In the first section, why don't you walk towards the enemies to kill them sooner? Wouldn't that make more enemies / the bosses spawn faster? Right now you stand still and wait for them to come to you, which seems like it adds a lot of time.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
TL;DR you do not play a person with average speed, strength etc. in video games, so your controller (ie you) shouldn't limit you to those options.
Now I'm curious what would happen if we had a portal TAS video played back on the rift.
I can't remember if it was Portal or Half Life 1 where a TAS (made by manually tweaking replay videos) gave me nausea due to the weird method of moving. Of course, that was just a 2D video; whether VR would be better or worse is hard to know without actually trying it.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I use a standing desk for most of my workday, every day. It took me maybe a week to adjust, and that was standing for much longer periods than you would normally game for. Standing is not a big deal for most people. Walking around a little is also not a big deal for most people.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Conceptually, a TAS that controls two programs simultaneously (one client, one server) is not impossible. We don't have the tools for it now, but there's no conceptual reason why we couldn't have a deterministic TCP/IP stack to go with deterministic game and deterministic server.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
The point is, is not the kill screen the de facto end of the game? If one reaches it, and I guess completes it as much as possible, doesn't that mean one has beaten the game, for all intents and purposes?
That's certainly one way of looking at it. My point was that declaring a game to be so hard it's TAS-proof is facile because if you've made a game that's impossible to complete by any means then of course a TAS is not going to be able to complete it. You can then redefine "completion" as reaching the point at which further progress is impossible...at which point the game becomes TASable, at least in theory if not in practice due to lack of tools.
Warp wrote:
It's hard to imagine such a game existing. I think it would have to be a kind of game that is not skill-based.
The closest thing I can think of would be a chess game. Although, in a sense, there are tools to help even a non-player to play at superhuman strength: Top chess engines. OTOH, beating such an engine would in itself be an almost impossible challenge, except by using another chess engine. So, depending on how you define TASing, this might not be an example.
Any game that does not rely on reflexes, is deterministic, and either has a small enough "game state" that a human player can readily figure out how various strategies would work, or has a built-in undo system. Basically, the vast majority of puzzle games confer no advantage to a TASer.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Any game that has a "kill screen" (thus e.g. Pacman) is immune to TASing. If the game is not actually possible to beat, then it can't be TASed in the sense that was used there.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I didn't bother reading most of this thread, but I saw the last few posts and decided I'd stop in. I know a few devs who have been working with VR kits (starting with early Oculus Rift kickstarter rewards, but also some more recent hardware). You should be aware that nausea is a huge potential problem with VR. Devs are having to learn techniques for ensuring the player stays "anchored" and doesn't get motion sick.
Some games work better than others -- in particular, games in vehicles seem to be less prone to causing motion sickness. But I'm fairly confident that you can't expect to just slap on a VR headset and play existing FPSes; they'd need extensive modification to work without the player getting sick.
And if you're talking about performing that kind of overhaul of the game, are you going to do it with an old game that everyone already owns, and where they'd expect to get the VR upgrade for free (or extremely cheap at least), or are you going to use it to help sell your new game as VR-capable? Not only is the new game going to be less work to update, since you already have the dev team assembled and they're familiar with the tools and environment, you're also probably going to make more money from updating said new game due to simple market economics.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Would automatically replacing each instance of the weapon switching keys (I guess L or R) by the corresponding menu presses (select, wait some frames, press left or right and select again) make a valid entry, or is it much complicated? If the RNG isn't influenced, wouldn't this be enough?
Even if this did work (and I don't know either way), it'd mean e.g. four pause screens if you hit L/R four times in rapid succession.
The big problem with using a non-cheated version of the game and then just making a special encode is that the music would end up skipping all over the place. Not that I disagree with the judge's decision otherwise.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Now I want to see an ACE that actually does write an effective program to feed inputs into the game to finish it within a reasonable timespan. Not as fast as a TAS does, of course, but because the TASer gets to hand off input as soon as the ACE section is completed, the TAS would still be "faster" than a manually-implemented TAS!
Obviously, simply encoding TAS inputs into the ACE program would only be effective if the TAS has copious lag frames or if the encoding can input more than one controller's worth of inputs per frame. Plus, that "solution" isn't very interesting except in a recursive kind of way, yo dawg etc.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
In other words, instead of trying to KO or kill your opponent, the goal of this game is to score points by performing successful attacks, with various types of attacks being worth different amounts of points.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
So, wait, Lord Tom: did you use savestates to make this? ;)
This was fun, and a good mix of showing the intended solutions played inhumanly perfectly vs. showing completely unintended solutions. Nice work!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Well, "artificial" just means something made by people instead of occurring naturally, right? And people are made by people, so by that logic, people are artificial (aside from whatever you decide the "first" people were, who of course occurred naturally). Assuming then that you buy that people are intelligent, people are artificial intelligences. ;)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Ahh, yeah, if it just flat-out isn't reproducible, then that violates one of the core rules of TASes on this site. That sucks. :( I'll still look forward to the completed work, though.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
TASVideos shouldn't have a mandate for absolute frame-perfection. IMO you should submit a run that is the best you can manage with a) your ability to commit time to it (subject to some reasonable minimum), and b) the quality of the tools you have to work with. Back when we didn't have frame advance, it would have been expected that you'd just lose frames here or there due to input inaccuracies; similarly, if you have an emulator that desynchs frequently, then you can't be expected to be as rigorous with your optimization.
I'm sure the run will still be great. Good luck!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
The PvP parts were great, the PvC parts were less so. Still, very nice work! I knew I was in for a good time when the game crashed on the first match. My favorite part was the ninja match, which was very...nonconventional. :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
You can still slide during the rail sections, if I recall correctly, and I think it deals damage. Good luck beating Seven Force that way though.
Some of the battles in the Dice Palace would also be problematic. I think everything else is doable assuming you allow partner throws; otherwise Golden Silver might be impossible.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Speaking as someone who hasn't played DQ3 before, there were some interesting parts to this run -- the setup for the item glitch and the item glitch itself in particular were pretty good. The level-up sequence was comedic at first but it just takes so long that it becomes dull again. And there's a lot of walking around with not much happening (and rightfully so; I daresay having random encounters during the walking would not improve it any).
I think maybe this game just doesn't make very interesting speedruns for the general audience, unfortunately.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
