Asus P5B-E 6400PRO or 6400C4?

[Darth Evader]

I just ordered an Asus P5B-E and TWIN2X2048-6400C4 memory because that is the memory the Corsair Memory Configurator listed for this board. Today I was looking at the Asus Qualified Vendor List for the P5B-E http://www.asus.com.tw/999/download/products/1347/1347_10.pdf and I see that they list TWIN2X2048-6400PRO as being the preferred memory for this board.

 

Now, I am concerned after reading other threads on here relating to the P5B-E motherboard and Corsair memory. Did I make a mistake in the model that I ordered?

 

If not, what do I need to do in order to get the C4 memory to work properly? I've read some posts saying that they P5B-E won't even POST if C4 memory is installed prior to changing the BIOS settings. Is this actually true?

[RAM GUY]

You are reading that wrong, that is not a prefered list just what they have tested!

[Darth Evader]

Cool, thanks. So, from reading your other posts, this is what I plan on doing before even trying to install on OS:

 

• Flash to most recent BIOS
  
• Set voltage to 2.1 Volts
  
• Change "Write to Precharge delay" in the BIOS (same page as for latencies) from default 10 to 15
  
• Set latencies to Corsair specifications
  
• Disable legacy USB
  
• Run Memtest86+ on each stick separately to make sure that neither produces errors.
  
• Run Memtest86+ on both sticks, in Dual-Channel mode.
  

 

Do you foresee any problems with this, such as the motherboard not being able POST on first boot-up due to the fact that the memory settings haven't been entered into the BIOS yet?

 

BTW...I went with the E6400 Conroe and a 750 Watt Thermaltake PSU.

[RAM GUY]

don't see any here...

[Mike Mendelsohn]

I just built this system and it booted right up with the default settings, but it occasionally locked up (couple of times a day). I was running on the default bios settings so decided to try setting them to the corsair specs.

 

I tried the following:

 

Set voltage to 2.1v - seemed OK, but still crashed

 

Set timings to 4-4-4-12 - wouldn't even boot

 

Followed the directions in the posting above (added USB disable and set write prefetch to 15) - booted, but failed in Memtest86+ within first couple of minutes (start of test 4)

 

I now have the machine back at the default settings and am wondering what to do to make it more stable.

 

I would also be interested in a utility that would report the [auto] setting for the voltage and other parameters.

 

Mike

[Darth Evader]

Mike, I throught I would follow up with my original post so you would know what happened to my setup.

 

I installed the system exactly as those above instructions outlined and I never had one problem or memory error. I ran memtest86+ for 24 hours with no errors and also prime95 for 24 hours with no errors. 4-4-4-12 memory and 2.1 Volts, etc. all worked perfectly for me.

 

I think you should try doing the following steps as a test:

 

-Remove one of your sticks of RAM

 

-Go go into the BIOS and configure for the above mentioned settings (also, make sure you have the newest bios)

 

-Run memtest86+ on the single stick and see if it makes it through without error.

 

-When the test is done, shut off the computer, remove the memory, and install the other stick.

 

-Repeat the memtest86+ on the other stick.

 

This will at least tell you if one of the sticks might be bad. Then, you can go from there.

[RAM GUY]

That is exactly what I would suggest as well. Thanks Darth Evader

 

If you do find you have a failing module, let's get them both replaced. Please use the RMA request found in TSXpress or follow the link in my signature “I think I have a bad part!” and we will be happy to replace them or it! And please note that you are from the forum!

[Mike Mendelsohn]

I followed Darth's suggestion and tested each stick individually under the tested settings listed on this web site. The only setting I couldn't do was disabling the Legacy USB since this shut off my keyboard, so I couldn't control the test.

 

I ran one module(X) overnight in channel A and it tested OK. I then tried the other module(Y) in channel A and the machine halted near the end of Test 5. I tried it again and it halted in Test 4.

 

I also tried testing both modules in channel B with similar results. The first module(X) made it past one full pass and is still running. The other module(Y) had a halt in test 4 in one run and failed with error bits ff000000 in two other runs in test 4.

 

Just for kicks, I tried both modules in dual mode and the test failed within a couple of minutes.

 

Sounds like RMA time to me. What do you think?

 

Mike

[DerekT]

I would perform the test with another keyboard. Legacy USB has proven to return errors on Memtest.

 

You could well find the same issue repeating itself when you insert the newly RMA'd module.

 

I have personally found this to be the case. I just mention this to hopefully save you the time on a RMA and still have the issue remaining.

[Mike Mendelsohn]

OK, I didn't understand that the legacy USB thing was a memtest issue. I'll try disabling it and running memtest again.

 

Assuming that the memory passes, I should reenable it before booting to Windows, right?

 

Mike

[Mike Mendelsohn]

That was not the problem.

 

One module tested OK with or without the USB Legacy enabled and the other failed either way.

 

It was worth a try anyhow.

 

Mike

[RAM GUY]

Let's get them both replaced. Please use the RMA request found in TSXpress or follow the link in my signature “I think I have a bad part!” and we will be happy to replace them or it! And please note that you are from the forum!

[Mike Mendelsohn]

I chose to have Newegg replace the memory since it was brand new and they would send me a replacement before I returned the others so I could continue to run my machine on one stick instead of having it dead in the water.

 

The new memory tested perfectly with the suggested settings and I'm a happy camper again. THanks for your help.

 

Mike

[RAM GUY]

NP Please let me know if you have any more questions!

[Darth Evader]

Sorry to bring my old thread back from the dead, but I have a question. Since I've installed Vista on this PC it's been freezing frequently when under heavy utilization (backing up hard drive, recording video while surfing web pages, etc). By freezing, I mean completely frozen..mouse freezes, keyboard doesn't respond, no errors are written to event viewer, no blue screen... Memtest86+ never generates errors nor does Prime95.

 

Anyway, after troubleshooting everything under the sun for the last couple of months (heat, drivers, etc.) I think I may be on to something. Back when I made this thread the P5B-E was new and at that time the agreed upon recommended settings were 4-4-4-12 at 2.1V. I've noticed that in recent posts regarding the P5B-E, the recommended settings are 5-5-5-15 at 2V.

 

Do you think that it might be helpful to relax to the 5-5-5-15 2V settings? I think Vista might be more aggressive on the memory than XP was. Here are the settings that I've seen most people agree upon:

 

CAS = 5

tRCD = 5

tRP = 5

tRAS = 15

DRAM Write Recovery Time = 2

DRAM TRFC = 42

DRAM TRRD = 10

DRAM Write to Read Delay = 10

Rank to write to Read Delay = 10

Read to Precharge Delay = 10

Write to Precharge Delay = 15

 

Does this look good?

[DerekT]

You may wish to check your hard drive(s), Video card, as well as power supply. Memtest shows no issues on the DRAM and Prime95 shows that your CPU is working within operational parameters. This does not mean the system is stable, just that the DRAM and CPU are so you need to test the rest of your system. Please enter Here and post your system hardware components being as specific as possible. 6400C4 version 2.xx rather than 6400, etc...

[Darth Evader]

I've updated my configuration on the settings page. When the freeze occurs, here are my stats according to Asus PC Probe II software:

 

Vcore: 1.20v

+3.3: 3.23v

+5: 5.04v

+12: 11.98v

CPU: 47 deg C

MB: 46 deg C

CPU Fan: 2537 rpm

 

I have tested both hard drives using the manufacturers diagnostics on the "long" test and neither have errors. I have tried different video and sound cards. I have tried different DVD burners. I've updated all drivers, including chipset, for all devices...to no avail.

 

As I mentioned, it is a hard freeze and Windows doesn't get a chance to log any errors in Event viewer nor are any logged in the Reliability reports.

 

So are you saying that the timings should work on either 4-4-4-12 or 5-5-5-15 in this motherboard? I switched to 5-5-5-15 last night and it hasn't frozen since then, but I'll post back if it does.

[DerekT]

So are you saying that the timings should work on either 4-4-4-12 or 5-5-5-15 in this motherboard? I switched to 5-5-5-15 last night and it hasn't frozen since then, but I'll post back if it does.

 

I am saying that DRAM is passive and it has two duties. To store and retrieve data. Memtest performs that function greater than any other software will and as such is the industry standard in memory testing. If the Memtest passes the tested functions of data storage and retrieval without errors, then any freezes will be external to the DRAM. And yes, that board should handle the 4-4-4-12 @ 2.1V of the 6400C4 DRAM. It seems to me the on motherboard memory controller is having issues with the ratio and timings, if by raising the timing, you are bypassing the issue. My advice? Well, if it were my system, I would RMA the mainboard. Your mainboard temperature is running rather hot as well. You are using the Intel stock cooler?

[Darth Evader]

Thanks DerekT. I am going to let it run for a week and see how she does at 5-5-5-15 2V. I've seen it go 5 or 6 days in the past with no freezes...so the fact that it made it for a few hours at 5-5-5-15 2V doesn't give me much confidence that it's fixed.

 

If after a week it seems to be stable, I'll probably leave it as-is and limp along until it's time for my next upgrade. If the problem still exists I might just purchase a new motherboard. As you noted, memtest86+ indicates that the memory itself is good, so it just might be the memory controller or something else on the motherboard. All of this hardware was purchased in October 2006, so the MB may be out of warranty for an RMA.

 

As for the MB temps, I thought they were pretty high too. I don't know exactly where on the MB this reading is taken from. Other than taking the sides off of my case, I don't know what else I could do to cool the MB itself. I'm not overclocking or anything and I always thought I had a nice Xaser III case, but maybe the case isn't doing a good job.

 

The CPU was running 50+ degrees with the stock HSF, so I replaced it with an aftermarket model (can't remember the name) and Arctic Silver applied per their instructions. The CPU temps now range between 40 to 47 degress.

[DerekT]

You are at the higher end in idle thermals, but not a drastically higher end. You might wish to look towards adding a few 120mm fans in your case just to move the air around and out. I think that it is a good idea to see if stability is gained in the long term. Has this issue (freeze) only been happening since an install of VISTA? All fine before the VISTA install? No changes to hardware and no freezes before VISTA? Such a hard lock is not easy to diagnose but if no hardlocks occurred before the install of VISTA, and no hardware changes were implimented, then I would look at a possible downgrade to Windows XP. It's an easy fix. VISTA reacts differently in that the thermal output is higher than on XP because of hardware requirements in Video, Audio, and idle processor usage and possibly there is a thermal issue, doubly so when you say in "Periods of Heavy Use".

 

So let's start the diagnosis there.

Download Core Temp

Here

Download OCCT Here

Here

Download CPU-z from

Here

.

Please post screenshots of CPU-z's CPU, Memory and SPD tabs. Run OCCT and enter the settings button, set the monitoring to "Built In" and choose 65C as the max CPU temp. Let OCCT Run for the one hour. Please post a screenshot of Core Temp when OCCT has run for ~20 minutes.

[Darth Evader]

Occasionally, I would experience this same freeze when using XP. It freezes more often under Vista...which I assume is due to increased system utilization.

 

I ran all of the tests you asked and the screen shots have been attached to this message. As reference, the CPU Core temp was at 39 C when it was idle before the test began.

[RAM GUY]

I would remove the sound card and use the On board sound for a few days and see if th system still locks. But as DerekT stated if it's passing http://www.memtest.org I really doubt its the memory that is causing the lock ups not to mention that the MB and the O.S. are designed to recover from a bit failure in memory and should not hard lock the system, normally you would get some type of Error or BSOD if it were a memory error. From what you have posted if the lock ups still continue I would change the MB or CPU.

[Darth Evader]

Hi Ram Guy. I failed to previously mention it, but I have tried removing/uninstalling the Sound Blaster and going with onboard sound but that didn't help.

 

I think your right that this may just be a motherboard issue. I'll give it a shot leaving the mem settings to 5-5-5-15 2v and see if it remains stable. If not, i'll look at getting a new motherboard...if that doesn't do it, then I'll go with a new CPU.

[RAM GUY]

Have we tried to replace the modules once? If not we can try that as well if you like?

[Darth Evader]

No we hadn't replaced them, they are the original sticks I got from Newegg. My original reason for starting this thread so long ago was simply to ask which was the best memory for the P5B-E board.

 

Now, after all of the testing I've done, I really don't think the memory is the problem...but if I end up with a new motherboard and CPU and the problem still happens, I might take you up on your offer for replacement. I'll check back in here when either A) it freezes again or B) it seems that 5-5-5-15 resolved the issue.