On switching to Game Maker 8.1... or Python.

[MedO]

Hi everyone. I recently played around with Game Maker 8.1 again, and found that it causes more problems than was realized before. What was already a known issue is that bitmap fonts are no longer supported. ruirize supplied a conversion to sprite font recently to work around this issue, but now attempting to host a game hangs with an infinite loop. Why? Because the md5 script depends heavily on string operations, and Game Maker 8.1 apparently changes the behavior of some functions (e.g. chr) based on the font selected for drawing. Using the new font-independend function ansi_char instead makes it not hang anymore, but now it spits out different md5 codes than before.

There is certainly a way around this, but this is just one more example of Game Maker-induced problems costing time and energy. Maybe we should just stay with GM 8.0, stop working on any big new features and concentrate on moving to PyGG2, to finally get to an (arguably) sane programming environment. That's just my thoughts on this for now, please feel free to comment and disagree.

[Orpheon]

Hi everyone. I recently played around with Game Maker 8.1 again, and found that it causes more problems than was realized before. What was already a known issue is that bitmap fonts are no longer supported. ruirize supplied a conversion to sprite font recently to work around this issue, but now attempting to host a game hangs with an infinite loop. Why? Because the md5 script depends heavily on string operations, and Game Maker 8.1 apparently changes the behavior of some functions (e.g. chr) based on the font selected for drawing. Using the new font-independend function ansi_char instead makes it not hang anymore, but now it spits out different md5 codes than before.

There is certainly a way around this, but this is just one more example of Game Maker-induced problems costing time and energy. Maybe we should just stay with GM 8.0, stop working on any big new features and concentrate on moving to PyGG2, to finally get to an (arguably) sane programming environment. That's just my thoughts on this for now, please feel free to comment and disagree.

[NAGN]

what real benefits does game maker 8.1 bring?

I'd recommend a change to pyGG2

[Lemonade]

Won't PyGG2 make modding a whole lot harder for beginners ?

[Nukleus]

The switch from game maker to python is very easy. Plus, the gg2 code is messy, so it's a lot easier once you learn python.

[Ruirize]

Won't PyGG2 make modding a whole lot harder for beginners ?

Most certainly.

The switch from game maker to python is very easy. Plus, the gg2 code is messy, so it's a lot easier once you learn python.

Not quite as easy as you'd like to think. There's a lot to GG2, and even if we have the code it's gonna take a long while to get it to the same level as GG2 is today.

what real benefits does game maker 8.1 bring?

Mostly speed. Candied GG2 uses it if you want to mess around with it.

[♫Sani Sani Disco♫]

GML is syntactically closer to C based languages and java, so if it's a port we're talking about I'd logically go with those. In the case of a remake my opinion stands.

[♫Sani Sani Disco♫]

I also like to think it'd run smoother if we just debloated gmlgg2. I know game maker is bad but relying on a port or remake is a bullshit excuse to not attempt to fix the flaws.

[NAGN]

GML is syntactically closer to C based languages and java, so if it's a port we're talking about I'd logically go with those. In the case of a remake my opinion stands.

there's no such thing as a port; the game has to be remade anyways

python is actually somewhat similar to GML in that it has loose syntax (somewhat) and its relatively accessible as its free and open source

[Orpheon]

GML is syntactically closer to C based languages and java, so if it's a port we're talking about I'd logically go with those. In the case of a remake my opinion stands.

there's no such thing as a port; the game has to be remade anyways

^

I also like to think it'd run smoother if we just debloated gmlgg2. I know game maker is bad but relying on a port or remake is a bullshit excuse to not attempt to fix the flaws.

Instead of trying to fix the ubiquitous flaws in GML-GG2, we're fixing those flaws by starting from 0 in something better (something free, and something cross-platform).

Won't PyGG2 make modding a whole lot harder for beginners ?

Most certainly.

I'm not sure it's THAT harder to mod. At least for anyone who knows Python, it isn't, and you have all of the benefits of Python over GML (better lists, better dicts, much better string stuff, control over everything, etc...).
Sure, modders who now know only GML will have to learn Python. That's true.

The switch from game maker to python is very easy. Plus, the gg2 code is messy, so it's a lot easier once you learn python.

Not quite as easy as you'd like to think. There's a lot to GG2, and even if we have the code it's gonna take a long while to get it to the same level as GG2 is today.

That's true, it's not going to be easy.
On the other hand, a good bit is already done, so yeah.

[♫Sani Sani Disco♫]

I also like to think it'd run smoother if we just debloated gmlgg2. I know game maker is bad but relying on a port or remake is a bullshit excuse to not attempt to fix the flaws.

Instead of trying to fix the ubiquitous flaws in GML-GG2, we're fixing those flaws by starting from 0 in something better (something free, and something cross-platform).

This is exactly the kind of stuff I'm talking about.

[Orpheon]

I also like to think it'd run smoother if we just debloated gmlgg2. I know game maker is bad but relying on a port or remake is a bullshit excuse to not attempt to fix the flaws.

Instead of trying to fix the ubiquitous flaws in GML-GG2, we're fixing those flaws by starting from 0 in something better (something free, and something cross-platform).

This is exactly the kind of stuff I'm talking about.

We could devote our attention to fixing some of the flaws (those that are possible to fix in GML) in GML. Or we could try to fix them all in something which can fix them all (or almost all).

[MedO]

We have been fixing the flaws for years. The thing is, apart from the Gml related problems, GG2 simply has a bad codebase. It was originally written by a total GM newbie in hacky, throwaway code since obviously nobody would care about it anymore two weeks after the demakes competition was over, and there was a deadline to meet. Only that people embarrassingly did continue to care, and so the code was extended, by developers of somewhat varying skill. As a result we have a huge mess. Some parts of it get cleaned up now and then, but whenever new stuff is added it has to exist in and interact with the remaining mess, so it becomes part of that mess and reinforces the existing bad structures.

Just as an example, take a look at how game modes are implemented. It's horrible imo, but it works, so and the next game mode will be created the same way, probably by the time-honored method of copy, paste, modify. This cements the existing mess, because now if you ever want to implement a new system for game modes, you have to consider and rework an additional one.

Imo, the only way to actually fix this is to start from scratch with a solid foundation. Whether it is worth fixing depends on whether the project will be worked on much in the future, but conversely the future of the project might also depend on whether this is fixed - having a solid codebase could make development much less painful and more fun.

[AlexTheFirst]

PyGG2 will make modding a whole lot harder for beginners

[Orpheon]

PyGG2 will make modding a whole lot harder for beginners

I still don't really see why. It's not like Python is immensely harder than GML, it's a lot cleaner and a lot easier to work with in most cases (just compare GML lists with Python lists).
The only things that are really more complex atm in PyGG2 than in GML-GG2 are the networking because of UDP (but it's still cleaner), and the engine because of the interpolation.

Also, remember that GMK is neither free nor as well documented as Python. Both count lots towards new modders.

It'll force all the current modders to learn Python. Yes. But I don't think it'll be that bad towards new ones.