#FGC Notepad, prototype and discussion

32 messages · Page 1 of 1 (latest)

oak quail
#

Post to discuss the tool I'm doing to help people share analysis on matches and scenarios, talk about the game and explore some interactions in detail.
This comes after a conversation in sf_coaching about the Scenario Analysis part of it.

Hopefully this doesn't count as spam as I just didn't want to have the coaching channel for the follow-ups of the conversation.

#

So, this thing already have some basic functionality right now, the Scenario Analysis is not the only part of it, it combines with other stuff.

#

I do think it's going to be tricky for people to use the analysis tool, and I do think it's way harder than in Poker but also a lot of the times when people are analysing pro level matches, the pause the replay to talk about a specific interaction that catches their eye, because sometimes pros choose an unintuitive response to a common problem

#

I do think there has to be ways to summarise that knowledge in small bits that can be written and saved.

#

Right now people learn through spreadsheets with okis, videos and twitter, which is fine but has its drawbacks. Beginners might not know key information that is not in oki sheets nor in basic guides (like what moves from your opponent are DIable or not), and I think the higher you're in MR the trickier these analysis become, so having a tool to help with them might be useful for everyone in the community

opaque cosmos
#

it kinda sounds like you're just trynna do everything

#

which btw, there's nothing wrong with having everything

#

but if you do go for that, I'd worry more about organizing the info

oak quail
#

In the backend it will be ok as it has a relational database to search and filter

#

In the front-end, I was thinking on different tabs for the different stuff people would want to search

#

The same that right now you go to the simple characters sheet and it’s organised by char, Basic guide, combos, oki, etc.

#

One tab for chars, other for moves, other for combos, etc.

#

I’m a bit more worried about importing the data, so people don’t need to rewrite their docs and stuff again

opaque cosmos
#

how does this compare to just the supercombo wiki that already exists

oak quail
#

It’s similar, but the wiki has some limitations.

  • Data isn’t easily searchable or filterable, you depend on the person doing the wiki organising properly.

  • It has duplicated info (if someone wants to reference the same combo twice, they need to copy-paste, in an app the use the same reference)

  • It has some problems with Seasons, as every season some data gets outdated and they have to manually check some things (in an app they can all be marked automatically)

  • It’s fundamentally different than a forum, a wiki is to gather important information. This app wants to promote discussion, it’s more similar to a subreddit that on the background has all the data from the wiki for people to reference and use

  • The wiki is in Wiki editor, the app goes in RichText with Markdown which has some benefits

  • Long term the app can do way more things like supporting videos / images for Match analysis and other functionality

#

I’m not saying the app is amazing (it has a lot of problems), or that this isn’t ambitious (it is), or that the current ways are awful (they’re amazing)

#

I just genuinely think this is reasonable to do in 1 year of my free time and wanted to try improving this

opaque cosmos
#

now I'm more confused tbh lol

#

how does a forum do all the things you're saying

oak quail
#

I say forum but it’s a general website

#

You can create posts but you can also import/save combos, oki, moves, analysis, etc.

#

Tbh I didn’t want to share all this until I had the Alpha version ready

#

So maybe it’s better to have follow up conversation then haha

#

Right now it’s just a draft and a long list of things to do

#

I think for August I will have a version ready for people to play around

opaque cosmos
#

aight, gl

#

cause yeah, right now it's hard to imagine it. So guessin you got some innovation cut out for yourself lol

oak quail
#

hey @opaque cosmos sorry for the ping, but a year has passed (wasn't in August as I thought, sadly) and now I have a... scuffed video of the app as it looks today, close to Alpha. I haven't talked about this with many people but I remembered having the conversation with you and I know you as a coach here and I know you play other fighters beyond sf, so I am interested in what you might think about it as a concept. Mind if I share a new video with you here on the update?

opaque cosmos
#

Yeah sure thing

oak quail
# opaque cosmos Yeah sure thing

thanks!, it's this: https://www.youtube.com/watch?v=NZmCsC2uqPE
The TL;DR (as it's 25 min and I know that's a lot to ask for), is:

  • It has an OKI analysis tool to find optimal Nash Equilibrium to calculate how much you should do each action in the given OKI
  • It has layers, to separate people that know how to execute all the defensive and offensive mechanisms from the game from the beginners, so not to overwhelm them
  • It allows for recursivity calculations (throw loops for example, the fact that loops generate value because they can be repeated)
  • It has a database of combos so you get accurate damage numbers depending on a given starter, and resources (so you can study an OKI if you have only 2 drive, or 3 super, or any given combination, it tries to search for the "best" combo in those circumstances)
  • Most of this I think it's applicable to other fighters, as you can get RPS scenarios outside oki, and this is basically a RPS analyser with specific features for fightings

Example for the open-source Wikipedia style website I'm building for SF6 (and potentially other fighting games). Any questions or feedback, I'm all ears!

Logo of the website is done by @miguel.type at instagram, spanish graphic designer!

▶ Play video
#

Obviously, it's still super Work in progress though. A lot of things are questionable, but any comment you might have I'd appreciate it. Like even if it's that you think no one would use this, I still would find it helpful if you explain a bit why