Guild Wars 2

I see in this forum and in any other gaming forum and technical video card forums that there is a spread issue with ATI 7xxx cards crashing in games.

Has anybody solved crashing ATI HD7850 during the game?
I have crashes all the time.

what system I have:
Processor: Intel Core i5-3470 CPU @ 3.20GHz
Gigabyte ATI radeon 7850 GV-R785OC-2GD
Motherboard- Gigabyte GA-H77-DS3H
16 GB corsair value RAM
SSD drive 128G corsair
seasonic bronze 650W powersource
windows7 64 bit SP1

The symptoms of how it is crashing for me- randomly inside the game:
a) I get black screen with the latest sound bit looping
or
b)orange screen with vertical lines with buzz sound.

when the crash occurs I cannot alt-tab or ctrl-alt-del or alt-F4 out of it, I have to shut down the PC with on/off button.

I don’t have any crashes in the benchmarks: Heaven DX11 Benchmark 3.0 and Future mark 3DMark 11 on max settings

I don’t play other games, so cant really test.
I don’t overclock the card!

there is no relevant error (or any except the force boot) in the windows event log
there is no guildwars2 dump file.
if I run the gw2.exe with the log parameter I get the usual missing skins and objects. nothing that can indicate a serious error in the log.

this is not overheating issue as the game crash at GPU temperature of 70c and 60c and even on 50c .
70c still considered normal gaming temperature of the GPU.

but it is related, because if I artificially keep high temperature (with no air circulation and disconnecting case fans) I crash more often than when I keep the GPU on low temp of 46c-50c.
When on high temperature the game crashed every 5 and 15 minutes.
On low temperature I could crash after 30 min to 1 hour
(degrees in Celsius)

what I have done:
1) I contacted Gigabyte support

they gave me the latest BIOS for the card: F5_A which increased the voltage for the memory.
they asked me to try the second (slower) PCI-E in the mainboard
both of their solutions didn’t help, they suggested to return the card to the place I bought it.

2) contacted ATI (AMD) support, I got no response.

3) searched the web and tried the following solutions:
a) disabled anti aliasing in GW2
b) disabled shaders in GW2
c) disabled HD sound in OS.
d) disabled completely the sound in BIOS
e) tried ATI drivers 12.8 -12.9 -12.10 and now I use the beta 12.11 from last week.
f) in BIOS changed the PCI-E (PEG) generation to 1
g) under clocked the GPU to up to 850mhz (default = 975)
h) disabled desktop background and screensaver.
nothing helped, nothing at all…

4) I read this in one of the related posts in guru3D forums and it partially helped.
the post said that there is a problem with the power saving mechanism of the cards that decide how fast the GPU should work and how much power it should be gaining, so sometimes when the load is high, the system thinks the card is idle and reduce the power gain so the card crash.
this mechanism should be disabled.

I installed third party overclock tool (from HIS)
and in this tool I disabled the 2D clocks and ULPS- as it was said in the post
from here:
http://forums.guru3d.com/showpost.php?p=4426512&postcount=323

when i do this, the crashes are much less frequent, I can play up to 2-3 hours with no crash, ( I can almost complete two sets of fractals! I start level 1 and can sometimes reach the second or third fractal of level 2, then I crash… )

but this solution doesn’t solve the problem for me completely
I couldn’t find the 12.6 driver for HD7xxx cards in AMD website, so i havent done this part.

So, my question remains, does anybody here had this problem and managed to solve it?

thanx

attached Guild Wars 2 Test for the support team.

Hmm, ok. First, have you also installed the latest CAP drivers? I would also recommend using something like Drivesweeper, do a clean uninstall of your drivers, then do a fresh install of the latest (12.11 and CAP 12.10). Another thing that comes to mind is to check you processor during playing and make sure it isn’t running too hot as well. I would also make sure ALL relevant hardware drivers are up to date to include your motherboard BIOS (it fixed my crashing issue).

Hi Lalangamena.

A further thing to check, please make certain your ram is running at the speed it should be. I had countless problems with ram running at 1333mhz instead of 1600mhz.

Also, make sure you have the latest firmware for your SSD. My Crucial M4 256gb was also giving me tons of issues until I updated the firmware.

Another thing to try is to change your sound setting in GW2 to lower settings.

Hopefully one of these things will help.

ty for replies

I indeed have the latest CAP 12.10 and the latest mainboard BIOS.
when i re-install driver I use “driver fusion” to clean all the files and registry entries.

to Myth:
my RAM indeed runs for 1333 instead of 1600, but it is because I use all the slots. (4×4)
when you use all the RAM slots, your system works on lower RAM speed.
I will remove two slots and try to play with 8 giga ram to test.

SSD is a “black box” for me, I looked for any driver or firmware update on corsair webpage but didn’t find anything, and Intel software don’t recognize my SSD, so I am not sure how to update the firmware…
I use corsair force series 3 SSD 120MB.
http://www.corsair.com/en/ssd/force-series-3-ssd/force-series-3-120gb-a-sata-3-6gbps-solid-state-hard-drive.html

GW2 crash even when I disable all the on board sound from BIOS (yes I played gw2 with no sound at all…), so i doubt its the sound settings but i will give it another try.

I have the exact same Gigabyte 7850 card but am not experiencing any crashes. I run the latest 12.11 beta 8 with the 12.11 CAP. My card is even overclocked to 1200mhz and is running at highest 56C when under heavy load. I know there was a few of the cards that problems but there is a bios update for those here: http://www.gigabyte.com/products/product-page.aspx?pid=4155#bios

update:
installed the latest firmware for the SSD drive and configured BIOS accordingly.
from here: http://forum.corsair.com/forums/showthread.php?t=110257

also removed two RAM stick and tried to go with 8GB but on full 1600mhz.

still crashed. doesn’t help.

:(

Not what you will want to hear but I have the same issue and people are suggesting to lower the core clock by 50-80 MHz. Currently trying it at 925MHz instead of 975 and so far it seems ok…. Need more time to tell.

Update: Tried 895MHz and 860MHz still having the issue after a few hours of playing. I uninstalled the drivers and reinstalled and now my computer continually crashes over and over if I leave the card in. Buying a new card monday and it won’t be ATI.

Hey Lalangamena,

First of all, I think you should have a look at this: http://www.tomshardware.com/forum/371181-33-crashing-temperature

Do not really have a solution but wanted to share a couple of things that might help. I have the exact same symptoms/crashes. I.e:

“The symptoms of how it is crashing for me- randomly inside the game:
a) I get black screen with the latest sound bit looping
or
b)orange screen with vertical lines with buzz sound.
when the crash occurs I cannot alt-tab or ctrl-alt-del or alt-F4 out of it, I have to shut down the PC with on/off button.”

I used to have this sort of crash randomly after playing most games for 1-2 hours. But I get it way more often with Guild Wars 2. So I started investigating. Turns out it is heat-related. After underclocking my gtx 550ti back to Nvidia’s factory specification and limiting frames to 30 FPS, the temps got lower and I rarely have this crash now. But I still get it occasionally especially in GW2 despite low/normal temps (50-64C during gameplay).

It never really made sense as to how this was possible. That is until I stumbled upon that Tomshardware’s forum post. It seems that GPU’s RAM is the one causing such crashes because it is the one overheating.

“It is almost certainly the GDDR that is overheating. Most cards do not have a monitor for the RAM temp on card.

The fan will increase the heat taken away from the RAM as well as the GPU.

Also I hav efound that if you are playing a modern game that is streaming textures from an SSD (if the game is installed on an SSD) it puts more heat into the GDDR.

My card has a MEMIO heat sensor and I can watch this happen in real time."

Seems like getting an aftermarket GPU cooler and getting a better case/better air flow is the only solution. But then, I have to try it to know for sure.

Here is my spec in case you need it:
Intel Q9650 (@stock)
Gigabyte GA-P45-UD3LR
Gigabyte GTX 550 Ti GV-N550OC-1GI
6 GB Kingston DDR2 667MHz (2GBx2, 1GBx2)
Hitachi Deskstar 1TB 7200 Rpm
Zalman 750 PSU
Windows 8 Pro 64 Bit

Needless to say, all drivers, etc, are up to date.

Good luck,
DS

I’m using a similar setup – while the imbalanced cards can sometimes be the answer, I have issues with Catalyst deactivating my CPU cores down to one, which kills performance (Obviously.)

Try seeing if some of your cores are being turned off for whatever reason.

DragonSoul.3208:

Hey Lalangamena,

First of all, I think you should have a look at this: http://www.tomshardware.com/forum/371181-33-crashing-temperature

Do not really have a solution but wanted to share a couple of things that might help. I have the exact same symptoms/crashes. I.e:

Here is my spec in case you need it:
Intel Q9650 (@stock)
Gigabyte GA-P45-UD3LR
Gigabyte GTX 550 Ti GV-N550OC-1GI
6 GB Kingston DDR2 667MHz (2GBx2, 1GBx2)
Hitachi Deskstar 1TB 7200 Rpm
Zalman 750 PSU
Windows 8 Pro 64 Bit

Needless to say, all drivers, etc, are up to date.

Good luck,
DS

DragonSoul hi,

ty for the reply.

It kinda make sense to me, because if I downclock the memory speed rather than GPU speed , I can play several hours without crash.
I downclock the memory speed from 1200 to 1000 (or how the HIS tool shows from 4800 to 4000)
but still it doesn’t solve the issue completely… maybe I should try to downclock it even lower.

my PC case is open and I have put two extra fans just below the graphic card, to keep it cooled constantly.. the idle GPU temperature is 38 degrees, and gaming GPU tempertaure dont pass even 65 degrees.

from my experience capping the FPS only increase the crashes…
I will try to look for an external thermal pad for the memory chips if i can find them…