disconnect windows drive

robertmo3 wrote on 2025-03-31, 12:45:

disconnect windows drive

I've had enough. I set it to boot off the USB. It reads it for a few seconds and then boots into windows. I enabled csm as well.

if you disconnect ssd with windows 11 it cannot boot into it
if you talk about win98 than hit f8 during boot and choose "boot to command line" from the menu

you can also force always displaying menu at start on your p4

robertmo3 wrote on 2025-03-31, 13:02:

if you disconnect ssd with windows 11 it cannot boot into it
if you talk about win98 than hit f8 during boot and choose "boot to command line" from the menu

My ryzen based machine reads the flash drive for a couple of seconds and then boots into windows. I need to test the flash key on my win98 machine to see if it works there.

It's entirely possible the key's not bootable

robertmo3 wrote on 2025-03-31, 13:04:

you can also force always displaying menu at start on your p4

To get the p4's results I had to boot straight into dos. The benchmark locked up after restarting in ms-dos mode.

don't restart in msdos mode, just choose boot into command line
you can even just have norton commander (nc.exe) in autoexec.bat before "win"

you can also add these lines in msdos.sys:
BootMenu=1
BootMenuDefault=1
BootMenuDelay=3

you will be given this choice
https://www.datarecoverytools.co.uk/wp-conten … e-safe-mode.jpg

robertmo3 wrote on 2025-03-31, 13:13:

you will be given this choice
https://www.datarecoverytools.co.uk/wp-conten … e-safe-mode.jpg

I'm familiar with the f8 menu and back in the 90s made my own.

But it's a distant memory now.

you don't need to remember anything. everything you need i just posted

it is a know problem that bios somehow sometimes chooses the latest os even if you force it to run older one
that is why you have to disconnect ssd with windows 11

robertmo3 wrote on 2025-03-31, 13:25:

it is a know problem that bios somehow sometimes chooses the latest os even if you force it to run older one
that is why you have to disconnect ssd with windows 11

I'm a bit beat at the moment. Will try it first thing tomorrow.

For me a bootable FreeDOS flash drive did it.

The only new issue was rendering VGA modes and the older ones, but obvio that's not the point here

and where are your dhrystone results 😉

Updated the post with the Ryzen FreeDOS score:
Re: RReady (Rendition Verite wrapper) Test Build (Alpha 1)

The problem was that my flash drive registered as a USB HDD and not a flash key (unless this is something different).

Pentium 4 3.4 GHz MS-DOS(Win98)

1Microseconds 1 loop:          1.06
2Dhrystones / second:      939794
3VAX MIPS rating:            534.89

Ryzen 2700X Dosbox (cycles=max 105%)

1Microseconds 1 loop:          4.36
2Dhrystones / second:      229555
3VAX MIPS rating:            130.65

[EDIT] FreeDOS Ryzen 2700X (native)

1Microseconds 1 loop:          0.17
2Dhrystones / second:     6015038
3VAX MIPS rating:           3423.47

[EDIT2] Ryzen QEMU (WHPX)
Ryzen 2700X QEMU (Windows Hypervisor Platform/WHPX)

1Microseconds for one run through Dhrystone: 0.13
2Dhrystones per Second: 7424594
3VAX MIPS rating = 4225.72

[/EDIT2]

accidentally while looking for houston i found this reliant park track:
https://www.icr2.net/forum/showthread.php?tid … ht=reliant+park
which may be nice for testing cars as it looks to be simplest track so far

robertmo3 wrote on 2025-03-31, 15:35:

accidentally while looking for houston i found this reliant park track:
https://www.icr2.net/forum/showthread.php?tid … ht=reliant+park
which may be nice for testing cars as it looks to be simplest track so far

Thanks. Qemu looks nice. Do you know whether it has support for pentium dma, for Rendition transfers.?

i have created a question here: https://github.com/kjliew/qemu-3dfx/issues/152
maybe kjliew won't close it 😉

robertmo3 wrote on 2025-03-31, 15:35:

accidentally while looking for houston i found this reliant park track:
https://www.icr2.net/forum/showthread.php?tid … ht=reliant+park
which may be nice for testing cars as it looks to be simplest track so far

Also, the memory access in dosbox is fairly optimal, I don't know how it can be made any faster. The cpu code is fairly rock solid. There are sleeps in there to control timing. I don't know whether they could be eliminated and whether that'd mess up the sound.