There goes that experiment. At work we have a dedicated machine that does nothing but output tens of tousands of PDF's based on foxpro data for our internal web based application. This process takes longer and longer as our database grows - up to 7 hours on a PIII 866. Had to do something, as the process is sneaking up to backup tape time on the server. Well, being the adventure some type, I decided to go ahead and give an Athlon XP/KT266A platform a shot. Well it's not really a test for the Athlon XP, but for VIA with the new KT266A chipset. And what better board to use then the perceived cat's ass, the Asus A7V266-E.
3 Different boards (all rev 1.07, w/ latest BIOS 1.05e from asuscom.de), all exhibiting the attributes I have grown to love from VIA. Random, HARD lock ups. No BSOD's, just a pure freeze that locks Win2K like I've never seen before. Sure, it would install fine and run for hours; procedure was install Win2k, apply SP2, apply latest 4in1, then the latest BM and AGP patches released 2 weeks ago. Then Windowsupdate, bigfix, blah blah every god damn patch and latest driver. Every so often, no matter what I'm doing, the thing will lock up HARD, at random. Disabled all extras, the RAID, sound controllers. Never oc'ed, and used the slowest, most conservative chipset/memory settings. DDR RAM (512MB) was Crucial CLS 2.5 - For the task we need this system to do, there is no room for failure or lock ups. Not one.
A7V266-E, 512 MB DDR, XP 1900+, Intel NIC, IBM 60GXP, TNT1 16MB.
The PDF output job was almost 90% done, in 2 hours 15 minutes(!!!), before it locked up. Too bad, would have been a record. Other attempts would lock up in less time.
Going back to the local vendor and getting a an AMD760 based board. I've wasted enough of my time trying to get something functional that should work out of the box. Our Foxpro guy of course thinks that we should just stick with an Intel platform, and I can't say I blame him, as our PIII never EVER EVER CRASHED, not once. I'm smart enough to know that it's not AMD but it's supporting cast at fault, so hopefully the AMD760 will calm nerves.
Thank you VIA for the memories, you've just reinstated my faith that you are indeed still the goatse.cx of chipsets.
I'm beggining to now understand what most of these asstards who write website reviews regard as "stable". As in, they really don't test real world office mission-critical stability. Running Q3 Demo1 in a loop for a day seems to be enough for them.
Notes: Yes, I'm aware that the latest, latest AGP driver may have issues due to it's register remapping. The hard locks occured with both the 4in1 AGP, and the newer AGP. Same goes for the BM driver. Yes, I'm aware the latest NVIDIA drivers are causing a stink with XP and possible 2k based via jobs. Only using the last WHQL released from them which is v21.83 not v23.11
3 Different boards (all rev 1.07, w/ latest BIOS 1.05e from asuscom.de), all exhibiting the attributes I have grown to love from VIA. Random, HARD lock ups. No BSOD's, just a pure freeze that locks Win2K like I've never seen before. Sure, it would install fine and run for hours; procedure was install Win2k, apply SP2, apply latest 4in1, then the latest BM and AGP patches released 2 weeks ago. Then Windowsupdate, bigfix, blah blah every god damn patch and latest driver. Every so often, no matter what I'm doing, the thing will lock up HARD, at random. Disabled all extras, the RAID, sound controllers. Never oc'ed, and used the slowest, most conservative chipset/memory settings. DDR RAM (512MB) was Crucial CLS 2.5 - For the task we need this system to do, there is no room for failure or lock ups. Not one.
A7V266-E, 512 MB DDR, XP 1900+, Intel NIC, IBM 60GXP, TNT1 16MB.
The PDF output job was almost 90% done, in 2 hours 15 minutes(!!!), before it locked up. Too bad, would have been a record. Other attempts would lock up in less time.
Going back to the local vendor and getting a an AMD760 based board. I've wasted enough of my time trying to get something functional that should work out of the box. Our Foxpro guy of course thinks that we should just stick with an Intel platform, and I can't say I blame him, as our PIII never EVER EVER CRASHED, not once. I'm smart enough to know that it's not AMD but it's supporting cast at fault, so hopefully the AMD760 will calm nerves.
Thank you VIA for the memories, you've just reinstated my faith that you are indeed still the goatse.cx of chipsets.
I'm beggining to now understand what most of these asstards who write website reviews regard as "stable". As in, they really don't test real world office mission-critical stability. Running Q3 Demo1 in a loop for a day seems to be enough for them.
Notes: Yes, I'm aware that the latest, latest AGP driver may have issues due to it's register remapping. The hard locks occured with both the 4in1 AGP, and the newer AGP. Same goes for the BM driver. Yes, I'm aware the latest NVIDIA drivers are causing a stink with XP and possible 2k based via jobs. Only using the last WHQL released from them which is v21.83 not v23.11