| |
crl
Registered: Mar 2020
Posts: 19 |
The first demo to feature...
There are quite a few landmarks in the demo history of the C64.
One of groundbreaking moments for me was the invention of FLI, which first appeared in Beyond Force's Charlatan.
Does anyone know who cam first with the following inventions, and what other inventions are in your opinion groundbreaking?
* Open sideborders
* DYCP
* DYSP
* DYPP
* Hardware scrolling |
|
| |
cba
Registered: Apr 2002
Posts: 932 |
Check http://www.atlantis-prophecy.org/recollection/?load=world_of_de.. |
| |
anonym
Registered: Jan 2002
Posts: 247 |
Great link, CBA. I was still writing this while you posted it.
To go even further back in time:
* FLD
* open upper and lower boarder
* use of $3fff in border
* fast loaders
* loading with running IRQ
* samples (and the evolution of them)
I think there might have been some FLI precursors, presented as "raster splits" at the time. Would be interesting to see if one could trace the timeline of some of those effects, too. |
| |
Cupid
Registered: Jan 2002
Posts: 83 |
ESCOS was another fun one. |
| |
Cupid
Registered: Jan 2002
Posts: 83 |
Then there is of course Crest with a new image format in every demo that nobody else pixeled in :) |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
There are several threads about this here.
Also Check here: https://codebase64.org/doku.php?id=base:demo_world_records_and_.. |
| |
TheRyk
Registered: Mar 2009
Posts: 2070 |
tech-tech is imho far older than 1988 as cba's links states, ain't the idea as old as Yeti Demo 5 by YETI from 1986? |
| |
Compyx
Registered: Jan 2005
Posts: 631 |
This is all very open-ended. I'd love to see a proper list of firsts and updated versions, but it's really hard to get right.
For example: I'm pretty sure I did the first DYCP over a FLI logo, but we waited to release it and someone else got it out before us. And that's just with the info I had at the time and the demos I got, mostly via swapping.
Another thing would be 'fixed' plotters vs 'variable' plotters. With fixed/precalc code, you can get quite a few plots, using a joystick controlled plotter you'd get a lot less plots, but at least it's realtime.
Then there's the issue of what constitutes a DYSP, does it mean sideboder magic? To me it does, sprites, even multiplexed, not going into the border don't make a DYSP.
Also, does one call a 'DYCP' using sprites over a charset or bitmap a DYCP, or a DYSP, and if the sprites don't go into the border, do we still call it a DYSP or just a multiplexer ? |
| |
Smasher
Registered: Feb 2003
Posts: 512 |
I'll now open a new topic "The last demo to feature..." :) |
| |
crl
Registered: Mar 2020
Posts: 19 |
Quote: Check http://www.atlantis-prophecy.org/recollection/?load=world_of_de..
Exactly what I was looking for, but as some of you mention, there could be other entries. |
| |
crl
Registered: Mar 2020
Posts: 19 |
Quote: Great link, CBA. I was still writing this while you posted it.
To go even further back in time:
* FLD
* open upper and lower boarder
* use of $3fff in border
* fast loaders
* loading with running IRQ
* samples (and the evolution of them)
I think there might have been some FLI precursors, presented as "raster splits" at the time. Would be interesting to see if one could trace the timeline of some of those effects, too.
* use of $3fff in border
What is that? |
| |
Zyron
Registered: Jan 2002
Posts: 2381 |
The ghostbyte you see when you remove up/down border. |
| |
Compyx
Registered: Jan 2005
Posts: 631 |
He has a point, who was the first to discover the ghostbyte? And who was the first to actually use it, ie not setting it to 0 ?:wq |
| |
chatGPZ
Registered: Dec 2001
Posts: 11114 |
i made the first knot raster. you can hardly beat that, ever. |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Quote: i made the first knot raster. you can hardly beat that, ever.
Indeed. Where is the line. Everything I code I always aim for some world’s first, like 50fps chess rotozoomer, portal engine, skew scrollers etc. But most often it’s just an optimisation of previous work or attacking it from another angle etc. |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Now that the pixel pipeline in VICE seems 100% accurate one could actually theoretically determine all VIC-tricks. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11114 |
Except that its not :P |
| |
Raistlin
Registered: Mar 2007
Posts: 557 |
Quote: i made the first knot raster. you can hardly beat that, ever.
First 50fps 304x192px (visible) 3-screen diagonal bitmap scroller (Delirious 11) ;-)
As JackAsser says, though, all these things I don’t see that we can claim - without having 10million different categories. |
| |
hedning
Registered: Mar 2009
Posts: 4595 |
Quote: Except that its not :P
:'( |
| |
Raistlin
Registered: Mar 2007
Posts: 557 |
Quote: Except that its not :P
It’s pretty damn close if it’s not. I code all my demos on PC in emulator and don’t believe I’ve seen an issue yet where something looked different on final testing on a real C64. |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Quote: Except that its not :P
Hehe ofc not! Would be nice to have a proper die-reverse engineering of it. Also the internal 40*12 bit memory and how that is updated needs proper explanatiom. |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Quote: It’s pretty damn close if it’s not. I code all my demos on PC in emulator and don’t believe I’ve seen an issue yet where something looked different on final testing on a real C64.
The pixel pipeline is what makes some fx 4px delayed etc. It’s notoriously hard to spot any differences and even harder to exploit it. Crossbow did however in his almost general upscroller in the lower border. Also effects like 49px or was it 50px wide sprites are subject to this etc. |
| |
tlr
Registered: Sep 2003
Posts: 1714 |
Quote: Hehe ofc not! Would be nice to have a proper die-reverse engineering of it. Also the internal 40*12 bit memory and how that is updated needs proper explanatiom.
That would be interesting, but a visual6502 style die-reversing isn't enough. A lot of the effects have to do with drive strengths of individual transistors and the fact that NMOS drives asymetrically (pull-ups). |
| |
The Human Code Machine
Registered: Sep 2005
Posts: 110 |
Quote: First 50fps 304x192px (visible) 3-screen diagonal bitmap scroller (Delirious 11) ;-)
As JackAsser says, though, all these things I don’t see that we can claim - without having 10million different categories.
Did that 2 years earlier with streaming data from disk in Concert by Performers. I did scroll the black borders too. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11114 |
Quote:It’s pretty damn close if it’s not. I code all my demos on PC in emulator and don’t believe I’ve seen an issue yet where something looked different on final testing on a real C64.
the things that are not perfect yet probably do not really matter to any real world code. and you see related differences between C64s (VICIIs) too. |
| |
Raistlin
Registered: Mar 2007
Posts: 557 |
Quote: Did that 2 years earlier with streaming data from disk in Concert by Performers. I did scroll the black borders too.
Ah, I actually thought that one was doing the "stepping" method of scrolling - 1 pixel horizontally one frame and 1 pixel diagonally the next? So that the "diagonal is aligned to the double pixels...
The harder hurdle for full 312x192 diagonal scrolling with infinite repeats is of course squeezing it all into memory. |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Infinite circular scroller, say no more! ;) |
| |
Krill
Registered: Apr 2002
Posts: 2839 |
Quoting JackAsserWhere is the line. Everything I code I always aim for some world’s first I'd say the line is between discovering/inventing (hardware-centric) techniques that serve as primitives/building blocks for effects, and those effects that "merely" build upon/combine these primitives.
So i'd file things like open (side-)borders, hardware scrolling (VSP), 4x4 chunky FLI, Graham-flex/zoom, Ninja NMI handler dispatch, N+1 sprites over FLI, diverse sample-replay methods, etc. under the former, and the various niche records and firsties mentioned here under the latter. |
| |
Raistlin
Registered: Mar 2007
Posts: 557 |
Quote: Quoting JackAsserWhere is the line. Everything I code I always aim for some world’s first I'd say the line is between discovering/inventing (hardware-centric) techniques that serve as primitives/building blocks for effects, and those effects that "merely" build upon/combine these primitives.
So i'd file things like open (side-)borders, hardware scrolling (VSP), 4x4 chunky FLI, Graham-flex/zoom, Ninja NMI handler dispatch, N+1 sprites over FLI, diverse sample-replay methods, etc. under the former, and the various niche records and firsties mentioned here under the latter.
Those are "just" hardware firsties though .. I'd also be interested to know the first DYCP, DYPP, etc .. it's not -all- about the VIC ;-p |
| |
Krill
Registered: Apr 2002
Posts: 2839 |
Quoting RaistlinThose are "just" hardware firsties though .. I'd also be interested to know the first DYCP, DYPP, etc .. it's not -all- about the VIC ;-p I specifically mentioned techniques not for VIC as well.
DYCP/DYPP etc. are instances of "soft" effects, as in of what you see, not how it's done under the hood. There are about a zillion ways to implement those, occupying different spots in the design space with the pros and cons those bring.
So yeah, that does escalate quickly when trying to enumerate any and all record and firsty. |
| |
macx
Registered: Mar 2002
Posts: 250 |
Hjørne Tilpasset Tekst Med Farve Cyklus (H.T.T.M.F.C.)
in Gammal Man from F*A*I*C. |
