Looking for Award Flasher 5.00 or can I use UniFlash?

Discusses BIOS flashers and utilities from Award, AMI and Uniflash
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Hi

I have a TMC AI5VG+ (now out of business) with a K2-2/500 processor and have downloaded Jan Steunebrink's modified bios M598U4_J so I can install a K2-3 processor.

The BIOS Utilities page says "Award Flasher v5.00 should be used with older BIOS versions (eg. 4.50)"

I cannot find AF v5.00 - all the Google links are dead. Even 5.xx links are dead (does it have to be a .00 version?)

I've been to UniFlash web site but there seems so many caveats, warnings and untested boards (I appreciate it's a mamoth task) that I'm unsure whether UniFlash will safely do the job. Advice is appreciated.

By current BIOS is
BIOS Manufacturer : Award Modular BIOS v4.51PG
BIOS ID : 11/19/1998-VP3-586B-W877-2A5LEM2KC-00
BIOS Date : 11/19/98
BIOS OEM Signon : AI5VG+-U2-1119
BIOS ROM Size : 256K
Chipset : VIA 82C597 rev 4
Super I/O Chip : Winbond 83877F (use 87h) found at port 3F0h

If I were to use AWDFlash then I'm unsure whether to use the following options or not:
/F = use flash routines in original BIOS for flash programming
/LD = destroy cmos checksum and no system halt for first reboot after programming


Note: to start with I am not inserting the K2-3. I will do that later once the BIOS update has gone OK.

Thanks for reading.
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

Uniflash should work OK with most (if not all) socket 5/7 boards.
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Denniss wrote:Uniflash should work OK with most (if not all) socket 5/7 boards.
Thanks, Denniss. I've just done a successful Bios upgrade.

:D

Next step is to try the K2-3....
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Please report back with how that goes.

Thanks.
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Hi KachiWachi
KachiWachi wrote:Please report back with how that goes.
The Bios upgrade has gone fine and so did the upgrade to the K6-3/400.

Some timings

Load Win98......Load FF1.5......Load Acrobat
AI5VG+ K6-2/350
03:00...............00:35...............n/a
AI5VG+ K6-2/500
02:45...............00:25............00:30
AI5VG+ K6-3/400
02:25...............00:22............00:23
AI5VG+ K6-3/400 + Faster memory timings
02:20................00:20............00:23

Not a great deal of difference but hopefully better with reading/writing to/from memory with the onboard L2 cache.

Cheers
edwin
The Hardware Archivist
Posts: 6286
Joined: Wed Mar 20, 2002 7:11 pm
Location: Netherlands
Contact:

Might be time for a better harddisk too then... when was the last time you defragged, win98 needs that about every 2 weeks in regular use. It messes up fast...
edwin/evasive

Do not assume anything

System error, strike any user to continue...
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Hi edwin
edwin wrote:Might be time for a better harddisk too then... when was the last time you defragged, ..
Defragged just before the upgrade. Even on my XP m/cs I'm still in the habit of regular defrags, but they are very quick on these SATA drives but then XP is not so fussy about eliminating all the gaps between the files.

With the new ATA-100 card I could upgrade the HD to 7,200 rpm / UDMA 5 drive which I think is the max. for a 33MHz bus. Say a 20GB WD Caviar WD205BB. This is a 2nd backup PC and is for my wife to do basic WP & email and a bit of internet. Though I'm thinking of taking it back for a little experimenting with Linux [grin]
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

To improve older VIa-based motherboards:
Improve the PCI performance with either the somewhat official RAID perfromance patch from viaarena.com or the unofficial from http://www.georgebreese.com/net/software/

The second link also has the memory interleave enabler
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Hi Denniss

Bad Hair Day here I'm afraid. PC wouldn't boot - ATA-100 saw no devices. So I reset SDRAM Cycle Length back to 2; Cache Timings back to Fast and now OK - maybe co-incidental, maybe not.
Denniss wrote:.... or the unofficial from http://www.georgebreese.com/net/software/
Thanks for that URL. I download WCPUID and WPCREDIT and installed WCPUID - it has a nice GUI. Then went and used my main PC.

Then noticed the K6 screen went into sleep mode and then a little later I heard strange clickings from the hard disk. Couldn't wake up the screen/PC so reset and rebooted. Windows fails to load as many files and directories have gone. Had to recover from image backup of C:. Other partitions seem OK.

Surely, WCPUID is read only - or could it cause the HD corruption problem?
Denniss wrote:The second link also has the memory interleave enabler
Thanks again, that utility seems useful and tempting. But with the above I'm going to give the system a little while to show it's stable before making any more changes. I did note that the Memory interleave enabler utility doesn't work with 16Mbit chips, so I need to check that out. But with 16 chips on 128MB DIMMS I make that 8MBit chips.

Cheers for now.
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

@Strange clicking - sounds like the HDD does not like to wake up from sleep mode anymore.

@Interleave - this can only happen if your RAM chips support this feature.

@WPCREDIT - do you have a PCR file (and datasheet) for your chipset?
Denniss
BIOS Guru
Posts: 3153
Joined: Thu Mar 21, 2002 8:16 pm
Location: Near Hannover (CEBIT) Germany
Contact:

Modifying memory and cache timings too far may result in data loss. Especially cache-timings are prone for this.

Test memory timings (or memory interleave enabler) always with Prime95, modified cache-timings should always be tested with Prime 95 and with a defrag program verifying what it writes. (I usually have Norton Utilities with write verification for this).

P.S.: 16 chip 128MB sticks consist of 64MBit memory chips, 16-chip 256MB stick consist of 128MBit chips. You seem to have confused MBit with MB.
cp
BIOS Guru
Posts: 1914
Joined: Mon Oct 21, 2002 9:07 pm
Location: Germany

http://www.mit.edu/afs/sipb/contrib/doc ... 82c597.pdf

check for device 0 offset 64, 65 and 66 (for bank0, bank1 and bank2).
Alan_UK
BIOS Newbie
Posts: 16
Joined: Tue Dec 05, 2006 6:16 pm

Thanks Guys for all your tips and advice.
KachiWachi wrote:@Strange clicking - sounds like the HDD does not like to wake up from sleep mode anymore.
I've written in another post that the HDD now it's on the ATA-100 doesn't go into sleep mode - no visible parking and winding down like when it was on the EIDE. All BIOS power mgt set off except:
..............ACPI Function: Enabled
..............Power Mgt: User Defined (but all PM Timers/PM Events are Off or Disabled)
..............PM Controlled by APM: Yes
I've a feeling the ATA-100 doesn't support it as the manufacturer does mention it on their 133 model.
KachiWachi wrote:@Interleave - this can only happen if your RAM chips support this feature.
Everest says:
Supported Memory Interleave 4-Way
Current Memory Interleave 1-Way
KachiWachi wrote:@WPCREDIT - do you have a PCR file (and datasheet) for your chipset?
Yes, I have the PCR for the 598 and 597 VIA chips which I have.
Denniss wrote:P.S.: 16 chip 128MB sticks consist of 64MBit memory chips, ....
Thanks for putting me right! :) and for the hint to use Prime85 - I'll hunt it down.
cp wrote:check for device 0 offset 64, 65 and 66 (for bank0, bank1 and bank2).
I see in the ref. doc. that those offsets ref to DRAM Timings.

Everest states:
Chipset Properties:
Motherboard Chipset VIA VT82C598 Apollo MVP3
Memory Timings 3-3-3-6 (CL-RCD-RP-RAS) [edited to correct]
[ DIMM1/2/3: Infineon HYS64V16300GU-7.5 ]
Memory Timings:
@ 133 MHz 3.0-3-3-6 (CL-RCD-RP-RAS)
@ 100 MHz 2.0-2-2-5 (CL-RCD-RP-RAS)

Memory Module Features:
Early RAS# Precharge Not Supported
Auto-Precharge Supported
Precharge All Supported
Write1/Read Burst Supported
Buffered Address/Control Inputs Not Supported
Registered Address/Control Inputs Not Supported
On-Card PLL (Clock) Not Supported
Buffered DQMB Inputs Not Supported
Registered DQMB Inputs Not Supported
Differential Clock Input Not Supported
Redundant Row Address Not Supported

The Bus is 100MHz - I went for 133MHz chips to provide some margin on quality.

My default settings are:
Memory Timings:
CAS Latency (CL) 3T [edited to correct]
RAS To CAS Delay (tRCD) 3T
RAS Precharge (tRP) 3T
RAS Active Time (tRAS) 6T
and I tried changing the CAS Latency (CL) to 2T but set it back - see earlier post. [edited to correct]

Originally the speed of the memory in the BIOS was 10ns. After I upgraded to Jan Steunebrink's BIOS the default was 8ns but for safety I reset to 10ns.

Any suggestions as to what is worthwhile changing *safely* appreciated.
cp
BIOS Guru
Posts: 1914
Joined: Mon Oct 21, 2002 9:07 pm
Location: Germany

*grrr* this forum just ate my post :(

okay, here we go again..first of all: set the turbo option in the bios at ram timings. use memtest86+ to check if everything is fine. the turbo doesn't set 4-way interleaving btw. to have 2-2-5-2.0 4-way interleaved timings write 00010010b aka 12h in byte 64, 65 or 66. you can set different timings for each bank (nice feature). if they would have enabled 4-way interleaving by default they would have had a 20% larger bandwidth on ram and would have blown away that i440BX.
you can also fiddle around with the pci registers at byte 70ff. most bios don't set them at all so that the pci transfers will suffer severe timing penalties. i always set 4-way interleaving on any board with via chipset and had only one board that wouldn't work (froze immediately on setting 4-way interleaving..had to go for 2-way instead).
KachiWachi
The New Guy
Posts: 1451
Joined: Fri Mar 29, 2002 10:32 pm
Location: Pennsylvania, USA

Infineon HYS64V16300GU-7.5 is 133 MHz, 16M x 64, 1 bank, 128 MByte SDRAM module (PC133-333-520).

You cannot interleave it.
Post Reply