Dilwyn Jones makes some magical things to make me read
}
} In terms of software drivers, what would Ethernet require to connect
} to other computers on a network? Would one driver be it (i.e. largely
} device independen to use QL-speak) or would it need various drivers
} depending on what it was hoping to connect to?
I mainly agree with Marcel: an Ethernet driver will not provide you with
TCP/IP by itself. Only with a frame-device (kind of like a serial port,
in raw mode).
The real question about Ethernet and our computers is : What do you intend
to use ?
If it's a dedicated hardware, that might be fine.
If it's the 'any PC-lan card that can be connected', then your driver
has a big problem: you cannot know every ethernet chip on the market...
Assuming an internal developpement of an hardware extension, you could fix
the chip... until it becomes unavailable (Gold Card anyone ?)
A few 'trap', somehow extended from the serial port one for example, could
be enough.
(You need to be able to deal with something harder: listening on multicast address on
demand)
The nice point of ethernet driver is flow control:
- when going down, if it is not possible to send a frame (or to buffer it), just
drop it!
(and same might be for the opposite direction... of course, the above layers
might not performed at 100% with such loss, but that's their jobs!)
}
} What I'm getting at (not being familar with PC networks) is would you
} need a driver to access Ethernet in general (or another QL on
} ethernet), another driver to access a PC over a network, another
} driver to talk to a printer hub, another to talk to a Mac etc etc. I
} am still smarting a bit from other QL hardware like Aurora which for
} all their promised merits (no blame fired at anyone here) were not
} fully usable until the drivers eventually came along (thinking of
} Aurora 256 colours which were not fully supported until Marcel took
} the law into his own hands).
Consider an Ethernet device with a driver just like the foundation
of a far heavier software. (really really heavier!)
First, you will probably want to have a TCP/IP stack on it (IP is simple,
when compared to TCP (I will leave the RARP/ARP/UDP out of scope, even if they are
nearly mandatory), then, above them you will want to run concurrently your
application protocol (such as NFS, SMB, SMTP(email)...)
But anyway, that's provide only the transport of your data, not their conformance to
what is expected at the other end.
If you have a PCL printer, you will still need to speak to it in PCL...
and for file-access... well NFS might be a little simpler than SMB, but both have big
trouble with disappearing machines, concurrent access and so on.
}
} You can probably see what I'm hinting at, if drivers are an issue, and
} we were asking Quanta to assist with development costs, need this only
} apply to the hardware or do we need to think of the software costs
} too.
The Ethernet driver might be cheap to make, if you do not want to driver
an universal ethernet chip. But the real cost will be on the IP and above
(and it might be very slow to build that)
