Asus P5E3 Premium Wi-fi-AP

[Kazi]

Hello all,

After i defaulted the bios of my P5e3 premium my raid0 array would not boot anymore.

I think the problem is that i saved (and exited) the defaulted bios, before changing the raidcontroller setting from ide to raid. So in other words. I booted my mobo with the incorrect raid controller setting (IDE). With the incorrect setting i did not try to boot my Vista 64bit, I just booted into bios again to change it from ide to raid. But after i set it to the correct setting and tried to boot into vista 64bit it did not work. It says i have to insert bootable media and press a key.

They array seems to be fine and the array is still listed in bios as a bootable device, but the system isnt able to boot from it anymore.

Can anyone confirm that booting the mobo with incorrect raid controller settings will messup the Raid0 bootdevice (partition)?
Is this a common issue? and is here a work arround?


Kazi

[Telo]

Hello all,

After i defaulted the bios of my P5e3 premium my raid0 array would not boot anymore.

I think the problem is that i saved (and exited) the defaulted bios, before changing the raidcontroller setting from ide to raid. So in other words. I booted my mobo with the incorrect raid controller setting (IDE). With the incorrect setting i did not try to boot my Vista 64bit, I just booted into bios again to change it from ide to raid. But after i set it to the correct setting and tried to boot into vista 64bit it did not work. It says i have to insert bootable media and press a key.

They array seems to be fine and the array is still listed in bios as a bootable device, but the system isnt able to boot from it anymore.

Can anyone confirm that booting the mobo with incorrect raid controller settings will messup the Raid0 bootdevice (partition)?
Is this a common issue? and is here a work arround?


Kazi

Sounds like the array might have been corrupted, or that you may be missing a setting somewhere. Is there a way you can post the settings for your boot up sequence?

[Puttz]

MemoryC.com great guys

I third it!

[nick2006]

Hey Nick

One other thing.

With Vista, I changed my power management settings to performance & found I've been able to pull my CPU voltage back 2 notches.

Continuing to remain stable - Of Course...

Thanx so much but I've been there while trying several combinations of settings (I can't even remembered exactly how many many times I've used the memtest86 tool, just to find errors after 4th test ALL THE TIME...) but it didn't help really. Now with the settings I posted I have at last something stable !!!

The only thing that bugs me is that system always hangs and reboots with a BSOD after "waking up" from sleep mode in 64-bit Vista Ult... Someone would advise.. just don't use the bloody thing, but it seems so useful power-wise.... My mind is fried on this one...

[SoulsCollective]

Is anyone else having odd reboot issues? My system is 1hr OCCT stable and the RAM will do three consecutive Memtest passes, but I can still get odd, totally random shutdowns just browsing the web. C1E and Speedstep are off, btw.

[Razmatazz]

Is anyone else having odd reboot issues? My system is 1hr OCCT stable and the RAM will do three consecutive Memtest passes, but I can still get odd, totally random shutdowns just browsing the web. C1E and Speedstep are off, btw.

I had similar issues when running 4 sticks of ram. No problem when scaled back to 2 sticks. When running 4 sticks, adjusting clock scew improved stability but I ran into just the problem that you are noticing.

[SoulsCollective]

Interesting, I'm using 2*1GB at the moment. I'll have a look at skew, any advice on what worked for you?

[nick2006]

I'm also fully interested in skew settings that may improve stability while o/c my 1800Hmz ram modules...

If it's not big trouble, friend Razmatazz, please share with us your skew settings...

THANK YOU.

[Mykou]

Is anyone else having odd reboot issues? My system is 1hr OCCT stable and the RAM will do three consecutive Memtest passes, but I can still get odd, totally random shutdowns just browsing the web. C1E and Speedstep are off, btw.

got one yesterday , bios 204 .

don't know why , wasn't at home and rig at stock speeds at that moment lol

[SoulsCollective]

Well, I just confirmed stability (at least to my mind). Prime95 v25.4 and OCCT 2.0 infinite loop just ran side by side simultaneously for eight hours without erroring; I felt sorry for my rig looking at the page file usage.

However, reboots persist. Has happened to me twice today. For some reason they seem to occur less frequently when my rig is under full load

[nunor]

Is anyone else having odd reboot issues? My system is 1hr OCCT stable and the RAM will do three consecutive Memtest passes, but I can still get odd, totally random shutdowns just browsing the web. C1E and Speedstep are off, btw.

I have exactly the same problem as you. I run prime 400*8(3.2Ghz) @ 1.20v for 12hours 100% stable. When opening some sites (ex: tvtuga . com) it just reboots after 5 secs with any browser except IE. I increased my vcore to 1.24 and doesn't reboot anymore. It's stupid but true.

[Mykou]

this board sounds to me like beta boards ROFL

no mature bios available , maybe rev 1 should unleash its full potential .

it is a shame to hear all that ram issues even I m still lucky running my cellshock beyond their specification .

[Razmatazz]

As I tried to push FSB and Dram frequency higher and higher, I finally ran into some memtest errors. The errors were quickly present with clock skew set to advance. After I set skew to delay (I used between 100 and 250, exact setting didn't seem to matter) the errors dissappeared. With skew set to auto or normal, I would get random crashes in 3D/Games and also random crashes with simply web browser. With skew to delay, 3D/Games were stable but I would still get random crashes with web browser.

Now, it would seem to make sense that skew channel A should be set more delay as compared to channel B (channel A is closer). I have been limited by not being able to post at higher FSB so I haven't been able to push DRAM frequency in order to "dial in" clock skew (ie dial out the errors). I believe it would be ideal to have each channel set to the optimal skew but I haven't a clue as of yet what that skew would be.

This hobby is still quite new to me and my learning curve is still incredibly steep. I have learned an ton of information thanks to all of you. Now that I have a little better understanding about FSB termination voltage and CPU pll voltage I am hoping that I will be able to push the DRAM frequency and find those elusive clock skews. Perhaps dual core users would have an easier time of it. If any of you have better luck, please post your expierences.

[Mykou]

Razmatazz

my only stable settings for now are with most of them on auto , just manually set vcore vtt and nb voltage and ram too .

As I'm not overclocking ddr3 actually I didn"t play with skew settings .

to list some issues , I noticed every time I use asus express gate , then boot to OS , it is impossible to get any web site , and I have to reboot directly to vista to get them again ...

weird isn't it ...

[Razmatazz]

Razmatazz

my only stable settings for now are with most of them on auto , just manually set vcore vtt and nb voltage and ram too .

As I'm not overclocking ddr3 actually I didn"t play with skew settings .

to list some issues , I noticed every time I use asus express gate , then boot to OS , it is impossible to get any web site , and I have to reboot directly to vista to get them again ...

weird isn't it ...

I've had issues with Express Gate as well as have several others. I just simply disabled it. It's just not worth the headache and extra hassle imo.

[metro.cl]

Hello all,

After i defaulted the bios of my P5e3 premium my raid0 array would not boot anymore.

I think the problem is that i saved (and exited) the defaulted bios, before changing the raidcontroller setting from ide to raid. So in other words. I booted my mobo with the incorrect raid controller setting (IDE). With the incorrect setting i did not try to boot my Vista 64bit, I just booted into bios again to change it from ide to raid. But after i set it to the correct setting and tried to boot into vista 64bit it did not work. It says i have to insert bootable media and press a key.

They array seems to be fine and the array is still listed in bios as a bootable device, but the system isnt able to boot from it anymore.

Can anyone confirm that booting the mobo with incorrect raid controller settings will messup the Raid0 bootdevice (partition)?
Is this a common issue? and is here a work arround?


Kazi

You need to select RAID for your Sata ports in order for it to "see" it again, because RAID is not a default config.

[soapman72]

Guys,

I gave up on the 1600mhz Platinum EB Kit. Wanting to stick to 4gb and try to find a 4gb kit, I decided to try the 1800mhz reaper memory by OCZ. Hopefully the board will like this memory better than those EB sticks. So far, the results are encouraging. I popped the new ram into the black slots, set the timings, set the vdimm to the specified value listed on the ram,and made it through 2 passes on memtest and got no errors. No additional tweaking, which is alot more than I can say for the platinum EB kit.

I am running Prime95 right now, and just before writing this I got an error after about an hour. I think, however, that I just need to bump the vcore a notch or two..........we'll see. When I first set the FSB to 450, I tried running prime, and got a rounding error almost immediately. I may be a noob, but I was suprised to find that running the exact same speed, I needed more vcore than I was running previously with the fsb set to 400. Moving up 2 clicks allowed prime to run for an hour and 15 minutes before erroring.

If anyone has any suggestions on what else to try, I am all ears.

Also, I was suprised to find out that my board will not post with a performance level of 6 at 450 FSB. I guess it is due to the 2x2gb memory. I am running the clock twister on stronger though.

[soapman72]

Well,

Bumping the Vcore did not help. I raised it 2 clicks,and got an error at 1hr 15 min, almost exactly the same time as before. I am at 1.51v to the north bridge. I am going to try 1.53v. Perhapse that is the issue. Anybody know if I should start thinking about mounting some sort of fan pointed at the NB heatsink? I have lots of case fans, including 1 that blows in from the side, but it does not point right at the NB. Opinions? Because this board does not report the NB temp, I am not sure if the NB is getting too hot, or if It just needs a touch more voltage. BTW, what is the limit for nb voltage for 24/7 use with stock cooling?

[Mykou]

Well,

Bumping the Vcore did not help. I raised it 2 clicks,and got an error at 1hr 15 min, almost exactly the same time as before. I am at 1.51v to the north bridge. I am going to try 1.53v. Perhapse that is the issue. Anybody know if I should start thinking about mounting some sort of fan pointed at the NB heatsink? I have lots of case fans, including 1 that blows in from the side, but it does not point right at the NB. Opinions? Because this board does not report the NB temp, I am not sure if the NB is getting too hot, or if It just needs a touch more voltage. BTW, what is the limit for nb voltage for 24/7 use with stock cooling?

Hi soapman

IMO you may raise your NB voltage up to 1.6v with good air flow .

that said , I'm very surprised to read on anandtech site how Chris Boughton ( freecableguy) got a stable rig with E8500 over 400 fsb with just 1.35v on NB with this board .

did mention priming stable . How come ?

I'm still thinking maybe asus sent him a "good" board and not a retail one...

I saw too some people finding hat board to overvolt the NB with their multimeter . so be careful raising your nb under air is my advice.

[SoulsCollective]

I'm now managing to run 450FSB stable with my Q9450 for 3.6GHz, using 1.3 VTT, 1.7PLL and 1.4vCore (~1.38 actual). Board settings are 450FSB, 333 NB strap, 1.4vNB. RAM settings are 1.9V, CAS-6, DDR3-1440, PL 7, Moderate twist, Channel A 50ps delay, Channel B 25ps delay. Due to dividers I can only get either 1440 or 1800+, and seeing as I'm yet to get back my other 2*1GB sticks from RMA and the known issues with running 4*1GB at over 1600 there's no point in me trying to get 1800+ stable yet.

[soapman72]

I am still getting errors when running prime. I tried bumping Vcore-NO HELP. Tried Bumping NB to 1.53-NO HELP. Raised FSBT to 1.2v--NO HELP. I am wondering if I just need more FSBT voltage. I am going to try 1.28v next. I also have the PLL at 1.58. Is that enough for an E8400 to run 450 fsb? It also could be that my CPU does not want to run 3.8ghz. I wouldn't think I have hit the wall yet, as it only needs 1.264v (load) to run stable at 3.6ghz with a 400mhz fsb, but I know things ramp up quickly from there. I am at 1.296v (load) and I would be willing to go higher, but it did not seem to help.


What should I try next? I would really like to get 3.8ghz stable. If this drags on for too long, I will back down to 8x multi and run at 3.6ghz

Should I try the 400mhz strap?

[SoulsCollective]

I am still getting errors when running prime. I tried bumping Vcore-NO HELP. Tried Bumping NB to 1.53-NO HELP. Raised FSBT to 1.2v--NO HELP. I am wondering if I just need more FSBT voltage. I am going to try 1.28v next. I also have the PLL at 1.58. Is that enough for an E8400 to run 450 fsb?

To run a higher FSB you are going to need to bump both vTT and vPLL - see just above the settings I needed for 450SB stable.

[soapman72]

To run a higher FSB you are going to need to bump both vTT and vPLL - see just above the settings I needed for 450SB stable.

Finally, some progress! With the FSBT at 1.28c and PLL at 1.58v prime is approaching 2 hours without an error. My question is if I do eventually get an error, how high can I go on with these 2 voltages for 24/7 use. I thought I read somewhere that 1.7v on PLL is too much for 45nm cpu's to be considered safe for 24/7

Again, Sorry to be such a noob, but I don't want to trash my CPU.

[SoulsCollective]

Intel lists the max safe vTT for 45nm chips as 1.35V, with the strong implication being that if you go over this sudden chip aspolosion syndrome may strike, so you may want to pay a little more heed to that than most other recommended limits. As for vPLL, the common understanding is that the danger level begins somewhere around 1.75-1.8.

[soapman72]

THANK YOU! Which variable should have a greater impact on stability? I just got an error after 2hrs 5 min.........GRRRRR

I am still encouraged as bumping the FSBT to 1.28v seemed to help because that made prime run twice as long as before. I just went up 2 clicks to 1.32v

I'll let you know if it does better after this. Gotta go to sleep now, but I'll run it to see the difference.