> I like your ideas about Fenix AES and as Christer said, it's quite close to what
> oAESis is
>
> Do you have a similar document dealing with the VDI ?
Before the fVDI implementation was begun, a design/implementation plan
was written. You can find that via a hyper link on the 'technical page',
which in turn has a link near the bottom of my fVDI page (see .sig).
I certainly didn't end up following that plan 100%. Some things in there
are not done yet (for example anything to do with MiNT/Fenix), and some
never will be (such as LineA emulation using the VDI).
fVDI really does work well with just about everything now, though. The
Eclipse/RageII driver is the best, but recently the monochrome one has
become nearly as good. The almost 100% C FalconTC example driver should
also be quite good.
> What was the plan for the FenixOS as far as the VDI is concerned ?
> Oh, fVDI I guess :)
Yes.
Most (all but a very few) VDI functions in fVDI already have versions that
are supposed to be called in the usual Fenix library fashion. Indeed, for
functions that are not too time critical, the Trap #2 versions call the
library ones internally.
> If not:
> Johan, what did you write fVDI from ? It there any document you would like to
> share ?
The documents mentioned above are the only fVDI specific ones available
besides the sources themselves and some recent device driver development
docs.
Aside from that, I've of course made use of all VDI documentation I've
been able to find. Unfortunately, that is not enough to make all programs
behave correctly, so a _very_ significant amount of work has gone into
tracing through various misbehaving applications.
--
Chalmers University | Why are these | e-mail: rand@cd.chalmers.se
of Technology | .signatures | johank@omicron.se
| so hard to do | WWW: rand.thn.htu.se
Gothenburg, Sweden | well? | (fVDI, MGIFv5, QLem)