Do we need tech levels?

If you are cybran you won't perform arty drops because their t1 arty has low DPS or would you argue that cybran arty drops are good compared to UEF and Seraphim?
PA is going to have 1 unit set. This means that YOU can't pick and choose tactics from all possible matchups to support your case. Choose one mirror matchup from FA and explain how and why 3 techs are so good.

3 T1 PD's doesn't stop 50 Loyalists. The T1 PDs can defeat Loaylists' for cost but Loyalists' are mobile. T1 PDs are not. Their range is also fairly similiar meaning that all your defending PDs won't be in range. While the Loyalist's can choose from which angle they attack. Depending on the map the Loaylists' could even attack where there are little or no defence and overrun the T1 PDs without few or no losses.

Also forgot about this little Gem, being a little mis-leading aren't you? your wording makes it seem like T1 Air scouts are the best scouting option when compared to other scouting options "until" T3 Air scouts come into play, this is a bad point to make to anyone that knows FA because;

A] There aren't other scouting options until The T3 Spy planes, there isn't a T2 scout plane, in fact there are only 3(THREE) Scout units(per Faction obviously), T1 Air, T1 Land and T3 Air. So what exactly are you trying to point out here???

B] It actually undermines you overall argument that higher Tier units don't make lower Tiers Obsolete. And to top it off in this case the only reason T1 is still viable even in the game has less to do with with it having an appropriate niche as it would in my 2 Tier system, but more so because AA options in Stock FA aren't all available(No T3 Mobile AA, which would shut down anything aside from MASS T1 Scouts, and even then leftover flak would still mess them up sufficiently)

Mike

I am sure I will draw a bit of flack for this; but can't you guys (namely OrangeKnight & ta4life) see that really what you are arguing about is a pretty minor detail? How effective T1 were against T3 in SC:FA or FAF has only limited relevance to this topic. Outside that argument at least you both agree that tech levels are required - though as to how many you obviously disagree.

What is more important IMO is that if people think that 3 tech levels is superior to 2 tech levels, then maybe they can suggest what can we do to address the issue - since Uber have already been quite specific about their intention of making this game with 2 tech levels. Not that they haven't been known to change their mind on certain aspects of the game to date.

Now it's very nice that ta4life has taken the time out to come down to the kindergarten and give all of us a 10 page lecture on how SupCom - particularly FAF - works and what the current state of the game is. But what I am asking is should it really be the core subject of 10 pages or should it simply be supporting evidence in his argument for employing a 3rd tech level in PA? My opinion is the latter.

What I would like to see is how the proposed two tier system could be implemented. That's why I commented a page or so back on what OrangeKnight was saying. I was pleased to see him trying to get the topic back on track and so far, there has been little or no negative comments about what I or OrangeKnight have discussed regarding a 2 tech level system. Though it has been hard for anyone to get a word in for the last 10 - 15 pages and I don't blame them if they lost interest.

@Ta4life. You repeatedly say about how good 3 tech levels are, but your comments on 2 tech levels and how they have been outlined in this topic have been minimal to say the least... And I think that is being generous. Why can't you be constructive instead of condescending? If 2 tech levels is wrong for this game then please point out why you think so. I agree that three tech levels has it's merits, but I think that 2 tech levels is the preferred option for PA. 3 was fine for SupCom, but that doesn't mean it will be right here.

I would say the majority of us have played SupCom, and although we may not be elite players, (I was about to say like ta4life, but you already dismissed yourself as an elite player and since you are so much better than us, where that leaves the rest of us I don't know) we do have an understanding of the game and how to play it. Perhaps a little humility would serve you well on the forum and maybe even garner you a little more respect and support. What we need to remember is that not everyone is a high level player. The game needs to cater for all of us, not just the hardcore. Not that those players aren't equally as important, just that there are other players out there besides the hardcore who will probably happen to make up the majority.

Other than the benefits that have already been outlined, a 2 tech level system should get the action started a lot faster. I think it alters the end game positively too. IMO it will be more befitting the scale and ethos of the game. Mainly because we are talking about massive armies and super-refined technology.

IMO we don't need to have 3 tech levels - we just need to have a lot more options available to us within the 2 tech levels. It will be interesting to see how a hundred or so units based in one pool, divided into 2 tech levels will be implemented by Uber. I think based around that it is more than possible to have a larger array of options available to the player than there were in SupCom.

Guys, I don't understand what exactly you are discussing. I have a feeling that it's something about "tools", not about real goals. Our real goal is "to make game interesting", and "diversity" is just one of the tools. Maybe it's very important tool and even required, but it not only tool.

Let me clarify my words with some fancy graphics.

But before that I want to introduce some additional terminology to prevent misunderstanding due to use of existing terminology.

Terminology:

Unit - a game object with 3d model, given stats fulfilling at least one role. Structures are also objects.

Role - a game design abstraction that serves as building block in producing complex winning strategies. In other words it's a particular possible "use-case", like anti-air, anti-pd, anti-exp. Roles are separated into all-rounder roles (which is generally useful in any strategy under any circumstances) and situational roles (this is hugely useful, but only in particular strategies under strict circumstances).

Player options - a number of all possible useful strategies that player may possible build-up with given set of roles.

Options availability - the amount of options given player may theoretically achieve at given phase of game.

Options affordability - the amount of options given player may practically achieve at given phase without stalling his economy.

Role efficiency - amount of weighted kills per unit fulfilling that role in head-to-head battle without accounting micro control.

With that definitions we may produce additional terms for generic teching mechanics:

Linear tech tree - model of units evolution when overall role efficiency is monotonically growing over time, but options count (count of roles) is constant.

Exponential tech tree - model of units evolution when overall role efficiency is constant, but options count grows exponentially (count of roles grow).


Graphics

Please note that all graphics are purely theoretical. They are here just to mark a problem we need to solve. They are not taking many things into account, including player skill and implementation. We will talk about implementation a bit later.

Total Annihilation
Let's start with local favorite, TA. It's obviously gravitate to pure exponential tech tree, but has slight deviations (as advanced factory is more expensive as basic one, so it's units are not available from t=0).



So, flat line means that there is no upgrades, no efficiency grow over time. Just as expected for pure exponential tech tree.



Aaaand. That's why we shouldn't blindly copy TA's system (IMO). This huge difference between availability and actual affordability is a bad thing. Why? We will discuss this later.

SupCom 1 series: Original, FA, FAF

SupCom is more of what we call "linear" tech tree, but more hybridish than TA.



Well, yes, that's what we expect for linear tech - each new tier boost up same roles to new level.

I've tried to make this steps huge to show that SupCom's efficiency model isn't
very balanced.



Ok, it's not flat-line as it should be in pure tech. as each tier also add new roles as well, but it's not that much growing as in TA. Worst part is last part when we actually start loosing viable strategies over time. And that's what is most frustrating for most pro players here. I also tried to show that in relation to TA, supcom has less roles and therefore (theoretically) less options. Actually, it's a little bit of confusion here, but we will discuss this later, into implementation section.

StarCraft

In reality, StarCraft is perfect hybrid solution. It could be not obvious from first look, but if you get a close look on it's upgrade system, you'll see a clear "overall efficiency boost" upgrades (like more armor, more shields, more DPS, more energy) and clear "options increase boost" upgrades (like siege mode, or fast move, or various spells for spell-casters). Also, availability of this options are properly guarded by complex dependency system. So it's graphics are as below:



Cool, isn't it? As a free bonus, starcraft also allow to actually variate from pure linear to pure exponential system. It's very flexible. So, why not copy starcraft? I'll try to answer this in implementation section, but in short: SupCom2 tried to.

This graphs are vastly inaccurate, cause StarCraft tree expanding system is not that streamlined as in SupCom or TA and highly depends on player choices and may variate from pure exponential to pure linear tech. But in most games it's something like that.

SupCom 2

It's actually a clone of StarCraft and it shared same graphics. Details below.


Goals.

Before I'll go into implementation details, let me clarify what IMO we should achieve.

There are few major groups of players and each of this groups is vital for healthy community.

1. Complete noobs - they never played in anything more RTS-ish than Angry Birds. This is fresh souls that are very hard to attract.
2. Newcomers - people experienced in other RTS games, but never played anything like TA/PA/SupCom before.
3. Beginners - people who just started to play game.
4. Mids - people who played game long enough to master most of options, know a lot of good strategies and have some knowledge about each unit's roles and their usage.
5. Pros - people who mastered every aspect of game. They know everything about units and how to use them. They are ones who actually are developing new strategies.
6. Cyber-sportsmen - koreans. They are just like pros, but they do most of ordinary tasks without any actual brain activity - it's just reflexes. They may fully concentrate on macro aspect of game.

So, why difference between affordability and availability, like in TA, is bad?
Because it rejects noobs (it's too much to master for a beginning), it's frustrating for newcomers and hard to beginners. Yet it doesn't affect mids and further, as they will group units into their own levels.
In other words, to enter a game (to became a mid) you'll need to learn a lot of information prior even doing anything. Otherwise any mid will crush you without blinking an eye. That will lead to community aging and death after some time. Just like it happened with TA and happening with SupCom.

Why fall of options in SupCom is bad is obvious. It's nothing bad for new players, but hit hard mids. Pros and koreans will handle that. Why? See below.

So, we need some solution with starcraft-alike graphics and good implementation.

Implementation

Why starcraft has such cool graphic, while it also have upgrade technically invalidating all previously un-upgraded units? That's easy - there is no un-upgraded units. Starcraft is using instant update model (just like SupCom2) and it's not giving you additional options while upgrading.

That is, one problem of SupCom is that it's actually giving additional options into intermediate phase of going from one tier to other, by mixing more efficient units with less efficient ones. After that intermediate phase is gone, this mixing options are gone too. This effect is less notable during T1->T2 and T2->T3 transition, but is very notable after T3 - there is no new options added, so we got this fall.

Second problem of SupCom is that it got more efficiency than new options per each tier, so raw power sometimes more viable than complex solutions. It's most hurting on mass-T4 level, which means even further fall of options count - now you have only one option "spam this T4's!!!".

Ok, but why SupCom2's idea of "let's just clone StarCraft" failed? Because starcraft flexible system is very flexible and therefore very fragile. It's very dependant on being well-balanced. But problem is - starcraft has smaller maps, starcraft has less units, starcraft do not calculate projectile physics (it has no projectiles, actually). SupCom2 do all this and much wider scale, that's why it was almost impossible to balance this solution. And it wasn't balanced, which lead to huge prevail of raw efficiency options over everything else. Do you remember commander rush "feature" from SupCom2? That's idiotic situation when you have only one option/one unit - ACU - and almost exponential grow of it's power.

I'm not even saying that Starcraft model is not suitable for streaming economy.

In other words, for PA scale it is insane to make anything too much customisable into game - it would be impossible to balance this out (single unit, ACU, is fine enough, but not more. Even SCU's in FA were a little bit OP).

How to achieve close-to-linear grow of both parameters, without big difference between availability and affordability, and with efficiency vs. options balance? I don't know, actually.

But I probably that just merging TA with SupCom won't help, after all.
It's obvious that we need some kind of tiers, or more dependencies than just two factories per type. It's obvious that we need some kind of efficiency grow.
But how to implement efficiency grow in balancable manner without loss-of-options effect?

Is it viable to mix research stations/dependency system with tiers?

I don't know, actually. Probably all solutions are possible. But I'm sure that just taking TA system is very bad choice. SupCom system could be balanced to make negative effects more soft, after all. TA system could not be (unless you'll make only a 5 advanced units, that is).

Glorious post nightnord. I'd never thought of it like that.

One way I can think of improving the gap that results between options available and economic power available is this: Have it so that higher tech mobile units are unlocked one at a time as resources become available. Of course, they would still have to be built in an Advanced Factory, so destroying the factory would still make them unavailable.

*face-palm.....

I am not even going to bother when people make up that much bias evidence to support their point.

Good lord....

It's got graphs, it must be true.

*edit* Having gone over it several times, I've gotta say, that's the biggest load of crap posted on this thread yet. It doesn't represent reality in any way. I suggest if you don't know about the games you're talking about, not to fill the thread with half a dozen oversized graphs.

Just my $ 0.02, but I agree with the idea of tier one being your main units, and T2 being specialists. An example could be, T1 you get tanks,bots, scouts, light AA. A good example of a tier 2 unit in my idea would be a sniper bot/tank. It would have long range and power, but poor LoS, which needs scout units to be used to its full advantage. Amphibious tanks,TMLs,Moblie Shield gens,etc also are definite T2. Please have amphib. tanks uber, they were a great raid unit in supcom,and there is no harm in adding them.

Yes, that's the point. Question is - how to do it? Simple "you should have X mass storages to build Y unit" (or X mexes) is something weird, IMO. Also, it doesn't add efficiency.

I thought about some kind of "tech buildings". You see, in SupCom your ACU got ability to build only very basic building. It were required an upgrade or engineer to build more advanced building. But in terms of animation, this "upgrade" (advanced bulid tech) wasn't much different from previous version - just more building rays. So we can assume that difference is in computational power and/or precision.

We can extrapolate this idea. Initially you may build only basic set of structures and buildings. To build more advanced units and structures you need to build tech center
(consider this as huge computational complex that is able to calculate complex production processes). Once it's built, you got more units and more structures available and you may upgrade it N times to got even more units and more structures.

Also at this point you got the ability to upgrade existing structures (this is worse than supcom model, as you can't rush eco before unit tech, but better than flat model) for more efficiency.

It's also possible to extend this to support unit efficiency as well - as soon as you built your tech center all existing units start to self-upgrade (to maintain unique L&F they may just transform into other unit upon upgrade end). All new units will be of new tech (difference between tech efficiency should not be as dramatical as it is in supcom).

You need only one tech center, but if it destroyed you - are back into the stone age.

I dunno, maybe this is bad idea (trying to hybrid research stations of some kind with TA model), just suggestion.

Where did you see "evidence"? It's an illustration, and you may try to argue with it's correctness. It shows tendencies more or less, not actual tech path.

Any specifics? Otherwise your words are empty, because you do always disagree and disrespect other's arguments, so it's nothing new

Not yet having read that giant post, I would like to know exactly what is wrong with it.

It makes no sense, and where I can makes some sense out of it I don't agree with it. Other parts (mainly the Starcraft related bits) don't seem too relevant to... anything.

I also fail to see how the area marked as bad on the second graph is infact bad. So you open a range of new options when you reach the second tier, so what? It's intended that you can't use them all at once. You have to plan your move instead of spamming everything.

Interesting. Now someone comes along and say PA should have teching similar to Starcraft. I remember people discussing this before. Propably in this thread .

I pretty much agree with nightord except that I don't think that the explosion of options when you open up tech 2 is that bad to the game.
Sure there is a steep learning curve that new players and new RTS player have to overcome which is I propably wont face because I have played TA and its' derivates since I was 12. There are several ways to overcome this like some good tutorials and a good campaign though. The same applies to having a flat tech tree.
In TA, Zero-K and alot of other Spring games, concentration of force and increase of weight means expensive units would be viable once the battlefield became more cramped even though for cost heavier units would be beaten by lower weight units. The increased range, HP and damage output, concentration of force, would make the heavier units win as the number grows. Typically heavier units would also be slower but there are still some heavy weight kiting units that are pretty fast but relatively weak. In that system light, medium, heavy and superheavy tanks all have their uses depending on the escalation of the conflict.

Come-on Zordon, take one for the team like the rest of us and get typing!

Mike

Okay, I read it. And I'm definitely siding with Zordon here.

Far more dirt than diamonds were to be found in that post. I think half of the graphs were either mislabelled, or irrelevant. I suspect that the utility of a unit is more influential than the availability of a unit. Yes, availability does go up in large steps, but the utility of what is added in that step is what causes problems.

Consider this slightly absurd hypothetical for FA: what if upgrading your factories didn't unlock T2/T3 units, but instead unlocked the T1 units of every other faction?

You will definitely see an increase in availability, simply because you have more things options at your disposal. But would you see much benefit from upgrading your factories? The more options really don't add a whole lot. At most you get floating artillery and floating tanks, T1 gunships, stun and AOE bombers.

So, did we see an increase in availability? I think ~4x the number of units is a massive increase in availability.

Now, did we see an increase in utility? Some, but not a lot. And what was added is mostly situational, like floaty tanks (useless on a desert, right).


The linear growth of unit power against economic power was never the problem. The amount of growth of both of them was the problem. And the disparity between unit power and economic power was never actually explained to be bad.



One of the things the analysis does is imply that having early-game units that become useless/redundant is bad. Along with this implication, is another implication; that this occurs largely because of increasing unit power/variety. While I agree with this, I do disagree with how to combat it.

As stated in the analysis, Sup2 and Starcraft solve this by increasing the power of early-game units, so they don't fall behind. I believe this is a band-aid solution: it mends the symptoms, but not the illness. To be more explicit in what I'm saying; redundant/useless units is the symptom, and the illness is the disparity between early-game units and late-game units.

So how do you treat the illness? It's pretty clear, you don't make units with such a large disparity.

One reason i like 3 tech levels as i have stated before is cost. The cost of the fusion power plant in TA (the t3 power plant in FA) is very expensive. You cant lump this in with other T2 buildings like t2 point defence. The same with the buzsaw style weapons (the t3 artillery buildings) you can’t put that next to t2 point defence or t2 factory’s it does not make sense.
New players are gonna be like this t2 building costs nothing and take 30 seconds where this t2 building costs an astronomical amount and takes 2 hrs.

I really liked TA but i never knew when i could afford the fusion because it was so expensive. I really like how FA introduced the T2 power plants and introduced t2 plane rushes and defences like shields without the need for the T3.

My biggest fear is that a solution would be to make the big stuff like the big bertha and t3 power much cheaper and turn it into T2 (mark my words this is what will happen) and this my friends will reduce the scale of the game dramatically and make it seem more like other RTS's such as star craft and less like TA or FA.

Think about it before you jump down my throat. It’s just a fear but after supcom2 anything is possible and i really want another good TA/FA style game!

I actually prefered the extreme jump required from solar panels/wind farms to fusion power in Total Annihilation; there was a huge cost risk vs payoff element to it which I felt was lost with the presence of the Tier 2 power generator.

The idea that everything in similar tiers is going to have to endure similar costs, or that less tiers equals less scale(The tiers themselves are just abstractions.) is frankly baseless(As is the idea that a new player can't figure out the difference between a 400 and a 4,000 metal costing unit.)

Why would we even need t2 or t3 power equivalents in PA? In FA higher tech power generators was simply more effective for cost. Why do we need the economy to be more complex without affecting combat?
PA could go with t2 power generators that aren't necessarily more effecient than their t1 equivalent but rather have different abilities. For example a t2 power generator could be mobile, could have cloak or it could turn itself into a giant lazor for example.

Why will that reduce the scale? Please justify why that will happen, because I can't think of any reason for it to happen.

You haven't read the post. It's no sense in discussing it.

There would be no campaign and I doubt that there would be good tutorials too. Tutorials of forums or youtube doesn't count.

You haven't read the post too, sorry. Particularly you missed graph titles and terminology part. Option is a building block for making viable strategies. Options based on roles.

In your example you do not increase amount of roles, nor increase efficiency, so there would be no options grow (both availability and affordability).

Not correct. I'm not suggesting StarCraft. Even more, I'm sure that such teching won't work on scale of PA and I wrote that down into my post.

Again, I'm not suggesting any solution in "big" post. I'm trying to outline problems, as people tend to argue about t1 viability in supcom which is quite incorrect direction of discussion. I suggested solution into my second post, but I'm not sure if it good enough.

What you are saying is diminishing efficiency grow out of the game. Efficiency grow is not a problem, so you are removing it just because. What you are proposing is simple TA-alike model, which is newbie-unfriendly.

If you can suggest some good mechanic to prevent that options burst than this solution would be good. But it would be even better if it would implement efficiency grow as well.

TL; DR Point of all my posts is: Problem with unit deprecation and problem with high learning curve are different problems. It's possible to solve both without losing good points of both models. That's what we should do. Discussing that SupCom model is worse than TA model is silly - they are different and they both are bad.

Fair enough. But would you say it would be the optimal solution for solving steep learning curves and tactical availability if it was possible to balance for PA?

Ability upgrades do give more tactical options like Blink-stalkers, siege mode and spell upgrades.

Anyway. You state that the steep learning curve is a big hurdle to overcome for a t2 system like TA got. PA is mostly going to cater to the TA lineage players so I'm not sure if that is going to be a big problem for its' popularity.