SE4 Strategy - An Art or a Science?

[Phoenix-D]

"my first question for Phoenix is "what do you do with all those extra worlds dedicated to research when you know it all?" I"

Scrap the research buildings and make something else. The speed advantage you gain in research is worth it.

EDIT: and I wasn't talking about *building* the ships per se. I was talking about having enough cash to build and maintain them.

Phoenix-D

[ 29 April 2002: Message edited by: Phoenix-D ]

[Talenn]

geoschmo et al:

Well, perhaps my wording was somewhat imprecise. You are correct that the weapons are somewhat varied in terms of research/kt/damage payoff, but from the games I've seen and played in, the players seem more or less pigeonholed into many decisions if they want to be competitive with other humans who are following the 'prefered' (for lack of a better term) strategies.

What I drew from Askan's post and from my own experience is that its mostly about Combat To-hit and Defensive bonus and Direct Fire weapons. Missiles dont even rate on the scale for the reason list below by Phoenix-D. The other weapons more or less depend on the stage of the game you are at and the amount of time you have to research rather than any real 'feel' of the weapons. By that, I would like to have seen more 'flavor' differences. For example, Torps that are monsterously damaging, but inaccurate. Some beams that could be extremely accurate, but low damage. Missiles could vary dramatically as well. Point defenses could have trade offs between range and accuracy.

I'm aware that some of these interactions are already present in the game, but generally not simultaneously. You have to focus down one path to optimize and that path will be the best one you can choose at that time. What I would have prefered is a selection of weapons down one path that would have strengths and weaknesses in relation to each other without the 'artificial' need to base it solely on tech level and damage/kt/turn.

Its hard to explain, I suppose, but I'd like to see more weapon be necessary for a variety of opponents. The little 'scraper' weapons might be useless vs a heavily armored/shielded rock, but could actually hit the fast little ships while the huge capital ship weapons could penetrate the toughest defense, but cant engage the small fries. In effect, you would need a combined arms approach.

In the current SE4, I dont see that. Usually you just stuff your hulls with most of the best weapon you know and be done with it. There is no trade-off in the weapons at that point. There is only a 'best' which is generally equally good vs all comers.

To me at least, a large portion of the tech tree seems unnecessary to be successful (at least on Medium sized maps...on large maps a lot changes, but its a VERY long process to play MP games on the larger maps). Most of our games center on developing beam weapons out the wazoo and going with the support of those weapons. No one goes down any of the interesting side paths as they simply dont provide enough return compared to players who 'go for the kill'.

Some other examples would be the 'Engine' techs. Every three levels you get something and the other changes are bare window dressing. My tech set added 'efficiency' at each level that wasnt giving a new movement bonus...ie, flavor.

Another example would be weapon mounts. In the base set, there is no good reason not to use the largest mount possible. That leads to less decision making and more 'pigeon holing'. I'd like to see a geometric increase in the cost (not size, as that changes the equation) or else 'to hit' penalties to make the ships have definate role rather than simply being better.

I guess that is my main 'complaint'...there is very little opportunity cost to most techs. They are simply better than the lower cost ones. IMO, increased tech should provide greatly increased options. But I dont see that here as often as I'd like. Generally, I keep the same 'decision cycle' and just the numbers change...eq, instead of a gun that does 30 out to range 5, it now does 35 and next level it will do 40 out to 6 and so on.

Hopefully that has explained my opinion a bit better than the previous attempt.

capnq:

Well, when I play singleplayer and just want to relax and 'role play' my race a bit, I tend to research the more esoteric techs too, but this post seemed to concern art or science in a competitive game. Everything changes when you are playing against people who dont fall for the same tricks time and again.

[Talenn]

My old mod (pre-gold) had many of those kinds of tweaks and choices in there, but they totally screwed up the AI. Given that my group cant often get enough people to play that we can totally dispense with the AIs, the requirement is that they can at least provide interesting opposition in the game.

I'd love to go back and work on a mod that totally cuts the AI lose, but with the number of times I'd actually be able to use it, I just dont think it would be worth all of the time and effort.

I really liked the Devnullmod as it coincided with many of my own mods (in fact, I even got a small mention in the readme. ). Perhaps now that the new Version is out, I'll give it another go, but the quality of the AI is generally not up to the TDM level for singleplayer play. Most of the folks I play with want to learn one mod and stick with it in multi and single player so that is why we had reverted back to the standard tech set.

Another change that I'd like to see again are real trade-offs in ship sizes. Larger ships should get expensive far more quickly. The way it is now, the larger ship classes are marginally more expensive due to all ships having the same general requirements...Bridge, LS, Crew, and ~6 Engines. The 'hull' cost is only equal to its size, and when a ship is totalling at 5700 minerals, the difference between a DS at 300 and a CA at 500 is miniscule. SE3 did a good job of showing the differences because of the Engine requirements on the larger vessels. Also, since they were build by component rather than cost, a larger ship took a LOT more time to build. In SE4 standard, a DS might take 3 turns and CA only 5. Thats just not enough incentive to build DS's once CAs or anything else are available. In my original mod, cost (and therefore build time and upkeep) was a huge factor in the larger hulls, but unfortunately changing a million AI construction files to 'understand' that was too much of a hassle for something I consider a pastime and a hobby.

Anyways, I'll probably get bit by the 'modding bug' again here in the near future. If I make any significant changes, I'll post 'em up here again for folks to see and piddle with.

Thanx,
Talenn

[geoschmo]

I think with the size of the user base on PBW, there may be enough of a demand for a "Human only" mod. I could be wrong, but I present as evidence that the "No-AI" mod is one of the most frequently used mods for PBW games.

Of course that is a far cry from the general use that a good single player mod like TDM gets, but I think it might be worth giving it a go.

I am getting a little off topic of this thread though so I think I will start a new discussion.

Geo

[PvK]

Talenn, have you looked at my Proportions mod yet? One of the main points (besides slowing down development of colonies and research) is to provide large numbers of valid design trade-offs. Small ships remain viable. Fighters and troops become almost necessary. Trade-offs appear between fast ships and long-range ships, firepower versus protection, quality versus economy, and so on.

I have adjusted the weapons and abilities around to make more variety of effective design. I had started a massive overhaul of the weapons and technologies, but these had to be cut from the original mod design for time reasons before the Gold deadline - just the manual work of re-doing the weapon entires was what took the most time, but eventually I might get around to it.

I too decided the AI had to come second. I actually have managed to get it to mostly work, but it takes way less time to set up interesting game changes that it does to tweak the AI to use them.

PvK

[Phoenix-D]

"No one goes down any of the interesting side paths as they simply dont provide enough return compared to players who 'go for the kill'."

The 'side paths' in this case are mostly more weapons; aside from a few specialized weapons like the Null Space and Shield Depleter, there isn't any reason to go down these, true. My point was more that you don't have to pick any specific weapon to be effective. My weapons selection generally depends on what I feel like using OR what will help me most.

-if I expect conflict very early, DUC
-a little later, PPB
-a little later, torps or meson
-fairly late
-if I am isolated, I research APB or Wave Motion

Phoenix-D

[geoschmo]

Talenn,

I agree that those kinds of choices would be nice to have, and would add even more flavor to the game. Adding those sounds as if it would be quite a daunting task. I know from experience that what you said about modding is correct. It's hard to do at all, even harder to do right, and then having to tweak things every time a new patch comes out...

In the Devnull mod we did something very similer to engines, making the mid-steps more efficent. My goal as to extend those out even further, for example, Ion-Engines out to level 6 or higher, where each level had increasing levels of efficency, or cheaper cost. Ad then when you switched to a new engine tech you lost all that in favor of more speed, and had to research down that thread. Never got aroud to that, and I was thinking it would have been hell to teach the AI how to use them.

Actually I am thinking about doing a new mod, and just conceding the AI from the start. That was always the limiting factor when working on the Devnull mod. We spent 3 hours in AI tweaking (work) for every one hour of brainstroming and designing new components (fun). In the end some ideas had to be scrapped not because they didn't work, or were unbalancing, but just because we couldn't get the AI to use them correctly. And when it comes to patch updates, that is almost always the majority of the work, making changes because now the AI stopped using something.

But if you think about it, how good is the AI ever? Even the TDM AI can't beat me unless I give them big bonuses or handicap myself in some way. But on the other hand I lose a lot more PBW games than I win. Not to mention I play so many PBW games I don't even have time for solo games, unless I am mod testing AI. So why bother right?

Something I'm going to have to think about some more.

Geoschmo

[Askan Nightbringer]

quote:

---

Originally posted by geoschmo:
I think with the size of the user base on PBW, there may be enough of a demand for a "Human only" mod. I could be wrong, but I present as evidence that the "No-AI" mod is one of the most frequently used mods for PBW games.

Of course that is a far cry from the general use that a good single player mod like TDM gets, but I think it might be worth giving it a go.

I am getting a little off topic of this thread though so I think I will start a new discussion.

Geo

---

I think a human only mod is an excellent idea. I was looking for your new discussion but couldn't find it
I build valuation models for a living and was speculating about using Present Values to analyse the tech tree. I could build a complex model using resource/research costs, current research/resource production and expected growth (and decline because of maintenance) of these parameters to work out say the present value of capital ship missiles 2 vs the present value of point defence 3.
Don't know if it would be useful but it would be way cool


Askan

[Atrocities]

Not to make light of this topic, but too me SEIV is neither a science or an art form. For me stratigy in SEIV is pure luck. Plain and simple, I role the dice everytime I process a turn. Will my race be conqured by a band of green pig people, will my ship make it through the warp point without getting obliterated on the other side? Stuff like that.

[Hank]

quote:

---

Originally posted by Atrocities:
Not to make light of this topic, but too me SEIV is neither a science or an art form. For me stratigy in SEIV is pure luck. Plain and simple, I role the dice everytime I process a turn. Will my race be conqured by a band of green pig people, will my ship make it through the warp point without getting obliterated on the other side? Stuff like that.

---

I strongly disagree. SEIV far from pure luck (otherwise I wouldnt be playing it!). SEIV is built on strategy. In strategy, you direct your forces to produce a favorable result from combat.

Probability is the "roll of the dice" you mention. The Science of Strategy is predicting the outcome of combat via probabilities. Calculating this involves math and while it can be tedious, it can be solved.

The Art of Strategy is plan or method by which you finesse the probabilities of combat to achieve that favorable result. The Art involves such things as trickery, feints, misdirection, essentially strategems. The Art is the human side of combat, where you win by out-thinking your opponent.

[ 01 May 2002: Message edited by: Hank ]