Beachwave is a reference client for live audio on ATProto. Identity comes from your ATProto account, rooms live as records in your own repository, and audio runs on LiveKit.
Reviews
(0)
—Be the first to review this product.
UpdatesView all
Developer's Notes 6.29.26
Jun 30, 2026
Playtesting Play To Clinch Tennis Pt. 1We already have two transfer windows, a Fall one and one after the season. I'm adding a third one that's the pre-season window that would simulate a late transfer period for tennis players. The reason? The game generates a lot of misallocated players early in the year and it's kind of tedious to move them.
Waiting for them to go in the Fall transfer window is fine, but it impacts the ITA event that we stage in the pre-season. I'll explain the uniqueness of the year-round tennis schedule after this, but this is a way to make sure things are at least somewhat resembling a real roster during the year before we get into the spring season.
There's a fall "indoor" season that in real life takes place in the winter, but for my sim, it easier to do this in the fall before we start the year since it's a simulation anyway. D2-D4 the Top 8 teams participate and are named Team Indoor National Champions (think indoor/outdoor track splits, essentially) and in D1, it's a 16-team bracket but they get there from 64 teams in my sim spread across 16 hosted pods that are single elimination within them.
So basically the pre-season NIT is what you have here. (I just renamed it that..it's a better name)
The fall portal will help me accelerate the talent development thing here at Monmouth, where I'll be starting with a fresh sim, maybe better prestige than last time but still bad in a conference that's not great, which is good for a competitive standpoint. We'll also build a strong non-conference. We will not be invited to the NIT, so that's not an issue.
The question is going to be how strong does our lineup need to be to compete at this level? Tennis rosters are only 6, but D4 rosters can be up to 16 players. D1-D2 can carry up to I think 12, D3, I think gets 14 or maybe they get 16 too, I'm not sure. At the D3/4 level with no scholarships, everyone is a walk-on, but the elite academic schools get something like scholarships for game-engine purposes to hand out, that makes players consider them over other programs.
Scholarship allotments are at least 6 for D1/D2 regardless of gender. I didn't feel like dealing with NIL, even though budgeting is wired into this sim because it's got a lot of the DNA of my baseball game embedded in it. (Like OOTP, a global universe of names)
Instead, think of scholarships like currency and each program has more than the base allotment. Players cost you a finite amount of "scholarship" if they sign with you. It used to be 1:1, but too many elite programs were maxing out early and stars were heading to too many mid-majors for my taste.
So now, the economy makes it a bit harder for those programs to accumulate anything besides the occasional 4-star.
Conferences used to determine scholarship bands, but I changed it so now it's based on your team's prestige and the conference just keeps you at a certain threshold. Coaches have more weight too now in the AI with where people end up, which I like compared to how it was when I started.
Okay, I got the agent to generate the actual scholarship/recruit figures:
```
_SCHOLARSHIP ECONOMY
_
Rosters are built by a scholarship-BUDGET economy, not a flat scholarship
count. A program HAS a budget (by conference tier) and SPENDS it on recruits,
who COST a scholarship-equivalency by star. To attract a tier at all, a program
must also clear that tier's budget FLOOR - that's what clusters blue-chips at
the blue-bloods.
RECRUIT COST BY STAR
Tier Stars Cost Floor to attract
--------- ----- ----- --------------------------
Blue Chip 5 7.0 >= 16.5 (blue-bloods only)
5-Star 5 3.5 >= 10.5 (major+)
4-Star 4 3.0 >= 5.0
3-Star 3 2.0 anywhere
2-Star 2 1.0 anywhere
1-Star 1 free walk-on depth
PER-PROGRAM BUDGET BANDS
(set by conference tier; funded within the band by prestige + a per-world jitter)
Tier Budget Typical core
----------------------- ------ -------------------------------------------
D1 Blue Blood (top) 16-26 ~3 blue chips - only blue-bloods stack them
D1 High-major (major) 9-16 a 5*/4* core, the odd blue-chip reach
D1 Mid-major (mid) 6-9 4*/3* core
D1 Low-major (low) 6-7 3* core, thin (the D1 floor)
D2 4-6 4*/3*; elite D2 (prestige >= 0.28) at top
D3 / D4 0 no athletic money - except gem alloc below
+- gem allocation 1-3 D4 academics >= 0.85 + Top-20 D3 by prestige
HOW A ROSTER GETS BUILT FROM THE BUDGET
1. Draw a budget from the conference-tier band, nudged by the program's own
prestige + a per-world jitter.
2. Sign the best recruit it can AFFORD (cost <= remaining budget) AND ATTAIN
(budget >= that tier's floor), repeating until the budget is spent.
3. Fill remaining seats with free 1* (or 2* at high-prestige D1) walk-ons.
So a budget-8 D1 program can't just buy a blue-chip (floor 16.5) - it builds
a 4*/3* core. Roster caps: D1 12 (8+4), D2 10 (6+4), D3/D4 16 (3+13).
Prestige is dynamic, so budgets drift both ways over seasons.
```
It's not a 1:1 though in terms of the AI knowing who is good. Instead, the actual OVR are hidden from the AI coach decison-making process. They get to evaluate players using several noisy metrics that don't always agree with each other. Each coach is tilted towards different metrics and it shifts year-to-year so that's what determines who they pick. My goal? Not just making the AI "smarter" but making it dumb like real life, able to make bad calls/decisions.
What really has to happen now in my evaluation phase is running multiple seasons and seeing how the game actually develops, I haven't done that yet.
We're going to try to do with Monmouth, not really just to tell a story, but to figure out how the thing works and how it feels under the hood.
Prestige this time is a 6, so a total doormat. Most of the D3/D4 programs that aren't academic elite schools are intentionally made to be very low prestige though so they can't recruit any too good, so that was a design choice, though the Top 20 D3 prestige programs and the D4 academic conferences get 1-3 "grant-in-award" awards to offer to players meaning they can't really get any blue-chips, but could get fallbacks.
The portal works as a circular economy, if a D1/2 program pulls a kid from one of the lower programs, they send a walkon downward to keep things balanced.
I forgot that Monmouth is in the MIAC not their normal Midwest Conference, while the MWC schools get a academic prestige bonus, which targets kids with high test scores, but none of them have the program prestige to target anything else.
I'm going to use the pre-season portal and see what kind of intake we can get, shedding the bottom walkons. I haven't decided what approach I'll take yet, I know getting a bunch of D1 fallbacks would probably be enough to win down here, but I honestly am not sure what the formula is because I've only ever been in "look at the results/testing/engine feel" mode not "roster construction" mode.
I know right now that a lineup that's just over 40 STR would put you in top tier across the divisions, because most of the top players are going to leave in the initial portal, but after that, I'm not sure what's going to actually work. I've only ever watched D1 so I'm not sure how variable things will be down here, and how dominant your lineup has to be to win at the lower levels.
So that's what we're trying to find out. In future test runs, we might be more organic about roster management, but this first run is really only to figure out 1) what does it take for the best team(s) to win 2) testing the pre-season portal which is new 3) clearing out overpowered talent in D3/D4 4) mostly importantly, play testing how realistic the results feel.
Jun 30, 2026Developer Notes: 6.27.26
Jun 29, 2026
