(top posting to make it shorter) I do need as much conventional memory as I can get, and even the it is a tight fit...
So everyone belives that JEMM386 X=TEST NOEMS is safer? if so, I will change... Thanks, Alain Em 17-05-2011 16:35, Rugxulo escreveu: > Hi, > > On Tue, May 17, 2011 at 10:45 AM, Alain Mouette<ala...@pobox.com> wrote: >> >> I am encountering a strange problem with UIDE. It happens with some >> machines but not all, one of the failing machine is nearby so I can make >> more tests easely... >> >> I am using UIDE and XMGR from 10-May-11, I made my config from your >> example (whole file attached), jemm386 is 572 dated 13mar2008 (complete >> config at the end): >> device=a:\dos\xmgr.sys /B >> device=a:\dos\jemm386.exe I=TEST NOEMS >> device=a:\dos\xmgr.sys >> device=a:\dos\uide.sys /D:CDROM >> >> It works ok booting from floppy and from hard-disk, but it does not work >> when booting from the CDROM. It stops with this screen and the HD led on: >> <http://suporte.cosmodata.com.br/downloads/moni/tmp/trava01.jpg> > > Definitely try latest JEMM386 5.74, but honestly, I think the problem > is two-fold (blindly guessing, so I could be wrong): > > 1). NOEMS is rarely used, and a lot of things don't handle it well. I > think this means VCPI is still available as well as UMBs but not EMS. > In other words, I guess? you're trying to save precious conventional > memory by using JEMM386 while only using XMS (no EMS). Honestly, > unless you really *need* that much conventional memory free (unlikely, > as FreeDOS is pretty efficient), I wouldn't do this. Just load a XMS > driver only (and not via "NOEMS"). > > 2). Actually, I've noticed JEMM386 will hang if you aren't careful, > esp. on newer machines due to memory conflicts. In particular, your > "I=TEST" parameter is pretty dangerous. I think typically it works > better if you do "X=TEST I=TEST" (or maybe even not at all). Lemme > search my old mails for a sec ... > > Apparently, BioMenace (DOS game, now freeware, see 3D Realms site) > needs both EMS (hence EMM386) and XMS and (at least for my other, > older P4) "X=TEST I=TEST NORAM". Normal JEMM386 loading just hung, and > anything else didn't give it enough EMS to load to E1L3 without > crashing the program due to insufficient memory. I blame ACPI, natch. > > P.S. Maybe I'm way off, and I know you're talking about a vaguely > different problem, but just keep in mind, EMM386 can be tricky on > newer hardware with weird memory configs (ACPI). That's the whole > point I was trying to make, however futile it may seem. (You can also > run "JEMM386 LOAD" at any time but that doesn't give you UMBs. JEMMEX > doesn't let you run as pure XMS only.) > > ------------------------------------------------------------------------------ > Achieve unprecedented app performance and reliability > What every C/C++ and Fortran developer should know. > Learn how Intel has extended the reach of its next-generation tools > to help boost performance applications - inlcuding clusters. > http://p.sf.net/sfu/intel-dev2devmay > _______________________________________________ > Freedos-user mailing list > Freedos-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/freedos-user > > ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
