In a full-game run, whenever someone clicks past the IGT before it can pop up (for example, they would have gotten a 15.62 on 1-2, but clicked to 1-3 before the 15.62 appears on screen), how should IGT be decided?
One way to solve this is to simply use the time in the top right of the screen at the end of the level (for example, if they had gotten a 15.62, the timer in the top right would have displayed 16), which, while unfortunate, seems like the fairest way to solve the problem, and it's what I've done whenever I've clicked past the IGT.
Another way might be to do a frame-count, but IGT might not cooperate with RTA frames.
Now, just to clarify, the options I picked are to get rid of the input delay completely, but they still cause some visual artifacts and frame drops.
The options that you originally used (in 0.2.0.1) easily look the best visually, as there's no visual problems. So, it depends on whether you want to remove a 4-frame input delay or have super smooth-looking gameplay.
The delay is definitely still there, and I've found an unsatisfying way to fix it.
To fix it, you have to go into options.ini and change AlternateSyncMethod from 1 to 0. This fixes the input delay, but it's unsatisfying because it adds back all of the lag problems from 0.0.1.1
Conversely, you can fix the lag and ADD the input delay to 0.0.1.1 by changing AlternateSyncMethod from 0 to 1.
So, that variable is the root problem, and, unless you have a different way to solve it, it seems like it comes down to the lesser of two evils: 1) Input delay, or 2) screen lag and lots of dropped frames.
EDIT: Enabling CreateTexturesOnDemand does reduce the lag that disabling AlternateSyncMethod creates, so it seems like the optimal settings are: CreateTexturesOnDemand=1 AlternateSyncMethod=0
No, I've never had that delay before the update, and if I switch back to the old version, there's no delay anymore. I'll try to get a video of it soon.
On an unrelated note, though, the update DID fix the excessive lag problems that I had with the old version. So, that's good.
Oh, thank goodness it's not a feature. I guess it's just my computer, then.
For me, whenever I press the spacebar to jump, there's a 1 or 2 frame delay between hitting the spacebar and when I actually jump, which may not sound like a whole lot, but it's extremely noticeable and throws me off quite a bit when speedrunning.
Okay, after playing the new version, I think I've changed my mind; there should probably be a variable added.
Also, dear god, the new version is absolutely horrible with the jump lag.
I disagree; I feel like this shouldn't be added. All it really does is make the leaderboards look messier.
However, if it IS added, please ONLY make it a variable. Don't rank runs differently based on the version; simply display the version number next to the run, and that's it.
If I may make a suggestion, I would like to request that the "RTA" subcategory for the Individual Levels be removed, as nobody is running them RTA and the in-game timer is perfectly fine for those (also, having subcategories seems to mess up the world record history in the statistics page, which makes me sad).
I don't know if I can edit the categories as just a verifier, and I also figure that I probably shouldn't just go around removing things, so I'm asking here to be safe.
I'll upload the video tonight if I remember. I only finished up until the Arctic, because I must say, I truly hate using the Wiimote after being used to using a mouse.
Okay, I've attempted a run on the Wii, and have come to the conclusion that using an emulator is definitely an unfair advantage. I played up to the end of the Arctic, and I was already over 2 minutes behind by PB, and while a decent chunk of that can be chalked up to me being bad with the Wiimote and making stupid mis-menus, there's absolutely no way that that lost me 2 minutes; at most, maybe 30-50 seconds was from that. I can only assume that the rest is from load times.
I did record the run with my laptop's camera, so I can upload it if you need it, but I don't see a way that it'd be helpful.
I think I'll do a run on an actual Wii tomorrow, to see how it compares. I don't have a way to easily record it, though, so hopefully you'll trust my word. It'll almost certainly be slightly worse because I'm not used to a Wiimote, but if there's a significant difference (say, if I get a 2:05 or higher), then that basically answers the question.
I only really used an emulator so that I could record it well without a capture card and easily record Livesplit alongside it.
Actually, I'm going to give an opinion, although you may want to take it with a small grain of salt, since I'm the one in question that's using the emulator.
Based on what I can see, the loading times on the emulator ARE slightly faster than on the Wii, but nothing super significant. And, if you want to be super safe, it's possible to add a "time without loads" bar on the leaderboards, although this makes timing runs much more of a hassle.
I do agree that using a mouse instead of a Wiimote allows for more precise movement, but in many cases precise movement isn't too necessary; You just have to get your character going in the right direction and then sit on it for a while. Granted, though, that is not always the case, which is why I think this is the biggest issue on the emulator discussion.
However, I don't think that it necessarily makes menuing easier. You can use the Wiimote's D-pad to navigate menu options, which I actually do very often in my runs, particularly when cycling between the pulsar, multisensor, and "return to boat". Granted, I don't use the d-pad on the vertical menus, like the Island's shortcut menu, but that's primarily because those buttons have different uses; The "up" button is never used in menuing, and the "down" button is used to open the menu when swimming, so I have that mapped to a different key that's less helpful in menuing.
The only other difference I can think of is that I can speed-click faster than I can spam the "A" button on the Wiimote; however, because this game has a set time limit for each dialogue box, that doesn't cause very much of a difference.
I suppose one could just watch both runs side by side to see where time is gained and whether it's attributable to the emulator or not.
Because every good game has a general discussion thread.
On a side note, I'll be uploading a 2:01:39 run tonight and submitting it tomorrow.
Becase this game is so short and there's not much to be optimized, accurate timing is absolutely necessary on the leaderboards. However, many of the times currently on the leaderboards are not retimed at all, and go by manual splitting, which is inaccurate.
FIRSTLY: HOW TO RETIME RUNS USING WINDOWS MOVIE MAKER
- Obtain Windows Movie Maker (it's a free download)
- Obtain the video of the run. If you can't download the video, you can just play the video while recording the webpage
- Place the video into Movie Maker
- Trim the run such that the first frame of the video is when the exclamation mark disappears from the door (because that's the first thing that happens upon interacting with the door)
- Trim the run such that the final frame of the video is the first frame that the "Day Completed" box begins to appear on-screen
- Find the length of the new video. THIS IS NOT THE RUN LENGTH. Convert the time into seconds, and multiply by 30 to get the approximate frame count.
- Because the frame count is a whole number in reality, round the approximate frame count to the nearest whole number to get the actual frame count. 8 ) Divide the actual frame count by 30 to get actual seconds, and then convert back to minutes:seconds
- Now you have the actual time of the run. NOTE: Windos Movie Maker is limited to 30 frames per second, which is not necessarily going to be accurate. I don't know if this game runs at 60 frames per second, but if it does, then you can use a video editing software that can handle 60fps and do everything listed above, except multiplying and dividing by 60 instead of 30.
SECONDLY: I'VE RETIMED THE NUGGET% RUNS Grayfoxgaming's - 3:47.366 (6821 frames) razziyi's - 3:47.700 (6831 frames) SubmarineCaptain's - 4:02.066 (7262 frames)
Are emulator runs acceptable? Specifically the Dolphin emulator, which is the only Wii emulator I could find.
Part of the problem with IGT for the years-long categories is that speedrun.com only lets you input times up to 999 hours, which makes it impossible for some IGT runs to even be completed. I'm pretty sure that even in Sandbox mode the only other planets that have a chance to be reached and landed on in under 999 hours would probably be Duna or Eve (so then also Gilly and Ike), which is why I didn't include IGT categories in the ILs.
You must show whether your game is using Earth Time or Kerbin Time. You can find this in the General Settings of your game.
1 Kerbin Year = 426 Kerbin Days 1 Kerbin Day = 6 Hours 1 Kerbin Year = 2556 Hours
1 Earth Year = 365 Earth Days 1 Earth Day = 24 Hours 1 Earth Year = 8760 Hours
EDIT: There is a limit to how long a speedrun can be on speedrun.com, which is 999 hours. That means if your speedrun is longer than 41 Earth Days (166 Kerbin Days), you won't be able to submit it unless I find some workaround