4th install. Progressive increase of BSOD over time

Well, I wish I could remember everything I had just written in this box, but it's gone now.

I had a stand alone video editor built. (See specs ) The first operation BSOD'd in about 4 hours, but didn't do it for quite some time after that. Then it started dropping the video, then it started crashing IE on opening, then it started crashing the editing programs. I searched this and other forums and eventually replaced the EVGA GT 650. That corrected the video drop and even the crashes for a while. Then the crashes began in earnest, escalating until the BSOD's joined in the fray. Since windows does not have the program needed to read their own dump files, and all other attempts at correcting failed, I did a "repair install" from the disk. That worked for a while, but the issue progressed in much the same manner as before. I finally did a reformat, which performed nearly identical to the initial operation, with a 4 hour onset of a BSOD, and then nothing. That is until the programs started crashing and the BSOD's appeared again. I then had the guru wipe the drive completely and reformat. The only difference in performance has been that it didn't blue screen after 4 hours. It did however crash IE on start-up, as well as the editing program.

At no time has this machine ever run fast. I have an XP Home machine next to it that would run circles around it in terms of video editing, both machines running Corel Video Studio ProX4. I have since upgraded to ProX6, which is a 64 bit program, but it runs even slower than the Prox4 did. It doesn't crash as often but a 1.5 second latency is outright annoying.

I have searched the threads here for hours trying to find a similar problem. I found recommendations for removing Pinnacle, "alcohol", ASUS Gamer, using scannow and some others, all of which I have tried to no avail. As to scannow, yes it was in the root, but it can't get past 80% of the initial scan. In safe mode, however, it goes 100% and finds no issues.

Many thanks to the Cap'n for the SF Diag program. I still don't know what I'm doing, but at least I have something that gathers info that a real person can look at!

Speaking of not knowing anything, I found it odd to find this string:
Built by: 7601.18409.amd64fre.win7sp1_gdr.140303-2144
That "amd64" in there may not be what I think, but this machine runs an Intel i5.

With best regards and hopeful anticipation,

Jeff

Thanks, I see!

I was waiting for some more replies before I went to the install forum, but it's been 3 days now. I guess I'll try posting there.

Thank you again!

Jeff

Hi there,

Sorry for the delay,

These are memory management BSOD's,

Code:

BugCheck 1A, {41790, fffffa8003be2030, ffff, 0}

Probably caused by : ntkrnlmp.exe ( nt! ?? ::FNODOBFM::`string'+35084 )

Followup: MachineOwner
---------

1: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

MEMORY_MANAGEMENT (1a)
    # Any other values for parameter 1 must be individually examined.

All I can suggest is testing your RAM until someone else comes along that can dig deeper than I can.

   Information

MemTest86+

Run a minimum of 8 passes, preferably overnight, or until errors occur using MemTest86+.

Use this tutorial to help you use MemTest86+:
RAM - Test with MemTest86+

Running memtest shows that it's rife with errors. Preparing for the next step, I got my Asus manual to check that they were installed in the proper slots to begin with. This is when I noted the "qualified vendors list" and compatibility listing. The Patriot PC3 is not on the list for 1600 MHz cards. I'll run the tests as prescribed, but I have to ask if it's possible that the card is good, but not compatible. Further, which frequency would be optimal?

(10 minutes after the above paragraph) Memtest immediately went in the red, piling up a list too fast to possibly read. I stopped the test and pulled one card. Started the test and it began running without any errors, so I stopped it and changed slots. Started, no errors. Changed card. Started, no errors, changed that card to the first slot. Started, no errors. Put both cards back in their original slots and started memtest again. No immediate red flags. I'll let it run all day, but it's looking as though a card wasn't fully seated or a bad connection somewhere. I'll know more at the end of the 7 recommended passes.

Sounds like some progress has been made, I'll get someone over here to have a look and answer your questions on the RAM.

Just a little add if you don't mind, make sure you're running Memtest86+ because I've seen false positives generated by the first memtest86 program where it generates errors when the RAM isn't actually faulty.

Quick update before hitting the hay. 28% into pass #7 with no errors

Doc, I am indeed running memtest86+. I just tend to abbrev.

S&B, I am quite limited on computer savvy. Quite limited indeed. My curiosity was to the particular ram not being on the list, therefore perhaps creating a potential. "Not likely" seems to be the call of the day! I'm running 2 identical sticks from the same package, so no cross contamination there. When the first attempt was immediately nothing but red, then I removed the B2 card and had no red, I suspected either that card or that slot. I figured it it was that BAD, then surely it will show immediately if I put that card in the A2 slot, but it didn't. I then pulled the B card and put the card from A2 in the B2 slot, and again no immediate errors. This made me think that it could very well have been a faulty connection due to my inept installation skills, so I placed each card back to the original configuration. It is currently on pass #7 with no errors. I'm letting it run all night as well. We'll see what comes to pass in the morning.

ICit2, I'm not in the market for RAM, unless, of course, I HAVE to!

Jeff

"Performance" is the pursuit of a toy that won't break.