ScotlandThere's nothing wrong with ties, and as lag is a thing using milliseconds is definitely a bad idea with the current timing. In terms of timing for ILs I do still definitely think IGT would be better though, for reasons I've described before (I have a thread about it).
What's the current progress with fixing runs?
It doesn't matter if things take a few days to get verified. Many of the mods are busy in general, and ESA is currently happening to add on to that.
Eh it's not too important, but if you know it it would be nice to include it. And I don't think it's entirely because Aether left as there have just been a massive amount of runs submitted due to it being the summer.
Yeah I guess it's debatable whether it should be allowed. Either choosing a cap at the start is allowed for a Yoshi run or just banned completely.I saw there were a couple of runs that had already done it when I started doing ILs (such as http://www.speedrun.com/run/zq632v8y) so I did it myself, and I do believe it should be allowed, as it means some cool strategies are viable such as this one: . It also means that for stage 100% runs as Yoshi you can choose a different character for each star, rather than likely sticking as Yoshi as caps would be too far out of the way to get, which I think is more interesting.
If either of you do feel really strongly about it we could have a vote in the Discord server I guess, but it would mean removing a good few times if we decided it wasn't allowed.
It would have been easier to delete the old one and submit a new one. If you're gonna do it like this you need to actually edit the time of the run and also the date it was set. Here's another one that wasn't changed properly: http://www.speedrun.com/run/m3d1g26z. Please check them all yourself then those ones at least should be sorted.
For many ILs using a cap with Yoshi is the best strategy though, and as you've already done the Yoshi runs there's no reason to not have them up on the leaderboards. Some will definitely need to be moved from Mario/Luigi/Wario over to Yoshi at some point or just deleted, as these few will be impossible to beat with the correct timing. Will honestly be easier to just fix them all now with the times I calculated, and then you can replace as many or as little as you want with new runs, whenever you want.
It would be better to change them now. Even if you think you can manage to redo ~50 runs in a weekend, the current runs can all be submitted under Yoshi. Will be a lot less confusing to just sort everything now than to try and replace and move runs one at a time as you do new ones.
Here are (I believe) all the edits required. If you click on one of your runs there is a symbol in the top right corner with three dots, which brings up a menu which includes editing your run. Then change the time and/or the category. I wasn't sure how you timed the Yoshi runs for stars where you had been using the cap method, so if it was fade in the times stay as they are, or if it was gaining control you need to add 2s. Also wasn't sure if cap method was used for TTC/RR so change those or not depending on if it was. And JRB, BBH and LLL are fine if I understood what you said about where you used the cap method, but if not I can figure out the time to add on later.
I hope you reclaim all these records with the correct timing and character soon! On the plus side you have lots of new Yoshi records :P
BoB 100%: fine WF 100%: fine
BoB: +1.6s gaining control & +1.4s cap = +3s BoB 1: 1:01 -> 1:04, Mario -> Yoshi BoB 3: 0:58 -> 1:01, Mario -> Yoshi BoB 4: 0:57 -> 1:00, Mario -> Yoshi BoB 6: 0:59 -> 1:02, Mario -> Yoshi BoB 7: 0:13 -> 0:16, Luigi -> Yoshi All other BoB runs -> deleted
WF: +1.6s gaining control & +1.4s cap = +3s WF 1: 0:40 -> 0:43, Mario -> Yoshi WF 2: 0:16 -> 0:19, Luigi -> Yoshi WF 3: 0:09 -> 0:12, Mario -> Yoshi WF 4: 0:52 -> 0:55, Luigi -> Yoshi WF 5: 0:19 -> 0:22, Mario -> Yoshi WF 6: 0:31 -> 0:34, Mario -> Yoshi WF 7: 0:21 -> 0:24, Mario -> Yoshi All other WF runs -> deleted
CCM: +1.7s gaining control & +1.4s cap = +3.1s CCM 1: 0:28 -> 0:28/0:30 , keep as Yoshi CCM 2: 0:26 -> 0:29, Mario -> Yoshi CCM 3: 0:40 -> 0:43, Mario -> Yoshi CCM 6: 0:12 -> 0:15, Luigi -> Yoshi All other CCM runs -> deleted
JRB: fine?
BBH: fine?
LLL: fine?
HMC: +1.7s gaining control & +1.4s cap = +3.1s HMC 1: 0:55 -> 0:58, Mario -> Yoshi HMC 2: 1:45 -> 1:48, Mario -> Yoshi HMC 3: 1:11 -> 1:14, Mario -> Yoshi
DDD: +1.8s gaining control & +1.4s cap = +3.2s DDD 7: 0:20 -> 0:23, Luigi -> Yoshi All other DDD runs -> deleted
WDW: +1.9s gaining control & +1.4s cap = +3.3s medium +2.4s gaining control & +1.4s cap = +3.8s low WDW 1: 0:13 -> 0:16, keep as Yoshi (low) WDW 2: 0:13 -> 0:16, Luigi -> Yoshi (medium) WDW 3: 0:51 -> 0:55, Mario -> Yoshi (low) All other WDW runs -> deleted
TTM: +1.6s gaining control & +1.4s cap = +3s TTM 4: 1:07 -> 1:07/1:09 , keep as Yoshi The other TTM run -> deleted
SL: +1.6s gaining control & +1.4s cap = +3s SL 1: 0:10 -> 0:13, Wario -> Yoshi All other SL runs -> deleted
TTC: +1.8s gaining control & +1.4s cap = +3.2s TTC 4: 1:44 -> 1:47, Mario -> Yoshi (Was incorrect timing/character actually used here? Old run and only one character)
RR: +1.7s gaining control & +1.4s cap = +3.1s RR 5: 0:40 -> 0:43, Mario -> Yoshi (Was incorrect timing/character actually used here? Old run and only one character)
Secret stars PSS Mario: +1.5s (0:20 -> 0:22) Over the Rainbows black box star: +1.6s (0:07 -> 0:09) Other secret stars: fine
Okay so were JRB, BBH and LLL timed correctly? Or did you still start timing upon gaining control, but use the actual character?
Also if you used the cap method within a course, was timing for the IL where you didn't select a cap at all done from fade in or control? For some levels (like Slip Slidin' Away) your Yoshi time is actually the fastest one so I'm figuring out what the correction should be to those.
If you happen to know the time more accurately than to the second for the ones where you didn't note it in the comment then that would also be helpful. Otherwise we'll just have to assume it was high: if you have a 14s time and we have to add 1.3s for gaining control and 1.4s for the cap (2.7s total) then we wouldn't know if it was 14.1 going to 16.8 or 14.5 going to 17.2, and would just have to assume it was a 17. Kinda sucks but there's no better way to do it :/
My guess is you just used the cap method timing for those stars, despite not actually putting on a cap. I'm gonna calculate what all the times should be right now, it will be a couple of hours until I have everything timed and checked. I guess a good start would be to delete your runs where you used multiple caps as Yoshi and leave only the fastest one. It can then be edited later to the correct character (Yoshi) with the correct time. So for Snowman's Big Head as an example you'd delete the Yoshi, Mario and Luigi runs, as the Wario run is the fastest. Then later I'll tell you the correct time and you can edit it to that time and also to Yoshi, and mods can verify it again.
Alright. And were all of the runs timed incorrectly? Or just the ones using the cap method? If they were all timed incorrectly then we'll also need to take into account the time taken to put on a cap, and make sure we do it for the right ones. Shouldn't be too much of a problem though.
Thanks for the explanation! It's fine that the runs were submitted wrong, they just need to be fixed. A list of all the runs that were timed incorrectly would be very helpful, then we can figure out the amount that each needs to be increased by, as it varies by stage. I'm not sure if I'm understanding you correctly but are you saying you entered a level as Yoshi and selected a cap, Mario's cap for example, and then started the timer when able to move? Selecting a cap like that is allowed, but would mean your run would be under the Yoshi category, for example like this run: http://www.speedrun.com/run/z5oj43jy. For any Mario/Luigi/Wario runs you need to actually enter the stage as that character. So if I'm getting what you said then some of your runs may need moved from Mario/Luigi/Wario over to Yoshi, along with being retimed. This is easily done but would unfortunately require you to redo quite a few runs :/ Selecting the Mario cap is often faster for Yoshi ILs so you'll basically just have lots of Yoshi times instead of Mario times it sounds like, but your Luigi and Wario runs will essentially need redone.
And most of all I'm glad you're not demotivated by this, it will be fun to have you to compete against! I hope you set lots of good times in the future :)
I have been thinking for a while that some of MaxDragonSoul's IL times are too good to be true. Unfortunately his lack of video has made it difficult to get any concrete evidence for this, but I believe I now have enough. However I am not accusing him of cheating: please skip to the final paragraph if you wanna see what I have to say about the evidence and don't care about the evidence itself.
A good example is Shoot into the Wild Blue (Mario) (http://www.speedrun.com/run/y67ldp1m), where his claimed time is 9.82. The strat described is ¤slower¤ that the one done in the 28 star TAS WIP ( at 8:58) but his time is only 0.09 slower than the TAS. Even with the same strat, getting this close to the TAS would be pretty ridiculous but with a slower strat it is frankly impossible. That TAS time was improved by 0.3, but it was using an even faster strat: . The 9.73 strat was perfectly optimised.
Another example is Over the Rainbow Black Box Star (Wario) (http://www.speedrun.com/sm64ds/Over_The_Rainbows_Black_Box_Star#Wario1). Unfortunately for this one Max did not provide the exact time, just that it was 7.xx. Using a ledge grab strategy I was only able to get a low 9s time. Admittedly I was grabbing onto the black box instead of the cloud but I think this should be very close to if not faster than grabbing the cloud as you can then ground pound instantly to get the star. I then managed to perfect a strategy that skips the ledge grab entirely, getting me 8.4: . This is still not 7s, even using a faster strategy than Max.
The final time that confirmed my suspicion that something was wrong was this one: Princess's Secret Slide Sub 21 Star (Mario). Here is my run, 21.9: http://www.speedrun.com/run/zxv204ky. And his run, supposedly 20.9: http://www.speedrun.com/run/m3dgok6z. My run is pretty clean and I doubted strongly that an entire second could have been saved. And it turns out that 20.9 is actually FASTER than the TAS time for this star: in the 28 star WIP the time is 21.1 (). So this time of his is evidently impossible.
I am not accusing MaxDragonSoul of intentionally cheating, and don't believe his runs should be removed. What I believe has happened is a timing error: he has started timing when he gains control rather than on the first visible frame of the level after the black screen. Using this timing method my ledge grab Over the Rainbows run would have been around 7.7s, and my faster strat one a low 7, both tying what he got. This error of around 1.5s, or slightly more or less depending on the course, would explain his other suspicious times too. This is hopefully what has occurred, as then all that needs to be done is the time from the first visible frame to gaining control for each level needs to be calculated, and then added onto each times. I will stress again this time will be different for each course. This theory seems the most likely and it would also explain him beating 3 of the Big Penguin Race times by 1s. If this was malicious he would not have provided explanations of strats used for most of his runs, so I do just think this is a misunderstanding with how runs are timed. Hopefully it can be resolved with a few hours of editing runs, and none need to be removed. Having a direct way to contact him would have alleviated the need for this thread, but sadly all he has is a dead link to a YouTube channel on his profile. Ideally he will see this thread naturally, and worst case scenario a mod simply links it within a rejection message. I hope no action is taken until he makes a response, as although I highly suspect there has been a timing error of some sort, it might not be exactly what I described.
Edit: I also wanna mention that some of his runs are almost certainly timed ¤correctly¤: his Goomboss Battle switch star time (http://www.speedrun.com/run/yl9v462y) was only a few frames slower than what I got today, 13.98 vs 13.8. And I changed up my strat a little to save a few frames at the end with a long jump, so his time seems like exactly the time I'd expect to see after grinding my old strat. So I'm a little more confused with what's going on now :P
MaxDragonSoul, you probably noticed already but Shasta verified all the runs a few hours ago. Runs are usually verified more quickly, I think the main problem was that SGDQ was happening so everyone was busy attending/watching that.
If you are wanting to be a mod then discussing it with the current mods would be better. This is meant to be a last resort thread, while all the current mods can easily be contacted through Twitter, Discord or even Twitch. Making an account for one of those if you don't have one (preferably Discord) would be ideal, and the SM64DS Discord server is linked in the Resources tab of the game's page. Regardless of moderation it would be nice to have a way to communicate with you, I'd like to discuss some of your times at some point :)
I'm pretty sure when ALAKTORN said about splitting the player base he meant meant between cyberscore and speedrun.com, not speedrunners switching to score runs.
The idea of integration between the two sites sounds like the best solution if it could actually work, but the two sites are designed fairly differently so it will be difficult.
Is there any info on this planned rework available? Haven't heard anything about it, sounds pretty cool.
And I was not aware of that side-effect, seems annoying.
Thanks to the front page showing the most recent runs, I have been seeing this a lot recently: runs that are not speedruns but are in fact aiming for highest score, using rules such as "Put your score in the milliseconds slot." speedrun.com does provide variables which can be set up to display a score along with a time (example http://www.speedrun.com/boxboy/Score_Attack_3#Bunny_ ) but it is not intended for the time column to be used for tracking score, as a run taking as long as possible to get the highest score possible is not a speedrun. Is there any intention for speedrun.com to accommodate score runs as well as speedruns in the future, or will the games doing this be dealt with?
Here are some examples of games doing this:
http://www.speedrun.com/flappybird http://www.speedrun.com/Happy_Jump http://www.speedrun.com/Crossy_Road#High_Scores http://www.speedrun.com/Ball_King#High_Score
http://www.speedrun.com/pt (All the Misc categories) http://www.speedrun.com/Piano_Tiles_2 http://www.speedrun.com/White_Tiles_4_Piano_Master (Arcade, Speed and VIP modes and ¤many¤ Misc categories)
http://www.speedrun.com/googlepacman (High score categories) http://www.speedrun.com/Google_Doodle_Soccer_2012 http://www.speedrun.com/chinesenewyear2013 http://www.speedrun.com/Pony_Express_Google_Doodle_2015 http://www.speedrun.com/Google_Doodle_ICC_Champions_Trophy_2017_Begins
These are just the ones I found after maybe 15 minutes and I'm sure there are many more. For at least some of these games a leaderboard for score runs exists elsewhere, for example https://cyberscore.me.uk/chart/168804 for Flappy Bird. As this is a speedrun site I don't think these games/categories belong on the site, unless one day there is an expansion of scope and score runs are also included.
