Lessons learned from TA and SupCom

This is meant as a replacement thread for Things TA did well that SupCom didn't since someone suggested strategic zoom was a bad idea. The majority of people here know there is no good argument to support that statement.

Let's document the things that worked well in Total Annihilation, Supreme Commander, Forged Alliance and SupCom II, and also the regressions introduced without dwelling too much on the obvious.

Personally I miss the wreckage blocking units and shots in Total Annihilation, but I don't miss the pathfinding. The only thing I liked about Supreme Commander II was the pathfinding, and although I never played anything more than the demo, I consider almost everything else in it a complete regression from what made all of the previous games so great.

Please don't bother mentioning mistakes that the devs have already addressed such as the non-streaming economy of SupCom II. Uber_neutrino has already said PA's economy will definitely be streaming, so it warrants no discussion.

Post away!

Something I think that is going to be interesting to address is how they go about tech. Personally, I think that both the ways that FA and SC2 did it were valid, but not ideal.

For example, the huge jumps in FA, and the fact that tech was made obsolete, wasn't that streamlined.

At the same time, as fun as the research system was in SC2, it had issues as well.

Perhaps an entirely different approach should be taken. I don't know if the Starcraft system would work either, or even make sense, to have some buildings required to be built first. You could simply offset tech with cost, but that implies that all tech is more advanced, when you could simply be adding another dynamic, like air.

I'm interested to see how the devs will approach this.


Well, let's see. Experimentals are a plus. They add an extra dynamic of epicness that I just don't get in other games. The AI capabilities in SC2 were actually quite amazing, especially how Sorian got them to build extra offensive bases. If we can get AI as smart as it was in SC2, I will be very, very happy. While I liked the research system personally, I don't think it will work with PA. I never played competitively, but from what I gathered on the GPG forums, Naval was pretty much broken, thanks to UEF exclusively getting subs, and Illuminate having no Naval, but most units could hover. So, that should probably be reworked.

One thing I especially liked in SC2 (not exactly liked by everyone) were the variances in experimentals. The difference between minis and megas were quite exceptional, and quite awesome. Minis weren't exactly good lone guns, but were great support units, and megas were great meat shields, but were fairly expensive to offset it. I liked this system.

In FA, there was lots of good stuff. The super expensive but super effective experimentals, for one. They were very endgame, but they simply inspired awe. Walls were interesting, and while they were ignored by most experimentals, I think having an effective wall system in this game could prove to be fruitful.

For anyone looking for more info on why the SupCom 2 AI worked so well: http://www.soriandev.blogspot.com

One thing I think made the first SupCom feel more epic than the second was the relative size of the units. Not only were experimentals huge, but more importantly the t1 units were tiny by comparison. The range of size made all the difference.

As far as tech goes, I'd like to see something of a refinement of TA's model. Commander can build basic factories, basic factories build T1 engineers, T1 engineers can build basic and advanced structures, including advanced factories. Advanced Factories build T2 engineers, which can build the top level structures.

So, Commander + T1 Engineer -> Basic Structures and basic factories
Basic factories -> T1 Engineers
T1 Engineer -> Advanced Structures and Factories
Advanced Factories -> T2 Engineers
T2 Engineers -> Top level structures.

As long as the number of units/structures in the basic, advanced, and top levels are kept small, they can be given unique roles without being displaced by higher level structures. Most of your army should be basic units, with a few advanced ones for support, and a top-level unit or two to anchor it.

I really think that Zero-K is the modern evolution of the TA philosophy. Take off the nostalgia glasses for OTA and play it on its own merits. It is better than TA overall. Unit costs are streamlined, there are more units, and more interesting unit diversity. Mex overdrive is brilliant compared to metal makers. And then there's the Lua UI which is quite advanced. You can draw on the map, or make points with text attached, to communicate with teammates- which is amazing. It is honestly mind-blowing how useful that feature is.

SupCom 1 did a few things better than TA. Graphics for starters- that game still looks good today. The AI is also brilliant. The UI is also much improved, and in general the game's fundamentals are much better than TA's. Strategic zoom was a game-changer for RTS games, and has become an industry standard.

TA is still the gold standard in terms of gameplay, though. TA really felt like a WAR. The ratio of units being built to units dying was about perfect, and kept the heat on at least a simmer the entire game, with big actions executed by the players to try and tip the scales in their favor. This might constitute a big push in one location with big units, constructing a bertha or other superweapon, or using a big air fleet to try and get a key structure like a fusion. Playing a skirmish against another player is a permanent fight, with continuous interplay of decisions, not peace with intermittent skirmishes and the occasional big battle that is over quickly.

Due to the interactions between unit movement, pathing through wreckage, and construction, the game creates this battle line dynamic which has not been replicated. Mounting a big push to try and advance the line, or attempting a breakthrough to kill economic structures behind the line of battle, are entirely emergent phenomena.

Almost forgot. Cheese. While in some cases it's hilarious, usually it's just not fun.

Example: I've seen replays of SupCom2 where someone just commander rushes a base, thanks to a close spawn point. That just...isn't fun. Obvious solution: far away spawn points. Of course, if you get to pick a spawn point, which is just awesome, you simply would have to account for this. Another solution is to just make repair and construction for bases work faster than the (initial) Commander can dish out, hence making this kind of play useless.

So yah. Cheese. If it's counterable, fine, but just...be careful about it, please.

What I learned from Supcom:

• -This projectile-calculation thing is awesome. I mean, if you miss something, it could still hit something else. also, its kinda boring, if you see a turret fireing and it only misses, because it has a set percentage of hits at this stage.
  -Performance is important. Playing with 6 Friends SupCom is NOT fun at all, if you allow 1000 units. The game becomes so slow, that it could take days to finish. Days.
  -Heavy machinery (really, really expensive ones) are multiplying the fun-factor ny 41³ by my calculations. Especially if they have huge laser (or whatever) beams swaying through the enemy lines
  -A huge battlefield is important to allow more than just a hand full of stategies.
  -Even though huge battlefields are nice, maps should also look decent. And they should provide much destructible stuff.
  -Upgrading Buildings. Dont leave it out. Preferrably more than just 3 tiers in one direction.
  -Assisting buildings with units is cool, but kind of messy. It'd be nice if one could build a network of multiple buildings to assist one building and upgrade those instead.
  -Don't make the factions too similar. Having different designs but almost the same features is kind of lame.
  -If you develop game-enders like an infinite-resources-building, make am disableable
  -Before SupCom was released, EARTH2150 fans were secretly hoping for battle below the surface. And customizable units.
  -I would have liked to see walls playing a bigger role.
  -I kind of missed real craters. Just scorch marks on the ground from a nuke or an exploding experimental were very dissatisfying.

This was a result of your CPUs. I have been able to play games of this scale without an issue.

It has already been confirmed there will only be 2 tiers upon release.

There will only be one faction.

An underground battlefield would only add another theatre to manage, resulting in 4 (maybe 5, with orbital) theatres on each planet you must be aware of.

This would result in lots of micro, remove the ability to instantly see what you're up against and add a lot to the complexity of balancing and development.

There was a mod for SCFA which actually added this, it was called Terrain Deform.

@thapear
about the performance issue.
Yes that may be. But my argument still stands. You can't say "hey, to have a 3vs3 match with full units faster than 5% gamespeed, everyone of you needs an ivy bridge quadcore"
I know, I'm over-exaggerating, but every pc was at least a dualcore, so that should suffice.

upgrades.
cool, I'll hope for more to come

only one faction.
well, if we had customizable units, we'll have no lack of variety. I hope they'll consider it.

about underground battlefield.
yes. and thats cool!

about the customizable units.
yes. and thats good. If I want rock-paper-scissors with a hand full of units, I'll play C&C.

craters.
omg really? thats awesome! I want this for TA!

I'm not sure if this has been posted elsewhere but I'll give it a shot anyhow.

When I was playing SupCom recently I noticed something that could be changed although I don't know how easy that would be.

It was the focus firing of defences on groups of units, particularly with air units and how the all the defences would fire on the same target resulting in a lot of overkill which meant that other air units could get through and do damage in the reload gap (mainly to do with Anti Air Missiles). I understand the merits of focus fire when it comes to big units as they will be more dangerous but larger groups mean some of the fire is wasted.

Obviously this would have to balanced otherwise rush tactics would just falter so maybe some sort of toggle option or something

Get thee hence, heretic!


Don't screw up the fluff with unauthentic gameplay mechanics. What reason could that have fluff-wise? One great thing about SupCom was, that it stuck to a fluff-concept and didn't break it, which made it convincing. I don't say keep teh SupCom tech-system unmodified, but don't bring in fluff-breaking things like that StarCraft tech system.

Image you had a factory that can build Gunships, Jets and Stealth Bombers. But the latter only if you had built the "stealth research facility"... bah! Don't even dare playing with this idea!

TA had interesting environments. Lots of asymmetric maps, and each map flavour was quite distinct. Forest with impassable dense trees and mountains, ice with largely flat terrain, city with few metal deposits, deserts with raised mesas, "beach" with large oceans, jungle with an excess of trees, cloud maps with barely any land or water, acid with damaging oceans and gasbags, lava with impassable seas, lunar with no wind, metal maps with ludicrous resources, martian with... okay it was like forest with no trees.

Compare supcom. Maps pretty much only differed by shape of land and oceans and metal placement. Sometimes there were mountains. Almost every map was symmetrical.

Sod the majority. It's just the SupCom fans on here outnumber the TA fans 10:1.

Personally I much preferred Supcom 2 to 1 (admittedly I gave up on the first game before FA came out). The resource model was too exponential. I felt it stressed economy building over offense too much.

I did however miss the sense of community you got from having gpgnet, replay downloads and such.

I'm all for symmetrical maps to be honest. Many of the maps in TA weren't symmetrical but it led to all sorts of problems. Some of the maps had more metal spots on one side. Or they'd more space. Half of the maps were completely awful. The ones that were popular in multiplayer were mostly the symmetrical ones.

Asymmetrical maps are the evil incarnate. As in PA maps will be spherical, symmetrical maps won’t look as artificial as in SupCom, I suppose.

That is the one biggest Issue FA fixed. You really missed out ;p

Mike

In fact something like this would be nice.
Let units do a simple check before they fire at a target:
If there are already enough shots flying at it to create overkill just consider that target dead already and attack something else.
That would be nice indeed.

Ok, even though I am probably best known for playing SupCom2 I am going to talk about how Uber Hack and TA did it and what it brought to the table.

I want to talk about Uber Hack, because Uber Hack was TA+ for me.

1. Uber Hack allowed you to choose the direction your factory faced. This is a feature that SupCom2 doesn't have and should.

2. TA had dragon teeth. Dragon teeth plus Light Laser Towers (LLT) were devastating to flash tanks and peewees.

3. TA had Medium Tanks. Uber Hack did a great job bringing Medium tanks up to par so they were useful for their range and their ability to kill LLT behind Dragon Teeth. They were slower, heavier, longer ranged and they shot in visible arcs that allowed you to attack from behind an allied unit and over small defensive walls. The down side was the tank speed and the projectile medium tanks fired was dodge-able by small fast enemies. This made Medium Tanks gain in popularity as the fight sizes increased, because larger groups had much less dodge capability. Keep in mind AOE on these was small and the projectile arc was about 15 to 20 degrees.

4. TA had rocket bots. Rockets were missiles that did not track their targets. They fired in a straight line and impacted the first unit encountered. Rockets were fun, because generally rocket bots were not super tough, but had more range and were effective at choke points where manueverability was limited or near defensive structures because they could add punch without getting too much into the enemy combat range.

5. TA had flees. Flees were tiny bots with a tiny laser. They were much in combat even against a peewee, but they had radar stealth, spider climbing and they had good vision and some radar. They were great.

6. Artillery Bots. Artillery bots shot a little bit further and a little bit higher than medium tanks. They were generally more reliable against towers and easier to dodge by mobile units. They were still useful in the big fights though due to their range.

7. Mobile Radar and AA. Good stuff for supporting ground forces.

There is no MML in Tech 1. I like this. It pushes players into fire fights under towers if they want a hard early push.

Generally speaking if the enemy builds dragon teeth first you can kill the constructor and build your own tower behind them. If they build the tower first you can attack the tower before it has a defensive wall around it. Also, with enough medium tanks or artillery you can wreck a good tower defense anyway with focused fire power.

Uberhack was one of the first mods to introduce AA units that couldn't fire on ground. This was very divisive at the time, but made room for the full game of TA to really shine.

One more thing to add that SupCom 2 did well, it's a small thing but when you move say 10 tanks somewhere, a circle with the number 10 appears above these 10 tanks when zoomed out. If you clicked on the number 10 it would select all 10 tanks. Of course in FA you could do something similar with Ctrl+1 but this did it automatically and you didn't have to worry about it. Of course I would still like to be able to use Ctrl+1 to make squads but I would like this as an option because if allowed us to select the squad in case we ran out of hotkeys for your squads.

I hated that, turned that stuff off as soon as the option was implemented, If I want them to be in any kind of control group of any kind I'll do it myself. The SupCom2 System just make it awkward to control them in some aspects from the zoomed out view.

this