*** Official ASUS P4C800/Deluxe (875P) Thread ***

[Xeon]

Originally posted by: DAVIDS

Originally posted by: Xeon
Interesting.

I'm running the 1013 and it's quite happy.

What kind of symptoms are they having DAVID? The boards just die?

After flashing 1013 and then resetting the CMOS, some people said their boards refused to POST. The rumor is that Asus is getting a lot of RMA requests. But the problem is not affecting everyone with a P4C800 Dlx or -E board, so I don't know what to tell you. BIOS 1013 was pulled from the US site a few days ago, and yesterday, it was also pulled from the German site which is very unusual. I can't recall a BIOS ever being removed from the German server.

Wow, this is good data.

Looks like I'll be flashing my board back to 1011.

Thanks for the information DAVIDS!

BTW, I do have the P4C800-E Deluxe so that could be why I didn't see any initial problems, but I'm not comfortable keeping 1013 in here now.

 

[DAVIDS]

Originally posted by: Xeon
Looks like I'll be flashing my board back to 1011.

Thanks for the information DAVIDS!

BTW, I do have the P4C800-E Deluxe so that could be why I didn't see any initial problems, but I'm not comfortable keeping 1013 in here now.

Do what you think is best. I'm not trying to start a panic. But people should know there's an apparent problem with 1013. Quite frankly, I'm a little annoyed with Asus for not offering an explanation.

 

[Xeon]

Originally posted by: DAVIDS

Originally posted by: Xeon
Looks like I'll be flashing my board back to 1011.

Thanks for the information DAVIDS!

BTW, I do have the P4C800-E Deluxe so that could be why I didn't see any initial problems, but I'm not comfortable keeping 1013 in here now.

Do what you think is best. I'm not trying to start a panic. But people should know there's an apparent problem with 1013. Quite frankly, I'm a little annoyed with Asus for not offering an explanation.

Oh no, not at all, I know your not.

I frankly appreciate the heads up for sure and I'm sure others do as well.

Thank you.

 

[shleepy]

What do u mean by low level formatting on HD, Xeon?

 

[Xeon]

Originally posted by: CannibalisticH0b0
What do u mean by low level formatting on HD, Xeon?

When you do a low level format you write 0's and 1's accross the entire Disk platter. This is the only real way to wipe a drive completely clean.

You can either boot from a Windows CD rom and do it through the installer, or in some cases you can do it through the bios as I normally do with my Dell Server.

In your case I would just boot from the windows CD or do it from creating a windows boot floppy and issue a command from the cmd line.

If you boot from the windows CD you will be presented with the options to Format and re-partition the drive, which is by far the easiest way to go.

You can also do this with 3rd party software that's available but for formatting a drive I would just use the built in Windows tools.

If this isn't clear then I would just suggest you do a search in Google and/or look in the Windows Help. It will tell you how to do it.

 

[jhites]

Many of the HD manufacturers offer utilities for complete restoration back to factory condition which also repairs and verifies the HD. I have only used this once on a Maxtor but it worked exceptionally well for me.

Never have seen a need to update my BIOS from 1008.004 and has been running solid since May. Guess there is something to be said for "If it ain't broke don't fix it". 😉

 

[Winuke]

I've been reading this thread with interest and this bit about these boards having unstable Vcore really hit home. I'm doing my first OC and been pretty successfull using the great input from this forum and others. My rig has this stuff:

P4C800-E Dlx (ver. 1.02)
P4 2.6/800
4x256 Geil PC4000 Platinum
Nexus PHT-3600 CPU Cooler
TT Tiger 1 NB Cooler
Seasonic Super Silencer 350W PS

My goal is to get to 1 Ghz FSB - if for no other reason then I like round numbers! I've slowly taken it up to 3250 Mhz (250 x 13) over that last week. It seems to run fine for normal messing around and will run SSandra burn-in forever without crashing. However, it won't run Prime95 for even a minute without an error. So as I learned here, I startup tweaking up the Vcore. Then I noticed that the vore drops by 0.1 to 0.12 v going from idle to full load! I increased the Vcore up to 1.75 v in bios which gives me about 1.74 v at idle and 1.62 v under load (using Asus Probe). With this setting Prime95 will run 10 to 20 minutes before getting errors. I feel like the rig could run stable if I raised the Vcore higher but I'm getting a little gun shy. I'd feel much better if I could set Vcore to 1.65 to 1.70 and have it hold there under all loads.

I have tried swapping out the PS with an Enermax I've got. Same story.

So I'm wondering if I just have a real bad board or is this typical behavior? What kind of Vcore fluctuations are others seeing? I've seen some talk in other forums that Asus may have a generic problem with their boards.

 

[DAVIDS]

Originally posted by: Winuke

My goal is to get to 1 Ghz FSB - if for no other reason then I like round numbers! I've slowly taken it up to 3250 Mhz (250 x 13) over that last week. It seems to run fine for normal messing around and will run SSandra burn-in forever without crashing. However, it won't run Prime95 for even a minute without an error. So as I learned here, I startup tweaking up the Vcore. Then I noticed that the vore drops by 0.1 to 0.12 v going from idle to full load! I increased the Vcore up to 1.75 v in bios which gives me about 1.74 v at idle and 1.62 v under load (using Asus Probe). With this setting Prime95 will run 10 to 20 minutes before getting errors. I feel like the rig could run stable if I raised the Vcore higher but I'm getting a little gun shy. I'd feel much better if I could set Vcore to 1.65 to 1.70 and have it hold there under all loads.

I have tried swapping out the PS with an Enermax I've got. Same story.

So I'm wondering if I just have a real bad board or is this typical behavior? What kind of Vcore fluctuations are others seeing? I've seen some talk in other forums that Asus may have a generic problem with their boards.

I would not run Vcore above 1.7v even if it drops under load. 1.75v is extremely high. A lot of people who overclock with this board run at default voltage. I would try to tweak your memory timings and your other BIOS settings a little more before boosting Vcore above 1.7v. I have a 2.6C running at 256FSB and I'm using default Vcore.

 

[DAVIDS]

Originally posted by: jhites

Never have seen a need to update my BIOS from 1008.004 and has been running solid since May. Guess there is something to be said for "If it ain't broke don't fix it". 😉

Yes, that is probably smart. But if you wish to run RAID 1, you have to have 1010 or higher. I'm running RAID 0 now, but am thinking of adding a couple more drives to form a mirror array.

 

[jhites]

Winuke
You will find that there is little to no gain in stability using a vcore setting above 1.685v and there is only marginal variants above 1.60v. The Prime errors may also be from your memory and not the cpu. You are correct about the drop in vcore when under full load. My system is set with 1.5875vcore and runs 24/7 with 100% cpu usage (SoB) and yeilds an vcore of 1.50v to 1.52v under this load. You did not state what you are using for memory timings and vdimm but if you have not done so, then set the vdimm to 2.85v and see if this might help. Also, set the vagp to 1.7v as this seems to be a sweet spot for most using this board.

 

[Winuke]

 

[Winuke]

My memory timings are default for these sticks 2.5 4 4 7. I found I had to raise Vdimm to 2.75 v in order to get memtest86 to run without errors. But at that setting memtest will run all night no problems. My rig is on a test stand with just an old pci vid card, so I haven't messed with Vagp yet. At default Vcore, I can only get to 235 or so without prime95 complaining. Perhaps that is all this cpu will do. I'll try 2.85v mem to see if that helps. Thanks for the advice, I will lower Vcore. But raising Vcore from default to 1.65 did get me stable at 240.

Another thing about my board is the temps that it reports. At idle the cpu temp is 3C below room temp! I have a little temperature probe that I use to measure the air temp exiting the cooler. Even at high load the cpu temp reading is 4C below the exit air temp! I understand enough about thermaldynamics to know that is quite impossible. So the bottom line is I have no idea what the real cpu temp is.

This is actually the second cpu I've tried. The first one OCed even less than my new one - but it also had the same problem with low temp readings. I figured that the cpu thermal diode was faulty - which is what I told Newegg when I returned it. But since both CPUs have read similar temps, it must be the MB that is the real problem.

I have seen in some forum or another that version 2.0 of this board is out and there is speculation Asus has fixed the Vcore droop issue. I'm really thinking of returning this board to Newegg and try to get lucky getting the new version. Have you heard anything about that?

 

[jcpiercy]

Hey Gang

Where in the Bios do I look for my AGP setting

Im experiencing some Choppiness , in UT 2003 and wanted to make sure Im on AGP*8

thanks

JP

 

[Xeon]

Originally posted by: Winuke
My memory timings are default for these sticks 2.5 4 4 7. I found I had to raise Vdimm to 2.75 v in order to get memtest86 to run without errors. But at that setting memtest will run all night no problems. My rig is on a test stand with just an old pci vid card, so I haven't messed with Vagp yet. At default Vcore, I can only get to 235 or so without prime95 complaining. Perhaps that is all this cpu will do. I'll try 2.85v mem to see if that helps. Thanks for the advice, I will lower Vcore. But raising Vcore from default to 1.65 did get me stable at 240.

Another thing about my board is the temps that it reports. At idle the cpu temp is 3C below room temp! I have a little temperature probe that I use to measure the air temp exiting the cooler. Even at high load the cpu temp reading is 4C below the exit air temp! I understand enough about thermaldynamics to know that is quite impossible. So the bottom line is I have no idea what the real cpu temp is.

This is actually the second cpu I've tried. The first one OCed even less than my new one - but it also had the same problem with low temp readings. I figured that the cpu thermal diode was faulty - which is what I told Newegg when I returned it. But since both CPUs have read similar temps, it must be the MB that is the real problem.

I have seen in some forum or another that version 2.0 of this board is out and there is speculation Asus has fixed the Vcore droop issue. I'm really thinking of returning this board to Newegg and try to get lucky getting the new version. Have you heard anything about that?

Hello Winuke

As DAVIDS has already pointed out I wouldn't push the VCore over 1.7 even though the way the P4C800-E Deluxe is currently working with their vcore ranges I still think it will be detrimental to your cpu at some point because it can spike into dangerous territory. I would definately crank up the VDimm voltage to 2.85 and just let it rip, unless your worried about warranty on your RAM of course. I know that OCZ the warranty is good all the way up to 3.1, but I'm not sure about the RAM you have there.

Yeah something isn't right with those temp readings but it isn't clear from your last post exactly how your probe is setup. I for example have a probe that is placed underneath the cpu between the cpu housing and it's core and it's not susceptable to air flow hitting it (Hardcano 9), which it sounds like it is with your setup. If your getting any kind of air flow over the probe I can see how you might get similiar readings to what you describe. Can you describe in a little more detail how this is setup on your box? What apps are you using to measure the temperature of the cpu core?

I'm sorry to report that the vcore droop as you put it, is still a problem on the REV2.00 boards because I have one. I'm assuming this is the REV your reffering to. I'm in the process of providing the data for these variances probably sometime tomorrow on this machine, as I'm traversing all the voltage settings (well up to 1.65 at least) to get their readings so that I can see what the ranges of the voltages are at Idle and under load. Suffice it to say, there is clearly a problem even at idle, with what is being set in the BIOS and what you end up getting when you measure these voltages in windows. This is a big problem for many people and I sit here and wonder why in the world they even provide a place in the BIOS to set these voltages when they are all over the map!!! They have to get this fixed if it's ever going to be a serious board for overclocker's.

At any rate I hope this helped.

 

[computer]

This is exactly why I did bring it up, to help my fellow users! I've spent 80+ hours testing to find the source of this problem and the entire reason I posted to even mention this issue early was so that all those users out here who are struggling to find the answer would know that an answer had been found and is coming, and that they wouldn't have to spend 80+ hours finding it themselves.

...and we've spent much longer than 80+ hours waiting! LOL! 😀 😀 😀

After flashing 1013 and then resetting the CMOS, some people said their boards refused to POST. The rumor is that Asus is getting a lot of RMA requests. But the problem is not affecting everyone with a P4C800 Dlx or -E board, so I don't know what to tell you. BIOS 1013 was pulled from the US site a few days ago, and yesterday, it was also pulled from the German site which is very unusual. I can't recall a BIOS ever being removed from the German server.

Do what you think is best. I'm not trying to start a panic. But people should know there's an apparent problem with 1013. Quite frankly, I'm a little annoyed with Asus for not offering an explanation.

That's for damn sure. SEVEN WEEKS AND COUNTING and I'm still not only awaiting an RMA, but the ba$tard$ won't even reply to 60+ emails I've sent and the 20+ times I've filled out the webform. I'm MORE than "annoyed" with them. They not only never offer any explanations, they ignore their "victims" of their products.

 

[Winuke]

Hi Xeon

As DAVIDS has already pointed out I wouldn't push the VCore over 1.7 even though the way the P4C800-E Deluxe is currently working with their vcore ranges I still think it will be detrimental to your cpu at some point because it can spike into dangerous territory. I would definately crank up the VDimm voltage to 2.85 and just let it rip, unless your worried about warranty on your RAM of course. I know that OCZ the warranty is good all the way up to 3.1, but I'm not sure about the RAM you have there.

The Geil sticks are warrenteed to 3.1v. Setting Vdimm to 2.85v doesn't change my results any.

Yeah something isn't right with those temp readings but it isn't clear from your last post exactly how your probe is setup. I for example have a probe that is placed underneath the cpu between the cpu housing and it's core and it's not susceptable to air flow hitting it (Hardcano 9), which it sounds like it is with your setup. If your getting any kind of air flow over the probe I can see how you might get similiar readings to what you describe. Can you describe in a little more detail how this is setup on your box? What apps are you using to measure the temperature of the cpu core?

I just use a little Radio Shack indoor/outdoor thermometer to measure ambiant room temp and the cooler exhaust. I'm not attempting to take a direct measurement of the cpu, my only point is that the cpu has to be hotter (much hotter perhaps) than the cooler exhaust. The measurement is really just a sanity check. My readings are clearly insane.

Sounds like the new version 2.0 boards are no better in terms of Vcore. I'll be interested to see the data table you're putting together.

Asus customer service is basically non-existant. They have stopped replying to email inquires. They shut down their own forum site last month ostensibly due to server problems. But everyone knows they were getting so much bad flack on the forum they shut it down because they couldn't stand the heat. They clearly aren't the company they used to be, both from a product quality and service standpoint. Perhaps Abit or MSI are the new standardbearers.

 

[jhites]

Hey Winuke
One thing that you might want to experiment with is using only 2 sticks of ram instead of using all 4 slots. There have been many that can get a better OC from using this configuration. The slots used from reports vary as to wheither the black or blue slots are used. From what you have been posting it really sounds like your cpu is just maxed out. Personnally, I would set the vcore to 1.5875v and using 250fsb as a starting point would drop 1Mhz at a time until Prime95 showed no errors. Using 1.5875v will give you close to default vcore under load with the voltage variances that this board yields and still be very safe at idle.

 

[computer]

Also Winuke, according to testing sites not all four slots can be use with PAT on. I've asked several times on this thread for someone to please verify that, but no one will. Using all slots with (double sided memory I think it was) disables PAT. (I use Geil PC4000 Ultra Platinum, but the 2x256mb which is SINGLE sided. Their 1gb kit is double sided).

 

[jcpiercy]

HI Gang

Im having some Vid card problems
I have a 9700 Pro on AGP slot

Can you tell me what PCI slot my Audigy 2 card , should be in

I have a feeling this sound card is messing with my Video card

Here is a pic , to tell the story better

PCI SLOT FOR AUDIGY 2

THANKS GANG

 

[shleepy]

Originally posted by: jhites
Many of the HD manufacturers offer utilities for complete restoration back to factory condition which also repairs and verifies the HD. I have only used this once on a Maxtor but it worked exceptionally well for me.

I have a Maxtor... where'd ya get that utility, jhites? Thx.

 

[shoman94]

Originally posted by: jcpiercy
HI Gang

Im having some Vid card problems
I have a 9700 Pro on AGP slot

Can you tell me what PCI slot my Audigy 2 card , should be in

I have a feeling this sound card is messing with my Video card

Here is a pic , to tell the story better

PCI SLOT FOR AUDIGY 2

THANKS GANG

Yes...put it in the lowest slot.....

 

[Xeon]

Asus P4C800-E REV2.00 VCore Voltage Variances Revealed

I've completed my testing on the fluctuations of the VCore voltages on this motherboard and it revealed some very interesting data.

You can find the complete spreadsheet here:

Asus P4C800-E VCore Voltage Test Data

Here is what they did, and I can tell you by looking at this data that it wasn't a mistake that the Asus P4C800-E Deluxe has this broad range of VCore values despite you setting it to a specific Value!! They engineered in a range of .096 for most all settings with this motherboard.

What does that really mean? Well if you set your VCore to 1.6 for example, it's actually going to have a range of 1.648 to 1.552 which is a .048 above and below the set 1.6!!!

This is the only voltage setting however that I tested that had an even split of the voltage range!!

Also, when you set it to 1.6 for example, it doesn't set you to 1.6 at idle, it sets you to the high end of the range, in this case 1.648! I believe the reason they did this was so that when your doing things that put the machine under load, it meant that you were getting closer to your 1.6 setting while doing work, but under heavy load you drop to the low end of the voltage range, in this case 1.552. This scenario is true for ALL the settings I tested!!

The other interesing point about this testing I did is that if you go anything above 1.6 in the VCore, your wasting your time and it in fact drops you below .096 range. So instead of having a .048 value above what you set your vcore to, you'll actually get something below the vcore value range ALWAYS!! Which basically means, you'll get a highter vcore range below a 1.6 vcore than you will if you set it to 1.625 for example.

I hope you find this data as useful as I have gathering it. There is clearly a good place to set your VCore to and a point where your actually setting it lower by picking a higher value!! Asus really needs to fix this problem if they are ever going to be taken seriously, at least from an engineering point as I do.

See ya!

 

[DAVIDS]

Originally posted by: jcpiercy
HI Gang

Im having some Vid card problems
I have a 9700 Pro on AGP slot

Can you tell me what PCI slot my Audigy 2 card , should be in

I have a feeling this sound card is messing with my Video card

Here is a pic , to tell the story better

PCI SLOT FOR AUDIGY 2

THANKS GANG

Definitely move the Audigy card away from the 9700pro. Video cards can get hot and the more air circulating around them, the better. What kind of video card problems are you having?

 

[ScissorClaw]

Interesting data XEON, if I could see it! Why don't you post it on this site or put a link to soemwhere whre you don't have to join or register to see it. I understand you'd like as many people to register for your site, but this is not the way to go about it. Find a free hosting site, there's plenty of them out there.

 

[Xeon]

Solution for MemTest-86 Hang using the Asus P4C800-E Motherboard

I received what I was waiting for today and I am now able to let you guys know the solution to this problem.

Now it's unclear from my testing whether this problem is wider issue. I personally believe it is, but I have no data to support it. For those using PC-4000 speed ram or Faster and are seeing MemTest-86 Hang during test 4, 5 or 6, this solution will most likely solve your problem. If your having this same problem and are using Memory that's slower than this speed and this solution takes care of your problem as well, I definately would like to hear from you so that I can confirm my suspicions.

All you need to do is to disable Legacy USB support within the BIOS and the Hang in MemTest-86 should disappear.

The other solution you can try is if your using a USB Mouse and/or Keyboard, disconnect them and using a standard keyboard and mouse that do not require a USB connection.

Sorry I couldn't provide this solution sooner, even though I've known about the solution for about a week now.

Because of the time it took for me to find this solution I was waiting on getting some things from Asus and OCZ regarding the solution I was ultimately able to find, before I could publish it to the world.

At any rate I hope it saves you the trouble and hassles that I went through to fix this problem.

Initially when I found this issue I thought it was the OCZ RAM. I must have swapped 6 different modules, in varying configurations and still had the same problem. I then looked to the motherboard itself to find the solution since after testing that many sticks or RAM it was pretty clear that the RAM itself wasn't the problem, since each RAM configuration I used still showed the hang.

Hopefully this will save many many people the trouble and you can just turn off this setting in the BIOS and be good to go. If you need this support within the BIOS, you'll most likely have to get yourself a different motherboard or try doing without the USB components that require you to need Legace USB support, as I don't see Asus fixing this issue anytime soon as their comments to me seem to indicate.

See ya,