home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!sgiblab!zaphod.mps.ohio-state.edu!cs.utexas.edu!uwm.edu!biosci!agate!doc.ic.ac.uk!uknet!yorkohm!minster!mjl-b
- From: mjl-b@minster.york.ac.uk
- Newsgroups: comp.sys.atari.st.tech
- Subject: Re: Falcon BUS..
- Message-ID: <721060362.188@minster.york.ac.uk>
- Date: 6 Nov 92 14:32:43 GMT
- References: <mpnolan.720881057@unix1.tcd.ie>
- Reply-To: mjl-b@minster.york.ac.uk (Mathew Lodge)
- Organization: Department of Computer Science, University of York, England
- Lines: 37
-
- In article <mpnolan.720881057@unix1.tcd.ie> mpnolan@unix1.tcd.ie (Mark Patrick Nolan) writes:
- >I believe that Atari have at last got their act together...
- >
- >Q: As the Falcon has (from what I've heard) one bus (32),
- >how will the custom chips (DSP,DMA Sound,Blitter,H-Scrolling,..)
- >interact, and will the system lose preformance as a result?
-
- It has ever been the case that the CPU must stall when DMA is taking place.
- The DMA chip grabs the bus for a short time and does a block transfer (if I
- remember correctly, the block size is 16 words for the ST's DMA transfers --
- it may have changed for the TT / Falcon). During this time, the 68k
- has to wait until the bus is released.
-
- The video system in the ST was designed so that processor access and video
- shifter access are interleaved. On the original ST, the memory subsytem runs
- at 16MHz, so that processor (8MHz) and video access can be interleaved. It's
- a bit more complicated than this (we had the discussion some months ago, and
- I forget the exact details -- sorry). Horizontal scrolling is simply some
- extra logic in the shifter and has a minimal performance penalty (you need
- one extra memory access at the end of a line).
-
- >Surely the DSP has it's own bus, as it does have 56k of it's own
- >memory!?
-
- The DSP is a member of the Motorola 56k series -- it has 96K of static RAM.
- The DSP has its own data and address buses, independent of the 68k address
- and data buses. I don't know how the DSP and 68K communicate, but I would
- guess that there's some sort of shared memory scheme (but that is just *my
- guess*)
-
- > Thanks, Mark Nolan
-
- Mat
-
- | Mathew Lodge | "I don't care how many times they go |
- | mjl-b@minster.york.ac.uk | up-tiddly-up-up. They're still gits." |
- | Langwith College, Uni of York, UK | -- Blackadder Goes Forth |
-