modding : Not enough 3d artists problem

It's almost more important in low poly to be able to design the mesh for easy texturing and animation, than the actual final poly counts. A good clean mesh with a bit more detail than necessary is a lot easier to work with than a mesh with messed up edge loops and broken quads.

Basically if you can model, low poly isn't going to be a problem. So learn how to model first (how to use all the tools etc) and then you move on to using less polys to do the modeling with.

And youtube is a great place for learning. So is 3dbuzz.com, or 3dtotal.com, or one of many other places CG artists gather.

Those issues only arose in Starcraft 2. It looks like you haven't played any BattleNet 1.0 game, ie Starcraft and Warcraft 3? I really recommend it, even just to see, as the system actually works better in those games than I believe your solutions would. (And besides, Starcraft 1 beats Starcraft 2 in everything but screen resolution.) In short: in Starcraft 1 all currently hosted custom games were listed equally no matter the popularity of the map, and all maps were 'open source' (if technically owned by Blizzard) so any player could edit one he just played and immediately host his own tweaked version. There was no Blizzard approval, rating system or any other middleman between designers and players; a good map would get popular immediately and many of the best ones infringed on copyright or had eeevil sexual content (like the one where zerglings are sperm cells rushing to fertilize an Overmind 'egg', that was fun!). You whippersnappers have no idea how far gaming has declined since the 90s

The Starcraft 2 custom map scene is a serious decline despite of the more powerful editor - or perhaps partly because of it, as the added complexity turns off the non-coders which used to be the lifeblood of the scene. Basically, anyone with the skill and time to make full use of the new editor could probably get a job in the industry, so why should they bother? The strength of the editor was always that anyone can use it without worrying about scripting or modeling. There's also a host of other issues like the stupid popularity based browsing system, the Blizzard censorship, the closed source style model and more. I made a thread dealing with the difference between BNet 1.0 and 2.0 and why Uber would do good to take inspiration from the former.

Is there any reason you can't do both?

This is the case for any game company vis a vis their mod scene, which is why it's so smart to focus on making it easy for players to reuse the included models, animations and sounds in new gameplay arrangements.

I don't think that many people were needed to make the 1998 Starcraft Campaign Editor. A PA editor really doesn't need to be any more complex than that to begin with. And now they have the advantage of a working model.

As far as needing a good editor, I dunno. I've always believed that if you're using the tools the original game devs are using, you're not going to get much better. I mean, we made mod

A much much more important factor, as mentioned already by lophiaspis, is how the mods are treated after they are created.

GPGNet fell into the same 'popularity' trap for maps which seriously damaged the map making scene - the 'most popular' maps were one or two maps that had the most downloads of all time, which meant that the top listing stagnated.

Whatever mod distribution system Uber comes up with, it needs to make it dead easy to get new mods, reduce any installation of mods to a single click, and automate updating of the mods so that everyone is always on the same version. So basically copy / use Steamworks as a base and make it better.

Do that, and the mod editor can be notepad - we can manage.

Techncially, SupCom-style is both already just missing the ability to 'lock' a mod-map combo, but you could still create specific maps for specific mods, LAB Wars had this.

Again, SupCom had this, it might not have been as easy as it was in the SCI editor, but the SupCom method wasn't limited in any way unlike the SCI editor, it was even easier to just change the stats on existing units in the game with a BP_Merge, even I can do those.

A map/campaign/unit editor would have to be many, many times more complex than the SCI editor, SupCom units had so much more depth to the variable fields when compared to Starcraft.

In the end, an "all inclusive" editor can be made for PA, it can be done for any game really, the problem is the time and effort it'll take, which isn't worth it considering PA is already going to be moddable on it's own.

Mike

Low Poly modelling isn't too hard, it more a mind set of understanding that you're creating an impressionist or slightly cartoony asset rather than a realistic one. Basically use primitive shapes to start with, avoid curves, avoid some of the more complex geometry types (which usual have their own propriety names in each program anyway).

Realize that as an RTS your model is probably going to be rarely larger than 100 pixels in size on the screen. You're going more for silhouettes than details. That's about all that I can really think of for low poly modelling, like anything it's an art form to do well.