AppHangB1 FarCry2/COH crash

Hey all,

If you've been reading the gaming forums lately, I'm sure you've seen a few of my posts...I'm having crashing problems with Ultimate64. I've updated my video driver, sound driver, chipset driver, BIOS, windows, game (COH), ran MEMTEST (passed), downloaded 3DMARK vantage and almost sacrificed a small goat to the computer gods...all in the wish that my computer would just WORK.

I've been having freeze (with audio loop/BSOD) problems in Company of Heroes (a million times), FarCry2 (once) and browsing on the internet (once). Every time it's the same: a random complete system freeze out of nowhere. Every once in a while my system will just randomly restart too. Sometimes (only if the screensaver pops on) AFTER benchmarking my system I'll get the same freeze. It never freezes during benchmarking (no matter how many times or how extreme) tests, but it has frozen twice after completion.

My temps should be fine (never seen the system past 55), and my system specs should be good too. This leads me to believe that I have some sort of DX10/Vista problem.

I've got an error code here, and was wondering if anybody knew a solution. I'm figuring that whatever shut down FarCry2 is shutting down my other apps (no error codes found for them). PLEASE HELP...for the goat's sake!! :D


Problem Event Name: AppHangB1
Application Name: FarCry2.exe
Application Version: 0.1.0.1
Application Timestamp: 4942e3f6
Hang Signature: d561
Hang Type: 2048
OS Version: 6.0.6001.2.1.0.256.1
Locale ID: 1033
Additional Hang Signature 1: 733c2cc5bd93d5915757eda66dc635ad
Additional Hang Signature 2: 9610
Additional Hang Signature 3: 3ff4596104d308ee985f9136cb173ec6
Additional Hang Signature 4: d561
Additional Hang Signature 5: 733c2cc5bd93d5915757eda66dc635ad
Additional Hang Signature 6: 9610
Additional Hang Signature 7: 3ff4596104d308ee985f9136cb173ec6

Extra information about the problem
Bucket ID: 413740200

First, nice case and your specs are good.

Even though your not overclocking, have you manually entered your settings in BIOS or have you left them at AUTO?

How many passes of MEMTEST did you run?

I've had similar freezing / restarts due to memory timing.

If it's freezing at the desktop (browsing) it's unlikely that DX10 has any bearing.

Does COH and FC2 work in DX9 mode or are you crashing there too?


....The Comptuer Gods....they want that goat

UPDATE: I just ran MEMTEST86+ for 10.5 hours with 12 passes and no errors. Should I even bother with trying to find out if I have a dead stick? That may seem like a silly question, but...yeah, well, it might just be a silly question, lol

Another quick (and noobish) question: when installing BIOS updates, should I get the very latest update, or just install one at a time, until I end up with the latest? (my MOBO has 3 updates). My buddy just told me to install the latest, but I want to be sure. Thanks.

My real concern now is my my memory timings and a 'memory leak'...any ideas?

Thanks again

*******

LOL, and the goat they shall have...I've about had it up to here with this problem...

my settings in BIOS are all auto. That could be a path worth checking...My MOBO says it handles 1600Mhz tri-channel, but I've heard of memory incompatibility before. In fact, that's what one of the BIOS updates was for. However, I'll dig a little deeper and see what I come up with. One of my buddies who works for a big networking company out here told me that it also may be a memory leak (ex: firefox is using far too much memory, etc)...I just tested this and found that Firefox runs one page at about 56,500k while Internet Explorer runs a page at about 8k. Huge difference! (At the time of me checking Firefox was taking up almost 200,000k). Could this be a problem?

Checking my 'performance' tab in 'windows task manager' I see that I have 6134MB of physical memory, with 4858MB cached, and 236 free (with 1.4GB usage) all while running two pages of Firefox. This does not seem right to me.

I ran MEMTEST86 for about 2.5 hours, and I believe it was only one pass. I'm a noob, so I'm honestly not really sure how it works...now that I'm reading up on it, perhaps I should have run MEMTEST86+...I'll go ahead and run it tonight and post back.

I havent tried DX9 on FarCry2, but COH will freeze if I flippin' look at it wrong...lol, so, yes, it does freeze in both modes.

So you're thinking that its a memory thing then, hmm? Timing, or a dead stick perhaps?

Thanks for your input, I appreciate the time

Sal

If the memory got through 12 passes of MEMTEST with no errors, then your RAM should be good. Highly unlikely that you have a dead stick

Just install the latest BIOS available. You can always install older versions if the latest one is no good.

It might be a good idea to manually enter your RAMS manufacturer rated timings (and voltages) while your at it. Some boards ideas of AUTO can be a little off.

The memory usage in Task Manager actually seems about right. For comparison, I have a total of 4094 / cached 3231 / free 33 and 1.12GB used with only firefox open.

The Firefox memory usage seems normal too. it is prone to leaking however. 200,000+ is normal if you have quite a few tabs open / closed in a long session.

If the RAM is not the issue, I'd take another look at your sound and settings.

Is this the driver you are using?

ASUSTeK Computer Inc.-Support-

:SoundMAX Audio Driver V6.10.2.6520

It looks like a multi driver so seeing *32 in MSCONFIG is normal.


It will be simple once you work it out. It usually is

Wow, no kidding.... Live and learn, lol...ok, I'm all over the map in this post, so forgive me if I don't make sense. I'll keep editing it untill I do, lol

The link you sent me just routed me back to the generic driver page (it always does), but I'm almost positive that you're right in saying that it's a multi-driver. The file that I downloaded was the only one I could download from ASUS, and it promises that it is Vista64 compatable. However, when you open it, it's a bit confusing for a noobish man like myself.

In the download there were two file folders: 1) DTS 2) NON DTS. Now, I have no idea what the difference between the two was (I tried Googling it), but inside the NON DTS folder were 3 other folders: (1) 32 (2) 64 (3) Vista. I installed the 'Vista' one, as it seemed that the 64 bit version was only for XP. (I read that online quite a bit). Indeed, it seemed pretty self explanitory: Install the Vista one. However, being a noob, I honestly cannot figure out how to find out exactly what driver I am currently using for SOUNDMAX.

What would be the harm in trying the 64bit XP version? Oh, by the way, I found a guy with almost the same rig spec as mine having the exact same problem on Asus' messageboards...hmmm....

Unfortunately, I have another few questions about the RAM issue (surprise surprise, lol): when I was running MEMTEST86+, the test showed my timing at 8-8-8-20. For some reason, I wrote that down. Now when I go to newegg/corsair to check out my memory specs, it says the timing is 9-9-9-24. Could this be the problem? MEMTEST86+ got all of my other information correct (i7, 6gb ram, etc), whereas MEMTEST86 did not (it said I had a Pentium 3, lol)...So I'm thinking that my MOBOs idea of 'auto' could be the wrong timing. Thoughts? (voltage is supposed to be 1.5 btw)...

If this is the case, how would I correct it? In the BIOS there are numerous settings, but I'm a bit scared to start punching in numbers. I'll investigate.

Anyways, thanks so much for your help once again

Sal

DTS stands for Digital Theater Systems. It's a higher quality (less compression loss) than Dolby Digital. The onboard soundmax on the P6T Deluxe should be capable of it. It depends on your speakers / amp whether or not it will work.

The worst that could happen by trying the XP drivers is that it doesn't install or the sound stops working. Just create a restore point before trying to install the drivers.

To find out what driver you are currently using go to:

Control Panel>Device Manager>Sound, Video and Game Controllers and right-click on your soundmax and select properties. Then select the Driver Tab.

When you had the crash browsing the web, were you listening to music etc? Anything using the soundcard?

***

Is this the RAM?
Triple Channel DDR3 (TR3X6G1333C9)

If so then the defaults are indeed 9-9-9-24 @ 1.5-1.6V


I'm not familiar with your particular BIOS but your RAM settings should be under the DRAM Timing Control section:

If you select manual they should currently appear as:

CAS Latency = 8
RAS to CAS Delay = 8
RAS PreTime = 8
RAS Act Time = 20

you can then change them to:

CAS Latency = 9
RAS to CAS Delay = 9
RAS PreTime = 9
RAS Act Time = 24


Before changing anything, you should be save your current profile in the O.C Profile section. That way if any changes don't work, you can quickly restore your working settings.

I'd double check the *exact* model numbers of your RAM just to be sure.

Run MEMTEST86+ again (2 passes should be enough) and if no errors are found, try it out in Windows. If it there are errors, then it's possible that other AUTO settings could be causing grief.

Since your using 3 DIMM slots, you may even need to bump up your NB (North bridge) voltage a tad to give enough voltage to all the DIMMS.

Check out some Overclocking Sites with your specs /board to get an idea of what settings others are using. It's a lot to take in at first, but you'll get there in the end

lol, no kidding...I tried googling how to find my current drivers and I never could find the answer...I always wondered why they weren't in the same spot! Thanks, that helps quite a bit.

Yes, when it crashed I was listening to a song on YouTube...and it has most often crashed while doing something that has to do with sound. However, I recently completely uninstalled my SOUNDMAX drivers in order to prep a clean install of new drivers. After doing this I left the computer running all night, and when I woke up this morning I had the same WSOD. (It was technically grey).

The only programs that were running were a few (maybe 5) Firefox tabs. This has happened before (my system would benchmark over and over, but if you let it sit after the benchmarking, it would freeze) but I had thought little of it...but now it has happened enough to think that this is the direction I should go in. Strangely enough, I can google, troll forums, or search through my computer for hours on end, but it *generally* only freezes up when I play a game, do something graphics intensive, or let it sit. Sound wouldn't have anything to do with that, correct? Overheating? (I don't really think so...temps are always around 40*, I have a great case, and great cpu cooler)... Or more likely, memory issues? Or even a GPU issue? Your thoughts on the direction I should go would be incredibly helpful.

That is my RAM, yes. One last RAM question: There always seem to be the 4 ram timing numbers: X-X-X-XX....however, when going into BIOS, there seem to be many number after those initial 4. Is this the same idea as PI is abbreviated as 3.14? Do I have to key in those numbers manually?

I was pounding the OC forums a little before all of this to find a good BIOS setting, but I was in way over my head...however, after all of your help I might go get a job programming! lol

Thanks again,

Sal

What benchmarking apps are you using? (the one's that freeze afterward) I remember stumbling across a thread where someone had trouble crashing after succesfuly running either stress testing or benchmarking. I *think* it was due to memory, but I can't be certain. It was a while ago when I read it.

When you uninstalled your sound drivers, did it crash with NO audio device installed, or did you reboot in the interim and Windows installed it's own drivers?

If it still crashed with no sound drivers/device then you can probably rule out sound as the cause ATM.

You can also pretty much rule out overheating. If you've got a HAF 932, then I know it's a good case. I've got one myself BTW, are those temps idle or load? If the CPU was overheating, then the PC would just shut down. No BSODS etc.


Those sub-timing in BIOS allow you to tweak your RAM even further, but generally you'll only need to adjust the first 4. (I've got my sub-timings set to AUTO)

If the PC is freezing whiulst gaming, I'd still be looking at RAM, because games will utilize more memory than normal desktop use would and therefore make you crash quicker




Even though you aren't overclocking, it might be a good idea to run some of these to see if those AUTO (including CPU) settings are causing the errors:

Prime95: (Blend mode checks RAM more. 8hrs+ for real stability
Prime95 Download - Softpedia

Intel Burn Test: (Good quick and Punishing CPU test. More CPU based than RAM, check if your CPU AUTO settings are ok. *Caution* Watch your temps with this )

RealTemp 2.9: (if you aren't already using it to monitor temps)
http://www.fileden.com/files/2008/3/...alTempBeta.zip

GPU-Z to monitor your cards temps. ( You can also log the temps while in game to see if it's spiking to high.)
http://www.techpowerup.com/downloads/SysInfo/GPU-Z/



CPU-Z (check the memory tab to see your timings)
http://www.cpuid.com/cpuz.php


These two stress testers are no magic bullet. They wont jump up and say "I am broken. Fix me!" but they will help narrow down the area where something is failing.

Also, do you have SP1 installed?

First for the sound: my system crashed when I had no sound drivers as well as when I turned my sound completely off in BIOS. So, I guess sound isn't the problem.

The benchmarking app I'm using is 3DMARK vantage...Apparently its the most advanced, just came out, etc etc...apparently the thing even requires vista and the i7 chip, if I remember correctly.

Memory seems to be the culprit, doesnt it? I mean, I'm by no means an expert, but if it was a GPU problem, it seems as though it would crash during the high stress of 3DMARK vantage...My rig dipped into the single digit framerates, but kept going and going, no crash. If it was a PS issue, my whole system would shut off, correct? I mean, I have 1000w, and I'm running only one GPU...can't be power...it just seems to make sense that my memory is being depleted faster than it can replenish itself...However, my system DOES seem to crash at random intervals, including right away. I don't know what to think.

LOL, yeah, it is a good case. I was thinking about the antec 1200, buuuut...I just thought this one was better. Not as pretty, but oh well. My temps (I use PC Probe II from Asus) at idle are around 35*, while in game are around 45-50*. I've got all the fans on max, just to be sure. I mean, this thing feels like an air conditioner!

I ajusted my latency times and my rig still froze up (with sound disabled). So I don't think that that was the problem.

I'm going to run XP on my 1TB, maybe by tomorow, and see if these issues keep happening. (I have Ultimate SP1, yes).

How can I check to see if my MOBO is reading my RAM properly? I'm starting to read a lot of distressing things about my MOBO online...

I'll try those apps you sent me, and get back to you with the results. Once again, your help has been invaluable, thank you.

Sal

3D Mark Vantage only needs a DX10 capable card and Vista because of DX10. It's not i7 specific The single digit test was probably just the CPU component test (planes flying through circles). All PC's dip in to single digits on that one.

Random crashing is common with RAM. When I built my PC about 2 months ago, my RAM timings were off just enough to crash randomly at the desktop (sometimes quickly and sometimes hours later) and almost instantly in games.

Ditch the Asus probe and use Real Temp. It's more accurate

CPU-Z is the one to use to check your memory in Windows.

The thing with building your own rig is that errors can be a PITA to track down (as you know) By troubleshooting and eliminating the main culprits, it helps track down the real culprit. Even though RAM seems to be the most likely suspect, it may turn out to be something completely different. Often it's something so simple, you kick yourself for not thinking of it.

I considered the 1200 as well. I'm glad I went for the HAF though. Your right, it's not pretty but it's huge inside, has airflow up the ying yang and is a joy to work with.

When your done, you should post some pics in the Show us your rig thread. It'd be good to see another HAF in there

http://www.vistax64.com/general-discussion/110939-show-us-your-rig.html

Ahh, I see. Is 3DMARKvantage a good GPU/CPU tester? It seemed like a good one to me.

So get this...I was checking out Asus' web site last night (right before I was going to download your apps/links) and my computer just simply restarts, completely out of the blue. The only programs I had open were Firefox sites. So, nbd, I get restarted, and get right back on to Firefox, and that Asus web site. BOOM. Restarted again! Third time's a charm, I thought, and I tried again. Restart. So I just shut the rig down for the rest of the night.

The only time it restarted like that (yesterday) was when I manually put in my 1333 Mhz in BIOS instead of leaving it on AUTO. I booted the thing up, and within like, 10 seconds it froze up on me (sound loop, orange [???] screen of death this time) while watching 'caturday' episodes on YouTube(lol, I know!). *sigh*

I'm on IE64 right now, and havent had a problem (yet). WTF is that? Ahhh!

I'll def get some pics of the rig up, for sure!

vantage is good to get an 'idea' of performance but not the best for stress/stability testing. It's also biased towards Nvidia cards score wise, so your in luck there. It's also good for e-peen bragging

Off the top of my head, I can't explain why IE seems more stable than FF. But the fact that after changing some RAM settings, you experienced similar crashes (even if they were worse ) still tends to point towards RAM. You may even need to adjust some different BIOS settings to compensate. I haven't really looked at the new i7 QPI settings to see how many of the rules have changed regarding strapping etc.

When you manually set the ram to 1333mhz, did you adjust the timings as well, or just leave them on auto?

You appear to be getting a few different colours with your freezes. What driver version are you using? Perhaps some of the drivers aren't playinf nice with your factory overclocked viedo card. I'd suggest trying a few difference forceware versions to see if that makes any difference. Assuming it doesn't crash while your downloading

I think that, yeah, it has to be either my memory or my GPU overclock. I'm wondering if my AUTO settings in BIOS are screwing with the correct voltage requirements for my GPU at load...To your knowledge, what other AUTO settings would need to be changed for an overclocked GPU?

I've done so many tests, I honestly dont remember if my timings were changed or AUTO when I manually inputed my 1333, lol...I thiiiiiiink they were, but I'm not 100%. I've been good about staying on top of the 'scientific method' of doing this, but I guess I forgot that one! I'll try 'em both.

I'm using the latest drivers on EVGA's web site. I WAS using NVIDIA's LATEST driver, but the crashes were still happening, so I went with EVGAs because of my factory overclock (I was wondering if my OC had something to do with it). (driver #: 7.15.11.8048).

Forgive me, what what is 'forceware'?

I downloaded those apps that you sent me btw (cpu-z, gpu-z, p95, crazy stress-tester thing for cpu) and they work GREAT. I love 'em. Havent run p95 or burntest yet (the download site said it would make my hair fall out!!! lol) because I'm a bit scared. My temps are currently in the low 30s for my cores, and the mid 40s for my GPU. I've googled 'acceptable i7 temps', etc etc, but I'm still concerned about it. In your opinion, how high should I let them go before I stop the tests? Intel's BurnTest seems especially dangerous...

Thanks again man,

Sal

Oh, and whats this I hear about Queensland wanting a tourist blogger? It's all over the news out here in California that you Auzzies are gonna pay somebody to chill out on a beach for 6 months and write about it!!! Sounds like I might have found my next career choice....:eek:

The overclock on your video card *should* be independant of your motherboard BIOS. I believe that the overclock is set in the cards own BIOS. TBH, I'm unfamiliar with OC'ed cards and their voltages

Forceware is just what nvidia calls their drivers. Just like ATI call theirs Catalyst. Do the EVGA drivers seem more stable?

The 'scientific method' is somewhat alien to me. I start out with good intentions but end up going 'crazy scientist' and wondering what the hell I've just tried and end up doing the same things 2-3 times

Definitely try manually entering the timings, speed and even voltage at the same time.

For 'safe' core temps, you probably want to keep them 70c and below. I believe that the i7 chips run a little hotter than C2D's. If your still using the stock cooler, then your temps are going to rise. For your piece of mind, keep them below 70c although a short little trip into 85-90c won't damage your chip. Permanent temps like that however....

IBT and P95 don't touch your video card, so you don't have to worry about that.

I don't know about that blogger job... I wouldn't mind it myself. I don't have to tell them I already live here

Ohhh ok, 'forceware' is kinda like 'sli' or 'crossfire'...confusing, fancy sounding terms to befuddle people like me...

Hmm. My CPU WAS running around 85 for the first few days I had it installed...I would say that, overall, the time it ran those temps would have been a 5 hours total...not bad, but not good. It ran fine, but kept giving us temp warnings. Eventually the whole rig just shut down..."well, time to get to work", my buddy said. We couldnt figure out what we did wrong, so we tore the whole rig apart, and found that we put on spacers that didnt need to be put on, which prevented us from screwing down the CPU cooler all the way. As soon as we got those spacers off, redid the thermal goo and slapped the Coolermaster V8 back on, temps were fine, with little variation.

Now the scary question: I've read that the i7s are pretty tough, but is it possible my chip is somewhat fried? I suppose those apps you gave me will test it out...*sigh* :huh:

I'm gonna run some of those tests right now, see how I do.

Oh, GPU question: GPU-z says my GPU fan is at 40%...would you happen to know how I can turn that fan up to 100%? I would imagine that it does it on its own, but I'm curious if I can manually do it. :sa:

Thanks,

Sal

PS...all you have to do is send in a video application for the Queensland thing! LOL, should be cake, theres only like, oh, 389475390753002378 people applying for it

PPS: just went into intelburntest (have NOT run it yet) and it said that I had 4147 MB of 'physical memory' avalable. I'm currently googling this, but just to make sure that I'm not missing something, would you mind giving me a brief explanation as to what 'physical memory' is? Is it saying that I'm missing a stick, or that I have 4GB of 'free' RAM (as in, not being used). Thanks again man

"Physical memory: Also referred to as the physical storage or the real storage, physical memory is a term used to describe the total amount of memory installed in the computer. For example, if the computer has two 64MB memory modules installed, it has a total of 128MB of physical memory."

Is this correct? If so, it seems to be saying that it only reads 4GB in my system.

SLI and Crossfire are dual / triple / quad Video card configurations. (more than one physical card installed and linked together)

SLI is the Nvidia version and Crossfire is the ATI version. Forceware is just 'series' name of drivers i guess you could put it.

5 or so hours at 85 shouldn't have hurt it. As you say, not bad but not good either

For the fan speeds, most drivers should automatically increase the speed when under load. It depends on the drivers. I haven't got an nvidia card anymore, so I can't say wheter there is a fan speed control in the Nvidia Control panel or not. There are apps like RivaTuner that allow you to adjust fan speeds I believe.

RivaTuner


yeah, I doubt I'd have much chance passing for a tourist anyway

Lol, yeah, I had to google what sli and crossfire were when I was building my rig. I guess they just wanted to give their multiple GPU configs a cool name

Anyways, its good to see you online! I won't take much of your time, I know you've got things to do, so I'll keep this short. As of right now I'm a bit concerned about the 'physical memory' being at 4GB (it keeps changing in value every time I open the program, from 4147 to 4182, 4233, etc etc)...So I think I just may run 95 all night and see how it goes. Ever since I turned off my all picture screen saver and started using Internet Explorer 64 I've been crash free at the Desktop, which is good.

I'll post any reports if need be, otherwise I'll just be checking the forums as always

The changing memory just means that the other 2GB is being used elsewhere. Nothing to worry about

If your going to try running P95 all night, try the blend test. In the menu select Torture test>blend (don't worry about the name 'torture')

I did indeed select the 'blend test'. It said it was a bit of everything, but more RAM intensive, correct? I do shudder at the "T-word" however...my rig is too pretty to be tortured! lol

As you can tell, I'm the worrying type, at least when it comes to 3k worth of parts sitting before me...ahh!

It's been going at it for a while now, somewhere around 2 hours or so. At first, my temps were pretty high, averaging in the 60s...at 99.7 load it got to 68 (core 0, 64* for the rest), but that was the hottest it ever got. Now the temps are in the 50s, and 40s...I had thought I cancled it by accident, but it seems like its still going. Does it just run over and over? How can I interpret its results?

Man, intel's burntest says temps average 20+ degrees over p95...that would place them square in the mid to high 80s, even 90s! Should I even run the test? keep in mind I have NOT yet adjusted anything in the BIOS...everything is still set to AUTO. (before I screw with voltages, I want to install XP).

I think I might be heading to bed while letting this thing run. I'm sure that will be ok, yes? I'm 99% positive I have warning beeps for overheating

Best,

Sal

I hear you about being the worrying type. 3k's worth of shiny new toy.... I was like that when I first started Overclocking. I didn't want to break my new toy while I was learning

The blend test is a loop test, you just let it run and stop it when you 'think' it's stable. 8hrs + is the recommendation.

You should have four 'worker processes' (8 with i7 hyper threading?) running through tests of between 8-12 passes per test. As each test passes it will say OK. If it's not stable, then the app will either crash or the PC will BSOD. Not fun to wake up to when you think you've got it stable

Down the track when your more comfortable, you can certainly work on the voltages to lower your temps. It's not so much heat that kills chips, but voltage. Lowering the vcore helps. For example, I originally left the BIOS vcore at auto and it was pumping much more voltage than was needed, even at stock speeds. (still safe, but still too high)

It's good to get everything stable at stock at the lowest settings possible. Then you can start overclocking