Spine is really lovely. It was great back in 2019 when I used it to release a game. They released the 4.0 version a year later with a lot of updates and while I have no plans to make another 2D side-scrolling type of game... this one piece of software is what makes it tempting.
The thing I REALLY appreciate is the lifetime license. Checking my emails, I bought the professional version in 2015 for $289. Now it's almost a decade later, still getting awesome updates, and I can still get the latest version. Totally unlike the current trend... I'm looking at you, $20 / month for Speedtree.
I mean, Spline seems to do the same thing if your company makes enough money:
>Spine Enterprise is required for businesses with $500,000 USD or more annual revenue. The license allows Spine Enterprise to be used by the specified number of users for a period of one year. All updates are provided during this time. After one year, Spine Enterprise must be licensed again to continue using Spine Enterprise.
common anywhere. Even if you paid only $3 a month 10 years ago, that subscription would still be more expensive to a lifetime professional as of now ($360) than your upfront price of $289. maybe a bit cheaper when you adjust inflation, but another year would close that gap.
So you get the allure from the business side, since they are also continually spending time and probably labor updating the tech.
Slightly different target use cases. You'd use Rive for your UI, but not necessarily your player character and other things that you want animated "with" the gameplay. For that, you'd use Spine.
Yeah, although there are Rive examples using it as a game it's really targeted to vector-art your UI and allows you to create animations inside their software, which you then just play in whichever engine (such as Unity etc).
Spine is fantastic. I'm using it for my latest game, Danger World. All of the animations my artists are working on are done in Spine and I'm able to use a bridge package that someone wrote to integrate what I need into Flame's engine.
It would be great if it was possible to develop a plugin for an engine to load spine in a game and let the decision of the license to the game developer (and not the engine developer). Right now you either are blessed that the spine devs decided to care for your engine or you aren't. It would be nice if a FOSS engine dev could develop a Spine plugin and then a game dev that was thinking about that engine could either opt to use it or not.
This is basically what the community is already doing. This is the package I've been using in Flame and its license (https://pub.dev/packages/flame_spine).
"License
The code and all associated assets of the flame_spine package is distributed under the MIT License (see the LICENSE file). HOWEVER, as flame_spine includes Spine Runtime libraries, your use of flame_spine in your games or other products is also subject to the terms and conditions of the Spine Runtime License Agreement (see the LICENSE file).
In particular, in order to use flame_spine you are required to obtain a valid Spine Editor License."
In the legalese text it's written in way that the person that is making the plugin must also hold a license which is too much, not going to pay to do work.
Anyone can distribute the Spine Runtimes without a Spine license, but they have to attach the Spine Runtimes license, which means the need for a Spine license is pushed down the line to their users. In this way someone without a Spine license can use the Spine Runtimes in game toolkits/etc without having a license themselves. https://esotericsoftware.com/spine-runtimes-license
Presumably, if you’re writing a Spine runtime, it’s because you’d like to integrate it into one of your own projects in an environment that they don’t officially support. I’m not sure why you’d take on that “work” if you didn’t already have a license.
>the person that is making the plugin must also hold a license which is too much, not going to pay to do work.
I suppose the idea is that you're paying to work on something that will make you money. And then if you have a FOSS mentality you can share that for others using your engine.
But I'm sure there's wiggle room if you contact. if you're an up and coming engine and are interested I'm sure Spine only has to gain by having its tech ready to go with it.
No, that's exactly what I meant. The author says they want to write a plugin which uses spine-cpp to render Spine skeletons.
A user of that plugin needs a Spine Editor license. The author of the plugin does not. Users of the engine not using the plugin do not need a license either. That's how integrations with e.g. GameMaker, Defold, etc. have worked for years.
Spine is fantastic for artists, and it's integrations with various engines is really good. My biggest complaint about it is that it's very CPU heavy. A spine runtime that has less features but uses GPU skinning would be excellent, especially if it runs on the web. Ages ago I had to write a JS shim to the C runtime compiled to WASM and it worked well, but I still wish it used a lot more GPU and less CPU.
The skinning could be done GPU side but with caveats: the number of bones that can influence a vertex would be limited and matrix palettes might have to be split up, inducing multiple draw calls. There actually was a GPU skinner at some point iirc, and the load for common example skeletons was pretty much the same sadly.
It's being used by a lot of companies with WebGL tech stacks mostly deployed on low-powered slot machines. Quite a few web games in the Asian market also use it. However, three.js likely won't add anything for a 2D game. You'd be better served with Phaser or PixiJS (which is the industry standard in slot games it seems).
It is a popular design convention, there are even generators to convert regular screenshots into this angled and depth-of-field style. There are 2 important properties this convention has:
First, it makes it obvious that the UI being presented is not part of the website, but a screenshot, so people don't trying to mess with it like it's an interactive widget.
Second, it showcases a general feature without inviting scrutinising the details of the UI, which a general product page isn't the best place to do so anyways.
---
(Personally I do love a good software Screenshots page, but that can be a separate thing IMO)
At one point in my life I did a lot of work animating comic book art for various projects. I wish this existed back then. It would have made the work so much better.
If you're making $500k/year in revenue, Spine's license cost is probably the least of your worries.
I tend to be averse to proprietary dependencies more from a maintainability/longevity aspect than a cost aspect. If Esoteric Software goes out of business, who develops future versions of Spine? Are they kind enough to open-source it before turning the lights off? Does it get sold to some other company that runs it into the ground? Does development just cease? Are there licensing servers that the tools and SDKs and such have to ping to be functional, and if so, what happens when those servers shut down?
To be honest I am surprised there isn't an open source alternative for this type of animation in 2D - I only know of blender and lottie as somewhat closer to it but nothing that is quite like Spine.
2D skeletal animation is an order of magnitude easier than 3d, but still a huge task in and of itself, with unique issues working with sprites compared to meshes. So it still requires very skilled understanding of art and physics, but there's a lot less demand as the AA/AAA industry is 99% 3d in the west.
It's not open source, but the best competitor to Spine is probably Live2D, which is used by a lot of Eastern studios (and basically dominant in the mobile space). You'd probably find some open source implementations over in Japan.
Spine is extremely polished, thought out and well written piece of software.
Expecting that somebody will do this kind of highest quality work for free is ludicrous.
We actually come from a strong FOSS background (Kryo, libGDX, and more).
For Spine, FOSS didn't make sense to us, as you either end up with open-core, which over time will shift incentives to work on the proprietary parts, or a donation based funding model, which is stressful and unpredictable.
I would not say same for Blender. Blender had absolutely god horrid UX for decades (it still has in many places) it is only very recently that they have gotten a bit better and polished some of the rough edges.
I don't know about Krita, but most open source software of considerable complexity and scope has god awful UX. It's no accident you named only two, of which Blender was infamous for it's bad UI and unintuitive nature.
Gimp is also the type of UX you generally get from foss.
> Blender had absolutely god horrid UX for decades
Yes, but it doesn't have it anymore: i personally know several 3D artists who switched from other (commercial, proprietary) programs to Blender and absolutely love it. You'll also find a ton of videos on YouTube from 3D artists praising it.
How it used to be is irrelevant because my point is that being opensource didn't prevent it to become a tool that is loved by its users for its UI.
It is also why...
> It's no accident you named only two
...i named Blender and Krita specifically (and intentionally, thus indeed not an accident) and did not brought up GIMP or anything else, because Krita is also loved for its UI (even if i personally am not a big fan of it, there are many artists who use it and like it).
My point was, again, that being opensource doesn't prevent having good UX - lacking good UX is not a byproduct of being opensource, it is a byproduct of the developers who wrote the software not caring about good UX.
Blender (via https://github.com/ndee85/coa_tools) and Godot's Skeleton2D node are both open-source. I'd assume Spine's advanced features far outpace them, but I also suspect the need for such features is rare.
It just seems like a "money grab" for them to claim a chunk of success for my game/product each year when I probably don't need any more support for the tool than when I was < $500k/year in revenue.
If you could make 500k/year (and lets say it reduces to a drastic 200k after taxes/platform cuts), I don't see why paying 1.25% (+0.2% per extra seat) is a hard ask for a core tech.
The enterprise license here seems pretty reasonable. $2.5k/yr+$370/yr per extra seat with no revenue share, and no cutoff for your existing products at the end of the year.
I love the developers are seeing that they deserve to be paid for their work and that corporations do not deserve to make money hand over fist for free.
More developers, especially “open source” ones, need to be charging corporations for their work.
Spine is used extensively in the gambling industry for 2d slot machines, so it makes sense to have a licensing model to accomodate these users, while still offering a tier for indie devs.
The Enterprise license doesn't claim any revenue share. It's a subscription model that triggers above a revenue threshold. It cross-finances the non-enterprise, pay once lifetime licenses for individuals, which get updates forever.
Isn't this trivial to get around? What if you worked with individual art contractors (which is not that uncommon in of itself), then the revenue requirement wouldn't manifest on your side, but theirs.
It's fine if they don't use the Spine Runtimes to playback animations, i.e. via sprite sheets. Most will want to use the Spine Runtimes though, which requires them to have a Spine license.
Close! It was 1 dev and 1 artist for a long time. Eventually the runtime and support load became too high and it's slowly grown to 9 people. Running lean was satisfying but exhausting.
The thing I REALLY appreciate is the lifetime license. Checking my emails, I bought the professional version in 2015 for $289. Now it's almost a decade later, still getting awesome updates, and I can still get the latest version. Totally unlike the current trend... I'm looking at you, $20 / month for Speedtree.