Printable View
just to add
you may try memset
sometimes a problem of high performance DDR 2 RAM and the P5 B bios
are the tRFC settings : 42 (default)
for high-speed RAM 1100 Mhz +... thats way too low ..try > 60
there is and sorry to say never existed an Asus bios that changed these way too low timings .
I was able to squeeze 1120 stable out of the 4096 Apogee 1066 set (not so easy with 2x 2GB RAM sticks)
with > 9100 read speed in Everest @ only 3 GHz
Best
mine
um...anyone care to answer my question? :)
snnoopypw: Sorry, I haven't tried running 8GB on a P5B at DDR2-800. You might have to be the guinea pig and give it a try. Even if some user posts that he did it, you might try it and it might not work with your ram.
I do know that the guy that said this:
was probably just running into the limitations of a 32bit OS.Quote:
I put 4 x 1Gb PC-6400 Rambo ram on my board, and it only detects 3Gb....
darn...well I'll try the web, maybe someone has more into on this.Quote:
Originally Posted by unclewebb
thanks anyway. :)
but of course... :DQuote:
Originally Posted by unclewebb
Hi unclewebb, thanks for the reply.Quote:
Originally Posted by unclewebb
When I say that the memory doesn't want to run at the rated speeds, they don't want to boot up at all - I have to unplug the power so that it resets to default settings.
I tried raising the NB to 1.65v, but it didn't seem to make a difference.
I've tried this ram with my e0 e8400 as well as a c0 e8400, no difference.
I've also tried increasing Vcore, but that doesn't seem to be my limiting factor.
I think your theory on the bios not being optimized for newer memory, seems to be the most plausible explanation - unfortunately, I don't have any other high performance ram handy to test it with.
Hey mineQuote:
Originally Posted by mine
I tried increasing the tRFC setting in windows via memset and then increasing the fsb via setfsb, the change did seems to make a slight improvement, I was able to increase the fsb slightly higher than with the tRFC at 42.
It's almost as if one of the recent bios updates has negatively effected ram compatibility. When I first got my 2x1GB Team memory it worked great but then it developed the problem where it would fail to boot sometimes. I thought one stick was bad so I swapped in a similar Crucial 1GB stick and it worked for a while but then it also developed the fail to boot issues. It got so bad at times that I'd have to swap in my 512MB Micron DDR2-667 to get it to boot up again so I could get in the bios. This was never a problem with my original E6400 but the 45nm E8400 - C0 seems a lot pickier when it comes to memory.
Since swapping in the OCZ Reapers, I haven't had this problem once. When I get the chance I'll try taking the SPD data from the Reapers and burn it to my Team memory to see if I can create some Frankenstein modules that don't have this boot issue with the P5B.
It should be here next week, thanks for the heads up. Ill post up what happens after its in.Quote:
Originally Posted by mine
@ unclewebb: I have an e8400 @ 4.2 w/1.38v DMM. Your pushing the NB up to 1.65? Have you taken off the HS assembly & redid the TIM? Also it is a very good idea to get some little washers to mount under the springs to give more pressure. I use caramiq on the mosfetts also, washers there too.
So far so good with 1.65 volts to the northbridge. If there was a bios setting for 1.75 volts I think I'd try that. My room temp is only about 20C so the NB temp isn't too bad.Quote:
Originally Posted by [XC] Hicks121
Pulling off the heatsink, squaring it up and using some AS5 was definitely a good thing to do. I haven't tried adding a couple of washers yet but that might drop the temps another couple of degrees. The OEM push pins are pretty cheap. I thought about adding a couple of bolts to it to tighten it up but it seems to work OK as is so I just left it.
I figure if I fry this board then I'll have a reason to get a P5Q Deluxe or similar. This P5B has had some quirks during its lifetime but at the moment it's running pretty good.
I was running perfectly stable(12+hrs) Orthos at 9*355 and I decided to try and
run 8*400 so I could run 1:1 RAM. When I applied the settings everything booted fine and loaded windows properly. Once I was in windows I started Orthos to test it at the new settings and it froze after ~20sec. Ever since it froze I have not been able to get it to boot again. I've searched a lot and tried numerous fixes posted by people with no results. I'll list what I've tried.
1. Initially just jumpered the reset CMOS for ~10sec, nothing
2. Removed battery and jumpered CMOS, nothing
3. Disconnected all CDROMS and hard disks, nothing
4. Removed all RAM and booted it with none, then rebooted with one stick, nothing
5. Left the power off and unplugged all night, nothing
6. The one last thing I'm trying is leaving the CMOS battery out and the jumper set all day while I'm at work.
Also, ever since it froze when it boots new it will boot the turn off for a bit then reboot. Almost like it does when you apply new clock settings. I'm really at a loss for what to do and would appreciate any help/advice. Do you think the board is dead and if so what new board should I consider without dropping $300. Thanks!!
WHat ram you running? Got a different stick you can try? One of mine last night, I came home to a frozen rig, rebooted, & was doing what your describing. Did the cmos & still had the prob, swapped out the ram & bam! it booted. Hope this helps.Quote:
Originally Posted by jgree32
Quote:
WHat ram you running? Got a different stick you can try? One of mine last night, I came home to a frozen rig, rebooted, & was doing what your describing. Did the cmos & still had the prob, swapped out the ram & bam! it booted. Hope this helps.
I had 2x1GB Corsair XMS2 DDR2800 installed and had just bought 2x2GB corsair xms2 DDR2 800 and was running with all 6GB installed. Bios and windows had recognized ram properly. When I started having problems I went to using just one stick of the 1GB in Channel A.
Try a different stick of ram then. What NB voltage you running, im not too sure on that amount of ram & this mobo, most I ever run is 2 gigs. It will end up booting for you, these boards are just plain buggy at times. I had to let one of mine sit for 2 days before i applied power to it & it finally booted.
Quote:
Try a different stick of ram then. What NB voltage you running, im not too sure on that amount of ram & this mobo, most I ever run is 2 gigs. It will end up booting for you, these boards are just plain buggy at times. I had to let one of mine sit for 2 days before i applied power to it & it finally booted.
Well, before it began locking up I was running Auto NB Voltage. I'm not sure
what Auto sets it to or I would have bumped it one notch maybe. Now that It won't boot obviously I can't change anything. Is a NB voltage increase typically required for 4 sticks? I agree, this board is very buggy at times. I've had problems where when upgrading the bios I could never get the BIOS to detect all my SATA drives. I had to downgrade to 1223 since that's the only bios that would work right. I haven't tried any of the new beta ones. What is the typical bios most people overclocking run on this board? I'm not truly convinced it's bad I'm just running out of options. Really appreciate you trying to help me out.
jgree32: My P5B also has a history of being buggy when pushing it hard and trying settings that it doesn't like. It gets into a mode where no matter what you do, it won't boot up. I'm using the most recent 12.37 bios and same thing.
The easiest thing that works for me is I have some Micron DDR2-667 2 x 512MB sticks of ram. They actually have D9GMH chips on them but only single sided. As soon as I stick in one of these sticks, it will immediately boot up so I can get back into the bios and go from there.
When trying to get it going again, use only a single stick of memory. Another thing that has worked for me in the past is briefly removing the graphics card. Power it on and then turn it off and replace the card again. You need to trick the board into thinking that something has changed so it dumps the bios settings that aren't working. I've found that 65nm Dual Core CPUs get into this "won't boot" mode less often than my 45nm E8400. I also have a cheap E2160 65nm Pentium CPU available to swap in to shock this board into booting up but I've never had to go that far.
I actually spent all day yesterday deliberately trying to force my board into this "won't boot" mode so I could try and understand it more and come up with some ways to fix this issue. I succeeded several times! :D
I found that having inadequate memory voltage when going for big DDR2 MHz numbers or tight timings was a good way to cause this problem.
I think having the northbridge locked at a minimum of 1.55 volts is a good idea. I wouldn't use any AUTO settings on this board because you'll never know how much voltage anything is getting. It gets pretty stupid when set to AUTO for CPU voltage so I don't trust it for anything else.
I've been using 1.65 volts for the northbridge lately. The advantage I've found with this is that when using SetFSB, I can jack the MHz up or down in one step by a huge amount without it locking up. With inadequate northbridge voltage, I have to move the MHz slider in baby steps. I also tend to get more lock ups when making adjustments in the bios when at a high FSB if the northbridge voltage isn't high enough.
Anyway, I decided to create a generic SPD setting for my old 1MB sticks that used to be kind of cranky. They are both double sided and rated DDR2-800 at CL4, one Ballistix and one Team, so I wanted to come up with a common SPD setting to use for both of them to try and cure some of the boot issues.
http://img352.imageshack.us/img352/8...ingsvm9.th.pnghttp://img352.imageshack.us/images/thpix.gif
More details in a minute or two about how that went.............
I should have posted more info about my system specs, I apologize for that. Just really irritated with this board right now. My specs are as followsQuote:
jgree32: My P5B also has a history of being buggy when pushing it hard and trying settings that it doesn't like. It gets into a mode where no matter what you do, it won't boot up. I'm using the most recent 12.37 bios and same thing.
The easiest thing that works for me is I have some Micron DDR2-667 2 x 512MB sticks of ram. They actually have D9GMH chips on them but only single sided. As soon as I stick in one of these sticks, it will immediately boot up so I can get back into the bios and go from there.
When trying to get it going again, use only a single stick of memory. Another thing that has worked for me in the past is briefly removing the graphics card. Power it on and then turn it off and replace the card again. You need to trick the board into thinking that something has changed so it dumps the bios settings that aren't working. I've found that 65nm Dual Core CPUs get into this "won't boot" mode less often than my 45nm E8400. I also have a cheap E2160 65nm Pentium CPU available to swap in to shock this board into booting up but I've never had to go that far.
I actually spent all day yesterday deliberately trying to force my board into this "won't boot" mode so I could try and understand it more and come up with some ways to fix this issue. I succeeded several times!
I found that having inadequate memory voltage when going for big DDR2 MHz numbers or tight timings was a good way to cause this problem.
I think having the northbridge locked at a minimum of 1.55 volts is a good idea. I wouldn't use any AUTO settings on this board because you'll never know how much voltage anything is getting. It gets pretty stupid when set to AUTO for CPU voltage so I don't trust it for anything else.
I've been using 1.65 volts for the northbridge lately. The advantage I've found with this is that when using SetFSB, I can jack the MHz up or down in one step by a huge amount without it locking up. With inadequate northbridge voltage, I have to move the MHz slider in baby steps. I also tend to get more lock ups when making adjustments in the bios when at a high FSB if the northbridge voltage isn't high enough.
Anyway, I decided to create a generic SPD setting for my old 1MB sticks that used to be kind of cranky. They are both double sided and rated DDR2-800 at CL4, one Ballistix and one Team, so I wanted to come up with a common SPD setting to use for both of them to try and cure some of the boot issues.
E6600
Vista 64
EVGA OC 8800GTS 640Mb
6GB Corsair XMS2 DDR2 800
I've had problems getting boards to boot before but this has proved to be the worst yet. I really don't believe it's dead but with all the things I tried I began to get a little worried. I don't have any old DDR2 lieing around. All I have is 2 2GB sticks and 2 1GB sticks so I guess I could try those one at a time to see if it helps. I haven't tried the video card thing you mentioned so I'll give that a shot. I do have a 7800 gt I could borrow out of my other Pc, do you think swapping that in would help anything? The other PC has DDR 1 otherwise I could try it's RAM and it's an AMD machine so no swapping CPU. Also is there a list available somewhere that shows what the AUTO voltage settings set to. I'm currently using vcore set to 1.35, Memory set to 2.1 and all else on auto. I didn't really know what to set the NB or FSB or any of those voltage to.
Set the NB to 1.45-1.55, unless your going crazy on the OC. Im good with 1.55 for 550fsb. You can run the vfsb all the way up & it will be fine. I run my e6600 @ 3.8 24/7/365. The memory is where you are having the issues. Got a local store where you can pick up a 512 stick to get it back up & running? Another option is to pop out the cpu & put it back in. Like unclewbb said, you need to trick the mobo, sometimes there is no method to this madness, sometimes some things work to bring it back, while sometimes what you did the last time wont work this time.Quote:
Originally Posted by jgree32
no method to this madness, sometimes some things work to bring it back, while sometimes what you did the last time wont work this time.
Nice
Same here ...
I have tried many of the numerous suggestions posted for my problems and still have had no luck. Anymore suggestions appreciated.Quote:
I was running perfectly stable(12+hrs) Orthos at 9*355 and I decided to try and
run 8*400 so I could run 1:1 RAM. When I applied the settings everything booted fine and loaded windows properly. Once I was in windows I started Orthos to test it at the new settings and it froze after ~20sec. Ever since it froze I have not been able to get it to boot again. I've searched a lot and tried numerous fixes posted by people with no results. I'll list what I've tried.
1. Initially just jumpered the reset CMOS for ~10sec, nothing
2. Removed battery and jumpered CMOS, nothing
3. Disconnected all CDROMS and hard disks, nothing
4. Removed all RAM and booted it with none, then rebooted with one stick, nothing
5. Left the power off and unplugged all night, nothing
6. The one last thing I'm trying is leaving the CMOS battery out and the jumper set all day while I'm at work.
Also, ever since it froze when it boots new it will boot the turn off for a bit then reboot. Almost like it does when you apply new clock settings. I'm really at a loss for what to do and would appreciate any help/advice. Do you think the board is dead and if so what new board should I consider without dropping $300. Thanks!!
Did you try removing your graphics card or moving it into the other PCI-E slot?
I had a nice Prime run today with my E8400 at 7.5 x 533 ~ 4000 MHz. I'll post some bios screen shots so if you ever shock your board back to life at least you'll have some good baseline voltages and memory timings that you can try.
I did try removing the graphics card and booting without it. I also tried removing the CPU and booting without it. I've also tried a different video card from another PC. I just can't believe that none of the things I tried worked. I guess maybe something was going wrong with the board or something for this to happen at settings which it booted at fine. I really have no idea at this point but I will try the other PCIE slot. Any suggetions on motherboards if I have to buy another without dropping $300. Thanks again! Oh, also forgot I tried another PSU.Quote:
Did you try removing your graphics card or moving it into the other PCI-E slot?
I had a nice Prime run today with my E8400 at 7.5 x 533 ~ 4000 MHz. I'll post some bios screen shots so if you ever shock your board back to life at least you'll have some good baseline voltages and memory timings that you can try.
It sounds like you've tried pretty much all of the normal tricks that usually work. I recently helped a friend with his new Asus P5Q Pro. It was inexpensive but overclocked well and had a decent cooler on the northbridge similar to the P5B. I also like Gigabyte boards but haven't used one with the newer Intel chipsets so I can't help you too much.
My P5B was rock solid today with some new settings. Here's how it did running Prime small FFTs for the day:
http://img396.imageshack.us/img396/5...3mhzig9.th.pnghttp://img396.imageshack.us/images/thpix.gif
This was with a bios setting of 1.55 volts for the northbridge. Here are the voltages and memory timings I used:
http://img378.imageshack.us/img378/8...voltagenn3.jpg
My E8400 (C0) runs OK but has always needed more voltage than most. The trips to 100C and beyond might have something to do with that! I think a fresh E0 stepping should run 4 GHz, Prime stable, with 1.30 volts or less.
http://img515.imageshack.us/img515/7...smemorysf3.jpg
Edit: I originally had the Static Read Control set to Faster but later learned that this can cause "unable to boot" issues when the FSB is at 333 MHz which is the default for an E8400. This changes the Performance Level as reported by MemSet. Not by much but enough to cause boot troubles with some combinations. At 533x7.5 on the Faster setting everything was fine but at lower FSB MHz it was problematic.
Nothing too complicated but check out the memory write bandwidth.
Newer DDR3 chipsets with aggressively overclocked CPUs would be envious of a number like that:
http://img407.imageshack.us/img407/2245/everestxh5.jpg
I was using my mismatched DDR2-800 ram but both sticks have D9GMH chips on them. I designed a new SPD table that I used so they would match up better and hopefully cause fewer problems at boot up. My Reapers have been pretty decent so far at booting up so I borrowed some of the settings from them and from the net and did lots of testing yesterday. The 965 chipset doesn't use the EPP settings so I got rid of all of them and I turned off the CL3 and CL4 SPD settings since I want this board always using CL5. Unfortunately, after a bunch of hard work, I learned that the P5B ignores most of the information in the SPD tables within the memory modules.
I thought boosting tRFC up to 56 and tRC up to 32 might help at higher frequencies. I created a SPD table with those settings as shown in my earlier post above but when I booted up, the P5B sets that to 42 so there is no advantage. You can adjust that with MemSet once you're in Windows but that isn't going to help any if you can't even boot up. My run above was with tRFC = 42.
I tried SPD timings of CL5-5-5-15 or -18. The P5B only sets CL5 and ignores the rest. Those two settings both end up as CL5-6-6-15 when running MemTest86+ or booting up into Windows. This might help explain why this board sometimes has problems booting up. It ignores most of the information in the memory SPD tables.
The modified SPD table I created is working OK for me today but with the P5B, tomorrow is another day. No lock ups recently other than ones I've deliberately tried to create. I might play some more tomorrow to see what else I can learn. I'll also post the SPD table I created if anyone wants to have a look at that. I'm not yet sure if it is helping me in any way since I had to use manually set timings when testing and most of the info is ignored by the P5B.
Edit: Resume from stand by is working 100% with these settings.
Nice and interesting and corresponding to my findingsQuote:
Originally Posted by unclewebb
you see the differencies in memory performance with
FSB @ 500 +
Here is mine with the Apogees @
FSB @ 348
http://img399.imageshack.us/img399/3...3869dc3ue4.jpg
Write peformance is clearly lower with the much lower FSB
But what about your latencies ?
Best
m
The L1/L2 cache latencies are different because your screen shot is for a 65nm CPU while my screen shot is for a 45nm. Intel likely changed the latency settings when they introduced the new CPUs for better stability at high MHz.
As for memory latency, usually the less the better. The most important numbers though are always memory bandwidth or how fast a CPU can read and write information to and from your memory. I'm not sure what spec, read, write or copy, is the most important to overall performance.
I tired to run my E8400 at 9.0 x 348 for a better comparison. My CPU is capable of that but my memory choked. I switched to the OCZ DDR2-9200 Reapers but they didn't have the balls to run at DDR2-1160 either. They are officially rated at DDR2-1150 but on this board they struggle and need too much voltage as soon as they go over DDR2-1100 speed. For my combination, I get excellent reliability and decent performance keeping them at DDR2-1066 without needing to go over their rated 2.1 volts.
When overclocking I get my best 24/7 results by finding out what my CPU can do and then adjusting multipliers and memory speed so that the memory is optimized. In real world use, there's not much difference between DDR2-1000 CL4 or DDR2-1100 or 1200 with CL5 timings. Most benchmarks that show a large difference that I've seen are usually when they are making a comparison with two totally different CPU speeds.
I'll have to go try some benches. Setting my CPU to 8x500 and using DDR2-1000 CL4 timings might be best for my combination. This CPU has seen 4500 or so MHz but it needs too much voltage to run reliably at much over 4 GHz for 24/7 use.
One thing I learned during today's testing was that the memory bios setting, Static Read Control can cause some boot problems.
I returned my CPU to default (9.0 x 333) and with the memory set in the bios to DDR2-667, it booted up fine. I rebooted and tried the next setting, DDR2-833 but it would not boot up. Static Read Control was set to Faster at the time. I tried using Fast but the same thing. It would boot up fine when this was set to AUTO or set to Disable but it didn't like Fast or Faster even though my memory was only trying to run at DDR2-833 CL5.
Just another P5B bug. I re-did my memory bios timings page in the above screen shot. If this item causes problems then I think it's best to leave it disabled.
Just for comparison, I ran Everest at DDR2-1000 CL4.
http://img529.imageshack.us/img529/5...1000cl4or5.png
DDR2-1066 CL5 is a hair better, not that anyone would ever notice.
CL5 also needs about 0.10 to 0.15 less memory voltage so I'll be sticking with that.
Edit: I also figured out that the Static Read Control setting in the bios changes Performance Level as reported by MemSet. When set it to Disable or AUTO, Performance Level gets set to 13. When set it to Fast, Performance Level changes to 12 and when it is set to Faster, Performance Level changes to 11. You get a couple of extra MB/s of memory bandwidth but that's about it. With the P5B, it's hard to notice a significant difference whether this is set to 1 or 15. If you have any "unable to boot" issues then set it to Disable and if you don't have any problems then I guess set it to Faster.
wow, i really like seeing the old chipset doing so well, have to say that my own p965 does just as good but it struggles with my sticks. so using a 45nm wuad wiht this board is kinda problematic?