Upgraded computer; Now all my games crash!

[antdude]

Rags: I could. Isn't 300 watts enough?

[Rags]

The wattage really means nothing. What matters is how well it performs. Some are better than others, you know. This is why AMD actually went through the trouble of showing approved power supplies. At any rate, I would be interested to know what happens if you swap out your power supply.

Before you do that, have you made sure that your AGP ratio is set properly in the bios of this motherboard? Have you tried puting the SDRAM latencies to 3? Have you tried bumping the voltages up to see if it helps? Does clocking the cpu back still solve your problems? What do your IRQ's look like now? When you installed your new motherboard, did you go through the trouble of booting into safe mode and deleting all devices in device manager?(you really should). Have you made sure that the heatsink is contacting your CPU properly, and the fan is working okay?


Rags

[antdude]

I will try the power supply later on.

AGP ratio. Umm, like what settings should it be at?

SDRAM latency is at 3.

Yes, I removed all devices in Windows in safe boot, but remember that DOS and Linux are puking as well.

Fans seem to be working. P3 CPU is warm.

IRQs:
0 System timer
1 Standard 101/102-Key or Microsoft Natural Keyboard
2 Programmable interrupt controller
3 Communications Port (COM2)
4 Communications Port (COM1)
5 Creative SB Live!
5 IRQ Holder for PCI Steering
5 Intel 82371AB/EB PCI to USB Universal Host Controller
6 Standard Floppy Disk Controller
7 ECP Printer Port (LPT1)
8 System CMOS/real time clock
10 3Com EtherLink 10/100 PCI TX NIC (3C905B-TX)
10 IRQ Holder for PCI Steering
11 Matrox Millennium G400 DualHead - English
11 IRQ Holder for PCI Steering
12 Diamond Multimedia FirePort PCI SCSI Host Adapter
12 IRQ Holder for PCI Steering
13 Numeric data processor
14 Primary IDE controller (dual fifo)
14 Intel 82371AB/EB PCI Bus Master IDE Controller
15 Secondary IDE controller (dual fifo)
15 Intel 82371AB/EB PCI Bus Master IDE Controller
9 (free)
Note: PnP in BIOS is disabled.

I will try the other ideas. I am going to sleep now :I. I will try these later.

[Brian R.]

AGP should be at 2/3 for 100 MHz fsb or 1/1 for 66 MHz fsb.

The power supply seems to be the most likely cause. Perhaps it is on the edge and not giving a clean enough DC. Might even be a periodic problem. Rectified AC can be ugly.

Sorry if this has already been discussed, have you removed all the cards except the G400 and tested the system stability? or do you need some other of the cards to do this?

If possible, try just a PCI video card and no drives except the floppy. Is this possible as a test mode? Adding additional cards and drives incrementally is instructive if you're able to test the system this way.

Another less drastic method would be to remove things (cards and drives) one at a time.

Have you tested the drive cables (replace)?


[This message has been edited by Brian R. (edited 25 September 2000).]

[antdude]

Some things I noticed overnight:

1) UT ran without problems during my downloads overnight during its demo loop. I ran this for about five hours.

2) I played D2 for about 15 minutes in Act 2. No problems. I usually get blue screen of deaths with sound being stuck in Act 2 in Dry Hills area on my old motherboard (with P2 CPU and P3 CPU), but not last night! Good sign I hope!

3) Quake 3 Arena demos (v1.17) seem to work fine. Only played for about five minutes. Is there a way to loop the demos?

4) burnP6/cpuburn -- http://users.ev1.net/~redelm/ -- hard locks up my computer after 20 seconds in Windows 98. Linux too.

5) AleGR memory tester -- http://www.aha.ru/~alegr/download/memtest_en.htm -- found problems and then found an error in a random test for test #2. I only ran this once though for. Only ran for 30 minutes.

I need to do more testings. I find it weird that the test programs barf, but my games and net surfing seem to work? Does that make any sense?

[Brian R.]

If this is different behavior than you were seeing with the old mb, the cause of the problems may be different now, or at least tied to a different pair of components.

Test programs may be more stringent in analyzing a particular mode or function. If the applications work, you may want to ignore problems with test programs.

[antdude]

Brian: I cannot seem to find the AGP ratio to put 2/3.

You can check the online manual here: http://www.asus.com/cgi-bin/dl.asp?f...f/p2bf-103.pdf (2 MB).

I cannot believe this manual is so thick unlike DFI's measely 20 pages.

As for test programs, maybe. It is way too early to say that my system is stable or not. We'll see! :I I was a bit surprised to see UT still running overnight! However, I do like to fix the test programs so in case I have problems in the future, I can run them.

[This message has been edited by antdude (edited 25 September 2000).]

[Brian R.]

Page 15 shows the jumper setting for AGP = 2/3.

[antdude]

Brian: I will check later. I am at work now. I think Sandra will tell me this information. Does it? I will check with that first when I get home.

[antdude]

Oh, I got an email from cpuburn author and he said:

"MS-Win98 gaming is not always very severe. I designed `burnP6` to be more severe, and most users report it hotter."

That is probably why these testers are crashing. I guess games are not really maximizing my CPU. Hmm?

[Brian R.]

Shows you're on the edge. If you don't lock up or get errors occasionally, I would live with it. Testing the ps is worth while IMO.

If it is bothering you, test the components and see if removing a specific card always allows you to run the test programs.

Have you tried AGP 1X for the G400 in this mb? (reg hack)

[antdude]

Brian: No for the registry hack because the testers fails in DOS boot. :I Doesn't help, eh?

ALL:
Here is another reply. This makes sense:

"Have you checked you grease job? Moving processors often ruins it, then you have to clean off the old, and reapply the new. You cannot stick back together because air-bubbles will form."

Has anyone had this problem when moving CPU around?

[Greebe]

I invariably refresh or replace the compound when swapping HS & cpu's. One can never be to safe.

[Brian R.]

I have never seen a problem with disturbing the heat transfer medium while removing or installing a processor (leaving the hs attached), but that doesn't mean it can't happen. I guess some forms of hs attachment may be more prone to this problem than others.

I think Greebe is talking about when changing heatsinks.

[This message has been edited by Brian R. (edited 25 September 2000).]

[antdude]

Brian: AGP bus speed is at 67 Mhz. I assume this is 2/3?

The version is v1.00.
Current Data Transfer Rate: 2x (134 Mhz)
Side Band Enabled: Yes
Fast-Writes Enabled: No
Aperture Size: 128 MB (was 256 MB earlier but didn't see anything differences).

I viewed this information from Sandra.