Newsgroups: alt.cd-rom,alt.sys.pc-clone.gateway2000,comp.sys.ibm.pc.hardware.cd-rom,comp.sys.ibm.pc.soundcard Path: cdrom.com!barrnet.net!autodesk.com!autodesk.com!apple.com!agate!howland.reston.ans.net!europa.eng.gtefsd.com!MathWorks.Com!news.kei.com!yeshua.marcam.com!charnel!olivea!news.bu.edu!att-in!cbnewse!ers From: ers@cbnewse.cb.att.com (eric.r.sterbenz) Subject: Re: HELP! Organization: AT&T Date: Fri, 11 Mar 1994 20:02:05 GMT Message-ID: Lines: 43 Xref: cdrom.com alt.cd-rom:1583 alt.sys.pc-clone.gateway2000:1441 comp.sys.ibm.pc.hardware.cd-rom:1721 comp.sys.ibm.pc.soundcard:8425 Folks, I posted an article earlier this week about a problem I was having with my Mitsumi FX001D and my Orchid GW32 card controlling it. In a nutshell the problem was that it had worked fine for several weeks and suddenly started not detecting the CD-ROM drive on power up. Well thanks to some helpful people at Micro Center, they jogged my memory and we found the problem. The basic problem was that whenever you change the GW32 configuration from GMIDI to MT32 or back, the configuration program re-edits your CONFIG.SYS file and puts the SWCD.SYS driver at the end of the file. In this case it was placing it after the Mitsumi driver's definitiion)! I remember noticing this before, but it was before I had a CD-ROM drive and thus didn't matter. Anyway, now I know (I won't forget again), and am passing this on to help keep anyone else from falling into this trap. Item 2: Thanks to a coworker I have also been able to benchmark the Mitsumi FX001D using the Mitsumi provided controller with and without DMA access. (There had been some postings in various newsgroups about the DMA channel mismatch between the Orchid cards and the Mitsumi driver). I am happy to report that Orchid was correct about DMA access (this should not be a suprise based on some recent postings about DMA). Anyway the performance on a 486 class machine using the DMA driver in interrupt mode is significantly better than using DMA mode. My coworker has a GW2K 4DX2-66V (he has a lot of stuff running). With DMA mode the using MPC Wizard 2.0 the best he could get at 100% cpu was about 305 KB/s. Using interrupt mode it jumped to about 365 KB/s. This is much more consistant with my results (GW2K 4DX-33V using the GW32 as a controller) of a consistant average around 380 KB/s. By the way I believe GW2K distributes their Mitsumi drives configured for DMA with the minimal amount of buffers. I hope that anyone trying to get the optimal performance out their CD-ROM finds this information usefule. Thanks, Eric ers@cbnmva.att.com