0

Hi, I was wondering if you could help me out with this issue. I have tried almost everything, even read similar posts but nothing did help at all.
Usually when I'm playing games like Battle For Middle Earth or Farcry, computer completely freezes up after some minutes of playing. And my computer should be able to handle these easily :sad:
I thought I was having temp issues, but borrowed a Heatsink for cpu and bought a new cooler, placed thermal compound on cpu, even replaced the case, and now temp has considerably slowed down, but I still have this annoying freezing up on games.
Yesterday I downloaded latest n-Vidia drivers for the video card, which is an 6600GT, and also tried installing Rivatuner, disabled AGP Fast Write in bios and Write Combining, as I read somewhere. But that didn´t help as well.
So plz help!!! Any ideas??

Computer specs:
AMD Barton 2500
Asus A7N8X-E Deluxe
1 gb DDR 400 mhz
Western Digital SATA 80 gb
Maxtor SATA 160 gb
Chaintech GeForce 6600GT 128 mb
PSU: Antec 550
Windows XP SP2

Thanx in advance!!

5
Contributors
23
Replies
24
Views
11 Years
Discussion Span
Last Post by Carebare47
0

I dont know about "Battle For Middle Earth", but I know Farcry.
I have A8N Sli/Athlon64 3700+/Gainward GeForce 7800 GT 512 MB/ 1 G DDR400 and I have (sometimes) poor performance with Farcry. Wooded areas and max. details are too demanding.

If it is not overheating issue, then I think it is memory related. Either voltage is not high enough, or the speed/timings are wrong.
Also, OCing your CPU and/or VGA would result in such behaviour.
I suggest that you set your BIOS settings to defaults (regarding CPU/AGP/Memory speed/timings). If those settings do not solve your problems, try manual increase of memory voltage. Some memory modules (like Gail) demand more juice than defaults.

Also, there are lots of bugs and patches for Farcry. Keep that in mind.

0

What Chaky ;) said plus:

Check the GPU temperature. If you use the new nVidia Control Panel, you can leave the temp. monitoring page open to see the graph, then start gaming but try to Alt-Tab to the monitoring page before the computer freezes. Back story: These GPUs change their temperature pretty fast, when you switch back to the monitoring page it has already lost 2°C. If you end the game normally, it's almost back to idle temp. (Ok, that was slightly exaggerated)
The net is full of reports of freezing 6600GT AGP but the nForce2 chipset has much less problems, so I suspect the card itself and some users report overheating probs with that card type.

@Chaky:
your PC/GPU have much more power than mine... it really shouldn't suffer performance in FC at any place since I even tweaked the draw distance and didn't have any slowdowns.

0

One more thing. Farcry settings:
1600 x 1200
max detail except AA is turned off. Seems that my system doesn't handle AA too lightly.

0

Ok...I ran FC in 1024x768, 4xAA, 16xAF, TAA on, all optimizations off, on a Barton 2800+ (2083MHz) and a 7600GS (430/900). Framerate never dropped below 25, even not in the heavy load levels at the end of the game (i.e. "Vulcan").
I ran highest possible image quality/low res and your rig had to render much more pixels, and this will have some impact, especially when using AA+AF, even on a 7800GT. I guess this can eat up more than the advantage of the 7800GT over the 7600GS. Your +200Mhz won't help that much since FC is very GPU dependent, so I guess everything's fine with your rig's performance...:cheesy:

0

You say your GPU is 430/900. Mine is 400/1000. 400 being core frequency.

"Aluminium Association" drops fps drasticaly on my rig. And not just FC. Doom 3, Quake 4 (although, I lack the system memory to run Q4 without extensive paging, Activision said "You have only 1 gig of memory"). FSAA doesn't do much damage, but I tend to turn that off too. Makes screen blurry.

0

Hmm... drastically is true, FPS can drop by 50%, but as long I get 25+ frames it won't harm. I played FC to check stability + OC capabilities of my new card, hence the 430/900 (stock is 400/800 on that card). But AF shouldn't make the image blurry? Its purpose is to rather sharpen (the mipmapped stuff)... Maybe you have the "optimizations" still turned on? They deteriorate IQ pretty much to the benefit of minimal FPS improvements.
But maybe we differ in preferences, too - I know that from my fellow flight sim friends, some of them hate the FPS impact and like the (subjective) perception of sharpness without AA+AF and don't care much about the resulting anthill flimmer on the ground. (I get headache from that :))

0

FSAA as Full Screen Antialiasing, not Anisotropic filtering.

Haven't run FC in a long time, but I think that there is no FSAA in FC. There is one in N4S most wanted. Made my screen blurry and my eyes hurt.

0

FSAA as Full Screen Antialiasing, not Anisotropic filtering.

I thought you made a typo because you mentioned AA and FSAA in that context.
FSAA is no function of the game engine - the ingame setting just tells the driver it's own idea of which type of AA is applied (and FC uses AA, too of course). If you turn AA off in the game, it takes the setting set in the driver control (which is recommended). But anyway - I can imagine now what happened:
I forgot that there are some AA types on nVidia cards that are blurry as hell indeed, namely the 2xQ and 8xS settings. Even letters look like "1 gal whiskey" then. I guess setting AA in NFSMW turned one of these on. That's why it's better to control that with the driver, not with the ingame menu - you never know what kind of AA it turns on.

0

Yup, (AA settings in control panel = 0ff - 2x - 2xQ - 4x - 8xS)

It seems to me that AA is not hardware-driven function, or my video drivers (latest) have HUGE bug.


Anyway, I think that we are miles off the topic here.

0

It seems to me that AA is not hardware-driven function, or my video drivers (latest) have HUGE bug.

http://en.wikipedia.org/wiki/FSAA (Bottom of the page) But I guess it's not all hardcoded and the driver contains the algorhythm for that part of hardware to be computed..(?)

Seems that mileage may vary a lot with nVidia's drivers if you have bad luck and IIRC you had your own special experience with their support already - me too. :cheesy:

Anyway, I think that we are miles off the topic here.

Sure, but we're just passing time until the threadstarter returns.

0

Ahh.... NVIDIA customer (so-called) support....

I'm still waiting for them to reply me on f6 drivers. $%#@*
2 weeks and counting.

I wonder if they would reply me on this "AA=major slowdown" thing.
Probably would instruct me to install the drivers I already have installed. Or say something like "we don't build graphic cards".
Right, they just build GPUs and make drivers for them. Nothing to do with video cards at all.

0

Probably would instruct me to install the drivers I already have installed.

That will most likely be happening. Sorry, can't resist to c+p that story here:

The problem I describe below appears on NV-based cards of numerous manufacturers and was already adressed by nVidia (new BIOS) and one manufacturer (Club3D) already offered 3 different ways to solve it at the time I contacted "Customer Care". It was never related to the driver and many GF7 (at least, prob. exists on GF6, too) users with sufficient PSUs bought new ones for no reason due to the false advice they found in forums and the "no-advice" by nVidia. Only a minority really had weak PSUs. The nV "Customer Care" was more clueless than I could stand. And it seems they never read mails:

I:
I have a problem with the 'nVidia Sentinel' reporting insufficient power after first (cold) boot on my new 7600GS (AGP). Like most of the other people with a 'sentinel' problem, I'm sure to use a quality PSU (Coba) with sufficient amperage (25A) on the +12V rails and have of course a power connector attached to the card. The sentinel pops up after a cold boot only, after a reboot everything is fine. I updated to the latest beta drivers (92.91) to see if the problem has been addressed, but no success. Looks like the card/driver checks a parameter too early during startup, not leaving enough time for the PSU to power up or something.
[...]

Kind regards,


NV: Please plug a power connector into your card

I: I did, as mentioned in the mail

NV: Please install the 91.47 driver

I: 91.47 came with the card, have tried NGO 91.47 and now use 92.91 beta. Haven't tried a Forceware 8 driver, since the 7600GS AGP isn't supported by them AFAIK.

NV: Please install 84.21

I: Driver didn't find nVidia hardware, since it can't know that card yet (Rem.: At least I knew that 91.47 was the first driver to support 7600GS AGP)

NV: Please install 91.33 :rolleyes:

I did. At this time I stumbled over the real cause for this problem and asked them to close the support ticket. I was sick of installing drivers. Don't expect support from nVidia. To be honest, I believe "Customer Care" is a bot, parsing mail titles for key words and replying accordingly. Worst artificial non-intelligence ever. :mrgreen:

0

Once, I got response: "please install our lates driver 91.47" and they gave me link to 91.33 beta.

0

Thnx for the help guys!
I was checking... Doom 3 appears to work just fine... no freezes at all for about 1 hour of playng. I also seemed to solved the issues with Battle For Middle Earth, I took the counsel of downloading all the patches, and now, though I have some slowdowns, the game never did crash, not even once.
But i'm still having some issues with FC, i have just check the GPU temp, was about 71º, whereas in idle stays at 56º.
I'm just believing right now it was a patching issue, but as i've been having some trouble with this pc i just had no clue on where to point at.
For example, right after having these games functioning properly, whenever now i turn the pc on, it stays on the motherboard screen, no error, no nothing, i have to reset it about 15 times or so to have it load Windows normally. Anyway, regarding that i think there may be a loose cable or a bad connection with one of them. Just gonna open the case and check if everything's fine.

I really appreciate your counsel guys! Thank u very much! You plainly rule.

0

71 °C would cause you slowdowns. Chips work slower when the temp goes up.

My VGA is set @ 115 °C to do an "emergency" slowdown, but it never gets beyond 50°C (GPU Fan = NV silencer ~$20 - $30).

And regarding that POST stall, good thing would be to disable full screen logo and quick POST in your BIOS, just to see on what stage this happens. My guess is that one of the IDE/SATA cables are making a bad contact, or one of the drives are starting to malfunction. But, it's just a guess. Could be anything.

0

Thnx a lot for the advice Chaky!
Regarding the Gpu temp I guess I'm checking now video cards coolers (the one you have is not expensive at all and appears to keep GPU rather cool). I guess 71º is too much indeed.

And also, perhaps you're more than probably right regarding the bad contact with the SATA connectors, because my case is a Thermaltake Armor Jr., and as the HDD's connectors are installed horizontally, the SATA cables do touch the case and stay somehow sticked to the case. Gonna see what can be done about that.

And thank you so much for the fine troubleshooting once more!!

0

I had the same problem.

I think it is something to do with the fact that your computer uses some HD space for extra memory if it is needed, try getting a bigfer hard drive, or playing with the settings on control panel or somewhere that determin how much HD space can be used as memory.

Good luck, let me know if this fixes the problem.

This topic has been dead for over six months. Start a new discussion instead.
Have something to contribute to this discussion? Please be thoughtful, detailed and courteous, and be sure to adhere to our posting rules.