| |
Wile Coyote
Account closed
Registered: Mar 2004
Posts: 646 |
Defining 'World Firsts'
It is obvious what a world first is when it comes to a single effect.
Combining 2 or more effects, does that count as a world first ?
I'd anwser yes. |
|
... 42 posts hidden. Click here to view all posts....
|
| |
Copyfault
Registered: Dec 2001
Posts: 468 |
Quoting GroepazQuote:Compared to IFLI ILM does offer smth new: amongst other things there's a write access to $d016 on every rasline. Ofcourse it is still based on FLI but with this point of view I'd say _every_ new gfx mode that was "invented" after FLI is nothing new.
- (i)fli routines that write to dd00 and d016 every line existed long before ilm
- and yes i agree, most of these "graphic modes" are nothing new afterall. to me after fli the only noteable one is shfli, because of how the sprite layer is done.
Ok, I really didn't know that there were IFLI-routines with $d021 AND $D016 writes every line. Do you have an example?
Quoting GroepazQuote:Concerning Mongobong: for me this one should also count as a "world first" because samples and FLI was not done before and it was a certain technical challenge to get it done.
read again :) mongobong contains _i_fli with samples. but ordinary fli with samples was done many years before that. and i dont consider adding lda $dd00 eor #2 sta $dd00 a technical challenge =)
Oops, I have to admit I didn't read this little detail :/
But while improving "samples with FLI" to "samples with IFLI" might not be too difficult there definately ARE technical challenges when trying to improve some fx with FLI to IFLI (UIFLImax springs to mind, up to now still not done!) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11293 |
Quote:Ok, I really didn't know that there were IFLI-routines with $d021 AND $D016 writes every line. Do you have an example?
dont recall any specific demo. i remember hcl did it a few times, but that was long after i first heard about the idea :)
(and btw its pointless to speak of ifli routines seperately from fli. its the same. really =)) |
| |
Cruzer
Registered: Dec 2001
Posts: 1048 |
Quote: sideborder sprites is harder though, since with fli you can be pretty sloppy with timing :)
didnt XAKK do it all before in Micro Sleep ? :) HALLO! YOU GOTTA DO THIS! =D
They sure did. Hadn't even seen that one before. Always nice to discover new oldskool classics. |
| |
Copyfault
Registered: Dec 2001
Posts: 468 |
Quoting Groepaz[...]
(and btw its pointless to speak of ifli routines seperately from fli. its the same. really =))
Yes and no! Concerning this FLI plus $d016 thing you're definately right, but there ARE effects or even "gfx modes" utilizing FLI that cannot be interlaced that easily. For example, try to make an interlace version of Ninja's 6 Sprites over FLI and you know what I mean ;) |
| |
Copyfault
Registered: Dec 2001
Posts: 468 |
Quoting algorithm[...]
Regarding MRFLI. Its similar to other FLI based gfx modes in that it generates a badline per line. A first release would really be when the gfx is first demonstrated. utilising the features.
For example with the invention of FLI (by Solomon / Beyond force?_ this was merely demonstrated as some amiga copper effects but realised and demonstrated into a gfx mode in Blackmails Dutch Breeze.
I have to agree that a gfx mode has to be demonstrated. What sense does a gfx mode without an example pic make anyway?
Until now I thought that at least the underlying routine was not done before the release date, but this is obviously not the case.
After all it's always good to have a tool for the real machine.
Quoting algorithm[...]
In regards to progress in C64 gfx modes, it has reached the stage where the potential (hardware wise) has been reached. The next stage to improve the quality would be the dither or/and mix color techniques used.
What do you mean by 'hardware wise'? For me there are two layers: the low level layer dealing with the gfx-routine running on C64 and the meta layer dealing with clever conversion algorithms. Almost every gfx mode comes with a special routine on C64 side, and many of the later gfx modes come with both a new gfx routine and an improved conversion algorithm.
Having NU(I)FLI in mind or the still to be done UIFLImax I'd rather say that on the low level side the limit has not been reached yet. I'm sure this holds true also for some of your still-to-come gfx modes.
Which brings me back to my actual question: does pure code count as a worthy release? |
| |
RaMoS
Account closed
Registered: Aug 2003
Posts: 37 |
When scroll appeared in demos?
|
| |
Wile Coyote
Account closed
Registered: Mar 2004
Posts: 646 |
I have seen horizontal FLI scroll, but not horizontal IFLI scroll.
I guess IFLI would be as easy as FLI, if what Groepaz says is correct. Not that FLI scroll is easy, as it is seen rarely. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11293 |
as copyfault hinted, whether interlacing something is easy (and the matter of toggling videobanks) depends on the memory usage of the respective gfx mode. it *is* quite trivial for most of them though :) |
| |
Wile Coyote
Account closed
Registered: Mar 2004
Posts: 646 |
Quote: as copyfault hinted, whether interlacing something is easy (and the matter of toggling videobanks) depends on the memory usage of the respective gfx mode. it *is* quite trivial for most of them though :)
I guess theres a (world) first right there, waiting to be claimed. Just needs someone with FLI and memory usage skills :D |
| |
algorithm
Registered: May 2002
Posts: 705 |
When i meant 'hardware' wise, i was referring to display quality. not size. ofcourse new gfx modes can be greated which can either increase the size of the visible display (eg my MUCSU-FLI) or to reduce the cpu time required for maintaining the display at high quality (MCIHD - not released yet)
At the time of writing, the ultimate display quality on a hardware point would be MUCSU-IFLI or HUCSU-IFLI allowing Hires FLI per line with multicolor sprite underlay.
To further improve the quality, this would need to be done via the conversion/dither routine
IFLI variations etc are trivial c64 code-wise but would generate a massive headache for a converter when it takes into account pal blending/color mixing / luma overload and as mentioned previously takes up a lot of the c64's ram |
| Previous - 1 | 2 | 3 | 4 | 5 | 6 - Next |
