Hypertransport sync flood error

[Eon the Wolf]

Sorry if this is the wrong section but not sure where it should go...I just got a crash of 'hypertransport sync error' while streaming FF14 (ironically about to get off...for a tiny bit.) I've never had this error before, and not even when I spent an hour in character creation last night (dont judge me)...and this is my first night actually playing FF14.

I don't overclock, and as far as I know all my settings are default...I've read this could be faulty ram, gpu, motherboard, heating...I really don't know....A few months ago I ran tests for gpu memory and furmark, nothing had any issues resulting. I will say I've struggled with this gpu since I got it in march, a full month after I requested it...It's supposed to be a replacement for my broken 5770, but it was a refurbished one that from the get go could not run the same games/settings as my original 5770 when it was new. I had to lower settings. But furmark n other tests dont seem to pull up anything. I ran a windows memory diagnostic a week ago on my RAM, and nothing popped up...

How can I check what is/is not causing the problem....I've read it may require me to manually adjust transport speeds or some such on certain bridges? I'll admit I'm not that knowledgeable/good at bridges so....yeah...what tests/things can I check? Any suggestions? Any other bits of knowledge you want provided I'm more than willing....I will say I should be getting my friends 560 TI within a few weeks hopefully, so if it IS the GPU itself...it wont be an issue much longer....

 

[Steevo]

You may have found the weak spot of your card, the reason that furmark didn't bring it up is furmark can reside entirely in vmem and doesn't stress the PCIe/HT busses.


Try to run F@H on your GPU and see if you get the same error.

 

[Eon the Wolf]

As in Folding@Home....?

Never heard of it, that's only result from google searches....its a bioresearch application, how would that stress test my GPU?

And if I may...if the PCIe/HT busses are the weak point...why has it not been the case that streaming or LoL or anything else in my 450+ video game library has caused it to crash...? Sure, I've had lag n some issues here n there, but I chalked it up to internet and/or just the GPU being of inferior quality...I fail to see how FF14 would make it BSOD/crash but not any of the other half dozen MMOs or intensive recently released games I've played....

If it is Folding@Home you're referring to, I can certainly try testing that....what exactly should I run?

EDIT - As a side note, how does one properly and fully uninstall prior GPU drivers before installing new ones? Cause I forgot how, and I may try that, never hurts...

 

[Eon the Wolf]

Sorry for double post but wanted update in new thing instead of just an edit; I downloaded n ran Folding@Home for about an hour....no crashes, no BSOD, no warnings or errors, nothing. Ran at Full. It said CPU was 1.89% in progress and GPU was 0.00. I was on project....8579 or 8759 or some such for cpu and 8900 for gpu. How long am I supposed to leave this running before I can say it has/has not found the issue....? i dont really wanna leave my PC on and running this for 3-7+ days non stop like some people say is needed to stress test stuff :/

 

[Steevo]

No actual work being done on the gpu?

 

[Eon the Wolf]

It listed both of them as green and "RUNNING"....I've never used F@H before, so I mostly just installed n ran it....

EDIT: Opened back....

Folding slot cpu:3 has status of RUNNING with progress of 1.4% with 0.00 secs ETA, --PPD, and 8579 (1, 2, 218) PRCG.

Folding slot gpu:0:Juniper [Radeon HD 5700 Series] has status of RUNNING with progress of 0.00% with ETA of 0.00 secs, PPD of --, PRCG of 8900 (645, 0, 267)

Running at Full with 'I support any disease resarch'. Says im contributing to both of them.

Ran it for an hour, nothing happened....no other processes or applications were open/running beyond bare minimum and my AV....

EDIT - And could I get an easy answer to my earlier question? Been meaning to do it for awhile now...how does one properly clean out all prior GPUs to do a fresh GPU driver install? Is there an option or a program or...?

EDIT - Screen/monitor just went completely black and said it had no connection. Mouse and keyboard still had their lights/power on, at least mouse did...though the xbox controller i just plugged in did not. No error message or anything. Happened right as I tried to launch dirt 3. PC is currently off. Xbox controller never received power, and it seemed to have happened just before I plugged in controller as dirt 3 launched or maybe as I plugged in xbox controller....any cahnce this could mean fauty mobo?
Friend is bringing me an 8800 gtx tomorrow hopefully to at least replace this 5770 with, if it is gpu hopefully that fixes it and lets me at least play until I get the 560 ti from another friend :X So done with ATI after this, giving me a faulty refurbished gpu to replace a broken new one within warranty period :/

EDIT - To add further, when that issue happened n everything went black, my friend told me I went offline on steam/skype, so....
I'm now struggling a bit to play videos in HD on youtube when ones of similarly prior content played just fine a day or two ago...

this one in particular, i keep selecting 720 and itll start lagging/skipping every once in awhile...It stops if I pause to let it buffer for a minute or 2, but I never used to need to do that, I have 35/5 MB down/up, a quad core 2.8 ghz cpu and a 5770 which im beginning to suspect more and more is just flat out dying...

 

[Eon the Wolf]

Last time to double post in this, but any additional advice/suggestions/tests to run would be appreciated. Kind of...a bit tight on time in coming weeks for things, and being unsure of whether or not this gpu is decidedly breaking/broken, or if its a fixable issue, would be really nice...because my other option is an 8800 gtx (or gts? i forget atm) that is obviously far weaker than a 5770, and I could be stuck with it for a few weeks to a month or 2 before I get the 560 ti from my other friend, which would put a screw in my plans :X

 

[Steevo]

download ghostbuster, and run it after uninstalling all the drivers, then go into the control panel and uninstall the card there too and check the box to delete driver software, then run ghostbuster to delete any additional copies of all unused devices. Make sure to not let windows reinstall drivers, so either download the latest drivers and then pull the network cable during the install.

 

[Eon the Wolf]

I take it that's for fully clearing out drivers before reinstalling them?

Is there a chance my current issues are driver conflictions/problems...?

EDIT - Or are these 2 crashes likely unrelated to any potential driver conflictions?
Doesn't ghost buster delete drivers for more than just GPU? Couldn't I accidentally delete drivers for USB hubs or something else needed? I've never heard of or used it before...can you give advice on how to double check/ensure that I only remove GPU related stuff?

 

[Eon the Wolf]

Could I please get some more answers than cryptic suggestions without explanation, please?
I don't know of ghostbuster other than you mentioning it now...do you mean ghostbuster.codeplex.com?

And after uninstalling the current driver n presumably safely clearing out remains of older drivers (Which I have no idea as I've never done it) and reinstalling the newest one(s)....how am I supposed to test the GPU/installation to see if that fixes my issue(s) or not? I'd rather do stress tests if it can be helped, instead of just having to play stuff n see if my computer breaks/crashes....

EDIT - Apologies if it sounds rude, perhaps cryptic was poorly stated...I'm just not 100% sure of how to use most of these, or how to avoid screwing something up when I do use them...so any additional info/advice/instruction would be helpful....and not having to wait a day between every reply would help :X

 

[Steevo]

Yes the duplicate device remover. Essentially uninstall all driver packages shown in add/remove programs, download the newest driver. Unplug from the network.


Go into device panel, find the video card, if it is still listed as anything other than standard VGA device, choose to uninstall it, when it asks if you want to delete device driver software check the box. Restart. Run Ghostbuster as a admin, then right click the classes and add the class for items listed as ghosts. Let it clean. Restart. Install the new drivers.

 

[Eon the Wolf]

Okay, and after I've done all that...how will I know if it has fixed my issue? I would prefer to find out by way of stress testing or doing something that will give me proof rather than playing a game n hoping my pc doesnt decide to start crashing :/

 

[itsakjt]

Stress CPU cores with Prime 95 with the Large FFT mode(max heat). Stress memory controller and RAM with Intel Burn Test and Memtest86. And if you can, clean your computer and reseat the RAMs and graphics card. Clean the gold plated contacts.

 

[Eon the Wolf]

Have an update. Ran system mechanic on advice of a friend and cleared 926 of 948 registry problems, cleared 240 something MB of IE temp files n 7.19 gb of windows temp, fixed 4 core windows communication instrastructure issues. It also told me, despite very little running, i only had 30% available memory (Eventually showed as 24 or 18% after a few programs n such were opened for a bit before rescanning)

Also ran memtest today. My standard 2 sticks of gb memory shut down (without me seeing an error code as I wasnt looking/in room when it happened nor with a beep). My friend helped me swap in a spare pair of 2 gb sticks that were slightly slower in speed. The first time it started, the %s, timer, and all else seemed frozen at 28%/4:26 while the white | where info/typing happens was blinking, though keyboard was unresponsive. The second time, it got at least as far as 58 or 60% and 12 minutes before the next time I looked it was shut down and black.

Does this mean mobo is at fault? Cause thats 2 separate pairs of 2gb ram that have crashed/shut down in the middle of mem test not to mention GPU seemingly having issue....or is it PSU failing to do power or is cpu somehow at fault or...?

EDIT - Currently responding from mothers laptop btw, pc is staying off until you can advise me further tests/things to change or until issue is found/fixed (or w/e have you)

 

[itsakjt]

Check the temperatures of the CPU and motherboard. And clean the system.

 

[Eon the Wolf]

The interior has been air canned/cleaned many times. CPU hovers between I wanna say....24 and 31 or 32 C is generally the range...I dont know if I can measure mobo's temp :X I know the voltages/temps are shown at start up, but i cant recall if mobo is ever listed...

EDIT - It'd be a rather large pain to try n remove/individually clean every part in my case and get it back in without issue, but I can definitely say safely that theres no obvious dust build up on parts or fans. Me n my friend both aired it all out a bit a few days between....he also checked to make sure everything was properly plugged in and tightly secured to mobo n power cords n such.