Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!samsung!usc!ucsd!ucbvax!bloom-beacon!eru!luth!sunic!mcsun!hp4nl!nikhefh!t68
From: t68@nikhefh.nikhef.nl (Jos Vermaseren)
Newsgroups: comp.sys.atari.st
Subject: Re: 40-Folder Bug
Summary: It ain't that easy
Message-ID: <805@nikhefh.nikhef.nl>
Date: 14 Mar 90 09:49:12 GMT
References: <2105@onion.reading.ac.uk> <1375@lzsc.ATT.COM> <1482@electro.UUCP> <1404@lzsc.ATT.COM>
Organization: Nikhef-H, Amsterdam (the Netherlands).
Lines: 35

In article <1404@lzsc.ATT.COM>, hcj@lzsc.ATT.COM (HC Johnson) writes:
> In article <1990Mar10.174051.24717@cunixf.cc.columbia.edu>, cmm1@cunixa.cc.columbia.edu (Christopher M Mauritz) writes:
> > Ignac does have a point.  If the entire world knew about the 40 folder
> > bug from day one (ok, maybe day two), why didn't you techno-programmo-
> > studs fix the thing in your subsequent 2 releases of TOS?
> > 
> Come on fellas, it was fixed. You just don't understand.
> 
> Before Tos 1.4, Every FOLDER that was touched counted as 1. If the limit
> was 40, then 40 touches and you started getting corruption and death.
> 
> Tos 1.4 counts 1 for every FOLDER CURRENTLY OPEN.  This is now a depth
> problem.  And you get an error message if the limit is exceeded.  Now only
> the USER knows how deep his folders are.  FOLDRXXX allows the system to 
> be TUNED to the users needs.  After all, there is a price paid for large
> XXX, namely lost RAM for programs.
> 
> Atari wins this one.  You lose.

Not quite. We have still had some problems here with TOS 1.4, necessitating
the use of FOLDRXXX.
FOLDRXXX also concerns the space available for Malloc headers.
Atari has improved that now also but at a very bad cost.
In one program I had Malloced more than 100 memory blocks of 16 Kbytes
(clearly a Mega-ST). Then I released these blocks. After that the computer
was for about half a minute nearly unusable, because the internal
garbage collection in GEMDOS to set all the blocks straight
`take some time'.
So Atari has solved the problem only half. There are fewer bugs but it is
still a very poor solution.

Another question: How comes that on MS-DOS the floppy is so much faster
(and noisier)?

Jos Vermaseren