What do you think about openra?

[Adolf Krause]

Hello everybody.

I would like to hear your opinion about the openra. Whether you like openra?

 

[Blade]

Are we taking about OpenRA the engine or the games built on top of it (of which there are several)? The focus of these forums is more toward the original games and most people here tend to prefer those over the games built to look like them on the OpenRA game engine. Emotions tend to run high when the OpenRA mods are presented as "replacements" or "improvements" over the Westwood games or exactly how the OpenRA mods relate to those games, but in general I don't think many people here have anything against the OpenRA project or the people who develop it.

[fir3w0rx]

Whether you like openra?

I like it a lot. I'd like it even more if they decide to continue working on and eventually releasing the RA2 & TS mods.

[Allen262]

People here seem split.

 

To me its fine. ALso with Blade in the "replacements" or "improvements" department. Also with fir3w0rx. I think they should be putting most of there time into TS/RA2/YR as they seem to more troublesome to get and keep running on newer systems.

[pchote]

I'd like it even more if they decide to continue working on and eventually releasing the RA2 & TS mods.

 

Work never completely stopped.  Progress is slow because it is a massive job with very few people working towards it.  If you want to see it happen faster, then come and help!

 

Despite that, a quick skim of our changelog and pull queue shows a bunch of changes related to the 'gen2' mods since the last release: #11058, #11091, #11120, #11124, #11211, #11216, #11291, #11299, #11356, #11375, #11380, #11426, #11445, #11483, #11498, #11522, #11558, #11565, #11441, #11484, #11501, #11536, #11572.

 

 

[Iran]

I watched some videos of the TS and RA2 mods for OpenRA and they look really cool. The C# code for OpenRA is also really nice.

[SiRaLeX]

Ugh, not very keen of it. Could have been really nice if they had made it in C++ and way more successful also.

 

It's a shame they went down the .NET route. But hey, anyone can "write" code in .NET in Visual Studio! So at least they got it to work, props...

 

 

 

[Watermelon]

Positively. Much work has been done.

[Tore]

I'm not keen on the gameplay of the TD mod or the general feeling of the game, but what they have done is impressive.

[Reynor]

Could have been really nice if they had made it in C++ and way more

I agree with you. C ++ would be a better solution.

[pchote]

I guess we should take it as a compliment if the worst thing you comment on is the use of C# over C++.  That has an insignificant impact on OpenRA as a game, and some significant advantages as a modding and development platform.

[Rampastring]

The engine is impressive and has massive modding potential. If I started a new C&C mod project today, OpenRA would be the engine that I'd use. TS and YR simply aren't even nearly as flexible, even with CnCNet's additions and Ares.

 

The name ("OpenRA") is slightly confusing (clearly meaning "Open Red Alert", while OpenRA is a general RTS platform), and I don't really enjoy the stock mods.

 

The code being C# is a good thing, like pchote already said.

[SiRaLeX]

I guess it should be renamed OpenRTS, then.

 

I guess we should take it as a compliment if the worst thing you comment on is the use of C# over C++.  That has an insignificant impact on OpenRA as a game, and some significant advantages as a modding and development platform.

 

Which advantages would that be? That it's more noob friendly? I didn't mean C++/CLI.

 

OpenRA is like 90 % declarations only, which, in itself is not a bad thing. I actually salute you, wherever you are on a daily basis for that, baby please.

 

But what the hell is a CPos, can't you just call it a CellPos or a Point?

 

 

[pchote]

Which advantages would that be? That it's more noob friendly? I didn't mean C++/CLI.

 

No, "noob friendly" is quite far down the list.

 

A few of the more interesting advantages are:

• Reflection: Code is just another type of data, and this is the heart of the OpenRA engine.  Mods can write their own actor traits, file format parsers, UI Widgets, Lua API actions, and a few other object types in C#.  The engine inspects itself and the mod dlls at runtime and makes these objects available for use by the mod yaml.  It then parses the yaml and constructs the objects that represent actors, the game UI, the Lua<->C# bridge, and other things... all without the piles of boilerplate that would be needed in other languages to register factory objects, define external type metadata, or parse data files.  This also makes it very easy to write static analysers and lint checkers that verify correctness in the core code and mod rules.
  
• Dynamic code generation: The engine automatically generates some performance-sensitive code for error and cheat-detection at runtime.  This again saves a bunch of unnecessary boilerplate and gives significant performance improvements over manually enumerating the relevant data.
  
• Cross-platform binary compatibility:  We compile the code once on Ubuntu, and then ship the same binaries (wrapped with platform-specific installation fluff) on Windows, OSX, and Linux.  We don't need to jump between a bunch of different OSes to build our installers or mess with cross-compilers.  If you were writing an OpenRA mod with custom code, would you really want to have to compile it for 10 different operating systems?  Or just the one that you use for development?
   
  
• Unsafe code operations (but only when we need them):  The majority of mod-level code gets to take advantage of the flexibility and friendlyness of the high-level "safe" parts of C#.  When we need extra performance or to interop with native code it is simple to drop into an unsafe{} context and hack around with the raw memory and pointers as much as we please.
  

 

[TaxOwlbear]

I like it, but there's no strong incentive for me to play OpenRA over the originals (originals with in high resolution, that is). That may change once Dune II is fully implemented.

[ANewWorldRecord]

I think the concept is quite novel but I'm not a big fan of it. I do like the fact that you can see defensive structure radius, queue units, and just some UI changes that better work with widescreen stuff along with just any performance/comparability fixes but that's really where my praise for OpenRA ends. When someone makes a fan-made "balancing" change to a game, more often than not it falls into the category of "Play the way I want or be punished" mod rather than something useful. The Flak-Track for the Russians has ridiculous range and one of the devs added in a five second wait time for engineers to enter a structure (at least for the RA module) after he got the tar beat out of him with an engineer rush. Like I get it some people don't like Engineer rushes, but god it already takes 3-5 engineers in Vanilla RA to capture a building. If you cant stop 3-5 engineers in a guarded base, maybe it's not the game that's the problem.

 

Anyway in summation, they changed too much and that's why I don't play it.

[Matt]

When I came to find out about the OpenRA project I really loved how easily you can get your hands on it. For example the re-balancing of the engineers was all done by community members. Capturing a construction yard or other vital buildings was a predominant strategy a few years ago. Not walling in your deployed MCV meant you already lost the game which scared away new players. So the Generals style external capturing was implemented and used in the Red Alert mod. It turned the game in favor for more interesting late game battles. For the Tiberian Dawn mod instead the sabotaging where engineers damage the buildings and can only capture badly damaged ones known from RA95 was added to allow more fast pace games, but still avoid game breaking overpowered engineer cheese tactics after just a few minutes of gameplay. The best of it all: the balancing isn't set in stone. To revert back to an instant capture is just a matter of changing a text file. You can also add the rule changes to a map that you can upload at http://resource.openra.net/ to test it against other players who will automatically download your custom map when they join the multiplayer lobby.

 

All in all this is a huge contrast to https://cncnet.org/ which tries to preserve the original game with patches being contributed by very few skilled people who cartographed its byte code and http://redalertpp.org/ which also goes for a conservative approach, but also is very secretive and follows a closed invite-only development model. If all the projects had the very same goals and design principles, they would in turn be quite redundant, so I quite like how the community ecosystem evolved with everyone finding their niche.