I used to run SORR under VirtualBox and had absolutely terrible loading times until I moved the program from a "shared drive" (between the host OS and the VirtualBox OS) to the virtual OS's C drive. Something similar to the shared-drive issue may apply when running in Hourglass, even when virtualization isn't an issue. No idea why; other programs didn't have similar issues. I'd guess SORR is doing something funky with disk accesses.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Yes, SORR is better. Has anyone investigated running it in Hourglass? Of course, it doesn't have the original SOR3 route; the best you can do is approximate it.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
That's the important word there. :) Guidelines are meant to generally be followed, but exceptions can always be made if people think it's worthwhile. Obviously you don't.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I'm with Brandon, frankly. Going for speed at all costs doesn't make the game interesting even when it's a more powerful system (c.f. arcade fighting games); it's not going to help on an NES game. Go for as much variety as you can, even if it's slower.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
You get the stars after scoring, I think, 50,000 points without dying -- each time you die, you lose a star. The stars give you upgraded blitz attacks, as Lorenzo mentioned. For example, Blaze's forward flip attack first gains more horizontal range, then turns into a backflip that hits more times, then gets a sliding dash added onto the end of it. I think all of the blitzes follow that formula -- first a faster/better-reach version of the basic blitz, then a brand new blitz, then the new blitz with an extra move on the end. Axel's Grand Upper transforms into a roundhouse punch and then a combination roundhouse-uppercut, for example.
As Lorenzo mentions, if you have a 6-button controller then you can execute any blitz attack at any time with the right button combination; it works a bit like street fighter games.
Streets of Rage 3 is honestly nowhere near as good as Bare Knuckle 3 (its Japanese counterpart). The increases in difficulty are completely over the top; not only is the Japanese Hard mode the American Normal one, but the enemies get even more health and deal more damage when they hit too. And of course you're required to play on "Normal" to proceed past the fourth stage! Playing Bare Knuckle 3 is fun; playing Streets of Rage 3 is just a slog.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Man, I'd forgotten just how much of a slog level 4 is. And it doesn't get better. Ninjas everywhere.
Thanks for the encode, though. The sequence where Zan was casually tossing enemies into the pit one after another was quite amusing.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Niiiiiiice. The extra chaos when you throw Skate into the mix is much better than having two ponderous muscleheads. I look forward to seeing the run on the workbench. :)
Thanks for the encode, Lorenzo!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
It occurs to me that this is a fighting game. Typically we don't go for the fastest possible result in fighting games, in the interests of variety. Perhaps this game would be better suited to a playaround?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I don't think so; mostly you'd just go slower in areas where you can't go around an enemy. As for subweapons, the game never makes your life difficult for not using them.
Feos's pacifist concept sounds more interesting, because the game is designed with the expectation that you'd be slaughtering everything in your path. Subverting that would be nontrivial, even given that you allow yourself to take damage. Ryu doesn't have unlimited health, after all.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Wouldn't memory.registerread need to know the address of the memory that you want to register to? That is, aren't you missing an argument to it?
I admit I haven't worked in lua in ages, but this seems somewhat underdefined otherwise.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Thanks for the encode, Patryk.
I have to say this got pretty repetitive. You did your best to vary things up and there were a few amusing bits, but on the whole it wasn't all that interesting. Voting meh for game choice.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I spent three years in Seattle and just couldn't hack it. The sun seriously goes away in September and you basically don't see it at all until April. For some people that's great, but not so much for me.
There are tricks you can pull to mitigate the effects without having to move -- full-spectrum lightbulbs, light boxes, etc. It's worth researching if you think it might be affecting you.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Well, I have to say I'm excited to see the 120-star run. But the reason I don't like the BLJ limitation is because you could come up with tons of self-imposed limitations on this or any other game's runs. Like if this run is accepted, what's to stop the site from accepting an Ocarina of Time "No Dungeon puzzle skips" or an NES Super Mario Bros. "No glitching through objects" run?
You seem to be laboring under the misconception that the site runs on strict rules regarding what a run must do in order to be acceptable here, and that any time an exception to those rules is created, it sets a precedent that allows other previously-unacceptable runs (or potential future otherwise-undesirable runs) to be accepted. That's not how it works though.
The "rules" for runs are guidelines, there to help runners figure out if a run they're planning would be likely to get good feedback or not. This saves them from spending a lot of time on a run that would definitely not get accepted, and saves the admins from having to spend a lot of time telling runners "That run might get accepted, but this one definitely won't be." However, being only guidelines, exceptions can and have been made to them for sufficiently entertaining runs. The most obvious simply being any run that has the tag "makes speed/entertainment tradeoffs".
As for precedent, while it's true that the standards for an acceptable run change over time, that has far more to do with changing audience standards than anything else. Generally speaking we're more easily entertained these days.
Your listed examples would probably be accepted if the audience deemed them entertaining. That's rule 0 right there: make an entertaining run.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
I think there's no reason to not allow BLJ anywhere in the run. Why not make a run entitled simply "70-stars" or "all required stars" and allow BLJ in the run?
Because such a run would be substantially different in a way that many people would find less entertaining to watch. By eliminating the BLJ, this run requires a large variety of solutions to problems.
If you want to see BLJs abused to their fullest, then the upcoming 120-star run is for you. It doesn't limit itself.
As for the strict categorization, I don't know that you can come up with a clean, concise category for this run. Who cares? It's awesome anyway.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
hypothetically, if we would be excessively strict on the speed rules, this should outrun the 70-stars route since they are both in the same category"BLJless"
To be excessively pedantic, the category on this is "any% BLJless", while the category on the other one is "70 star BLJless". Different categories, ergo one can't obsolete the other.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
First, it just collects 70 stars instead of 120, so the run feels incomplete.
That's because the minimum amount to beat the game without a major break (BLJing) is 70 stars. It's not some random number.
I know. All I'm saying is, I wished I could have seen all 120 stars, not just the required subset to finish the game. I'm pretty sure if this run is published that a 120 stars version will obsolete it right away.
The upcoming 120-star run is being made with BLJs and will thus look completely different from this. I would be incredibly surprised if it obsoleted this run.
This isn't a 100% run; it's an "any% with no BLJs" run. We have similar runs on the site; the one that leaps to mind is the Super Metroid "any% with no X-Ray Scope" run, but there are other examples. Basically any time a glitch completely breaks a game, there's room for an "as fast as possible, without completely breaking the game" run.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
It'll be interesting to see what happens when you do finally get an upgrade. Will you get back to the power you normally should be at at that point, or will you just get the +1 power that the fingernail and wings would normally have you at?
That is, are the upgrades "Set Firebrand's jump height to X", or are they "Increase Firebrand's jump height by 1"?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Great job, all of you! I've been looking forward to this run for quite awhile (evidence of how much effort went into making it), and it did not disappoint. Plenty of novel-to-me strategies and unexpected shortcuts all over the place. Way better IMO than just accelerating to lightspeed and phasing through all obstacles.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Ilari has it. If I recall correctly your average PC can't guarantee you better than ~10 milliseconds' accuracy when trying to do something at a specific time. If you need to respond to polling events at 60Hz, then you need accuracy down to 16.7 milliseconds -- and if you screw up for even one frame then your playback will desync. You need reliable response times from your controller.
The Arduino used for the NESBot can be plugged into a computer for ease of use (loading new programs onto the card, etc.), but it also has its own onboard clock for realtime (i.e. very low reaction time) activity. I assume it would be possible to write a program that would let you send controller input with your keyboard, through the Arduino, and on to the console, allowing you to play a "real" console game with your keyboard. I couldn't guess what kind of additional input latency that extra translation layer would impose though.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Well, fair enough then.
And I suppose you could access the full range in any direction by unplugging and replugging the controller repeatedly, though doing that with frame precision would be a neat trick.
Thanks for answering my question.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
