home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.hardware
- Path: sparky!uunet!mcsun!fuug!prime!mits!ppihkala
- From: ppihkala@mits.mdata.fi (Petri Pihkala)
- Subject: Re: a1200 ide problems
- Organization: MITS, Helsinki, Finland
- Distribution: comp
- Date: Mon, 28 Dec 1992 04:48:59 GMT
- Message-ID: <1992Dec28.044859.18085@prime.mdata.fi>
- References: <1992Dec26.4509.27108@dosgate> <1992Dec26.191926.16917@crash>
- Sender: usenet@prime.mdata.fi (Usenet poster)
- Nntp-Posting-Host: mits.mdata.fi
- Lines: 20
-
- In article <1992Dec26.191926.16917@crash> uzun@crash.cts.com (Roger Uzun) writes:
- >Many IDE drives have buggy firmware, which does not allow xfrs greater
- >than 65535 bytes, set your maxtransfer back to something less than
- >65535 byytes. Quantum for example, acknowledges this bug in
- >their 120 and 240M IDE drives.
-
- I have had a custom-build AT-IDE controller in my A500 for some years with
- QNTM LP52AT, and I detected this problem already then. One can read 1-255 blocks, but not 256 as
- AT -specs mention as max. I received new EPROM that works ok, ie allows 1-256
- blocks to be read at one command. If Amiga software asks for more than 256
- blocks (128kBytes), then IDE-driver asks repeating for 256 blocks until done.
- The reason for this bug is in AT-BIOS that asks for max 64kB (128 blocks) at a
- time due to segmented memoryhandling. It's pity they have not corrected this
- problem in 120/240 which was reported to them as bug in PL52/105 code. This makes GVP
- unusable, as it asks for 256 blocks, unless of course application/DOS
- (maxtransfer) asks for less.
-
- --
- ppihkala@mits.mdata.fi
-
-