I'm not purposely trying to be a downer here, but I'm pretty skeptical that just playing several normal TASes back to back will hold many people's attention.
Isn't that what the live commentary is for? The commentary is what makes the run interesting to the viewers.
One avenue of ACE that is untapped so far is sending information back to TASBot via controller pings.
I think that's precisely what "they have specifically asked us to avoid focusing on technical shenanigans for this event" is referring to. In other words, they want just speedruns, not technical shenanigans like that.
I'm not purposely trying to be a downer here, but I'm pretty skeptical that just playing several normal TASes back to back will hold many people's attention.
As Warp already said, these runs wont be uncommentated. Infact the right commentary should make the key difference between holding peoples attention or not. Remember SMW Maker at AGDQ2016? That had almost no commentary in the long run and felt dragged. And most of the GDQ viewers have a very distorted picture of TASing in general. Which is mostly the fault of showing to much technical shenanigans, which then may cause potential future TASers away from the site, because they think they can't hold up to our standards, they think if you can't use debuggers and lua before making your first TAS project you're impractical as a TASer. But in fact the site has a low standard for the technical part.
A movie tag for TASes shown at GDQ's is something that is not in my favour, it doesn't show any relevant information about the TAS. The only information this tag has is were it was first shown, which is irrelevant.
Isn't that what the live commentary is for? The commentary is what makes the run interesting to the viewers.
TASeditor wrote:
As Warp already said, these runs wont be uncommentated. Infact the right commentary should make the key difference between holding peoples attention or not.
My point was meant to be that even with live commentary it won't be very interesting without something interactive, sorry this wasn't clear. Right now we're talking about 3-4 people sitting on the couch for ~30 minutes playing pre-recorded inputs off someone's laptop. There won't be a 'runner', no one will even have a controller, and TASBot doesn't even need to be there. Sure this will be more representative of typical TASing, but to me it doesn't sound very fun.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Warp wrote:
Alyosha wrote:
One avenue of ACE that is untapped so far is sending information back to TASBot via controller pings.
I think that's precisely what "they have specifically asked us to avoid focusing on technical shenanigans for this event" is referring to. In other words, they want just speedruns, not technical shenanigans like that.
Just to refine this a bit more, they are asking us to not focus on technical shenanigans but they are not telling us we have limits on what we can do in the name of entertainment and polish. Thinking back to AGDQ 2017, virtually every single run was there because it was technically hard. Making console verification work on the (very technically challenging) NES Classic? Technical in nature. High datarate in SMB3, MM1, and especially LoZ:LttP requiring new hardware and supporting software built just for the event? We were insane. Assembly language level optimization to enable not one, not two, but three consoles to work in concert to pull off the effect? Technical beyond anything that was sane to try, which is exactly what motivated us to try.
The thing is, they don't want us to focus on doing things that are hard, they want us to focus on doing things that are entertaining and well polished. If we take advantage of the fact that we already have the ability to control things like the latch pin we may be able to pull off this suggestion without a large amount of development work compared to previous events. Even if we were to do something like this it would be the only part of the TAS block like it, i.e. AGDQ 2017 was a tech fest from start to finish, whereas this would be a single piece and the only thing like it in the block.
Alyosha wrote:
My point was meant to be that even with live commentary it won't be very interesting without something interactive, sorry this wasn't clear. Right now we're talking about 3-4 people sitting on the couch for ~30 minutes playing pre-recorded inputs off someone's laptop. There won't be a 'runner', no one will even have a controller, and TASBot doesn't even need to be there. Sure this will be more representative of typical TASing, but to me it doesn't sound very fun.
This description would be awful. I'd hate to watch a block like that, let alone be part of making it. I touched on this a few posts ago but I'm convinced that what we need to focus on is variety, entertainment, and polish. We need one run from each category, without having too much of one thing. We're allowed, as a concession, to show a run or two that can only be played back on an emulator but I would definitely not want to see an entire TAS block full of that type of content. I'd like to see a 3D or disc-based console run or two, definitely, but I'd also like a couple of console verified runs in there as well with one of them probably doing something glitchy for the people that like to see that.
Our goal over the next two weeks is to identify a variety of entertaining content. The goal over the next two months after that is to add polish to said content, practicing it regularly. Right now we're doing the right thing by refining what we want to put in the block. I mentioned this a few times but I don't plan to be "that guy in charge" this year - I'll keep guiding this conversation in a direction I think is right but I'm not going to be authoritative and force anyone to do something specific. Feel free to disagree with me and keep the conversation going - thanks all!
There's been some progress with this idea recently.
Post #451802 and most of that thread.
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.
There's been some progress with this idea recently.
Post #451802 and most of that thread.
For a live event, maybe a camera pointed at the screen and some basic image recognition would work for a bot to wait for loading times (assuming a permissive enough RNG)?
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Demon Lord wrote:
feos wrote:
dwangoAC wrote:
disc-based console run
There's been some progress with this idea recently.
Post #451802 and most of that thread.
For a live event, maybe a camera pointed at the screen and some basic image recognition would work for a bot to wait for loading times (assuming a permissive enough RNG)?
What you are describing has been approached in a different way by using a vsync detect chip (with mixed success) but that extra information doesn't resolve the fundamental issue with console verification of disc-based consoles, namely that variations in disc loading times impacts RNG and results in non-deterministic behavior. Console verification requires absolute reproducibility. Good thought though!
There's been some progress with this idea recently.
Post #451802 and most of that thread.
For a live event, maybe a camera pointed at the screen and some basic image recognition would work for a bot to wait for loading times (assuming a permissive enough RNG)?
What you are describing has been approached in a different way by using a vsync detect chip (with mixed success) but that extra information doesn't resolve the fundamental issue with console verification of disc-based consoles, namely that variations in disc loading times impacts RNG and results in non-deterministic behavior. Console verification requires absolute reproducibility. Good thought though!
Just as a bit of fantasy, I'm imagining an idea of rewiring a disc-based console to a device which could simulate it's CD drive allowing to provide loading at the same speed as the emulator does while using some flash card as the actual media. Although it depends on how the game actually accesses the disc. If it does that on purely software level, without trying to do driver-specific tasks like navigating the disk or data buffering, an emulation could be possible. A semi-real console for a semi-alive player :) Or a real console with pirating features...
Totally irrelevant link
My consoles: PS2 (PS1 as bw\c), NDS Lite (GBA as bw/c), 3DS XL, Wii U (vWii theoretically), PS4 slim.
I figure no one actually read the thread I linked?
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.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
feos wrote:
I figure no one actually read the thread I linked?
I did but the last activity was in April and the project seems stalled. With AGDQ 2018 submissions opening this Saturday there is no hope of having it ready enough to be something we should commit to. It'd be great for a future event, though!
Discussions in #tasbot on Freenode IRC have gone well but there are more things to iron out so feel free to swing by (http://chat.TASBot.net). Thanks!
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
Edit: For anyone coming from Twitter, the rest of this thread may have AGDQ 2018 spoilers but read below anyway. There's not much time for this, but GDQ may do a marathon this weekend: https://twitter.com/GamesDoneQuick/status/902773537802514432
In brief, I've offered to do TAS stuff for a Games Done Quick online marathon for Hurricane Harvey relief that is slated to run this weekend. We only have a few days to pick content. What do we have *right now* that would be a good showing? I have access to literally the entire NES library aside from obscure games thanks to BigJohn82 (a local collector) and I can spend the next couple of nights doing sync tests so we can stick with console verification for now. Should we instead re-show something that went badly in a previous GDQ and provide it with either better commentary or some other context?
There's not much time so respond soon, every suggestion and offer of help is useful. We can split this post off in its own thread as needed but for now I think it makes the most sense to leave it here. Thanks for all the support!
Just as a bit of fantasy, I'm imagining an idea of rewiring a disc-based console to a device which could simulate it's CD drive
I'm not sure that physically modding the console goes with the spirit of TASing.
If we go that route, what else could we mod, that helps with TASes?
It's not about modding the console itself for easier TASing but rather modding its data input hardware. I don't really see much fundamental difference between the device used to acquire data from the player and the device to acquire data from an optical media. Both cases are about handling an unstable source of information which provides data at an undeterministic rate and the console has to get along with it. Sorry if I'm off topic.
Totally irrelevant link
My consoles: PS2 (PS1 as bw\c), NDS Lite (GBA as bw/c), 3DS XL, Wii U (vWii theoretically), PS4 slim.
It's not about modding the console itself for easier TASing but rather modding its data input hardware. I don't really see much fundamental difference between the device used to acquire data from the player and the device to acquire data from an optical media. Both cases are about handling an unstable source of information which provides data at an undeterministic rate and the console has to get along with it. Sorry if I'm off topic.
The difference is that one requires physically modding the console, the other doesn't.
I'd say it's a justified modification (for the science) if it can't really be restrained otherwise. I still hope it can be.
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.
[21:15:16] <dwangoAC> Someone want to update the AGDQ thread to note that we're on the schedule for Battletoads and SMBall2 tomorrow at some odd time?
Just transmitting a message from IRC.
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.
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.
Moderator, Senior Ambassador, Experienced player
(898)
Joined: 9/14/2008
Posts: 1007
I'm happy to report that everything went quite well during Harvey Relief Done Quick / #HRDQ2017 despite being taken off guard by the event being ahead of schedule and not realizing I was on an open mic (twice) during setup times. There are some things that weren't perfect due to the rushed nature of the event - the schedule was running so far ahead I ran out of time to post an announcement on Twitter/IRC/Forums and I'm still not happy that attribution was yet again messed up (none of the author's names of the TAS's we showed were displayed on the layout like I expected) but overall it was a reasonably smooth showing.
It was good having TheAxeMan there to help physically get things set up and to have on hand during Battletoads (and being able to console verify the game-end glitch on the picture-in-picture camera). TheMexicanRunner and JC provided some fun banter for Battletoads with TheAxeMan throwing in technical details here and there and Geoff and Cyclops were on hand with some really good commentary for Super Monkey Ball 2. Audience response was reasonably good, although my daughter was resting her head on my shoulder and chat fixated on WrenTendo possibly being a vampire and the conversation was much more confusing after that.
So, some numbers - at the start of our block when they switched our mics on the total was $83,987 and at the beginning of Undertale (the run following the TAS block) the total was $86,207 for a total of $2,220 raised during the TAS block itself. Viewership hovered around 30k (I think - that's what chat said, please correct me if I'm wrong). After I had recovered from running the TAS block I made a donation of $176.36 during LoZ:BotW which is every cent I've ever made from Twitch.
That's all I have to say about that, I did my best given the very limited amount of time available to practice and I have no regrets. This gave me an opportunity to analyze what I still need to work on for AGDQ 2018 if I present and it also gave me an opportunity to gauge audience reaction to running on an emulator and I think we're safe there as it wasn't nearly the backlash I expected. Speaking of AGDQ 2018, submissions have been pushed back a week and I'll be chatting with folks in #tasbot and working out what to submit for that this coming weekend. Thanks again to everyone who supported the efforts so far!
The idea to verify the glitched run in the back was perfect, it came in a a usual TASBot mindblower that while you guys are enjoying this "skill ceiling" TASBot has already beaten the whole game. The shock the commentators (that are absolute batlletoads masters FYI) expressed was the best part of the show to me. Awesome job!
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.
I don't know when the time list is coming out but the game list is out if you want to see what else is being shown off.
It starts from Friday 8pm and ends on Sunday 8pm EDT.
Schedule isn't out yet.
