Live data from Hacker News

The Battle for Wesnoth needs help

wesnoth.org

141–150 of 155 posts

Re: The Battle for Wesnoth needs help

#141

Earlier quoted context omitted.

Wait, how on earth do you backport blackbox improvements?

So, mainly thinking of usability improvements. See Fallout New Vegas, which had iron sights. Sure you wouldn't have the code (in a universe where Fallout 3 was open source), but you would know how it works by treating it as a black box.

Right - gotcha.

Re: The Battle for Wesnoth needs help

#142
post #90

Earlier quoted context omitted.

Put this another way: lots of people don't have a second thought about telling this /extremely successful/ project how it really should have been run.

It took me a while to realise, but HN renders text in between stars as italics. So * italics * Becomes italics Indented text is rendered in verbatim monospace

The text processing of Hacker News is described at https://news.ycombinator.com/formatdoc.

HN only supports the two formatting types you described, plus auto-linking of URLs written out in full, such as the one in the previous paragraph.

Re: The Battle for Wesnoth needs help

#143

Earlier quoted context omitted.

> Rewriting any game in javascript is imho a great idea. How do you deal with the fact that the only native numeric type is a 64bit float? Space Engineers currently uses floats as the basis for their world coordinate system. Things get really bumpy when you get far away from [0,0,0].

32/64 bit float is the native float type on the cpu you are running on.. hence you can't really claim "native" support for larger types.. if you mean native in terms of being able to write a+b for a numeric type, then that eliminates all languages that don't have operator overloading..

I mean "native" in terms of what's supported by the language itself, out of the box.

C and C++ offer -at a minimum- the following numeric types:

* signed and unsigned integers in a variety of widths

* 32 and 64-bit floating point numbers

JavaScript offers:

* 64-bit floating point numbers

When performing integer math, JavaScript's numeric type is only safe for arithmetic operations between -((2^53)-1) and +((2^53)-1).

Now that we have that out of the way, my original question remains unanswered. :)

Re: The Battle for Wesnoth needs help

#144

Earlier quoted context omitted.

32/64 bit float is the native float type on the cpu you are running on.. hence you can't really claim "native" support for larger types.. if you mean native in terms of being able to write a+b for a numeric type, then that eliminates all languages that don't have operator overloading..

I mean "native" in terms of what's supported by the language itself, out of the box. C and C++ offer -at a minimum- the following numeric types: * signed and unsigned integers in a variety of widths * 32 and 64-bit floating point numbers JavaScript offers: * 64-bit floating point numbers When performing integer math, JavaScript's numeric type is only safe for arithmetic operations between -((2^53)-1) and +((2^53)-1).…

In general 54-bit signed ints are enough for games. Floats are more common for math.

Re: The Battle for Wesnoth needs help

#145

Earlier quoted context omitted.

I mean "native" in terms of what's supported by the language itself, out of the box. C and C++ offer -at a minimum- the following numeric types: * signed and unsigned integers in a variety of widths * 32 and 64-bit floating point numbers JavaScript offers: * 64-bit floating point numbers When performing integer math, JavaScript's numeric type is only safe for arithmetic operations between -((2^53)-1) and +((2^53)-1).…

In general 54-bit signed ints are enough for games. Floats are more common for math.

Cool.

Many contemporary games are multi-threaded out of necessity. How would you deal with JavaScript's single-threaded, single-process model of execution? Experimental features like Web Workers?

Re: The Battle for Wesnoth needs help

#146
post #130

Earlier quoted context omitted.

I'm confused. Why did you post this?

Not familiar with the game, was delighted to see that it's in the repository.

Ah. Your comment would have been better received if you'd said this, rather than posting output from aptitude.

Re: The Battle for Wesnoth needs help

#147

Earlier quoted context omitted.

In general 54-bit signed ints are enough for games. Floats are more common for math.

Cool. Many contemporary games are multi-threaded out of necessity. How would you deal with JavaScript's single-threaded, single-process model of execution? Experimental features like Web Workers?

No disagreement on that one! If your application must be multi-threaded, it does make sense to write it in a language with proper multi-threading support.

Re: The Battle for Wesnoth needs help

#148

Earlier quoted context omitted.

I think it's undue ego. If a developer gives you code but imposes conditions that the code must not be modified because this damages their artistic integrity, (and I have seen people make this claim), most people would consider that ridiculous, because code is functional and usable. In the context of a videogame, artwork is no less functional and usable. Your ego is not more important than the people who are using yo…

That's, in the end, their prerogative. They can choose to give, but under a no-derivative licence, if they so wish.

They can have all the prerogatives they want as long as they understand that their prerogatives have no place in a free game. Their artwork may be scaled, resized, and cropped in order to fit into a playing field. The colours may be changed in order to indicate teams. Additional animations may be required for certain actions. The style might need tweaks to fit the existing style of the game.

Unmodifiable artwork poses many practical problems for a free game. It's simply unacceptable.

Re: The Battle for Wesnoth needs help

#149
post #137

Earlier quoted context omitted.

The problem there, though, is not the maintenance, it's the actual reimplementation of the thing. At this point it's probably man-years of work to do that sort of a port to a new engine, and getting that sort of commitment doesn't strike me as easier than finding one or two folks to help out.

It'd probably be far easier. Plenty of people would be willing to learn UE4 or already know it. Who wants to learn Wesnoth Markup Language?

Perhaps the cold reality is Wesnoth 2.0 will need to take radical steps away and use some rather modern engine (Unity, MonoGame, etc.) But keep in mind, right now we’re looking at a fairly successful, high production-value FOSS project. it needs short term answers as well as long term ones.

Re: The Battle for Wesnoth needs help

#150

Earlier quoted context omitted.

Orders of magnitude more code written in it than Java? Hohoho good one. HTH: http://www.tiobe.com/index.php/content/paperinfo/tpci/index.... http://githut.info

Java may have been a stretch.. but think about how many web apps have been written.. and consider that a large %age of user time is spent in the browser who's language IS javascript.. Ive removed java from my systems since oracle ganked it. Java is used in lots of business/banking software, true.. but even the tiobe charts u linked show it declining in popularity.. contrasting w js being language of the year 2014. As…

It wasn't a "stretch." It was flat out wrong.
Post reply on HN