Assembly Summer 2017 oldskool platform lottery
category: parties [glöplog]
It's time for the annual oldskool platform lottery. The sooner, the better.
https://www.youtube.com/watch?v=AUQdWKMcp04&t=15s
I wonder if we could have the same oldskool PC spec two years in a row. :)
Personally, I'd like to keep this year's 1997 setup as-is. As opposed to, let's say, downgrading to 486 or something. Well maybe downgrade the ET-6000 to a run-of-the-mill S3 or something, because that would be more in line with the actual 1997 specs.
Here's my reasoning
* Making changes every year makes things worse.
* Like always, this year I started too late and ended up spending most of my time setting up a working dev/test system, and I got parts for a slow Pentium system. I'd like to use that instead of throwing it away.
* 1997 is oldskool enough. That was 20 years ago.
* Some stuff in the recent Assembly 2016 oldskool demo compo would have needed more than 133 MHz, not less.
* Do you remember back in the day, you could always trust that next year the compo PC will be at least equally powerful. If you make a good effect or framework or something, it will be usable next year.
* Claiming that Pentium 133 is too powerful compared to Amiga 060 ... well that was the situation in 1997.
* For me personally, the Pentium 100-150 class DOS PC is the finest oldskool PC, and the one I have the fondest memories of. Lots of power, but with Amiga-style low-level control. No Windows with its myriads of drivers and abstraction layers, let alone 3D acceleration. In other words, it's the finest oldskool PC platform, the way I see it.
To put the question the other way around, what BAD might happen if the oldskool platform just stays the same?
Any ideas? Comments?
https://www.youtube.com/watch?v=AUQdWKMcp04&t=15s
I wonder if we could have the same oldskool PC spec two years in a row. :)
Personally, I'd like to keep this year's 1997 setup as-is. As opposed to, let's say, downgrading to 486 or something. Well maybe downgrade the ET-6000 to a run-of-the-mill S3 or something, because that would be more in line with the actual 1997 specs.
Here's my reasoning
* Making changes every year makes things worse.
* Like always, this year I started too late and ended up spending most of my time setting up a working dev/test system, and I got parts for a slow Pentium system. I'd like to use that instead of throwing it away.
* 1997 is oldskool enough. That was 20 years ago.
* Some stuff in the recent Assembly 2016 oldskool demo compo would have needed more than 133 MHz, not less.
* Do you remember back in the day, you could always trust that next year the compo PC will be at least equally powerful. If you make a good effect or framework or something, it will be usable next year.
* Claiming that Pentium 133 is too powerful compared to Amiga 060 ... well that was the situation in 1997.
* For me personally, the Pentium 100-150 class DOS PC is the finest oldskool PC, and the one I have the fondest memories of. Lots of power, but with Amiga-style low-level control. No Windows with its myriads of drivers and abstraction layers, let alone 3D acceleration. In other words, it's the finest oldskool PC platform, the way I see it.
To put the question the other way around, what BAD might happen if the oldskool platform just stays the same?
Any ideas? Comments?
Sounds like an good idea. Thumb!
I'm fairly sick today but throw a quick comment or two.
Let me play the devil's advocate.
Pentium 133 is a lot faster than 68060 66Mhz. GUS can sound better than Paula. Super VGA is more flexible than AGA. And to add to that PC platform has way more memory than Amiga. Isn't the compo fairly unfair once the PC guys actually build a decent engines and get their design to the same level as the current amiga demos?
Yeah I get that yeah P133 and AGA are the platforms that were the main platforms then but putting them against each other can be unfair. Wouldn't bringing the platforms more closer to the same specs make more sense in competition wise?
Now that's not a terribly serious issue for me and I'd be happy to stick to the current rule set for now but that is a concern of my especially in the long
How about in addition to that allowing platforms that are kind of similar by asking compo organizers if they think they fit the compo in spec and era, like some old macs or consoles like 3DO or Jaguar just to spice things up. But then there is the risk that it requires way too much research time from the organizers to make the call and risks allowing hard to judge platforms in the compo since ~nobody in the audience knows their spec well enough to know if they are good or not.
Let me play the devil's advocate.
Pentium 133 is a lot faster than 68060 66Mhz. GUS can sound better than Paula. Super VGA is more flexible than AGA. And to add to that PC platform has way more memory than Amiga. Isn't the compo fairly unfair once the PC guys actually build a decent engines and get their design to the same level as the current amiga demos?
Yeah I get that yeah P133 and AGA are the platforms that were the main platforms then but putting them against each other can be unfair. Wouldn't bringing the platforms more closer to the same specs make more sense in competition wise?
Now that's not a terribly serious issue for me and I'd be happy to stick to the current rule set for now but that is a concern of my especially in the long
How about in addition to that allowing platforms that are kind of similar by asking compo organizers if they think they fit the compo in spec and era, like some old macs or consoles like 3DO or Jaguar just to spice things up. But then there is the risk that it requires way too much research time from the organizers to make the call and risks allowing hard to judge platforms in the compo since ~nobody in the audience knows their spec well enough to know if they are good or not.
Quote:
Yeah I get that yeah P133 and AGA are the platforms that were the main platforms then but putting them against each other can be unfair.
Yes, I think it's unfair that just by mentioning that btw, this is an amiga demo you gain points =)
Yeah, once the PC guys just catch up with the Amiga guys who had 20 more years to polish their engines… Totally unfair, I'd say.
What are all these "engines" that people supposedly polish for years?
Seriously, with the exception of 2, maybe 3 groups, people doing Amiga/060 stuff have engines at the level of "get me a framebuffer and play some music".
Just be honest and admit that it takes a bit of time to build your *skills*, without trying to camouflage it as some sort of arms race.
Seriously, with the exception of 2, maybe 3 groups, people doing Amiga/060 stuff have engines at the level of "get me a framebuffer and play some music".
Just be honest and admit that it takes a bit of time to build your *skills*, without trying to camouflage it as some sort of arms race.
On "engines". I think that stuff made with an "engine" tends to be boring. Demos should be coded, not made with a demo maker.
On "fairness". It's not fair that Amiga guys have so many registers. So rules should only allow using d0-d3 and a0-a3 on the m68k.
Branch: are you speaking as an Amiga coder yourself, or in behalf of some Amiga folks who feel that the scene was unfair in 1997?
On "fairness". It's not fair that Amiga guys have so many registers. So rules should only allow using d0-d3 and a0-a3 on the m68k.
Branch: are you speaking as an Amiga coder yourself, or in behalf of some Amiga folks who feel that the scene was unfair in 1997?
what is this shit? a proper amiga demo runs on a500/68000 anyway.
To be honest, after watching the current state of DOS demos I don't think the Amiga guys have much to be worried about. Using compo machine specs from 20 years ago makes a lot of sense to me.
Yzi neither.I'm speculating about the future.
yzi: so, you will volunteer to handle the oldskool compo next year?
and in general, how about contacting the people who organize the compo FIRST instead of ranting on pouet/irc/whatever. you might actually get better results.
Sorry.
and there's nothing oldskool about pentium. the entire 95-98 PC period can confirm it.
Quote:
Pentium 133 is a lot faster than 68060 66Mhz. GUS can sound better than Paula. Super VGA is more flexible than AGA. And to add to that PC platform has way more memory than Amiga.
This is my view as well. I was always under the impression, then and now, that a 486/66 with GUS and a slow video card was roughly equivalent to '060+AGA. Even though I'm a PC guy, it seems pretty unfair that '060+AGA is competing with a P133. Pentiums have a PCI bus, fast floating-point instructions and wide floating-point registers, and two pipelines...
I leave the allowed platforms for Assembly 2017 to others..
But... Can there be an oldskool intro compo and an oldskool demo compo ?
Just my 2 cents..
But... Can there be an oldskool intro compo and an oldskool demo compo ?
Just my 2 cents..
magic: most likely not
I think the P133 platform was too 'newskool'. When I think of oldskool, I think of having to work inside tight limitations, and breaking those limitations with exceptional code.
A P133 allows you to get away with some simple C++ code doing standard chunky effects.
I guess that may also be the reason: Doing a 'proper' PC demo with tight code on an 8088, 286, 386 or even 486 requires a special mix of skills and experience, which probably limits the entries to 0 :)
A P133 allows you to get away with some simple C++ code doing standard chunky effects.
I guess that may also be the reason: Doing a 'proper' PC demo with tight code on an 8088, 286, 386 or even 486 requires a special mix of skills and experience, which probably limits the entries to 0 :)
Scali: are you saying that because everybody will be making just bog standard not-really-worth-watching chunky effects anyway, then at least it should be made hard to make the crap?
Quote:
Scali: are you saying that because everybody will be making just bog standard not-really-worth-watching chunky effects anyway, then at least it should be made hard to make the crap?
No, I'm saying that if you pick the right PC platform, you can't get away with making simple chunky effects, and you need to write REAL effects (pretty much like the Amiga 500).
Downside there is that that it takes a lot of time and knowledge to pull that off, and not many people in the scene are capable of doing that, or are interested in doing such a demo. So the chances of even one such demo being ready for Assembly are slim. Let alone enough demos to have a proper competition.
I guess I'm saying the oldskool PC platform is dead, unlike the Amiga.
Quote:
No, I'm saying that if you pick the right PC platform, you can't get away with making simple chunky effects, and you need to write REAL effects (pretty much like the Amiga 500).
For one very subjective definition of "real" effects.
In my opinion, 1996-1998 was the golden era of PC demoscene. I'm very interested in revisiting that, mostly from design point of view. I promise to write fast code though.
[I guess I'm saying the oldskool PC platform is dead, unlike the Amiga.]
... which of course isn't filled with bog-standard uvtable effects and slow chunky texturemappers...
Quote:
For one very subjective definition of "real" effects.
Effects as in using the hardware in ways the designers never imagined, to pull off things people didn't deem possible.
That's what demos historically are, so that's oldskool.
Quote:
... which of course isn't filled with bog-standard uvtable effects and slow chunky texturemappers...
Amiga 500 isn't, because well, the platform isn't really fast enough for such effects. If you pull off such effects on Amiga 500 anyway, you're doing something really clever, hence 'real effect'.
I would also prefer to have a slightly lesser PC in the compo, but unfortunately based on comments I've received that would result in even less entries. But then again, even back in the early 90's the PC-scene needed beefed up machines to even match A500 stuff, so it kind of goes with the oldskool spirit ;)
Quote:
I would also prefer to have a slightly lesser PC in the compo, but unfortunately based on comments I've received that would result in even less entries. But then again, even back in the early 90's the PC-scene needed beefed up machines to even match A500 stuff, so it kind of goes with the oldskool spirit ;)
Yup, PC demos didn't become serious competitors to Amiga demos until 1992-1993, when 486, VLB and GUS arrived.
There's a number of reasons for that, such as:
- PCs were much more expensive than C64 or Amiga, so not many sceners had a PC, let alone a high-end one.
- Even if you did have a PC, C64 or Amiga had more interesting hardware to code for, so why go PC?
- People didn't have any experience coding demos for PC yet, so they didn't really know how to make the best use of the hardware yet.
As a result, not many demos aimed at 8088, 286 or 386 exist. Personally I think it would be interesting if people explored these platforms more. They still have a lot of untapped potential.
Coming back to the Assembly oldskool rules and specs, one way to convince us to keep the current specs is to release a few awesome DOS-demos aimed at Pentium-class hardware :)