft Enterprises Ltd.