Face-Off: ZX Spectrum vs. Commodore 64
I found an old oldish, but very informed and interesting article on the ole Spectrum vs. Commodore 64 argument.
The article explains the technical difficulties associated with programming on both machines and there are some really interesting view points from several industry figures too.
http://www.eurogamer.net/articles/digitalfoundry-face-off-zx-spectrum-vs-commodore-64
The article explains the technical difficulties associated with programming on both machines and there are some really interesting view points from several industry figures too.
http://www.eurogamer.net/articles/digitalfoundry-face-off-zx-spectrum-vs-commodore-64
Post edited by sj_howlett on
Comments
I'm surprised by that. I thought it was quite balanced. I found the parts where programmers of both systems explained the challenges associated with the systems was really interesting.
That's exactly my opinion, from the time when I first read this article.
If this text has incorrect facts or complete lies, then they should be accurately labeled and explained.
I think it definitely shouldn't be ignored, its a good article very nicely written, without much passion, pretty objective and fair.
http://zx-pk.ru/
I really dont think the cost of the c64 compared to speccy is justified at all though.
It's definitely not 3x better. Maybe 1.3x (if the game is a scroller)
I only skimmed it but it acknowledges the Speccy was better for isometric, massive sprite, and wireframe games.
At launch spectrum cost 175 GBP, about 300 US dollars.
C64 has cost about 350 GBP (595$)
These are the prices at launch, only a year later, Commodore drops the retail price twice, first at $400, and then to $230.
So, C64 has never cost three times more, even at launch and later the difference was even smaller.
All this of course, has nothing to do with this article where spectrum isnt presented as an inferior machine..
http://zx-pk.ru/
I.e. everything but games with lots of small sprites and scrolling.
To give the C64 its due, it got the S.C.U.M.M. games (Maniac Mansion and Zak McKraken), disk based RPGs (notably most of the Ultima series), Pirates!, Prince of Persia, and M.U.L.E, none of which ever made it to the Speccy (not counting unlicensed Russian versions).
But could have done as the russians showed ;)
The only thing missing is the SCUMM system-based games, we have played around a lot with the idea here but in the end nobody takes the first step, although it would be a hugeee Project :)
There are some SCUMM-like games for the Pentagon, along with disk-based RPGs. The thing is, the Pentagon has a standard disk system. That's really what set the C64 apart from the Speccy.
New projects for the Speccy with that kind of depth are almost unheard of these days. The only one I can think of is Los Amores de Brunilda, which is probably the best JRPG ever written for the 128.
Can you really say that the C64 had superior hardware when on average the Spectrum performed tasks at twice the speed of the C64? Are you not letting the lure of the support chips cloud your judgement? Admittedly, the C64 did have a glorious sound chip. But which are better - apples or oranges?
Both were and still are great machines, although I consider myself extremely fortunate to have learned to program on the Spectrum, it was and is a genuinely blank canvas whereas the C64 tended to corral the developer more. I don't know how other developers feel, but I don't know if I'd have had quite the same experience had I grown up with the C64 instead. Would the C64 have made me a better programmer or game designer? Probably not.
Keep it quiet, but I think those of us who grew up programming the Sinclair machine had an unfair advantage. ;)
Egghead Website
Arcade Game Designer
My itch.io page
same goes with gfx in such games as magnetic scrolls, again due to the gfx modes that the atari / c64 could do.
I do think however some games suffer from sloppy conversion, that simple.. I can list many offenders of very spectrum centric companies just porting something that was sub standard.
an interesting one though, hall of the things.. like for like spectrum port that has no sprites in game (which is highly unusual for a c64 game) .. but scored high in zzap solely down to game play.
...to the MC68K based Atari ST and Amiga, and the x86 based PC in 1988?
Indeed. Lack of a standard disk system meant that all but one of the Magnetic Scrolls games were +3 only, and the Spectrum got two cRPGs of note (Ultima rip-off Ring of Darkness and Origin's Times of Lore).
I think the best games were ones that were written specifically for a given system. I can't think of one original Spectrum game that was improved by porting it to the C64. Same goes for Atari ST games ported to the Amiga.
Agreed. This is the bottom line.
and I'm sure as you are aware.. there are different revisions of the engine.. hence why certain games were on amiga / st and NOT c64/apple or atari.. and partly down to the market as the US had embraced the amiga quickly.
monkey island being the ones that comes to mind for starters.
For me, the key passage is where Gary Liddon explains that innovation in 3D gaming came more readily on the Spectrum, because its lack of video features ensured developers had a "blank page" to work with, and were not led down a certain path.
Quote: "If the hardware isn't heavily suggesting side-scrolling 2D games then you may as well try anything, and I think that's a pretty good reason why the Spectrum was the birthplace of that kind of gaming."
Or, if the hardware cannot do something (scrolling, sprites, colour), then you are somewhat forced to do something else, not try anything (because anything is not possible)
I don't think it is fair to say the spectrum was a "blank page" in this regard. It was a very small page with not many colour crayons to play with. That may have well sparked creativity in certain directions (what can you fit on this page?), but of course, that's different to having a truly blank page where only your imagination is the limiting factor.
There were some truly imaginative games on the C64 and NES that could never have been realised on the spectrum for example (and vice-versa of course) - and maybe programmers where not even looking down those avenues, because they were closed.
Having options allows the imagination to roam: I think the classic example is Ocarina Of Time on the N64, an imaginative experience that had to wait for the technology. In that example, the hardware allowed a game world, and so a type of gameplay, that was hitherto impossible
Even on today's consoles, because of the hardware, there are games that could not have even been imagined before. And this will no doubt continue.
Obviously, the C64 and Spectrum were produced to very different budgets, but it was pretty much the central dilemma of home computer design in the 80s. Do you invest in a sophisticated GPU with accelerated graphics features, at the expense of having to take a cheaper CPU, or do you go for a simple graphics solution and the best CPU you can get for the money?
I don't think there's a simple answer to that. On the one side, you've got machines like the Apple II, Spectrum, Amstrad CPC, BBC Micro, Atari ST and pretty much all PC compatibles right up until the 3DFX era. On the other, you've got the Atari 8-bits, C64, MSX, Amiga, and most consoles. They're all machines capable of great games, after all.
Straight-up factual errors? Uridium on the Speccy runs at 25fps, not 20. Joffa's scrolls use 2-pixel pre-shifting, not 1, and because of the simplicity of the backgrounds take up very little extra memory. It's pre-shifted sprites that hog the most memory.
They trivialise the processing speed differences too; it may not be a straight 4:1 comparison but it's still nearer 2:1 than even.
They do point out that the C64 hi-res doesn't look good as it's too high a resolution for PAL TVs to display accurately, but then what they fail to mention is the C64 was designed for NTSC, which could barely show a lo-res image properly. And then saying the Spectrum's screen memory layout gave it an advantage shows a little naivity as they've obviously never had to try and calculate a screen address.
As for criticising the Spectrum palette compared to the C64; every time a retro magazine prints a page full of C64 screenshots they look like the printer is missing one of its colour inks.
And on Paradroid, it's not that difficult a game to port, even with scrolling. The backgrounds are hardly the greatest works of art. But if they'd released anything looking that bland on the Spectrum by the time they got round to it, no-one would have given it a second glance. It needed the isometric view of Quazatron just to get noticed.
As for saying you couldn't do Encounter on a Spectrum - why would you want to? Like Quazatron, it could be done a lot better, using the likes of the scaling object routines from Space Harrier or Savage. Like Glass, for example.
- IONIAN-GAMES.com -
err, no.
You can go so far with software, if the computer is powerful enough (the concept of a Turing Machine).
But to say, you just do it all in software (on a spectrum), is absurd; you are not going to get C64 Turrican (for example) on a spectrum. Not going to happen. Nor Uridium - yes it was decent on the spectrum, but you are not going to get something as good as what can be done on the C64.
If it could "all be done in software", it would be. The multi-coloured graphics you mention are a programming marvel, but there are severe hardware limits to what can be done.
hmm, it had a natural colour palette.
Excellent for creating natural environments and skin tones. Brown and green trees, flesh coloured people - thats what the world looks like.
And it wasn't that difficult to turn the colour up on the TV if you wanted Spectrum colours.
Of the things the spectrum was better at than the C64, colours was not one of them.
- IONIAN-GAMES.com -
The critical difference is that the Spectrum screen is directly addressable, and the C64 one isn't. Despite the non-linear layout, that still gives you a huge advantage when doing things like wireframe 3D.
"A Z80's bigger register set and relatively efficient 16-bit math probably helped quite a bit for this sort of programming." - see, it's all speculative, it's not from experience.
- IONIAN-GAMES.com -
That's not quite right. The C64's video memory is directly addressable by the CPU. The main reason why it struggles with wireframe 3D is that the CPU is too slow.
If you're talking about machines like the MSX and NES though, where it's all got to be shoved through a port to get it into video memory, such games become a nightmare. That said, they still got a few good ones.
Well, obviously you're not necessarily going to be able to port specific games to the Spectrum. As has already been said, most of the best games on 8-bit systems are very heavily optimized for their hardware. Indeed, despite the fact that the MSX and NES support hardware sprites and scrolling, I'd struggle to think of too many C64 games that were successfully ported to them either.
So, unsurprisingly, to find games that use such features well on the Spectrum, you've got to look at the games that were designed with the machine in mind. Then you can see how you go at porting any of them to the C64, and see how useful its features are when you're doing that.
Elite on the NES was a fantastic port.
Even in the hi-res graphics modes the screen is directly addressable by the CPU. It's just that you've got a lot less cycles per frame to draw them in, a CPU that's not terribly efficient at 16-bit maths, and a screen layout less conducive to calculation of co-ordinates by bit-bashing.
That said, there are still some pretty decent games. Mercenary would be the obvious one, but it also got ports of things like Space Rogue and Battle Command later in its life.
Yes, but I think we can safely say that it was made in spite of the machine's hardware capabilities rather than facilitated by them. It came out at around the same time as Elite Plus on the PC too.
I think this sums up C64 folk fairly well. On the C64 you have to use the hw to accomplish what it did -- there is little choice because the processor is too slow to adequately replace what the hw does. Because of that I think those with little experience outside the C64 have a hard time understanding that substitution can occur. You can replace dedicated hw with general cpu and vice versa.
With a fixed target price, you might choose to emphasize dedicated graphics hw over cpu if games is what you want to focus on. The question is how much an advantage that will get you for games. The sky is not the limit -- dedicated graphics have to stay within a development & BOM budget, be on time to market and have a limit in number of transistors according to the technology of the day. And after looking at the spectrum sw catalogue, I think the answer is that the dedicated hw the C64 got didn't give as much an advantage as what the original designers might have thought.
One advantage it did get is mainly anything to do with scrolling. The spectrum is not fast enough to do the equivalent. 50fps is nice but it is not necessary for a smooth scroll and when the spectrum attains 25fps there are restrictions on what can appear in the background, which the C64 hw does not impose. Whether that manifests in the game depends on the game (see Turrican - yes - and Green Beret - no).
Another is increased colour resolution. The cpu speed to display file size ratio the spectrum got is probably in some sort of sweet spot. If you added higher colour resolution the display file size increases and what you can accomplish reduces. If you decrease pixel resolution and increase colour resolution such that the ratio stays the same, you lose on features not related to gaming (ie the spectrum's intended market).
All you have to do is look at the game catalogues of both machines to see where the C64 game hw allowed games to be noticeably better to see what that gaming hw got the machine. It works the other way too. In game types that do not involve 2d scrolling and sprites cannot directly apply, the C64 suffers in comparison.
Just having game hw does not automatically make a better game machine. You need to look at the actual games to judge that. The one thing I have noticed is that many five minute reviewers place the highest importance on colour resolution. It looks good, that's all that matters. How it plays, how smooth it is, if it's a good game are all secondary.
Write games in C using Z88DK and SP1