Xref: utzoo comp.sys.ibm.pc:37467 comp.binaries.ibm.pc.d:5031 comp.misc:7266 comp.periphs:2233 Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!agate!e260-1g!c60a-1bd From: c60a-1bd@e260-1g.berkeley.edu (Jeff Davis) Newsgroups: comp.sys.ibm.pc,comp.binaries.ibm.pc.d,comp.misc,comp.periphs Subject: Re: Don't buy from HDI Summary: my experiences with st296N Message-ID: <1989Nov2.175945.20234@agate.berkeley.edu> Date: 2 Nov 89 17:59:45 GMT References: <754@dekalb.UUCP> <4439@cbnewsc.ATT.COM> Sender: c60a-1bd@e260-1g (Jeff Davis) Reply-To: JCDAVIS@LBL.GOV (Jeff Davis) Organization: University of California, Berkeley Lines: 69 In article <4439@cbnewsc.ATT.COM> wolfordj@cbnewsc.ATT.COM (j.w.wolford) writes: >From article <754@dekalb.UUCP>, by greg@dekalb.UUCP (Greg Philmon): >> >> A little note of warning for any in the hard drive market. Several months ago >> I purchased the Seagate ST296N/ST01 SCSI drive kit from Hard Drives >> International, a company I had heard good things about. Their advertisement >> promised "1 to 1" performance. The purchase was made, largely, on this claim >> as I needed a larger, faster hard disk. > >You have to be kidding right.... what do you want from a $49 disk >controller.... yes thats right I can even get a ST01 for $39 if you want... >do you think it has any memory to do caching.... Sigh... And yet a >WD1007 SCSI controller is > $200 and some places are trying to get > $400 >there has to be something diff...... Sneering WITH you not at you... > >The ST01 is a 8 bit controller and you are hooking this up to a 25Mhz >386..... Your $$$ "aint" in the right place..... It don't matter how fast >your CPU is if you can't get the program off of the HD then things are >going to be slow unless you only run that one program all day... and I have >yet to run into someone that runs only one program all day on a 386... > [ some stuff deleted] >> Telling DiskManager to format it 1:1 (as promised in their ads) brought >> terrible results. It seemed that 2:1 was the best it could do. But I >> foolishly refused to believe it, instead wondering what I was doing wrong. My >> old hard drive was already spoken for and I sold it when the 296N arrived. > >Have you tried a interleave test program like HTEST/HFORMAT or SPINWRITE >to check out what the optimum interleave is.... Programs like Htest/Spinrite etc. WILL NOT WORK WITH SCSI! Does anyone know of programs which do this sort of thing for SCSI? I just reformatted my drive with a 2:1 interleave and am now getting ~500K transfer rates as opposed to 320K with 3:1 and about 100K with 1:1. This is on a 20MHz 386 with an st02 controller and the bus at 10MHz. [more stuff deleted] >TAKE THE RMA..... get rid of that ST01 slow controller buy yourself a REAL >controller ESDI or SCSI.... ie WD1007 or a ADAPTEC.... > >> --------------------------------------------------------- >> | Greg Philmon ...gatech!dekalb!greg CIS: 72261,1724 | >> --------------------------------------------------------- > >Jeff Wolford uunet---+ >att!iwsag!jww | >att!iwsag!iwtjw!jww ucbvax--+---- att --+ iwsag!jww > | > decvax--+ After speaking with tech support at seagate (well first I had to figure out how to get a person rather than a tape) they said the best interleave you could hope for is 2:1 although if you are "lucky" 1:1 might be better. I am not entirely sure that the bottleneck is the ST0[12]. I think it may be the smarts built into the drive which prevent you from running at 1:1. I am rather disappointed by the performance of the st296N but considering the price I think I can't complain too much. I am collecting stories about this drive and would appreciate anyone having used it to send me a note with benchmarks and so forth. Has anyone used it with one of the better controllers? Does it help much? Please send me mail and I will post a summary. -- Jeff Davis JCDavis@LBL.GOV or on bitnet JCDavis@LBL