Adrian D. Ives wrote, On 29/10/09 08:59:
Tony,
I'm able to do the hardware side, but my area is really in the
software. If you were able to knock up some kind of interface between
the superHermes (which I have) and the USBWiz I'm prepared to put in
the time to look at the software.
As I'd do this work on an Aurora QL, the ideal "prototype" for me
would be something I could just plug into one of the 9 pin D
connectors on the superHermes-enhanced standard serial ports, but if
it needs to be connected straight to the SH inside the case that's no
problem. It doesn't *sound* like a huge job but, as we know, there's
always something... :(
Do you really have a 9D lead connected to your sH?
I thought I made only 25D.
It will be great to see sH serial working at its full speed.
I am a mite rusty, but I think that the highest ser3 baud rate is
actually 230400. USBwiz can go much higher than that I think.
Input throughput is controlled by how fast the QL can process the data.
57.6K seems the maximum to RAM (with SGC).
Mind you that was with qtpi, so maybe higher speeds are possible.
For me, the first step in the software would be to knock up a simple
"Development Workbench" application that would send commands to the
USBWiz and display the responses. From there, I'd need to develop some
kind of QDOS API. Then, if that all worked, my next step would be to
modify the QUBIDE source to invoke the USBWiz API.
It all sounds so easy when put like that, but I haven't even looked at
the QUBIDE sources yet to see the extent of the task.
Phil Borman should be able to help here. Quanta will have contact
details. He was incredibly helpful (twice) when I had dead hard disks.
We recovered the file system manually.
I'm also
starting to think of things like getting the drive characteristics and
whether the USBWiz has direct block/sector access or is just a high
level file-oriented interface.
High level I am sure - they were talking about FAT 8.3 which suggests that.
Mind you maybe it is hackable (8-)#
If the latter, it might be possible to
use a virtual filing system in a container file. Anyway, I'll start
looking at the QUBIDE stuff over the weekend.
So, if you're interested in collaborating on a prototype, I'm up for
it. My idea would be to do this with as much reuse and as little
bespoke hardware development as possible. That probably makes
superHermes a prerequisite, as developing a ROM port card, or
something for the expansion bus would likely be prohibitively
expensive.
Yes sH is essential. Imagine having to work at 4800bps (std QL) or
roughly 14000+ (Hermes).
I bought the USBwiz ages ago, announced it here, but had no response.
It is now certainly worth getting a hardware mockup.
Come to think of it, as the USBwiz interface is TTL, initially it could
be all done using the TTL serial from sH. I will see if I have a sH I
can modify to send TTL on its ser3..
It is devious for sH to create RS232, and for usbWIZ box to convert back
to TTL. This is a bit like using a serial mouse which has to do
precisely that.
I see sH has the nominal +-10V and GND on its low speed serial ports,
and 5V/GND on its spare port. That would be a great way of getting
power to the USBwiz in the final version.
.... or I might make up a vero circuit and lead from the start and use
RS232.
We ought to go private on this now to avoid boring everyone (8-)#
I am very busy right now (Lucerne for the next three days and bathroom
building in London).
Realistically it will be a few weeks for a prototype.
Tony
--
QBBS (QL fido BBS 2:257/67) +44(0)1442-828255
[email protected] http://firshman.co.uk
Voice: +44(0)1442-828254 Fax: +44(0)1442-828255 Skype: tonyfirshman
TF Services, 29 Longfield Road, TRING, Herts, HP23 4DG
_______________________________________________
QL-Users Mailing List
http://www.q-v-d.demon.co.uk/smsqe.htm
