Teach Yourself Demoscene in 14 Days (opens in new tab)

In this context, the deadline section is referring to being ready for a demo party, possibly as a compo entry, which happens on a specific date, so what would even be an alternative to having a deadline? The writing wasn’t urging the reader to choose whether to use deadlines, it was saying pay attention to the fact that your demo showdown is coming up on a day picked by someone else, so plan accordingly. Right?

Personally I find them paralyzing - so I'll just progress at my own speed - nobody is forcing us to use the timings proposed in the repo.

If I want to tinker or aimlessly play with something,sure,artificial deadlines may not suit me.

But if I want to even remotely accomplish an actual goal,or build something,or even learn something concrete,I positively need a firm and ideally external and enforced deadline - otherwise I'll meander,change scope and direction,redo from scratch,try a different tool / framework / system / method /whatever...and generally never ever finish

Time to deliver is just another constraint, which tends to force all decisions be made quickly and to err on the side of forward progress vs. some theoretically ideal option.

Having participated in a few demo parties back in the 90s, I have a deep appreciation for the scene. But I think we tended to dismiss the negative consequences of sleep deprivation, considering what we know now about the brain and sleep.

Edit:

Also, in terms of deadlines in general, not just the insane 24/48hr demoscene/gamejam style competitions, it can be rather difficult to prioritize a project receives your precious bandwidth on a given day if there's no deadline. It just keeps getting pushed aside by the tasks that do.

For actual deadlines often features get cut to meet them - and said features may return in future versions.

That’s all good and well, but my initial thought is simple. Just remember the title. The author specifically designed this course to teach Demoscene ‘in 14 days’.

If you don’t mind taking longer, then the deadlines don’t apply. But the course is designed to help you do it in this timeframe.

Maybe a bit like schoolwork, what you get out of it is what you're able to put into it.

I've been running a node-ytdl-core server alongside it to stream audio from youtube videos, which is my primary issue right now: making something interesting for others seems like it needs to be able to play the audio they want, but most people don't have folders full of mp3's anymore, and both youtube/soundcloud/others and the modern cross-site web security model make it incredibly hard to do something as simple as decode the audio signal the browser is playing.

Regardless, your project looks super cool! Using online radio seems like a neat way to workaround the problem I've been having.

Learned a lot of really hard math and parallel GPU concepts that I would not have been able to without the ability to ask experts questions.

https://developer.nvidia.com/gpugems/gpugems/contributors

The Cg book although outdated provides good information about shaders,

https://developer.download.nvidia.com/CgTutorial/cg_tutorial...