Quoting: TheSHEEEP

Quoting: flesk

Quoting: F.Ultra

Quoting: TheSHEEEPIf I ever make a game of my own, I'll only actually distribute a launcher to each platform. That launcher will then take care of actually downloading/patching the game. Sure, it would mean having to maintain my own server(s), but I've been using Amazon S3 & AWS (and other providers) for quite a while now... it's not that difficult.

Seems to be the only way to prevent having to maintain different builds across the various platforms, each with their own uploading, etc.
Also has the added benefit of being able to serve as a built-in mod-manager.

Or perhaps why not create a cross-distribution build platform that can build and handle the distribution to steam, gog, itch.io and so on. Could potentially by worth some real money for game devs?!

Such things already exist and is known as continuous integration and deployment (CI/CD) in "regular" application development. The idea is that you use software (Jenkins, GitLab CI/CD, Travis, Circle CI, etc.) to pull code from a repository when you push changes, and then build and deploy to the store fronts you want your game on, eg. whenever you version tag a commit. For this to work, however, the stores must have an API or software that makes it possible to automate the process. Itch has Butler, GameJolt has a similar tool that's in beta, and Steam has SteamCMD, though I've not used the latter.

Which is a very large amount of work to set up and maintain. You have to do configurations per distribution platform per OS per version per <possible other configuration>.
And then you need to react to API changes, etc.
In large companies, you have people who just maintain the CIs as their main job, because while it does make things easier, it is still far from easy. I've been using a few of those at work and none of them are easy to use if you need to do some more configurations.
And then you still have to upload every single build to the distribution platforms without such an API. So you have to do it anyway if you are on more places than just Steam.
Then you have to make a news post on all the platforms as well, etc.

It might still be in total less work than writing a launcher yourself, but a launcher comes with a LOT of added benefits: News posting within the launcher, version management, mod management, crash reporting (can be done inside the application as well, but is MUCH easier to do in an external application), various settings that don't really fit in the game options, etc. And all of that fully independent from the distribution platform...

Of course, such a launcher would then have to be maintained itself per distribution platform per OS, but launchers do not change often once functional.

The biggest downside is likely the cost. After all, you already pay the distributor cut (which WOULD include the traffic cost for your game bytes), and in addition to that you have to pay whatever traffic is needed so people can download/update your game via your launcher. So you likely need a CDN as well. And so the costs rack up.

You'd have to do the math per case, but in the end, it might just be cheaper to not let your launcher do the downloading/patching. That doesn't mean you can't have a launcher anyway, of course (see Paradox games).

Yes, I agree that it might be a lot of work if you don't have a skilled developer on your team, which might be the case for lots of indies. I'm not sure if dedicated software is necessary though, when providing example templates for each combination of engine and CI/CD platform would be enough. Of course, it might be easier to market and charge for a software solution, if that's the goal.

View this comment - View article - View full comments

Quoting: F.Ultra

Quoting: TheSHEEEPIf I ever make a game of my own, I'll only actually distribute a launcher to each platform. That launcher will then take care of actually downloading/patching the game. Sure, it would mean having to maintain my own server(s), but I've been using Amazon S3 & AWS (and other providers) for quite a while now... it's not that difficult.

Seems to be the only way to prevent having to maintain different builds across the various platforms, each with their own uploading, etc.
Also has the added benefit of being able to serve as a built-in mod-manager.

Or perhaps why not create a cross-distribution build platform that can build and handle the distribution to steam, gog, itch.io and so on. Could potentially by worth some real money for game devs?!

Such things already exist and is known as continuous integration and deployment (CI/CD) in "regular" application development. The idea is that you use software (Jenkins, GitLab CI/CD, Travis, Circle CI, etc.) to pull code from a repository when you push changes, and then build and deploy to the store fronts you want your game on, eg. whenever you version tag a commit. For this to work, however, the stores must have an API or software that makes it possible to automate the process. Itch has Butler, GameJolt has a similar tool that's in beta, and Steam has SteamCMD, though I've not used the latter.

View this comment - View article - View full comments

Quoting: liamdaweAlso, to those asking about A Hat in Time, VP simply told me to speak to the publisher which I've done so. Waiting on a reply.

I wonder what that's supposed to mean. If there's some sort of NDA in place, you'd think they'd broken that already by accidentally leaking the game twice now, and shoveling inquiries over to the publisher. It doesn't seem like a secret they're keen on keeping at least, if there's anything to it at all.

View this comment - View article - View full comments

Looks fun! I hope we don't have to wait too long for the Linux version.

View this comment - View article - View full comments

I love the first game and still enjoy playing it every now and then. I'm super excited about the sequel! :D

View this comment - View article - View full comments

Now that's just mean if it turns out there's nothing to it. :'(

View this comment - View article - View full comments

Wow! I'm super excited about AHiT coming to Linux!

View this comment - View article - View full comments

I also loved Fran Bow, so this will be an instant buy for me, when available for Linux.

View this comment - View article - View full comments

Looks very fun and well made. :)

View this comment - View article - View full comments

Definitely A Hat in Time or one of the LEGO games this time.

View this comment - View article - View full comments