Want to share an interesting article!

Link: https://www.pcgamer.com/modders-resurrect-forgotten-first-party-playstation-rpg-with-a-pc-port-thats-already-better-than-the-original-and-promise-4k-and-60fps-in-the-next-few-months/

Original Article:

A team of modders have breathed new life into cult Playstation RPG The Legend of Dragoon, a first-party game for the original console developed by Sony Japan. The game was released in Japan in 1999 but wouldn't reach North America until 2000 and Europe by 2001, and the combination of a mixed critical reception and a new generation of hardware saw it overshadowed.

Some players, however, have always held a candle for this most unusual and somewhat pioneering title. The Legend of Dragoon may have arrived nearer the end of the PlayStation's life cycle, but that meant Sony Japan really knew what it was doing by this point. It produced one of the machine's most visually impressive games, a huge singleplayer adventure, a clever turn-based battle system where characters could turn into the titular Dragoons, and realtime elements (early QTEs), all packaged-up with elaborate and fully voiced cutscenes.

Sony has kept the game available (with a few gaps), and a recent port to Playstation 4/5 even added a few features like quick save, but this is one of those games where you'll always be able to find the die-hards begging for a proper remake or even an actual sequel. Neither seems likely, which may be why a group of modders are going way beyond that and have created an unofficial PC port of The Legend of Dragoon, which is currently playable but "in a few months" will add no less than 4K resolution and 60 fps to the 25-year-old game.

The port is called Severed Chains and per lead coder Monoxide is "a reverse-engineered (unofficial) PC port of Legend of Dragoon and a full reconstruction of its game engine [that is] pushing the boundaries of traditional emulation software." Their point is that this "rebuild" is not based on another piece of software mimicking the PlayStation hardware, but standalone and "for that reason the game runs faster, leaner, and has breakout modding capabilities."

"I started by helping moderate and improve the subreddit [in 2016], then I created a Discord server so we could have a place for live chat," says Drew, a Dragoon fan who marshalled the game's fans. "By 2018 I was focusing on archivism, sharing rare concept art and promo images most fans had never seen before. Around 2019 I started our unified website."

Drew's passion for the game and keeping Dragoon alive led to fan events and podcasts, but when Monoxide turned up on the scene things got serious: "we were used to reversing being really intensive and difficult in general," says Drew. "Turns out, [Monoxide] was a senior engineer in real life. In under two years the whole game was decompiled, and he started converting it to Java. We now have additional team members and are making great strides together."

This is, as Drew will tell you, "the wayyyy shortened version." The achievement here is not that something came from nothing, but from years of hard work, dedication, and perhaps some childish love of the very idea of transforming into Dragoons.

Drew's glad that the pair's work is getting some attention and hopes this is "just the beginning" but cautions that the 4K 60FPS and upscaled graphics are "runnable now in our devbuild which can be downloaded, but the stable release will take a few months. For now, we're still on software rendering with RecBuild 2.1. We don't want to mislead anyone."

The project has been available in a beta form but Monoxide reckons the newest build is where they really start cooking, with among other things the music now working. They acknowledge a few minor graphical bugs in combat but, outside of that, "the game is fully beatable with no issues that negatively impact gameplay."

What's seriously impressive is the list of improvements made to the base experience: after all, this is a 32-bit RPG released in 1999, and 25 years is a long time. This version of Legend of Dragoon features instant loading, the option to skip FMVs, max item slots, unlimited save slots, tweaked controller input, mouse support in menus, higher quality audio, and "we’ve even fixed many retail bugs." There are a whole bunch of optional enhancements like text scrolling speed and automating certain regular commands, and then debug options that let you skip around the game's battles and maps at will.

-- I am a big fan of 60 FPS, so this will be huge once released!

1. You can see in-game time everytime you open a menu and time the rest of the duration manually. Like last menu before Doel -> Save Screen or whatever.

2. Last opened menu before Melbu Frahma -> his death. This has been the standard for all segmented speedruns since 2000s.

3. Yes, perhaps. You can still see the in-game time no matter where you are from last opened menu and time the rest manually. Also a lot of work, so I get it. Why the change to in-game time. It's only convenient in one regard. For those who save after they get a new PB to see their time easily. Some people (extreme minority) might not use a splitter and just play as is. They don't have to time anything and still see their time at the end.

All of these options were presented by GB. 2 options are basically the same as the current rules. Just that it would go over to in-game time (not a requirement). Can still be timed with real-time if it's prefered by everyone instead of in-game time. If people don't want a change at all, that's cool too. It's only a hassle to go over to in-game time if the majority prefers to not save, and if so just let it be real-time.

This one is for everybody!

The best way to change things are to vote for them. For 10 days you can vote which rules you think is best suited for the individual discs category. You can share it with other people you might think are interested in voting!

A link to the poll: https://strawpoll.com/NoZr3WPrey3

After the poll is over I will change all the submissions to fit the new rules. No changes will be made to the leaderboard during this voting period. If you wish to run during these 10 days; just make sure to save if you get a new PB (that will help you submit no matter the outcome).

Just want to add that changing to In-game time would most likely eliminate full game runs since I don't think any runner who do a full run wants to save their game during each disc swap just for this category. You could still use real-time for yourself to see where you are in terms of speed, but you submit your in-game time at each disc swap. This also uses the added times together (disc 1 + disc 2) automatically when you save after beating disc 2 since it's in-game time that matters. It would be easy on the rules too. The downside is that it requires a memory card. It was possible before to submit without one... That's the only downside I can see.

So my biggest complaint is that it removes all RNG. First time you run Disc 1 you just care about getting the Bandit's Ring and chill and take it easy to make sure everything is great for disc 2 start. Then you re-run disc 1 and play fast (don't get the ring) and get shield skip and some guard skips. And use the disc 2 start with Bandit's Ring.

The funny thing is that it was supposed to be in-game time we went by originally. I think it's far superior than timing it in real-time. The only problem was that full runners can't submit their best times if they get a new PB while doing a full run. I would actually want it to be in-game time that matters. The downside is that the board's times need to be updated and some will be removed. But in the long-run that would be a nice compromise. Maybe we should check or vote what the others think. It would make the administration of the leaderboard much easier too.

I will check every single submission and let the best one stay. There aren't any new rules it was just common sense. Also the rules were in a thread but we transfered the rules over to the leaderboard since it needs to be clear without browsing through the Forum threads that are 5 years old.

Full game runs have an insane disadvantage in the individual category because someone can grind disc 1 until it's perfect whileas a full runner do all individual discs in one attempt. The reason why the category was added was so that you can compare different strategies to find out which one is fastest. While one strat can be faster in disc 1 (like Rose only through disc 1) but can be slower on consecutive discs it was easy to spot the difference. Also to allow emulator runners to have somewhere to post their runs.

I'm sure most people won't be too affected by the updated rules. And yes, every submission will be thoroughly checked before 2024 begins. I can promise that. I'm starting with top ranked players and continue downwards.

It wasn't an issue before at least, it may have become an issue this year. It's clear you can't use a shield skip'd disc 1 together with a non-shield skip'd start of disc 2. :) I could level to 60 in disc 1 and save for disc 2, and use that save file for disc 2,3,4 and re-do disc 1. That wouldn't be fair. Your inventory, gold and levels may differ.

The rules aren't that harsh. Once you have gotten a good enough time in disc 1 for instance. You NEVER have to re-run disc 1. You can rerun disc 2,3,4 with that save file over and over and over again in the future. So always saving once you PB a disc is a sure-fire way to ensure there won't be any issues. And don't save if you don't PB for instance. No need to wipe the board, because previously there were no issues. As long as people save after a new PB there isn't an issue. Just overwrite your old disc 2 start with the new PB of disc 1 or whatever disc it may be.

Happy holidays, everything will be fixed at the start of 2024. :)

Greetings everyone!

We have now decided that we allow emulator to be used in all categories. The reason it has taken so long is that we needed to ensure that the emulator we decided on is stable, and that it is fair for those who have run on console to not feel disheartened by emulator being faster. With these settings you can compete, as close as possible, to real hardware. The difference is hard to determine exactly since emulator load battles faster, and console load screen transitions faster. In the end the difference should be minimal since I did extensive testing against my PS2 turbo footage.

If you are interested in speedrunning this game, in any category you wish, just make sure to download the emulator provided below (or in the resource section) and follow the settings provided! While submitting a run you need to select "PC" category and "Emulator was used".

Enjoy!

Duckstation 0.1-4918 now available in the resources tab or download below!

Duckstation (55.7 MB Zip): https://www.speedrun.com/static/resource/e4e23.zip?v=9fa38ff

Note: Please use the Duckstation version provided with the settings included in this thread. Red means it is mandatory and Green means it's recommended. While submitting a run you need to select "PC" category and "Emulator was used". Proof of settings isn't needed, but if you want to show them at the end of a run feel free to do so for easier verification process.

Duckstation vs. PS2: I have done some extensive testing on the Duckstation emulator, and this is the closest I could get it to while tuning it. It won't mimic the PS2 exactly, but it should be very similar. So console benefits everytime the screen needs to load a new area, whileas emulator will get into; and out of battle faster to mitigate the timeloss from screen transitions.

Turbo settings on emulator: Putting 1 frame on X button is optimal. You can have a higher number if you want a slower turbo speed (to make shopping easier). Turbo is optional; so if you want to play in the Legit category it is totally fine!

Duckstation 0.1-4918 now available in the resources tab!

I have done some extensive testing on the Duckstation emulator, and this is the closest I could get it to while tuning it. So it will be imbalanced, but in the end they should be very similar. So console benefits everytime the screen needs to load a new area, whileas emulator will get into; and out of battle faster to mitigate the timeloss from screen transitions. Putting 1, 2, 3 frames on X button was only way to make the text progress as a normal turbo controller for some reason. You can have 2 buttons for X at maybe 2 and 3 frames (or more) if you want slower turbo speed (for shopping).

I had even more accurate settings but then it crashed on Feybrand intro every time when I had 2x readspeed, otherwise we would have almost perfect settings. Sadly... 3x speed was absolutely faster than console in every regard (even if slight difference).

If you get a 3:02 disc 1 on emulator, you would get something similar on a PS2 console with errors being about +/- 1 minute. So expect to get 3:01-3:03 on real hardware. In which direction, I don't know yet. The point is that it's very close to real hardware.

So, what is better than text trying to justify things? A video of course!

^ This video uses the settings provided in this thread.

There is another test (forbidden one) that is running with 3x speed (no comparison, but it was faster than real hardware when I compared it off-screen).

^ this is the FORBIDDEN TEST footage.

So now the obvious follow up question... what do we do with this? I expect the console to be faster in disc 2 and 3 while disc 1 and 4 might be faster on emulator(?) It will be comparable to a real run within the margin of +/- 5 minutes, from what I can see. That's pretty accurate. The best would be obviously if a top runner does a full disc 1 run to get an idea of how well balanced this is. However, the way I see this is that if you are serious about getting good. You will know approximately what your time is compared to real hardware.

I will run the full disc 1, one day... I promise.

BDD, can I please have the notes of Addition%, I can fix the layout to my liking afterwards as long as all the strats are down on the document. Thanks!

I guess Lasagna's Lv 60 Dart could be useful too. :) Thanks!

I'm so proud of you BDD! First to break sub 10 is an eternal legend. ;-------------)

I think he meant Any% as finishing it as quickly as possible. That's at least how I took his question, but emulator usage like Duckstation finally feels like a viable way to play. However, I need to make sure it's as close to PS2 as possible without being better. Once I have the settings up I will be able to upload the emulator and its settings automatically there so people don't have to fiddle with it more than the controller or whatever they choose to use as well as resolution.

We are looking into it mate. I need to test the emulator: duckstation with certain settings. Need to test a bit more. It's on the way. Expect to be able to run it on emu at least at the beginning of the summer.

The time has come my dear friends... THE TIME HAS COME!

We will birth a category from this moment on... A non-throwables category will be born (All damage dealing items selected from the ITEM menu). I'm torn about whether SACHET will be counted towards this or not. Would be nice to have VS Jiango and Windigo. Thoughts? Ban or allow sachets? Only used for those two encounters. Also TOTAL VANISHINGS deal no damage but instant death so is it also allowed or banned? I'm going to be torn about these things. The Total Vanishings are only used to remove the pets in the battle second battle against Fruegel (and also one for a BOGY in disc 2).

Sachets and Total Vanishings are extremely limited and cannot be bought, so maybe it can be allowed. I don't know.

I'll be creating it this month. I will upload the guide for it as well. I hope I get time to run it soon to show case it.

I'm super-interested in this category where Miranda will be useless. No more OP strategies and you get to use Additions instead of the magic spam.

These categories have been added:

• Any% (AKA Classic)
• Magician Faust

These are now possible to run on PS4 and PS5 to everyone under the label "HD CONSOLE".

I have verified that the DEBUG ROOM will not work as far as I could test. If I close the game before the save is completed there will be no save file in the slot. If there is another method I cannot think about in the future to enable DEBUG ROOM being possible, then we will change it to add a CLASSIC category as well. We just rename Any% to Classic if so, but as it looks now. I doubt it will be possible to reach the debug room.

Exactly, so forbidding users of using rewind won't really hurt good players anyway. Meaning it's a great tool for noobs who want to rewind on Virage deaths, drop rates or guard skips/shield skip.

Anyone curious if the Popo skip works on PS4/PS5? I am... Anyone brave enough to try it? I want to see if it works. The trigger seems to be by grinding UP/RIGHT after stardust and then release up and press X from the video (see the exact spot he gets it at every time). I don't think I have a save anywhere near so not sure if I wanna play 3 hours to get there again. :D

1. Load speeds are similar to PS1 in some places but cutscenes load significantly faster. I would say it's about 1 hour slower than a PS2 run, but faster than PS3.
2. I think the RNG is the same, but I got Bandit's Ring and a Magical Hat from the Bogy (solo battle). Which made me wonder if they had been slightly altered, but I could have gotten lucky.
3. You can rewind to each turn so if you missed a Crush Dance you can rewind and repeat it. Or if you didn't get the Bandit's Ring drop you just rewind and mash the throwable item to another % to alter the RNG (it changes for each % you mash so if you got 210% on last mash you can mash 212%, 214% or 216% to see if Bandit's Ring drops).

Here comes the idea. Since the game have this rewind feature, should we allow it to be used? I think so. It will cost time to manipulate so you get Bandit's Ring on your 3rd or 4th rewind most likely if you are unlucky. A pro would probably restart after 1 rewind. Also... you can rewind so you get the guard skips/shield skip guaranteed in Black Castle. It's going to be more noob friendly and you cannot do anything else beneficial as I can think of.

The only thing I haven't thought about yet is how to abuse the Debug Room (it has save states and hard saves just like a ps1). Can you get there is the question, I would think you can and then I don't know if the rewind feature will have an advantage or having a save state at the loading screen to faster load the debug save or something.

I don't think you can cheat grossly at least. Save states + Rewind feature.

Congratulations! :)

I saw the TAS video and I do not mean it could be improved or not, just to talk about the points where I think there could be something overlooked.

1. Not getting Bandit's Shoes (it puts you in Area 1 if I am not mistaken if you purposely get caught into the whirlwind on Area 3 (final area where we are initially when coming to the desert). Even if it directly costs time to grab the boots; instead of back-tracking all the way to Power Down. It could serve as a reverse diminishing return to grab it to get closer to the Power Down? Bandit's Shoes is normally very far away since we start in Area 1 of a Classic run, but now we already are in Area 3 as starting position, meaning we do not waste anywhere near as much time grabbing it as in a Classic Run is basically the big point I'm making. Honestly, I get why it would not be super-beneficial in the TAS anyway since you can manipulate turns to the fullest of the solo character fights on Moon. Swapping around the boots for each male char's solo battle is probably something only worth doing in a RTA. So my point for this could be no time save at all in this section, just making sure it was not overlooked if it was.

2. Melbu Frahma Manipulation Getting Albert (or Dart albeit more difficult) sucked into the armor of Phase 2 of Melbu Frahma. This prevents the phase shift. The concept video is below, altough it is stolen from a classic run. It seems more feesable in a TAS seeing how much damage Miranda deals. You bring Phase 2 down to its threshold (18500 HP? or close to this number of dealt damage) then you let Albert (or preferably Dart) get sucked into his armor to prevent phase shifting. You would then need cast 8 Psyche Bomb Xs and a Gust of Wind Dance(?) to kill. Not sure if this is possible since you do not have access to ALL the booster items. Just something I thought could be possible in a TAS to speed up at least 3 minutes (probably more).

Nice man, would be nice with a Disc 1 speedrun of any sort to sort out how good/bad the settings are compared to real hardware. ePSXe sucks because even with a turbo controller you can't hit 268%. So Duckstation is the future when it comes to these things. I played FF9 on it and it felt better than ever (even better than a PS2 due to faster menus and instant inputs). Once someone has played through on Duckstation we can confirm if this game works or not (it should). Then we can see what to do. If you end up testing Disc 1 with video footage for research, then go ahead and do so once you've had time to play.