NES vs 7800

[supercat]

But I would never ever claim that an equally clocked 6502 is equal or better to an equally clocked 68000. Only if you construct some special cases, which theoretically are possible, but practically, you again end up limiting yourself on what you can do on the 6502, while on the 68k, you can accomplish much more with much less effort.

 

The 68000 is the winner, but in some cases it doesn't win by such a huge margin as you might expect. For example, to copy a lot of bytes as quickly as possible from one part of memory to another, the 6502 will average 9 cycles/byte, while the 68000 will take 20 cycles per four bytes (5/byte). Actually, the 68000 may be able to do a skosh better if one doesn't mind trashing a bunch of registers. Moving 32 bytes at a time (trashing eight registers) would take 152 cycles (4.75/byte). A win versus the 6502, but by less than a factor of two.

 

If one were not allowed to keep values in registers between samples, the BTP2 music player in Stella's Stocking would be:

; 68000
12  move.l #table1,a0
12  move.l #phases_and_results,a2
04  moveq  #0,d0

08  move.w (a2),d0
14  move.w (a0+d0),d1; Amplitude
18  move.w (a1+d0+2),(a2)+; Phase
08  move.w (a2)+,d0
14  add.w  (a0+d0),d1
08  move.w d1,(a2)+
 
08  move.w (a2)+,d0
14  move.w (a0+d0),d1; Amplitude
08  move.w (a2)+,d0
14  add.w  (a0+d0),d1
08  move.w d1,(a2)+
; Total time 28 + 70 + 52 = 150 cycles

I'm being generous and only updating one of the four phase values (in the BTP2 player, I have four different macros which are invoked one per line). Perhaps my code is not optimal, but I think it's pretty good. The BTP music player runs at a rate of 46 cycles per sample, less than a third the cycle count of the 68000 version. Even if I assume registers A0, A1, and the top half of D0 will be left alone between samples, the cycle count would still be 2.6 times that of the 6502 version. Even if I assume I'm going to unroll the loop and make maximum use of registers:

18  move.l (a0+d0+n),d4; Two amplitudes
20  add.l  (a0+d1+n),d4
12  move.l d4,(a0)+
18  move.l (a0+d2+n),d4; Two amplitudes
20  add.l  (a0+d3+n),d4
12  move.l d4,(a0)+
; 100 cycles per two samples, not counting other overhead

That's doing pretty well, actually, but measuring the BTP2 player by the same standard would yield a time of 26 cycles/sample (rather than 46).

Edited August 8, 2008 by supercat

[Vigo]

That sort of tells me you might not know how to efficiently code for a 65x processor then. It's called optimize accordingly. I've been coding on a 65x for years and I haven't had a problem with the lower register count. Never had a problem with the 256 ZP registers either

 

Okay, another pissing contest. My statement neither reveals my skills at 6502 optimizing nor does your statement invalidate the FACT that the 6502 has one of the most crippled register architectures out there, and all 6502 optimization out there is to bend yourself coping with these limitations, while on 68k, you simply need not to think about those issues.

 

And I never said that you did. The discussion is in the context of consoles, not computers, and as such is optimized/written differently.

 

Have you ever written a Genesis game? Who are you to say how Genesis games (or any other game on a 68000 platform) should be written?

 

No, that wasn't my point. Geez. My point was that optimizing for 32bit operands represents such a small percentage of the overall CPU resource for 68k on the Genesis. And that you keep repeating generic examples of LONG operands as if it's strength of the 68k and the sole exertion power in context of the Genesis (and even in general). It's not. But you keep repeating it...

 

I repeat it, because that was the line you misquoted from my post in your very first post and taken out of context. I only used this particular quote for this particular example, but YOU are the one now starting to cram another useless discussion down my throat about things I never said, nor ever disputed. And besides, read your first posts again where you clearly stated that in no case, 32bit optimization is useful for the Genesis, which is simply horse pucky!

 

If you knew how to effectively code for the 65x, you'd know optimizations are totally different than coding for a 68k. That's not hard to understand.

Limited registers means a different approach to coding and problem solving than a processor with a larger amount of registers. Some solutions are less (than) graceful, but the faster instructions times help expedite this. Never underestimate small LUTs on a 65x either . There are also other memory/test(compare) effective instructions besides using load/store - sparing register usage. Game logic is pretty simple. It's not complex. It doesn't require overly complex code. 8bit elements easily fit into the core design logic. In the context of the PCE and the MD.

 

It quite suits you indirectly calling me dense, but starting a whole discussion by misquotinq my first statement, and discussing again, points which I already made myself clear. Your assumptions are vague and pretty general ("a game does not need this"). If I have an 68000, who stops me from using the benefits of this architecture? Are you coming to me now, telling me I should not, because that's not what the Genesis is supposed to do?

 

Narrow minded? I prefer the term "focused"

 

Lol, focused on getting on my nerves.

 

I'm not saying that the 68k is crap. It's not. The 68k has lots of positive attributes. It has a great instruction set, a roomy register set, easy to code efficiently for, and my favorite - linear address range. It's just not exaggeratingly "much more powerful" as you say.

 

From my standpoint, it is much more powerful, but if you would have read the thread, it would have become quite clear what I meant.

 

Seems to be a trend with a lot of 68k old schoolers, that nothing can touch 68k.

 

How do you know where I come from? I have coded on many different CPU architectures, and yes, the 68k was indeed the most pleasent for me, it's a beautiful CISC design. I'm sorry if my remarks awake "bitness envy" in you.

 

Performance is relative to context. And in the context of the 16bit consoles, the PCE's huc6280 shines spectacularly.

 

Yeah, whooopie doo, exactly the same I posted in every 2nd post up before you came up with this useless provocation. I already wrote software and designed hardware for the PC-Engine, so don't tell ME how much it shines.

 

 

My self made Flash card development system for the PC-Engine. Completely with 1MB ram, file browser, and loads files at about 350-400 KB/s. Currently not operative because I am using the SRAM's and the flash chip in other projects.

Edited August 8, 2008 by Vigo

[malducci]

The 68000 is the winner, but in some cases it doesn't win by such a huge margin as you might expect. For example, to copy a lot of bytes as quickly as possible from one part of memory to another, the 6502 will average 9 cycles/byte, while the 68000 will take 20 cycles per four bytes (5/byte). Actually, the 68000 may be able to do a skosh better if one doesn't mind trashing a bunch of registers. Moving 32 bytes at a time (trashing eight registers) would take 152 cycles (4.75/byte). A win versus the 6502, but by less than a factor of two.

 

Though not a standard 6502, the HuC6280 does block copying at 6cycles per byte - with a 17 cycle setup overhead but it pushes and pops all registers for you. Block copy up to 64k similar to the 65816's block copy opcodes.

 

I'm curious about your music player. Are you handling the volume of the samples in software? I have something similar using 262 scanline interrupt system for MOD, but I'd like to do something else (chiptune style). What are some of the type of effects you plan or have implemented (LFO, FM, AM, etc)?

 

 

Okay, another pissing contest.

Sorry to disappoint, but it's not a pissing contest. End it at that.

[Vigo]

Okay, another pissing contest.

Sorry to disappoint, but it's not a pissing contest. End it at that.

 

It has become on the moment you started, for no good reason, calling my experience on 6502 coding into question. YOU started it, YOU end it.

[supercat]

What are some of the type of effects you plan or have implemented (LFO, FM, AM, etc)?

 

Effects? Effects!?

 

The BTP2 music player, while amazing by 2600 standards, is really very crude. Algorithmically:

; Line 0
AmpL = aptr0d[phase0] + aptr1c[phase0]
AmpR = aptr2b[phase2] + aptr3a[phase3]
phase0 = phptr0[phase0]
; Line 1
AmpL = aptr0a[phase0] + aptr1d[phase0]
AmpR = aptr2c[phase2] + aptr3b[phase3]
phase1 = phptr1[phase1]
; Line 2
AmpL = aptr0b[phase0] + aptr1a[phase0]
AmpR = aptr2d[phase2] + aptr3c[phase3]
phase2 = phptr2[phase2]
; Line 3
AmpL = aptr0c[phase0] + aptr1b[phase0]
AmpR = aptr2a[phase2] + aptr3d[phase3]
phase3 = phptr3[phase3]

Different volumes are handled by using different wave tables. Crude, but reasonably effective. For each volume there are twelve wave tables, one for each pitch. The lowest one, C, is 60 cycles long and is padded by 31 bytes; the others are padded by 15 bytes. The effect of that is that by changing pointers I can achieve any pitch in a five octave range. The only "effect" available, however, is wave-table selection.

 

Still, managing real-time four-voice 15.75KHz output stereo music at 60% CPU utilization on a 1.19MHz 6502 is nothing to sneeze at. A 68000 machine could do better than the Atari 2600 by virtue of having more RAM to play with, but I think the 6502 really holds its own. BTW, about 15 years ago I wrote a 68000 music player which generated 256 samples at a time. That one processed each voice separately, summing them in the output buffer. As with the BTP2 player, amplitudes were handled by using different wavetables. From memory, the loop was something like:

lp:
 add.w d0,d1 ; D0=freq; D1=phase, LSW
 addx.b d2,d3; D2=freq; D3=phase; MSB
 move.b (a0+d3.w),d4
 add.b d4,(a1)+
; Unroll loop a few more times, then
 dbra d5,lp

Not quite as efficient as the BTP2 player, but it allowed for 24-bit frequency accuracy while only needing one wave table per amplitude (instead of twelve). In retrospect, I could perhaps have built up 16-bit waves (even though the output will only be 8 bit) and then post-processed to convert 16 bits to 8. That would have not only allowed a slight speedup in the generation (at the expense of extra post-processing), but also allowed improved precision.

lp:
 add.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d4
 swap
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d4
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d5
 swap
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d5
 add.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d6
 swap
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d6
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d7
 swap
 addx.l d0,d1 ; D0=freq; D1=phase, sorta
 and.w d2,a1 ; A1 holds constant $000001FE
 move.w (a0+d2.w),d7
 addx.w d2,d1; D2 holds constant zero
 add.l d4,(a2)+
 add.l d5,(a2)+
 add.l d6,(a2)+
 add.l d7,(a2)+
 dbra d3,lp

Thinking about it, the code could even have been adapted to use tables of 16-bit samples if bit 0 of d3 was clear but bit 0 of d0 was set. That might have been a nice approach. Oh well... For timing, let's see. Each pair of samples would need:

  addx.l reg,reg  ; 2x6 = 12
 and.w reg,reg  ; 2x4 = 8
 move.w (indexed),reg; 2x10 = 20
 add.l reg,(mem)+; 2x20 = 40
 swap ; 1x4 = 4 
[code]
So that's 84 cycles per pair, plus loop overhead of
[code]
 addx.w reg,reg; 4
 dbra reg,dest ; 12?

So 8 samples would average 44 cycles each, per voice. Note that the code stores the two LSB's of the phase and frequency in the upper word of the registers, and the MSB in the bottom, so as to allow the latter to be used directly for indexing. Bit 0 of r0 must be set to allow carries to propagate to bit 1.

[DracIsBack]

What was GUMBYs specs?

 

Good question. I don't know, to be honest. They talk about it briefly in the 7800 reunion MP3 that Curt Vendel has posted but I don't remember them getting into technical details.

[oky2000]

Can you imagine if Jack was working for FOX studios when George Lucas came in with a script for a movie called STAR WARS? He probably would have said: "Hey, we don't need your movie, we'll just colorize some old b&w Flash Gordon serials!"

That's basically what a bunch of studios did. He shopped Star Wars around to numerous studios and they all declined. Until George met Alan Ladd Jr. at 20th Century Fox.

 

 

It's amazing some of the morons that get to run companies they have NO business running.

 

I have to give the Tramiel's a big fat red FU for the most

idiotic decisions a company exec could possibly make.

 

How do you buy a game company and abandon the games?

 

Horses toots! All of them!

 

I think that's unfair. You have to remember at the time he bought Atari the technology that came with it was practically worth sweet FA. From this he got millions of people buying the ST (which again had to be made from scratch). Jack was an exemplary, but particular, type of cut-throat business man. He inherited the debts of Atari as well as the deeds to the company. He inherited outdated technology (2600, A8) when better things were already getting ready to be released on the market.

 

The C64 is his successful legacy, and only happened because he bought MOS Technologies...and from that day on it was game over. The reason the C64 was cheaper than the Atari 800/Apple II is NOT because it is inferior (there are things the C64 does the A8 can never do as well and vice versa but please no fanboys replying to this bit!) but because he didn't have to pay a middle man for the technology and got it at cost as he owned MOS. The C64 started life as a development project for a traditional early 80s arcade board NOT a home computer (which probably explains the crappy basic and overly complex sound and graphics chips compared to most other 8 bit micros)

 

What happened when Jack took over Atari was he didn't have available the advantages his business strategy relied upon. There was no easy way to make something both better AND cheaper to produce than the Amiga simple as that. The ST was a compromise, still better than the shitty mono MAC so history could not repeat its self a la C64 business model AND he didn't have Irving Gould's deep pockets to finance such strategies. I think Jack did the best he could with what he had. People forget Commodore was Irving Gould's financial resources AND Jack Tramiel's business acumen. Look what Commodore pissed away with the Amiga A1000's technical advantage in 1985 for god's sake, never had there been such a huge gap between the established systems and the new kid on the block in technical terms. You think Jack would have marketed it as a $1500 machine for rich clueless people. It was a games machine, only an idiot at post Tramiel Commodore would market the Amiga as anything else!

 

Please don't get me wrong, I would rather I walk into a shop and my only two choices of computer were still Commodore or Atari not the pieces of shit running Windows or piece of shit over priced Apple rubbish. How the hell we ended up with the two most pathetic designs as our only choice is just beyond me and is why computing is just zzzzzzzzzzzz

 

Edit: Don't forget the price didn't include the Atari arcade division, that was way out of his price range.

Edited January 23, 2009 by oky2000

[oky2000]

Probably because no one at the time thought that Nintendo was going to revive the dead American video game market (and interest in video games) back around the mid-1980s until they saw NESes and games really selling in the late 1980s?

 

Frankly, I'm not sure how well the Atari 7800 would have sold if it had been released back in 1984, given the state of both Atari and the video game market at that point. It could have wound up as another Vectrex, leaving Nintendo and Sega as the only real contenders for the "king of the market" throne with the following generation.

 

I have to agree with you on this for the crash knocked out three of the four gaming systems of the time and shops in the Uk couldn't offload the stuff fast enough. Had the 7800 been released in 84 I doubt many places would have stocked it and aside from die hard Atari fans I question whether anyone else would have bought it?

 

I think the clever company (ie the one who would have escaped the crash) would have released the 7800 as a home computer. What killed off the consoles was the £30 cartridge games weren't better than the best of the £6-£8 cassette games or even £12 discs for more affluent buyers. Congo Bongo on the Colecovision was 3x as much as the enhanced Sega developed C64 disk version in price and looked like it was running on hardware technically a generation apart!

 

I would rather die than be forced to play the childish NES games only as my only source of 8bit gaming entertainment. It was too childish and toy like, certainly I have no idea how such games became cool, but then there is no accounting for taste in some markets is there

[oky2000]

You and I simply disagree on what constitutes good games.

 

My Personal Opinion is that the NES has terrible, poorly colored and flickery graphics, even in the so-called "good games". and simply not enough games that I like to play.

 

And I'd argue that the Sega Master System is better than either of them.

 

I think it's safe to say that your opinion is very much in the minority. Just the number of A+ titles for the NES was probably about the same size as the entire 7800's library. The SMS had plenty of good games as well, but saying it had a better library of games than the NES is ridiculous.

 

I can't stand Zelda or Mario games...now tell me the NES had better games? I'd rather play Gauntlet Hang-On and R-type than most NES games. In fact I only have 2 NES games..Galaxian+Galaga Cart and Donkey Kong because they are both arcade perfect...(but then again so was Donkey Kong by Ocean software on the C64 for 1/6 the price)

 

If the Famicom is a 1983 machine though the fact the SMS is technically better doesn't mean anything, the Amiga was only 3 years later than the C64 to put that into perspective.

[supercat]

I think the clever company (ie the one who would have escaped the crash) would have released the 7800 as a home computer.

 

For computing purposes, how would the 7800 be better than the Atari 8-bits? I suppose if one could have afforded the RAM, a 320x200x4-color screen mode might have been nice, but MARIA really seems better suited for gaming than computing.

[gdement]

I can't stand Zelda or Mario games...now tell me the NES had better games? I'd rather play Gauntlet Hang-On and R-type than most NES games. In fact I only have 2 NES games..Galaxian+Galaga Cart and Donkey Kong because they are both arcade perfect...(but then again so was Donkey Kong by Ocean software on the C64 for 1/6 the price)

The NES has games of those types as well, but of course not everybody prefers the NES library. Just most people.

I don't see what makes the SMS any less childish than the NES. Nintendo didn't get tagged that way until much later. Except maybe by the computer gamers, who looked down on consoles in general, not just Nintendo.

 

If the Famicom is a 1983 machine though the fact the SMS is technically better doesn't mean anything, the Amiga was only 3 years later than the C64 to put that into perspective.

Regardless of when they were built, the important result is the SMS had a technical advantage when they went into competition outside Japan in 1986/87.

The Amiga wasn't just 3 years newer, it also cost a fortune compared to the C64.

 

I would rather I walk into a shop and my only two choices of computer were still Commodore or Atari not the pieces of shit running Windows or piece of shit over priced Apple rubbish. How the hell we ended up with the two most pathetic designs as our only choice is just beyond me and is why computing is just zzzzzzzzzzzz

Probably because the real money was in computers for practical business use, not games. Thus graphics and sound didn't matter much, Commodore had questionable quality, and Atari was known as a game company.

C64 did well because it was cheap enough to buy for kids and they could play games on it, but it's hard to trust them for serious work, and the floppy drive was horrible. Maybe they should have released an upscale business model much earlier, comparable to IBM but cheaper.

 

Still, I think Commodore could have had more staying power if they only had been able to design a significantly more powerful computer that retained compatibility. Instead, C64 had a great run but became obsolete and disappeared. They didn't turn it into a lasting architecture.

Edited January 24, 2009 by gdement

[oky2000]

I guess all the arcade conversions of the SMS library gave it a slightly less childlike atmosphere as arcades were really populated by teens and adults not young kids in he 70s/early 80s, Marketing was also a part of it sure. The Genesis/Megadrive took that one stage further in it's UK advertising, purposefully distancing itself from the SNES cutesy games.

 

The C64 was artificially cheap because MOS made the custom chips they designed and Commodore owned MOS. You can add $200-300 to the price of a C64 if say Atari wanted to build it. That was Jack's genius at work. The C64 was selling well right into the late 80s too. The cracks started to appear in the early 90s when they were still trying to sell it. Not much choice though as the Commodore 16/Plus4 were disasters waiting to happen and the Commodore 128 and disk drive combination was perilously close to the price of an Amiga 500 in the 80s so 8bit games didn't have a massive choice really.

 

Also Commodore were very respected by the business world with the PET series in the very early days of computing. I totally agree with you on compatibility though, the Amiga should have shipped with a DOS emulator to atleast run Word Perfect/Dbase/Lotus 1-2-3/MS-Dos etc. Even I thought that when it was launched. Most people who bought a 'multimedia' PC around the time of the Falcon/Amiga 1200 had no choice. Universities and businesses were using Microsoft Office etc and so they needed to be able to work on such files at home hence they plumped for a PC (the massive advantage of the PCs byte per pixel 256 colour VGA meant 8 times less work was required to run games like Doom which helped a lot). As for build quality the Amiga 1000 was beyond reproach, I still have my original 1986 Amiga 1000 to this day which has never ever let me down once. Most A500s I have had problems with were down to cheap internal drives.

 

It's all water under the bridge now though really.

[oky2000]

I think the clever company (ie the one who would have escaped the crash) would have released the 7800 as a home computer.

 

For computing purposes, how would the 7800 be better than the Atari 8-bits? I suppose if one could have afforded the RAM, a 320x200x4-color screen mode might have been nice, but MARIA really seems better suited for gaming than computing.

 

Well the Atari and Commodore with the exotic custom chips weren't really the norm, competitors had to make do with just programmable character sets and attribute colour control like MSX with no hardware pixel scroll etc. A colourful 7800 based machine with a basic rom and some extra ram would have made more colourful games of the style popular at the time, in terms of on screen colour resolution available and sprite flexibility. The best home computers in terms of sales were always the games machines disguised as home computers (Amiga 500/C64 etc etc).

 

So the 7800 system's technology would have been a good choice I think given that it was probably cheaper to produce than the Atari 800 motherboard say. Add a pokey and ram and some Atari basic and away you go. I don't want to open up another pros/cons debate on the A8 though. And I don't know how much the 7800 would have cost to build in that form in 1983 either so hard to say for sure. But for classic 80s arcade games having 16 colours in 160x200 with a much more flexible sprite system would have served Atari well for porting classic games.

 

Think of it as if the Atari 800 was Americas version of the UK's Acorn BBC Micro then the 7800 system would be the US equivalent of the Acorn Electron (cut down/ half the price/ not 100% compatible without some effort by the software houses)

[Gorf]

Hmmmmm.....

 

I assume you are talking 6502 vs 68k at there respective clock speeds?

 

I'd say the 68k will win in math like multiply and divide but that little 8

bit 6502 probably has better gains in indexing...again... If it were possible

to have a 6502 at the same 8-16 MHZ of a typical 68k, I think you'd have

to break things down to category. At the same clock, the 6502 may have

an edge since the cycle efficiency per instruction is like tht of RISC but

you have the advantage of large register banks, multiplies and divides

on the 68k even if they take a bunch of cycles more. I bet over all the 6502

at the same clock would beat the 68k in a good deal of things but the 68k

would show its strengths where mult/div math and a lot of registers would

come in handy. The addressing modes of the 68k are very nice even if cycle

greedy.

 

 

NES vs 7800

 

So far, only coding the 7800, for what little I've learned I am quite impressed

with the possibilities of this machine.

 

Not having ever coded for an NES, I can only go by what I see on the screen.

The NES looks dull and seems like anything past a few sprites causes a lot

of flicker. However, I can't honestly judge the hardware by this. I'll have to

take the word of the veterans on this one.

 

The truth is, I just never had an interest in coding the NES. Perhaps I need to

take a look at it.

 

Having just figured out a nice method of creating a RAM buffer back drop

for the 7800(thanks to folk like PacMan Plus and Crazy Ace for invaluable

help), I think there is some interesting possibilities for more detail in 7800

games.

 

Of course you have to be mindful of all the graphical fetches that will be

required. Like just about every Atari console, coding is a balancing act.

 

My guess is this is not much different on the NES.

[Gorf]

I think the clever company (ie the one who would have escaped the crash) would have released the 7800 as a home computer.

 

For computing purposes, how would the 7800 be better than the Atari 8-bits? I suppose if one could have afforded the RAM, a 320x200x4-color screen mode might have been nice, but MARIA really seems better suited for gaming than computing.

 

Yes....Maria is a game graphics chip for sure. If it were used as a sub system

along side a nice high column text mode chip, it might have been pretty sweet.

[DracIsBack]

I don't see what makes the SMS any less childish than the NES.

 

I think it's a perception of the games from the parent company. Whereas NES would have stuff like Zelda, Mario and Kirby, Sega would bring out stuff like After Burner, Golden Axe etc.

 

Across 3rd parties though, not sure it the NES or SMS really gravitated to one audience or another. Lots of computer titles in Europe I suppose, but the NES also had a batch.

 

Regardless of when they were built, the important result is the SMS had a technical advantage when they went into

competition outside Japan in 1986/87.

 

It's not really that black and white as the NES also had technical advantages. Where the SMS shined was the number of colors it could display. It could have 16-color tiles (more than NES or 7800) and bright multicolor sprites (I believe 16-colors as well).

 

Now the resolution was a bit lower than NES and the sound wasn't quite as adept so it's not universal either.

 

Having said that, I'm off to fire up my SMS. It's been a whilte!

[atarian63]

I can't stand Zelda or Mario games...now tell me the NES had better games? I'd rather play Gauntlet Hang-On and R-type than most NES games. In fact I only have 2 NES games..Galaxian+Galaga Cart and Donkey Kong because they are both arcade perfect...(but then again so was Donkey Kong by Ocean software on the C64 for 1/6 the price)

The NES has games of those types as well, but of course not everybody prefers the NES library. Just most people.

I don't see what makes the SMS any less childish than the NES. Nintendo didn't get tagged that way until much later. Except maybe by the computer gamers, who looked down on consoles in general, not just Nintendo.

 

If the Famicom is a 1983 machine though the fact the SMS is technically better doesn't mean anything, the Amiga was only 3 years later than the C64 to put that into perspective.

Regardless of when they were built, the important result is the SMS had a technical advantage when they went into competition outside Japan in 1986/87.

The Amiga wasn't just 3 years newer, it also cost a fortune compared to the C64.

 

I would rather I walk into a shop and my only two choices of computer were still Commodore or Atari not the pieces of shit running Windows or piece of shit over priced Apple rubbish. How the hell we ended up with the two most pathetic designs as our only choice is just beyond me and is why computing is just zzzzzzzzzzzz

Probably because the real money was in computers for practical business use, not games. Thus graphics and sound didn't matter much, Commodore had questionable quality, and Atari was known as a game company.

C64 did well because it was cheap enough to buy for kids and they could play games on it, but it's hard to trust them for serious work, and the floppy drive was horrible. Maybe they should have released an upscale business model much earlier, comparable to IBM but cheaper.

 

Still, I think Commodore could have had more staying power if they only had been able to design a significantly more powerful computer that retained compatibility. Instead, C64 had a great run but became obsolete and disappeared. They didn't turn it into a lasting architecture.

in that line of thought all 8-bits would fall into lacking a lasting architecture.

[gdement]

Regardless of when they were built, the important result is the SMS had a technical advantage when they went into

competition outside Japan in 1986/87.

 

It's not really that black and white as the NES also had technical advantages. Where the SMS shined was the number of colors it could display. It could have 16-color tiles (more than NES or 7800) and bright multicolor sprites (I believe 16-colors as well).

 

Now the resolution was a bit lower than NES and the sound wasn't quite as adept so it's not universal either.

 

Having said that, I'm off to fire up my SMS. It's been a whilte!

 

Naturally none of the 3 systems was universally superior, but the SMS had the overall technical advantage IMO. It isn't huge though, and the sound does suck, at least subjectively if not technically. I might prefer the buzzing TIA over that annoying ding dong bell in the SMS.

 

The difference in resolution is negligible. If I remember right there's only a slight difference in the vertical, which makes me wonder if there's really any difference at all in what actually fits on the TV. Even if it's truly different, it's so small as to be meaningless.

 

Besides the graphics, SMS also has a somewhat faster processor and twice as much built-in RAM.

[gdement]

in that line of thought all 8-bits would fall into lacking a lasting architecture.

True, and I think it's odd it happened that way with companies that had an established 8-bit userbase.

The Apple IIGS is the only exception I can think of, and they didn't even put much emphasis on that machine.

 

Commodore should have been in the best position to design a compatible upgrade, since they owned MOS and could have designed whatever CPU or other chips they needed. Not sure when the 65C816 came out but it obviously wasn't good enough, seeing how almost nobody used it despite being the clear choice for compatibility. Did MOS just get no funding from Commodore, what happened to them? Such a 1-hit wonder.

[CV Gus]

A problem I had here with Vigo's posts earlier can best be described by the CV Pac-Man Collection.

 

Since a CV can only show one color/sprite, to have a 2-color character (assuming you cannot combine a sprite with bit-mapping or custom characters) you will need 2 sprites.

 

The problem is this: if more than 4 sprites line up left-right (why do I keep thinking it's 3?), you will get flickering and disappearing. This problem can be seen in Burgertime.

 

Since "stacking" (look at the enemy soldiers in Front Line) cannot be used in the Pac-Man games, in order to have a 2-color ghost you will need 2 sprites. Therefore, anytime any 3 characters line up, there must be some nasty flickering. Figure in Ms. Pac-Man, and the problem is worse, since it could then be up to 11 or 12 (depending on how the prizes look).

 

Therefore, if you just look at the stats of the CV on paper, you could only conclude that you'd best stick with single-color ghosts and prizes (for Ms. Pac-Man). Otherwise, you'd get an intolerable amount of flickering.

 

But the CV moves fast enough for the "flicker/alternating" trick to work. Thus, you CAN have multi-colored characters in the game, with occaisional minor flickering. It's actually less than what one might guess from a normally-programmed SINGLE-colored version. The sprite-specs made it impossible, but the TOTAL abilities of the CV made it quite possible.

 

 

The point I'm making here is this: it is not enough to simply keep repeating some tech-specs about a system, since such may not fully represent a system's abilities. If I went back in time and described Opcode's CV Ms. Pac-Man itself to most people, even programmers (at Coleco, no less!), they would've told me that it couldn't be done. But it has been done.

 

 

So it isn't likely we know what the 7800 could have really done. Vigo, you kept describing the different way a 7800 puts images on the screen as opposed to the NES, but all that meant was that the 7800 had a different way of doing it. What if a 7800 has such an ability to put and change objects on screen that it could handle scrolling backgrounds to match an NES? The key question is this: what limitations- including a similar "flicker" method- does the 7800 have? If it can move fast enough, then maybe it could overcome the limitations you pointed out with Sirius. What can a 7800 do if ALL of its abilities are brought into play, used in a manner not normally considered, as Opcode did?

 

The NES had the overwhelming advantage of a top-notch team of developers who never stopped pushing the limits of the NES. As did the 2600. But if the NES was backed by a group of 800-pound gorillas, the 7800 seemed to have been stuck with a few lazy monkeys. Until a 7800 Opcode appears, we may never really know.

[Vigo]

So it isn't likely we know what the 7800 could have really done. Vigo, you kept describing the different way a 7800 puts images on the screen as opposed to the NES, but all that meant was that the 7800 had a different way of doing it.

 

Look, what are you trying to do here by warming up the same tiresome discussions over and over again? Your relativism doesn't make it any better. Obviously you somehow have gotten by now that the 7800 and the NES use VERY different methods of handling graphics. But that fact alone is no single way of comparing both machines. Nor indicating they achieve the same results with different methods. They are very different, and have advantages and disadvantages in almost the opposite kind of fields.

 

That's why, again, it is more than stupid to keep this discussion alive if the 7800 could have done NES games equally well. No, it can't. It can do other types of games better, where lots of objects are moving in front of a simple background.

 

What if a 7800 has such an ability to put and change objects on screen that it could handle scrolling backgrounds to match an NES?

 

"What if" discussions are pretty much irrelevant if they describe non-existing hardware. If you want the 7800 concept of displaying graphics pushed to the extreme, refer to the Atari Jaguar. Unlike the 7800, the Jaguar has enough horsepower to handle this kind of flexible grqaphics architecture efficiently.

 

The key question is this: what limitations- including a similar "flicker" method- does the 7800 have?

 

And obviously, you still haven't gotten the fact that displaying many (and huge) objects is NOT a weakness of the 7800. No amount of flicker will overcome the fact that the 7800 isn't able to display a fully tile based scrolling background with individual color attributes for each block. Yeah, of course let's make the whole screen flicker and enjoy seizures.

 

I am repeating myself endless of times, and yet, I still have the feeling I could very well be talking to the kitchen oven with the same results...

 

If it can move fast enough, then maybe it could overcome the limitations you pointed out with Sirius. What can a 7800 do if ALL of its abilities are brought into play, used in a manner not normally considered, as Opcode did?

 

You make again and again and again the mistake of dreaming a dream where the 7800 is a NES. No, it is not.

 

The NES had the overwhelming advantage of a top-notch team of developers who never stopped pushing the limits of the NES. As did the 2600. But if the NES was backed by a group of 800-pound gorillas, the 7800 seemed to have been stuck with a few lazy monkeys. Until a 7800 Opcode appears, we may never really know.

 

Want a banana?

Edited January 26, 2009 by Vigo

[GaryH917]

So it isn't likely we know what the 7800 could have really done. Vigo, you kept describing the different way a 7800 puts images on the screen as opposed to the NES, but all that meant was that the 7800 had a different way of doing it.

 

Look, what are you trying to do here by warming up the same tiresome discussions over and over again? Your relativism doesn't make it any better. Obviously you somehow have gotten by now that the 7800 and the NES use VERY different methods of handling graphics. But that fact alone is no single way of comparing both machines. Nor indicating they achieve the same results with different methods. They are very different, and have advantages and disadvantages in almost the opposite kind of fields.

 

That's why, again, it is more than stupid to keep this discussion alive if the 7800 could have done NES games equally well. No, it can't. It can do other types of games better, where lots of objects are moving in front of a simple background.

 

What if a 7800 has such an ability to put and change objects on screen that it could handle scrolling backgrounds to match an NES?

 

"What if" discussions are pretty much irrelevant if they describe non-existing hardware. If you want the 7800 concept of displaying graphics pushed to the extreme, refer to the Atari Jaguar. Unlike the 7800, the Jaguar has enough horsepower to handle this kind of flexible grqaphics architecture efficiently.

 

The key question is this: what limitations- including a similar "flicker" method- does the 7800 have?

 

And obviously, you still haven't gotten the fact that displaying many (and huge) objects is NOT a weakness of the 7800. No amount of flicker will overcome the fact that the 7800 isn't able to display a fully tile based scrolling background with individual color attributes for each block. Yeah, of course let's make the whole screen flicker and enjoy seizures.

 

I am repeating myself endless of times, and yet, I still have the feeling I could very well be talking to the kitchen oven with the same results...

 

If it can move fast enough, then maybe it could overcome the limitations you pointed out with Sirius. What can a 7800 do if ALL of its abilities are brought into play, used in a manner not normally considered, as Opcode did?

 

You make again and again and again the mistake of dreaming a dream where the 7800 is a NES. No, it is not.

 

The NES had the overwhelming advantage of a top-notch team of developers who never stopped pushing the limits of the NES. As did the 2600. But if the NES was backed by a group of 800-pound gorillas, the 7800 seemed to have been stuck with a few lazy monkeys. Until a 7800 Opcode appears, we may never really know.

 

Want a banana?

[DracIsBack]

Look, what are you trying to do here by warming up the same tiresome discussions over and over again?

 

I'm a bit lost myself. It's been a week since he's started up the approach of "7800 vs. Colecovision" in a new thread. I suppose we should look for it shortly, complete with lengthy - and historically innacurate details - of the looming 7800 Colecovision showdown.

 

 

They are very different, and have advantages and disadvantages in almost the opposite kind of fields.

 

There was a quote I saw here recently ... VDUB Bobby's. 7800: Strong on sprites weak on tiles. NES: weak on sprites, strong on tiles.

 

I am repeating myself endless of times, and yet, I still have the feeling I could very well be talking to the kitchen oven with the same results...

 

You're not the only one. Maybe we should start a support group.

 

 

 

The NES had the overwhelming advantage of a top-notch team of developers who never stopped pushing the limits of the NES.

 

This quote of his I actually agree with. This isn't getting into how well one equals the other, but how well the talent actually tried to push the architecture.

 

 

Don't get drawn in Vigo. He won't go away ...

[xZanU]

The NES had the overwhelming advantage of a top-notch team of developers who never stopped pushing the limits of the NES. As did the 2600. But if the NES was backed by a group of 800-pound gorillas, the 7800 seemed to have been stuck with a few lazy monkeys. Until a 7800 Opcode appears, we may never really know.

As entertaining as it is true. The rest of the stuff is a bit over my head : )

[Vigo]

Naturally none of the 3 systems was universally superior, but the SMS had the overall technical advantage IMO.

 

You can easily create games on the SMS which will look worse on both NES and 7800.

 

It isn't huge though, and the sound does suck, at least subjectively if not technically. I might prefer the buzzing TIA over that annoying ding dong bell in the SMS.

 

The SN 76489 is nothing more than a 3 channel noise/square wave generator. Very boring stuff. It's a shame they did not put in the YM2419 FM chip in all SMS, not just the later japanese ones.

 

The difference in resolution is negligible. If I remember right there's only a slight difference in the vertical, which makes me wonder if there's really any difference at all in what actually fits on the TV. Even if it's truly different, it's so small as to be meaningless.

 

Agreed.

 

Besides the graphics, SMS also has a somewhat faster processor and twice as much built-in RAM.

 

A 3.58 Mhz Z80 is outperformed by a 1.78Mhz 6502 in almost all cases.