jerry@polygen.uucp (Jerry Sheckel) (12/29/90)
whitewolf@gnh-starport.cts.com (Tae Song) writes: >jerry@polygen.uucp writes > >I have to disagree with on the 8088 been faster than the 65816. The 65816 can >retrieve 8-bit data in a single cycle instead of 4. 65816 has pipe-lining for >higher efficency and execution speed. > I could be wrong here, but I believe the 8088 also has pipelining. Retrieve 8-bit data in a single cycle? Are you sure that's valid for anything but zero-page addressing? > >You try using Windows 3.0 and Finder... >but that's of course apples and oranges, but imagine trying to write Finder in >8088/86 code... > Oh, I have tried Windows 3.0 and Finder. You're right, the comparison is apples and oranges, because Windows 3.0 is much more than the simple graphical program launcher that the Finder is. Windows supports multitasking, inter- process communication, etc. I agree that even something as simple as the Finder would be slow on a stock PC, but I find it *very* slow on a base-model Apple IIgs. By the way, Windows is most probably written in a compiled language like C. It just ain't the most efficient piece of code. Take a look at GEOS. It is a multitasking window system that runs on the 8088 much faster than Windows and requires only 512K RAM. It would be fair to compare that with the GS Finder. > >What >do you base functionality on? There isn't anything that you could do on an >8088 that you couldn't do on a 65816. > Yeah? How about loading 4 general purpose registers with values? How about being able to write completely relocatable programs without additional memory management hardware? > > >The Gs has only been out for 4 years, >now... how long has the IBM PC been out? > That's my point. You're comparing a 4-year-old GS with a 10-year-old PC. Is that fair? Even if its performance were similar, the GS would still be too little too late for way too much money. > >On the point of comparing two packages... take the example of AppleWorks and >some popular IBM package comparison several years ago... Both persons were >considered experts in using their software, having written books on using the >packages... AppleWorks won running an accelerated Apple IIe (3.6Mhz) I think >against a maybe AT class IBM... They were working on a spread sheet. Now, you >will agree a 286 machine is MUCH faster than a PC or even an XT... so much for >the MIPS and what have you. I believe the IBM also had a math co-processor. > You're saying that a 286+287 at 8MHz is slower than a 6502 at 3.6MHz in terms of number-crunching speed? In terms of floating point? That's so ridiculous I don't even know what to say about it, except that maybe you should give it a try for yourself. > >whitewolf@gnh-starport!info-apple > -- +-------------------+----------------------+---------------------------------+ | JERRY J. SHEKHEL | POLYGEN CORPORATION | When I was young, I had to walk | | Drummers do it... | Waltham, MA USA | to school and back every day -- | | ... In rhythm! | (617) 890-2175 | 20 miles, uphill both ways. | +-------------------+----------------------+---------------------------------+ | ...! [ princeton mit-eddie bu sunne ] !polygen!jerry | | jerry@polygen.com | +----------------------------------------------------------------------------+