Detecting OpenMSX or RealMSX?

Sooner the later this is going to be abused for whatever reason and within time false positives will start to show up as the emulation improves. The same happened in android and rooting.

@Manuel:

Here is the code for MSX2 (and I guess msx2 and up, tested on Sony HBF..... and OpenMSX) :

        ORG   &HC000
START:
        DI
        LD    A,245
        OUT   (&H99),A
        LD    A,15+128
        OUT   (&H99),A
        NOP
        NOP
        LD    C,0
        LD    B,255

TEST:   IN    A,(&H99)
        OR    A
        JR    Z,JUMP
        INC   C
JUMP:   DJNZ  TEST
        LD    A,C
        LD    (WTF),A

        XOR   A
        OUT   (&H99),A
        LD    A,15+128
        OUT   (&H99),A

        EI
        LD    A,(WTF)
        OR    A
        RET   NZ
        LD    A,1
        LD    (EMULAT),A
        RET
WTF:      DB    0
EMULAT:  DB    0

build/compile it and run it!
defusr=&hc000:a=usr(0)
?peek (emulat)

result: 1= in emulator! (MSX2 and up version)

Manuel, listen carefully!!! You presume all emulators are perfect, they aren't! Even if you look at the code of MAME and MSX VDP emulation, some is missing.... The goal is to emulate 100%, but you need some stupid fuckers like me, among, to show you out of the box possibilies... Only after reporting, out of the box, or showing posibilities it can be detected, change is happening. I prefer 100% emulation, however if I like to test some undocumented (combinations) or want to have the result from the VDP without the necessary waiting time,... or what happened if I poll a register (or combination) of registers too often (can for example the vdp handle all requests), or polling unspecified vdp registers (47-63), or a command related to vdp 32-46, that is not menioned, or even vdp status registers, (officially, only a few, but 245 shows my proof!!!)

Food for thought!!!

PS: I have some drive to find ways, over and over again to detect emulation...... You should be happy, because I am reporting....

@Wlcracks, did the increase to 32 work?
emulatwl.bin
64 times check refresh for you specially!

Can people confirm that reading non-existing VDP status registers always results in 0? If yes, then we can fix that in openMSX. We somehow thought they returned 0xFF.

@Wlcracks,

you used the new files in the post mentioned above?
I'll get back... (after returned from drawing table)

I created also another file:
version2
10 bload "emulat2.bin"
20 poke &Hc0b6,128
30 defusr=&Hc000:a=usr(0)

@Wlcracks, I will create a file for you and will show all retrieved data on which i base my assumption.

Sorry to jump in without reading all the history, but I think that best way to detect OpenMSX vs RealMSX is to measure the exact CPU speed against some other device... by default OpenMSX runs at 3579545Hz speed and that is theoretically correct answer, but in reality there are tons of things that put it off a big way starting from room temperature. -> If the speed is always same 3579545Hz it must be emulator.

Sure, you can get around the detection very easily, but I'm not really sure, what was the original purpose.