I didn't realize you only had just a couple of days to prepare for this DwangoAC, I never checked the date and just assumed it was at least a couple weeks away.
Under the circumstances I think that went very well, nicely done. Sorry Battletoads gave you so much stress thuogh, I guess True has the magic touch for that game.
Aside from that have you followed the recent Atari 2600 Dragster discussion at all? There's some real interest in having console verification of some A2600 runs. I'm not sure it's interesting enough for any kind of event, but it might raise the profile of TASBot a bit in some other circles. Just a thought since it's current.
The content for HRDQ was quite enjoyable. The commentaries were fun and explained the games well.
The commentaries for every GDQ were to much game focused, never explained TAS techniques in detail. The commentary shouldn't be a tutorial, like it has been shortly done for SGDQ2015, which was no good. For a good commentary of that sorts a game which is self-explanatory, but still has hidden details (stuff that RTA runners usually don't attempt to optimize), is needed. ACE/heavy glitching is not allowed. The commenter should be a very good TASer who never [sic] worked on this game before, the only information he should know about the game is from submission texts/game resource pages. The lenght for a run would need to be about 10 minutes, the run should be played back while commenting. The commenter should talk about optimization (level of detail, differences between RTA runs), tools (explain what they do, evolution of tools) and the art aspect of TASing. I only have two runs that be pontential canditates for such commentary, although one is in the works and is unlikely to be finished in time. But I would gladly held to write down the commentary for the presenter.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
The commenter should be a very good TASer who never [sic] worked on this game before, the only information he should know about the game is from submission texts/game resource pages.
I have put together a poll of all of the games suggested.
Please, rate as many as you have time for, but don't feel pressured to answer them all.
https://goo.gl/forms/Mm00XEkECZ3Wr8GE2
Joined: 10/12/2011
Posts: 6337
Location: The land down under.
1) Who are you?
2) No.
3) dwangoAC or link_7777 (or both) are fully suitable on their own to actually make the decisions with what we already provided by themselves.
WebNations/Sabih wrote:
+fsvgm777 never censoring anything.
Disables Comments and Ratings for the YouTube account.Something better for yourself and also others.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Spikestuff wrote:
1) Who are you?
2) No.
3) dwangoAC or link_7777 (or both) are fully suitable on their own to actually make the decisions with what we already provided by themselves.
Uh, calm down Spike - I asked for help making a poll and Deef volunteered to do this. link_7777 and I are fully aware he's made this for us. This is to help us get more datapoints on what people would like to see in preparation for a submission by Friday (or Saturday at the latest). Does that help?
https://goo.gl/forms/Mm00XEkECZ3Wr8GE2
I feel like I need to say SOMETHIG here, so I might as well introduce myself.
Hi. I go by deef. I have been hanging out in the TASbot IRC channel for some time now and have been helping dwangoAC et al. out with twitch stuff and whatever needs being done.
Glad to join the community.
I can understand where Spike is coming from though. It's a new account, getting input about a recent thing that happened, with speech that is moreso informal. It definitely at least seems potential troll. That it is not is nice to know, but I can certainly understand where Spike is coming from.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Habreno wrote:
I can understand where Spike is coming from though. It's a new account, getting input about a recent thing that happened, with speech that is moreso informal. It definitely at least seems potential troll. That it is not is nice to know, but I can certainly understand where Spike is coming from.
Fair enough, I was honestly surprised he didn't ping me on IRC. :) I'm not worried, my tone may have come across as harsh but honestly I was more amused than anything.
A decent number of people have taken the poll, thanks! Keep it coming!
First of all, great job on HRDQ! That was thoroughly enjoyable.
Second, I mentioned this in the survey but I'll also mention it here: I would love to see MrWint's GBC Pokémon: Yellow Version "Arbitrary Code Execution" in 05:48.28 at AGDQ 2018. I think it hits that sweet spot with enough ACE WTF?!? without going too long or overboard.
It would help to explain to the audience that the only inputs to the run are pressing up, down, left, right, A, B, select or start at specific moments. There are no other inputs or data feed than those eight buttons.
Also, to make it more of a surprise, you could run it as if it were a real timed run, starting the clock on power up. Then you can finish by calling 'Time!' when the Yellow credits start to roll.
This isn't a bad suggestion but the challenge is we can't currently console verify it. We'd have to show it in an emulator and for an ACE like that I think it's better to wait until we have the tools we need to try this on real hardware. Eventually we may just give in and do it in an emulator because I do think this one is quite fun.
Thanks for the feedback and suggestion, and for taking the poll!
I have a few questions before filling the form...
1. Is the list related to the AGDQ subs? Like movies picked based on that list?
2. Is the AGDQ already "final" list of all available subs?
3. Should I skip rating my own ones?
4. How does weighting or whatever the analysis reflects this?
My concern for the final one is that I saw a few NES titles that probably I'm part of it (author), but I wouldn't want to watch edit: the other, not known games' whole TASes that I.... don't care?
PhD in TASing 🎓 speedrun enthusiast ❤🚷🔥 white hat hacker ▓ black box tester ░ censorships and rules...
This isn't a bad suggestion but the challenge is we can't currently console verify it. We'd have to show it in an emulator and for an ACE like that I think it's better to wait until we have the tools we need to try this on real hardware. Eventually we may just give in and do it in an emulator because I do think this one is quite fun.
When it comes to GDQs, I really don't think console verification vs. emulator is much of an issue. Most of the viewers are coming to the TASBlock to see something cool and out of the ordinary, to be blown away by things that humans are simply incapable of doing. As long as it's entertaining, it'll draw people to it, and thus draw donations. I have a feeling most people watching online wouldn't know or care about the difference.
So my opinion is that we showcase the most entertaining runs we have, to draw the most viewers, to bring in the most donations, as well as to secure TASBlocks for every GDQ in the future!
I have a few questions before filling the form...
1. Is the list related to the AGDQ subs? Like movies picked based on that list?
2. Is the AGDQ already "final" list of all available subs?
3. Should I skip rating my own ones?
4. How does weighting or whatever the analysis reflects this?
My concern for the final one is that I saw a few NES titles that probably I'm part of it (author), but I wouldn't want to watch edit: the other, not known games' whole TASes that I.... don't care?
1. Yes, we are trying to make use of that list to narrow down what to submit.
2. In theory. The list came from suggestions in this thread, but it isn't binding. We could ignore it and do something else if it was compelling.
3. No.
4. We are looking at the results both weighted and unweighted (to account for people that only rate a limited sub-set), but the results actually aren't much different either way
When it comes to GDQs, I really don't think console verification vs. emulator is much of an issue. Most of the viewers are coming to the TASBlock to see something cool and out of the ordinary, to be blown away by things that humans are simply incapable of doing.
When at that one GDQ they showed the Ikaruga (I think it was that) run as a video, rather than being run in real time in a console, it received a lot of criticism from the viewers.
Granted, perhaps being run in real time in an emulator is ok with the public, but I have the feeling that being run on an actual console is preferred. Being run on an emulator always gives the feeling of it not being completely "legit" in the eyes of the wider public (who may not be intimately acquainted with how TASes work).
I agree that it's better to replay the Pokemon ACE on hardware if it's possible even in the far future. But if the future is too far, it might be good enough to just run it on an emulator.
Warning: When making decisions, I try to collect as much data as possible before actually deciding. I try to abstract away and see the principles behind real world events and people's opinions. I try to generalize them and turn into something clear and reusable. I hate depending on unpredictable and having to make lottery guesses. Any problem can be solved by systems thinking and acting.
Some of my thoughts on the poll results:
The Sorted By Averages look to me as if people didn't even watch what they were rating. Apple II The Oregon Trail is 23rd out of 30, it is one of the least entertaining TASes to watch. People probably only remembered the submission text and based their rating on it. The Oregon Trail has an entertainment rating of 2.2 here on tasvideos, the TASes 24th to 30th have much higher entertainment ratings than The Oregon Trail. It is one of these movies were the "book" is good, but the actual movie is outright trash. If you were to show it just read the submission text to the audience without showing the TAS.
Same with N64 Magical Tetris Challenge, it shows nothing what a human can't do and it's rated 18th.
And some outdated runs are rated very high as well.
The top rated TASes are all from famous games as expected.
Many of the comments don't have actual thought to it, a few show lack of understanding.
But some show concerns about the commentary potentialy lacking proper explanation on TASing:
I feel like TASBlock is becoming a hard sell at *DGQ mainly due to the general audience's lack of understanding with the technicalities involved.
I feel like runs that will be the most immediately gratifying to people who don't have a good tech understanding are runs that completely break the game, with the Mortal Kombat playaround run being an excellent choice. It's got the nostalgia factor and its also completely off the wall. Same deal with the F-Zero X run.[...]
I think Mortal combat is the worst choice to explain TASing, F-Zero X is good, but not the best choice.
would be good to get live commentary from regular runners of these games
is in contrast to
[..]Commentary from the TASers would bump any score up a couple points.
Commentary from the RTA runners are a bad idea, just remember the SM64 this year, other commentary were RTA runners commentated were also just rambling. Letting the TASers from the game might be a good idea, if they don't explain like commentators on RTA runs do.
Feedback from the AGDQ 2017 thread shared the same worries as there were no proper explanation on TASing.
From the Poll Responses:
For rating where there are 10s as ratings and the rest empty indicates that (s)he only rated TASes of games (s)he like without ever looking at others. Some with some of the 1s ratings, people were like "I don't know the game, so I rate 1". One rated only 10s, while others rated nothing, these probably didn't watch any of the TASes. Compare the 10s ratings of The Oregon Trail to the others and count how many times there's a more entertaining run rated with 1.
Otherwise no abuse seen, one person rating only for one run multiple times or instructing people to vote for one run.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
As was noted above, the poll results are in for your perusal. There were over 380 responses and for the most part it's actually high quality data in the sense that people don't appear to have been trying to game the results. Yes, some people only voted for just their favorite game but removing those results, according to deef, didn't substantially change the results due to the large sample size. I am impressed with the number of man hours from this and I will probably do it again next time, although probably with a *much* smaller list of options. Without further ado, here's where discussions are heading:
Based on the results of the poll and discussions in #tasbot we've narrowed down to submitting the following games:
- SNES Kaizo Mario 3, hopefully with console verification (to be tested as soon as possible, potentially as early as tonight)
- GC Super Monkey Ball with console verification via replay files (already confirmed working on two setups)
- Atari 2600 Dragster with recent console verification (it may not be accepted but we've been told we absolutely must submit it)
- SNES DKC2 out of our back pocket, an unpublished run by p4plus2 and assistance from Masterjun (we can consistently crash the game and we think we can finally get an ACE to be consistent on console but aren't there just yet. This one is risky because we're running out of time.)
- Door #5. Please help define what this is, because there are too many good options.
I'm quite fatigued from the last several weeks of work and some very challenging issues in real life that have me beaten down but I will endeavor to keep moving forward. If anyone wants to take a crack at a draft submission text from any of these runs please reply with one. If you have a strong opinion after reviewing the poll results of something you think really ought to be the 5th submission, *please* speak up, we don't have much time left. Thanks!
- SNES Kaizo Mario 3, hopefully with console verification (to be tested as soon as possible, potentially as early as tonight)
Great choice! Lots of fun play around and split-second timing. Truly entertaining!
dwangoAC wrote:
- GC Super Monkey Ball with console verification via replay files (already confirmed working on two setups)
Saw this run on HRDQ and although it's extremely impressive, each level goes by so quickly that there's no time to understand what just happened before the next level is upon you. Unlike Kaizo above, there's no down time to process and appreciate what's going on, so after a while it just feels like you're getting bombarded with input. I think what would make it better is if the full replay of every level could be shown so the audience gets a second look at the tricks, however that would go against the spirit of the speedrun, and may cause desyncs requiring the whole run to be done over...
dwangoAC wrote:
- Atari 2600 Dragster with recent console verification (it may not be accepted but we've been told we absolutely must submit it)
Meh. It's neat, but not something I'd think is very GDQ entertaining (unless there's some big history I don't know about). But if we absolutely must submit it, then I guess we have no choice?
dwangoAC wrote:
- SNES DKC2 out of our back pocket, an unpublished run by p4plus2 and assistance from Masterjun (we can consistently crash the game and we think we can finally get an ACE to be consistent on console but aren't there just yet. This one is risky because we're running out of time.)
Ooh! Sounds intriguing! Good luck!
dwangoAC wrote:
- Door #5. Please help define what this is, because there are too many good options.
I've already voiced my candidate and reasons for Door #5. It'd be a shame to hold back such a great run indefinitely. However, in lieu of that choice, Family Feud is always good for generating laughs. :)
(Were it not so long, I'd also recommend N64 Super Mario 64 (USA) "70 stars, no BLJ". There are some amazing tricks in there that would seriously wow people, much more than what the BLJ does. Maybe just run part of the run?)
Edit: Okay, I just finished watching through the first 30 stars of N64 Super Mario 64 (USA) "70 stars, no BLJ" and was once again floored by all the amazing tricks that are pulled off in that run. In spite of its length, that has got to be a main contender IMO for Door #5.
