コメント
スレッド: Diversity
CanadaCavin8562 months ago

As of the time writing this, the as-of-now new WR for the Glitchless (No Preserve) category has led me to have some concerns. These concerns have led me to conclude that the only way to maintain a competitive and fair leaderboard is to require stricter verification.

Thus, as of July 22nd, 2024, any WR run in any category will now require showing the "stat.playOneMinute" value, immediately after the run. Please ensure that it is clear which world the file is coming from.

Additionally, for the Glitchless (With Preserve) category, as of July 22nd, 2024, all runs faster than 3:55 (3 minutes, 55 seconds), will also require showing "stat.playOneMinute".

Additionally, for the Glitchless (No Preserve) category, as of July 22nd, 2024, all runs faster than 18:50 (18 minutes, 50 seconds), will also require showing "stat.playOneMinute".

For more information on how to find this stat, please refer to the December 3rd, 2022 Timing Policy Update, found here: https://www.speedrun.com/mcm_diversity/forums/yok89

It should be noted that if you load the world and don't immediately start your run (to let the world render in, as timing starts on first input, officially), you should immediately pause and show "stat.playOneMinute". This will allow moderators to subtract the frames from which you loaded in before starting the run off of your final time. Not doing this may result in losing some frames.

As stated in the previous timing update, the rule changes may apply to other categories if the top level times because close enough to warrant it. At the time, this was mainly in reference to the Any% category, which currently employs a benchmark threshold requiring all runs to show "stat.playOneMinute". I believe it would be valuable at this time to apply this same benchmark protocol to the Glitchless (With Preseve) and Glitchless (No Preserve) categories. Following the precedent in the previous update, all new runs faster than that categories benchmark that does not show "stat.playOneMinute", will have their time set to the benchmark. All runs currently on the leaderboard that do not show "stat.playOneMinute" will be grandfathered in by having their milliseconds set to 999, or according to the in-game time if an in-game timer is shown in the video.

The reason for this change is that in the aforementioned WR run, a major source of time save over the previous WR run was load times between branches. Although I didn't track exactly now much time was saved from load times alone, it's clear that it's certainly possible for a potential future run to be submitted where the only reason it might be WR is because of faster load times.

I should emphasize, this is NOT to say that the current new WR run is invalid or undeserving of it's placement. On the contrary, it was clear that it did perform better than the previous run in several places. Rather, that there exists the potential for load times to be the deciding factor in future WR runs. Thus, these rules changes are designed to try and eliminate that.

ラン: Seedling
CanadaCavin8561 year ago
このコメントは削除されました
スレッド: Diversity
CanadaCavin8561 year ago

Due to the optimization of the Any% category turning into a frame war at the top of the leaderboards, it's been determined that a standardization is needed to accurately compare times that are merely frames off.

Thus, the following rule change for Any% is in effect immediately:

All runs submitted from December 3rd, 2022 onwards, faster than 2m 25s 000ms will now REQUIRE you, both before and after the run, to show your stats .json file and show the value for "stat.playOneMinute".

To find this .json file, navigate to "C:\Users\YOUR_NAME\AppData\Roaming.minecraft\saves" (or wherever your saves folder is located, if different). Find the Diversity world (should be called "Diversity (v1.3.4)"), then open the "stats" folder, and then open the .json file inside (you can use any basic notepad program to open this).

Runs slower than this benchmark will not require this, and instead a simple timer can be used, HOWEVER if the milliseconds cannot be determined from the video, then the milliseconds will be set to 999ms. If the milliseconds are shown, they will be rounded up to the nearest multiple of 050ms.

Runs faster than this benchmark that do not show the stats .json file WILL BE REJECTED (or have their time set to the benchmark).

The value shown in the stats .json file is in in-game ticks. There are 20 minecraft ticks per second, so take the number after the run, subtract it from the number before the run, and then divide by 20, which gives you your time in seconds. For example, the current WR run goes as follows: 2767 - 1 = 2766 2766 / 20 = 138.3 138.3 is 2m, 18s, 300ms.

Your milliseconds will always be either X50 or X00, with X being any integer (number) from 0-9.

Current leaderboard runs have had their milliseconds already updated to reflect the new changes, with runs faster than the any% benchmark being adjusted as if they were slower than the benchmark, to avoid having to remove old runs. This is a one-time grace that was only granted to old runs. New runs are subject to the new rules.

If other categories ever become as competitive or have top level times within seconds of each other, than these rules will apply to those categories as well, with appropriate benchmarks being decided if needed. However, regardless of category, showing the stats .json file is still preferred, in addition to showing on-screen timers.

Regarding on-screen timers, Any% currently will not require an on-screen timer if one shows the stats .json file, as the presence of Save + Quit and the shortness of the category would invalidate the precision of most timers.

ALL OTHER CATEGORIES will currently only require an on-screen timer for WR runs regardless of if the stats .json file is shown or not (however, also showing the .json file will help with insuring a precise and verifiable IGT). For non WR runs, only one of these timing methods will be required. Again, this is subject to change should other categories become more competitive.

It is important to also note that: For the stats .json folder to count as a valid timing method, IT MUST BE SHOWN VISIBLY in the video. The ideal order of things would go like this:

  1. Start the recording and open the stats folder
  2. Enter the Diversity map world
  3. Immediately pause and show the "stat.playOneMinute" in the .json file (it should be pretty much the only stat there)
  4. Do the run
  5. Immediately pause and show the "stat.playOneMinute" again in the .json file. Now there should be several stats from the course of the run. Make sure that "stat.playOneMinute" is visible. If you want further verification for safety, scroll through the rest of the file on the recording, and in a dispute, the mods can compare the stats with the recorded run (although this is not required in any category currently).

I hope these rule changes are not strenuous for the community, and I thank you for your patience in this.

RespectYou これを好き
スレッド: Diversity
CanadaCavin8564 years ago

On screen timers are a way to improve the validity of runs. They are harder to fabricate in general principle, and also help with accuracy and proofcalling. If you need timers, you can check out the guide in the Java Edition Minecraft pages. It is highly encouraged to have visible on screen timers, especially for WR runs.

スレッド: Diversity
CanadaCavin8565 years ago

Lets not make a mountain out of a mole hill. The point of speedrunning is as a community effort to lower the time taken to complete various categories decided by the community. Without the boat OoB, the category would be some combination of Survival Branch Only and Glitchless no Preserve. The point of the way the categories are split is to have as few redundancies as possible. Each category has a fairly different route, and that's the beauty of it. We could add 10 more categories and have all of them be combinations of the other ones, but that's just silly. The rule is vague enough as it is, so instead we'll use common sense and reason to not box ourselves in with hard, fast, rigid rules that are black and white, rules of our own creation. The verified runs exist, and I'm also not about to delete them just because they violate some sudden rule change. Although the rules do seem to include boats, by doing so we'd cause much many more issues to crop up. The point of the category is to showcase the preserve primarily, and I feel it does this fairly well. So in conclusion, I see no reason to change what we've already established. The categories are good enough as is, as changing them would cause many more problems to crop up.

スレッド: Diversity
CanadaCavin8565 years ago

I'd thought you could do 2 freebies like that, but I tried for quite a while and every time, I was teleported before the second wool was given. Either way, for the purposes of not having too many redundant categories, we're just going to keep it at 1 freebie. And I don't know if there is anything to do in dropper for preserve.

スレッド: Diversity
CanadaCavin8565 years ago

Not only is it redundant for the rules to say that multiple freebies aren't allowed in an All Branches run where any freebies aren't allowed, multiple freebies aren't allowed in any run except any%.

Obtaining multiple freebies is due to a minecraft glitch rather than a map exploit, due to it requiring a S+Q. In the glitchless categories, S+Q is banned, thus multiple freebie wools are not allowed.

In All Branches, no freebie is allowed because you are getting all the wool from every branch,

Additionally, infinite freebie wool it just any% anyway, there was an old any% speedrun from a while back that used infinite freebies. So even if there was a way to get them without S+Q, it is still banned in that category due to it essentially obsoleting the category itself.

スレッド: Diversity
CanadaCavin8565 years ago

you are looking in the wrong spot. It's under no exploits because that is the only applicable category

Timing starts on movement and ends upon respawning on top of the wall facing the Diversity logo. You must get the wool from their respective spots in each branch, you may not skip to the end or take the freebie wool.

You may not use any glitches in Minecraft, nor use any exploits in the map mechanics.

The following exploits are forbidden in this category: -- Skipping portions of the labyrinth branch due to damage boosting or surviving a dip in lava. -- Jumping into the redstone below the Boss Battle Branch and altering the contents of the hopper to skip fighting the Witherboss. -- Preserving items from any branch outside of that respective branch. -- Activating command blocks from outside of the bedrock box in the Survival branch. -- Obtaining multiple freebie wools. <----------- HERE

This is not an exhaustive list, these are just the currently known exploits. Any new exploit not on this list is also considered banned.

  1. All map rules must be followed. Generally speaking, map rules trump all.
  2. No altering gameplay affecting options (difficulty, open to lan, etc.) after loading the map, unless instructed to do so by the map rules.
  3. No mods are allowed except optifine (assuming zoom is not utilized) and mods specifically required by the map.
  4. Use the correct Minecraft version for the map version you have.
スレッド: Diversity
CanadaCavin8565 years ago

It was added.

スレッド: Diversity
CanadaCavin8565 years ago

Yes you probably could. And All Branches remains a misc category for now because there are no runs.

スレッド: Diversity
CanadaCavin8565 years ago

yeah, therein lies a major problem, it would basically invalidate all previous runs, thus we cannot do it.

スレッド: Diversity
CanadaCavin8565 years ago

I was thinking it might be best to re-time IL's, starting on first movement and including getting the white wool etc. But currently, no preparation is allowed for IL's.

スレッド: Diversity
CanadaCavin8565 years ago

Yeah actually, that is a good idea. Hmm... yeah i guess in theory you could preserve stuff to other branches... Well, if we need another category, we'll make it when the run shows up that doesn't fit well into other categories. For now, we have plenty imo.

スレッド: Diversity
CanadaCavin8566 years ago

OK, the categories have been clarified. We now have All Branches, which allows all glitches and exploits, and a version which allows none. Additionally, No Exploits has been removed. This is done to remove as much overlap as possible, because personally I think it's kind of silly if we have two categories that are virtually identical save one gets the freebie and one does the arena branch instead.

ThatRaisinTho これを好き
スレッド: Diversity
CanadaCavin8566 years ago

This would involve only the survival branch and the branches that are faster than survival, like trivia, etc. I dont think this would be a good category as it would overlap too much with other categories and cause confusion.

スレッド: Diversity
CanadaCavin8566 years ago

Not sure how you could preserve ender pearls, as again we can't alter command blocks into the hub, but submit a run and we'll check it out. We can take a vote if we want glitches/exploits allowed for No Freebie/All Branches, or we could even make 2 versions of it.

スレッド: Diversity
CanadaCavin8566 years ago

Correct. The first player to move is when timing starts, not when player 1 starts. There is no player 1.

スレッド: Diversity
CanadaCavin8566 years ago

These sorts of runs aren't often done, and are generally kept to a separate leader board for big games etc. This won't be added, but definitely go for it.

スレッド: Diversity
CanadaCavin8566 years ago

Rules have been updated to reflect this. This is something I hadn't thought about before. I wonder how/if it would save much time somehow, by using the Wither status effect somehow.

スレッド: Diversity
CanadaCavin8566 years ago

All runs must have a visible timer, or some proof thereof. Going into the world files and showing the exact tick count would be acceptable, that can be converted into real time (20 ticks per second, 60 seconds per minute, etc). Doing this, however, you must make sure that you show that the world you played in and the world of the stats file you opened are the same, and that is has been completely unedited (so don't cut the recording etc, just keep recording until you find the tick count). Having a timer visible is still highly recommended. There are other accurate timers, like WSplit for instance.

Cavin856について
参加日
8 years ago
オンライン
today
走行
26
プレイしたゲーム
Minecraft: Java Edition
Minecraft: Java Edition
最終ラン 6 years ago
7
走行
Harvest Moon
Harvest Moon
最終ラン 4 years ago
6
走行
Seedling
Seedling
最終ラン 6 years ago
3
走行
Mario Party 64
Mario Party 64
最終ラン 4 years ago
3
走行
There is no Learning Curve
There is no Learning Curve
最終ラン 6 years ago
2
走行
The Legend of Zelda: A Link to the Past
1
走行
Diversity
Diversity
最終ラン 8 years ago
1
走行
The Legend of Zelda: A Link to the Past (NES)
1
走行
フォロー中のゲーム
Minecraft: Java Edition
Minecraft: Java Edition
最終訪問 1 year ago
490
訪問
Seedling
Seedling
最終訪問 10 months ago
95
訪問
Diversity
Diversity
最終訪問 1 year ago
1,019
訪問
Harvest Moon
Harvest Moon
最終訪問 1 year ago
322
訪問
The Legend of Zelda
The Legend of Zelda
最終訪問 1 year ago
76
訪問
Egg Hunt 2018: The Great Yolktales
Egg Hunt 2018: The Great Yolktales
最終訪問 1 year ago
64
訪問
Switchball
Switchball
最終訪問 11 months ago
10
訪問
Super Mario 63
Super Mario 63
最終訪問 3 years ago
8
訪問
モデレートしているゲーム
Diversity
Diversity
最終アクション 2 months ago
44
アクション