What suggestions and ideas do you guys have? What keeps you going through your projects?

My biggest problem usually is getting started on a given day, especially if I know there's a huge task to be done. So I tend to keep one or two small tasks around, like something simple to refactor, or any editor-comfort feature that is not working as needed. This encourages me to just pop up visual studio and get going, and once I'm at that point, its not hard to keep on.

Then, as others have said, if its a hobby project and you just don't feel like going on, just take a break, however long it will take. This has happened to me on multiple occasions for I'm really having passion for, sometimes I just need some time off regardless. Some people would take this as a breaking point and kill the project, but I just say to myself "I'll come back here at some point", and it always works out. (Ofc this can lead to huge downtimes in extreme cases like myself, so if you have a better method to continously keep going I would take that; this is just advice for the case where you are literally before burning out - just put the project on halt with the plan on continuing it soon, instead of stopping with it altogether).

Also regarding time to make: Don't think about all the tasks that lie ahead. If I think about the final result, this can surely feel nice and all, but if I think about it in terms of tasks to be done until I reach the final goal, this can quickly kill my motivation. So I tend to think in small portions, what has to be done next and what can be done with the smallest possible effort. Then gradually, step by step it will evolve to the final state, without getting overwhelmed by the amount of work to be done.This is especially important when you start something big - like at the beginning of a game/code project, or when implementing a huge feature.

I agree with the comments about the hobby thing. There's not much wrong in not finishing a project. Most of my projects are unfinished.
With that said, a method for me to avoid burnout is to jump regularly between the tasks, and whenever I get stuck with something, I pick another task instead (I'm not talking about debugging, one must fix a bug if it's jamming the project). This way I can keep the flow mental state.

Another thing that helps me if the program is actually being used (by me or someone else). This is encouraging and pressing at the same time. And in this case, you are forced to set shorter term goals, because of the timelines.

No matter if it's a hobby or professional work of any scale, my method is that I usually try to make the new major feature (or set of features) work in maximum 3 days even if that means a preliminary, not entirely fool-proof or even a placeholder solution (GUI, algorithm etc) which will be fixed/enhanced/completely changed later. This will produce a pretty bad architecture and some ugly hacks (actually the code looks like a prototype, since the my method is more like prototyping), but usually either I'm the only one who will ever gonna work with the code or I refractor the code from time to time. Also with more and more projects behind me, the better architectures I will begin with and less likely I will need architecture refractoring.

I can't imagine starting a project that won't have a usable/playable 0.0 version in a week's time...

Take a big big sheet of paper, painter style, on the left write start, on the right write your goal, and try to figure out the bare minimum to go from start to your goal (will likely branch and merge back), then follow that to the letter without constantly changing technologies on a whim, you're better of sticking with something subpar than switching to something better as long as the former "will do" for your goal even if it's not optimal, you'll learn more that way and what you've learnt will help you make other décisions for future projects.

Do try to solve EVERYTHING (to maximise your own potential) but also DO STOP trying to solve something after a reasonable amount of time was spentn reasonable time depending on the task, if you're stuck on a bug, look for a couple hours, if nothing comes up, ask for help, if instead we're talking about designing a large system, give it a few days alone before asking for help.

Release something, no matter how shitty it is, as early as possible, and once that's done keep a "my project must be releasable at a no worse state at any given time", sticking to this will help you avoid technical debt pretty naturally as you won't put down bugs for later while working on the next cool feature.

Physical exercise. Energy exertion. Take a sabbatical. Revitalize. Renew your mind. Make it feel new and exciting again.

For one of the crunches I found that thinking capabilities recede to zero. If you think that its not pushing any further, but your manager thinks its possible, then rather take a 10 -30 minutes break. If its overloading in the wee-hours then just sleep when if you need. If the burnout is causing a level of unfavorable pain which is mental or physical, just pause everything.

No manager is gonna stop you from taking a power-nap during a burnout. But if you take a complete day off and other don't then its not really appreciated and can go to the unofficial records.

Passions for games is the first thing that gets destroyed if there are too many crunches and burnout.

There are so many professions in the world that use the same skill as developing games but work life balance is perfect bliss. If there were such burnouts in software engineers who are working in the automotive industry or oil industry think about the mistakes developers would make, and how many accidents would have been caused.

Fact fully ,if anything relatively deviant is to be created or innovated its has to take its right amount of time. Time like 2-8 years is what it would take to make a perfect next gen game. Think about the solar powered aircraft, or the fastest Bugatti. Same is the case with games. they take their time.