Re: NOS for DJGPP...

[mikebw@bilow.bilow.uu.ids.net (Mike Bilow)]

Russell Nelson wrote in a message to Mike Bilow:

 RN> No, I'm serious Mike.  I bought the software from Clarkson,
 RN> and they asked me not to call them the Clarkson Packet
 RN> Drivers.  They've already threatened to sue other people
 RN> infringing on the Clarkson trademark, so I'm sure they're
 RN> serious about it.

 RN> They're the Crynwr Packet Driver Collection.

I did not know the history.  I was aware of the change, but I was unaware of
the circumstances.  I'll be careful on this in the future.  I guess I just got
used to saying "Clarkson."  I still say "kilocycles" on occasion, too.

    RN> DJGPP has packet driver support already built into its DOS
    RN> extender. The Crynwr packet drivers work just fine with DJGPP.

 RN>    My understanding was that this imposes constraints of some
 RN> kind on the packet driver, and that it must be written so that
 RN> it will work with Windows, DJGPP, or some DPMI system running
 RN> on top of it.  In other words, my understanding is that some
 RN> packet drivers will work, but others may not. Is this wrong?

 RN> Yes, this is wrong.  Anything that goes through DPMI to
 RN> access packet drivers (which DJGPP will do) will work over
 RN> any packet driver.

That's interesting.  Why do some packet drivers require a special argument to
support running Windows above them?  So the packet driver need not be able to
respond to multiplex broadcasts using Int 2Fh?

-- Mike