home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.hardware
- Path: sparky!uunet!mcsun!Germany.EU.net!mpifr-bonn.mpg.de!speckled.mpifr-bonn.mpg.de!mlelstv
- From: mlelstv@speckled.mpifr-bonn.mpg.de (Michael van Elst)
- Subject: Re: My A4000 HD performance
- Message-ID: <1992Nov10.034248.12352@mpifr-bonn.mpg.de>
- Sender: news@mpifr-bonn.mpg.de
- Nntp-Posting-Host: speckled
- Organization: Max-Planck-Institut f"ur Radioastronomie
- References: <1992Nov1.030034.20922@beaver.cs.washington.edu>
- Date: Tue, 10 Nov 1992 03:42:48 GMT
- Lines: 20
-
- In <1992Nov1.030034.20922@beaver.cs.washington.edu> rrogers@cs.washington.edu (Richard Rogers) writes:
- >Something very strange seems to be going on with my 4000's hard drives!
- >DiskSpeed 3.1 reports that my Maxtor 7120 / 2091 is doing 29 K/s
- >(Yes, that's 29000 bytes) reads with all contention turned off. The internal
- >floppy is managing 21 K/s. This seems a bit ridiculous.
-
- No. It is pretty normal. The A2091 cannot DMA to any memory but chip
- memory in your A4000. Thus the filesystem will allocate a single buffer
- (512bytes) in chip memory to do file transfers. This lets your drive
- transfer a single sector per disk revolution. Theoretically one can
- some patched driver from C= that turns DMA completely off and enables
- PIO mode for the A2091. This is of course CPU burning and slow but still
- much faster than this last chance method used by the filesystem.
-
- Regards,
- --
- Michael van Elst
- UUCP: universe!local-cluster!milky-way!sol!earth!uunet!unido!mpirbn!p554mve
- Internet: p554mve@mpirbn.mpifr-bonn.mpg.de
- "A potential Snark may lurk in every tree."
-