Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!pasteur!cory.Berkeley.EDU!navas From: navas@cory.Berkeley.EDU (David C. Navas) Newsgroups: comp.sys.amiga.hardware Subject: Re: Bridgeboards taking 2M of autoconfig space Keywords: why? Message-ID: <22410@pasteur.Berkeley.EDU> Date: 25 Feb 90 18:45:50 GMT References: <1827@esunix.UUCP> Sender: news@pasteur.Berkeley.EDU Reply-To: navas@cory.Berkeley.EDU.UUCP (David C. Navas) Organization: University of California, Berkeley Lines: 35 In article <1827@esunix.UUCP> blgardne@esunix.UUCP (Blaine Gardner) writes: >I've heard that they take 2M out of the autoconfig space. > >Is this true? I have heard similarly. >Why? >Isn't the shared RAM only 128K? And doesn't the autoconfig protocol >provide for a minimum size of 64K? Unless there is something I'm missing >(quite likely), the Bridgeboards should only take 128K, not 2M. There are something like four modes of addressing the BridgeBoard. Remember that Intel structure ints are "backwords", that IBM bitmaps are interleaved, etc. The different modes of addressing are accomplished via using additional address lines, so you need at LEAST four times the autoconfig space. >Does the Bridgeboard actually USE 2M of space (for a 640K computer?), or >was it just easier to build it this way? My guess is that it was easier to build it this way... I would have rather had some global register settable to one of the four transfer modes -- but there is an efficiency cost/speed problem, as well as a re-entrant question. [Which might have been avoided with tc_Switch/tc_Launch, but oh well...] >-- >Blaine Gardner @ Evans & Sutherland 580 Arapeen Drive, SLC, Utah 84108 >Here: There: (My Amiga running uucp) >blgardne@esunix.UUCP blaine@worsel.UUCP >{decwrl, utah-cs}!esunix!blgardne utah-cs!caeco!i-core!worsel!blaine David Navas navas@cory.berkeley.edu "Think you can, think you can't -- either way it's true." Henry Ford