klieb@tekigm2.TEK.COM (Kurt Liebezeit) (03/16/88)
In article <3777@cup.portal.com> Chad_The-Walrus_Netzer@cup.portal.com writes: > NO! NO! NO! NOT the NuBUS!!!!! It is pure crap for a an >highspeed/highpower computer, or even a lower speed/lesser power one! >At least, Apple's implementation of it is... A good BUS is ESSENTIAL to >the throughput of a machine, and the NuBUS is a MEGA blunder in this >respect... It will only slow things down. I recently heard an > [ some interesting stuff about alternative architectures deleted] I am curious why someone would label Apple's implementation of the NuBUS as crap. From what little I have read about the NuBUS it seems to make sense: the assertion/sampling edges would appear to eliminate hold time problems, and I like the way that the bus decides on cycle N who has control during cycle N+1. What are the serious shortcomings of the NuBUS? The only thing that I didn't care for was the fact that the Mac II NuBUS has no DMA; cards are expected to supply their own. Kurt Liebezeit tektronix!tekigm2!klieb
daveh@cbmvax.UUCP (Dave Haynie) (03/25/88)
in article <2669@tekigm2.TEK.COM>, klieb@tekigm2.TEK.COM (Kurt Liebezeit) says: > Keywords: NuBus > I am curious why someone would label Apple's implementation of the NuBUS > as crap. From what little I have read about the NuBUS it seems to make sense: > the assertion/sampling edges would appear to eliminate hold time problems, > and I like the way that the bus decides on cycle N who has control during > cycle N+1. What are the serious shortcomings of the NuBUS? The only thing > that I didn't care for was the fact that the Mac II NuBUS has no DMA; cards > are expected to supply their own. The idea of the way NuBus clocks things is really elegant. The problem is that it's slow; some 3 or so wait states for the 68020 in the Mac II. Last I checked, 3 added wait states means that all NuBus access has you '020 running at half it's potential speed. And the one way around this, the nybble burst mode, isn't supported by Apple. It was nice of Apple to add an interrupt to it, but I still have problems with it. That's not to say Apple won't be successful with it, they didn't do so poorly with the Apple II bus, which was neither fast nor elegant. > Kurt Liebezeit > tektronix!tekigm2!klieb -- Dave Haynie "The B2000 Guy" Commodore-Amiga "The Crew That Never Rests" {ihnp4|uunet|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy "I can't relax, 'cause I'm a Boinger!"