This game does not quite fit in with the standard definition of having Individual Levels and a Full-Game. Having granularity allows more people to participate, potentially finding the "perfect speedrun", but comes at a cost of more maintenance and may potentially be redundant. (Speaking of which, the game appears to place some importance to getting no-hit, which seems very grindy with the number of levels you have to do, but may be possible.)
Individual Levels can be classified as one category for: IL1. Each (non-)spell IL2. Each scene IL3. Each character played through
I would not recommend IL1 as starting position is important with some bosses. IL2 presents a slight issue where difficulty can 't really be discerned except through boss health and damage from certain attacks, and there aren't great cues for determining when timing should start since the boss could have been played on first load or after a reset.
Full-Game can be classified as one category for: FG1. Each character played through FG2. All characters played through (Flandre A or B or C) FG3. All content played through (Flandre A and B and C)
The problem with FG2 and FG3 is that the game does not seem to handle resources well and that there appears an inconsistent impact of load screens, keeping in mind that the value of tracking real time mostly comes from the existence of tricks where you spend more real time to reduce in-game time (restarting Flandre's tutorial 10 times in 1.04). That said, it is also appropriate to do FG1 as it is now but with an All Characters category as well.
Thank you for your valuable feedback. I read the sentences with the help of machine translation in some places, so I'm sorry if there are any discrepancies in the fine nuances.
I personally feel that it is controversial as to whether in-game time or actual time is better. It's a very delicate issue with rankings, so if we change it, we should discuss it in an open space like this forum, not in a closed space. If there is no objection to maintaining the status quo, I think it is appropriate to maintain it, but if anyone thinks that it is better to change it, I would appreciate your opinion.
Apart from that, I feel the need to create a category for All Characters. In fact, ello had run many times with that regulation last December.
I find it difficult to handle Flan in that case. ello voluntarily chose C to make the true ending appear, but of course A and C have very different difficulty levels and times.
There is also a glitch that makes A's in-game time much shorter than it really is, so there's the question of what to do with FG1. In addition, Kanako and Yorigami are definitely faster at 1.02 than at 1.04, so if you're aiming for a record with FG1, you won't be able to use the ending skip at 1.04. It simply increases the burden on the runner and is not desirable. For the above reasons, I personally think that FG2 or 3 is good.
FG3 is good for fairness, but it certainly makes one run longer. I think it's okay to prepare both categories 2 and 3 for this, but I would appreciate the opinions of the runners.
My sincere apologies for the delay in addressing this matter. I intended to create a thread to determine the community's opinion on "100%" or "All Routes" last month. However, I have been very busy. I have not even had time to play. Moreover, communicating in English is very difficult and time-consuming for me. I wanted to be inclusive of all members of the community, but as a result, I was far too slow to fulfill my responsibility.
Regarding categories. I believe it is fine to have both "FG2" and "FG3". "All Characters" (FG2) can coexist with "All Routes"/"All Endings"/"100%" (FG3). If there is interest from players. Personally, I intend to do "FG3" when I have time.
Regarding in-game time / real time. I have two concerns with the usage of real time. ① Inconsistent load times may cause players to have a competitive disadvantage. ② For "FG2/FG3", it would save time to restart the game to skip the ending between routes. However, I have observed issues with OBS detecting the game after restarting. I think it may be better to not skip the endings.
This is the extent of the my opinion. Of course, any decision should be based on the community opinion.
Hi. Personally I like IGT better for individual scenario runs. It removes a lot of external variables and keeps the timing focused on the game, not on mashing text. As for full game runs, the option I see viable is adding up all IGTs at the end but disallow stalling at menus/text. While Menuing tends to be a skill that many speedruns have, the menus here are not extremely interesting, but we don't want people just staring at the menus or pause screen while they plan their next move. The other thing to take into consideration is the order in which to play the scenarios, a full game run opens the door for playing the scenarios in any order, we will have to consider if that is allowed as well. On my end I'd rather it be in the intended order, but I don't oppose any order to be allowed.
Hello sorry for the late reply, but i welcome the idea off adding full game runs whether it be one ending or 100% both are nice. As for timing, I have just been using real time since i've have just been comparing against myself and that is the simplest way. However if it is made a category i can understand timing by IGT so everyone is on the same footing. I just don't want to jeopardize what makes a speedrun a speedrun. If we we're to use IGT i prefer people use 1.04 so you don't have to wait 10 minutes for endings or close the game and have trouble with OBS recording the game. As for order of story i have just gone in order but i'm not against going in any order but i would prefer if it was done in a proper order, as in the way you unlock them. That being said i would rather not have to do the tutorials and focus on just completing all the stories so whatever people prefer.
RE: real time vs. game time, the main objective in speedrunning is to have fast gameplay based on a given approach. This game has intermissions between the actual gameplay, so in an ideal situation where all runners can be trusted, game time ought to take precedence in ranking with real time being a sort of tie-breaker in a given category.
For the sake of convenience and impartiality (and lack of a visual for the in-game timer) though, I find it acceptable to collect both and rank by real time, as it follows lower game time would result in lower real time. I'd estimate that on average non-standard buffering time plus mistakes in skipping stuff would only make at most a difference of 10 seconds in a character run, which is rather negligible and offers an area to improve on. Ultimately, the viewer can be the judge of what's more impressive if top runs do end up being affected, as unlikely as it is to occur.
However, it becomes more of a problem in multiple ending categories solely because you're exposed to more load screens. Remapping the pause key via config may help somewhat, but I think it should be the game offering things like an autoskip setting to keep such things uniform.
It's been more than a week since last opinion came out, so I'd like to share my personal opinion.
There is no doubt that it is not desirable that the Full-Game category is not created as it is, so I would like to proceed with the creation. To sum up the ideas of everyone who gave us their opinions on this occasion, IGT will be used for measurement.
All characters: All characters played through (Flandre A or B or C) 100%: All content played through (Flandre A and B and C)
I'm thinking of creating the above two categories.
Also, the CD media product I purchased in Japan the other day is version 1.03, which is the same as 1.04 except that tutorials for characters other than Reimu and Marisa have not been introduced. It seems that. If you decide to include the tutorial in Full-Game time, getting this 1.03 is the best solution. I don't like that because it's a game that you can easily buy on Steam. Therefore, I would like to exclude the tutorial from Full-Game and measure the total time of each simple character's story.
Regarding the categories by character, we will continue the specifications of IGT as it is, considering the comprehensive opinion. If you have any opinions or differences in my perception, please feel free to contact me.
The idea of separating categories by version was proposed by myself before, but I later reconsidered that it was not desirable to add too many categories. However, I also think it is important to create an environment where people can compare different versions. I'm doing a run of Kanako, which is best played in 1.02, so I feel even more strongly that we should have an environment where people can compete with their 1.04 times. Since I'm not an experienced moderator, I haven't figured out how to create a comparative environment without separating categories, but I'm going to try to find a way to do that.
I also find the idea of creating a category for tutorials interesting. I'm thinking of treating it as Misc since it's out of the mainstream, but I think I'll create one for each character.
I'm also thinking of creating a dash-only category for Marisa in Misc as well.
If there are no objections, I'll add the Full-Game category, the Misc tutorial and Marisa's dash-only in the next few days.
With the board format now it is possible to use the filters and look at each version individually, but it has its disadvantages:
- Newcomers are likely playing on the most current version
- People might not submit runs if they see that they won't be ranked
- People who have played with a significant advantage on an older version will instantly have their current version runs obsoleted(?)
With the suggested current version-compatible format there is the risk that today's runs may end up lumped into the incompatible category, but there is not much that can be done with the game being technically unfinished. The worst case of this that I'm aware of between version 1.02 and 1.04 is the Yorigami route not having to play through 3 of Niwatari's spells. This also allows for people to submit runs on the demo(s), which is most likely going to be fastest because there is not as much content as current, although with the current rules they are disqualified as they don't have Toutetsu.
Of course, people are free to submit what they like and can have it moved to an appropriate category once it gains enough interest, so it's not particularly urgent.
I think Yorigami is particularly problematic because of the inter-version issues, and I'm struggling with that myself; Kanako is also having a very hard time beating 1.02 record due to the spec changes for herself and the Battle of Kutaka, but the impact is limited compared to Yorigami.
I'm considering separating the pre-1.02 and post-1.04 categories for Yorigami, since the threshold for new players is getting higher and the update has changed what Yorigami can do. I'm also considering doing the same for Flan's Any%, which has undergone a major spec change with the update.
There is a principle issue to consider with the creation of version categories. If Yorigami has v1.02 and v1.04, one version will be considered default. I do not believe people will compete on both versions. The question: is the faster version better, or is the newest version better? Each community has its own preference. It's a difficult question. Unfortunately, there was no response to SAS_rta's thread regarding this subject. If the community had a clear opinion, I would have made a decision then. However, no opinion was offered, so it was hard to decide. That being the case, I thought the status quo was okay for the time being. At least, it is possible to filter by version currently. I have no objection to creating a v1.04 category for specific characters if that is desired.
A different category does not necessarily suggest that one is more valuable than the other, only that the things inside them can not be compared. However, the nature of this game seems to apply more focus on mastering the movesets and patterns than on reaching the ending of the story, so I would lean more towards versions that contain all intended content.
Even if new versions were considered default, I think people would still play with characters they like on incompatible versions provided they have access to them and don't mind the differences, especially if it does not take as much effort to rank.
This is only a matter of perception. If people are aware of the circumstances, it does not matter, so it is good to have a discussion about it.
I've added the FULL-GAME category, as well as Marisa's dash-only and each character's tutorial in the Misc section. If you have any requests, please contact me.
I also feel that the versioning issue is a very difficult one. This game will probably not end with ver 1.04, and there will be more updates in the future. Because as you all know, the "剛欲な挑戦" at the bottom of the chapter selection has not been implemented yet.
Therefore, I fear that if we default to the latest version every time a version is updated, it may affect the motivation of the runners, because it also means that their records will be reset almost every time there is an update. I'm thinking of keeping the current system for the time being, but I'd be happy to hear the opinions of the runners if you like.
Thank you very much for your valuable feedback too. I would like to continue to study English and do my best.
Come to think of it, I think the two best ways to handle Flandre in the Full-game are to effectively clear only the A route or all of the ABCs.
The situation where the A route ends up being the last one, where Yuuma stands up and cannot be defeated, or where you have to play Flandre three times in a row, might be a little awkward for the category of playing the whole game, especially when there are opportunities to play the whole game in speedrun events.
So I feel that it might be worthwhile to create a True End category, which is somewhere between All characters and 100%, and ends with the right route being played only once. If anyone has any opinions on this point, I'd appreciate your input.
I also find it awkward for Flandre to fight Utsuho 2 times, Yuuma A 3 times and Yuuma B 2 times for the full 100%, but I believe that's the requirement to get rid of the exclamation bubble when you have a new game. A way to remedy it somewhat is to do A and 3B+C (where one does 3 stages of route B before finishing the run in route C). This way one will only have to fight Yuuma A 2 times if you are not going to employ something like quitting mid-run.
On the justification of using specific routes for one ending:
- Route A is a technicality with reaching an ending. It is the fastest because there is less content to play through.
- Route B is fine since Yuuma B is where the conventional combat ends and matches the character progression (i.e. Kanako, Murasa, and Joon also had to fight Yuuma B )
- Route C unlocks Yuuma C, so it is more complete than the other two routes.
For a compromise, you can have it as a 3A3C4B (since B is the middle door, one must not fight Yorigami until it's forced) run, so the only content that is not played is Reimu C and Yuuma C, which is probably worth the sacrifice to play the other bosses. Although as that is a disproportionate amount of time spent playing Flandre over other characters, I think just route C is fine too.
In this game, you can fight all enemy characters in the process of making Flanders playable. If I am correct, the attack patterns of the opponents are basically the same for each enemy, even though our characters are different. However, Reimu C and Yuuma C have a distinct peculiarity in that their attack patterns are very different from the previous ones.
There is a lot of value for players in "using Flandre to fight all opponents," but I recognize that there is a 100% category for that.
The reason I suggested the True End category was simple: I thought there was a certain value in fighting Reimu C and Yuuma C without being redundant, and making Flandre's best ending (which is basically skipped) appear.
The suggestion to play against as many Flandre opponents as possible for one ending comes from seeing that people seem to prefer specializing on A or B or C, so to be neutral it would involve combining them like that or having 3 separate categories in the All Characters run.
Favoring one over the others is also acceptable with consensus.
