unknown@ucscb.UCSC.EDU (The Unknown User) (01/17/90)
In article <759@cs.wmich.edu> dodgson@cs.wmich.edu (Harry Dodgson) writes: > I didn't find out what language was supported, but a 4 MHz 65C02 >can compete with an 8 MHz 68000 in execution cycles, so don't sell it short. >A few 16 MHz coprocessors don't hurt either. As far as a simulator, the board >executes all the code, so simulation on the Amiga screen isn't needed. > >notice: I got this info direct from the Chicago Atari office - no speculations, >no rumors, this is what the person told me. Wow, someone's finally giving the 65c02 some credit. This is a possible correction though. Other people have mentioned that the Lynx is 16 bit. This makes me think that it is run off of a 65c816 (yippee! yeah! same processor as my Apple IIGS!). -- If you want info about a 20 meg "floptical" drive mail me. YOU CAN ALSO GET ONE FOR --$440-- (NEW **LOWER** PRICE!) BY MAILING ME. [WE HAVE TO GET ENOUGH PEOPLE THOUGH..Slightly higher for less people but not much] unknown@ucscb.ucsc.edu APPLE II FOREVER APL24VR GS tips? Mail me. unknown%darkside.com@ames.arpa Please use the former address for regular use.
dnelson@mthvax.cs.miami.edu (Dru Nelson) (01/17/90)
> Wow, someone's finally giving the 65c02 some credit. You just having been looking hard enough. A lot of people love this chip. > This is a possible correction though. Other people have mentioned >that the Lynx is 16 bit. Since you don't have one, you would say this. Only one person mentioned that it is 16 bit. It is a 6502. The super famicom is the only I know that acts like your gs. >-- >If you want info about a 20 meg "floptical" drive mail me. YOU CAN ALSO >GET ONE FOR --$440-- (NEW **LOWER** PRICE!) BY MAILING ME. [WE HAVE TO GET >ENOUGH PEOPLE THOUGH..Slightly higher for less people but not much] >unknown@ucscb.ucsc.edu APPLE II FOREVER APL24VR GS tips? Mail me. >unknown%darkside.com@ames.arpa Please use the former address for regular use. You should really use your real name here and cut down on the commercial announcements. -- %% Dru Nelson %% Miami, FL %% Internet: dnelson@mthvax.cs.miami.edu %%
karl@sugar.hackercorp.com (Karl Lehenbauer) (01/18/90)
In article <7243@lindy.Stanford.EDU> unknown@ucscb.UCSC.EDU (The Unknown User) writes: >> I didn't find out what language was supported, but a 4 MHz 65C02 >>can compete with an 8 MHz 68000 in execution cycles, so don't sell it short. >>A few 16 MHz coprocessors don't hurt either. As far as a simulator, the board >>executes all the code, so simulation on the Amiga screen isn't needed. > Wow, someone's finally giving the 65c02 some credit. 65C02? That's the fast CMOS 6502, right? A 6502 with a few new instructions like zero a register and push x and y, right? The rest of this article assumes that this is the same 65C02 I had on an accelerator card in my ][+ a few years back, and not something like the 65816. Whimper. Look, the 6502 has, basically, three general purpose registers, A, X, and Y. (It also has a status register, program counter and an eight-bit stack pointer). You cannot use register indexing to go more than 256 bytes past an address. You cannot do indirect addressing unless the address you want to indirect through resides in the first 256 bytes of memory. The stack is limited to 256 bytes. A MULTIPLY INSTRUCTION?? PSHAW. On the 6502 you will do it with eight-bit registers and single bit shifts. A 16 X 16->32 bit multiply in 6502 assembler is *pages.* (It's ONE instruction on the 68000.) A 16-bit move from memory to memory is four instructions, to do with indirect addressing requires something like twelve. Although the 6502 can execute instructions in fewer clock cycles than most other cruddy old eight-bit CISC microprocessors, those instructions typically don't do as much. From a programming standpoint the architecture is so totally brain-dead that even an 8080 looks heavenly by comparison. The scalable sprites and stuff, though, sound really interesting. -- -- uunet!sugar!karl "It takes a smart man to know when he's stupid." -- -- Barney Rubble -- Usenet access: (713) 438-5018
portuesi@tweezers.esd.sgi.com (Michael Portuesi) (01/18/90)
>>>>> On 18 Jan 90 05:41:45 GMT, karl@sugar.hackercorp.com (Karl Lehenbauer) said:
karl> A MULTIPLY INSTRUCTION?? PSHAW. On the 6502 you will do it with eight-bit
karl> registers and single bit shifts. A 16 X 16->32 bit multiply in 6502 assembler
karl> is *pages.* (It's ONE instruction on the 68000.)
The Lynx has a hardware multiplier circuit.
--M
--
__
\/ Michael Portuesi Silicon Graphics Computer Systems, Inc.
portuesi@SGI.COM Entry Systems Division -- Engineeringsdh@flash.bellcore.com (Stephen D Hawley) (01/18/90)
In article <4952@sugar.hackercorp.com> karl@sugar.hackercorp.com (Karl Lehenbauer) writes: >In article <7243@lindy.Stanford.EDU> unknown@ucscb.UCSC.EDU (The Unknown User) writes: > > [ranting about the 6502 removed] Yes, it is a small processor. It is also very old, and designed for other things than 68000's. You have recognized this, but went swinging out into space in your fury. The main difference between, which I imagine is being exploited, is that the 65C02 is CHEAPER than any 68000 variant, and it is CMOS (remember the Lynx is a handheld). 4 MHz? Not too shabby. Recall that some very impressive things were done on 1 MHz 6502's and 6809's. This is because the programmer ends up handcrafting the code to a large extent. The programmer must take different approaches to problems. For example, I wrote a program to display "star trek" stars zooming by in one-point perspective on an Apple ][. Instead of using division for perspective calculation, I used a table lookup. I wrote the same program on a Macintosh using the exact same techniques to address the frame buffer, but used division for the perspective calculation (written in a mix of C and assembly). Guess what. The Apple version could display more than twice as many stars before it was slower than the Mac version, and even with all the tables it was aboit half the size. What's wrong with this picture? The Apple runs at 1Mhz and the Mac at 8MHz. You can't blame it on the Mac since I never touch the operating system -it's the 68000. Kinda makes you wonder... Steve "Push the hardware limitations back with your mind" Hawley sdh@flash.bellcore.com A noun's a special kind of word. It's ev'ry name you ever heard. I find it quite interesting, A noun's a person place or thing.
karl@sugar.hackercorp.com (Karl Lehenbauer) (01/19/90)
In article <19142@bellcore.bellcore.com> sdh@flash.UUCP (Stephen D Hawley) writes: >You have recognized this, but went swinging out into >space in your fury. Hardly fury. Hey, I could program an 1802 with toggle switches and get it to do something more or less useful. That doesn't mean I want to. I acheived a near maximum guhuri level on the 6502, IIMSSM. At one point on the Apple ][+ I had hacked up a two-user Forth environment with a tree- structured filesystem and multiple assembly tasks running under a realtime exec, and the code in that exec was damn near optimal. Oh yeah, I wrote some massively hacked bit-blit routines, too. (Yuck, this all keeps coming back) a-and my own loop-while-speaker-toggling sound-playing routines that I tuned up with a strobotuner. My point is that *I've been there.* And I'm not going back. I finally learned the futility of trying to reinvent the wheel without massive resources. On the Apple that was more like "invent the wheel," 'cuz most of that stuff wasn't available. But on the Amiga, I have all these guys at Commodore (and elsewhere) writing that stuff for me. Tree-structured filesystem. Realtime exec. Bit-blit routines. Yep, it's all there, plus groovy stuff like C compilers, sample-playing hardware, audio.device, serial.device, BitBltMapRastPort or whatever it is, the blitter, windows, screens, fonts, fish disks, IFF (!) &c &c &c If you choose to develop for the Lynx, you are choosing to assume some major burdens, to wit: small developer community, little "garage" development (high entry price); you'll be insular (little free code) and poorly supported. Good luck, and this is *not* a flame. It's just that *I* want more of an environment and ever-higher performance, not less, so I see moving towards (Indeed I am already there) protected-mode 32-bit systems adhering to open systems standards (Unix) on the one hand, and realtime multitasking on an affordable machine, hopefully on the same hand, but if not, as for now, on the other hand, i.e. the Amiga. -- -- uunet!sugar!karl "It takes a smart man to know when he's stupid." -- -- Barney Rubble -- Usenet access: (713) 438-5018
chad@ucscb.UCSC.EDU (darknight) (01/20/90)
>In article <foo@blah.blah.com> sdh@flash.UUCP (Stephen D Hawley) writes: [ Much bragging about his programming accomplishments that have little or nothing to do with the Lynx machine, and discussion of the Ultimate Computer Environment deleted] >Good luck, and this is *not* a flame. It's just that *I* want more of an >environment and ever-higher performance, not less, so I see moving towards >(Indeed I am already there) protected-mode 32-bit systems adhering to >open systems standards (Unix) on the one hand, and realtime multitasking on >an affordable machine, hopefully on the same hand, but if not, as for now, >on the other hand, i.e. the Amiga. PRESS RELEASE -- 1/19/90 Atari(tm) has just announced an upgrade of their Lynx(tm) system known as the "Lynx Too"(tm), and the subsequent discontinuation of the original model Lynx, due to some humbling comments made by one Stephen Hawley(tm) on Usenet. It seems that they've determined with help from Stephen(tm) that people do not really want a hand held game machine with lots of Neeto(tm) colors, or lots of Neeto(tm) sounds, or lots of Neeto(tm) sprites, or lots of Neeto(tm) blits, or lots of Neeto(tm) fun. Nope, they want a hand-held game machine with a Well Designed Architecture(tm), Structured Gameplay(tm), Expert System Capabilities(tm), Demand-Paging(tm), a Rexx Port(tm), Full Extensibility(tm), a Graphical User Interface(tm), an Advanced 32-Bit Bus(tm), and Unix(tm) SysV Rel.4(tm)... So Atari(tm) whipped it all together and is releasing the "Lynx Too" instead, which has all the above features that Stephen(tm) asked for, plus a Neeto(tm) gold finish. Upon seeing it, Stephen(tm) remarked that it now fit within his realm of consideration as a true Handheld Video Toy(tm), and was quoted as saying, "It [shines] like [gold]!" Atari(tm) said they were very pleased with Stephen's(tm) suggestions, and they were quoted as saying, "Oh yeah? Well he [Stephen(tm)] [shines] like [gold], too!" List price of the "Lynx Too" is slated at $9999.95, but with a special Trade-In-Your-NeXT(tm)-Get-A-Lynx_Too-For-A-Mere-$3000-Trade- In-Offer, Atari(tm) clearly has the student market in mind for their new HandHeld Video Toy(tm). Atari(tm) did note that because of the licensing fees for Unix(tm) they were unable to develop any games for the machine, but they don't expect that will deter Stephen(tm) from buying "Whopping great wodges of 'em". NEXT WEEK -- The Amiga 3500 takes over the Home-Recipe-Database market with it's N^Googolplex color pallete, N^Googol bitplane graphics, and infinite user capability. Stay Tooned(tm). -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ INTERNET: chad@ucscb.ucsc.edu Chad 'The_Walrus' Netzer->AmigaManiac++ ------====== "The memory of other cannot be trusted." ======-------
daveh@cbmvax.cbm.commodore.com.commodore.com (Dave Haynie) (01/23/90)
in article <4952@sugar.hackercorp.com>, karl@sugar.hackercorp.com (Karl Lehenbauer) says: > Xref: cbmvax comp.sys.amiga:50282 rec.games.video:5610 > Although the 6502 can execute instructions in fewer clock cycles than most > other cruddy old eight-bit CISC microprocessors, those instructions typically > don't do as much. From a programming standpoint the architecture is so totally > brain-dead that even an 8080 looks heavenly by comparison. That's somewhat debatable, but it doesn't really matter. The 6502 is the defacto standard in the video game world. If you're a games programmer, you know 6502. Here's the list: Atari 2600, etc.; Atari 800, etc. Commodore VIC 20, C64, C128 Apple II/IIe/IIc, etc. Nintendo (actually uses a 6502 clone with all the MOS patented things cut out). So you would think, if you're trying to get games ported to any small, cheap game machine, a 6502 compatible CPU is probably an awfully good idea. > -- uunet!sugar!karl "It takes a smart man to know when he's stupid." -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy Too much of everything is just enough
chad@ucscb.UCSC.EDU (darknight) (01/23/90)
In article <7380@lindy.Stanford.EDU> chad@ucscb.UCSC.EDU (darknight) writes: > >>In article <foo@blah.blah.com> sdh@flash.UUCP (Stephen D Hawley) writes: > >>Good luck, and this is *not* a flame. It's just that *I* want more of an >>environment and ever-higher performance, not less, so I see moving towards >>(Indeed I am already there) protected-mode 32-bit systems adhering to >>open systems standards (Unix) on the one hand, and realtime multitasking on >>an affordable machine, hopefully on the same hand, but if not, as for now, >>on the other hand, i.e. the Amiga. > > [Much brilliantly conceived humour deleted] The above attribution is wrong! It was NOT said by Stephen Hawley, but by Karl Lehenbauer... As you can see from the nesting level, Stephen was responding to Karl (whose name I foolishly erased). Hey, it wasn't me, my Nesting-Level-Sensitive-Warning-Glasses failed to operate... That's the last time I wear them in the dark! But seriously, I apologize to Stephen for this mistake, as well as the net for making such a basic error... (I guess this just proves that it DOES happen to the best of us. :-) Finally, this means that the humor and cynicism in the article should have been vented on Karl instead of Stephen. (MWUUU-Ha-Ha-Ha [text version of devious laughter]). In my previous posting (or in you head) please change all the "Stephen"'s to "Karl"'s...[ie. Karl(tm)] Flame me by mail, if you must... -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ INTERNET: chad@ucscb.ucsc.edu Chad 'The_Walrus' Netzer->AmigaManiac++ ------====== "The memory of others cannot be trusted." ======-------