I have used the MSI K7T Pro2A with an AMD 1GHZ chip and about 1/3 times burning a cd would cause a blue screen of death no matter what drivers I had nor switching the cd-r to it's own channel did not do the trick. I would experience frequent blue screens when doing intensive I/O traffic. After countless hours trying to solve the problem I decided to sell it and get an Asus A7V with the promise controller onboard. This, so far, has solved all my problems, and I have never had any problems with the A7V. I liked the MSI mainboard and thought the quality was excellent, although I have to relate the problem with the Via Drivers. I believe VIA may have rushed the development of the SouthBridge without providing the proper drivers. Almost always these problems exist in newer technology and will be ironed out in the future.
I remember in late 99 when i bought an fic-sd11, DOA, replacemant, doa..the entire batch that was sent here on the East Coast of Canada were bad!!
The gigabyte 7ix boards came in about a week later..i bought one and it was as solid as a rock...still solid to this day!!
BTW..the fic boards had a VIA southbridge!!
The gigabyte 7ix boards came in about a week later..i bought one and it was as solid as a rock...still solid to this day!!
BTW..the fic boards had a VIA southbridge!!
Leo,
I also got MSI K7T Pro2-A with Thunderbird 1 GHz. The first board was defective (to my surprise) and it failed to even install Windows 2000 properly. I didn't realize this and a couple of hours later, it even refused to boot up to the BIOS screen.
I saw the 4 red LED lights, so the next day I got an exchange from the store. The store manager looked at me funny when I said that I brought back a defective MSI board. Now everything seems to be running fine.
I also got MSI K7T Pro2-A with Thunderbird 1 GHz. The first board was defective (to my surprise) and it failed to even install Windows 2000 properly. I didn't realize this and a couple of hours later, it even refused to boot up to the BIOS screen.
I saw the 4 red LED lights, so the next day I got an exchange from the store. The store manager looked at me funny when I said that I brought back a defective MSI board. Now everything seems to be running fine.
The Pro2A definitely has it's draw backs, such as only 2 3-pin power connectors, virtually no memory tweaking from the BIOS, and seemingly, inferior memory bandwidth from the KT133 standard. But it is stable, for me anyway. The only crashes I've had were from programs that didnt feel like closing, or my undervoltage testing. As for ATA100, I cant compare my benchmarks to a Pro2, but looking only at HD Tach, I score well above average. My only complaint of MSI is how they play big brother by not including such standard settings as Bank Timings in the BIOS. I mean, my oldest computer had Bank Timings (Fast, Turbo, Normal, etc), CAS Latency, etc. MSI says "I will check the module for you, and if you dont like it you can use the slowest settings possible: x/3/3/6 w/o any interleaving, punk." So I really hope they address that in a BIOS update, but I've stressed my system with Sandra burn-in, hours of CS, DivX set to level 4 DVD quality, and all passed...
I agree that MSI is a good company, I've used the BX master and 6905 slotket and they both performed much better than their Abit counterparts. I dunno what went wrong with the K7T Pro2A.
My only question is...
Whats the best BIOS to use with this board? I need something with a high range of voltages that will do better with DMA enabled. I hear the newest one has fried a few boards so I'm reluctant to try it.
My only question is...
Whats the best BIOS to use with this board? I need something with a high range of voltages that will do better with DMA enabled. I hear the newest one has fried a few boards so I'm reluctant to try it.
that tech support stuff sucks man. I have to say I've lost some faith in Anandtech for recommending this board. And Modus, I've had two of these boards...both have had identical problems. My first one fried after a bad BIOS update and my second one is plagued with all the same problems as the first.
Adul
Elite Member
Just wanted to add that i buld many systems and the first few pro2a's were good but the last 2 have the exact some symptoms decrbed by leoV and others. Random crashes bsods etc.
I'm starting to loose faith as well.
I'm starting to loose faith as well.
Thanks for the heads up, guys. A couple of more comments:
1) This sounds like a batch problem, either bad boards or possibly a bad run of the 686B southbridge, since there are people out there with functioning boards. I'd tend to lean toward the southbridge because other manufacturers don't seem to be having this problem and MSI and Epox were the first out of the gate with 686B southbridges on their boards. Considering that VIA is known for improving quality and reliability on later batches of their chips, and MSI and Epox I'd guess were most likely to have gotten the first production batches of the 686B, then I'd expect them to have the problems first. BTW, isn't it kind of curious that Abit and Asus waited and didn't drop the 686B immediately into their KT133 products? There's already rumblings out that there's a FSB limitation (won't go above much 145MHz FSB) on the first batches of KT133A chips as well, but later batches will hit 160MHz. That may be the problem with Anand's KT7A-RAID board. A suggestion - if there isn't a sticker over it, perhaps everyone with a problem could compare the date codes on their 686B and see if there's any consistency. Your chip should look like:
VT82C686B
XXXXCD TAIWAN
The "XXXX" should look like "0043" for a 2000 Week 43 chip.
2) For those of you who are unhappy with the untweakability of the MSI BIOS, I've heard H. Oda's WPCREDIT now supports the KX/KT133 chipset. I haven't used it, but there's a PCR file for the chipset available.
A general rant - I'm a bit irritated that VIA intentionally crippled the KT133 so that it wouldn't support a 133MHz FSB. I'm not absolutely positive about it, but everything points in that direction:
1) The KX133, the KT133's predecessor could frequently hit the 120s and even 133MHz FSBs. It's virtually unheard of for a KT133 to reach the 120s, let alone 133Mhz.
2) My Abit KT7's BIOS sets the PCI bus at 41MHz (/4 divider) at the 133MHz FSB. Doesn't seem unusual until you consider that the KT7's clock generator also supports a 33Mhz PCI bus (/3 divisor). Not that it can reach it, but why isn't the 33MHz PCI bus the default at 133Mhz? The /4 divisor doesn't kick in until the BIOS reaches 136MHz. Just looks like another artificial barrier.
3) If KT133s were regularly hitting the 133MHz FSB, how would AMD's DDR push have looked? As it is, DDR got a huge publicity boost and a couple of weeks later suddenly here comes the KT133A.
1) This sounds like a batch problem, either bad boards or possibly a bad run of the 686B southbridge, since there are people out there with functioning boards. I'd tend to lean toward the southbridge because other manufacturers don't seem to be having this problem and MSI and Epox were the first out of the gate with 686B southbridges on their boards. Considering that VIA is known for improving quality and reliability on later batches of their chips, and MSI and Epox I'd guess were most likely to have gotten the first production batches of the 686B, then I'd expect them to have the problems first. BTW, isn't it kind of curious that Abit and Asus waited and didn't drop the 686B immediately into their KT133 products? There's already rumblings out that there's a FSB limitation (won't go above much 145MHz FSB) on the first batches of KT133A chips as well, but later batches will hit 160MHz. That may be the problem with Anand's KT7A-RAID board. A suggestion - if there isn't a sticker over it, perhaps everyone with a problem could compare the date codes on their 686B and see if there's any consistency. Your chip should look like:
VT82C686B
XXXXCD TAIWAN
The "XXXX" should look like "0043" for a 2000 Week 43 chip.
2) For those of you who are unhappy with the untweakability of the MSI BIOS, I've heard H. Oda's WPCREDIT now supports the KX/KT133 chipset. I haven't used it, but there's a PCR file for the chipset available.
A general rant - I'm a bit irritated that VIA intentionally crippled the KT133 so that it wouldn't support a 133MHz FSB. I'm not absolutely positive about it, but everything points in that direction:
1) The KX133, the KT133's predecessor could frequently hit the 120s and even 133MHz FSBs. It's virtually unheard of for a KT133 to reach the 120s, let alone 133Mhz.
2) My Abit KT7's BIOS sets the PCI bus at 41MHz (/4 divider) at the 133MHz FSB. Doesn't seem unusual until you consider that the KT7's clock generator also supports a 33Mhz PCI bus (/3 divisor). Not that it can reach it, but why isn't the 33MHz PCI bus the default at 133Mhz? The /4 divisor doesn't kick in until the BIOS reaches 136MHz. Just looks like another artificial barrier.
3) If KT133s were regularly hitting the 133MHz FSB, how would AMD's DDR push have looked? As it is, DDR got a huge publicity boost and a couple of weeks later suddenly here comes the KT133A.
It must be really hit 'n miss then because my Pro2A works like a charm. Never once have I got a BSOD and it boots very fast all the time. The onboard sound works superbly and I have never got any coasters (due to it).
My only complaint is that I can't get the fsb above 103, but my Duron600 can't do much more than 1030 anyways (I originally had it on a KT7 which was a much more troublesome board).
The thing is Leo V, I got my board from John too ...and it works perfectly ...so I don't think it can be just called a bad batch like RD hinted at.
My only complaint is that I can't get the fsb above 103, but my Duron600 can't do much more than 1030 anyways (I originally had it on a KT7 which was a much more troublesome board).
The thing is Leo V, I got my board from John too ...and it works perfectly ...so I don't think it can be just called a bad batch like RD hinted at.
"The "XXXX" should look like "0043" for a 2000 Week 43 chip."
That's exactly what mine says under the ATA100 sticker--it's a 2000 Week 43 chip. I appreciate your diplomacy in calming everyone down, but it doesn't change the fact. VIA/MSI released a defective product into the market. Nothing has been done about it. Whatever theoretical solutions VIA might produce in the future are NO excuse for that.
I'm seriously considering jumping the AMD ship altogether. I've been one of AMD's strongest defenders, but I'm NOT going to be an AMD apologist. In the rare cases when Intel released defective products, they were immediately recalled and replaced. This is the least I expect of the AMD/VIA alliance if they're to keep my business.
That's exactly what mine says under the ATA100 sticker--it's a 2000 Week 43 chip. I appreciate your diplomacy in calming everyone down, but it doesn't change the fact. VIA/MSI released a defective product into the market. Nothing has been done about it. Whatever theoretical solutions VIA might produce in the future are NO excuse for that.
I'm seriously considering jumping the AMD ship altogether. I've been one of AMD's strongest defenders, but I'm NOT going to be an AMD apologist. In the rare cases when Intel released defective products, they were immediately recalled and replaced. This is the least I expect of the AMD/VIA alliance if they're to keep my business.
Leo V,
You really don't have any basis for claiming the whole product is deffective. The one YOU got might be deffective, but to indict three companies because of your own bad luck is a bit absurd.
Return the board. Get a new one. And have a beer or something.
Modus
You really don't have any basis for claiming the whole product is deffective. The one YOU got might be deffective, but to indict three companies because of your own bad luck is a bit absurd.
Return the board. Get a new one. And have a beer or something.
Modus
IMHO, a 900 tech support line is unacceptable. Personally, I would never buy anything from a company with a 900 tech support line. No 800 number is no big deal - I can buy 3 cents/minute long distance phone cards from a local convenience store 
WetWilly, maybe it does have to do with the batch # afterall then. I just peeled the ATA100 sticker back and mine is a week 41 (0041CD TAIWAN). My board works flawlessly.
In regards to a couple other complaints I missed above, I have had no trouble with the LiveBIOS (used twice without a hitch) and to the guy who said the memory bandwidth sucks well I have to disagree. I only have a Duron 600 and 128mb ram and here are my scores ---> Sandra mem bench for 2a
In regards to a couple other complaints I missed above, I have had no trouble with the LiveBIOS (used twice without a hitch) and to the guy who said the memory bandwidth sucks well I have to disagree. I only have a Duron 600 and 128mb ram and here are my scores ---> Sandra mem bench for 2a
i have 2 problems with my Pro2-A:
1. my factory unlocked Duron 600 will run at 818mhz/1.7V if i go 7.5x109, but not if i do 8x100 :|
2. if i use 2 stick of ram(both of which work, and i tried with 4 other sticks as well) i can't install anything large/run them, like UT, Theif II and the like, which is really pissing me off because i payed good money for 384mb of ram and can only use 256 :|
if anyone knows how to fix either of these problems/a way to contact MSI, i'd greatily appreciate it.
1. my factory unlocked Duron 600 will run at 818mhz/1.7V if i go 7.5x109, but not if i do 8x100 :|
2. if i use 2 stick of ram(both of which work, and i tried with 4 other sticks as well) i can't install anything large/run them, like UT, Theif II and the like, which is really pissing me off because i payed good money for 384mb of ram and can only use 256 :|
if anyone knows how to fix either of these problems/a way to contact MSI, i'd greatily appreciate it.
Modus, about a dozen Pro2-A owners have reported the same problems. Just how low is your standard of failure? No offense, but you don't happen to be associated with one of these companies, hmm?
I just wanted to add that I've experienced some random crashes with my Power Management functions in Windows 2000. This only happened with my MSI board. The Gigabyte board seems to be rock stable.
FYI, my VIA chipset is from week 50 of 2000.
FYI, my VIA chipset is from week 50 of 2000.
Leo V,
A dozen failures is absolutely meaningless. How many hundreds of others have no problems with the board and felt no need to complain as you have? I can guarantee you that the silent majority who have used the board successfuly are not going to bother to chime in and report it. Forums like this tend to amplify only the bad results, while people such as myself and Red Dawn and Patrick at PC Resources who have a combined experience with hundreds of these boards report no abnormalities.
As for the tech support number not being toll free -- welcome to the real world. Check around with other Taiwaneese manufacturers. Good luck finding 1-800 numbers.
GET A REPLACEMENT from your reseller. Only then can you even begin to contemplate pronouncing the entire line deffective.
Modus
A dozen failures is absolutely meaningless. How many hundreds of others have no problems with the board and felt no need to complain as you have? I can guarantee you that the silent majority who have used the board successfuly are not going to bother to chime in and report it. Forums like this tend to amplify only the bad results, while people such as myself and Red Dawn and Patrick at PC Resources who have a combined experience with hundreds of these boards report no abnormalities.
As for the tech support number not being toll free -- welcome to the real world. Check around with other Taiwaneese manufacturers. Good luck finding 1-800 numbers.
GET A REPLACEMENT from your reseller. Only then can you even begin to contemplate pronouncing the entire line deffective.
Modus
Hmmm... I've been having some problems with my Pro2a also. The onboard sound doesn't work properly for some reason. It would work for a random period of time, then it will just go out and start making a high pitch squeal. It didn't really bother me that much since I was planning to replace it with a SBLive that week anyways. My SBLive btw is working perfectly, after struggling with it for abit to get rid of IRQ conflicts. Another problem I'm having, but I'm not sure if it's motherboard related. Whenever I watch DVD movies, the system will lockup sometimes with a BSOD, random reboot, or black screen with a green line on top. I'm not sure if it's fixed or not with some driver updates. The leaked Nvidia drivers seemed to have fixed it though. I'm not 100% sure though, but a movie hasn't locked up on me since then. The way I see it is that it'll eventually get fixed with proper drivers... I Hope...
i had problems with the abit kt7 and went with this msi board and I'm all good.
you can see my link to my system with an ata100 hard drive. My system seems all good to me. Had it for a couple months now and only had to reinstall win98se once after a corrupt registry when trying to get sacrifice to work with my voodoo 5500 and not readint the readme. my cousin rebooted like 10 times when the game failed to start and this error started so I reinstalled and thats where Im at now(I had to change a setting in the control panel)
I can also burn cd's at full speed with burn proof and do other things at the same time. I simply haven't had hard freezes when say installing things or playing games or surfing and listening to music.
one thing I have noticed that keeps crashing or hanging the computer is windows media player7. for some reason it takes forever to load some video clips and will hang in the background or screwup the computer. I have since ditched that piece of crap for whatever reason.
you can see my link to my system with an ata100 hard drive. My system seems all good to me. Had it for a couple months now and only had to reinstall win98se once after a corrupt registry when trying to get sacrifice to work with my voodoo 5500 and not readint the readme. my cousin rebooted like 10 times when the game failed to start and this error started so I reinstalled and thats where Im at now(I had to change a setting in the control panel)
I can also burn cd's at full speed with burn proof and do other things at the same time. I simply haven't had hard freezes when say installing things or playing games or surfing and listening to music.
one thing I have noticed that keeps crashing or hanging the computer is windows media player7. for some reason it takes forever to load some video clips and will hang in the background or screwup the computer. I have since ditched that piece of crap for whatever reason.
Modus, I wouldn't mind calling coast-to-coast long-distance so much, if somebody picked up at the other end. MSI's answering machine with unreturned messages is absolutely the most pathetic excuse for customer service possible.
Your suggestion that my problems, or those of many fellow Anandtechers, are nothing is preposterous. The first step is contacting the motherboard manufacturer to verify that this is, in fact, simply a defect (seems like a rather common defect to me!) I'd rather not waste a month, or however much longer MSI will take, to obtain a replacement with the same flaws. The problem is, MSI doesn't find it expedient to pay for "tech support". But perhaps you'll claim that I called MSI at the most inopportune moments, when all their phonelines were tied up? I mean, how could it possibly be MSI's fault?
Your suggestion that my problems, or those of many fellow Anandtechers, are nothing is preposterous. The first step is contacting the motherboard manufacturer to verify that this is, in fact, simply a defect (seems like a rather common defect to me!) I'd rather not waste a month, or however much longer MSI will take, to obtain a replacement with the same flaws. The problem is, MSI doesn't find it expedient to pay for "tech support". But perhaps you'll claim that I called MSI at the most inopportune moments, when all their phonelines were tied up? I mean, how could it possibly be MSI's fault?
i dont care what you do or where you go but to be so loyal to one company is insane. i mean..i prefer amd, but if someone gave me a pIII ill take it with no questions asked. id be excited to learn more about intel cause im getting bored with al the amd boards and such. and honestly, you cant say thats its VIA's, MSI's, or AMD's fault. you had a few problems..and so what?
<< 'd rather not waste a month, or however much longer MSI will take, to obtain a replacement with the same flaws >>
shipping takes time, as does rma'a and anything else you do via mail order. it sucks but thats the real world buddy. took me three weeks to get my abit kt7raid from the MO place but hey, i got it and now my board runs flawlessly. if you didnt want to deal with delayed shipping and rmas and such, you shouldve bought it from a local dealer and make them deal with it. i dont know msi boards, but they seemed to have proven themselves quite well and if you got one bad board you have to understand that SH!T HAPPENS!!
my 2 cents
loo
<< 'd rather not waste a month, or however much longer MSI will take, to obtain a replacement with the same flaws >>
shipping takes time, as does rma'a and anything else you do via mail order. it sucks but thats the real world buddy. took me three weeks to get my abit kt7raid from the MO place but hey, i got it and now my board runs flawlessly. if you didnt want to deal with delayed shipping and rmas and such, you shouldve bought it from a local dealer and make them deal with it. i dont know msi boards, but they seemed to have proven themselves quite well and if you got one bad board you have to understand that SH!T HAPPENS!!
my 2 cents
loo
I hate to drop in late, but has anyone tried the IDE-cable trick yet ? You know, use an old IDE-cable with only 40 pins instead of 80 to force it to do ATA-33 ? It just might work
As for Via chipsets... I have a mobo based on one, and I have to say I'm not too happy about it. This one is an Apollo Pro 133A. You can't imagine the things I had to do to get semi-decent performance out of it.
A few examples :
- Due to slow BIOS-updating on AOpen's behalf, I had to use those funky programs from H.Oda (Wpcredit and Wpcredit) to hack the chip in hex-modus to get the memory performance to levels that are just better then the BX-chipset. Granted, if you use a mobo from Asus and MSI, a 4-way mem-interleaving setting is already incorporated in the BIOS, but if it isn't, you're stuck with not so stellar performance And I think Via is to blame for this. Asus, generally known to tweak everything of the chipset to squeeze that last bit of performance out of it without sacrifising stability, was one of the first to offer the IOQ & 4-way mem-interleaving settings. We had to wait almost an eternity to get those. Now, either Via doesn't have the people & resources to offer acceptable support to the motherboard makers, or they simply don't care.
- The 686A-southbridge. WetWilly said the peformance of the southbridge was acceptable on his motheboard. Now, I can garantee you that it isn't up to snuff on my board. I run the latest BIOS & drivers (official, but beta's don't add anything more).
- Non-functioning AGP4x support for my vid-card (TNT2), that supports AGP4x according to Nvidia.
- Sub-par support in Linux for the chipset. This is changing due to the new 2.4 kernel, but still.
All die-hard Via fans just have to agree on this point : Intel chipsets don't have this kind of problems ! My older BX-mobo, a simple no-name board with integrated graphics, beat my new board in almost every memory benchmark, without even tweaking it !
I'm sure if Intel would've released such a buggy and slow performing chipset, this board would scream of murder and rape !
Via 'may' release those flawed chipsets. You know why ? Since they give the public what it wants. Nobody wanted Rambus, and Via was there with an alternative. The die-hard AMD fans also must defend Via, since it's the only chipset maker that can produce up-to date chipsets. All problems associated with it are happely taken. Although this last point isn't all that valid anymore with all the new chipsets for the Athlon processor.
Anyway, I think that it has been proven that there is something really wrong with the 686B southbridge. Leo V is a well-known member of this board, and I'm sure he knows what he's doing. As is Tomshardware.
Maybe Tom should write an article about it. Perhaps it will triger the recall of all faulty chips, just like the PIII1.13Ghz
As for Via chipsets... I have a mobo based on one, and I have to say I'm not too happy about it. This one is an Apollo Pro 133A. You can't imagine the things I had to do to get semi-decent performance out of it.
A few examples :
- Due to slow BIOS-updating on AOpen's behalf, I had to use those funky programs from H.Oda (Wpcredit and Wpcredit) to hack the chip in hex-modus to get the memory performance to levels that are just better then the BX-chipset. Granted, if you use a mobo from Asus and MSI, a 4-way mem-interleaving setting is already incorporated in the BIOS, but if it isn't, you're stuck with not so stellar performance
And I think Via is to blame for this. Asus, generally known to tweak everything of the chipset to squeeze that last bit of performance out of it without sacrifising stability, was one of the first to offer the IOQ & 4-way mem-interleaving settings. We had to wait almost an eternity to get those. Now, either Via doesn't have the people & resources to offer acceptable support to the motherboard makers, or they simply don't care.- The 686A-southbridge. WetWilly said the peformance of the southbridge was acceptable on his motheboard. Now, I can garantee you that it isn't up to snuff on my board. I run the latest BIOS & drivers (official, but beta's don't add anything more).
- Non-functioning AGP4x support for my vid-card (TNT2), that supports AGP4x according to Nvidia.
- Sub-par support in Linux for the chipset. This is changing due to the new 2.4 kernel, but still.
All die-hard Via fans just have to agree on this point : Intel chipsets don't have this kind of problems ! My older BX-mobo, a simple no-name board with integrated graphics, beat my new board in almost every memory benchmark, without even tweaking it !
I'm sure if Intel would've released such a buggy and slow performing chipset, this board would scream of murder and rape !
Via 'may' release those flawed chipsets. You know why ? Since they give the public what it wants. Nobody wanted Rambus, and Via was there with an alternative. The die-hard AMD fans also must defend Via, since it's the only chipset maker that can produce up-to date chipsets. All problems associated with it are happely taken. Although this last point isn't all that valid anymore with all the new chipsets for the Athlon processor.
Anyway, I think that it has been proven that there is something really wrong with the 686B southbridge. Leo V is a well-known member of this board, and I'm sure he knows what he's doing. As is Tomshardware.
Maybe Tom should write an article about it. Perhaps it will triger the recall of all faulty chips, just like the PIII1.13Ghz
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter