Recent content by stynx

You can solder a wire to the 486Now oscillator clock pin on the bottom and plug that wire into the clock pin on the inboard386 oscillator socket. I doubt, that this will change the memory throughput (a lot) but it is worth try.

Your version of the 486Now! might not have them. This would mean that you can only tinker with the oscillator and software. Based on the other boards, the module supports 16, 20 and 25 MHz 386 clock rate. 20-33 and 25-33 is very asynchronous and i suspect that there are wait-states involved to...

Regarding the low memory throughput of the Kingston 486Now! module: I have read through some documents and there seem to be switches on some models of the accelerator. These switches configure mostly the wait-states for the accelerator to access the main memory. Maybe there is something to test...

Its nice to see that my initial research which i handed over to Harrison, who finalized the 2Mb version, did come in handy. By putting our heads together we were then able to get the 4Mb running (with a small mod to the 2Mb design). The work was just too much a hassle for me and since Harrison...

I have looked a bit for my dos-config but the XT with the inboard386 is currently out of order because the PSU had a smoking accident when we reached 36°C earlier this year... The boot drive (SCSI2SD) took a hit and i can no longer access the data. You may have to look up AMD register setup...

Regarding low memory transfer results: I suspect that either register settings are not optimal or your cpu timings are off. The Kingston module most likely uses buffers or a small cache to connect the two clock-domains (386-486). This approach may result in missed clock-cycles since they do not...

Normally you would have to leave the first page (64K) on the mainboard for compatibility. (There was DMA mentioned somewhere i think?). The Inboard ram does indeed replace all the system memory once it is initialized. -Jonas

I found a similar card with the same ASIC (D-82 / D-802). https://th99.classic-computing.de/i/I-L/50254.htm The JUKO M16-X has a jumper block (J10) for rom address-selection that my card is missing. The configuration Is a bit different as well. J6 and J7 are swapped and the method of...

The settings are: COM1: 3F8h LPT1: 278h gameport: 201h clock: enabled bios: CA00-CC00 J1&J2 seem to be LPT related J3&J4 are most likely COM related J5, J6, J7, J8? J9: enable floppy controller? The bios has a soft-set for the floppy-type (360k, 1.2m, 720k, 1.44m). it remembers the setting...

There is a VXD for windows 3.11 and Windows 95 to use the memory above 16M for these accelerators. But it may be needed to start windows to get the VXD to work. It may be needed to habe 8M so start Windows 95 to be able to load the VXD and access the additional memory. The DOS-expansion...

I got the Kingston 5x86 upgrade to work with the 2x Adapter on the inboard at 21Mhz bus clock and 8x total multiplier. The kingston module is modded to use Write-Back cache. Doom would not run reliably but i once got 15.13 fps. -Jonas

I have received info that the PK-586X4 adapter for the PC-9801 will work on the Transcomputer 486 module. It will allow 6x or 8x mode based on 3x or 4x multiplier with 5x86 CPUs. With a 20Mhz Inboard bus clock it would allow up to 160Mhz CPU clock. Maybe it is worth a try? The PK-586x4 is...

My results with 50Mhz oscillator on the inboard (=25Mhz bus-speed) and about 8Mhz bus-speed on the 5160 Mainboard. The 4Mb custom expansion was used for these tests. I had to solder the pin B13 on the AMD CPU to VCC to get Write-Back cache instead of Write-Through. It gives a bit more...

Another working prototype (low effort) without resistors and capacitors. 4 Simms directly soldered to the card. PS2 modules would be nice but they will not work since the have bidirectional parity lines. The Inboards needs separate parity input and output lines. It may be possible to take a...