...more like...what can this motherboard do...M5ATC Rev 1.0

How-to identify your motherboard ?
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Hi All !!

Just got this board from a friend...Biostar M5ATC Rev 1.0. Already downloaded the manual, the latest *official* BIOS, and the patched one from here, but haven't flashed it as yet. Wanted to make sure these will support this Rev. of the board. It is going to replace my girlfriend's now defunct SuperMicro P5STE (i430HX).

My question centers around what this guy can support. I'm not familliar with the ALi 1531 B1/1543 B1 chipset features. This board only has SIMM sockets, but there are spots to install DIMM sockets, if I can find them. What are it's memory limitations? It also looks like it can run asychronous with the bus...can it support 83MHz with the PCI bus still at 33MHz? Couldn't find anything this old on the ALi site...info-wise.

As for the CPU, it does have a built-in VRM...looks like it can be set from 3.5V to 2.1V per the manual...is 2.0 V possible? What kind of current can this thing support on VCore? Can the BIOS recognize an AMD K6-2/+? If not, I'm sure it can be patched... (Hi Jan!!).

One thing I did notice right off is that the memory test programs I have show the memory speed to be much higher than on my DFI 586IPVG (i430VX, Cyrix MII, 66MHzx4, 128MB EDO, 512k)...currently the Biostar board is set properly for a 200MMX (66MHzx3, 64MB FPM, 256k) so I would have figured the test scores would be the about the same...does the ALi chipset have features that increase the memory timings/speed on the memory bus? Both boards seem to be setup the same in the BIOS for that, although there are a few more settings there that I'm not sure what they do...the difference is like at least 10MB/sec if I remember right...but I can post those if necessary later on.

OK...hope I can get some good info here, or at least some links to check out that may help me in setting this guy up right.

Thanks All !!
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

That board will behave exactly like the PCCHIPS TXPROs (M560, M565, M575).
And will very likely have the UDMA33 bug... :(
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Kinda saw some of that here and at the PCChips site...but wasn't sure the chipset was the same...mine do not say *TXPro* on them...just ALi and the ID numbers (mine were mfg'd week 40/44 1997).

But I'll check them out...please...continue to post if you have any more info.

UDMA33 bug...even in the patched version found here??

:D
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

TXPRO was a PCCHIPS "copyright"... :wink:

Kachiwachi... UDMA33 has nothing to do with the 32GB bug... :roll:
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Thanks AJZ...found that out while reading *alot* of stuff over at the PCChips site. I'll just have to remember to set the WD AC28400 to PIO-4 when I install it (is UDMA-66 capable) so I don't mess up her data.

Any more insight from anyone??
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Problem...

The VCore on this board is set like most others...4 jumpers that can set the voltage from 2.0 to 3.5 Volts. No problem here...

There is also a jumper JP15 which is only installed to set VI/O to 3.45 Volts when required. For all other VI/O, 3.5 or 3.3 V, it is off. I see no jumper to set VI/O from 3.3 to 3.5 V...how can this be??
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

My opinion:

When voltages 3.2V and below are selected, the I/O voltage will stay at 3.3V always, having detected dual-voltage mode.
When 3.3V upto 3.5V are selected (CPU voltage), then the CPU voltage takes over the I/O voltage, so it'll go from 3.3V to 3.5V depending on the CPU and how you may be "forcing" it.

JP15 might only be a tool for raising the I/O voltage only when in dual-voltage mode, or when using a single-voltage processor at less than 3.5V, wanting to raise the I/O for any particular reason.

I may be well off-track...
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Thanks AJZ for the insight...I'll have to fiddle some later.

As for the UDMA/33 bug, can that be disabled in the BIOS? (I didn't see any feature to do that.)

The drive that will go in as I said can support UDMA/66, and I've downloaded the WD tool for setting that...however it makes no mention of disabling *all* UDMA modes and limiting the drive to just PIO Modes. Does anyone know how/if that is possible?
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Um...guess I should re-phrase...

Not sure which BIOS setting to change...there seems to be two...

IDE Primary/Secondary Master/Slave PIO - Auto, 0-4
IDE Primary/Secondary Master/Slave UDMA - Auto, 0-4

Which should be set to limit UDMA from turning on?

(Just the UDMA (set it to PIO-4) and leave PIO in Auto?)
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

If it were that simple, then people wouldn't have had such a terrible time with ALi's IDE drivers and the M1543 southbridge.

When it's an AMI BIOS, the problem is even worse, because there is literally NO way of telling the BIOS to deactivate UDMA completely, so if you try to install a UDMA-capable HDD and click the DMA checkbox in Windows, then you're only looking for trouble (hangs, data corruption, etc).

On the other hand, I've seen some AWARD BIOSes (don't remember the boards) on M1543 boards that did have a "UDMA enable/disable" feature, so the solution was rather simple, because it could at least work correctly in MWDMA2 mode (traditional PIO4+DMA). Most AMI boards are limited to plain old PIO4.

BTW, changing the PIO won't solve the problem. Do you have any BIOS entry that makes reference to UDMA? Or is UDMA just displayed at POST?
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

UDMA works nice on my PC Chips M726 board (M1621+M1543)... I have a Matsonic MS6260S here but have not tested UDMA yet.
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

That's most probably because the BIOS has support for both the 1543 and 1543C... or was that from the release sheet of the 726mrt?
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

I have to correct myself...

IDE Primary/Secondary Master/Slave PIO - Auto, 0-4
IDE Primary/Secondary Master/Slave UDMA - Auto, 0-2, Disabled

So I guess by setting the UDMA to Disabled, no UDMA transfer modes will be available via the BIOS...or in Windows (98SE in my case).

I did confirm this last night with the Seagate 40GB drive I used to test Jan's patch for my DFI, which previously posted with UDMA5 on the M5ATC (if I remember correctly). By setting the UDMA to Disabled, the drive posted at PIO-4 (PIO was set to Auto).

Even though the Seagate posted with UDMA5, the ALi chipset cannot support this...it could only support it as UDMA2...if it worked...correct :?:

On a working UDMA chipset, the same applies...it could only run at UDMA2, even though it was showing UDMA5...correct :?:


So to be sure I understand this...

1) Hard Drive transfer modes are set in the BIOS...PIO or Ultra-ATA. You can set a lower transfer rate than the drive is capable of.

2) Windows DMA (checkbox) sets SW or MW DMA if drive is not Ultra-ATA compatible. If you don't check the Windows DMA box, PIO drive will stay in PIO mode.

3) If you don't check the Windows DMA box, Ultra-ATA drive (set as UDMA-Auto in BIOS) will stay in PIO mode :?:

4) Windows DMA (checkbox) sets UDMA if drive is Ultra-ATA compatible, or SW/MW DMA Mode only :?:


Numbers 3 and 4 above are what confuse me now...

Along these lines... PIO-3 is 11.1MB/s transfer rate. SW DMA2 is 8.3 MB/s transfer rate. If Windows DMA (checkbox) is checked, this drive will actually run SLOWER...correct???
ajzchips
El cheapo dude
Posts: 3048
Joined: Sat Mar 23, 2002 12:41 pm
Location: Madrid, Spain

KachiWachi wrote:I have to correct myself...
IDE Primary/Secondary Master/Slave PIO - Auto, 0-4
IDE Primary/Secondary Master/Slave UDMA - Auto, 0-2, Disabled
So I guess by setting the UDMA to Disabled, no UDMA transfer modes will be available via the BIOS...or in Windows (98SE in my case).
Exactly. It's good you have that option. AWARD BIOS?
Even though the Seagate posted with UDMA5, the ALi chipset cannot support this...it could only support it as UDMA2...if it worked...correct :?:
On a working UDMA chipset, the same applies...it could only run at UDMA2, even though it was showing UDMA5...correct :?:
Exactly. If and only if your BIOS is recent/intelligent enough to not try to force a mode the chipset isn't capable of (a typical AWARD BIOS bug).
So to be sure I understand this...
1) Hard Drive transfer modes are set in the BIOS...PIO or Ultra-ATA. You can set a lower transfer rate than the drive is capable of.
2) Windows DMA (checkbox) sets SW or MW DMA if drive is not Ultra-ATA compatible. If you don't check the Windows DMA box, PIO drive will stay in PIO mode.
3) If you don't check the Windows DMA box, Ultra-ATA drive (set as UDMA-Auto in BIOS) will stay in PIO mode :?:
4) Windows DMA (checkbox) sets UDMA if drive is Ultra-ATA compatible, or SW/MW DMA Mode only :?:
Numbers 3 and 4 above are what confuse me now...
Along these lines... PIO-3 is 11.1MB/s transfer rate. SW DMA2 is 8.3 MB/s transfer rate. If Windows DMA (checkbox) is checked, this drive will actually run SLOWER...correct???
1) Yes, but as I said, if your BIOS isn't modern enough (or intelligent enough) to adapt itself to the maximum setting the chipset supports, you may need to use the HDD utility to lower its UDMA setting to your chipset's.

2) Exactly. Disabling UDMA in BIOS (wherever possible), setting PIO4 and activating the DMA checkbox will "usually" get you MWDMA2.

3) Right. It's the checkbox that actually activates MWDMA2 or UDMA.

4) The DMA checkbox will activate UDMA if it hasn't been BIOS-disabled. It will activate MWDMA2 otherwise.

I have never seen a SWDMA2 device... I'd benchmark to find out, If I were you.

Note: we're presuming that the motherboard/chipset are UDMA capable.
Rainbow
The UniFlasher
Posts: 3122
Joined: Wed Mar 20, 2002 4:16 pm
Location: Slovakia
Contact:

ajzchips wrote:That's most probably because the BIOS has support for both the 1543 and 1543C... or was that from the release sheet of the 726mrt?
Don't know if there's M1543C because something like TXPro-II is printed on the chip (and covered by a label with BXCel). The BIOS does not enable UDMA, I did it manually by editing chipset registers (using WPCREDIT/WPCRSET).
Patched and tested BIOSes are at http://wims.rainbow-software.org
UniFlash - Flash anything anywhere
Post Reply