Article View: comp.sys.ibm.pc.hardware
Article #19329Re: SCSI Transfer Rates ??
From: Morgan Stair
Date: Fri, 22 Jan 1993 15:11
Date: Fri, 22 Jan 1993 15:11
41 lines
3079 bytes
3079 bytes
>>Didn't he say it was a VL card? That means that the CPU can't do >>anything anyway, since when the disk controller has the local bus the >>CPU can't talk to memory or anything anyway. This is why I don't see >>VL as a real win for anything except perhaps video; it will give you >>great burst mode numbers, but it impacts overall system performance to >>do that. > >That doesn't make any sense. The only way to minimize the hit the CPU >takes when doing DMA is to make the DMA go as fast as possible, which is >what VLBus does. There's no free lunch. :) I'm on the verge of upgrading my 486DX-33 ISA machine to a 33MHz VL-ISA motherboard that can handle a DX2 chip (To head off the tangential questions now, there's a problem with my motherboard, and the dealer's letting me swap up to what I think is a better motherboard for a little more money). Is the VL bus an improvement or not? I thought VL-ISA would be superior to an EISA bus for my needs. I mean, aside from video and hard disks (2 VESA slots) what COMMON applications are there for DMA transfers > 5MB/sec? I suppose loading data into a soundcard's buffer is one, but the industry is just starting to take advantage of the 16bit DMA channels (4-7) on the ISA bus (Gravis Ultrasound Card). I never put together that VL cards lock up the local bus to do their transfers, whereas DMA transfers to/from an ISA card DON'T!! Is this true? I know MS-DOS and MS-Winows (I think) aren't smart enough to be impacted by this, but can anyone comment on whether any PC UNIX (Linux?), Windows-NT, or OS/2 operating systems are intelligent enough to be impacted, and how bad would it really be? Morgan -- ************************************************************************* * W. Morgan Stair Phone: (617) 552-8783 * * Boston College FAX: (617) 552-8778 * * Institute for Space Research Email: morgan@dl5000.bc.edu * * 4 Alfred Circle * * Bedford MA 01730 * *************************************************************************
Technical Details
Message-ID:
<MORGAN.93Jan22101105@dl5000.bc.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!howland.reston.ans.net!spool.mu.edu!uwm.edu!ogicse!hsdndev!news.cs.umb.edu!morganReferences:
<1993Jan17.134956.3123@solace.hsh.se> <1jdb71INN989@mudos.ann-arbor.mi.us> <1993Jan19.211953.26368@mksol.dseg.ti.com> <C16GHF.34r@unix.portal.com>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!howland.reston.ans.net!spool.mu.edu!uwm.edu!ogicse!hsdndev!news.cs.umb.edu!morgan From: morgan@dl5000.bc.edu (Morgan Stair) Newsgroups: comp.sys.ibm.pc.hardware,comp.os.linux,comp.os.ms-windows.programmer.win32,comp.os.os2.programmer Subject: Re: SCSI Transfer Rates ?? Message-ID: <MORGAN.93Jan22101105@dl5000.bc.edu> Date: 22 Jan 93 15:11:05 GMT Article-I.D.: dl5000.MORGAN.93Jan22101105 References: <1993Jan17.134956.3123@solace.hsh.se> <1jdb71INN989@mudos.ann-arbor.mi.us> <1993Jan19.211953.26368@mksol.dseg.ti.com> <C16GHF.34r@unix.portal.com> Sender: news@cs.umb.edu (USENET News System) Distribution: all Organization: Institute for Space Research, Boston College Lines: 41 In-Reply-To: danb@shell.portal.com's message of 21 Jan 93 00: 10:26 GMT Nntp-Posting-Host: dl5000.bc.edu