Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Samsara wrote:
Having done this, I would recommend that the submission file be replaced with EZGames' resync that goes through the first time setup, both for consistency with the currently published DSi runs and to prevent sync issues down the road. I'll do this as soon as I get permission from theballaam96.
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
101% TAS WIPs
Start - Enter Aztec 1: https://youtu.be/1ZdjCLoFe-Y
Aztec 1: https://youtu.be/FXqkua8SA8g
The route does exhibit a fair resemblance to the 2019 LOTAD (4:11) route, but will have some ik-level rerouting and Owl Race Skip in it (which affects the level order and does do some interesting stuff to route this in) https://youtu.be/YiGA7NyNQoQ
Owl Race Skip saves about 1m, some level rerouted should save another minute. The obvious goal is for the optimisation to bring the TAS down to sub 4. Given that the 2019 LOTAD doesn't use angle scripts to optimise movement direction, and some minor movement techniques like roll wiggling and ostand jumping weren't known either for the majority of development if the 2019 lotad or at all, I remain hopeful
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
I can't say I'm too well versed with Layton speedrunning, so I may be missing a subtlety.
Whilst it seems like this movie displays good optimisation (which you should be proud of), I feel like it isn't hugely entertaining after the first few puzzles.
I feel like this is a good candidate for standard
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Not wanting to come across as combative here, just want to justify the choice of this category and the decisions made:
ais523 wrote:
I also think a no-SRAM run might be more interesting. There'd be more downtime, but being able to start with (nearly) all the moves removes one of the main aspects of the game (i.e. buying the moves and becoming more powerful over time).
The no-SRAM equivalent of this run would minimise moves purchased. You'd basically get the following moves:
- DK's Gun (To free Diddy & Lanky)
- Diddy's Gun (Diddy Phase of K. Rool)
- Chimpy Charge (To Free Tiny)
- Simian Slam (To free Chunky, Mad Jack and Chunky Phase)
- Trombone Tremor (Lanky Phase)
- Hunky Chunky & Primate Punch (K. Rool)
Get the Golden Bananas that you can acquire with these (Given that you can get 165+ Golden Bananas without any moves, as evidenced here: https://www.youtube.com/watch?v=PsmC1GpEW-Y&t=0s&ab_channel=Ballaam . Fulfilling the GB requirements isn't really a task). So irrespective of the SRAM state, that aspect of the game is almost completely gone from a No Levels Early run.
ais523 wrote:
However, if you go that far, you'd probably want to also ban Troff 'n' Scoff skip (i.e. "no bosses early" in addition to "no levels early") so that everything that was meant to cost collectibles would actually cost them.
I think it could be debated on quite a fair bit on what would make the category appeal to a person more as everyone has their opinions on what would make the most satisfying run. However, to give insight, the RTA community came up with the category for No Levels Early to be similar to the "70 star category" in Super Mario 64 (In the same way that "70 star" is really a way to prevent levels being entered earlier than they should, with the ban on BLJs helping that, therefore adhering to SM64's primary mechanism of adhering to the game's intended structure). In DK64, the game's primary mechanism of preventing early level entry are the B. Locker requirements and K. Lumsy opening each lobby.
Troff 'n' Scoff skip was allowed in the category for the reason that whilst the act of collecting coloured bananas may offer interesting route options, turning in the 1680 coloured bananas would add around 11 minutes of cutscenes onto this. Whilst this is arbitrary, it's safe to argue that the majority of categories that aren't Any% or 100% are somewhat arbitrary. The design is intended to produce the most entertaining category to run for "middle categories", hence the ruleset being as it is
ais523 wrote:
I also think that any run that relies on things unlocked on a previous playthrough should be marked as newgame+ (or "glitched newgame+" if the previous playthrough isn't meant to matter but has been made to matter, as here).
Yes this is NG+, however if "glitched newgame+" would be the goal, then the run would follow a similar route to the one done by RTA runners of the "Any%" category. Which has been deemed as not as entertaining as regular Any% ( http://tasvideos.org/5376S.html ), a ruling I agree with. The differentiation NLE brings compared to Any% is grounded in adhering to the surface level requirements the game intends to place on the player (Keys & GBs), thus showing off more of the game than Factory, Helm & K. Rool (& Japes given the modern Any% TAS)
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Alyosha wrote:
Well, the benefits you are getting from SRAM come from previously entered levels. To me this seems counter to the goal.
If necessary, I can provide a movie which generates the SRAM state but with the SRAM generated movie also complying with the NLE ruleset. If that is what you are getting at
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Alyosha wrote:
How much does it cost not being able to press start?
The inability to press start honestly doesn't impact the run as much as you'd think. Since we're ending each level in an area where it's easy to void, most of the difference is the difference between pause exiting and voiding & leaving via the DK portal. It's probably around 45-60 seconds if you add it all up.
Alyosha wrote:
I don't really like using SRAM here, seems to clash with the other stringent restrictions, but whatever.
Mildly curious, what do you think the clash is with the category restrictions and the use of dirty SRAM?
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Thank you for the positive feedback :)
TwistedTammer wrote:
Hopefully a second video with the TBS camera hacked visible can be published alongside, if at all possible.
Unfortunately a straight conversion isn't possible. Not only would the inputs have to be changed to be relative to a different camera angle, but also the hack you are referring to causes a little extra lag from the game having to render level geometry. As such, the inputs would desync super easily. (Note that the game just desyncs very easily from slight changes. The translucent DK here took several takes from visifying DK every 2 or 3 frames to account for the desyncs that occurred: https://www.youtube.com/watch?v=1kEe0ZOiKy8&feature=youtu.be&ab_channel=Ballaam)
Yeah, I knew a bunch of LOTAD's had been made, but I was thinking of making a full on TAS by tasvideos rules.
I'm glad that you are interested in the world of TASing Donkey Kong 64. As someone who made the LOTAD Blooperio linked, it's good to see new people looking into full-game TASing outside of myself and RingRush. However, for 101%, I feel like I should give you the warning that if you want to go for it, it's a HUGE task for a few reasons:
- The run would be 4 hours long, where TASing DK64 usually is 1 year ot TASing for 60-90m of gameplay (if you go for TAS quality). Obviously this means a huge time investment, but people do make long TASes and spend years on them. However, those games don't suffer from the biggest problem DK64 has. DK64 tends to have huge game breaking discoveries that happen every year or two:
- 2012: Lag clipping
- 2014: Tag Barrel Storage
- 2015: Telegrabbing, other tag barrel related stuff, Dogadonequick
- 2017: Phasewalking, Phasefalling, Spawn Snagging, Orangstand TBS Noclip
- 2019: Crash Climb
These are just the big discoveries that either save minutes or are widely applicable that would affect routing. There are lots of things that get discovered semi-frequently too. The longer your run, the bigger the risk of some-to-most of your run falling victim to this.
This is some of the reason why 101% has only been LOTAD'd at this point in time (People don't tend to have 2 years minimum to dedicate to one TAS when there is a large risk of being "glitch-invalidated". 7 months to make a 101% LOTAD (The length of time it took to make the 2019 LOTAD) gets relatively close to TAS-level execution for lower risk and more time to do other things).
- In terms of TASVideos rules, you're running into the problem of spawn snagging and optimal TBS movement. I'm working on a "No Levels Early" (NLE) TAS at this point in time and if you want it to be remotely optimal, there are a tonne of sections which you will have to do under locked cam.
Examples:
https://www.youtube.com/watch?v=UKTxvVsfxq0&feature=youtu.be&ab_channel=Ballaamhttps://www.youtube.com/watch?v=Q9IKUBaTJHk&feature=youtu.be&ab_channel=Ballaam
These add up and generally people prefer to see what is going on, where you are going etc. For NLE, there's going to be at least 10 sections like this of varying length. For 101%, given how much spawn snagging helps with the slow GBs, you're talking at least 25 (Factory has 5 just off the top of my head).
ROUTE
In terms of route, whilst there have been a few new tricks since the 2019 101% LOTAD, it won't be massively dissimilar route wise. The whole "LOTAD" thing with the 2019 LOTAD is just that the execution isn't 100% in-line with TAS expectations and also that there are a few hacks used for viewer watchability (note that whilst hacks are used, the sequence of events are exactly what it would be if the hack weren't used).
If you are serious about wanting to TAS 101%, you can DM me on discord @ theballaam96#6738 and I will help you get started with the basics. DK64 TASers don't tend to check TASVideos forums much, so any WIPs or questions won't be answered until several days later if you decide to post in here.
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Great TAS as always, what the OoT community has done this year with SRM/ACE and optimisation is incredible. Really appreciate the level of detail in the submission notes too. Yes vote
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Certainly an interesting TAS, the "every block warps to the nearest warp zone" code certainly brings more originality to the table than the other two codes. Good play outside of the codes.
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Memory wrote:
The only reason somebody should veto a team name is if it is massively inappropriate. "Team 5" is far from that.
I second this. I can understand the confusion between "Team 5" and the unnamed team currently referenced as "Team Number 5". However, since TN5 is unnamed, why not give them a new temporary name until they come up with a name?
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
This is a bug with the interpreter/pure interpreter core types. You need to use the Dynarec core type in the plugins settings. If you are using a 2.x version of BizHawk, then Dynarec won't be in the list.
As such, you need to downgrade to 1.12.x or 1.13.x in order to have dynarec to fix this bug.
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
Barn to be Wild (BtbW) 100% (In the context of the 100% TAS. Doesn't make much of a difference aside from it being better to void at the end): https://www.youtube.com/watch?v=W3dJFn2LIns&feature=youtu.be
Uses "Perspective Force Slide" to move across swamps up to 5.7x as fast as the game normally allows.
As a refresher, BtbW 100% requires you to collect all 100% marbles and 7 udders. The 7 udders being tied to:
- Crossing the swamp at the start
- Destroying the 4 cameras
- Opening all the assorted barn doors
- Reaching Colonel Pluckitt
- The Helium Balloon Task
- Killing all enemies in "Bootcamp" in under 2 minutes
- Delivering Pluckitt's note to the crow
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
I can confirm that limiting how much you read and where you read stuff is best for keeping sanity. Every news site wants to post the worst about any situation because it's what gets the clicks and spins every story to fit what they want to say.
Something which has also helped me is also trying to look for the positives of the situation, not only is China really low on new case count (whether you choose to believe this or not is up to you, but I like to employ the idea of "believing something which could happen until I see evidence suggesting otherwise"), but Italy is also starting to see some form of decline in cases over the past few days. It feels like as long as you employ good social distancing, the worst of it will pass potentially by the time we hit Early-Mid June.
Whilst you have people being very greedy with stockpiling, you also have a lot of people offering to shop for others who are unable (for one reason or another) to go outside. This will undoubtedly raise awareness of not spreading around even small viruses. If this does become a seasonal thing or another virus occurs, hopefully the governments of the world will know how to react and what measures to take to reduce the spread.
And as Dacius said, employing the rules that have been set out will mean that are far less likely to pass on anything.
If you haven't seen, I feel like this video by Mark Rober helps with explaining some things: https://www.youtube.com/watch?v=I5-dI74zxPg
TL;DR: Don't look at new sources, there are positives to the situation, there is an end to this.
Experienced Forum User, Published Author, Player
(226)
Joined: 3/26/2019
Posts: 41
It's also best to start with just 1 console and make sure you get that to an accurate standard (since you are posting this on TASvideos, I assume that you are aiming for accuracy rather than PJ64-lag-emulation style "inaccuracy to make the games better") before working on more (and showing similar processes for adding more consoles to your emulator).
Starting with making 13 sounds great in design, but for a beta, it really needs to be less than 3 consoles. Otherwise you could get in a situation where you have builit an emulator for 13 consoles, and no-one is using it because it doesn't stand out in a particular field in any one of them.
EDIT: Not saying that PJ64 aims for inaccuracy, but more the fact that they are lacking accuracy in the lag department whilst working on more recent developments that are there to fix visual emulation glitches makes it seem that accurate lag emulation is not top of their priority list.
