Great work. I'm already using Autosplitter and it really helps my run a lot.

I should mention something related to load-removed time (LRT): for the latest version, it is faster to turn off retry menu than turn it on in Rikyu and my environment. There's about 0.4s difference for each stage. (In my past comparison video with older version, this result is different from my old computer, so computer spec might affect with this issue)

In Rikyu's prototype load remover, we can only detect loading flag when we turn on retry menu. For current specification afaik, the flag becomes active when we click the arrow in retry menu. If we want to use the load remover in Any% without retry menu, we need to find out other flags (e.g. stars' appearance ~ transition of background color) and add information to Autosplitter too, and this may take more time. (I think Rikyu already working on to solve this issue, according to his comment on Japanese speedrun community)

To add to this, in prototype load remover, we cannot detect the loading time of the first stage. If we were competing for accuracy to the nearest tenth of a second, this issue is also worth considering. However, in the current leaderboard situation, I think just reducing the overall delay in the other stages would be sufficient.

I personally think fine about merging category and adding version variables when the load removing features becomes available. Thank you for responding to runners' requests every time.

I did additional comparison about delay difference about PC spec/environment. (Thanks @RiKyu for sharing recording data!)

https://twitter.com/monochroparadox/status/1700469431074488762?s=20

In conclusion, My recordings took more loading time than Rikyu's. Also on my computer, the delay differences increased in Deep Clean (about 0.2s for shortest, and over 4s for longest / in legacy there were less than 0.2s). I hope this test helps for retiming discussion or establishment of new regulation.

• As I wrote in my previous comparison video, the transition delay is not always same for each stages/runs. Also, my videos are 30fps and Rikyu's are 60fps, there may have another few flames' differences.

In 100% category I also agree that using retry menu in Deep Clean makes our run much faster and comfortable, even they have longer transition delay than legacy (I haven't checked how effective it actually is in DLC 100% yet, but I think it works better for same reason as original 100%). To talk about future runs, based on the information afaik, currently I am planning to use Deep Clean for 100% and legacy for Any%.

[Add] I checked DLC 100% with Deep Clean, and felt that the effectiveness changes by player's environment, but not much as original 100%. On my computer, waiting time for level menu is approximately 30sec, and I had at least 1.5sec delay in each stage for selecting menu and stage transition (except multiverse drawer: 28 transitions), so on the calculation my Deep Clean DLC 100% should be a bit longer than legacy.

For current/future runners who want to know how it actually differs, here's a sample comparison clip for DLC on my PC. Using retry menu makes Deep-Cleaned run a bit faster. (link to my Twitter/X post)

https://twitter.com/monochroparadox/status/1700089833837019518?s=20

@TFGA_tFoMS: Thank you for the patch update and the information about cursor assist! In speedrun I have same idea about cursor assist as you mentioned.

Yeah I agree that bugfixes make our speedrun stable. To talk about only trophy cabinet puzzle, personally I've lost nearly or more than 1 minute for several times because of the bug (as I wrote above) and I understand that it is safe to use latest version. (I haven't run DLC categories after update, so I don't reap its benefit yet, will try and check)

At the same time, it is not always sure that which loss (transition delay or bug loss time) will make our run slower until we try, because it is not always that bug occurs. The ideas for the speedrun depend on each runner: who want to run faster anyway even they have risk, or who want to do stable speedrun even it can be a bit slower than risky version etc. If we add same seconds for down patched run and put them in a same category, it is much more easier and simple way to organize, but it possibly means that we ignore the risk they took. (Of course I can't tell how many active or near future runners having former idea, and that also connects my idea of "premature". I knew that I am too careful person, but I think it's better to have time and make decision after finding out how many run with down-patched version submitted/planned). (Just to be sure, my comments does NOT mean I totally disagree ideas for retiming, it's kind a matter of choice!)

By the way again, How about other new options (cursor assist, effect etc) for you all? I'm dull and honestly still haven't find out what is changed, is there any good improvement for speedrun?

Also I just tried (and submitted) Archive 100%, and felt that it is effective to set version variables on 100% categories. I've turned on retry menu at every levels and thought that operation accuracy for selecting retry/level/continue is also can be a part of competitive element. (*anyone please let me know if there's any other faster way)

Thank you for the really important announcement. Exactly I am also one of the runners who is interested in the level transition delay and new categories. (I've just posted about this topic on my Twitter/X few days ago)

These are my personal opinions (and impressions) about latest update:

• Currently I prefer to make subcategories (version variables) and leave old records (or down-patched records) as they are.
• One thing I wonder: even ALttL doesn't require higher PC spec, runner's environment may possibly effects for longer/shorter transition times. (I haven't checked exact differences, so it there's no problem about that, please ignore this comment) If runners want to strictly compete their time and if we have differences as I mentioned by our environment, what we need to think about is reducing loading time (by load remover/AutoSplit or something), not to gain additional time and make our speedruns slower. (A topic about additional tools should be on another note and I will say no more here)
• As a runners who's using Steam, I personally feel that it's not complicated to down patch. (I am doing speedrun with down-patched version in other games and tried down patch of ALttL including DLC today. Only thing we need to do is prepare data for previous version and exchange them with current data files. It will complete less than 10 minutes, and I can share how-to if there's anyone who needs help) One problem that came up in my mind was the numbers of Daily Tidy and other achievements, but fortunately it didn't lost after down-patched.
• I think it is bit premature to exclude choice of faster down-patched runs. If it becomes sure that there is no runner who runs with older version in the future, then it will be proper (and not too late) to have discussion again about retiming or deleting subcategories. (I knew and sorry for mods to make extra tasks if it goes as I said...)
• I've wrote down a lot, but I'll respect all opinions from every runners and finally I'll follow decision by moderators.
• By the way, I am so happy and welcoming about the trophy cabinet patch! This trouble often makes my DLC run ruined and now I can play it calmly!
• As an aside by a runner of miscellaneous categories, I don't mind removing my misc. records (Good Tidyings & Something Eggstra) and unify them as Archive 100% category. (Actually I am going to practice Archive 100% run right after this post lmao) Please feel free to clean them up if it's needed (both now or in the future)!

Last but not least: thank you so much for all moderators and runners who support this community. English is not my mother tongue and I hope my translation is making sense. Please tell me if there's anything I can help for ALttL speedrun (including Japanese translation, retiming, verifying or moderating).