Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!kuhub.cc.ukans.edu!maverick.ksu.ksu.edu!unlinfo.unl.edu!fergvax!231b3678
From: 231b3678@fergvax.unl.edu (Phil Dietz)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Xetec Fastrack
Message-ID: <1991Apr18.080111.20882@unlinfo.unl.edu>
Date: 18 Apr 91 08:01:11 GMT
References: <22887@know.pws.bull.com> <71795@microsoft.UUCP>
Sender: news@unlinfo.unl.edu
Organization: Comp Sci and Engr, Univ. of Nebr.
Lines: 52
Nntp-Posting-Host: fergvax.unl.edu

In article <71795@microsoft.UUCP> iank@microsoft.UUCP (Ian KENNEDY) writes:
>In article <22887@know.pws.bull.com> C506634@UMCVMB.MISSOURI.EDU (Eric Edwards) writes:
>>Recently there was a post from somone who had a Xetec Fasttrack for a couple
>>of years.  He said that the driver disabled interupts for an unreasonably long
>>time.  Does anyone know if the current driver shipped still does this?  I do
>>an awful lot of up/downloading........
>>
>>Also, the AmigaWhirld article states that the Xetec is a DMA controler.
>>Is this really a new controller or did amiga or did AmigaWhirled screw up
>>again?
>The XETEC controler uses what the manufacturer calls DMAx, which they claim
>produces DMA like speeds without the DMA bus contention when working
>with hi-res screens. However, I believe that the controler is really NON-DMA
>and uses really efficient driver software.
>
>I own a XETEC FastTrack controller driving a 48 MB 28ms seagate drive.
>Even with the pathetic Seagate I get throughput of about 400 Kps. The Claims
>of 750 Kps (Kilobytes per second) can only be realized with a really fast
>hard disk. I have also added the 4MB fast ram daughter board for $360 from
>the manufacturer with 80ns simms.
>
>I'm running it on a circa 1987 rev 5 A500 and it works great. I've never 
>experienced any waiting while the hard drive is accessed. I've even been
>able to work on one partition while de-fragmenting/optimizing another.
>
>Hope this helps...
>
>Ian.
>>
>>Eric Edwards: c506634 @    "The 3090.  Proof that by applying state of the
>>Inet: umcvmb.missouri.edu   art technology to an obsolete architecture,
>>Bitnet: umcvmb.bitnet       one can achieve mediocre performance."

I'm even worse.  I bought the Xetec and put in a Conner cp-340.
When I bought the Conner, we tested it out on the sellers Hard Frame.
It got 600+k/sec (with no HD noise!)
 
When I got it home though, I was dissappointed to find that the max 
the Xetec can do for my A500 is a VERY pathetic 150k/sec!
 
I tried everything....but no go.  The Conner is weird.  It works with the
the Miniscribe asyncronous driver AND the Quantum syncronous driver.  I
bet that the Conner needs it's own syncronious driver to perform it's max....
 
oh well
Phil Dietz


-- Subliminal Signature
                                                            Phil Dietz 
   A      M      I      G      A              231b3678@fergvax.unl.edu       
                                                     Univ. of Nebraska