Slashdot Mirror
Mass Motherboard Review
Niscenus writes "The folks at AnandTech are buzzing about this article originally posted here at CS. It's an extensive comparison of motherboard manufacturers, which is great for people who want to build their own system."
All the biggies look to be here. Good timing for me since I'm planning on
building a system in another month or so.
This is really a key issue that people miss most of the time. My Dell system is now 2 years old and I want to replace it. I've been happy with the system and all I really need is a new MB, CPU and RAM so I was just going to put it all in the Dell case and call it a day. I've walked out of 2 retail stores in the past couple days because thier RMA policies pissed me off. If I have a MB fail I need it fixed asap not replaced as soon as it can be delivered from Tiwan via sailboat and bike courier. All the places I've been to only offer a 60 or 90 day quick replacement after that you have to deal with the manufacturer directly.
The only other option I can think of is to go buy a whole new system, even barebones that'll be twice what I would need to spend otherwise.
Mna this sucks. I've become really used to that next day courier replacement parts service Dell has.
Does anyone else experience problems with the P2B series over time? We have about 30 asus p2b based computers in our department, and they start needing replacement at an alarming rate.
The problems show up as solid lock-ups, at first maybe once a month, and then escalating to several lockups a day. The problems can be reproduced with the memtest.exe produced by alegr software (free download) as a lockup on pass 4 (always pass 4). Already tried exchanging all other parts except the motherboard so Im pretty sure Ive narrowed it down.
My theory is that the capacitors dry out over time and thus fails to meet their spec.
Out of thirty wewe so far had five go bad, and the problem is difficult enough to reproduce for the vendors to blame it on software problem.
I used to think of asus as good products, now Im not so sure anymore.
One simple basic thing works fine on all the Asus, Epox, Gigabyte, and Tyan motherboards I've ever worked with, and failed on all the Intel motherboards I've worked with. That thing is the PS/2 keyboard port when either not connected at boot/reset time, or connected to an electronic KVM switchbox which has not selected that machine at boot/reset time.
If the keyboard is not plugged in at all, and then you plug it in later, it doesn't work. Linux sees a keyboard device, but no keypresses ever get through until you reboot (and Ctrl-Alt-Del doesn't work).
If the keyboard is plugged in to the electronic KVM switch, and the switch has not selected that machine at the time, then it gets even worse. Something (BIOS or chipset) just hangs. Video sync comes up but the screen is blank except for a cursor. When the KVM is switched over to that machine, still nothing. If you move the mouse, sometimes it will start up and BIOS will initialize and boot the system. However, the keyboard and mouse ports are now transposed electically or logically in the chipset. Mouse movements go in as garbled keypresses, and real keypresses occaisionally tickle mouse logic.
It's not a fundamental chipset flaw that I can see, as I have some cases of the same chipset on both Intel and ASUS boards, and the ASUS boards work fine. It might be how the chipsets are configured by the engineers, though I personally suspect the BIOS is the major culprit.
As a desktop, this is OK. As a server, this sucks. Even the rack mount Intel ISP-1100 (TX440 motherboard, BX chipset) has this problem, and that machine is clearly intended for the server market. I have contacted Intel support and after the issue was bounced around several engineers, it finally came back as "Not supported with a KVM switch" and they just dropped the issue.
now we need to go OSS in diesel cars
It seems to me that Elitegroup Computer Systems bought PC Chips a while back. PC Chips (aka Eurone, but they had other aliases too) was the undisputed master of really crappy ultra-cheap mobos, with names like "TXPro". PC Chips would take random chipsets, notably ALi's, and relabel them to sound more like Intel's. The board quality was awful -- performance wasn't bad when it worked, but they had no quality control. El cheapo white box dealers moved a lot of them.
ECS, in those days, made a higher quality, if boring, board. Some respectable oems used them. I suspect that some of the ECS boards today are pretty stable, coming out of "old ECS", while others are warmed-over PC Chips (and probably better than the old PC Chips).
It looks like the reviewer was tainting all of ECS with the record of PC Chips.
Have a comment? Have something worth mentioning?
You're free to come and talk kietch with the author of the article at either #queenofgeeks or #aselabs at the chat.planetz.net server on port 7000
__________
"Yeah...it was the numbers that were irrational, not the murderous cult of vegetarians...." -- Hippasus of Metapontum
Commodity PowerPC mobos were available back in the mid-1990s.
Nobody bought them, which is the main reason you can't buy one now -- that and the fact that nobody wants to pay to maintain the Windows PPC port (which is a majority of the desktop market and a good chunk of the server market).
Both IBM and Moto expected great things in the desktop market, didn't find them after spending an assload of money, and retreated to safer markets.
I was a little dissapointed that the article didn't mention some of the higher end motherbard manufacturers like SuperMicro. If you want to build a really stable and reliable system most of these Tiawanese specials are a little lacking.
I was trying to get the onboard via8233 sound chip to work, and not getting anywhere.
I emailed Epox tech support, and someone emailed back in about 4 hours with ALSA drivers attached and instructions on how to compile and configure. Crank ON!
Admittedly, the sound chip appears to suck anyway (mp3s are fine, but sound in Tribes 2 is horrible - on my Athlon 1600XP). But you have to congratulate their tech support for jumping in with the ALSA drivers. I was expecting to be told to use the kernel drivers (which don't work) and seek support in newsgroups...