Thanks Wlcracks,
A quick fix is:
quickfix
if you can do me a favour again,
Could you run also the following file?
version next
It will not detect MSX1, however, after running could you
bsave "msx1-v2r.bin",&Hc000,&Hc1e8
and send to me? I added an extra command for timing, and need timing results for MSX1..
To add to nyrriki's comment: could team openmsx not add some variance to the clockspeed? That changes so once in a while?
Are you really sure with the quickfix version?? check if 'No' is before Z80.!!
The only thing I modified is, when results are like you send me, an addition to the check, as you provided.
For me it is almost impossible it does not work...... the results you provided are almost 'too stable'.
My REAL MSX and OpenMSX show fluctuations, your results don't...
Even running an emulator (VM) on a z80 is possible and could influence the results...
get back to you!
To add to nyrriki's comment: could team openmsx not add some variance to the clockspeed? That changes so once in a while?
See Wouter's comment at the bottom of https://www.msx.org/forum/msx-talk/openmsx/detecting-openmsx... I think.
I remember again. Sorry for bringing that up again.
I think I have found the solution to detect any emulator. But, have to admit, false positives and false negatives could occur,
MSX1, is tricky, however, not impossible, MSX2 there for I am using 2 types of MSX2(an up) tricks. MSX2 and up must work 100$, MSX1, I git it a 70/30 % to work, If the MSX/2/2+/up has exactly the same clock, if would fail, however I have seen 0 emulation with the exact same clock. Will check on 2 more machines and when I am satiated... publish...
At latest few days before MSX Nijmegen,......
New version, MSX2 is working, MSX1 is more difficult I noticed...
Maybe this version works:
bload "xxxxxx",r
@Wlcracks , try this version (MSX1 and up)
New version...
This version works only on MSX2 and up:
V9938 and up
peek &hc251 -> 1 and up = no emulator
peek &hc250 -> 0 -> v9918/v9928/v9929 1 -> v9938+
I have come to senses
It is always is a vicious circle, however there is always, and really always a way to detect..
IFF faults can be overcome by emulating, I R eally am convinced at the end emulation can be even better than the original, as it was intended.