How to get Windows to recognize Dual-core

KimB

Legend
Okay, I just installed an Athlon 64 X2 3800+ (2GHz, dual-core), and Windows doesn't recognize it right away. I'd really like to get this working without reinstalling Windows, so does anybody know of some quick way to get it working?

In the meantime, I'm compiling a new kernel for Fedora, hopefully it works this time (for some reason, I appear to have had some configuration issues when compiling an SMP kernel earlier....wonder if there's options other than just enabling SMP and NUMA that Fedora likes? Well, no matter, I'll figure it out....worst that'll happen is that I rip the config file from one of Fedora's own SMP kernels and modify it to my tastes).
 
Last edited by a moderator:
John Reynolds said:
Probably need to update the BIOS on your motherboard.
Nope, got SMP working just fine in Linux (had updated the BIOS previously anyway for the purpose of overclocking).
 
Argh, now this is f'ed up.

SMP seems to be working great, but now my audio is just gone. I mean, I've uninstalled and reinstalled nVidia's motherboard drivers, but there's just no sound device to be found in the device manager. I may have to reinstall Windows after all :(

Oh, well, back to Linux for the time being.
 
Pete said:
Chal, I wonder if this will help you avoid a reinstall.
Actually, turns out I just needed to do a BIOS update after all. My old BIOS supported dual-core CPU's, but not the latest one that I got (the 3800+ that was just released).
 
Chalnoth said:
SMP seems to be working great, but now my audio is just gone.
It may be worth noting that for like the longest time ever, creative's soundblaster live/audigy drivers didn't support multiprocessor systems. Not that they didn't take advantage of multi-processors like any other program not specifically coded for it, they plain didn't WORK.

Actually, this might still be the case.
 
Chalnoth said:
Actually, turns out I just needed to do a BIOS update after all. My old BIOS supported dual-core CPU's, but not the latest one that I got (the 3800+ that was just released).
So I guess you got a E6 stepping then (afaik the X2 3800+ exists with both E4 and E6 stepping)?
 
Chalnoth said:
Actually, turns out I just needed to do a BIOS update after all. My old BIOS supported dual-core CPU's, but not the latest one that I got (the 3800+ that was just released).

Heh, when I reviewed the X2 3800+ I dropped the A8N's 1011 BIOS, which properly ID'd the new CPU. I then told AMD about it since their comments to reviewer's contained a plethora of information about ensuring that the CPU was running correctly since they had no idea what the various mobos being used would set it at.

If you fire up EverQuest 2 and notice it starts crashing a lot (happened with my X2 4800+), set the affinity to just one processor and it'll then run fine.
 
John Reynolds said:
Heh, when I reviewed the X2 3800+ I dropped the A8N's 1011 BIOS, which properly ID'd the new CPU. I then told AMD about it since their comments to reviewer's contained a plethora of information about ensuring that the CPU was running correctly since they had no idea what the various mobos being used would set it at.
Well, the 1008 BIOS ID'd the CPU just fine, which is why I didn't look at the BIOS update first. But when both Windows and Linux no longer saw the sound, I felt something had to be up. A check at the boot menu (just found out that pressing "pause" lets you actually look at that initial status screen), there was no sound device there, either. So, BIOS update it was.

If you fire up EverQuest 2 and notice it starts crashing a lot (happened with my X2 4800+), set the affinity to just one processor and it'll then run fine.
Hrm, I'll have to try EQ2, I suppose. I've basically stopped playing it, but I still have a while left on my account.

Edit:
Wow, you weren't kidding. I recently reformatted, so had to reinstall. The game didn't even wait until I got into it to start crashing, though: the LaunchPad.exe program kept crashing! I had to execute it like 5 times before it finally got to the point where it was able to load up the game update window.
 
Last edited by a moderator:
Chalnoth said:
Wow, you weren't kidding. I recently reformatted, so had to reinstall. The game didn't even wait until I got into it to start crashing, though: the LaunchPad.exe program kept crashing! I had to execute it like 5 times before it finally got to the point where it was able to load up the game update window.

My crashing didn't occur until the game was fully loaded, and varied from immediate to 10-15 minutes. SOE tech support says a forthcoming patch will address this, but they just recently got dual core systems into their test labs so who knows how long it'll take for them to incorporate this update.
 
John Reynolds said:
Heh, when I reviewed the X2 3800+ I dropped the A8N's 1011 BIOS, which properly ID'd the new CPU. I then told AMD about it since their comments to reviewer's contained a plethora of information about ensuring that the CPU was running correctly since they had no idea what the various mobos being used would set it at.

If you fire up EverQuest 2 and notice it starts crashing a lot (happened with my X2 4800+), set the affinity to just one processor and it'll then run fine.

Do you have any idea why so many games are having problems with dual core systems? World of warcraft had major problems, massive frame rate loss and errors, they fixed it in the 1.6 patch though. But I've read about quite a few other games playing like they are way too fast.

Do these games have problems with multi processor systems, or does dual core add something that needs to be handled differently in the programming?

Counter strike souce also has problems with obtaining clock speed correctly on my dual core also.

Examples:

http://homepage.ntlworld.com/brokenhope/Misc/lol.png
http://homepage.ntlworld.com/brokenhope/Misc/lol1.png
 
Broken Hope said:
Do these games have problems with multi processor systems, or does dual core add something that needs to be handled differently in the programming?
Probably race conditions between two threads that didn't begin failing until there were two processors executing the separate threads.
 
Back
Top