Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!pacbell.com!tandem!netcom!wolf
From: wolf@netcom.COM (Buckskin Tech.)
Newsgroups: comp.windows.ms.programmer
Subject: Re: installation problem
Message-ID: <26180@netcom.COM>
Date: 28 Feb 91 09:16:41 GMT
References: <1386@ssp18.idca.tds.philips.nl>
Organization: Netcom - Online Communication Services  UNIX System {408 241-9760 guest}
Lines: 25

In article <1386@ssp18.idca.tds.philips.nl>, mm@idca.tds.PHILIPS.nl (M. Molenaar) writes:
> 
> Conclusion: the (operating) system does not detect a change of
> a floppy.
> 
> The problem does not arise on all 386-machines and never if the
> tool is started up from the harddisk.
> All floppydrives have 'change line support'. We suspect it is a
> timing problem were possibly the hardware interrupt 0Eh has no
> effect. Is there anyone who can confirm or refute this suspicion?

Well, I can tell you that some early sub-versions of Windows had a BUNCH of
problems with floppies, and they all seemed to be timing-related.  You might
try finding a brand-new Windows to try out.

Also, are you REALLY sure that the change-line is supported?  You can remount
the floppy and FORCE change-line support with a DEVICE=DRIVER {blah blah}
command (the exact format of which escapes me at the moment...help me out,
folks).  I once saw a 386 clone that would NOT recognize changed disks until
this was done.

If none of this helps, you might be able to force a floppy read of the FAT.
I'm reaching here, but doesn't an attempted write force a FAT read? 
 
 - Phil @ Buckskin Technologies