VOGONS


First post, by Indrid Cold

User metadata
Rank Member
Rank
Member

I think it's an interesting configuration, especially regarding to the cause of the problems I encountered while reconfigure and cleaning up everything, as well as guessing correct identification of the CPU.

The case is a nice full tower, vintage but also very robust - the thing that immediately (as I have already shown in another section of the forum) intrigued me was the presence of a tape drive for backup, but not the tapes that I've seen before, but apparently common cassette tapes, with its proprietary controller board. The motherboard is a PC Chips M919, with PCI bus, ISA, VLB, 16MB SIMM RAM, and processor with active heatsink, which immediately made me think of a fairly-high-clocked 486. The CPU then has proved to be an AMD AM486DX4-100 v3.3 - a CPU with good performance and very competitive at the times, able to overcome in the performance the very first Pentium CPUs - in contrast to common DX4 486, the multiplier for this CPU is x3, so with FSB @33MHz, although versions released after had FSB jumped to 40MHZ. The strange thing is that POST signaled the CPU as DX2-100 o_o

After careful investigation and monitoring of the jumpers about configuration of CPU/FSB clock, I found that who has mounted originally this computer had incorrectly configured CPU with FSB @50MHZ, setting x2 as multiplier -___-

So, for all the time of its use (accounting backup - installed operating system: ScoUnix), this PC was used with CPU set with higher FSB (50 instead of 33 MHz) to that for which it was created.

Other present drives are a 5 1/2 floppy drive, 1.44MB floppy, 1.2GB Seagate hard disk (replaced with IDE2CF adapter and 2GB CF with MS-DOS 6.22) to which I later added a Philips DVD, with yellow bezel which seems to be comfortable with the rest. The video card that was present when I was given the computer was REALTEK Quadtel 87-90 RTG 3105 IEH, that I replaced with a common but compatible S3 Virge DX. The audio board was not present, so I chose one of my SoundBlaster 16 models, the CT2890.

IMG_3996.jpgIMG_3995.jpgIMG_3976.jpgIMG_3978.jpgIMG_3981.jpgIMG_3982.jpgIMG_3983.jpgIMG_3984.jpg
IMG_3985.jpgIMG_3989.jpgIMG_3992.jpgIMG_3993.jpgIMG_3997.jpg

Last edited by Indrid Cold on 2016-03-15, 10:07. Edited 1 time in total.

Reply 1 of 17, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++

Does the PC Chips board have real cache installed?

All hail the Great Capacitor Brand Finder

Reply 2 of 17, by Indrid Cold

User metadata
Rank Member
Rank
Member

Yes, I found a bank already installed, POST reports 256kb

Reply 3 of 17, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++
Indrid Cold wrote:

Yes, I found a bank already installed, POST reports 256kb

It's PC Chips, so be sure to double check with independent tools. They're one step away from the guy in the white van in the corner parking lot. 😐

All hail the Great Capacitor Brand Finder

Reply 5 of 17, by Indrid Cold

User metadata
Rank Member
Rank
Member
gdjacobs wrote:
Indrid Cold wrote:

Yes, I found a bank already installed, POST reports 256kb

It's PC Chips, so be sure to double check with independent tools. They're one step away from the guy in the white van in the corner parking lot. 😐

???

Sure, I'll do more testing as soon as possible

Reply 7 of 17, by mrau

User metadata
Rank Oldbie
Rank
Oldbie
Indrid Cold wrote:

What kind of module is it?

its cache that usually isn't fake

Reply 8 of 17, by Skyscraper

User metadata
Rank l33t
Rank
l33t

The utility cachechk will show if there is any L2 cache in the system.

The cache sticks are normally real but I have heard about fake ones, the soldered cache on many PC-Chips 486 boards are often fake.

New PC: i9 12900K @5GHz all cores @1.2v. MSI PRO Z690-A. 32GB DDR4 3600 CL14. 3070Ti.
Old PC: Dual Xeon X5690@4.6GHz, EVGA SR-2, 48GB DDR3R@2000MHz, Intel X25-M. GTX 980ti.
Older PC: K6-3+ 400@600MHz, PC-Chips M577, 256MB SDRAM, AWE64, Voodoo Banshee.

Reply 10 of 17, by Skyscraper

User metadata
Rank l33t
Rank
l33t
gdjacobs wrote:
Indrid Cold wrote:

???

Sure, I'll do more testing as soon as possible

https://en.wikipedia.org/wiki/White_van_speaker_scam
https://www.youtube.com/watch?v=i3B_KKyntQE

PC Chips and the "white van scams" have little in common. PC Chips motherborads with fake cache might be slower than boards with real cache but we are talking about ~10% in performance difference not people selling useless junk as high end audio gear.

There is a reason PC Chips boards are so common, there were cheap. 😀 The M919 is considered to be one of the best 486 overclocking motherboards ever made and with a real cache stick it's also pretty fast.

New PC: i9 12900K @5GHz all cores @1.2v. MSI PRO Z690-A. 32GB DDR4 3600 CL14. 3070Ti.
Old PC: Dual Xeon X5690@4.6GHz, EVGA SR-2, 48GB DDR3R@2000MHz, Intel X25-M. GTX 980ti.
Older PC: K6-3+ 400@600MHz, PC-Chips M577, 256MB SDRAM, AWE64, Voodoo Banshee.

Reply 11 of 17, by Indrid Cold

User metadata
Rank Member
Rank
Member

Thanks for your great information! I hope it is real cache then...

Reply 12 of 17, by leileilol

User metadata
Rank l33t++
Rank
l33t++

You can usually make up for it by putting an AM5x86-P75 in and clocking that 40x4 and setting your memory wait states to 0 or 1 (which can cost you the ability to soft reset, so don't set that until your drivers are all installed 😀 ). At that point it's a 59.79 in speedsys, and Quake is definitely in the realm of being playable to the end.

My personal bad experience with M919s had to do with faulty serial support so my mouse didn't work half the time, and I think that trumps the cache issue.

Last edited by leileilol on 2016-03-15, 17:09. Edited 1 time in total.

apsosig.png
long live PCem

Reply 13 of 17, by mrau

User metadata
Rank Oldbie
Rank
Oldbie
leileilol wrote:

My personal bad experience with M919s had to do with faulty serial support so my mouse didn't work half the time, and I think that trumps the cache issue.

you could've just went on and installed a usb controller card ;p

Reply 14 of 17, by Skyscraper

User metadata
Rank l33t
Rank
l33t
mrau wrote:
leileilol wrote:

My personal bad experience with M919s had to do with faulty serial support so my mouse didn't work half the time, and I think that trumps the cache issue.

you could've just went on and installed a usb controller card ;p

A serial port card would have been a better choice as I think he wanted to use the mouse in DOS. 😉

I use a serial port card all the time when testing/benching late Socket-3 and Socket-5/7 motherboards as I cant be bothered with finding a working serial port bracket, there are at least two different pinouts.

edit spelling

Last edited by Skyscraper on 2016-03-15, 17:05. Edited 1 time in total.

New PC: i9 12900K @5GHz all cores @1.2v. MSI PRO Z690-A. 32GB DDR4 3600 CL14. 3070Ti.
Old PC: Dual Xeon X5690@4.6GHz, EVGA SR-2, 48GB DDR3R@2000MHz, Intel X25-M. GTX 980ti.
Older PC: K6-3+ 400@600MHz, PC-Chips M577, 256MB SDRAM, AWE64, Voodoo Banshee.

Reply 15 of 17, by mrau

User metadata
Rank Oldbie
Rank
Oldbie

is a serial mouse not a bigger burden on the system than a usb or ps/2 one?

Reply 16 of 17, by leileilol

User metadata
Rank l33t++
Rank
l33t++
mrau wrote:

you could've just went on and installed a usb controller card ;p

easier said than done, and it didn't work.

(If anyone can get a VIA USB card going on a M919, with both usb storage and usb mice, i'd like to know what you did because i've lost my patience years ago)

Last edited by leileilol on 2016-03-15, 17:08. Edited 3 times in total.

apsosig.png
long live PCem

Reply 17 of 17, by Indrid Cold

User metadata
Rank Member
Rank
Member
mrau wrote:
leileilol wrote:

My personal bad experience with M919s had to do with faulty serial support so my mouse didn't work half the time, and I think that trumps the cache issue.

you could've just went on and installed a usb controller card ;p

USB mouse working under MS-DOS? I don't think so... FreeDOS supports them instead, by emulation. I'm currently using a serial one, Logitech model