overwhelmed

[smacksim]

This is what you do: Clean Shell Units and upgrade only as needed.

1-1-2 Clean (and possibly trained as well).
1-1-1 Clean Police (usable as is in conquest as drone-quellors, upgradable for heavy assault if needed)

For air units they will often be able to reach their targets soon enough that upgrading isn't worthwhile ..... and......

a 4-1-8 Clean Air-Superiority noodle will always be useful against enemy air and for ZOCing or providing cover for your units.

a few 8-1-10 Clean / Trained Choppers will be able to take out any ground unit of the pre-photon wall era, so they don't need upgrading

etc..

95% of battles should be well understood before the military even gets created: IE, you don't need 20+ military units in the early game unless you are going to go for the jugular of an enemy. If you are, pre-clean units, 20+ 4-1-2 rovers will take any AI out ---> plasma armor.

A player should consider if they are going to get advanced weapons/armor before even making a military. If they will be getting better abilities then its better to make 15 1-1-2 rovers for upgrading to 15 4-1-2s using cash than it is to use those same resources for 20+ 2-1-2s, not be able to win, and be broke.

The better you get with military, the less units you'll need. In the rare cases where you want to ship a bunch of early game units half-way around the world you have to consider the expense of support and later upgrading, the danger of the journey, etc.. Most players simply wait till air-power before attacking at all (unless its very early) for these reasons.

Clean Shell units are cheap and pretty much no-risk to make. Having 50 clean 1-1-2s sitting around because you overdid it isn't going to hurt you all that much (lost production, yes, support, no).

Another thing to consider is simply not making units that will quickly go obsolete, hence the value of 1-1-1 clean police. These can later become 10-1-1 drop and 1-5-1 AAA Trance if you want, or anything else.....

The chassis never go obsolete, so neither should the units. This is why actually building anything but shell units should be undertaken only when you're sure to be using them immediately, IMO.

Edit: I should add though that making assault units (air, rovers, etc..) clean is something of a waste againt stiff opposition. This is fine vs. the AI or when your production is high. Against equal opposition or with lower production (a crawlerless game perhaps) simple 1-1-1 trained shells are a better use of resources. Any unit that has a questionable viability shouldn't be made clean if another choice is available. This is why generally Formers are the first to be made clean.

[d=me]

But what if the enemy also upgrades? When you build units do you expect to use them immediately and let them die? If you want to have a tech advantage over the others so that your units are stronger don't you have to upgrade?

And what if you started beside mariam?

[GeoModder]

If you start next to Miri, it is important to attack first for the 25% attack bonus those Believers have.
I've never seen the AI upgrade its existing units.

[Senethro]

Sounds like you have production problems d=me. How many bases do you have, whats your typical governments and how many supply crawlers/formers do you typically have?

[smacksim]

Shell units are for upgrading.......didn't I mention that? You just upgrade them at the last minute. It costs that unit a turn (if using ctrl-u and not mass-upgrading), so it takes planning.

Starting next to Miriam (or Yang or Sparta for that matter) can be death for even the most veteran player. If you're right next door then ignore the 'clean shell' technique unless you are sure to be getting a new tech (and prototyped unit) very soon.

What do you do if Miriam is right next door, you have a huge but obsolete military and no cash to upgrade more than a handful of units at the moment?

I'd say: Try to buy some time. The Believers will be able to field more units more quickly than anyone but the Hive in the early game, and its perfectly reasonable to aquiesce to Miriam's demands and make a strategic retreat to regroup in many early game situations.

To your other questions: Yes, of course you want to upgrade to have a strategic advantage when you actually go into combat. On the other hand, its sometimes not affordable, useful, or possible to upgrade every unit. Consider:

You have a stack of 4 2-1-2 rovers sitting outside a Believer base guarded by 2 1-1-1s and a 1-2-1. In this case, upgrade 1 rover to a 4-1-2, kill the 1-2-1, then use 2 other rovers to kill the 1-1-1s and move in. No need to upgrade them all, right? And why should you if you think you'll soon be getting Superconductor or Syth Fossil Fuels, in which case you'll want to be upgrading again?

So generally you only want to upgrade a military unit once: To your most current technology that will serve for the war at hand. This is why going from 1-1-1 --> 6-1-1 is far better than 1-1-1 ---> 1-2-1 ----> 2-2-1 ---> 4-2-1, etc.. That is very expensive!

It comes down to math as far as upgrading goes. But it sounds like your strategic situation is more complex than merely wondering at the utility of upgrading 'obsolete' units.

Crush the Believers at any cost I say! Sometimes that means taking a step back until you can field vastly superior units. If you have the possibility to field a bunch of 4-1-2s via upgrade but Miriam has thousands of 1-3-1s around, is it worth it? Or should you give her some gifts and seek a truce of some kind? Hell, I've had games where I had to put a few colony pods on ships and move away altogether when the going got rough. Its just not worth it sometimes to fight 'even' wars, especially against such a mean ***** like Miriam in the early game.

[Chaos Theory]

Don't put clean on shells unless you plan to stockpile them! Once a unit is upgraded, its special abilities (except Trained) cease to have any effect, and to get them on the upgraded unit, you must pay for them again. Instead of a clean-trained 1-1-1, build a trained 2-1-1 or whatever for the same price, and save some ECs when upgrading.

[Kataphraktoi]

Starting next to Miriam (or Yang or Sparta for that matter) can be death for even the most veteran player. If you're right next door then ignore the 'clean shell' technique unless you are sure to be getting a new tech (and prototyped unit) very soon.

i wouldnt say so...i convinced my dad to try a LAN game and he started next to spartans and held em off anyhow(he never played before)

don't put clean on shells unless you plan to stockpile them! Once a unit is upgraded, its special abilities (except Trained) cease to have any effect, and to get them on the upgraded unit, you must pay for them again. Instead of a clean-trained 1-1-1, build a trained 2-1-1 or whatever for the same price, and save some ECs when upgrading.

the idea was, stockpiling clean units for later. although,production tends to go up later,not down, so id still build them on-demand

It comes down to math as far as upgrading goes. But it sounds like your strategic situation is more complex than merely wondering at the utility of upgrading 'obsolete' units.

i would say there is far to much emphasis on upgrading here,for what has to be more complex,yes...

you really only need half a dozen units to take on the fiercest AI,just get say 4-1-2 in your city with sensors and attack first.most early stacks arnt enough to worry about.a forest walls and 6-1-1 work against masses of rovers later,when a mobile defence wont cut it(without alot of production)

[smacksim]

Originally posted by Kataphraktoi
i wouldnt say so...i convinced my dad to try a LAN game and he started next to spartans and held em off anyhow(he never played before)

So you would say:

Starting next to Miriam (or Yang or Sparta for that matter) can't be death for even the most veteran player.

Really? It never happens that a player gets killed by a rough start next to an agressive AI? Or are you just being ornery?

[Kataphraktoi]

Really? It never happens that a player gets killed by a rough start next to an agressive AI?

im sure it happans,but even rough spots are manageable. played ''things that go whump in the night''? the believers are about as nasty as they come,and blake is doing an ok job fighting them.

if i came across as 'ornery' that was not my intention

[Flubber]

Originally posted by smacksim
.

Starting next to Miriam (or Yang or Sparta for that matter) can be death for even the most veteran player.

While yes it CAN mean death (ie it is possible) in 99% of cases against an untweaked AI, it simply does not mean death. Either you suck up nad buy peace, or probe away a good unit, or use superior production to beat them OR OR OR.

I have lost to the AI precisely once in the last 3 years and it was because I got stuck on a small island as the drones and made some big strategic errors. Please please give me an adjacent faction and any veteran can turn an AI neighbor to their advantage.

The AI is , quite frankly, tactically stupid and very very easy to defeat with basic tactics.

[Flubber]

On upgrading, I will frequently have very obsolete units among my empire. Even in the very late game some of my garrisons may be 1-1-1 clean/police although anything on the frontier will likely current.

I don't TYPICALLY build a lot of units pre-airpower, counting on agressive scouting/patrolling to give me notice of opposing forces and fast production to allow me to beat back the attack

[Blake]

I remember I've lost a couple of games to AI's. Once a Spartan Impact rover appeared from nowhere and took University Base (I only had one other), game over.

I lost an OCC due to a planetbuster.

I've also had games when an AI COULD have killed me in one move, but declined to. Like when the Caretakers declared ventta, parked their Ogre outside Morgan Industries (defended by a single scout patrol), then, sat there. (I eventually won that game, it was pretty tough and required very creative strategies)

[Flubber]

Originally posted by Blake
I remember I've lost a couple of games to AI's. Once a Spartan Impact rover appeared from nowhere and took University Base (I only had one other), game over.

I lost an OCC due to a planetbuster.

I've also had games when an AI COULD have killed me in one move, but declined to. Like when the Caretakers declared ventta, parked their Ogre outside Morgan Industries (defended by a single scout patrol), then, sat there. (I eventually won that game, it was pretty tough and required very creative strategies)

Don't misunderstand . . . I lost a bunch of games when learning . Now . . I don't know why, but I haven't faced those early rushes. If I meet the Spartans or catetakers early, they were always willing to make peace long enough for me to beeline to weapons or build a cash stockpile to probesteal a good unit.

I also agree that there are games that I could have lost if the AI did something simple (like suicide some needles to get through a trawler barrier and PB me) but thats the point really. The AI often doesn't do what it should to win

[James M]

Originally posted by smacksim

So you would say: Really? It never happens that a player gets killed by a rough start next to an agressive AI? Or are you just being ornery?

Once upon a time, I was testing something-or-other (my memory fails me) using the drones, and since it didn't require an empire I just built my first city, garrisoned my other cp and scout, started building energy and hit next-turn around 40 times.

As it turned out I had started out sandwiched between the two aliens. Once the test was done I decided to actually play the game out, began expanding, and very swiftly got into a war with the Usurpers.

I won that one, with laser infantry vs plasma steel and impact.

So yeah, barring something really astonishing, a veteran player should win every game, regardless of start position, neighbours etc. It might look dicey at times, but the AI doesn't know how to win.

[Flubber]

Originally posted by James M


Once upon a time, I was testing something-or-other (my memory fails me) using the drones, and since it didn't require an empire I just built my first city, garrisoned my other cp and scout, started building energy and hit next-turn around 40 times.

As it turned out I had started out sandwiched between the two aliens. Once the test was done I decided to actually play the game out, began expanding, and very swiftly got into a war with the Usurpers.

I won that one, with laser infantry vs plasma steel and impact.

So yeah, barring something really astonishing, a veteran player should win every game, regardless of start position, neighbours etc. It might look dicey at times, but the AI doesn't know how to win.

Not that surprising really. When I play challenges such as what you did, I generally find that being weak and pathetic can yield huge dividends in the diplomacy world. Pretty much nobody views you as a threat . . . so you get peace and assistance from lots of folks.

You can even base anti-AI strategies on it since things like massive mineral production, terraforming, hovering probe teams and infrastructure don't seem to count in the AI assessment of how powerful you are. Its fun to milk the AI for credits and tech before going on a proberaping rampage for more tech and credits and then cashing 5 AAs while beginnin a pop boom.

I do it for fun . . . Intentionally build a super strong empire that an AI would see as weak. You can almost feel their collective " How the heck did that happen?" as their diplomacy gets less and less friendly.

The fun part is that since I wasn't a threat, their armies are almost all deployed elsewhere and it takes them a while to catch on.

While in a multiplayer game, most people will advocate energy strategies since getting behind in tech can be fatal, in an SP game, a very mineral heavy strategy can work wonders. The AI will gift and trade you important techs in a way a human never would and the AI never seems to realize the explosive power of those 40 minerals a turn bases