@2:56 'Can't you see how far ahead of its time this computer is' ... Thank you! To this day I cannot believe how everybody left the amiga to get a PC or an apple, or play games on a snes. That bein' said, my fixation on the amiga wasn't a good thing either, because, let's face it, Commodore just blew it.
Very valid information. In the 70s, the acronym MIPS was often taken to be short for Meaningless Interpretation of Processor Speed. I prefer computing primes or Machin's formula for computing Pi as a better benchmark .
I picked this example for several reasons. First, it was something you actually did on both architectures. Second, it's something simple enough to completely implement, start to finish, on both CPUs within the obvious constraints of a TH-cam video. Last, it's something that the 6502 has a fighting chance to win, even if under *extremely* favorable assumptions (displaying 8-bit, 68000 at same clock speed, available zero page addresses, it's okay to unroll the loop).
@@CompuSAR I agree completely and you picked an excellent example to illustrate the situation clearly. I'm just reiterating that in general, solving a problem which takes many instructions will always produce a more realistic figure than raw clock speeds.
As an original Amiga owner I have innumerable memories of trying to explain to other computer owners just what my machine was and what it was capable of.
@@CompuSAR I often wonder what the landscape of home computing would be like now if Amiga had not been bought by Commodore and then been mismanaged so badly or if Jack Tramiel had not purchased Atari and gone on the War-Path. So many ifs. I've collected old computers since 1990 and my thirteen Amigas are the pride of that collection. The original 1000 I bought in 1985 still complete in its box is the one I will never part with.
@@netcreature Steve Jobs gave an interview a short time before returning to Apple. You can find it if you search for "Steve Jobs lost interview". In it he is asked about why IBM won, and he says they had the wits to create an echo-system where other companies had an interest that their platform will succeed. He is wrong, of course. It was not an IBM decision, but was forced on it. Still, I think he's correct in analyzing why the winner was who it was. (which makes you wonder about his iPhone decisions, but that's another story). So I don't think there would have been any long term differences had the Amiga been better managed. So long as you're based on proprietary chips that your competitors can't copy, you'll always eventually lose against an open architecture.
@@CompuSAR The PC based systems would most likely have still grown to dominate but the Amiga could have made a much larger impact. If Jay Miners' team had more money earlier on and the Atari suite had not happened perhaps the 1000 could have been released as early as the Macintosh and caught some of the hype it created for a GUI and mouse driven OS. If Amiga had been properly marketed as well things would in all likelihood have been much more interesting and I've no doubt we would not have a mere two platforms today.
Your comparison is wildly generous to the 6502. DIVS and DIVU are 32 bit / 16 bit operations OTOMH and while a 6502 isn't too bad processing 8 bits, it's horrible processing 16 and 32 bit data.
Deliberately so. I'm trying to refute a claim that A is 2.5 times faster than B. So I take terms that are exceedingly favorable to A, and show that even under those terms it only comes in par, or very slightly ahead.
I... have not. I've added it to my list, but my time is so full of personal projects (of which this channel represents two), that I highly doubt I'll get to it before Netflix start adding commercials, at which point I like to tell myself I'll cancel my subscription.
THANK YOU! Sorry the "8 bit guy" just gets on my nerves with his claims. This and the "I need a SNES port for more than 1 button" is a personal gripe of mine (9 pin D-Sub has many implementations of more than 1 button, even if he didn't know the technical aspect of it. The examples should have been enough). Someone like him should know better. He should know that not every instruction set and architecture is the same, so MIPS is a very bad way to compare. Especially when you consider that the moment you move out of ZERO page, the 6502 becomes less efficient. It's a really nice CPU, but it is a product of it's time, a far cheaper competitor to the 6800 that also fixed some of it's major flaws. It's not even in the same generation of a 68000 let alone same market. It's like comparing a Z80 with a 286. Sure both come from "Intel architecture". But they are two very different architectures not just in market, but in time of release, capabilities and even design. Just because there might be ONE thing a 6502 does in less cycles, the practicality of that one thing is negligible, especially when compared to the benefits of the newer architecture. Sorry I'm ranting, it just was a claim that made me question myself and go "Wait, why is nobody else mentioning this?! Am I going crazy here?". So I am a year late to your video but still wish to thank you for it.
@@CompuSARWait? he did it again? I mean this guy should know better. Edit: Just watched your video, and YUP, he did it again. Due to the more modern instruction said it wouldn't even run 7 times faster but a multitude of that in everyday workloads. I mean he seems oblivious to the fact that 16 bit numbers exist and that a 68000 internally is a 32 bit architecture for it's calculations is a given. And I doubt he's that "dumb", he must know that the moment you move past 8 bit numbers and more advanced addressing, the 68000 just leaves the 6502 in the dust. And that's not a shame, it's a far newer architecture made for a completely different market when it was launched. It's just infuriating to also hear him talk about the Apple 2 GS, I mean try doing a spreadsheet on a 6502 based system vs a 68000. The latter can pull it off way more efficient and with less cycles than a 6502 ever can. Sorry heh ranting again.
Very interesting, even if I dont understand assembly,. I just compared famous basic benchmarks programs (rugg feldman, sieve of erathostene, creative computing benchmark etc.) in commodore basic 2.0 and in Amiga Basic, and saw Amiga Basic (that was not a very optimized basic) was much fasterr, i know basic is not a good stone of paragon between machines, but all those benchmarks really give the impression that the Amiga is.a much much faster machine then a c64, and we know amiga basic was not the best basic ever
I don't know the 65c816. If we were to use the same program as for the 6502, however, I suspect we'd get similar results. The 6502 really is very efficiently built. It might have new commands that would make things different. Either way, architecture counts for a lot. The 68000 and 65c816 are both 16 bit CPUs. The former, however, is a 32 bit chopped on the outside to 16 bit, whereas the later is an 8 bit CPU expanded to 16 bit. I suspect a performance comparison is still going to go the 68000 way.
Yeah, I know. It's one of those things you prepare in advance, and then you get to editing and curse. Mind you, for someone whose native tongue is not English, I think I'm doing okay. Words that belong to a technological or mathematical niche that I originally studied in a different language do trip me up. I try to look them up ahead of recording, but that doesn't always 100% work out. I think I use the same word in another part of this video where I do pronounce it correctly (or, at least, more correctly).
@2:56 'Can't you see how far ahead of its time this computer is' ... Thank you! To this day I cannot believe how everybody left the amiga to get a PC or an apple, or play games on a snes. That bein' said, my fixation on the amiga wasn't a good thing either, because, let's face it, Commodore just blew it.
Very valid information. In the 70s, the acronym MIPS was often taken to be short for Meaningless Interpretation of Processor Speed.
I prefer computing primes or Machin's formula for computing Pi as a better benchmark .
I picked this example for several reasons. First, it was something you actually did on both architectures. Second, it's something simple enough to completely implement, start to finish, on both CPUs within the obvious constraints of a TH-cam video. Last, it's something that the 6502 has a fighting chance to win, even if under *extremely* favorable assumptions (displaying 8-bit, 68000 at same clock speed, available zero page addresses, it's okay to unroll the loop).
@@CompuSAR I agree completely and you picked an excellent example to illustrate the situation clearly. I'm just reiterating that in general, solving a problem which takes many instructions will always produce a more realistic figure than raw clock speeds.
As an original Amiga owner I have innumerable memories of trying to explain to other computer owners just what my machine was and what it was capable of.
I know, right?
At least for me, that was whether they were interested in the subject or not.
@@CompuSAR I often wonder what the landscape of home computing would be like now if Amiga had not been bought by Commodore and then been mismanaged so badly or if Jack Tramiel had not purchased Atari and gone on the War-Path. So many ifs. I've collected old computers since 1990 and my thirteen Amigas are the pride of that collection. The original 1000 I bought in 1985 still complete in its box is the one I will never part with.
@@netcreature Steve Jobs gave an interview a short time before returning to Apple. You can find it if you search for "Steve Jobs lost interview". In it he is asked about why IBM won, and he says they had the wits to create an echo-system where other companies had an interest that their platform will succeed. He is wrong, of course. It was not an IBM decision, but was forced on it. Still, I think he's correct in analyzing why the winner was who it was.
(which makes you wonder about his iPhone decisions, but that's another story).
So I don't think there would have been any long term differences had the Amiga been better managed. So long as you're based on proprietary chips that your competitors can't copy, you'll always eventually lose against an open architecture.
@@CompuSAR The PC based systems would most likely have still grown to dominate but the Amiga could have made a much larger impact. If Jay Miners' team had more money earlier on and the Atari suite had not happened perhaps the 1000 could have been released as early as the Macintosh and caught some of the hype it created for a GUI and mouse driven OS. If Amiga had been properly marketed as well things would in all likelihood have been much more interesting and I've no doubt we would not have a mere two platforms today.
Your comparison is wildly generous to the 6502. DIVS and DIVU are 32 bit / 16 bit operations OTOMH and while a 6502 isn't too bad processing 8 bits, it's horrible processing 16 and 32 bit data.
Deliberately so. I'm trying to refute a claim that A is 2.5 times faster than B. So I take terms that are exceedingly favorable to A, and show that even under those terms it only comes in par, or very slightly ahead.
have you seen the The Billion Dollar Code series from Netflix? your video remind me that series for some reason!
I... have not. I've added it to my list, but my time is so full of personal projects (of which this channel represents two), that I highly doubt I'll get to it before Netflix start adding commercials, at which point I like to tell myself I'll cancel my subscription.
THANK YOU! Sorry the "8 bit guy" just gets on my nerves with his claims. This and the "I need a SNES port for more than 1 button" is a personal gripe of mine (9 pin D-Sub has many implementations of more than 1 button, even if he didn't know the technical aspect of it. The examples should have been enough). Someone like him should know better. He should know that not every instruction set and architecture is the same, so MIPS is a very bad way to compare.
Especially when you consider that the moment you move out of ZERO page, the 6502 becomes less efficient. It's a really nice CPU, but it is a product of it's time, a far cheaper competitor to the 6800 that also fixed some of it's major flaws. It's not even in the same generation of a 68000 let alone same market. It's like comparing a Z80 with a 286. Sure both come from "Intel architecture". But they are two very different architectures not just in market, but in time of release, capabilities and even design.
Just because there might be ONE thing a 6502 does in less cycles, the practicality of that one thing is negligible, especially when compared to the benefits of the newer architecture.
Sorry I'm ranting, it just was a claim that made me question myself and go "Wait, why is nobody else mentioning this?! Am I going crazy here?". So I am a year late to your video but still wish to thank you for it.
*This* video may be from a year ago, but my second video on the precise same subject is from last week!
th-cam.com/video/yqqrKww6wfE/w-d-xo.html
@@CompuSARWait? he did it again? I mean this guy should know better.
Edit: Just watched your video, and YUP, he did it again. Due to the more modern instruction said it wouldn't even run 7 times faster but a multitude of that in everyday workloads. I mean he seems oblivious to the fact that 16 bit numbers exist and that a 68000 internally is a 32 bit architecture for it's calculations is a given. And I doubt he's that "dumb", he must know that the moment you move past 8 bit numbers and more advanced addressing, the 68000 just leaves the 6502 in the dust. And that's not a shame, it's a far newer architecture made for a completely different market when it was launched.
It's just infuriating to also hear him talk about the Apple 2 GS, I mean try doing a spreadsheet on a 6502 based system vs a 68000. The latter can pull it off way more efficient and with less cycles than a 6502 ever can.
Sorry heh ranting again.
Very interesting, even if I dont understand assembly,.
I just compared famous basic benchmarks programs (rugg feldman, sieve of erathostene, creative computing benchmark etc.) in commodore basic 2.0 and in Amiga Basic, and saw Amiga Basic (that was not a very optimized basic) was much fasterr, i know basic is not a good stone of paragon between machines, but all those benchmarks really give the impression that the Amiga is.a much much faster machine then a c64, and we know amiga basic was not the best basic ever
amazing. thank you. I am now a subscriber.
Thank you. Always great to hear people appreciate your work.
Now do the same benchmark with the 65C816. 🙂
I don't know the 65c816. If we were to use the same program as for the 6502, however, I suspect we'd get similar results. The 6502 really is very efficiently built. It might have new commands that would make things different.
Either way, architecture counts for a lot. The 68000 and 65c816 are both 16 bit CPUs. The former, however, is a 32 bit chopped on the outside to 16 bit, whereas the later is an 8 bit CPU expanded to 16 bit. I suspect a performance comparison is still going to go the 68000 way.
fyi, it's pronounced "quo-shunt" not "quoy-tant"
Yeah, I know. It's one of those things you prepare in advance, and then you get to editing and curse.
Mind you, for someone whose native tongue is not English, I think I'm doing okay. Words that belong to a technological or mathematical niche that I originally studied in a different language do trip me up. I try to look them up ahead of recording, but that doesn't always 100% work out. I think I use the same word in another part of this video where I do pronounce it correctly (or, at least, more correctly).