This topic is for the purpose of discussing #6968: zggzdydp's PSX Suikoden II "best ending" in 6:48:37.71

Man. I love December submissions of long PS1 titles. Claiming when it's accepted before the New Year. ----- Edit: YouTube version for those having issues with bilibili. Link to video Came to my attention audio borked, somehow. I'll switch it around in a bit.

Still have yet to finish the movie, but from the first few minutes, the run looks pretty optimized. Yes vote!

I didn't watch the movie yet, but I want to ask if these glitches could save time, and if they have been used: The instant 999,999 potch glitch: http://www.suikosource.com/games/bugs.php?game=gs2&bug=8 The instant level up 99 glitch: https://www.youtube.com/watch?v=KQFJNR6a_lE Recruiting members at level 99: http://www.suikosource.com/games/bugs.php?game=gs2&bug=15 This one seems to be more frequent on emulators, but there are also reports of happening on real console. I also found someone mentioning that Omnigamer performed some researches about it: https://gamefaqs.gamespot.com/boards/198844-suikoden-ii/75946098

I don't think the potch and level up glitch save any time, because there is no need for potch after getting your HQ, and the level up glitch takes place in the last dungeon. I think the manner in which the hero gets to level 64 in one battle uses the same principle though, just way earlier.

I have same view with Chanoyu. For the level 99 guys, in fact I have used high level Gerog and Pesmerga in normal playing who were strong indeed. However, they join very late and only can attach one rune, compare to 3 rune guys, they are not so strong as they look. For other lv99 guys, they are weaker than that 2 guys above. Overall, these guys are not worth to use.

Thanks to both of your for the response. Yes, I thought that these glitches could not be useful, but I wanted to ask anyway, just to be sure.

This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [4369] PSX Suikoden II "best ending" by zggzdydp in 6:48:37.71

I only saw this submission once it was up on the TASVideos YouTube, but I have a lot to add about its optimization and planning. The briefest way I can summarize this run is that the planning it used was naive, or at least quite outdated as compared to current RTA strategies. This has a significant impact on the entire course of the run. For comparison, consider the current RTA record in the same category: https://www.youtube.com/watch?v=7SrKeb3AlKU With a comparable route and further luck abuse I think a more optimized TAS could be about 10-15 minutes faster than this submission. I've listed some specific points for improvement below. General Comments: -The recruitment order is fundamentally different in this run from very early on, so it is difficult to directly compare the strategy. However, I believe that the order chosen in this submission is not optimal because of several forced revisits, party composition issues, and some wasted opportunities for multitasking. -Early-game movement is not optimal. About halfway through, it looks like the author noticed that proper-timed diagonal shifting can save frames over normal non-bounce movement. However, they did not use this in earlier parts of the run and only started appearing about 3 hours in. -One of the largest and most beneficial bugs is the Kindness underflow, which I believe is completely absent from this run. This is the keystone glitch that is abused in RTA for significantly faster boss fights. In a TAS, this is even more potent since you can manipulate critical hits and significantly cut down on preparation time vs what's done in this submission. -Another critical bug that is not utilized is the Rune Unite bug, which lets characters piggy-back off of a single Level 4 spell to cast the powerful Unite magic. I don't think this is used a single time over the course of the run, and would stand to significantly speed up boss fights from the midgame onward. -Speed is seldom utilized in this run. One key component of speeding up boss fights is making sure you can attack before they can. For many characters this isn't practical, but the Spark rune helps with this by matching all following characters turn orders to the Spark user's. I don't think this was utilized anywhere. -Somewhat related to the prior two points, a special bug for abusing Rune Unites to grant a 30% speed boost is not used. This is a very tricky bug to explain, but it is understood, and is planned for in the current RTA routes. It does require careful control of item and equipment slots. -Spell animation times are a significant source of time loss throughout the run. The current RTA route heavily optimizes for this, and in fact leans on combining the Kindness underflow with a specific, very fast rune ability to clear out the enemy mobs quite a bit faster. -Experience routing is a critical element in this game, throughout the entire course of the run. While the author does abuse some power-leveling in the earlier portion, they miss other significant opportunities that would have cut down on the number of battles, and also had benefits in later boss fights with a slimmer character pool. -Because this route relies heavily on magic for raw damage, there is a huge amount of inventory and character management going on that eats up a lot of time. It could be significantly improved if the strategy were changed. -Using L or R buttons in War Battles lets you jump between units. In several cases, the author would have saved time from using these instead of manual cursor navigation. Specific points of improvement, separate from route: -1st fight in Rakutei Mountain: could have finished the fight sooner by manipulating a counter against the Commander. -Farming double beats is generally inefficient compared to power-leveling and using the Kindness underflow. -Picking up the Peeing Boy statues serves an additional purpose for recruiting Gordon, but I still question whether there would have been a better opportunity elsewhere. -The very first War Battle can end sooner if Solon Zhee is taken out. It requires some mild cooperation and luck from the NPCs though. -In RTA routes for all categories, the standard is to prepare Jowy to get a power-leveled boost in Sindar Ruins to level ~30. This significantly speeds up the Double Head fight, and reduces time spent in some of the other encounters. -It is hard for me to imagine the early Matilda Gate being beneficial overall. The Harpy fight is painfully slow, and you have better opportunities to power-level in the region later. The actual strategy of letting the Birdmen carry members away was clever, but still wholly too slow compared to other options. -Bosses throughout the entire midsection are quite slow. This is from relying on magic, rather than optimized Kindness and Violence strategies. -Some recruiting order choices are questionable. Again, it's hard to compare the strategies, but for example getting Sid on the return trip from Greenhill is a costly choice for just one additional castle level 2 star. -There was an instance to save several seconds in Banner Pass with an Escape Talisman on the last screen. -Ridley War Battle could have gone somewhat better, considering there's fewer things to control and manipulate for. -The Luca fight was extremely slow, due to relying heavily on magic. There are also new-ish bugs abused in the RTA route that significantly speed up this fight and abuse certain characteristics. -The author could have skipped the entire Sierra introduction cutscene. This is a new trick as of ~2 years ago, and saves a minute at almost no cost. -The author does not utilize an Escape Talisman in Tinto Mines after the Golem for the "Zombie Skip." It's estimated to save about a minute. -The author gets into an extra battle in Greenhill that can be avoided with just movement. -All battles in the latter half of the game are quite slow, including regular enemy battles, due to not abusing faster spells, rune unites, or Kindness underflow. -Could have removed Hoi during the Vincent recruitment to remove an additional cutscene in the Castle. -Could have included Pasmerga in the party instead of upgrading weapons with the Silver Hammer to recruit Gen. There's plenty more I can say, but with the above comments it should be clear that this run is not optimized, even to the best of current knowledge. This is a very tough game to optimize since there are so many decisions that have impacts throughout the run duration, but I would at least like to see the run be up-to-date with the current best known strategies, glitches, and movement. One last thing that I'd like to add is that the author should connect with the rest of the speedrunning community for this game if they'd like to improve this submission. As far as I know, despite being credited in the notes, myself and the active portion of the RTA communities were not consulted for this TAS which would have aided its quality and planning significantly. As it stands, it needs a full rework from beginning to end. That said, the few runners of this game have invested years in understanding all the different mechanics, time tradeoffs and otherwise. I'd really love to see what the limits of running this game are, and am happy to provide my knowledge towards that, but this run is not it.

I only saw this submission once it was up on the TASVideos YouTube, but I have a lot to add about its optimization and planning. The briefest way I can summarize this run is that the planning it used was naive, or at least quite outdated as compared to current RTA strategies. This has a significant impact on the entire course of the run. For comparison, consider the current RTA record in the same category: https://www.youtube.com/watch?v=7SrKeb3AlKU With a comparable route and further luck abuse I think a more optimized TAS could be about 10-15 minutes faster than this submission. I've listed some specific points for improvement below.

That would still be less than 4% of the whole duration of this movie.

There's plenty more I can say, but with the above comments it should be clear that this run is not optimized, even to the best of current knowledge. This is a very tough game to optimize since there are so many decisions that have impacts throughout the run duration, but I would at least like to see the run be up-to-date with the current best known strategies, glitches, and movement. One last thing that I'd like to add is that the author should connect with the rest of the speedrunning community for this game if they'd like to improve this submission. As far as I know, despite being credited in the notes, myself and the active portion of the RTA communities were not consulted for this TAS which would have aided its quality and planning significantly. As it stands, it needs a full rework from beginning to end. That said, the few runners of this game have invested years in understanding all the different mechanics, time tradeoffs and otherwise. I'd really love to see what the limits of running this game are, and am happy to provide my knowledge towards that, but this run is not it.

Since implementing the modern strategies would require redoing the whole thing, and the resulting time difference is not dozens of percents, I don't think accepting this run was a mistake. Otherwise I agree that when planning to TAS some game, one's best bet is getting in contact with the community that speedruns it, tool-assisted or not.

I think in general it bears noting that RPG movies, particularly ones as long as this one, are enormously tough to judge. Many strategy or routing choices look obvious to an expert of the game, but are hard to notice or grasp to anybody that isn't. All a judge could really do in this case is either give the author the benefit of the doubt, or to search out for other experts of the game to get them to weigh in on the movie. The latter is not always an easy or even viable option. Despite many of the criticisms noticed, I don't think this movie would necessarily be perceived as suboptimal to a more casual onlooker without the inside knowledge of Omnigamer's post. That still fulfills one of the key aims of the judgment process: we cannot expect every movie to look perfect to everybody, but they need to look clean to the general audience. If this movie passes that criteria, then this is not necessarily a wrong judgment. As for the experts, since everything that's needed to improve the movie is known and documented here, one of them could make a new submission that fixes all these issues and obsoletes this movie.

To be clear, I'm not saying the movie should be taken off the site whole-sale. But it should not be regarded at Moon level, at least. Being only 3.5% faster than an RTA run which cannot manipulate perfect luck, have perfect dialogs and menus, and have ideal movement alone is not distinguishing it from the best of RTA.

That would still be less than 4% of the whole duration of this movie.

OK, I was being fairly conservative with my initial estimate. Summing up just the improvements from small errors and strategy omissions in the current run would net around 10 minutes without changing anything about the route. More realistically, after building on an improved route, the total improvement could be 30 minutes or more. I also just want to say that it's somewhat deceptive to just look at the percentage of total runtime, especially for RPGs. Some 25-50% of the total runtime is unskippable dialog, cutscenes, and other non-optimizable content. I recognize that the full runtime is all an end observer has to go on, but it misses the point that saying 15 minutes of gameplay optimization is actually a significant overall improvement over the portions of the game that the player can influence. I also want to note that my comments were not a sleight on the judges for accepting this initially. A lot of this was very specific information that couldn't have been known or followed without staying on top of what the RTA communities had been developing for the game, especially in the last several years. You went on your best judgment, and parsing 7 hours of content for optimization is rough, with or without the extra nuance from domain knowledge. In the end, I just want to see a great TAS for this game. That starts with listing the known flaws in the current one, so that either the author or somebody else can recognize opportunities for improvement later on.

First off, thanks zggzdydp for doing this! I should mention back in October, zggzdydp and I did exchange a couple privates messages, since I used to TAS this game and maybe I could be helpful. Contacting Omnigamer or posting on the forum would have been helpful, yet I was quite busy IRL at the time(I had an important injury and even now I may have an another operation later). Also, I'm not fully up-to-date with all the time-saver on this game. Some other consideration to think about: wouldn't it be more entertaining by not going thru with the early Matilda glitch as opposed to other ideas which might be more enjoyable to watch? This is certainly questionable as the former seem to loose of time in exchange for security in battle(being over-level). Omnigamer technical mention are on-point. There's definitely several good chunk of optimization for this run, as some mistakes could be prevented with better planning, research or redoing parts to skip an event, etc. Stilll, diving in a long RPG isn't easy task. It take a lot of determination. Now, IMHO sometime I would rather see a RPG TAS made with a naive/simple/decent route first while being clean enough on the "overall execution", then see how it can be massively improved afterward.

I understand everyone's comments, different people have different views on a matter from different angles, for the general audience and expert of the game, their opinions are very different. It was my fault that have not communicated to other people before did the run. We should stop argument, I think the more positive thing is researching new technology and doing a brand new run. Maybe I'll consider about that.

Maybe I'll consider about that.

I would be looking forward to it. You'll need a setup for lua script, so it get much easier to manipulate luck, if you were doing it manually. Or if you already have lua scripts, it could be worth to share it, same goes for omni if he has any(sadly mine got lost in a hard drive failure). On another note at some point, getting a subtitle/audio commentary in order to explain the technical aspect, plot or explaining the original story about "Water Margin"as well would be awesome, since it's quite a long run to watch.