home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!rutgers!cbmvax!daveh
- From: daveh@cbmvax.commodore.com (Dave Haynie)
- Newsgroups: comp.sys.amiga.hardware
- Subject: Re: Problems with the A4000
- Message-ID: <35127@cbmvax.commodore.com>
- Date: 15 Sep 92 16:05:50 GMT
- References: <1992Sep12.083819.17967@news.iastate.edu> <92258.123734LEEK@QUCDN.QueensU.CA>
- Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
- Organization: Commodore, West Chester, PA
- Lines: 47
-
- In article <92258.123734LEEK@QUCDN.QueensU.CA> LEEK@QUCDN.QueensU.CA writes:
-
- >MB seems to overlook what's going on in terms of hardware area as usually
- >. The existing SCSI interface on the A3000 is a SCSI-I 8-bit interface.
- >While it is nice and quick -I saw someone mentioned 3.9 meg/sec DiskSpeed
- >4 on those really FAST drives, it is a bit out of date.
-
- Nobody's supporting 32-bit SCSI on PCs. There are no drives that support
- it on the market, only RAID boxes. No one's going to buy a $25,000 RAID
- box for a $3000 computer.
-
- The performance advantage of SCSI-2 would be fast synchronous mode, which
- can give you 10MB/s transfers, well in excess of anything you'll find in a
- hard drive for many, many years to come. Other advantages might be had in
- dumping the WD33C93A chip for something more modern. While that part can
- manage reasonably fast transfers, as any DiskSpeed report can tell you, it
- isn't all that efficient in protocol transaction speeds compared to more
- modern parts.
-
- >(I said A3000 is SCSI-I because of the passive 220/330 terminating
- >resistor vs SCSI-II's active termination. That doesn't limit the
- >A3000 issuing SCSI-I command, rather the upper speed that data can
- >be transfered reliably and lower power consumption.)
-
- The 220/330 termination is still a part of the SCSI-2 specification, last I
- checked. The 110 ohm regulated termination is shown as an option. This isn't
- really the speed limitation -- fast synchronous dies due to single-ended
- drivers over anything but very short cable lengths, regardless of the
- termination (incidently, the aforementioned active termination scheme is only
- for single-ended cable anyway). Additionally, the Mac-style D25 external
- SCSI connector is too noisy for Fast Synchronous, SCSI-2 does specifiy a 50
- pin mini-D-shell-ish connector for better noise immunity.
-
- >But a multimedia machine using the extra resolution & colours is going
- >to need a lot more hard drive bandwidth for full animation. The
- >A3000 SCSI interface (DMA part is fine) is out of date pretty soon.
-
- To date, we've found one drive, the Maxtor Magic, that might actually run
- sustained transfers a tad faster than the A3000 can handle. And the A3000
- goes faster with that drive than any system Maxtor has yet tested, supposedly,
- as the A3000 is mentioned in their docs. It's going to be a long time before
- the A3000's SCSI is noticably out of date.
-
- --
- Dave Haynie Commodore-Amiga (Cool Advanced High-End Systems You Can't See Yet)
- "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh BIX: hazy
- "We're not hitchhiking anymore, we're riding" -Ren Hoek
-