bb_netbsd problem

Hi folks!

I'm a NetBSD Newby, although with experience in var. Linux-distributions and 
(of course) RiscOS machines. I was wise enough to browse to this mailing list 
and the arm32 one before installing. So I got the newest bootloader and the 
machine specific install install / partitioning instructions from the ftp 
given.

The hardware I'm trying to get NetBSD 1.5.2 running is an RiscPC 600, 64+1 MB 
Ram, StrongArm, 10GB harddisc and CDrom via the internal IDE Port, RiscOS 
4.28.

My 10 gig HD is splitted in 6 GB RiscOS and the rest for BSD, max. cyl count 
is 19590, where the RiscOS partition ranges from 0-13520. So I thougt this 
would be the right number to state in bb_netbsd. I suppose something is 
written on HD, as the LED flashes (a second time after backing up the old 
sector). When I tried to boot NetBSD the first time with the installation 
kernel, the (well known) error message appears:

8

bad disklabel

If I am entering the cyl. number calculated by bb_netbsd which is -3123 
(minus ??), though the size of the partition is calculated correctly! I get 
the folling using the bootloader:

Disc error 16 at :4/77FEE620

and if I start the loader second time:

abort on data transfer at 8039CE4AC, and then "File Core in use" if I try to 
access to any disk.

The only thing I changed in FastBoot was following:
adjusting the path for Monitor-Defs
Set NetBSD$Kernel ADFS::4.$.NetBSD.RPC-Inst (where this is the location of 
the unzipped installation kernel)
Run <BtNetBSD$Dir>.native.moutufs ADFS 4 (I left unchanged as this sounded ok 
to me)

My actions were in oder of appearance:
  using bb_netbsd
  edit FastBoot
  run the Bootloader
  ... well that was all

Ahm, so what may be the source for my problem and how could I get rid of it?
 
And in future realeses a less cryptic installation procedure so far would 
sure be appreciated by many possibly NetBSD-interested people.

regards,

Erhard Furtner