[CSDb] - User Forums - World Record Database

You are not logged in - nap

CSDb User Forums

Forums > C64 Coding > World Record Database

2023-11-03 03:48

NoiseEHC

Registered: Feb 2005
Posts: 51

World Record Database

Hi!

Is there anybody who knows if there is some kind of world record database? If there is none then are there people interested in creating one in a wiki? I am thinking about creating a paragraph of the effect description of "every" demo ever released (which have world records or new effects of course), and to group the effects into categories. For example "sprite multiplexers" would be a category, listing record numbers/years going up to Crest's 120/170 sprite records (each one a link to the respective demo).

Anybody interested?
Thanks,
Noise/EHC

... 33 posts hidden. Click here to view all posts....

2023-11-04 11:58

Krill

Registered: Apr 2002
Posts: 2982

Quoting NoiseEHC

Quoting Krill
And the advances are getting smaller and smaller the closer we asymptotically approach some hard theoretical maximum.
You are sooooooo wrong about that... ;)

Just for the record: there is still some $d011 stuff have not been done.

How does that contradict my assertion?

Quoting NoiseEHC

Probably I could have been more clear: by "World Record" I mean not only something which has a numerical value which is increasing in every iteration, but all "World First" effect, which have not been done before. I thought that it was quite obvious.

I must have missed the "or new effects of course" bit in the OP.

That would boil down to a database of general demo effects and their underlying technical details.
Who made them first isn't so relevant (except for the respective creators' egos) as the information that it has already been done at all.
Of course, such a database will never be complete, thus may always yield false negatives.

So in any case, whether or not your grand idea has been done before, why not just enjoy typing it in for an actual .PRG, then see how people react?
Just don't get carried away, claiming unproven facts ("Worldfirst!") in the scrolltext or so. =)

2023-11-04 13:11

Oswald

Registered: Apr 2002
Posts: 5095

there is no wrong or right way of doing a demo, there are only opinions. but I must agree the creativity is often lacking, the same effects are repeated ad nauseum. even in tier A demos.

2023-11-04 17:08

Frostbyte

Registered: Aug 2003
Posts: 183

I've been often pondering about such effects database. It'd be great for random "Wow, such many dots, but is this the all-time record or has anyone displayed even more dots at some point" wonderers, as well as people who enjoy (and have the skills for) breaking such records. :)

However due to the many factors (e.g. other effects on screen at the same time) that may affect the interpretation of what can be considered the best, imho a wiki would be one of the least useful places for it. To me such database would be all about attribution, and powerful search, filtering and ordering utilising it.

For example: If you want to buy a t-shirt from an online store, you don't just go to the store and search for t-shirt, and buy the only hit (or let alone google for "ACME Online Store t-shirt"). Instead, you may e.g. select a main category (men's clothing) and then filter down the t-shirts subcategory by selecting certain attributes or attribute ranges, e.g. size: L -> fit: regular or loose -> material: cotton -> colour: blue, and then sort the results by e.g. their rating, price etc. Or you may search for t-shirt and drill down from there.

As far as I can see, such demo effect database would only be useful if it would have enough metadata (attribution, properties, parameters, whatchamacallit) against the records to narrow down exactly what you're looking for.
As an example, "most sprites on screen at once" record(s) could be accompanied by several attributes:
- Number of sprites (integer)
- Crunching (boolean)
- Stretching (boolean)
- Top/bottom borders (boolean)
- Side borders (boolean)
- Framerate (float)
- Released (date)
etc etc.

Then people could search for (or navigate to using filters) exactly what they want, for example

- Most sprites
- Most crunched or stretched sprites
- Most sprites in all borders, sorted by no of sprites and framerate

Of course the possible combinations of attributes is nearly endless, but at least the attributes would be there and it would be up to you to determine if they're relevant for your upcoming record breaking case or not. You probably get my drift. :)

In order to actually implement this, I'm sure there are plenty of open source search engines that could build their index based on dynamic, typed attribution, as well as provide filtering and sorting dynamically, and maybe something suitable for the frontend too for both users and the ones who contribute. Not saying this wouldn't be a pretty large task though. And we haven't even talked about CSDB integration yet, as of course we'd want to do that. ;D

2023-11-05 02:54

NoiseEHC

Registered: Feb 2005
Posts: 51

Quoting Frostbyte

Of course the possible combinations of attributes is nearly endless, but at least the attributes would be there and it would be up to you to determine if they're relevant for your upcoming record breaking case or not.

I think you are overengineering it. In this I agree with Krill that there are too much variety to have meaningful categories. So we have to go with broad categories.

What you seem to overestimate is the number of entries. If we ignore filler parts, then my guess is that there are <1000 effects in <200 demos (considering only modern demos and not every raster+logo combo). So if you have 100 entries in a broad category, ordered by year then you can just use your brain to filter the rest.

It can be done with GitHub Pages and Jekyll blog -> HTML conversion. Because what is needed is an (blog) entry for all "serious" demos (with pictures and link to csdb), and tags/categories, which the blog engine already supports. Then you can filter to tag/category, which are the effects. So this can be done without any software development or anything. I will set up a repository and fill up the Graffity and Crest groups. Then anybody can contribute, hopefully every proud coder will write about their brilliant Word Record... :)

2023-11-05 02:59

NoiseEHC

Registered: Feb 2005
Posts: 51

Quoting Krill

How does that contradict my assertion?

It does not. These are just opinions, there is no One Real Truth (tm) which can decide whether one opinion is right and the other is wrong.

Quoting Krill

I must have missed the "or new effects of course" bit in the OP.

You have not missed anything, it was a bad word usage from my side. We used to call everything a World Record 3 decades ago, I did not know that there is an accepted word usage now which is different.

Quoting Krill

So in any case, whether or not your grand idea has been done before, why not just enjoy typing it in for an actual .PRG, then see how people react?

That would just destroy the best thing in demo making: surprise. The ???HOW??? factor.

2023-11-05 05:27

Krill

Registered: Apr 2002
Posts: 2982

Quoting NoiseEHC

Quoting Krill
How does that contradict my assertion?
It does not. These are just opinions, there is no One Real Truth (tm) which can decide whether one opinion is right and the other is wrong.

On that we agree. But what was i "sooooooo wrong about" then? :)

Quoting NoiseEHC

Quoting Krill
I must have missed the "or new effects of course" bit in the OP.
You have not missed anything, it was a bad word usage from my side. We used to call everything a World Record 3 decades ago, I did not know that there is an accepted word usage now which is different.

No idea whether scene-wide accepted standard or whatever, but to me "world record" and "world first" were always different things.
You can be the first to surpass someone else's record-holding effect, in which case both terms conflate,
but when coming up with an entirely new thing, you're first all right - but only technically hold the record, as the one and only contestant thus far. =)

Quoting NoiseEHC

Quoting Krill
So in any case, whether or not your grand idea has been done before, why not just enjoy typing it in for an actual .PRG, then see how people react?
That would just destroy the best thing in demo making: surprise. The ???HOW??? factor.

Not sure i understand. Destroy the surprise on whose side? The demo maker's? The audience's? And why would it?

If you drop some novel effect or technique just like that, why would there be no surprise or "htf does this work" with the audience?
And if in another case that effect turns out to be not so novel after all, perfectly surprising for the creator, no? :D

2023-11-05 05:28

Mr SQL

Registered: Feb 2023
Posts: 137

Quote: Quoting Krill
And the advances are getting smaller and smaller the closer we asymptotically approach some hard theoretical maximum.

You are sooooooo wrong about that... ;)

Just for the record: there is still some $d011 stuff have not been done.

Quoting Krill

And breaking some "world record" (previously held by another implementation of the same concept) is by definition the exact opposite of inventing a new effect.

Probably I could have been more clear: by "World Record" I mean not only something which has a numerical value which is increasing in every iteration, but all "World First" effect, which have not been done before. I thought that it was quite obvious.

I agree there are still plenty of discoveries waiting to be made, $d011 and beyond. Some intentional and some accidental by those willing to explore the old technology with a fresh perspective pushing the boundary of innovation and imagination.

Looking forward to seeing your idea developed. csdb seems like a good place to discuss regardless of the competition venue or itinerary you select.

2023-11-05 08:29

NoiseEHC

Registered: Feb 2005
Posts: 51

Quoting Krill

But what was i "sooooooo wrong about" then? :)

You are wrong on that we are nearing some limit. The demo JustinBlue was released this year, and it still had 3 (or 4 depending on who you ask) completely new effects from 30 years ago. And I have a ton of new ideas. Maybe coders just became lazy, as good design seems good enough?

Quoting Krill

Not sure i understand. Destroy the surprise on whose side? The demo maker's? The audience's? And why would it?

The audience's. And by "audience" I mean other coders who read these forums.

The coder's surprise about recreating a record will be hopefully fixed by the World Record Database... :)

2023-11-05 11:08

Frostbyte

Registered: Aug 2003
Posts: 183

Quote: Quoting Frostbyte
Of course the possible combinations of attributes is nearly endless, but at least the attributes would be there and it would be up to you to determine if they're relevant for your upcoming record breaking case or not.

I think you are overengineering it. In this I agree with Krill that there are too much variety to have meaningful categories. So we have to go with broad categories.

What you seem to overestimate is the number of entries. If we ignore filler parts, then my guess is that there are <1000 effects in <200 demos (considering only modern demos and not every raster+logo combo). So if you have 100 entries in a broad category, ordered by year then you can just use your brain to filter the rest.

It can be done with GitHub Pages and Jekyll blog -> HTML conversion. Because what is needed is an (blog) entry for all "serious" demos (with pictures and link to csdb), and tags/categories, which the blog engine already supports. Then you can filter to tag/category, which are the effects. So this can be done without any software development or anything. I will set up a repository and fill up the Graffity and Crest groups. Then anybody can contribute, hopefully every proud coder will write about their brilliant Word Record... :)

I do agree that such attribution would need some effort to implement (and take longer to enter the data, and require more discipline to maintain it), but at least some would be needed to make the top lists somewhat meaningful. A 1002 dot plotter isn't a world record if it is updated every 2nd frame and another 1001 dot plotter is updated every frame. :)

What I do disagree on is that we have to go with broad categories. You only need to do that if you don't have enough variables to categorise your entities on. E.g. a "fastest car" category would be only half the truth, as a fastest rally car and a fastest street car are two completely different things, both fastest in their own subset of cars.

Anyway, I don't have time to develop such database, and even if I would the scene would need to be fully behind it and collaboratively fill the data in and maintain it, so take everything what I just said as just a fellow scener's brainstorming session (or brainfarts session if you wish) rather than me trying to be the infamous Ideas Man who throws ideas at people and expects them to do the actual work. :)

2023-11-05 11:32

Krill

Registered: Apr 2002
Posts: 2982

Quoting NoiseEHC

You are wrong on that we are nearing some limit.

Disagree.

I didn't say that there aren't any new discoveries to be made, but that they are finite, thus ever fewer and farther between as we go along.
It's just a fact after dabbling for more than 40 years and counting on the same machine.

Quoting NoiseEHC

The demo JustinBlue was released this year, and it still had 3 (or 4 depending on who you ask) completely new effects from 30 years ago. And I have a ton of new ideas.

So? Fine! :)

Quoting NoiseEHC

Maybe coders just became lazy, as good design seems good enough?

Matter of taste.

Personally, i enjoy complex "math-based" routines more than early-90s-styled raster-only effects, while the underlying raster techniques of the latter are perfectly suited as building blocks to achieve this or that visual thing in the former.
(Not saying that there can't be an overlap between the two! Or that there are just the two, for that matter.)

Now, if you invent a new low-level rendering or sample replay technique, hats off to you, good sir!
But don't berate others for using it to create different styles (with entirely different audiovisual impact), refining it but without coming up with new basic techniques on their own. =)
It's a bit like fundamental research vs. applied engineering, or the now-forgotten difference between Technologie und Technik (sorry, that doesn't work so well in English).

Quoting NoiseEHC

Quoting Krill
Not sure i understand. Destroy the surprise on whose side? The demo maker's? The audience's? And why would it?
The audience's. And by "audience" I mean other coders who read these forums.

Still not sure i get it. We're talking about just coding whatever you have in mind and releasing it, without asking "Is this new" beforehand, yeah? Then no beans are spilled.

Previous - 1 | 2 | 3 | 4 | 5 - Next

Refresh

Subscribe to this thread: