Gigabyte P35 motherboards and DPC

ORIGINAL: Desperate Dan

My temps are about 20c cooler than they were with the stock cooler at 100% load in Prime 95.

I Have found that TAT (thermal Analysis tool) From Intel stresses the CPU more than Prime 95 does and 100% CPU on prime doesn't give the Same Temperature that TAT generates. Also it's made by Intel so it should be accurate, also you can have the stress levels set to 50% for 20 seconds then 60 etc etc to stop borderline overheating before it get's to CPU shutoff temperature. I mistakenly set my 2.4 CPU to 3.6 and it just ran a little bit hot but I could see it approaching the 63 Degree limit and went back and checked my settings and found incorrect Multiplier values.

I'm aware of the TAT, but my goal in testing with Prime95 is to check for stability at 100% load, and not to drive the CPU to thermal shutdown temps. Under normal usage, my CPU temps never come anywhere close to what I see with 100% Full load under Prime95.

Do you know what acceptable motherboard temps are with the P35 chips? I'm testing with 1600 FSB now, and my motherboard temps read by the onboard sensor are about 33c when under full load, after about an hour. That's about 3c higher than 1333, and 5c higher than 1066. Does that seem OK? The heat sinks aren't very warm.

Re: BIOS settings not working / sticking after a flash... sometimes BIOS controllers like to be reset after a flash. If the Gigabyte supports it (and you haven't already tried it), you may want to attempt a CMOS Reset. (For the ununitiated: Be sure to follow the MOBO manual EXACTLY on the reset procedure. Some MOBOs can fry if you don't do it right!)

battery out for 30 minutes, or as I read here, put in the opposite way for 5 seconds does the trick. I had the bios not sticking problem and this worked like a charm.

I was curious, so here's your MOBO manual. You can clear the CMOS (reset the BIOS) either by the battery method (page 30, section 19) or with the jumper method (page 29, section 18). It looks like both do the same thing. The jumper method saves wear & tear on your battery holder. Whichever you do, be sure that 1) the power is off and the PSU is unplugged from the wall before you do it, and 2) you put the jumper or battery back the way it was before you plug back in and power back up.

Can some of you guys who have had success with the Beta Bios please confirm that after resetting the CMOS everything in the BIOS is functioning normally, including vcore settings and everything else?

ORIGINAL: dewdman42

Can some of you guys who have had success with the Beta Bios please confirm that after resetting the CMOS everything in the BIOS is functioning normally, including vcore settings and everything else?

I flashed back to the last official BIOS for my board, F3, and everything is working as it should.
EDIT: I think I misunderstood you. I haven't tried resetting the CMOS with the beta BIOS yet, but I will.

I'm aware of the TAT, but my goal in testing with Prime95 is to check for stability at 100% load, and not to drive the CPU to thermal shutdown temps. Under normal usage, my CPU temps never come anywhere close to what I see with 100% Full load under Prime95.

Hi Dan, No I didn't mean push it into the red temp wise,Mine happened to go there because a) It was a real hot day and b) I had overclocked it to 3.6 without realising it. I just meant that you can set your CPU load and that 100% on TAT is more than 100% on Prime 95, that's at least according to my calculations. I Can runother programs with Prime95 running, I can't run anything with TAT speeding along at 100% load on the CPU cores.

Do you know what acceptable motherboard temps are with the P35 chips? I'm testing with 1600 FSB now, and my motherboard temps read by the onboard sensor are about 33c when under full load, after about an hour. That's about 3c higher than 1333, and 5c higher than 1066. Does that seem OK? The heat sinks aren't very warm.

The Board temperatures vary from board to board, my previous Intel board never shut down but the ASIC chips were reporting 50 Degrees C in the Zone 2 which was right next to the CPU it never shut down but was always too hot under load, I figured a Dud ASIC chip as the Board was really well cooled and all the rest of the temp zones were way within bounds.

Now back to Gigabyte being such a nice company and replying to my request for a new Bios, the F3e version they gave me the download link for actually made my DPC latency even worse, needless to say I'm somewhat dissapointed. I Have spent some time stopping and starting different services and have found a few culprits that were causing some spiking and have removed them.

With regards resetting the CMOS in the Gigabyte manual they reccomend shorting the Two battery terminals with a flat screwdriver which won't wear out the Battery compartment, although how many full resets does one actaully have to do?

Hey Lanceindastudio, I see we're using the same board. Keep us posted as to how you fair with the beta BIOS, I'm knee-deep in some projects and can't do any tweaking. That being said I did roll-back this past weekend to F3 as I stated upthread and all is well, the spikes are gone. I had a session last night where I was tracking a guitarist, bassist (both through Amplitube) and myself on keys at the same time with a soft synth. I was at 2.2 ms on my Edirol UA 101 and no glitches or dropouts at all. I didn't really get them before, but things seemed more stable.

hmm ok well F3 where the heck do I get that? Is it on the site? Also, do you overclock?

I lowered back down to 3ghz as 3.2ghz was running the cpu hot with stock cooling during Prime95 torture test

Yea I tried all kinds of different ways to go higher than 3.0ghz and every time the CPU just got a little bit too hot for comfort in Prime95 or else if I lowered vcore it would not pass the tests for very long. The Vcore requirements start to jump up considerably beyond 3.0ghz. My conclusion is that stock cooling is meant for no higher than 3ghz. I fully plan to get better cooling and go to 3.2 or 3.4 later, but for now I'm pretty happy with the stock cooling except for the noise, at 3ghz.

There is a huge difference in performance as measured by benchmarks between the stock CPU speed and 3.0ghz(with optimized FSB and ram speed also). Not as much of a noticeable difference between 3.0 and 3.2, but the difference is definitely there. I want the better cooling anyway just be quieter, so once i have it I will bump up to 3.2 or maybe even 3.4 and call it a day. But 3.0ghz is like the easy everyone-should-by-default-do-it-OC for the Q6600.

yeah well I was going from 1.8 on my e6320 and 3.0ghz all day hehe

Hi Lance....No I do not overclock...yet. I've only had this rig for about 2 months and so far it more than meets my expectations at 2.4ghz. The F3 BIOS was listed on the BIOS page for the board. I read either on this thread or the other one that it worked for someone with our board so I tried it out Sunday. So far so good.

Hey there. Got a GA-P35-DS3L here and getting the same spikes that other users of this motherboard are getting. Sent a request to Gigabyte to look into it. Will report back if any news.

EDIT: On F5, by the way...

Mahlon

Ok cool man. My spikes are pretty low overall, spiking in the green still close to 500. Im using bios F8a

ORIGINAL: Mahlon

Hey there. Got a GA-P35-DS3L here and getting the same spikes that other users of this motherboard are getting. Sent a request to Gigabyte to look into it. Will report back if any news.

EDIT: On F5, by the way...

Mahlon

http://forum.cakewalk.com/fb.asp?m=1365807

Thanks Steve,
Rolling back to F3 -- I'll try that. But I want to upgrade my processor soon... so I hope we get a fix.

Mahlon

Thanks Steve,
Rolling back to F3 -- I'll try that. But I want to upgrade my processor soon... so I hope we get a fix.

What CPU are you using now? I have the Q6600 and have had no issues with this BIOS. As far as I can tell the new BIOS revs are for the Penryn series that Intel has come out with.

ORIGINAL: emwhy

Thanks Steve,
Rolling back to F3 -- I'll try that. But I want to upgrade my processor soon... so I hope we get a fix.

What CPU are you using now? I have the Q6600 and have had no issues with this BIOS. As far as I can tell the new BIOS revs are for the Penryn series that Intel has come out with.

Yes... Q6600 processor here as well. Mahlon... Unless you're think you might end up with a Yorkfield or Wolfdale processor... f3 BIOS will suite you just fine.

Here's an update:

I tried flashing to the beta BIOS, then clearing the CMOS. The BIOS still would not respond to changes in vcore. I flashed back to the last official release BIOS and set up a 3.2 GHz (8x400) overclock. This time I flashed to the beta BIOS and didn't clear the CMOS, and as in my previous testing, it held my overclock settings. So I ran the Sonar 6/7 benchmarks:

3.2 GHz (8x400) RAM at 800

Beta BIOS

WDM
64 samples = 62 multibands
128 samples = 101 multibands
256 samples = 122 multibands
512 samples = 134 multibands

ASIO
64 samples = 38 multibands
128 samples = 68 multibands
256 samples = 87 multibands
512 samples = 134 multibands

Official release BIOS with DPC Spiking

ASIO
64 samples = won't play at all without pops and clicks
128 samples = 52 multibands
256 samples = 86 multibands
512 samples = 108 multibands

WDM
64 samples = 62 multibands
128 samples = 101 multibands
256 samples = 122 multibands
512 samples = 132 multibands

ORIGINAL: SteveD

ORIGINAL: Mahlon

Hey there. Got a GA-P35-DS3L here and getting the same spikes that other users of this motherboard are getting. Sent a request to Gigabyte to look into it. Will report back if any news.

EDIT: On F5, by the way...

Mahlon

http://forum.cakewalk.com/fb.asp?m=1365807

Hey Steve thanx for that link. I hadnt caught that. I will try the F3 bios tonight!

Cheers! Lance

I have not read this entire thread, so if this has been stated already, please forgive me. But the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes. My other ASUS board with a different chipset (VIA with a similar CPU/Bios) does not.

I personally believe it has to do with DMA memory management for devices such as IDE/SATA. The regularity of the spikes suggest its occuring when the system polls a device such as a drive. This would explain why disabling every other driver but essential does not seem to have an effect.

There are periods where I get dropouts in music if I have something going on that is accessing a drive such as a SATA based DVDR. Most annoying and I have yet to actually spend any time troubleshooting it as I tend not to have any problems with recording or working in the DAW and that is what I would rather be doing heheh.

I am using a Q6600 @ 1200mhz FSB OCed to 2700mhz. I could take it higher but I do not like how much voltage it takes to get it to go higher, and keeping memory synchronous is more important to overall performance, than having a faster CPU thats asynchronous with memory in a Vista environment (with Vistas new preloading cache that really requires memory bandwidth to be smooth).

I told gigabyte about people here thinking it is the southbridge part of the shipset, or something ot do with intel's P35 setup, and they said I should try rolling back drivers for the chipset by downloading older ones from Intel's site.

I was like, gee, thanx. lol It's there board, they should solve the problem. This was after they had me try a different bios. Apparently this guy isnt actually aware of any chipset problem.

But, I will try F3 tonight. If it solves the problem, this would tell me that it is probably the bios that is a problem afterall. Then again, it could be a combination of chipset/bios problems, like, perhaps, the chipset is set up in a way that the bios has to be written a certain way for it to work correctly. Otherwise, maybe the chipset is "picky".

ORIGINAL: foxwolfen

the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes.

Foxwolfen,

Which Asus board do you have that is showing DPC spikes?

ORIGINAL: Lanceindastudio
If it solves the problem, this would tell me that it is probably the bios that is a problem afterall.

On my system, with all the same hardware and drivers, just a change to the beta BIOS, all of the spiking is gone.

That seems to prove that it can be solved with a BIOS fix. Maybe different drivers from Intel might solve the problem with the BIOS that has the spiking, but I spent quite a bit of time disabling things and trying different drivers to no avail.

But I am going to see if Intel has any older drivers posted, and I'll try them out. I'm currently back on the official release BIOS.

ORIGINAL: Dan DAmico

ORIGINAL: foxwolfen

the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes.

Foxwolfen,

Which Asus board do you have that is showing DPC spikes?

ASUS P5K with AMI Bios (V0901 - 01/02/08) P35 CS with Intel 82801B (ICH9) southbridge - spikes go as high as 1012us with one spike going up to 2000+ during the run.

If you got the spikes out with the bios, then I am probably stuck as ASUS is really slow updating bioses these days, and intel has not released an update for this chipset for well over a year.

ORIGINAL: foxwolfen

I have not read this entire thread, so if this has been stated already, please forgive me. But the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes. My other ASUS board with a different chipset (VIA with a similar CPU/Bios) does not.

Well, please read the entire thread.

If you do, You will find some screen captures I posted of my ASUS P5K which does not have any DPC spikes whatsoever. I even include a screen capture of all the processes I had running on the system, as it is an internot computer of mine, not a DAW. My ASUS P5K-SE does not have any spikes either, nor does my ABIT IP35 Pro. The only mainboard of mine that has these spikes is my Gigabyte board (but no longer with the beta BIOS update I'm running).

...Just wondering, how can you have 270 posts in less than a month around here?

-mr moon

ORIGINAL: foxwolfen


ASUS P5K with AMI Bios (V0901 - 01/02/08) P35 CS with Intel 82801B (ICH9) southbridge - spikes go as high as 1012us with one spike going up to 2000+ during the run.

Is it simply a P5K or one of the P5K-* boards? The reason I ask is I have a P5K Pro sitting here, brand new and unopened, which I purchased to replace the Gigabyte board. I was under the impression that the Gigabyte boards don't suffer the problem, and some users have posted DPC latency results that don't show spikes with P5K boards. I can still return it, and probably will.

I think we need to see what we all have in common. Aside from two Seagate SATA II drives on the Intel controller, I also have a PATA DWD-RW on the onboard Gigabyte (jMicron) chip. I disabled it in the BIOS and still had the spikes. I also disabled it in Device Manager and still had spikes. Are you using any PATA drives?

I'm on Windows XP Pro 32-Bit. I see you are on Vista. Have you ever run the DPC test on your motherboard with XP?

Dan, Read my post above. I have 2 ASUS boards; a P5K and a P5K-SE, *neither* of which have DPC spikes. I have posted screen captures in this thread to demonstrate this.

-mr moon

ORIGINAL: Dan DAmico

ORIGINAL: foxwolfen


ASUS P5K with AMI Bios (V0901 - 01/02/08) P35 CS with Intel 82801B (ICH9) southbridge - spikes go as high as 1012us with one spike going up to 2000+ during the run.

Is it simply a P5K or one of the P5K-* boards? The reason I ask is I have a P5K Pro sitting here, brand new and unopened, which I purchased to replace the Gigabyte board. I was under the impression that the Gigabyte boards don't suffer the problem, and some users have posted DPC latency results that don't show spikes with P5K boards. I can still return it, and probably will.

I think we need to see what we all you in common them. Aside from two Seagate SATA II drives on the Intel controller, I also have a PATA DWD-RW on the onboard Gigabyte (jMicron) chip. I disabled it in the BIOS and still had the spikes. I also disabled it in Device Manager and still had spikes. Are you using any PATA drives?

I'm on Windows XP Pro 32-Bit. I see you are on Vista. Have you ever run the DPC test on your motherboard with XP?

Ditto - I have a P5KC with a Q6600 and it's solid as a rock. I've never been a fan of Gigabyte BIOSes since ages ago, and I'm very glad I went with the ASUS board (again - my previous board was an ASUS) rather than listen to some of the pro-Gigabyte people here.

ORIGINAL: Dan DAmico

ORIGINAL: foxwolfen


ASUS P5K with AMI Bios (V0901 - 01/02/08) P35 CS with Intel 82801B (ICH9) southbridge - spikes go as high as 1012us with one spike going up to 2000+ during the run.

Is it simply a P5K or one of the P5K-* boards? The reason I ask is I have a P5K Pro sitting here, brand new and unopened, which I purchased to replace the Gigabyte board. I was under the impression that the Gigabyte boards don't suffer the problem, and some users have posted DPC latency results that don't show spikes with P5K boards. I can still return it, and probably will.

I think we need to see what we all have in common them. Aside from two Seagate SATA II drives on the Intel controller, I also have a PATA DWD-RW on the onboard Gigabyte (jMicron) chip. I disabled it in the BIOS and still had the spikes. I also disabled it in Device Manager and still had spikes. Are you using any PATA drives?

I'm on Windows XP Pro 32-Bit. I see you are on Vista. Have you ever run the DPC test on your motherboard with XP?

PK5 SE

I am beginning to agree that it might be peripheral related.

Here are the Spec of my whole system
As mentioned, I have an ASUS MB with a Quad Core OCed
4GB DDR2 OZC RAM
250GB PATA HDD (Seagate), one SATA 160GB (Seagate) and one SATA DVD-RW (LG).
Creative Platinum Fatality with Breakout box (2 Midi in and one mic) ASIO driver
Nvidia GF 8800GTX (768MB)
Network is onboard (ANTASTIC) IGB PCI-e
Mon is Apple 23" Cinema HD @1920x1200 - with 2USB and 2Firewire running a 20GB external USB HDD and a CF Drive
Wacom Tablet (USB)
Logitec G15 (134 keys - 16 programmable macro - great to automate the DAW by the way)
Korg USB Midi controller/keyboard (off during test)
M-Audio PK 88 USB (not pluged in during test)
Vista Ultimate SP1

No, I have not tried the test in XP.

Cheers
Shad

Mr. Moon and kp,

Have either of you run the Sonar 6/7 benchmarks with your P5K boards? Can you post the results?

ORIGINAL: mr. moon

ORIGINAL: foxwolfen

I have not read this entire thread, so if this has been stated already, please forgive me. But the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes. My other ASUS board with a different chipset (VIA with a similar CPU/Bios) does not.

Well, please read the entire thread.

If you do, You will find some screen captures I posted of my ASUS P5K which does not have any DPC spikes whatsoever. I even include a screen capture of all the processes I had running on the system, as it is an internot computer of mine, not a DAW. My ASUS P5K-SE does not have any spikes either, nor does my ABIT IP35 Pro. The only mainboard of mine that has these spikes is my Gigabyte board (but no longer with the beta BIOS update I'm running).

...Just wondering, how can you have 270 posts in less than a month around here?

-mr moon

1) With almost 200 posts in the thread it got to be a bit impractical.

2) Unless you are calling me a lier I, then what is it your are trying to say? I posted the screenshot of the output of my machine... it has the spikes. So if you have none, then the next thing to do is to try to find the common denominator and see if we can solve it. That is my only intent here... what is yours?

3) Who cares how many posts I have? I fail to see the relevance to this problem. But if you must know, then I will tell you. 90% of my posts over the last month have been supporting the work of other song writers/producers by listening to their music and then providing a hearty slap on the back and a "well done". Go visit the song forum and you will see I have provided (for the vast majority of the cases) positive feedback to every song posted since I came. I do this because I like getting feedback on my songs (which I have produced two since arriving) and others have asked for it.

Most of the the rest have been trying to help others in various forums with technical problems to the best of my ability.

a couple were philosophical debates.

So can we get back to solving this problem now?

Guys...settle down. This has been a productive thread up until now. Please don't muddy it up with pettiness.

ORIGINAL: dewdman42

Guys...settle down. This has been a productive thread up until now. Please don't muddy it up with pettiness.

Exactly. Thank you.

That being said, a lot of information has been discussed in this thread and if these issues apply to you foxwolfen, I would heartily suggest you take 15 minutes and scan through the entire thread.

Sounds like a bios version thing again of some boards work and some dont. We have report of an asus with spikes, as well as abit I think I read earlier.
We also have reports of all brands, gigabyte, asus, and abit working. It has to be a matter of quality control ot bios version, and I think the bos/chipset relationship is the determinging factor based on what I have read here. Again, Ill report my results after trying F3 on my board later tonight.

This thread is worth the discussion. I agree, lets keep it helpful, and not "bash"full lol

Wow. Sorry if I came off as such a dick, but I guess I felt that it was rude of you to just jump into the thread without reading the reams of posts and information we've put hours and hours into. I apologize if I came across terse ...I meant to be more amusingly sarcastic than as an arse.

1. If you are going to post information in a thread, it is proper "netiquette" to read the thread through enough to understand where the thread is at in relation to the topic. I feel it was rude of you to just pop in and say "hey, I don't care enough to read through all the other posts, but look what I have to say".

2. I am not calling you a liar, I did not do so. All I'm saying is that your results are completely opposite to what others are experiencing.

3. All I was saying is that in the less than 4 weeks you've been here, you have posted a lot. That is interesting. That's it.

4. There is no #4.

Peace out.

-mr moon

ORIGINAL: foxwolfen

ORIGINAL: mr. moon

ORIGINAL: foxwolfen

I have not read this entire thread, so if this has been stated already, please forgive me. But the problem is with the Intel P35 chipset and not the motherboard or bios. My Asus board shows the same problems with the chipset with the same spikes. My other ASUS board with a different chipset (VIA with a similar CPU/Bios) does not.

Well, please read the entire thread.

If you do, You will find some screen captures I posted of my ASUS P5K which does not have any DPC spikes whatsoever. I even include a screen capture of all the processes I had running on the system, as it is an internot computer of mine, not a DAW. My ASUS P5K-SE does not have any spikes either, nor does my ABIT IP35 Pro. The only mainboard of mine that has these spikes is my Gigabyte board (but no longer with the beta BIOS update I'm running).

...Just wondering, how can you have 270 posts in less than a month around here?

-mr moon

1) With almost 200 posts in the thread it got to be a bit impractical.

2) Unless you are calling me a lier I, then what is it your are trying to say? I posted the screenshot of the output of my machine... it has the spikes. So if you have none, then the next thing to do is to try to find the common denominator and see if we can solve it. That is my only intent here... what is yours?

3) Who cares how many posts I have? I fail to see the relevance to this problem. But if you must know, then I will tell you. 90% of my posts over the last month have been supporting the work of other song writers/producers by listening to their music and then providing a hearty slap on the back and a "well done". Go visit the song forum and you will see I have provided (for the vast majority of the cases) positive feedback to every song posted since I came. I do this because I like getting feedback on my songs (which I have produced two since arriving) and others have asked for it.

Most of the the rest have been trying to help others in various forums with technical problems to the best of my ability.

a couple were philosophical debates.

So can we get back to solving this problem now?

Yes they have and having scanned the thread (I scanned the thread, which meant I tended to skip over posts that stated they had no problem as they did not seem particularly useful) I did not see any real solution had been found and it seemed that the consensus was that the problem was with Gigabyte. But that is not the case, so for me the solution of a bios update does not apply - so perhaps the problem can be solved another way, especially if people are going out and spending money on parts they may not need, or that may not solve the problem.

If the solution had been found, then please direct me to the relevant post because with some 200 of them, its a bit hard to find.

Cheers
Shad

I figured it out! Dont get P35!

Thats a solution! And in the meantime, us here with a P35 will look for other solutions!

Thanx!, Lance

ORIGINAL: mr. moon

Wow. Sorry if I came off as such a dick, but I guess I felt that it was rude of you to just jump into the thread without reading the reams of posts and information we've put hours and hours into. I apologize if I came across terse ...I meant to be more amusingly sarcastic than as an arse.

Sigh - there is an old saying - never assume or you make an ASS out or U and ME. You assumed that because I said I had not read every post, that I did not read any? I read the majority, but when you are scanning some 200 of them, it gets hard to absorb everything posted. I stated that I had not read them all because when you do arrive at a thread that is applicable to yourself, and there is 200 posts, its nearly impossible for any human being to scan them all and not miss anything. This is not unusual.

I thought I was being polite.

Shad

guys. Please. Drop it or take it offline to Private email. Gracias.

I apologized to him. There's nothing more I can do except leave this thread. So I will do that now.

Bye.

-mr moon

ORIGINAL: dewdman42

guys. Please. Drop it or take it offline to Private email. Gracias.

Mr Moon,

Please don't leave the thread because of this misunderstanding. I appreciate your contributions, and your valuable insight.

Sigh - there is an old saying - never assume or you make an ASS out or U and ME. You assumed that because I said I had not read every post, that I did not read any? I read the majority, but when you are scanning some 200 of them, it gets hard to absorb everything posted. I stated that I had not read them all because when you do arrive at a thread that is applicable to yourself, and there is 200 posts, its nearly impossible for any human being to scan them all and not miss anything. This is not unusual. I thought I was being polite. Shad

Shad; When Mr. Moon gave his reply I was thinking the same thing. This thread is full of detail information. It really can't be scanned with out lot of valuable data being missed. I would go as far as to say if one has an interest in this it would be best to print out the entire thread and study it. It is that complicated. There are a lot of conflicting points be made. Sorting out all of this is not going to be easy.

This is in no way a rebuke or a criticism of your posting it is simply how I view how important it is to anyone looking at this thread to understand all the issues.

I can fonfirm that F3 bios fixes the dpclatency spikes issue on the GA-P35-DS3L board. See screenshot below.
Also, my overclock works perfect, 1.8ghz stock to 3.0ghz.

Lance have you lost any functions by rolling back the BIOS?

You wanna hear something truly whacked with ASUS? I can not roll my bios back to an earlier version. No in Windows, not from their recovery CD and not even when I made a old style dos boot disk... each time it says that there bios I am trying to replace is newer than the one I am replacing it with... yes I know...

This never used to be a problem...

With ASUS's famous tech support (or should I say - non- support) I will never get a solution to this problem.

I had the same problem with the last Asus board I used, an A8N-SLI Deluxe. They must have so much confidence in their BIOS releases that they assume no one will ever need to go back to a previous version.

I had several problems with that board, and contacted Asus tech support 5 times over a 3 month period and never received a reply. I finally called their RMA dept. on the phone and requested an RMA. After about a month they told me the board was beyond repair, and they replaced it with an A8N-SLI Premium.

I have lost nothing. The only thing is, this bios does not support Yorkield 45nm cpu's. Luckily I dont have that and this might persuade me to go Q6600 instead of Q9300 0r Q9450.

The newer bios releases support 4nm CPU's but I like my dpclatency check to pass ;)

And overclocking works absolutely perfect.

Well I am still struggling with my Gigabyte EP35-DS3R rev 2.1. I Have spoken to the RME Agents and done everything they suggested. My latency is fine now, I can play a 10 Track song at 256 Samples with Hypersonic at 10% or less CPU, but I am now getting the Occasional dropout about maybe every thirty seconds. It doesn't spike the DPC Latency Tool, or the Process monitor or Disk monitor, so it's not Disk, Process or CPU related but something else.

I tried changing my TI Based FW800 Card to a VIA chipset Firewire 400 card, made no difference. I read and re-read my RME manul to see if I missed something there. I tried Underclocking the Video card, no joy. My Playback in Sonar is now a lot smoother and doesn't bang when I hit stop, like it used to. The Gigabyte F3e BIOS they emailed me didn't do much at all, in fact it could be a little worse. I also removed my Raid drivers as advised and found some Useless Services and processes that I took out. It feels quick and streamlined now but the occasional dropout still bugs me.

Lance your screen shot looks almost identical to mine after I rolled-back to F3, glad to see it worked for you. I've had two heavy-duty sessions since then and no problems, although things were stable for me before. I guess I just feel better now. BTW I vote you get Q6600, you can overclock it to 3.0 with ease, but you might not need to.

Well I Have finally fixed my problem with my New PC and Fireface 800, Someone suggested I download a Microsoft Fix which seems to have worked, even with slightly Bumpy DPC latency I can now get my Fireface down to 64 Samples, which is not neccessary but it is nice to know it's capable of low latency work at reduced bandwidth. I will keep updating the BIOS and maybe they get that right for my DS3R board as well, which will be a bonus.

Dan,
Which MS fix did you download? I was going to suggest the dual core and Intel CPU MS fixes, but I assumed you already had those. Is there something else?

I have lost nothing. The only thing is, this bios does not support Yorkield 45nm cpu's. Luckily I dont have that and this might persuade me to go Q6600 instead of Q9300 0r Q9450.

The newer bios releases support 4nm CPU's but I like my dpclatency check to pass ;)

And overclocking works absolutely perfect.

Lance, this is comforting, since my new set up is very close to yours. Same board and sound card. I have the Q6600. Currently on board Bios ver. F7, and will probably run the DPC latency test this weekend. If I see spikes I'll try the F3.

Curious, what are you all using to benchmark number of tracks, etc. Is there a standard SONAR file somewhere to use for this? Sorry if I missed that somewhere. Thanks.

Dan,
Which MS fix did you download? I was going to suggest the dual core and Intel CPU MS fixes, but I assumed you already had those. Is there something else?

Hi Dan

I Just started typing a reply to your message telling you how everything is Hunky Dorey and wanted to post a screenshot of DPC Latency and I launched Sonar and all it does it Motorboat on the Right channel until you hit the Reset Exclamation mark. Damn I'm getting tired of this especially when I thought I had it beat.

Here's the Screenshot, check the CPU2 in the Right hand corner all I can think of is that I have just swopped the 702 for the 703 Hotfix.

Mark, I don't use any standard SONAR benchmarks so to speak. Mostly what I do is load-up a full mix of about 40 tracks or so with various plug-ins and soft synths and see 1. how low I can get my latency before I start getting drop-outs. 2. See if I can jam in any more soft synths or multi bands into the project. My last machine was a 3.2 P4 with Hyper threading with a 400 mhz FSB and it would just choke compared to what I can do now. I've been running between 2.2 & 3 ms of latency and haven't had to go any higher with an Edirol UA101 since I put this new build together at the end of January. So the only issues I have had have been caused by 3rd party plug-ins, mostly the KORG MS-20 USB controller. It just doesn't like me for some reason.

BTW guys I've been running XP Pro SP3 (RC1) on this DAW since I built it without any hitches.

Dan

Sorry I didn't answer your Question after the Sonar Freaked out. I rebooted and it's back to performing well now, don't have a clue what caused that. There is a process that keeps popping up called winlogon.exe, but I dont logon or use those facilities. Sorry back to the Point, the Microsoft Fix is called KB904412 ver 2.0 and it is now on my list of must have MS fixes. I always install MS, Firewire,Microcode and dual core fixes on all the PCs I build but this 904412 is new to me, it was pointed out to me by my RME rep. I am going to post a separate thread on this subject as I think it's important.

Here's my DPC Latency now which doesn't bother me in the least.

ORIGINAL: mwall

Curious, what are you all using to benchmark number of tracks, etc. Is there a standard SONAR file somewhere to use for this? Sorry if I missed that somewhere. Thanks.

There is a Sonar 6/7 Benchmark Test you can run. That is where the figures I have posted in this thread have come from.

You can link to it here. Scroll down to post #206 for instructions.

ORIGINAL: Desperate Dan

Dan

Sorry I didn't answer your Question after the Sonar Freaked out. I rebooted and it's back to performing well now, don't have a clue what caused that. There is a process that keeps popping up called winlogon.exe, but I dont logon or use those facilities. Sorry back to the Point, the Microsoft Fix is called KB904412 ver 2.0 and it is now on my list of must have MS fixes.

Thanks. I'm downloading that fix now.

BTW Dan thanks for a great thread!