atc@waikato.ac.nz (01/13/91)
In article <2509@sparko.gwu.edu>, kilian@seas.gwu.edu (Jens Kilian) writes: > In Article 1068 of comp.sys.atari.st.tech, kenc@vaxb.acs.unt.edu writes: > >> Chris, >> This is a no go, because of the way that the CPU and the MMU swap cycles. >> >From what I understand, the system clock ALREADY runs at 16 Mhz, but it gives >> half to the 68000, and half to the MMU. (I may have the wrong chip...it may be >> the glue...) >> Anyway, since THAT chip, and all the support chips are not rated at anything >> like 16 Mhz, and all the timing is sensitive, you try to double the speed of >> the system clock, and you have an expensive paperweight. >> I think the speed up kits available use a caching technique, so that the new >> chip DOES run at 16 Mhz, for it's cache, and 8Mhz for anything else. That way, >> you do get the speed improvement, and the system timing doesn't change. Only >> problem, is that that usually requires a caching chip, and another >> daughterboard, making the cost more than $100. >> >> If I'm wrong here, please jump in, I'd LOVE to be able to speed up my ST for >> <$100....(grin) > > The cheapest speed-up board I've seen was a project by the German magazine > "c't" (one of the fall issues, but I can't check it now). It should be around > $100 or cheaper, IF you're prepared to put it together yourself. > As far as I remember, it only speeds up accesses to the ROMs and internal > CPU operations, in RAM it runs at the normal speed (you can't speed up RAM > accesses because the Shifter needs to access video RAM at fixed intervals). > It has no cache, so it isn't as effective as the commercial boards, but it's > CHEAP. > > Jens Kilian > -- Why not just take out the 68000 and put in a 68010, most software should run and should run about 10% faster. People do this to thier amigas,but they have a piece of pd software that gets over some minor incompatability problems. P.S. Its a very cheap upgrade too. Not sure if this will work but it should Andrew Chambers Computer Consultant University Of Waikato New Zealand
tony@tacky.cs.olemiss.edu (Tony Reynolds) (01/14/91)
In article <1991Jan13.180628.2676@waikato.ac.nz> atc@waikato.ac.nz writes: >Why not just take out the 68000 and put in a 68010, most software should run >and should run about 10% faster. >People do this to thier amigas,but they have a piece of pd software that gets >over some minor incompatability problems. >P.S. Its a very cheap upgrade too. >Not sure if this will work but it should > >Andrew Chambers >Computer Consultant >University Of Waikato >New Zealand No Go. I desoldered my 68K and put in the 68010. Turn it on, the screen clears, everything looks great! The drive light comes on... Then, bombs, the screen clears, and more bombs! It keeps bombing till you turn it off! I looked through ST Internals and found some MOVE SR type instructions, in the boot-up sequence, which are privileged on the '010, and not on the '000. Now, I want to know if its my silly TOS version, that is, 1.0, which causes this problem, and if it's been fixed for TOS 1.2 or 1.4. Theoretically, TOS 1.6 should run.... I have some patches that I never implemented. Perhaps I'll get the time oneday.... +=/----\=+ Get up, get,get down \ Tony Reynolds |/ Byte \| I.B.M. is joking your \ tony@tacky.cs.olemiss.edu |\ Boys /| town. \ /usr/games/fortune: +=\----/=+ <<Appearing NetWide>> \If I had any humility, I'd be perfect. Starring MC 68000, Tony RISC, and V.G.A.\ --Ted Turner
steve@thelake.mn.org (Steve Yelvington) (01/15/91)
[tony@tacky.cs.olemiss.ed writes ... ] > I looked through ST Internals and found some MOVE SR type instructions, in > the boot-up sequence, which are privileged on the '010, and not on the '000. > > Now, I want to know if its my silly TOS version, that is, 1.0, which causes > this problem, and if it's been fixed for TOS 1.2 or 1.4. The Jan/Feb issue of Atari Explorer says TT TOS includes an exception-handler in the BIOS that traps this error and replaces it with a move from CCR. Apparently TOS ROMs are not the only software with this instruction. (I don't know whether TOS 1.4 and 1.6 have this error-handling, or whether they include the offending instruction.) ---- Steve Yelvington / P. O. Box 38 / Marine on St. Croix, MN 55047 USA Editor of the MAST (Minnesota Atari ST Users) newsletter - $15/year INTERNET: steve@thelake.mn.org UUCP: plains!umn-cs!thelake!steve
pegram@kira.UUCP (Robert B. Pegram) (01/16/91)
From article <1991Jan13.180628.2676@waikato.ac.nz>, by atc@waikato.ac.nz: > Why not just take out the 68000 and put in a 68010, most software should run > and should run about 10% faster. > People do this to thier amigas,but they have a piece of pd software that gets > over some minor incompatability problems. > P.S. Its a very cheap upgrade too. > Not sure if this will work but it should > Andrew Chambers > Computer Consultant > University Of Waikato > New Zealand Nope, won't work. As I understand it, the stackframe size is different and corners were cut in TOS. The line F stuff may interfere with things too - don' t know enough about '010's to tell. In general, if you want to run a 680X0 where X>0, you need a fixed version of TOS, either a german patched eprom version or TOS 1.62 from Atari. Also, remember that the 68000s are soldered into Atari STs. Bob Pegram pegram@griffin.uvm.edu or uunet!uvm-gen!pegram (line eater food)
klamer@mi.eltn.utwente.nl (Klamer Schutte -- Universiteit Twente) (01/16/91)
AAron@sun.soe.clarkson.edu writes: >From article <1991Jan14.060452.878@cs.olemiss.edu>, by tony@tacky.cs.olemiss.edu (Tony Reynolds): >> In article <1991Jan13.180628.2676@waikato.ac.nz> atc@waikato.ac.nz writes: >>>Why not just take out the 68000 and put in a 68010, most software should run >>>and should run about 10% faster. >> No Go. I desoldered my 68K and put in the 68010. Turn it on, the screen >> clears, everything looks great! The drive light comes on... >> >> Then, bombs, the screen clears, and more bombs! It keeps bombing till you >> turn it off! But please, a 68010 != a faster 68000! There is a different expection/trap format for the both chips. Tos assumes it is a 68000 and thus will fail. Same problem occurs with a 68020 and a 68030. (probably 68040 as well -- haven't seen any spec's about him). Klamer -- Klamer Schutte Faculty of electrical engineering -- University of Twente, The Netherlands klamer@mi.eltn.utwente.nl {backbone}!mcsun!mi.eltn.utwente.nl!klamer
marten@feki.toppoint.de (Marten Feldtmann) (01/24/91)
In article <klamer.664014474@mi.eltn.utwente.nl> klamer@mi.eltn.utwente.nl (Klamer Schutte -- Universiteit Twente) writes: >AAron@sun.soe.clarkson.edu writes: >>>>and should run about 10% faster. No, that's not realistic. The most you can get is about 3% to 5%. 5% with programs, wich do *much* arithmetic! >There is a different expection/trap format for the both chips. >Tos assumes it is a 68000 and thus will fail. Yes, that's the problem. But one can get here in Germany a patch-program for some money from a magazine called "c't", which reads the built-in TOS and produces a new TOS (one has to burn it into EPROMS of course) working with 68010/20. I did this work with the TOS-1.2. But some people are waiting for the release of a (not official) TOS-version called KAOS-TOS, which is said to work with the 010/020/030 and bases on TOS 1.4. It's said, that it's compatible with ABC GEM 2.2, has no Line-F Code and improvements an the AES and GEMDOS. One can even improve this program by using the KAOS-TOS together with a new programm call NVDI (= new vdi). Results from Quick Index for KAOS and NVDI: TOS text: 286% TOS string 1075% TOS scroll 132% GEM dialog 367% (numbers from 'ST-Computer, 1/91, page 21) NVDI is a commercial program. KAOS was offered to Atari, but Atari-US rejected it. (One can read in our magazines, that Atari-Germany is very interested in this new TOS-version.....) So the developers are still looking for another way to publish it. Some thought of the release via book, with a disk with a patch-program on it........ >Same problem occurs with a 68020 and a 68030. (probably 68040 as well -- >haven't seen any spec's about him). I thought, that the 040 is the first processor, which doesn't support the variable bus-range (8-, 16- or 32-bit), which allows the 020 or 030 to work with a 16-bit architecture like the ST oder Amiga(500/2000). Marten Marten Feldtmann, Kieler Str. 29, W-2300 Kiel 14, Germany, Tel.: +49 431 731916 DNET/EUNET/USENET/SUBNET: marten@tpki.toppoint.de Please keep your replies short - I have to pay for them