3 ms·
> I think the only reason Unity went this route is because they don't want to be seen to publicly break the promise of no revenue ... Not quite. Unity merged w
by philipov 3y ago
> I think the only reason Unity went this route is because they don't want to be seen to publicly break the promise of no revenue ...
Not quite. Unity merged with a company that sells mobile ads: ironSource. They are direct competitors of AppLovin, the dominant player in that market. The terms and backroom deals that Unity is offering devs will waive these fees if they use Unity's ad products instead of AppLovin. This is clearly an attempt to abuse their position in one market (game engines) to bolster their position in another market (mobile ads).
- and0 3y agoThat's fascinating, thanks. None of the various analysis I've read so far have hit on this point but it makes sense.
- philipov 3y agoI think HoegLaw did a good rundown of the legal side of it: https://www.youtube.com/watch?v=rGMrebXypJo https://www.youtube.com/watch?v=rGMrebXypJo
- antonjs 3y agoKind of feels like this is the Boeing--McDonnell Douglas of game engine acquisitions.
- nugget 3y agoUnity seems to be attempting this in the most deceptive and deceitful way possible, establishing the new Runtime Fee and then offering a temporary 100% "waiver" of the fee if you use their other (presumably inferior) products. As soon as the pressure fades, the waiver will be reduced to 50% and then eventually dropped completely - but of course the new fees will remain. They must think the average game developer has no business sense whatsoever. Based on the backlash, my prediction is that Unity either quickly reverses course (damaging their brand a little and perhaps costing the CEO his job) or stubbornly doubles down (damaging their brand a lot and giving Godot and others an opening to eventually rival them).
- KronisLV 3y ago> Unity seems to be attempting this in the most deceptive and deceitful way possible, establishing the new Runtime Fee and then offering a temporary 100% "waiver" of the fee if you use their other (presumably inferior) products. I looked into it a bit more and unless I did some bad maths or misread their terms, the whole Runtime Fee looks like a badly disguised sales funnel to me: https://blog.kronis.dev/articles/unity-runtime-fee-a-look-at-some-numbers https://blog.kronis.dev/articles/unity-runtime-fee-a-look-at... The Personal and Plus tiers in particular now need to basically find additional 50 or so cents per install (factoring in platform fees and publisher fees), whereas for Pro and Enterprise tiers that figure is closer to under 10 cents). In other words, once you start having to pay the Platform Fee on the Personal or Plus tier, it very quickly becomes cheaper to just get a Pro subscription and have the Platform Fee go away for 800'000 more installs on Pro (on top of the 200'000 you get without the platform fee on Personal/Plus).
- WillPostForFood 3y agoHow are you calculating .50 cents personal vs .10 cents pro?
- KronisLV 3y agoClosest to 50 cents and 10 cents was at around 1'600'000 installs I think (some graphs at the bottom of this comment). Let's see how I got there, let's take their Runtime Fee rates: https://unity.com/runtime-fee https://unity.com/runtime-fee For Personal tier you start paying after 200'000 installs (threshold), whereas with the Pro tier you start paying after 1'000'000 installs (threshold). That gives us 1'400'000 installs that we need to pay for with the Personal tier and 600'000 installs that we need to pay for with the Pro tier, for which the runtime fee will be calculated and which are above the corresponding thresholds. For the Personal tier, which has a fixed rate, the platform costs are then: 1'400'000 * 0.20 = 280'000 USD For the Pro tier, with the volume discounts, the platform costs are then: 100'000 * 0.15 = 15'000 USD 400'000 * 0.075 = 30'000 USD 100'000 * 0.03 = 3'000 USD TOTAL: 15'000 + 30'000 + 3'000 = 48'000 USD (they're lower for emerging markets, but using the main prices here) With 1'600'000 units sold, the overhead per a single unit then becomes: Personal: 280'000 / 1'600'000 = 0.175 USD Pro: 48'000 / 1'600'000 = 0.03 USD If we assume that a publisher might take around 50% (just an example value in the spreadsheet I used; though in practice can be as low as 20%), then that figure becomes: Personal: 1/(1-0.5) * 0.175 = 0.35 USD Pro: 1/(1-0.5) * 0.03 = 0.06 USD On top of that, there are also the platform fees (like Steam might take 30% of your revenue straight off the bat, other platforms might take less), so the figure then becomes: Personal: 1/(1-0.3) * 0.35 = 0.5 USD Pro: 1/(1-0.3) * 0.06 = 0.085 USD Sanity check (doing the same in reverse): Personal full cost (overhead): 0.5 USD Personal after 30% platform cut: 0.5 - (0.3 * 0.5) = 0.5 - 0.15 = 0.35 USD Personal after 50% publisher cut: 0.35 - (0.5 * 0.35) = 0.35 - 0.175 = 0.175 USD Pro full cost (overhead): 0.085 USD Pro after 30% platform cut: 0.085 - (0.3 * 0.085) = 0.085 - 0.0255 = 0.0595 USD Pro after 50% publisher cut: 0.0595 - (0.5 * 0.0595) = 0.0595 - 0.02975 = 0.02975 USD (close enough to 0.03 USD) While the exact percentages might change, Unity asking for say 0.20 USD per copy (or effectively 0.175 USD in the example, because the first 200'000 don't have the runtime fee) means that you'll need to make more gross revenue per copy than that, because your publisher and the platform will both take some of that for themselves. Here's the graphs that I came up with, first how the total runtime fee changes based on the tier: https://blog.kronis.dev/images/1/4/-/u/n/14-unity-pricing-graph-e868be73.jpeg https://blog.kronis.dev/images/1/4/-/u/n/14-unity-pricing-gr... Also, approximately how the runtime fee overhead changes with publisher/platform fees included (with the example percentages above): https://blog.kronis.dev/images/1/6/-/f/e/16-fee-per-install-graph-dc7488d8.jpeg https://blog.kronis.dev/images/1/6/-/f/e/16-fee-per-install-...
- ToucanLoucan 3y agoI could see that, but based on the massive stock unloads from numerous high level execs right before hand, it seems more likely to me that it's just another company that's going to be cratered inside of a year, stir up tons of controversy and legal problems with stupid decisions, and then everyone responsible will leave after the bankruptcy up millions of dollars. And another perfectly viable business will be stripped and sold for parts on the market because that looks better on balance sheets than just making and selling a product these days.