Discuss Scratch
- Discussion Forums
- » Advanced Topics
- » Is scratch 3.0 unsuitable for creating advanced/complex projects?
- badatprogrammingibe
-
Scratcher
500+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
I have seen a variety of views both for and against scratch 3.0 being useful for advanced and complex projects.
I would love to hear about all of your thoughts on this topic.
If anyone has examples of advanced/complex projects created in Scratch 3.0, please post them here.
I would love to hear about all of your thoughts on this topic.
If anyone has examples of advanced/complex projects created in Scratch 3.0, please post them here.
- infinitytec
-
Scratcher
1000+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
I think it depends on what you are doing. Some things obviously are not fit for it.
It is well known that the memory management has issues. When this is fixed it will be improved a lot.
The other thing that will be important to complex projects is third-party extension support. When this gets added there is a possibility for some really advanced functions to be added.
There are obvious weaknesses compared to 1.4 and 2, but I think it still has a lot of potential. It does of course have a learning curve, but I think as time goes on it will improve.
It is well known that the memory management has issues. When this is fixed it will be improved a lot.
The other thing that will be important to complex projects is third-party extension support. When this gets added there is a possibility for some really advanced functions to be added.
There are obvious weaknesses compared to 1.4 and 2, but I think it still has a lot of potential. It does of course have a learning curve, but I think as time goes on it will improve.
- badatprogrammingibe
-
Scratcher
500+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
I think it depends on what you are doing. Some things obviously are not fit for it.What do you think could be added that would help advanced/complex scratch projects with third-party extension support?
It is well known that the memory management has issues. When this is fixed it will be improved a lot.
The other thing that will be important to complex projects is third-party extension support. When this gets added there is a possibility for some really advanced functions to be added.
There are obvious weaknesses compared to 1.4 and 2, but I think it still has a lot of potential. It does of course have a learning curve, but I think as time goes on it will improve.
- bybb
-
Scratcher
1000+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
I have seen a variety of views both for and against scratch 3.0 being useful for advanced and complex projects.I've been trying to implement a really bad attempt at OOP in Scratch using lists to store the class data and the class instances, the goal wasn't speed, but rather to make it easier to port an OOP project. In Scratch 2, accidentally calling a atomic define block that got in to an infinite loop wasn't that bad, Scratch froze (not the browser though), Flash crashed, I reloaded and continued.
I would love to hear about all of your thoughts on this topic.
If anyone has examples of advanced/complex projects created in Scratch 3.0, please post them here.
However in Scratch 3, the same mistake pins my CPU at 99% usage, freezes my entire browser and all other Windows (something multiprocess browsers are supposed to stop so I haven't a clue how that's happening) and takes a good few minutes of not being able to do anything (close the tab included) before my browser decides to let me have the option to end the script.
There should be at least some protection to stop me from crashing my entire browser tbh.
Last edited by bybb (April 12, 2019 00:19:35)
- infinitytec
-
Scratcher
1000+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
Do you mean Scratch 2?I have seen a variety of views both for and against scratch 3.0 being useful for advanced and complex projects.I've been trying to implement a really bad attempt at OOP in Scratch using lists to store the class data and the class instances, the goal wasn't speed, but rather to make it easier to port an OOP project. In Scratch 3, accidentally calling a atomic define block that got in to an infinite loop wasn't that bad,
I would love to hear about all of your thoughts on this topic.
If anyone has examples of advanced/complex projects created in Scratch 3.0, please post them here.
- bybb
-
Scratcher
1000+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
Yes I did, thanks for spotting that mistake.Do you mean Scratch 2?I have seen a variety of views both for and against scratch 3.0 being useful for advanced and complex projects.I've been trying to implement a really bad attempt at OOP in Scratch using lists to store the class data and the class instances, the goal wasn't speed, but rather to make it easier to port an OOP project. In Scratch 3, accidentally calling a atomic define block that got in to an infinite loop wasn't that bad,
I would love to hear about all of your thoughts on this topic.
If anyone has examples of advanced/complex projects created in Scratch 3.0, please post them here.
- -Rex-
-
Scratcher
500+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
The list length limit of 200,000 (and the removal of dynamic list creation and access using hacked blocks) makes pseudo-OOP systems slower and restricted to a small amount of “memory” (list entries).
- _nix
-
Scratcher
1000+ posts
Is scratch 3.0 unsuitable for creating advanced/complex projects?
The only technical reason for Scratch 3.0 supporting less complex projects is the 200,000 list limit mentioned elsewhere here. Otherwise, it's a pretty fair battle between 2.0 and 3.0.
Personally, I find Scratch 3.0 much, much, much better for developing complex projects - the improved user interface (zooming, scrolling, etc all being more natural and powerful than in 2.0), the sorted dropdown menus and custom block list, and the new blocks like “item # of” and “costume name” are all ridiculously useful improvements, among many other changes in 3.0. There are definitely a few things 3.0 lacks presently, like the abilities to jump to a custom block's definition and to rename broadcasts, but everything else outweighs those considerably, and the features that aren't here yet will be coming eventually.
Here are a couple of my projects which are considerably complex and made entirely in 3.0 (plus make use of 3.0's new features): HK Scenery, which shows off a particularly complex camera system based on adapting to where you are and what you're doing, and its more recent version, which dynamically creates and deletes clones to avoid the 300 clone limit (see its notes and credits section); and Deltarune Battle Engine, which is a full-on recreation of the battle engine in Deltarune, very closely matching its visual feel, and is implemented in a modular view, making it straightforward to add new enemies and features and the like.
It's definitely worth noting that Scratch 3.0 doesn't run well on lower-end devices, though – many of which didn't struggle nearly as much with Scratch 2.0. But on devices which run 3.0 well, it's capable (sans the list limit) of as much as 2.0 and has several features that make development a lot easier.
(Seriously, dropdowns and custom blocks being sorted alphabetically are, at least for me, a world of difference in programming with Scratch…!)
Personally, I find Scratch 3.0 much, much, much better for developing complex projects - the improved user interface (zooming, scrolling, etc all being more natural and powerful than in 2.0), the sorted dropdown menus and custom block list, and the new blocks like “item # of” and “costume name” are all ridiculously useful improvements, among many other changes in 3.0. There are definitely a few things 3.0 lacks presently, like the abilities to jump to a custom block's definition and to rename broadcasts, but everything else outweighs those considerably, and the features that aren't here yet will be coming eventually.
Here are a couple of my projects which are considerably complex and made entirely in 3.0 (plus make use of 3.0's new features): HK Scenery, which shows off a particularly complex camera system based on adapting to where you are and what you're doing, and its more recent version, which dynamically creates and deletes clones to avoid the 300 clone limit (see its notes and credits section); and Deltarune Battle Engine, which is a full-on recreation of the battle engine in Deltarune, very closely matching its visual feel, and is implemented in a modular view, making it straightforward to add new enemies and features and the like.
It's definitely worth noting that Scratch 3.0 doesn't run well on lower-end devices, though – many of which didn't struggle nearly as much with Scratch 2.0. But on devices which run 3.0 well, it's capable (sans the list limit) of as much as 2.0 and has several features that make development a lot easier.
(Seriously, dropdowns and custom blocks being sorted alphabetically are, at least for me, a world of difference in programming with Scratch…!)
Last edited by _nix (April 12, 2019 21:22:14)
- Discussion Forums
- » Advanced Topics
-
» Is scratch 3.0 unsuitable for creating advanced/complex projects?