Oklahoma, USAI started streaming and speedrunning about three years ago because I felt that the WR run in my primary game, Legend of Legaia, was not very good at 12:22:08. I set out to improve it and have continually worked to bring the time down a little at a time. Right now, the WR for any% is 7:26:00, a nearly five hour drop over the past three years, and with new glitches being discovered, any% NG+ (sort of a misnomer, but the glitch relies on using pre-made save files) now exists with a WR time of 6:52:05.
With all the new skips that have been found since the start of the year (and the year's only 13 days old), I'm actually excited to see what the next few months will bring, and really do hope to see the run in Limit Break this May.
The basic premise starts with a glitch called incense glitch which I have known for a year or so; if the 'Incense has lost its effectiveness' text box pops up while you would hit a loading zone or other trigger, instead the text box cancels out the trigger, potentially allowing you to walk around through out-of-bounds. I had hoped that some day, this would allow us to walk around through out of bounds and get into Nivora without breaking the ice, and potentially skipping Usha, Uru Mais, and even (some of) Sol. Sadly, that doesn't seem to be possible.
However, during testing, Trance found that it was possible to skip all three dream sequences in Uru Mais, and found RTA-viable setups which I confirmed worked (with minor changes) on a slim PS2 console. It's also possible to skip the Jette fight completely by having a text box pop up when he would show up to fight you, as shown in the video above.
Altogether, this is expected to save about 11:30 in any% and 7:30 in 100% (can't skip Jette because of Master Course glitch, and also because by fiat I decided 100% should include all bosses, including Lapis and other "optional" bosses). There may be other possible skips yet to be determined.
Still holding out for an item cloning trick, but I'll take what I can get. (Being able to easily and quickly clone Power / Miracle Waters would be the holy grail of the run.)
I would request a change of name to Dr Skies (DrSkies is fine if names can't have spaces; please no underscore.)
I've really only run the category once to prove I could do it, so I don't have notes on no slots runs.
Also, the route will really change whether you are playing US or JP/PAL versions, since the JP and PAL versions have enemies that give 133% EXP and 200% compared with the US version, so your route will have to account for that in some way. You'll get more levels and money from forced encounters in the JP/PAL versions, and the run is faster/easier overall.
I will say that the routes are identical through Zeto, except that you want to pick up the Tempest Jewel and Burning Jewel in the rooms right before Songi for an extra 2,000 G. You also want to get Theeder, since that's a cheap AoE spell that can help clear out mobs for quicker EXP.
Berserker is Earth element, so Earth Jewels will halve damage output on that boss.
Xain can be fought at L14. Here's the strategy: Earth Jewels still work on normal attacks, but Bloody Horns and Terio Punch both deal non-elemental damage (like Cort's Guilty Cross). I've tried it at L13, but Noa can't reliably survive Bloody Horns and Gala can't reliably survive Terio Punch without Spirit until 14.
"I think that idea from Werster would be good if coupled with the ability to see how long ago they last verified a run."
I don't know how much I like that as a metric, because it's incomplete data.
For example, I'm on here pretty regularly, but I think I've only verified one run that's not my own. In the case of Quest 64, we have enough moderators to verify runs within a few hours. In the case of Legaia, no one else runs the game, so I guess I could create dummy runs to verify them and make me look more active, but it's pointless.
I'm the only person who's done runs of Legend of Legaia in the past year. Mostly because the slot machine timing is a real pain in the butt, and the run is a bit longer than 8:30:00.
A lot better than the nearly 12:30:00 that it was at this time last year.
I don't know if I like the fact that game mods can submit runs "on behalf" of someone without that someone's consent to appear on the leaderboards here.
I doubt that everyone here wants every run they do submitted here, particularly first runs that are really bad.
Both of these glitches are known. The first glitch mentioned is not useful in any way that we are presently aware. The second glitch is something known as the "initiative glitch", because it allows Brian to "cheat" the first turn in the battle. This has potential uses, but because it is, to our knowledge, frame-precise (or at least no more than 2-3 frames), it's pretty much impossible to trigger on command - just a sort of nice accident if you get it.
Honestly, I don't care about requiring a video. -shrug-
I think that not requiring a video is fine, but I don't feel comfortable accepting top times that don't have video evidence. For the time being, I removed the "require a video" requirement, but I'm also going to require video evidence for claims of top times and reject runs faster than 10 hours (for any%) that don't have a video.
Of course, if you take a picture of the in-game time on the stat screen of the end of the game, that would be useful. Not 100% ideal because of save scumming, but it does help validate.
I like the use of category variables here for uses OoB / uses Spirit Cloning. And yes, any% should include all the glitchless runs (but, to the extent that people want to do a glitchless category, the reverse will obviously not hold. :p )
Also agree with the delineation of any% vs. glitchless any%, as opposed to any% and glitched any%. Glitched runs appear to be the favored categories to run at this time, so the any% should assume that as the default.
It's also worth noting that there's an alternate method of fighting Fargo from Out of Bounds, and this method only takes about 2 minutes total to fight him, instead of 4. I do think that this is worth doing, now that it only costs a couple of minutes to do the fight and get his 20 HP bonus.
Some other things:
If you do OoB, going back into Boil Hole to fight Fargo takes about 4 minutes, compared with going directly into Baragoon Moor from Boil Hole without fighting him.
It's about 19:00 from Shilf to Fargo using no OoB.
By comparison, I did a save from the Dries Hut to Baragoon Moor. I was able to get from the Dries Hut to Baragoon Moor in about 10 minutes, and it was about 14 to get back to the Moor if I fought Fargo.
Fighting him increases your max HP by 20. This turns out to be quite a lot, as it means that:
Guilty doesn't kill you with Short Range + Long Range at 25 defense (pretty typical amount); his min is 129 and max is 144. (It's more likely than not that you don't die at 140 / 25, but I'd prefer not to take any chances)
Beigis doesn't kill you with 2x Short Range (86 x 2 at 26 defense means you take 172 max at 26 defense, compared with ~167 HP, assuming you didn't fight Fargo.)
Mammon can't kill you with double Flame Wall (212 HP / 27 Defense means at most 105 from Flame Wall).
It would require about 32 defense to survive double Flame Wall (195 HP vs. ~98 damage max?)
Fastest place to grind is probably Skelebats in Blue Cave 2? I think you can get hit 40 times per minute, and you need about 200 hits to get from 27 to 32 defense. That would cost about 5 minutes as well.
If we were willing to risk Flame Wall not happening twice in a row, we'd need to get to about 29 defense for Beigis to not die to 2 Sword Slashes. This would require about 1:40 in additional Skelebat grinding (to about 102 HP), but we would need to accommodate for less time spent taking hits in Boil Hole, so grinding to 103 or 104 HP for Nepty is probably ideal to accommodate this, and makes this and the Shilf fight safer. This makes OoB about 5 minutes faster than no OoB under realistic scenarios, meaning that 2:15 or so is probably on the table, but would require an insane amount of optimization.
There is also the fact that you will be down a bit of power due to fewer spirits acquired (2-3 battle spirits), and the damage multiplier is dependent partially on Brian's max HP. Cloning the Baragoon Moor spirit on the bridge and getting all six Spirits in Glencoe (I only got the first four, all cloned) will give 5-ish more Spirits. Can also get the 1st spirit in Brannoch Castle (~18 seconds?) or the Shilf Antechamber spirits as well (~20-ish apiece) to make up some damage output. This means that more realistically, we'd be looking at shooting for a high-2:16 to low-2:17.
So, yeah, here's some theorycrafting about actual routing instead of categories.
Yeah, pretty much: should have any%, all spirits (both glitchless and normal, no save corruption), save corruption run, and possibly 200 spirits (US only) if anyone cares about the category.
The top times for the glitchless categories should still be listed as their respective category WR's until supplanted by a run that does use glitches - it's somewhere between disingenuous and outright wrong to say that my 2:30:xx run is the any% record when kirkq's glitchless run is a full 8 minutes faster.
This is what I feel as far as categories are concerned.
Save corruption should NOT be allowed EXCEPT as a separate category; it feels one step removed from the OoT crooked cartridge trick; this really feels like hardware manipulation to me. Now, it's an interesting run / route and I feel we should consider it, but I don't know of any game where save corruption is part of any% without being a separate category from no save corruption.
Spirit Cloning should be allowed in all categories (except if a glitchless category is developed).
OoB should be allowed, but I feel we should have at least one category with no OoB, because the routes seem like they would heavily diverge. I personally (read: personal bias at work here) would care to see any% allow it and all spirits to disallow it, so as to let All Spirits be the run that showcases the majority of the game, and any% is "beat the game as quickly as possible, utilizing whatever techniques, glitches, tricks, etc. enable that goal."
Now, as far as routing is concerned, if you are doing Save Corruption, it almost seems like you might just go all-in on Fire at the start of the run since you won't need Healing magic, though you would lose access to Exit (and cloning). It seems like, because of cloning, it would be fastest to do 24 Water (Return, Escape, and Exit spells) -> 50 Fire -> 50 Earth; you can probably get past Shilf while only burning ~50 Spirit Lights and you can easily get to 20-25 Fire by Cull Hazard and 32-34 Fire by Zelse for damage. Fire Ball 3 pretty cleanly destroys Nepty and Shilf as well, and you'd skip Fargo (and probably have Avalanche by Guilty/Beigis anyhow).
Will probably have better things to say later when it is not-Midnight.
Stickied thread to request name changes? Given the number of threads for such requests on the first page here, seems like it might be a good way to cut out some of the clutter.
I checked a couple of times during the stream - at about 30 minutes, an hour, and an hour-and-a-half into the run - and it was not visible.
Could it be an issue that Twitch just doesn't report streams unless said stream has enough followers / viewer count?
I checked an hour into the stream and nothing. There was no [nosrl] in the title, and my Twitch is setup. There may be an issue where Twitch likes to randomly log me out for no reason because I have some privacy block software that filters out third-party cookies; that would be the most likely thing that I can think of that would cause it not to show up. If there are any domains here that I need to whitelist so my Twitch page appears, let me know and I can whitelist the appropriate trackers.
I'm having the same issue; I was directed over to this thread from my post in "The Site" thread, so I'll crosspost to make sure it gets seen. My Twitch page is linked into the site, I'm on OBS as the others are, and I checked the stream page while I was an hour into my stream, so I don't think it's a caching issue. I haven't really changed any Twitch settings from their default values, so I don't think there's something crazy out-of-the-ordinary that is unique to my Twitch page, but I'm more than willing to fiddle with things on my end and see if that helps to try to help debug.
