lupi

Lupi

@lupi

cow of tailed snake (gay)

avatar by @citriccenobite

you can say "chimoora" instead of "cow of tailed snake" if you want. its a good pun.​

i ramble about aerospace sometimes
I take rocket photos and you can see them @aWildLupi

I have a terminal case of bovine pungiform encephalopathy, the bovine puns are cowmpulsory

they/them/moo where "moo" stands in for "you" or where it's funny, like "how are moo today, Lupi?" or "dancing with mooself"

᠎
Bovigender (click flag for more info!)
bovigender pride flag, by @arina-artemis (click for more info)

log inask
twitter
twitter.com/awildlupidragon
twitch
twitch.tv/lupidragon
lupi
@lupi
MisfitMouffette
@MisfitMouffette
mcc
@mcc

Unity Corporate Self-Immolation Fest 2023: Final(?) update

Tuesday before last, Unity out of nowhere dropped an announcement that they were changing their pricing structure, killing the Plus tier and introducing a new, completely unworkable pricing scheme based on "installs". After a series of meaningless "clarifications", today Unity issued a partial walkback. You should read that, I'm not really going to summarize it.

I have three thoughts.

The Runtime Fee policy will only apply beginning with the next LTS version of Unity shipping in 2024 and beyond.

Thought 1: The fact that the new fees only apply to future versions is the critical change in this latest announcement, and probably(?) avoids a lawsuit. This will have interesting ecosystem effects since many people will now want to stick on 2022.3 LTS indefinitely. Possibly, some libraries will latch to 2022.3 and never update.

(Note that 2023 LTS will be released in 2024 and thus the new fees will apply. Yes, this is confusing.)

Thought 2: Which is more expensive now? Unity or Unreal?

So this gets grits teeth complicated.

The new policy is a bit confusing, but much easier to understand than last week's. They now claim that, once you hit $1m revenue, they will charge the lower of 2.5% revshare, or a charge per "number of new people engaging with your game" that they "calculate" based on your "self-reported data".

The part after the "or" there is the remnant of their baffling "install"-based pricing from last week. This new, "new installs" number is still untrustworthy gibberish. But! Now you can ignore it. So the "price" for using Unity is now effectively 2.5% of revenue over $1M, possibly reduced arbitrarily.

2.5% is less than Unreal's revshare (5% over $1m). But Unity is also still charging a per-seat fee, which Unreal doesn't. (In today's announcement they raised the revenue threshold for the per-seat fee to $200k— but they also killed the Plus tier, meaning for the smallest shops the per-seat fee just increased 5x.)

A per-seat fee plus a revshare is sour tasting. ("They're getting you coming and going.") Per-seat fees are really inconvenient for small operations like mine where some or all of your devs may only use Unity part time. But never mind how it feels. If you combine the revshares and seat prices, which costs more now?

Under $200k revenue, both engines are free. Under $1M revenue, Unreal is free. From $200k to $1m revenue, Unity costs $2050 for one dev or $41k for 20 devs. After 1M revenue, Unreal rapidly starts to cost more, although even with as few as 20 devs you need to make $2.6M before Unity is a cost save. With $3m revenue and 1 dev, Unity costs $52k and Unreal costs $100k. With $3m revenue and 20 devs, Unity costs $91k and Unreal costs $100k.

Prices USD. This assumes both Unity and Unreal charge revshare on marginal revenue over 1M. It assumes one year of seat licenses and assumes all revenue comes in over a single year. It also assumes PC or mobile, not console. This image and explanatory paragraph are released under CC BY-NC 4.0; Grapher and Pixelmator files available on request.

Fans of Homestuck will recognize this diagram immediately.

Okay, this is a mess! But here is my takeaway: Unless you are making a lot of money, Unity is the more expensive option. If you're making a lot of money— multiple millions in revenue— then Unreal's higher revshare fee very rapidly dwarfs the price of Unity. But you have to make $1.08 million on a single-dev project— or $1.41 million on a 5-dev project, or $2.64 million on a 20-dev project— before Unity starts to look like a better deal than Unreal.

Is 2.6 million dollars a lot of money? Well, here's what's weird. The range where Unity looks like a worse deal is between $200k and $2.6m in revenue. Make no money and it doesn't matter, make over $2.6m and Unity is a money save. But that "mid-range" where Unreal does better? That's supposed to be Unity's target market. That's where Unity originally made its name. If Unity is now trying to abandon the midrange and make a play for the high end, there's a different problem: Unity isn't as good as Unreal on features. Maybe the idea is to abandon the midrange and the high end, but milk high-revenue, low-quality mobile developers who don't care about features— but if they don't care about features, then instead of paying Unity why not use all that money to retarget to (say) Godot?

Thought 3: This is manageable, but Unity is dead anyway

My gut is that if this open letter's policies were what Unity had announced to begin with, like last week, then people would have been annoyed, as they were with the 2021 changes, but you wouldn't have seen a community collapse or mass exodus.

However, that's not what happened. I do not think this open letter will halt the mass exodus. The core problem is that Unity can no longer be trusted. You can consider the new 2024 prices acceptable. But now what you really have to worry about is how they will change in 2026.

EDIT: Then there's this

MisfitMouffette
@MisfitMouffette
This page's posts are visible only to users who are logged in.
log in
You must log in to comment.

in reply to @mcc's post:

mcc
@mcc

3 more bonus thoughts

  • All above analysis assumes that the things said in the "open letter" today are actually true. It's very possible the letter is lies, or else will be walked back in two days.
  • Did you notice how the folksy "open letter" is written by a manager from engineering? So the C-suite made this huge mess, but when it came time to clean it up, they send engineering out.
  • This took way longer to make than I expected and I want to send Unity an invoice for my last 3 hours
login to reply
LotteMakesStuff
@LotteMakesStuff

Just a point of clarification, the guy who wrote that letter was the guy who made all the calls on the new business model changes. Even the original ones. He is the C-Suite executive in charge of the 'Create' business unit - which is the one that interfaces with game developers and owns the engine toolset. This was all his call, so yeah he wrote the letter. You can verify that yourself on the corporate webpage if you check out the leadership section https://unity.com/our-company

That said, John Riccitiello's silence over the last two weeks has been deafening, and I think his position in the company is untenable and he needs to go - but I've been saying that for years.

login to reply
mcc
@mcc

Also, a point made by Maddy Thorson on Mastodon: Above I assume the revenue all comes in over one year and the per-seat license is for a single year. But as long as you qualify for the revenue threshold, you have to pay the per-seat license every year. Your next game will probably take more than one year to finish, but you're not going to make as much revenue the second year after your last game's release. In a whole-project graph, the per-seat cost would probably be multiplied by a factor of 2 to 5 vs above. In other words, even though this graph really doesn't make Unity look like a good deal, I'm actually being unfairly generous to Unity!

I guess if you're making microtransaction-pump mobile games it's plausible you could develop a game in a year and you could keep up sustained income from a game 1 year post release. Once again, rapidly developed microtransaction-pump mobile games seem to be the only market segment Unity's interested in selling their engine to.

login to reply
mcc
@mcc

That makes sense. But surely crowdfunding is not an equity/debt situation? What about, say, small business development grants? In lots of places it's expected very small indie shops are receiving significant help from local-government media development funds, and only some of these funds are structured in a way they take equity.

login to reply
jackie-stackie
@jackie-stackie

I've seen the term crowdfunding used to refer to three different concepts:

  • Fundraisers, like GoFundMe, which I don't think are relevant here.
  • Broad but nonpublic equity raises, like WeFunder, under the SEC's Regulation Crowdfunding. These are generally not considered revenue
  • Presales of a product, with the idea that the money raised will help develop the product, like KickStarter. These are considered revenue, but the manner and the timing with which that revenue is booked will depend on the accounting method used.
    Definitely a confusing topic, and just a couple weeks ago I was talking to a credit underwriter who didn't understand the differences

I'm not sure about small business development grants! I haven't worked with them before, and I'm not sure how they're structured. My general understanding is that, at least for tax purposes, debt is typically not booked as income or revenue unless/until it is forgiven. Even a zero-interest loan, which has significant economic value in the current environment, is not considered to be income or have value in the accounting sense. So:

  • government purchases an equity stake: not revenue
  • government loans money, even at favorable terms: not revenue
  • government forgives a prior loan: definitely revenue, might or might not be attributable to a specific game?
  • government gives unconditional money as part of a grant program: definitely revenue

I'm not an accountant, and while I am trying to learn accounting principles, I'm still very early in that journey. This represents my best understanding of how I would expect accountants to treat this, but I could definitely be wrong.

login to reply
mcc
@mcc

Okay, having had a night to think about this there's an obvious way that this graph is still "usable": Instead of treating the Z+ axis* as seats, treat it as person-years. Multiply the number of people on your team by the number of years the team worked on it, and look that up on the Z axis.

In order for that to be fair to Unity, you'd need to pay the per-seat fee every year, and there would need to be no tail on your game, like, you'd need to make the entire million dollars (or whatever) in a single year. But that's not an uncommon case.

* Unity axes

login to reply
holden
@holden

This is a really helpful writeup. I agree with your conclusion, I don't think we can trust Unity, and my studio is preparing to move to Unreal for our next project.

For people who are stuck with Unity in the near term, though, the install fee model is probably worth another look because it might be a better deal than revshare.

They changed the install fee policy so that you can just count unit sales instead of installs if you make a premium game, and you no longer have to worry about installs on multiple devices or any of that stuff. Revshare is only a better deal if your average sale price is under $6/unit (assuming Unity Pro, since 2.5% of $6 is 15 cents.). You can also change between choosing revshare or install feels on a month-to-month basis and only ever pay the cheaper of the two.

So if the average sale price of your game is $20, you can basically treat the install fee as a 0.75% revshare instead of paying 2.5%. That could change the math for some folks in the short term until they inevitably change it again and make it worse.

login to reply
cidolfas
@cidolfas

I do have to wonder how you factor in lump-sum deals (like putting your game on Game Pass). When you get that money and when the game is getting new "first-time interactions" from them can be pretty spread out - so do you get to pick revenue split when it launches on ther service but go against purchases the month you decide to include the payment in your revenues and essentially exempt those deals from the cut?

Also do the fees (and the $1m threshold) apply per-product like the Unreal ones do - or is it for anything your'e selling as an organization so long as the org has crossed the rolling 12-month/$1m revenue threshold? Those can wind up with very different outcomes for studios with multiple smaller titles out.

Still a lot of unresolved questions.

login to reply
holden
@holden

Yeah, tons of questions still. Do upfront GamePass deals count as sales revenue? Because I assume if you get paid before launch, you can't just pay $0 in install fees instead of 2.5% of the revenue...but when does that start getting counted?

Does publisher funding or grant money count if GamePass funding does?

😬

login to reply
mcc
@mcc

The way I look at it, regardless of whether the new-install fee is a good plan or not, it is difficult to predict (also, it's not clear to me as of this blog post what the per-unit new-install fee is, since they changed several things and only some of them are reflected ). So to me it's easier to reason about the revshare, since I can at least think of that as a ceiling. My engine cost will be that or lower.

login to reply
holden
@holden

That's fair, but even if you do just want to plan on revshare, the change in how they are counting installs does matter because it means none of this kicks in until you cross 200k or 1M units sold, depending on which Unity plan you have. So in theory, a game that has made $2M in revenue would pay 0% revshare and no fees if the average sale price was over $2.

They went into a lot more depth in the new docs: https://unity.com/pricing-updates

It was very unclear (and often contradictory) in the previous versions but they now say that for premium games, you should use "number of units sold" and subtract refunds as your estimate for monthly new installs, and ignore everything else.

login to reply
Pinned Tags