Very true. If what I wrote sounded confrontational, my appologies. Most of what I was describing is utterly useless and counter-productive in many cases for a 'video filter' that wasn't actually doing compression. :-)
That might or might not reduce the video size, depending on how exactly zlib does its work.
Then what you were describing wasn't actually a compression technique, like I read it as, but a pre-process technique that was relying on zLib to do the actual heavy-weight compression. If you're relying on something like zLib, then yes, keep EVERYTHING byte-aligned. :-)
Though to be honest, a modern BWT-prewashed Adaptive Arithmatic coder would perform significantly better than zLib does. But things like BWT didn't physically exist when zLib came out in the first place. It's damn good tech, but there's better stuff available now. :-)
The previously-posted synopsis of tile-based compression has a few subtle but important changes that could be made to be, ideally, to likely improve compression rate if M-Mike decides to re-implement such a codec. You might have done these, and simply not mentioned them explicitly in your synopsis before, Warp. If so, I appologize.
First, and this is the simple trick, merge all the tiles that move the same amount into a single 'macro tile' when there are more than a certain amount of them. Considering that a bit-mask of the entire screen on NES (256x240) would only be 30 bytes (16x15 blocks, 240 bits, 8 bits to the byte) and the vast majority of the screen IS going to move in lock-step, this is very likely to reduce the space needed for matching block-moves dramatically.
Second, also simple, don't try to keep things aligned on the byte-order. For example, if you only need (-16,15) range, don't use 8 bits. Use 5 bits.
Third, and this would be simple but processor-intensive (what compression isn't?), try all possible alignments of the grid on the new screen. Don't lock them to the boundaries of the screen, so to speak. This would allow better 'locking' with two-axis-scrolling games, and moderately improve the lockdown on single-axis-scrolling games.
Fourth, something of an amgamation of the above... a possible (and completely different approach) would be to only make bit-masks covering portions of the screen with motion-vectors attached, then simply encode the 'unvectored pixels' directly. The bitmasks can be trivially encoded with run-length compression run through some type of Huffman compression as a trivial solution, which could also work quite well with the limited-selection 'unvectored pixel' values on highly-limited-palette systems like the NES games since such a technique would cause a LOT of chaotically-random sequential pixels that might not compress well because their localized relevance was dramatically reduced, so Huffman's advantage (localized differences aren't any harder to compress than distant differences) might give it an edge over more modern LZ-based compression. For truly ridiculous compression ratios though, a BWT pre-filter on the data would likely 'cure' almost all of these problems, since a full-bitstream BWT would be easilly possible on the unvectored bits.
And yes, I do way too much research into compression in my spare time. :-)
*bzzt*
That's like saying you can't rebind the keyboard in Quake before making a speedrun.
Many turbo controllers were manufactured by the original hardware manufacturers. That, by definition, makes the turbo feature a built-in component available for the system.
That's like arguing that if your original pack-in controller broke, you can't make a speed-run on that console anymore. And forget about making a two-player or four-player speed-run if you didn't buy a package deal that included all four controllers in the original box!
Hell, you could buy an NES with two NES Advantage joysticks included at one point, without any 'original' NES gamepads included.
Re: McBain
You just answered your own question with your statement.
Most people here like as fast as possible runs better.
I'm not one of those people. I'm one of the people that prefers seeing Quake Done 100% Quick instead of just Quake Done Quick, and has pulled off 100% Average Rank on MMZ3 on a real GBA.
And it seems like there are plenty of folks handling the QDQ speedruns, so I might as well tackle my first love, 100% runs.
Re: Mikwuyma
Odd... I always seem to get the Active Form first. Probably because I tend to walking slash-dashing slash combo the 'two hit' enemies I come across. And I hadn't known the dashing slash was that much slower than the jumping slash, again probably because of my play style leaving me falling off a ledge before the dash-slash terminates whenever I can manage it, and on the 'test runs' I've been doing I just flick-dash-jump off the wall as I fall down those first few pels.
This is why I'm discussing things here though, I'm learning things along the way and improving my plan for the run more since a lot of my initial play-time was on a RL GBA so I'm still getting into the statistics-side of planning a run.
And I think it's safe to say that we've both brought up bosses where each of the two forms would shine. On anything remotely lethargic, Rise would definately win, but on the hyperactive bosses Active could give an edge. It may well be that I'll just eat the 'switch delay' between them. It's not like it's hard to test both approaches on any given boss. :-) I'll start experimenting with the Rise Form more.
Re: Zurreco
Good luck with whatever you're working on, too. Look forward to seeing anything that comes out of your neck of the woods.
Pros of the Rise Form:
More damage from a complete 'standing three-slash combo' (14) attack than the Active Form mid-air (6) or dashing (8) two-slash combo.
Cons of the Rise Form:
To complete the three-slash combo takes far, FAR longer than the respective Active Form attacks which take around 10 frames to complete their damage sequence versus 10 frames without even completing the first slash of the Rise Form attack sequence.
This results in many situations where the combo has to be aborted, resulting in 8, or even only 4 damage, to give time to dash back away from a moving boss or enemy. Try to land the three-slash combo routinely on the Train-stage boss, for example. It's very hard to actually get that supposed 'extra damage' on that boss for example, when it reduces your mobility that much.
The Active Form attacks, by comparison, are all done while moving, allowing you to combine them with dodging incoming attacks.
That's my reasoning for favoring the Active Form over the Rise Form. Theoretical damage is wonderful, but with how hyperactive many of the bosses are, and the amount of jumping and dashing most of them involve, I feel that having to 'restrict' myself to only immobile attacks to deal more damage is a waste.
And Re: Zurreco, I'm going for a type-2 video. Nothing but 100% Rank, with no concerns about anything else.
100% Rank, to start with.
100% Items would involve a LOT of boring menu-shuffling and sitting there 'mining' cash (energy crystals) then upgrading all the Elves, then USING all the blasted things. Maybe feasable for a 'second quest' run through the game as a kind of continuation to the first 100% Rank run, but definately NOT something I'll do as an initial goal.
And actually, I've found the Active Form much more useful because of the two-hit jumping slash, and tumbling slash. The standing slashes take too much time and freeze your movement compared to the jumping or dashing slashes.
I see topics about 1, and 4, but 2 and 3 are my personal favorites, though I only have 2 copied from the cartridge yet as my friend is borrowing the programmer/reader for the next month or so. So... this is mostly to announce that I'm working on a 100% run for 2 at the moment, and to open the floor for discussion and possibly planning.