Before I start, I don't want to discourage any of you from creating a big game, if you have the know-how, I'd suggest porting your progress over to a real game engine.

Scratch is fun to use, no doubt about it, but it's not a game making software in any way, while yes you can make games within it, it's not designed to do that, it's specifically designed to be a teaching tool. one of scratch's biggest faults at creating games with is RAM, I'll give you an example, once I had a project that was about 9MB big, should be no problem right? Well, when loading the project up, it takes up nearly 3/4 of a gigabyte of RAM, and when going around it and loading assets in, it almost reaches a full megabyte. The second problem with trying to make a large scale game in scratch is performance, Scratch (Specifically Ver. 3.0) was built using things (can't think of the word) designed for web applications, and web applications aren't usually that complex, when making games on scratch on the other hand, you're going to have lots of complex scripts doing all sorts of things, sure there are Scratch mods that optimize scratch, but there's only so much optimization you can do at the end of the day.

That's probably enough rambling for now, before you ask I won't help you try to port over you existing games over to a game engine, I only really wanted to talk about the problems when creating a big game on scratch , maybe it is possible to do so, but it's probably only something a professional could do.

say [Have a good day and remember to scratch on! :)]

but creating large projects is fun. it feels like getting doom to run on a toothbrush in a way, in that you're creating and running something spectacular with elementary tools. and the community is fantastic.

everyone knows

Da-thing wrote:

The second problem with trying to make a large scale game in scratch is performance, Scratch (Specifically Ver. 3.0) was built using things (can't think of the word) designed for web applications, and web applications aren't usually that complex

Really?

Web tech these days is actually quite fast (browsers have been compiling JavaScript for years- it's not interpreted anymore) And for an example of something complex created using web tech: Google Earth is a good example.

JavaScript hasn't just been used for logins and a few if checks for years. It's quite fast (and gets faster all the time), has a multitude of useful data structures, and when combined with Web Assembly, it can be used with native applications.

And that's just the tip of the iceberg. The possibilities expand greatly with browser extensions and PWAs, and Electron, which can be used to run web tech natively.

The main reason Scratch is so slow is because it's interpreted, and it's not particularly efficient at it. In addition, the entirety of one project is run on one core of your CPU, so it doesn't even try to take advantage of modern processors' multiple core structure, which makes it run slower. For example, on @ChromeCat_Test's triangle filler, I get an advantage of 27,000 triangles per second, using one core of my iPad Air (5th generation), which has an Apple M1 chip. I did the math, and if all of it's processing cores were used, I could get something closer to 200,000 tps.

Da-thing wrote:

Before I start, I don't want to discourage any of you from creating a big game, if you have the know-how, I'd suggest porting your progress over to a real game engine.

Scratch is fun to use, no doubt about it, but it's not a game making software in any way, while yes you can make games within it, it's not designed to do that, it's specifically designed to be a teaching tool. one of scratch's biggest faults at creating games with is RAM, I'll give you an example, once I had a project that was about 9MB big, should be no problem right? Well, when loading the project up, it takes up nearly 3/4 of a gigabyte of RAM, and when going around it and loading assets in, it almost reaches a full megabyte. The second problem with trying to make a large scale game in scratch is performance, Scratch (Specifically Ver. 3.0) was built using things (can't think of the word) designed for web applications, and web applications aren't usually that complex, when making games on scratch on the other hand, you're going to have lots of complex scripts doing all sorts of things, sure there are Scratch mods that optimize scratch, but there's only so much optimization you can do at the end of the day.

That's probably enough rambling for now, before you ask I won't help you try to port over you existing games over to a game engine, I only really wanted to talk about the problems when creating a big game on scratch , maybe it is possible to do so, but it's probably only something a professional could do.

say [Have a good day and remember to scratch on! :)]

Sorry if this counts as necroposting but I actually consider doing this with one of my series (generic the cube). I highly recommend doing this if you want your game to be good and reach a new audience. Indie Games number one!

#5

I mean until any of the above reasons become obsolete, I don't think it counts as necroposting