toddpw@tybalt.caltech.edu (Todd P. Whitesel) (09/01/90)
I most humbly apologize, Matt. I always thought that AFE was an Apple project. However, I maintain that AFE Mac->Prodos sucks raw goose eggs. Chan Wilson's program performs the same function much more quickly -- my mentioning the P8 block driver was a misguided attempt to stay technically accurate while describing the program's translation speed. What I really meant to say was that Chan Wilson's program performs the HFS to Prodos transfer at the fastest rate supported by its O/S environment, and that Apple File Exchange currently can't make anything near that claim. Todd Whitesel toddpw @ tybalt.caltech.edu
cwilson@NISC.SRI.COM (Chan Wilson) (09/07/90)
In article <1990Aug31.205240.16853@laguna.ccsf.caltech.edu> toddpw@tybalt.caltech.edu (Todd P. Whitesel) writes: >However, I maintain that AFE Mac->Prodos sucks raw goose eggs. Chan Wilson's >program performs the same function much more quickly -- my mentioning the >P8 block driver was a misguided attempt to stay technically accurate while >describing the program's translation speed. > >What I really meant to say was that Chan Wilson's program performs the HFS to >Prodos transfer at the fastest rate supported by its O/S environment, and that >Apple File Exchange currently can't make anything near that claim. Hmm, just have to step in here and drop my .02 on the floor. :) a2fx achieves the speed it does because it's doing straight block reads off of the disk. Once it's found where the file is, away it goes, only stopping to write out the buffer when it gets full. Pretty simple, so it's pretty fast. The thing that really astounded me was the speed at which it reads the directory... considering all the code I wrote, it's still bloody fast... unless the catalog happens to be scattered all over the disk, like one that someone sent me. I don't know if there's such a thing as READ_BLOCK in the MacOS world. It'll be real interesting to see what the speed difference is when I get around to writing a GS version of a2fx, which will use the fst calls... Speaking of a GS version, what's the interest level on something more akin to AFE (Mac), where you can have different translators and such? Considering that I haven't even written a single line of GS-specific code, it'll take a day or two, but is the interest there? Plug-n-play translators... Unix troff to Appleworks. Microsoft Word to Appleworks. Lotus 123 to Appleworks.... food for thought. Feedback? >Todd Whitesel >toddpw @ tybalt.caltech.edu --Chan ................ Chan Wilson -- cwilson@nisc.sri.com <!> I don't speak for SRI. Janitor/Architect of comp.binaries.apple2 archive on wuarchive.wustl.edu "a2fx it!" ................
russotto@eng.umd.edu (Matthew T. Russotto) (09/07/90)
In article <20804@fs2.NISC.SRI.COM> cwilson@NISC.SRI.COM (Chad Wilson) writes: > >a2fx achieves the speed it does because it's doing straight block >reads off of the disk. Once it's found where the file is, away it >goes, only stopping to write out the buffer when it gets full. >Pretty simple, so it's pretty fast. The thing that really >astounded me was the speed at which it reads the directory... considering >all the code I wrote, it's still bloody fast... unless the catalog >happens to be scattered all over the disk, like one that someone sent >me. > >I don't know if there's such a thing as READ_BLOCK in the MacOS world. >It'll be real interesting to see what the speed difference is when >I get around to writing a GS version of a2fx, which will use the fst >calls... Yes, there is a READ_BLOCK in the MacOS world. And Apple File Exchange MUST use it, as otherwise, there is no way to get the data from this disk (the file system won't handle it, remember?). >Speaking of a GS version, what's the interest level on something >more akin to AFE (Mac), where you can have different translators and >such? Considering that I haven't even written a single line of >GS-specific code, it'll take a day or two, but is the interest there? Yuck-- AFE on the mac side is a kludge. IMO, Apple should have external file systems (equivalent of FSTs), so ANY program can read Prodos and MS-DOS disks (Dayna does this for MS-DOS, no one does it for prodos). Something like AFE should do file format translations only, NOT disk format translations.
joshuat@pro-grouch.cts.com (Joshua M. Thompson) (09/08/90)
In-Reply-To: message from cwilson@NISC.SRI.COM I would have a definite interest in an AFE-type program for the GS, but more for just the OS-transfer capatiblities, not as much for converting WP files and such. One thing that would be nice about it though is programming it with a nice, well-documented interface for adding modules to the program. That way anybody with a new idea could put that idea into a code. UUCP: crash!pro-grouch!joshuat ARPA: crash!pro-grouch!joshuat@nosc.mil INET: joshuat@pro-grouch.cts.com
bchurch@oucsace.cs.OHIOU.EDU (Bob Church) (09/09/90)
In article <1990Sep7.152829.15533@eng.umd.edu>, russotto@eng.umd.edu (Matthew T. Russotto) writes: > > Yuck-- AFE on the mac side is a kludge. IMO, Apple should have external > file systems (equivalent of FSTs), so ANY program can read Prodos and MS-DOS > disks (Dayna does this for MS-DOS, no one does it for prodos). Something > like AFE should do file format translations only, NOT disk format translations. Just my opinion here, but I think Apple has taken the best option. They give you a way to convert from Apple to Mac with the System tools. It may not be the best but it works. More importantly, it encourages individual vendors to create their own xfer programs. Since these people have access to all of their source codes this should cut down on bugs etc. If Apple were to try to provide FSTs for every program a couple of things would happen. Lots of bugs because they are not as familiar with the programs, and new developers would be discouraged from coming up with their own formats. Apples policies have always been ( once again, IMHO ) to provide a minimum software package with the computer and to encourage 3rd party people to develop by providing needed information and keeping software competition (from Apple, that is) to a minimum. When I look at Publish-IT!, Copy II+, Beagle Bros/Timeout, AE, etc. I can't help but think that it's a policy that works. bob church bchurch.oucsace.cs.ohiou.edu