In message <
ant14130...@client.cjemicros.co.uk>
<snip>
> >
> > I was only half paying attention at the time, but there was something
> > recently about recent versions of HForm using fragment ID sizes that
> > old versions of RO (e.g. in the Iyonix) can't cope with.
>
> Druck mentioned it as possible problem, Sprow pointed out the latest copes even with RISC OS 3.1 which we have been able to confirm is o.k.
>
> The thread is here:
https://www.riscosopen.org/forum/forums/5/topics/15945
>
> > I apologise if this turns out not to be relevant.
>
> IDE to SATA interfaces seem very varied and we've seen them seem to work but
> then report a drive as unformatted that it had read o.k. before!
>
>
> Chris Evans
>
I haven't seen any unreliability in re-reading yet.
Having a reason to, (To let the IDE/SATA adapter have it's own IDE
channel) I found that the sony CDROM would not write a CD as slave.
I configured it back to master, and my boot IDE drive to slave and
they co-exist that way OK. Also the ide drive stays at 4.
Now the boot drive is set to slave, I tried Alan's suggestion of
trying the IDE/SATA as master on the same IDE channel.
The result was that the SATA drive was found but the IDE (boot) drive
was lost, regardless wether I configured drive to 5 or not.
It looks like these IDE/SATA adapter(s?) need their own IDE channel, but
I haven't tried other HDD's, multiple adapters, or other types.
Ronald May