Posts for Derakon


Post subject: Re: I don't think we'll be visiting minetown...
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
dwangoAC wrote:
I don't think we'll be headed to minetown as there's not much there we really need.
I meant more "if you happen to have a couple hundred spare turns, this is something you could do with them". I don't really have a feel for how tight the timing is before turn 2000 comes around.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Hooray, angry shopkeepers. :) You should make the entirety of Minetown angry with you (without actually killing anyone, of course), if you have time. Maybe leave Izchak's shop intact.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Warp wrote:
Heisanevilgenius wrote:
Just go with what the popular expression is so there isn't confusion.
I can't help but to notice the irony here. It was precisely the confusion that the term "hack" caused that prompted me to start this thread...
But I've never seen anyone else get confused. Sure the terms may not be perfectly accurate, but people adjust to the language as it is used, instead of worrying about the precise meanings of words.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I dimly recall that some enemies can drop blaster powerups -- at least, I think I remember grinding on womp rats to get high-powered blasters. Maybe it was extra lives though. It might be worth seeing if their drops can be manipulated, or if it's one of those "every 20th monster drops something special" kinds of things. If you could reach Plasma during the first level I imagine that would speed things up a lot.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Hooray! Congrats on finishing; I look forward to watching it. :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
GeminiSaint wrote:
Why not good old Snes9X?
Not only has SNES9x been deprecated, it also can't handle the Super Star Wars. Why do you think there haven't been movies published for them before? EDIT: thanks for the encode, feos! On the whole, it looks pretty good, though it'd be nice to have some explanations for seemingly-suboptimal parts. Particularly, the Sarlacc Pit Monster fight. Also why do you sometimes shoot at random (especially when jumping over pits in the hangar bay or up slopes in the Princess Leia level)?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
TASVideos is not outside the ROMhacking community -- there's substantial overlap between the people who play hacks and the people who watch TASes. Switching up our terminology for no good reason will just create confusion.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Looks great! Thanks for the encodes, Snufflimz. When you dive-kick against Rahab, does the kick damage increase while you're underwater even though you're slowing down? I seem to recall that kick damage is based on the amount of time you spent in the kick animation, so it's hilarious if a comparatively minor kick is greatly enhanced just because water resistance made it take longer to reach its target. :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Yep, that was fun. Nice work! And thanks for the encode, oneeighthundred!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
The screenshot appears to be from an A2600 game.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Having read the text, my understanding is that even understanding the RNG only buys you comparatively minor gains (in terms of a 2-hour movie) without more work. The main potential for improvements is in "RNG pathing" -- where you go more slowly early on in exchange for payoffs down the road. Bobo's bot does not try to go very deep into the potential search space because it rapidly becomes prohibitively expensive CPU-wise (similarly to trying to find optimal inputs via brute-force). Apparently (assuming Bobo's estimates are accurate) a run that used one more level of optimization could be generated by someone willing to dedicate a week or two worth of computer time.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Rolanmen1 wrote:
This is a basic rule of Lua, if you want to write a "\" inside a string you have to use "\\". So the correct syntax is "C:\\pics\\map.png"
It's true for practically all programming languages. They use \ to indicate the beginning of a "control character" (for example, "\n" starts a new line), which of course makes it difficult to directly represent \ itself. The solution is to make \\ the "control character" that means "\".
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Not that voting is really relevant here, but I do agree that the screenshot should try to show the viewer why they want to watch the movie, not recommend against watching it. It'd be like having the description say "This is a terrible movie; you shouldn't watch it." Even if you're joking, it's a bad idea.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I don't see why this is such a big deal. There are multiple valid points at which you can end input, and "soonest point at which no further input is needed to end the game" is one of those. Not everyone likes that, but enough people do that it's an accepted approach -- and it can lead to interesting strategies like in the Gradius (grab a shield and park in a specific position) and Gimmick (arrange a shot to hit the boss 3 times) runs. It's perfectly valid here. So, why are we having this argument again, over this run in particular?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
What a shame that MM9 doesn't have quick weapon-switching. Still, I think your plan sounds like a good one, and the two levels you've completed both look good too!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I feel sorrier for Gergoth. Spends all that time crashing through tower floors, and what does he get? A devil-enhanced boot stomping on his face. He doesn't even have any skin there! Ouch. Looking excellent, Kriole! Keep it up. :) And thanks for the encode, Mothrayas!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Well, they'd be marginally easier to hex-edit, wouldn't they? :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Going OOB in a Knytt Stories level is just a sign that the level wasn't competently put-together (though apparently that's been patched?). Using that to break the game is nothing special.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
During the landing sequence, why do you have to maintain such a slow speed for so long? Couldn't you approach the carrier faster and then slow down right near the end? Some explanation of the mechanics of the missions might be nice too. As far as I can tell, you just fly out in an arbitrary direction, destroy targets as they pop in front of you, and eventually go down to land again; does the game not track your X/Y position at all?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Ah ha, thanks for the clarification, MESHUGGAH. I should have realized that the run was over as soon as the dragon fired! Silly me.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Hey, it's an improvement to a movie I liked to begin with. One question, though -- wouldn't it be faster to end the Dragon fight with the dragon as close to its tile location as possible? The game spends a long time slowly moving it over to the other side of the battlefield. (And to a lesser extent, the same question applies to the Goblin fight)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Since there's no way to confirm that the person voted the way you want them to, they're always free to vote as they like and then lie. Not that that justifies the initial threat, but it's a pretty toothless threat. Regarding the crypto discussion we were having: thanks for the explanations. I admit that I can't see an obvious way to subvert individuals' votes under that regime. That doesn't allay my other concerns: * I don't trust the creators of voting machines to actually make a verifiably secure design * Paper trails are fundamentally harder to subvert than bits -- even if the paper trails just replicate the bits! * I see no pressing need to have instant results (indeed arguably the current system is devaluing Alaska and Hawaii's votes as the election is largely called before their polls close!) * Specifically for internet voting, there's no way to make it anonymous (someone could always be looking over your shoulder)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Twelvepack wrote:
Derakon wrote:
But the second issue is that there's no paper trail. If someone does find a way to change your vote, then there's no record of what your vote was originally, no authoritative, trustworthy fallback option.
This problem has a simple solution; using digital signing. This would prove the identity of the person who submitted your vote was you, and it would be impossible for anyone to change the vote.
I will admit to not being a crypto expert, but from what I understand you to be saying, the idea here would be that each voter would encode their vote with a private key, and the public key would be held by the government? I.e. the machine can count votes but cannot change them, because it would need the private key? Or of course sufficient computing power to crack the crypto, but we can assume enough bits are used to render the encryption uncrackable in the timeframe allowed. So, two issues: 1) The machines running the voting are still subject to tampering and are functionally impossible to validate. The creators of voting machines are notoriously closemouthed (relying on security through obscurity to hide how bad a job they do in the design phase). Even assuming you could design an open-source voting machine, how do you prove that the software on the machine is the same software that passed review? An untrustworthy machine can cast votes however it pleases, and there's no way for the user to detect it; how are they supposed to know that the random string of hex the machine generated is the proper encryption of their vote? Are you expecting every voter to learn how to perform their own encryption? 2) Nothing stops a malevolent actor from scattering extra votes in. How do you prove the votes are fake? You'd have to decrypt them -- which requires the private key.
Besides, what makes a piece of paper authoritative but an unalterable, traceable computer record wouldn't be?
The fact that manipulation of data on a grand scale (as necessary to throw an election) is much harder if that data exists in the real world instead of as a bunch of bits. Like it or not, a lot of voting security relies on it simply being harder to undetectably do things in the real world than it is on a computer system. So no, paper ballot are not 100% authoritative -- they can be altered. But to do so on a large enough scale to steal an election, you need a massive number of people, spread over a huge area (so the tampered ballots aren't all obviously in one place) which makes it much harder to hide.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I do have to admit to being curious why the manticore isn't moved onto the top power point before the dragon kills the wizard instead of afterwards. Presumably that results in an extra non-dragon fight (thus negating the gains of avoiding the dragon-unicorn fight), but it's not discussed in the writeup.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Dooty wrote:
I also do not understand why, but in the 16-bit era it was common for enemies to explode to show that they were dead or blink to show that they were receiving damage.
It saves the developers from having to create more complicated death animations. It's reasonable to not expect to see an enemy any more if they explode (the remaining pieces would be too small to be worth drawing), but otherwise you end up having to either show the corpse, using up memory and processor cycles, or have the corpse mysteriously vanish, which feels odd to the player. Oddly, making every enemy out of explosives feels less odd... One of the big things I was impressed by when I first played Symphony of the Night was how all the enemies had customized death animations. Which, granted, still often included explosions, but hey, explosions are cool. :)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.