Wiki: issues + quick questions/answers
Moderator: Moderators
Re: Wiki: issues + quick questions/answers
http://sourceforge.net/p/spring1944/cod ... icense.txt seems to be correct link to file.
Well if you want then not only fix the link, but also fix the link-text to "license=various" or something. Not sure what else you mean or is to be done.
Well if you want then not only fix the link, but also fix the link-text to "license=various" or something. Not sure what else you mean or is to be done.
Re: Wiki: issues + quick questions/answers
Agreed. Clearer licensing is always a plus, but I'm not sure you can describe games art/code under just one license each. There's usually a number of them involved. Anyhow, see what you can do fix it.hokomoko wrote:https://springrts.com/wiki/Games
I think there should be a clear template of licensing for each game, so it's obvious how the code is licensed and how the content is.
One can be easily confused into thinking S44 isn't FOSS.
This ^ . Manual downloads + installers is not an issue (those games work fine). Please don't bring politics of what "should be" to this.Silentwings wrote:Some players do manually download files (of their own choice), and obviously some games do offer/require their own installers.
Agree with regards to ordering on https://springrts.com/wiki/GamesDownloads . Feel free to change the order but don't delete "one-game-installers".8611z wrote:Imo the universal lobbies (springlobby, weblobby) should be listed above the one-game-installers etc
Re: Wiki: issues + quick questions/answers
The issue is larger than switching the order of lobbies and game-installers on https://springrts.com/wiki/GamesDownloads
To make sense at very least https://springrts.com/wiki/Download would need changing too, from point players towards "Universal Lobbies" and not "Games Downloads"
All the texts and guides on downloads/playing are in need of remake and many more pages need slight rewording. For example on maps, it begins with instructions from what site to download and into which directory to copy the files.
Those texts about moving files around are worth a footnote at most, every lobby has automatic mapdownload.
But if this is too touchy politics then players are stuck with instrutions thought up in 2006.
To make sense at very least https://springrts.com/wiki/Download would need changing too, from point players towards "Universal Lobbies" and not "Games Downloads"
All the texts and guides on downloads/playing are in need of remake and many more pages need slight rewording. For example on maps, it begins with instructions from what site to download and into which directory to copy the files.
Those texts about moving files around are worth a footnote at most, every lobby has automatic mapdownload.
But if this is too touchy politics then players are stuck with instrutions thought up in 2006.
Re: Wiki: issues + quick questions/answers
No, https://github.com/spring1944/spring194 ... icense.txt is the correct link, (notice I already updated it in the wiki)8611z wrote:http://sourceforge.net/p/spring1944/cod ... icense.txt seems to be correct link to file.
Well if you want then not only fix the link, but also fix the link-text to "license=various" or something. Not sure what else you mean or is to be done.
Re: Wiki: issues + quick questions/answers
Started the 101 wiki cleanup.
A few questions (more to come as I go along):
1) How to add new templates? I created a tempalte for https://springrts.com/wiki/Template:Removed but it's not automagically being added here: https://springrts.com/wiki/Modrules.lua#los (even though I added the "removed" tag) Nvm, figured it out.
2) How to mark changes in default values? Example: ! change default airMipLevel 2->1 & radarMipLevel 3->2
A few questions (more to come as I go along):
2) How to mark changes in default values? Example: ! change default airMipLevel 2->1 & radarMipLevel 3->2
- Silentwings
- Posts: 3720
- Joined: 25 Oct 2008, 00:23
Re: Wiki: issues + quick questions/answers
I suggest only marking changes in the wiki for the rare cases where is a good reason that someone reading carefully through the changelog wouildn't spot that a change had occured.
Currently, its random whether previous behaviour is mentioned in wiki or not, and there are simply too many changes to record all and maintain removing them after some time lag, etc etc. The assumption has to be that active game devs should use the latest engine and read the changelog, afaics we don't have manpower (or reason) to do otherwise.
Currently, its random whether previous behaviour is mentioned in wiki or not, and there are simply too many changes to record all and maintain removing them after some time lag, etc etc. The assumption has to be that active game devs should use the latest engine and read the changelog, afaics we don't have manpower (or reason) to do otherwise.
Re: Wiki: issues + quick questions/answers
Wiki should reflect current engine, previous versions can suck it.
Not so unfeasible, I have updated wiki with every relevant changelog entry since 90.0
Not so unfeasible, I have updated wiki with every relevant changelog entry since 90.0
structure of wiki
Roger :)
I'm thinking that eventually we could have multiple wiki versions for each old engine version (this is something a number of projects I use do), and basically only allow editing for the stable/dev version. But for now I'll just update the new stuff.
I'm thinking that eventually we could have multiple wiki versions for each old engine version (this is something a number of projects I use do), and basically only allow editing for the stable/dev version. But for now I'll just update the new stuff.
Re: Wiki: issues + quick questions/answers
https://springrts.com/wiki/FAQ:technica ... u_cores.3F
The FAQ question "Why doesn't the spring engine use all my cpu cores?" needs a better answer. Currently it is awkwardly written and does not answer the question in way that players can understand. That is why the question keeps coming up. I do not understand what the text wants to say either. It seems to be more about making the question look stupid instead of answering it.
("From now on we'd recommend refraining from saying: 'Spring should be more multithreaded'" - except the question never asked that? '..make you look less foolish..' etc) I think the question not as foolish as the FAQ tries to make it appear and needs better answer.
The FAQ question "Why doesn't the spring engine use all my cpu cores?" needs a better answer. Currently it is awkwardly written and does not answer the question in way that players can understand. That is why the question keeps coming up. I do not understand what the text wants to say either. It seems to be more about making the question look stupid instead of answering it.
("From now on we'd recommend refraining from saying: 'Spring should be more multithreaded'" - except the question never asked that? '..make you look less foolish..' etc) I think the question not as foolish as the FAQ tries to make it appear and needs better answer.
Re: Wiki: issues + quick questions/answers
what do you suggest to phrase it better in few words? explaining players technical stuff is difficult, imo the answer can't be improved much.8611z wrote:The FAQ question "Why doesn't the spring engine use all my cpu cores?" needs a better answer. Currently it is awkwardly written and does not answer the question in way that players can understand. That is why the question keeps coming up. I do not understand what the text wants to say either. It seems to be more about making the question look stupid instead of answering it.
Re: Wiki: issues + quick questions/answers
Why doesn't the spring engine use all my cpu cores?
Spring could use all your cores, and development-Versions of spring exist that explicitly use all cores available. However, measuring the performance on those cores, yielded in the somewhat counter intuitive result - that spring is slower with multiple cores.
Modern computers are very fast, when it comes to the the cpu, but compared to those cores- your memory is very slow (1/300) of the speed of your core, and it is this memory that generates the biggest performance bottlenecks.
As tough as it is to imagine- your cpu has periods during a full fledged spring game- where it has near nothing to do - but wait for memory to load- and/or the graficcard to receive send data and complete draw operations.
In fact, as you can see in this graph of total time spend during a frame, the time spend in the cpu (busy) is rather small.
<img></img>
So you claim your engine is perfect?
This does however not mean, that our engine is perfect. If you want to optimize things, feel free to step up and put your work into the basket. Be warned though, its complicated, really really complicated.
Companys like UbiSoft, EA and Activision get it done without problems, why cant you?
Because those companys have not limited ressources, require you to have the best hardware to play there games, and work excessively with some approaches, spring does not yet fully support.
If you want spring to be more efficient regarding the usage of multi-threading, moving more of the work to the gpu - with OpenCL would yield much bigger performance improvements. Many - non-gameplay related tasks (like particle physics, physics, rendering etc.) can thus be moved from the cpue onto the GPU and this one can execute those instructions blazingly fast. Spring is not there yet, but will be one of these days.
We understand that computer science seems like arcane magic sometimes, and part of this magic show by others, is creating the illusion that infinite performance improvements are possible. But this thing you are using - its a finite machine. Open a calculator, and multiply 1 * 0.9 20 times. Now multiply it back- value * (1/0.9). You wont get back the 1 you started from. Why? Because, this is a machine. A fast machine. A technical marvell. But it has limitations, who are tricky to circumvent.
But open https://github.com/spring/spring/blob/0 ... eading.cpp
and build your own better version, and let it race against our version in the performance benchmarks.
We would be surprised if you managed it to squeeze more then 1 % improvement from this stone. Suprised in a good way. Welcome to Spring.
Spring could use all your cores, and development-Versions of spring exist that explicitly use all cores available. However, measuring the performance on those cores, yielded in the somewhat counter intuitive result - that spring is slower with multiple cores.
Modern computers are very fast, when it comes to the the cpu, but compared to those cores- your memory is very slow (1/300) of the speed of your core, and it is this memory that generates the biggest performance bottlenecks.
As tough as it is to imagine- your cpu has periods during a full fledged spring game- where it has near nothing to do - but wait for memory to load- and/or the graficcard to receive send data and complete draw operations.
In fact, as you can see in this graph of total time spend during a frame, the time spend in the cpu (busy) is rather small.
<img></img>
So you claim your engine is perfect?
This does however not mean, that our engine is perfect. If you want to optimize things, feel free to step up and put your work into the basket. Be warned though, its complicated, really really complicated.
Companys like UbiSoft, EA and Activision get it done without problems, why cant you?
Because those companys have not limited ressources, require you to have the best hardware to play there games, and work excessively with some approaches, spring does not yet fully support.
If you want spring to be more efficient regarding the usage of multi-threading, moving more of the work to the gpu - with OpenCL would yield much bigger performance improvements. Many - non-gameplay related tasks (like particle physics, physics, rendering etc.) can thus be moved from the cpue onto the GPU and this one can execute those instructions blazingly fast. Spring is not there yet, but will be one of these days.
We understand that computer science seems like arcane magic sometimes, and part of this magic show by others, is creating the illusion that infinite performance improvements are possible. But this thing you are using - its a finite machine. Open a calculator, and multiply 1 * 0.9 20 times. Now multiply it back- value * (1/0.9). You wont get back the 1 you started from. Why? Because, this is a machine. A fast machine. A technical marvell. But it has limitations, who are tricky to circumvent.
But open https://github.com/spring/spring/blob/0 ... eading.cpp
and build your own better version, and let it race against our version in the performance benchmarks.
We would be surprised if you managed it to squeeze more then 1 % improvement from this stone. Suprised in a good way. Welcome to Spring.
Re: Wiki: issues + quick questions/answers
Wiki icon ala my avatar in 135px
- Attachments
-
- SpringWiki1a.png (17.22 KiB) Viewed 8971 times
Re: Wiki: issues + quick questions/answers
thanks, done :)
Re: Wiki: issues + quick questions/answers
Hey hey wikki hey wikki hey.
Re: Wiki: issues + quick questions/answers
Nuke https://springrts.com/wiki/Maps:Compiling (outdated - last update in 2012), in favor of https://springrts.com/wiki/Mapdev:Tutorial_Finalizing (less outdated)?
Re: Wiki: issues + quick questions/answers
I wrote that page a few years ago, not sure if its still accurate.gajop wrote:Nuke https://springrts.com/wiki/Maps:Compiling (outdated - last update in 2012), in favor of https://springrts.com/wiki/Mapdev:Tutorial_Finalizing (less outdated)?
Re: Wiki: issues + quick questions/answers
I found the page when I googled "springrts map compile"
@enetheru: good to see you're still around :)
@enetheru: good to see you're still around :)
Re: Wiki: issues + quick questions/answers
Does anyone know what's up with the wiki formatting? It seems to have broken most code tags.
Do we need to manually edit those tags (e.g. remove the "pre" part) or is there a better solution?
Also, stickied this.
Do we need to manually edit those tags (e.g. remove the "pre" part) or is there a better solution?
Also, stickied this.
Re: Wiki: issues + quick questions/answers
i've noticed that, too. could be the last wiki update maybe?
https://github.com/spring/spring-websit ... e910801666
https://github.com/spring/spring-websit ... e910801666