Biostar M7VIZ mobo OK or junk?

[larry moe 'n curly]

Is BioStar mobo quality OK?

I have a used BioStar M7VIZ v. 8.0 Socket A mobo with VIA VM400 chipset
and integrated video. It was given to me because it wouldn't boot, but
the problem turned out to be just a loose or corroded BIOS chip or
socket. However I noticed that there's only a single 20-pin power
connector, even though the CPU is powered from the +12V, and every
other mobo I've seen that used +12V for CPU power also had a 4-pin
ATX12V connector. Is the lack of that connector going to cause
problems? I don't plan on running a super-fast CPU with it.

The rest of the build quality seems better than PC Chips or ECS. For
example, the toroid coils have double-wound wire, there's a fuse
instead of a soldered jumper for the rear I/O, and the heatsink for the
north bridge is a lot bigger.

 

[kony]

Is BioStar mobo quality OK?

Normally I'd call them crap, but about a year ago I got one
for a song and it's still running. PCChips sometimes last
over a year too though, take it with a pinch of salt as it's
bios is buggy to the point where memory timings are very
strange.

I have a used BioStar M7VIZ v. 8.0 Socket A mobo with VIA VM400 chipset
and integrated video. It was given to me because it wouldn't boot, but
the problem turned out to be just a loose or corroded BIOS chip or
socket.

Well in their defense, they probably use a similar if not
indentical PROM socket to many of the budget brands, simply
plastic with nickeled copper contacts. Couldn've been the
environment the board was stored or used in? Wierd
placement of the chip though, if yours is to the left of the
NB like in a pic I saw.

However I noticed that there's only a single 20-pin power
connector, even though the CPU is powered from the +12V,

Are you certain it's powered from 12V or are you assuming
it?

I've never seen a board that used 12V rail for CPU without
*Some* kind of aux connector. I'll bet it uses 5V rail.
Measure the input voltage to the fets with a multimeter, or
right before the first inductor in the VRM circuit, on it's
leg.

Looking at a picture of the board, it does appear to use 5V
for CPU. Not big deal, especially with integrated video it
probably won't need a very hefty supply, maybe good 250-300W
with 180W combined 3V/5V rating.

and every
other mobo I've seen that used +12V for CPU power also had a 4-pin
ATX12V connector. Is the lack of that connector going to cause
problems? I don't plan on running a super-fast CPU with it.

I think you're making assumptions, that perhaps socket A
usually uses 12V? IIRC there is that vague reference to it
in the AMD literature but it is only a passing suggestion
still left up to board manufacturer's discretion. Most
socket A boards use 5V for CPU, it's only the nForce2 boards
were the average started turning towards 12V power, a few
KT600 and KT880 (or whatever that last Via Skt A chipset
was) and even then the biostar nForce2 board I have uses 5V
for CPU. Not a big deal really, several Asus boards did too
and were fine... 12V power for CPU becomes more of an issue
once CPU starts pushing 100W, too much heat density then for
the VRM circuit without a 3 stage setup.

The rest of the build quality seems better than PC Chips or ECS. For
example, the toroid coils have double-wound wire, there's a fuse
instead of a soldered jumper for the rear I/O, and the heatsink for the
north bridge is a lot bigger.

I wouldn't consider it a "lot" better, but I think that
might also have a bit to do with the market-segment for a
KM400 board- it's targeted at the cheapest system build
possible and towards that end it might be a reasonable
choice over ECS/PCChips. It's just hard to look favorably
at the KM400 since even if it's an office box without
signficant video memory bandwidth needs, the Via PCI bus was
slower (less efficient) than nForce is/was.

 

[larry moe 'n curly]

On 16 Jul 2005 04:44:10 -0700, "larry moe 'n curly"


Well in their defense, they probably use a similar if not
identical PROM socket to many of the budget brands, simply
plastic with nickeled copper contacts. Couldn've been the
environment the board was stored or used in?

This is a refurbished mobo, and maybe it was sent through a bath of
harsh chemicals because the metal tabs of the MOSFETs and AGP voltage
regulator are unusually dull, and the stainless steel covers for the
rear NIC/USB, keyboard/mouse, and audio connectors have mottled stains
that I couldn't remove with any chemical. The BIOS socket pins were
shiney, but the chip pins were not.

Are you certain it's powered from 12V or are you assuming it?

I've never seen a board that used 12V rail for CPU without
*Some* kind of aux connector. I'll bet it uses 5V rail.

I discovered this while checking the accuracy of the BIOS voltage
readings, and I was surprised when I saw 12V on the metal tabs of some
of the MOSFETs (other tabs measured 1.5V), so I took a look at the
board and found that the capacitors closest to the rear were rated
2200uF and 16V. The capacitors closer to the CPU were rated 3300uF and
6.3V but measured only 1.5V each. I also measured about 2.5A through
the yellow +12V wire but about 0A through the +5V.

I think you're making assumptions, that perhaps socket A
usually uses 12V?

This is actually my first +12V mobo, and even my P4 uses +5.0V (ECS
P4S5A2).

I wouldn't consider it a "lot" better, but I think that might also
have a bit to do with the market-segment for a KM400 board- it's
targeted at the cheapest system build possible and towards that
end it might be a reasonable choice over ECS/PCChips. It's just
hard to look favorably at the KM400 since even if it's an office
box without signficant video memory bandwidth needs, the Via PCI
bus was slower (less efficient) than nForce is/was.

I just hope it works better than my ECS K7VTA3 v. 8 (VIA KT333), whose
BIOS somehow got flashed when I was trying to flash the NEC DVD writer
in the same machine. I first thought that it had blown a regulator,
but all the voltages were normal, including ripple, and after
unsoldering the BIOS and using Uniflash to back up the chip, I verified
that it contained the DVD drive's BIOS.

 

[Jan Alter]

I'm running a Biostar M7NCD now through its 2nd year and I believe it's
about the same vintage as the M7VIZ. It's only been a couple of years that
all the cpu's started using the 4 pin 12 volt cpu connector. At first those
connectors were associated for only P4 processors and then the use migrated
to the AMD chips. It should be no big deal not using it. As far as Biostar
goes I have found them pretty solid little boards unless you try
overclocking them with by upshifting the fsb. That's when you may hit a few
hurdles. I've used 3 M7NCD boards now in builds with no problems.
I tried pushing an Athlon XP 2500 from its normal fsb (166 mhz) to 200
mhz and everything ran fine except when I tried rendering movies to DVD.
Never could finish any rendering at all. When I reverted to 166 mhz fsb
there was no problem completing movies. If you're interested to hear what
other folks say about that particular board read some of the reviews by
consumers who bought it at Newegg.com. Just input the model number to get
some views.

 

[kony]

This is a refurbished mobo, and maybe it was sent through a bath of
harsh chemicals because the metal tabs of the MOSFETs and AGP voltage
regulator are unusually dull, and the stainless steel covers for the
rear NIC/USB, keyboard/mouse, and audio connectors have mottled stains
that I couldn't remove with any chemical. The BIOS socket pins were
shiney, but the chip pins were not.

I'd be surprised if they were harsh chemicals as that would
be a bad thing to do to capacitors, but guessing at this
point wouldn't be too productive or useful to further any
goal.

I discovered this while checking the accuracy of the BIOS voltage
readings, and I was surprised when I saw 12V on the metal tabs of some
of the MOSFETs (other tabs measured 1.5V), so I took a look at the
board and found that the capacitors closest to the rear were rated
2200uF and 16V. The capacitors closer to the CPU were rated 3300uF and
6.3V but measured only 1.5V each. I also measured about 2.5A through
the yellow +12V wire but about 0A through the +5V.

I'm not sure that 16V caps are necessarily any indication of
anything as two lines making different boards might be
dipping into same stock of caps, though the 12V reading is
quite puzzling, even moreso when you report 0A 5V as the
board should certainly be using 5V rail regardless of
whether CPU used 12V or not. I hate to second-guess your
measurements but I suspect your readings are wrong.

I just hope it works better than my ECS K7VTA3 v. 8 (VIA KT333), whose
BIOS somehow got flashed when I was trying to flash the NEC DVD writer
in the same machine. I first thought that it had blown a regulator,
but all the voltages were normal, including ripple, and after
unsoldering the BIOS and using Uniflash to back up the chip, I verified
that it contained the DVD drive's BIOS.

That is odd, never heard of it happening till now. It's sad
that ECS couldn't put another 20 cents into an EPROM socket
either. I could see if a video card couldn't spare the
real-estate but ECS... gotta wonder if someday they'll just
start sending out some glue and a ball of twine and your'e
supposed to build a board out of that. I never did
understand why K7S5A sold so well except for the low cost.

 

[larry moe 'n curly]

On 16 Jul 2005 07:15:27 -0700, "larry moe 'n curly"


I'd be surprised if they were harsh chemicals as that would
be a bad thing to do to capacitors, but guessing at this
point wouldn't be too productive or useful to further any
goal.

Another thing I don't understand is why the conformal coating and all
the plastics seem unaffected.

Are you certain it's powered from 12V or are you assuming it?

I've never seen a board that used 12V rail for CPU without
*Some* kind of aux connector. I'll bet it uses 5V rail.

I'm not sure that 16V caps are necessarily any indication of
anything as two lines making different boards might be
dipping into same stock of caps, though the 12V reading is
quite puzzling, even moreso when you report 0A 5V as the
board should certainly be using 5V rail regardless of
whether CPU used 12V or not. I hate to second-guess your
measurements but I suspect your readings are wrong.

Here's what I measured a second time:

Q24: 9V gate, 1.5V drain, 0V source
Q23: 0V g, 12V d, 1.5V s

Q20: 3V g, 12V d, 1.5V s
Q21: 9V g, 1.5V d, 0V s

+3.3V: 3.6A +5.0V: 0.3A +12V: 2.7A

The amp readings are probably accurate to only 0.2-0.3A because of the
difficulty of zeroing the DC clamp-on probe (OK, I should have flipped
the probe and averaged both readings) and the 2-digit resolution of my
multimeter.

That is odd, never heard of it happening till now. It's sad
that ECS couldn't put another 20 cents into an EPROM socket
either. I could see if a video card couldn't spare the
real-estate but ECS... gotta wonder if someday they'll just
start sending out some glue and a ball of twine and your'e
supposed to build a board out of that. I never did
understand why K7S5A sold so well except for the low cost.

I flash the DVD drive in Windows because I didn't learn until later
that a DOS-based flasher was available. And I think that Windows
screwed it up because the flash program probably goes through it to
talk to the peripheral rather than talk to the hardware directly;
that's the only reason I can think of for it not working in Safe Mode.

One funny thing is that I probably wouldn't have gotten this mobo for
free if its BIOS had been soldered in rather than plugged into a
socket.

 

[David Maynard]

Another thing I don't understand is why the conformal coating and all
the plastics seem unaffected.

Why would that surprise you?

If they really did do anything to it, as opposed to the rather common case
of returned boards being OK, they might have run it through a flux cleaner
bath but one would hope a manufacturer would know better than to douse it
with "harsh chemicals."

 

[kony]

Here's what I measured a second time:

Q24: 9V gate, 1.5V drain, 0V source
Q23: 0V g, 12V d, 1.5V s

Q20: 3V g, 12V d, 1.5V s
Q21: 9V g, 1.5V d, 0V s

+3.3V: 3.6A +5.0V: 0.3A +12V: 2.7A

It's early so my math might be off but that appears to be
under 50W total. It seems far too low unless you had
absolutely nothing but a _bare_ board, and even then it'd be
puzzling as it wouldn't account for the 2.7A of 12V power.
Is it possible your multimeter is fubar?

The amp readings are probably accurate to only 0.2-0.3A because of the
difficulty of zeroing the DC clamp-on probe (OK, I should have flipped
the probe and averaged both readings) and the 2-digit resolution of my
multimeter.

I suggest that the readings are incredibly inaccurate. I am
at a loss to explain 12V on the fets but would check
continuity between the inductor immediately before them and
the 5V rail. I did not have a high enough resolution
picture to see the board's copper in detail but I still
suggest it uses 5V rail for CPU even though I have no
explanation for the readings. Is the board running now?
Was this with drives or other peripherals also powered? If
you felt daring you might pull the 12V lead out of the power
supply plug then try powering up the system without any 12V.
Naturally the fans shouldn't spin but could be powered by
one of the 4 pin plugs instead.

One funny thing is that I probably wouldn't have gotten this mobo for
free if its BIOS had been soldered in rather than plugged into a
socket.

Yep, it's amazing how much hardware is now abandoned for
various reasons, but then again I can see a board like that
not being worth troubleshooting to some techs, they'd rather
just sell a new set of parts.

 

[larry moe 'n curly]

It's early so my math might be off but that appears to be
under 50W total. It seems far too low unless you had
absolutely nothing but a _bare_ board, and even then it'd be
puzzling as it wouldn't account for the 2.7A of 12V power.
Is it possible your multimeter is fubar?

It's Tenma. I couldn't afford Fubar. But the voltages straight
from the PSU read normal.

The CPU is only a 1.6 GHz Duron -- my fastest CPU -- and I used
just the bare board plus a floppy drive to run MemTest86 and Gold
Memory (yet another Kingston ValueRAM DIMM failed).

I suggest that the readings are incredibly inaccurate. I am
at a loss to explain 12V on the fets but would check
continuity between the inductor immediately before them and
the 5V rail. I did not have a high enough resolution
picture to see the board's copper in detail but I still
suggest it uses 5V rail for CPU even though I have no
explanation for the readings.

I've measured other mobos that definitely used the +5.0V for CPU power,
and their MOSFETs never read higher than +5.0V on any pin. I don't
have a puller for those square Molex pins, but I measured about 800
ohms between each inductor and the +5.0V pins, and the polarity of the
meter leads didn't matter.

 

[kony]

I've measured other mobos that definitely used the +5.0V for CPU power,
and their MOSFETs never read higher than +5.0V on any pin. I don't
have a puller for those square Molex pins, but I measured about 800
ohms between each inductor and the +5.0V pins, and the polarity of the
meter leads didn't matter.

Hmm, maybe it does use 12V. Bad design IMO, though with
only a Duron 1.6 it may be ok, as that's probably less than
5A 12V power.

Usually you can get pins like those out of you stick a
needle on both sides (the sides perpendicular to the long
side of the plastic connector block) then slightly wiggle it
while you pull. I keep meaning to get an extractor myself
as I already have a round one but it seems I keep forgetting
till next time I fool with one.