Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Weatherton wrote:
I'm willing to provide commentary. It would help to use runs that have detailed submission comments to ensure we can review and give the TAS justice.
I also think it would be fun to do something like the Choco Mountain Gauntlet (but for another game). So, something where a bot sets up a difficult to complete scenario without total control.
You have less than two weeks to identify a game in which to do so. I do like the idea and hope you can find a game.
We will probably have to set up a poll or two to decide what the majority of people think will work. I like a lot of the suggestions but we're going to need to narrow them down a bit first.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
EEssentia wrote:
I always find SM64 to be entertaining and Super Metroid, especially the Reverse Boss Order one, although it may be a bit long.
Still, I find that it just feels wrong to take an off-the-shelf run and play it back at GDQ. What's different from doing it at GDQ than simply watching the run at youtube? Each runner carefully practices their runs before the event and has to pull it off realtime. But a TAS has none of that at least not premade. So it feels wrong to pick an off-the-shelf one. It feels more "right" to make a TAS explicitly made for GDQ.
The point of TASes, I think, is not just completing the game as fast as possible, but to pull off stunts to gives off "awesomeness" to the public, especially under a GDQ setting. Anyway, those are my thoughts.
And those are all very valid thoughts indeed! The main answer is commentary, i.e. taking a run that already exists and showing it off with properly practiced commentary, well polished and tailored for the viewers of a GDQ. Think of it as the value-add. Keep in mind that while many of us here on the site know what the gems are most GDQ viewers probably don't and it's up to us to dig them up and show them how they sparkle in the spotlight.
I mentioned this in IRC but I think the best choice is to get a diverse grouping of different game types, say one racing game, one platformer, one 3D game, etc. I'd love to have an ACE if it makes sense to cap off the block but again I don't want to do that if it isn't highly entertaining in and of itself. I'd really appreciate it if someone can tell me (privately or in this thread) about a run they are working on that would be a good fit for the marathon as it'd be awesome to unveil something new that hasn't been seen before. This is all good stuff!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Spikestuff wrote:
Arc wrote:
SMB3 beaten in 2 seconds is a good example.
Super Mario Bros. 3 beaten in 2 seconds is a TERRIBLE example.
dwangoAC wrote:
The AGDQ 2018 games submission committee has asked us to submit TAS content. They have specifically asked us to focus on highly entertaining (and reasonably short or otherwise varied) content, and they have specifically asked us to avoid focusing on technical shenanigans for this event. Console verification is preferred where possible but not required meaning running TAS's in an emulator is acceptable for this event (although playing back videos is not acceptable). Good, practiced commentary is a must and having the authors of the run present is considered a bonus.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Warp wrote:
dwangoAC wrote:
(and reasonably short or otherwise varied)
...Yet if a TAS goes on for 10 minutes, it somehow becomes boring.
The chaotic nature of the human mind will never cease to amaze me.
This is because they want to do several different TAS's, i.e. we can do an hour long TAS or we can do a handful of ~15 minute TAS's but probably not both. If we want more time we'd probably be scheduled at an inconvenient time is my guess. There's always RPG Limit Break though...
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
The AGDQ 2018 games submission committee has asked us to submit TAS content. They have specifically asked us to focus on highly entertaining (and reasonably short or otherwise varied) content, and they have specifically asked us to avoid focusing on technical shenanigans for this event. Console verification is preferred where possible but not required meaning running TAS's in an emulator is acceptable for this event (although playing back videos is not acceptable). Good, practiced commentary is a must and having the authors of the run present is considered a bonus.
As I've noted in a few places I cannot guarantee it will be possible for me to attend the event. Likewise, I have completely shifted my leadership role away from being "the guy in charge" to being more of a facilitator or guide, simply asking volunteers how things are going and helping them stay unblocked. This change is needed for a wide variety of reasons, ranging from how I feel I personally mishandled certain aspects of past events to the current amount of free time I am likely to have for the foreseeable future due to real life events. At this time I hope to be able to attend the event but I've asked Jaku to stand in for me if I cannot be there in person.
With all that said, we have a limited amount of time to identify content that meets the game selection committee's requests as submissions start September 2nd and close September 13th. Can you, yes you dear reader, think of a run that would be a good fit based on past content already shown at GDQ events and based on length and entertainment? Again, I'm expecting we will not be doing console verification but we will attempt to do so wherever possible; I realize this may be an unpopular viewpoint and there will be some that will be very upset by this but it's the only reasonable way to allow us to show a wider variety of content than we have been able to show so far and the benefits outweigh the downsides in my opinion.
So, what should we pitch? Reply here, come to #tasbot / http://chat.tasbot.net or something else but don't assume someone else will step up. Thanks to everyone in the past who has supported TAS at GDQ efforts!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Kurabupengin wrote:
If we're really gonna use TASBot as the site's mascot, then at the very least we should design him in a way that Nintendo's ninjas don't come after us, since ROB has been recently used as a Nintendo character in recent games like Smash Bros.
Or at least that's what I think anyway.
We put a lot of thought into determining whether the artwork was acceptable back in 2016 when Ange made the image for The Yetee to use for SGDQ 2016. It was acceptable to The Yetee and was well-received publicly with no backlash from Nintendo. One part of it is that this was clearly an artist's imagining of a physical object (as opposed to, say, the exact pixel art ripped from a ROM file) and further the image contains no trademarked words or other legally concerning elements. Here's a full version that more clearly shows the uniqueness of the image and TASBot in context (click to embiggen):
Ange has all the original files and this image can be cleaned up with a different background color, no text identifiers or lower-right box area, and turned into a single piece of art that would be usable as a mascot. Note that I haven't recently spoken to Ange about doing additional work on this but I think this would be a small amount of time for him and he'd probably be willing to do it as it would simply be pruning away elements that are not needed for a mascot image. Thoughts?
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
So there are a bunch of interesting things that Ange Albertini made for the site that as far as I'm aware haven't yet been put to use which I think is a shame. They are possibly poorly scaled due to being in .svg format (update, changed to 150x150) so it is probably best to view them directly by clicking on them.
Capcom style TASVideos logo:
Megaman style TASVideos logo:
Super Metroid style TASVideos logo:
SMB1 style TASVideos logo:
Megaman slide:
Now, I'm definitely biased, but I happen to like the idea of TASBot as a mascot - Ange made several SVG versions including the one at the top of http://TASBot.net but here's just his head in .png format:
I also like this little 48x48 guy (he's almost a bit too small, but he's cute and was made by endrift):
Again, I'm quite biased so feel free to ignore me but i hope this art is of use to someone. Enjoy!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
I may do some tests with a couple of games that have historically been poorly emulated and report back. I'm not sure when this will happen as three are a lot of new shiny things to chase (C64 support in Bizhawk being one of them, console replay advancements on several fronts being others, etc.) but I'm definitely intrigued by this emulator.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Today I stumbled on Nintaco (http://nintaco.com) which is a new cross-platform open source cycle-accurate NES and FDS emulator and I was really floored with how many features it has. The main interface looks fairly standard:
It has full TAS features, as well as a TAS Studio style input method referred to as the History Editor:
The FAQ indicates that the emulator seems to cover... literally every peripheral and configuration I've ever heard about the NES/FDS supporting. There's even an API and examples of how to use it to, say, display messages on screen using various languages like Python and C#.
There are several other things of interest such as the fact that it uses the LGPL 2.1 license, the fact that it is written in Java, and the fact that it adds a virtual R.O.B. window for the two games that use it. One of those three things may be a problem for some but I'm quite surprised at how much the author has accomplished.
This leads to an obvious question - if it proves itself as a reliable emulator that's in many ways more feature rich than even FCEUX, what's the possibility of allowing the emulator and its movie files to be accepted on the site for submissions?I have to say, before I saw this I wasn't thinking to myself "What the world needs is another NES emulator" but now that I see what it purports to have support for I'm rather intrigued. What are your thoughts? Discuss. :)
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
This was a lot of fun - I was quite surprised by the ending. I would love to be able to console verify this and I have every belief that if I had all the right hardware that it would succeed. The only reasonable case to reject this run is that it does not use pure controller input but I think this limitation is unwise in this case as the expansion port is a valid port exposed to the end user. Therefore, this gets a Yes vote from me.
I've made my case repeatedly and I will continue to state that we need to stop being afraid of having multiple branches and we need to accept this *without* obsoleting any previous run. For this scenario, because this is in a fun gray area between just a game-end glitch and an ACE and it includes unique content that makes it stand out as a work of art on its own merit I believe it should be submitted as its own branch, separate from existing runs. That is all.
Great work, Masterjun!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Honest question - did this not qualify for vault for some reason? (I admit I couldn't stomach making it through the first rap, which is still rattling around disturbingly in my head in a way that concerns me that it is growing on me, but I digress.) I mean, sure, this is clearly an awful game, but was there a technicality I missed? Just curious.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
I had this game for the Xbox and seeing it destroyed like this is awesome. Sure, compared to game has fewer tricks but I love the atmosphere. Yes, I have a soft spot for the music. Please don't judge me. :)
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Invariel wrote:
I feel that TASVideos should withdraw from *GDQ events for a couple of years, to give people the opportunity to seek us out on their own, and if and when we return to *GDQ events, we should focus on accurately explaining what TASes are, as well as showcasing runs from the previous year instead of engaging in more esoteric pursuits.
Thank you for your feedback. The earliest I will return to organizing a GDQ event is SGDQ 2018. The content that link_7777 has submitted for SGDQ 2017 is more suitable for your goals and if the TAS explanation portion is accepted it should help fill in several gaps that have been widening over the last few GDQ events. I also feel that the recent Californithon runs (viewable here) were much more in-line with what the site needed for proper exposure of what kind of content to expect. I agree that GDQ needs a fairly long break from the types of work we were doing with large teams killing themselves to create over-the-top content; pretty much everyone involved was burned out and it was time for a break. I'm glad to have a bit of time to breathe, I have to say that much. :) Thanks again for your thoughts!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Alyosha wrote:
SM64 commentary though was not so good in my opinion.
So about that... Long and short of it, the first resource I identified for SM64 commentary wasn't as knowledgeable as I thought and we agreed to tap someone else but that person didn't show up in the (admittedly cramped and packed) stream room. In between MK64 and SM64 the audience identified someone who was knowledgeable about speedrunning SM64. Unfortunately, he was seriously nervous on mic and choked, something I didn't expect at all. I rolled with it but I'd agree that the commentary wasn't all that great for SM64.
Anyway, it was a good experience for our family. It was a fair bit of driving but it was nice being able to come back home to our own beds. The kids had fun at the park we stopped off at and my wife enjoyed our trip to Ikea on the way home. Compared to the stress of AGDQ 2017, this was a positively pleasant experience - it was perhaps 30 hours of work ahead of time to prepare and it only cost us about $100 total after fuel, bridge tolls, parking fees at the event, and parts.
I'm continuing to ponder what I can do to get better at commentary. A lot of it comes down to practice, and it strikes me as somewhat like trying to prepare for a piano recital. You can practice on a piano at home (in the same way I can practice commentary while streaming on Twitch) but it's not quite the same thing as practicing in a high-energy room in front of a live audience. Comparing my own confidence at Californithon as a presenter to the early GDQ events I can tell I *am* making progress but I can always be better. I'll continue to practice however I can while balancing time with my family and hopefully I'll be ready to go by SGDQ 2018.
Thanks again to everyone for the support, especially Alyosha and weatherton ahead of time and Jaku and MeGotsThis at the event, as well as Jeff_In_3D for inviting us to submit in the first place.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
I feel good about most things but I'm sure commentary will be a lot more free-flowing than scripted; scripting it just didn't feel right. I'll be directing people to the submissions instead of trying to highlight every last nuance of whether SM64 is a parallel universe or not. Here's to a good run, I've made a post on the front page so I'm hoping we have some representation there. Catch you guys on the flipside!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
With the help of Alyosha we now have all four Mario Kart 64 files and I'm testing the last two in just a few minutes at http://twitch.tv/dwangoAC (on my laptop that I'll be using during the marathon tomorrow). I've read through Weatherton's notes but I need additional review.
I'm the most worried about what to say during SM64; there were concerns in the AGDQ 2017 thread that the commentary was not adequate or accurate and if someone who actually plays and speedruns the game can't provide proper commentary I have no idea how I'm ever going to measure up to their expectations. I'll do the best I can with the resources I've been able to commit to this endeavor, i.e. the amount of time I've been able to take away from my family to work on this event. I think everything is going to be just fine because this is not an event with the same expectations as a full GDQ but I still want to do the best I can.
I'll be focusing on just run commentary today but I plan to incorporate a lot of the "intro to what a TAS is" stuff during the actual presentation. We'll see how things flow in today's practice and I'll take it from there. Thanks again for all of the support and here's to a great showing tomorrow!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
New problem: The connector ends showed up yesterday right on time but the extension cables are inexplicably delayed until next week, so we're back to getting a 1-controller version working. Having said that, #1533: Weatherton's N64 Mario Kart 64 "Any%" in 06:43.23 (yes, that one all the way back from 2007, nearly a decade ago) synced! I'm streaming right after I write this post (http://twitch.tv/dwangoAC) to see if I can do something like hex edit the start of a working movie in or use a version that Weatherton just sent me; we'll see how far I get tonight but there's renewed hope in this one.
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Solution (maybe) - I was able to order four N64 extension cables and some 2-pin female headers with fast shipping and they should arrive on Tuesday. I have to be careful to not throw too much money at this (one of the major reasons for having to take a break is donations to cover our costs substantially dried up for AGDQ 2017) but if this works it'll be worth it. It's still very unclear if a resync of the existing run would even be possible and it's better to just do this the right way if at all possible.
I remain a bit worried about commentary - if you read the earlier posts in this thread you'll see that one of my main goals was to correct the bad commentary from AGDQ 2017 (for my part, I was completely annoying with my commentary during Portal). My other goal was to ensure that working on this event would not take time away from my wife and kids. If you've been watching my recent Twitch streams you may have noticed a trend, namely I've been playing games with my family rather than working on TASBot stuff.
All that to say I think I've prioritized the right things but I have to make sure I really think through what I'm going to say and how I'm going to say it this week. I'm the most confident about being able to describe the basics of making a TAS and I'm the least confident about exactly how some of the more technical glitches in SM64 are performed. Perhaps, just perhaps, the Californithon audience doesn't actually need a high level of detail and a simplistic "At this point Mario is going through a door he shouldn't be able to using a TAS-only technique" and make sure to point people at the submission notes for further details. Just know that I'm pondering this a bit right now.
More updates probably on Tuesday but feel free to post any ideas or feedback. Thanks!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Problem - I only have one N64 controller cable but Weatherton's run requires four of them. Alyosha was a huge help in creating a script to convert from BizHawk format to .m64 format which allowed us to at least test getting something sane to play back but the problem is more than one controller is used in the menuing system to make things go faster. I haven't yet investigated the rest of the run to see if additional controllers are used during races and I have no idea if a resync with one controller is possible. I'm considering MK64 a risk at this time.
If I don't (or can't) do MK64, there are a number of interesting things I could show at Californithon. I'm curious if Lost Levels would work and there will be an FDS there but I'll have virtually no time to confirm that. We could possibly show one of the longer SMW runs that haven't been demonstrated yet. There are plenty of "easy" things, including showing the audio playback from GeekPwn. The only thing I'm (as always) a bit concerned about is commentary so I'll be working on that while streaming in the evenings this week. More updates soon!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Quick update - after confirming the 1-key SM64 run I now need to work out MK64. I've spoken to Weatherton about how best to handle the run because the cups are all separate from one another and the submitted run does a reset. We'll need to segment each cup into its own movie file which is probably not too bad. The harder challenge is we'll need to convert the BizHawk file into a format that does not include lag frames. The Mupen format was nice for console verification because it didn't matter too much how many lag frames there really were on the console so long as the game always polled for input when it needed it the same way and we need to get either into that format or get into some similar format that we can adapt the script to read. Alyosha or rcombs might be willing to tackle this which would be really helpful, otherwise I'll scratch my head a bit over the weekend.
My plan is to stream at http://twitch.tv/dwangoAC on Saturday morningish on my attempts. If anyone wants to lend a hand, please feel free to stop by #tasbot in IRC / http://chat.TASBot.net to discuss. Thanks!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
link_7777 wrote:
Doing a sort of TAS review of RTA tricks seen in the marathon is also something I want to make sure gets submitted, but I need to sync up with dwangoAC on that topic.
I don't think this is something that's going to be submitted but I need to hear back from staff on that. It certainly wouldn't be an actual run in the marathon as it'd be filler in-between runs. Seeing as registration filled up *really* fast I should work this out as soon as possible. I'll ping my contacts again and see if something breaks loose.
My take: Don't bother with console verification for this marathon and instead focus on providing game-length commentary. Do try to do something that can be played back in an emulator rather than showing a video, though - we got a lot of flack for that...
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
As I noted in the update I made to the first post, I do not currently plan to attend AGDQ 2018 and it is possible I will not attend SGDQ 2017 either in order to spend more time with my family and work on self-improvement. I will not lead or otherwise attempt to organize a TAS / TASBot block for either event, but I will fully support anyone who chooses to do so. I will gladly provide TASBot as needed and I can assist with PR during the event, but I cannot commit to the grueling process of organizing another event on the scale that past events have been.
The only thing I'm considering doing just for SGDQ 2017 is providing Madden-style analysis between runs to praise a runner who has just performed a difficult trick by showing how the trick was executed using an emulator and TAS tools. The idea is similar to using slow motion to replay a sporting event while describing what took place, but in this case with the benefit of the entire arsenal of TAS tools in an emulator. This is not something I am committed to doing and I'd like to try it out on my stream (http://twitch.tv/dwangoAC) but it seems like an interesting idea.
If anyone wants to step up for SGDQ 2017, now is the time to do so; submissions are only open from March 6th to March 11th and there is not much time left. If someone is interested in taking this on please post about it here. Thanks!
Experienced Forum User, Moderator, Published Author, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Some status updates - first, I'm happy to report that with a minimum of work (mostly involving sorting out a bad flash) I was able to get micro500's N64Bot up and running and I was able to use a US resynced file by rcombs to console-verify [3264] N64 Super Mario 64 "1 key" by Tyler Kehne, mkdasher, sonicpacker, snark, SilentSlayers, ToT, Plush, Gaehne D, Eru & sm64expert in 04:21.30 - here's a highlight video. The full VOD from last night also includes a verification of the 120 star run.
I've made some major changes to how I'm doing things - for one thing, I'm in my livingroom, with my kids, doing something a lot less stressful than trying to organize a huge event in secret. It was fun, even for my kids. After the kids were in bed my wife and I enjoyed a relaxed session of Assassin's Creed II (and a bit of knitting) with the remaining viewers, and the worst thing I can say is we stayed up too late. This has been the least stressed (as in, I'm not stressed at all) that I've ever been prior to a marathon, for which I'm glad.
I would like to get my hands on a copy of Mario Kart 64 if I can. I think it would be fun to show Weatherton's all cups run but I need a cartridge to borrow to test with. After I nail down which runs to show I'd like to get some commentary help. I know boct1584 has made an attempt at creating a brief explanation on what a TAS is and what the process is like which I may refine and adapt to be in my voice and modify for use in a live setting. For the 1-key run I need to fully understand the correct terminology or otherwise draft someone to help with commentary and practice it several times, probably publicly in realtime while streaming to make the time pressure a thing. The runs are never long enough on stage to say all that needs to be said.
Finally, there are possibly a few minutes left in the schedule for "one more thing"; I'd prefer to stick to N64 content since that's the replay device I'm bringing, any thoughts on what to show? Thanks for the support, all!
Ange has all the original files and this image can be cleaned up with a different background color, no text identifiers or lower-right box area, and turned into a single piece of art that would be usable as a mascot. Note that I haven't recently spoken to Ange about doing additional work on this but I think this would be a small amount of time for him and he'd probably be willing to do it as it would simply be pruning away elements that are not needed for a mascot image. Thoughts?
Megaman style TASVideos logo:
Super Metroid style TASVideos logo:
SMB1 style TASVideos logo:
Megaman slide:
Now, I'm definitely biased, but I happen to like the idea of TASBot as a mascot - Ange made several SVG versions including the one at the top of http://TASBot.net but here's just his head in .png format:
I also like this little 48x48 guy (he's almost a bit too small, but he's cute and was made by endrift):
Again, I'm quite biased so feel free to ignore me but i hope this art is of use to someone. Enjoy!
It has full TAS features, as well as a TAS Studio style input method referred to as the History Editor:
The FAQ indicates that the emulator seems to cover... literally every peripheral and configuration I've ever heard about the NES/FDS supporting. There's even an API and examples of how to use it to, say, display messages on screen using various languages like Python and C#.
There are several other things of interest such as the fact that it uses the LGPL 2.1 license, the fact that it is written in Java, and the fact that it adds a virtual R.O.B. window for the two games that use it. One of those three things may be a problem for some but I'm quite surprised at how much the author has accomplished.
This leads to an obvious question - if it proves itself as a reliable emulator that's in many ways more feature rich than even FCEUX, what's the possibility of allowing the emulator and its movie files to be accepted on the site for submissions?I have to say, before I saw this I wasn't thinking to myself "What the world needs is another NES emulator" but now that I see what it purports to have support for I'm rather intrigued. What are your thoughts? Discuss. :)