The Unofficial ASUS P5N-E SLI 650i Board Thread

[Nevets Drassel]

Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

On a side note, is there anyway to flash back to 0202? I seemed to get better memory tweaking with that bios compared to 0401. I haven't been able to keep 1T very stable in 0401 and it was just fine using 0202. I tried to flash back but it wouldn't allow me.

Yes trhere is but you have to force it in DOS because using the ASUS BIOS flash routine does not allow backward compatible flashes.

All you need is a boot disk (no system or autoexec file), AWDFLASH and the bios.BIN file that you want to use.

Start your machine using the boot floppy and then type a>AWDFLASH NAME.BIN (where NAME is the BIOS.BIN named file like 0202.bin or 0401.bin)

If all goes well, you won't have to RMA your board due to a bad flash.

You can use either a floppy, or USB flash drive (which I prefer but do not know how to set up), or leave it as it is.

 

[dlxmax]

It looks like we might have a vdroop mod. Check here: http://www.xtremesystems.org/forums/showthread.php?p=1968396#post1968396

One guy seems willing to jump. If someone here with a multimeter could confirm his measurements I'm sure he'll test it out. I'd do it, but I left my multimeter in 'merica.

Anyway, my E4300 is able to boot into windows at 3.6ghz, and probably beyond. It's not stable, but with the vdroop mod, who knows!?

 

[Core2]

Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

On a side note, is there anyway to flash back to 0202? I seemed to get better memory tweaking with that bios compared to 0401. I haven't been able to keep 1T very stable in 0401 and it was just fine using 0202. I tried to flash back but it wouldn't allow me.

Did you run Memtest 1.7 to check your ram for any errors? Check that they seated in the slots and locked in tight. Heat can cause them to loosen or rise up. Make sure you do a full reset on the CMOS after flashing

 

[Core2]

Originally posted by: dlxmax
It looks like we might have a vdroop mod. Check here: http://www.xtremesystems.org/forums/showthread.php?p=1968396#post1968396

One guy seems willing to jump. If someone here with a multimeter could confirm his measurements I'm sure he'll test it out. I'd do it, but I left my multimeter in 'merica.

Anyway, my E4300 is able to boot into windows at 3.6ghz, and probably beyond. It's not stable, but with the vdroop mod, who knows!?

One is already done VMod

 

[zingero84]

core and adder...

I got into the BIOS and changed the memory timing as well as voltage. I also turned off the Jmicron controller and also enabled RAID.

Windows loaded very well without a single problem and I installed a few software packages as well as the required drivers.

After transferring "My Documents" from another networked computer in my house, it had frozen up at the end of the file transfer. I let it sit for a few minutes then did a hard reboot.

Upon rebooting, the hard drive was checked for errors (blue screen diagnostics). Now when I boot up, the mouse and keyboard will work for about 10 seconds of the initial desktop screen and then after that the computer completely freezes. I cannot do anything. The hard drive blinking light will blink about every second, but that?s about it. I don?t know what the problem could be since everything was fine until I added my documents to it. Should I change some more BIOS voltages? try and use the repair utility on the cd? reformat again?

 

[Core2]

I got into the BIOS and changed the memory timing as well as voltage. I also turned off the Jmicron controller and also enabled RAID.

Were you running in a Raid before you made those changes. I don't use a raid setup. I know you have to F6 and load the drivers before you install windows. I have had the JMicron controller disabled from day 1. Ryan can help you out better in that area then I can

Edit: I always use special software to transfer files that may contain sys files or files that are meant for that specific machine. Windows does not like you transfer those files. I guess they think your going to run multiple instances of windows on more than one machine

 

[zingero84]

No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

 

[Core2]

Originally posted by: zingero84
No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

I always thought that the raid had to be initiated with clean drives and the drivers had to be installed before you install windows

 

[Core2]

Wait for adder to chime in before you do anything. He seems to understand the Raid setup pretty well.

 

[zingero84]

Originally posted by: Core2

Originally posted by: zingero84
No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

I always thought that the raid had to be initiated with clean drives and the drivers had to be installed before you install windows

I did start with two clean drives and the Drivers and RAID was enabled before installing. That is not the problem. Something went corrupt after I transferred my documents from another networked computer. It was working fine for the hour before I did this after the winxpx86 install.

 

[adder1971]

Originally posted by: zingero84

Originally posted by: Core2

Originally posted by: zingero84
No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

I always thought that the raid had to be initiated with clean drives and the drivers had to be installed before you install windows

I did start with two clean drives and the Drivers and RAID was enabled before installing. That is not the problem. Something went corrupt after I transferred my documents from another networked computer. It was working fine for the hour before I did this after the winxpx86 install.

i would try to have native command queuing disabled. seemed to work for some people here. my drives don't support it so i never had issues with it. also you may be having an issue with memory timings at this point. i would set the timings intentionally loose as well as maybe bumping up your vcore to rule out any cpu instability. in bios what does hardware monitor say your cpu temp is?

 

[Core2]

Originally posted by: zingero84

Originally posted by: Core2

Originally posted by: zingero84
No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

I always thought that the raid had to be initiated with clean drives and the drivers had to be installed before you install windows

I did start with two clean drives and the Drivers and RAID was enabled before installing. That is not the problem. Something went corrupt after I transferred my documents from another networked computer. It was working fine for the hour before I did this after the winxpx86 install.

You have me confused here.

You said " got into the BIOS and changed the memory timing as well as voltage. I also turned off the Jmicron controller and also enabled RAID"

Which I understand as meaning the Raid was never enabled (or disabled from Flashing or clearing the CMOS). Right??

then you said "I did start with two clean drives and the Drivers and RAID was enabled before installing"

So when was the raid disabled?? When you flashed the 0401 bios? Maybe when you cleared the CMOS. Can you rebuild the raid in the bios?? Like I said I haven't much experience with raid setups so wait for someone that can help you better.

 

[zingero84]

When I get home tonight (730pmest... yep, long day...) I will let you know. I do remember seeing it before and it was quite cool, but I don't remember the exact value.

What timings and voltage core do you recommend? Currently everything is set to auto except the timing at 4-4-4-12-2T@2.08V.

Also, what is native command queuing and where can I turn it off/on at?

EDIT

I don't know if this might be an issue but after installing windows and some software, I change the outlet location of the tower by going to another receptacle. The BIOS battery was out during this time since I was setting the BIOS and didn't want to deal with taking it out all of the time. I received my locking up error and the blue screen HD checking only after rebooting it on another receptacle and transferring my documents. Only after seeing this error, did I remember to put the BIOS battery back in.

FYI, I did reload the same exact BIOS configuration that I had when I got XP to install in the first place. Could the battery being out and the outlet receptacle being moved had a problem with BIOS and the configuration of Windows?

 

[adder1971]

Originally posted by: dlxmax
It looks like we might have a vdroop mod. Check here: http://www.xtremesystems.org/forums/showthread.php?p=1968396#post1968396

One guy seems willing to jump. If someone here with a multimeter could confirm his measurements I'm sure he'll test it out. I'd do it, but I left my multimeter in 'merica.

Anyway, my E4300 is able to boot into windows at 3.6ghz, and probably beyond. It's not stable, but with the vdroop mod, who knows!?

thats a fantastic overclock you have there. those e4300 and e4400 are going to be somthing when the prices drop.

 

[aintitthelife98]

Originally posted by: Core2

Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

On a side note, is there anyway to flash back to 0202? I seemed to get better memory tweaking with that bios compared to 0401. I haven't been able to keep 1T very stable in 0401 and it was just fine using 0202. I tried to flash back but it wouldn't allow me.

Did you run Memtest 1.7 to check your ram for any errors? Check that they seated in the slots and locked in tight. Heat can cause them to loosen or rise up. Make sure you do a full reset on the CMOS after flashing

yep I used memtest 1.7 to test any changes I made to RAM. It ran without problem at 1000mhz and 5-5-5-15 2T. But I wasn't able to get 1T to work at 800mhz and the 0401 bios and I had it working at 860mhz and the 0202 bios.

Anyone know if my SuperPi time is good for 3.2ghz and Ram at 1000?

 

[zingero84]

Originally posted by: Core2

Originally posted by: zingero84

Originally posted by: Core2

Originally posted by: zingero84
No, I was not running RAID before those changes. That was done before I loaded the RAID driver in the Windoes setup and followed through from there.

I really have no idea what could be wrong. With the disks mirroring, does it take some time after the transfer of a large amount of files (circa 20gb) to mirror or not? I assume that this is done realtime on both disks?

I cannot even do anything in Windows to run diagnostics, so I either have to change BIOS, repair Windows using the utility (which it will probably say everything is fine), or reformat...

Would bumping up the voltages be of any use? Change the RAM timing?

I really got no clue here...

I always thought that the raid had to be initiated with clean drives and the drivers had to be installed before you install windows

I did start with two clean drives and the Drivers and RAID was enabled before installing. That is not the problem. Something went corrupt after I transferred my documents from another networked computer. It was working fine for the hour before I did this after the winxpx86 install.

You have me confused here.

You said " got into the BIOS and changed the memory timing as well as voltage. I also turned off the Jmicron controller and also enabled RAID"

Which I understand as meaning the Raid was never enabled (or disabled from Flashing or clearing the CMOS). Right??

then you said "I did start with two clean drives and the Drivers and RAID was enabled before installing"

So when was the raid disabled?? When you flashed the 0401 bios? Maybe when you cleared the CMOS. Can you rebuild the raid in the bios?? Like I said I haven't much experience with raid setups so wait for someone that can help you better.

Yes, raid was enabled, but I could not get XP to load until I used the NVIDIA driver. I started with two clean drives, turned RAID on, enabled NVIDIA controllers in XP setup, and went from there. I don't think that RAID is the problem here. I have that down pretty good now.

The problem is locking up in Windows Desktop about 10 seconds after the intial viewing of the desktop.

I'll wait on adders suggestions of timings and voltages.

 

[adder1971]

Originally posted by: zingero84
When I get home tonight (730pmest... yep, long day...) I will let you know. I do remember seeing it before and it was quite cool, but I don't remember the exact value.

What timings and voltage core do you recommend? Currently everything is set to auto except the timing at 4-4-4-12-2T@2.08V.

Also, what is native command queuing and where can I turn it off/on at?

EDIT

I don't know if this might be an issue but after installing windows and some software, I change the outlet location of the tower by going to another receptacle. The BIOS battery was out during this time since I was setting the BIOS and didn't want to deal with taking it out all of the time. I received my locking up error and the blue screen HD checking only after rebooting it on another receptacle and transferring my documents. Only after seeing this error, did I remember to put the BIOS battery back in.

FYI, I did reload the same exact BIOS configuration that I had when I got XP to install in the first place. Could the battery being out and the outlet receptacle being moved had a problem with BIOS and the configuration of Windows?

i wouldn't think so but it would be wise to double check all the bios settings.

here's a link to command queuing location. it is a feature that speeds up drive access by basically queuing a bunch of stuff and then the drive picks which data is quickest to access.

Command Queuing Location

 

[adder1971]

Originally posted by: aintitthelife98

Originally posted by: Core2

Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

On a side note, is there anyway to flash back to 0202? I seemed to get better memory tweaking with that bios compared to 0401. I haven't been able to keep 1T very stable in 0401 and it was just fine using 0202. I tried to flash back but it wouldn't allow me.

Did you run Memtest 1.7 to check your ram for any errors? Check that they seated in the slots and locked in tight. Heat can cause them to loosen or rise up. Make sure you do a full reset on the CMOS after flashing

yep I used memtest 1.7 to test any changes I made to RAM. It ran without problem at 1000mhz and 5-5-5-15 2T. But I wasn't able to get 1T to work at 800mhz and the 0401 bios and I had it working at 860mhz and the 0202 bios.

Anyone know if my SuperPi time is good for 3.2ghz and Ram at 1000?

sounds about right. superpi 1M gives the e6600< chips an advantage due to increased l2 cache. my best is in the 16.xxx range with 3.5ghz 500x7 1000mhz mem 5-5-5-12 2t. but it's not very stable for much of anything else.

 

[Genuine Imitation]

Another quick update, the evercool VC-RE is on my board but it didn't go without a fight. It took MASSIVE modification with the dremel to clear all 3 caps close to the SB ALONG with the two little silver colored things between the SB and the IDE ports. clearing the two little silver guys was an absolute biatch. But its on there. I'll try to get a good pic of it. I've got 3 more connections to make on the WC before I can fill and leak test. That will happen tonight. Tomorrow morning will be the power on event, followed by much overclocking and rejoicing

 

[Core2]

Originally posted by: Genuine Imitation
Another quick update, the evercool VC-RE is on my board but it didn't go without a fight. It took MASSIVE modification with the dremel to clear all 3 caps close to the SB ALONG with the two little silver colored things between the SB and the IDE ports. clearing the two little silver guys was an absolute biatch. But its on there. I'll try to get a good pic of it. I've got 3 more connections to make on the WC before I can fill and leak test. That will happen tonight. Tomorrow morning will be the power on event, followed by much overclocking and rejoicing

Good Luck

 

[Core2]

Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

Here is mine at 400x8 3200

and one at 450x8 3600

 

[Nevets Drassel]

Originally posted by: Core2
Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

Here is mine at 400x8 3200

and one at 450x8 3600

So it is the cache size that allows SuperPi to run faster. I got 16.109 on my ES at 3000. This being the case, is a better SuperPi time at a lower multiplier and lower OC better than a same time in SuperPi with an OC of 3600?

Do you know what effect the SuperPi times have on overall program response? Is it the standard "better SuperPi times = faster program response" or are there other factors that also control the overall performance?

Thanks

 

[Genuine Imitation]

Originally posted by: Nevets Drassel

Originally posted by: Core2
Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

Here is mine at 400x8 3200

and one at 450x8 3600

So it is the cache size that allows SuperPi to run faster. I got 16.109 on my ES at 3000. This being the case, is a better SuperPi time at a lower multiplier and lower OC better than a same time in SuperPi with an OC of 3600?

Do you know what effect the SuperPi times have on overall program response? Is it the standard "better SuperPi times = faster program response" or are there other factors that also control the overall performance?

Thanks

There are literally tons of different variables thats effect overall performance. Technically, your hard drive is the slowest part of any computer, but once you get the data off of it and into memory, that's where overclocking effects the performance. Super Pi just gives you a relative snapshot of what the performance of your pc is when running math heavy operations. Other programs might not be math heavy but involve tons of memory reads and write, which would behave much differently.

 

[Nevets Drassel]

Originally posted by: Genuine Imitation

Originally posted by: Nevets Drassel

Originally posted by: Core2
Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

Here is mine at 400x8 3200

and one at 450x8 3600

So it is the cache size that allows SuperPi to run faster. I got 16.109 on my ES at 3000. This being the case, is a better SuperPi time at a lower multiplier and lower OC better than a same time in SuperPi with an OC of 3600?

Do you know what effect the SuperPi times have on overall program response? Is it the standard "better SuperPi times = faster program response" or are there other factors that also control the overall performance?

Thanks

There are literally tons of different variables thats effect overall performance. Technically, your hard drive is the slowest part of any computer, but once you get the data off of it and into memory, that's where overclocking effects the performance. Super Pi just gives you a relative snapshot of what the performance of your pc is when running math heavy operations. Other programs might not be math heavy but involve tons of memory reads and write, which would behave much differently.

Thanks for that explanation. It clears some things us. Does SuperPi run solely in the CPU cache or is memory also involved?

 

[adder1971]

Originally posted by: Genuine Imitation

Originally posted by: Nevets Drassel

Originally posted by: Core2
Originally posted by: aintitthelife98
Looks like I've got it stable at 3.2ghz. Also I have my memory running unlinked at 1000mhz and 5-5-5-15 2T.

My current best SuperPi (1M) with this config is 18.250s.

Here is mine at 400x8 3200

and one at 450x8 3600

So it is the cache size that allows SuperPi to run faster. I got 16.109 on my ES at 3000. This being the case, is a better SuperPi time at a lower multiplier and lower OC better than a same time in SuperPi with an OC of 3600?

Do you know what effect the SuperPi times have on overall program response? Is it the standard "better SuperPi times = faster program response" or are there other factors that also control the overall performance?

Thanks

There are literally tons of different variables thats effect overall performance. Technically, your hard drive is the slowest part of any computer, but once you get the data off of it and into memory, that's where overclocking effects the performance. Super Pi just gives you a relative snapshot of what the performance of your pc is when running math heavy operations. Other programs might not be math heavy but involve tons of memory reads and write, which would behave much differently.

absolutely agreed. hdd is the weakest link. can't wait for 100gb highspeed mem drives.