specialist and city radius

[Fitz]

quote: Originally posted by The diplomat on 01-29-2001 08:32 AM It seems to me that it is better to put crawlers outside the city radius or is it better to put crawlers inside a city radius so as to get specialists sooner?

The way you phrased this question, it sounds like you are unaware that you can change workers to specialists anytime that you want to. All you have to do is click on the worker icon, or the tile being worked, from inside the base screen. Then you get a pop-up letting you select a specialist.

If you missed this, did you also miss that you can rearrange the tiles being worked? If so, try drag and drop from one tile to another in the base screen.

[Blake]

Two tricks using specialists, one dubious and another a cheat.

In the early game (before crawlers, but generally assuming WP or nutrient restrictions lifted) if you have a base at size 6 or 7 (working kelp or other high food stuff) then turn it's entire population into libarians. It's food plummets and will probably completely deplete the supplies. Next turn back to workers, so your base produces atleast 1 excess food. Next turn back to libarians, once again it depletes it's food supplies, but as long as there was atleast one food in the nutrient box the previous turn your city doesn't starve. You can repeat this as long as you like. OR if you need a really big wad of cash turn your entire population into technicians for a turn, next turn restore them to workers. This can create obscene amounts of labs and cash every second turn.


The second thing I've found would be a really lowlife thing to do in a multiplayer game, infiltrate the datalinks of an AI, then hit F4 for base screen, select the faction you infiltrated find one of there nice big cities in the base screen then right click on the workers... despite the fact they are not yours they get turned into doctors! (empaths or transcends later) turn entire population into doctors, repeat next turn, city starves, loses all production and all energy production except base square. Repeat, repeat repeat until every city in there empire is size one, or more likely, you get bored.

Because it is so obviously cheating I've never used it even against AI's, except for amusement value (which I assure you, was very shortlived). This bug is in both SMAC and SMAX, incidentely I found it when I was studying the progy citizens for the first time, I was amusing myself with the way they change color slighty as I scrolled, I clicked on them to see if it would shuffle all the colors then I noticed the bug. Damn lousy programmers.

[Ogie Oglethorpe]

Damn,

Your away for a couple of days and next thing you know you've missed everything.

My thoughts on specialists and mid game. Much depends on the pace of the game firstly. The longer you have the ability to effect some significant t-forming the more effective you will be at growing population centers whilst still removing citizens for specialization. Your growth rate is most important as each citizen represents energy. Growth rate normally becomes limitted via ability to prepare crawlers/trawlers as well as t-forming improvements.

Regarding pace of game,consider if you are running a blind research game wherein restrictions aren't lifted for some period of time. Then all of a sudden case 2 with WP can break your game wide open with respect to tech advancement etc. as condensor farms allow 4 nutrients pre gene splice.

Some simple calculations (all assume no +2 econ, no trade energy, and assume no efficiency loss)

Case 1

8 square calcs.

Forest Treefarm/hybrid configured.
Nutrients = 24 Minus 16 to support workers = 8 available or 4 specialists
Mins = 16
Energy = 16 + 4 specialists at 3,5, or 6 equivalent energy = 28, 36, 40 base energy respectively

Former time required = 8 forest times 4(3) per square = 32(24 turns w/ WP) (assumes no forest growth)

Case 2

Contrast Condensor farm

Condensor farm 8 squares all crawlered
Nutrients = 32 or 16 specialists
Mins = 0
Energy = 16 specialists at 3,5, or 6 equivalent energy or 48, 80, or 96 energy respectively.

Former time required = {8(6) + 4(3)}* 8 squares or 96 former turns (72 if WP)

Mins required = 8 times 30 or 240 mins for crawlers.

Case 3

Kelp farms and Tidals plus Aquafarm & Thermocline

Nutrients = 32 minus 16 for workers = 16 for a total of 8 specialists
Mins = 0
Energy = 32 base + 8 specialists at 3,5, or 6 or
56, 72, 80

Former time required = {4(3) + 6(4)} * 8 or 80 (56 if WP)


Option 4 Advanced Mid game

Condensor/farm/soil enricher 6 squares
Borehole 2 squares

Nutrients 32 - 14 specialists with 4 nutrient surplus
Mins - 12
Energy - 12 = 14 @3,5,6 = 54, 82, 96

Mins required = 6 crawlers at 30 or 180 mins

Former time = 72(54) + 6 soil enrichers @8?(6) + 2 Boreholes @24(16) or 144(122 w/WP)

Any amount of subsitution/scenarios may be considered but the anlysis remains the same.

Realistically one needs to look at the pros and cons of each faction. Contrasting Case 1 to Case 2 and 3 the benefits in the mid game pre fusion (pre engineers at 5 energy i.e. librarians and technicians at 3 energy) is at best a break even proposition with a lot of t-forming required and a lot of mineral expense in those cases:

Case 1 is 16 mins and 28 energy
Case 2 is 0 mins and 48 energy but expended quite a bit of mins and former time

Case 3 is 0 mins and 56 energy expended less mins and less former time than case 2 but significantly more than case 1

The case becomes much more strongly made tho' once Engineers become available. In these cases then comparison becomes

Case 1 16m/36e
Case 2 0m/80e
Case 3 om/72e

All of a sudden case 2 and 3 become much more palletable and need to be seriously considered despite the investment in terms of former time and/or crawler mins.

One aside for case 3 one nice aspect is that half your citizens are still on the ocean as workers and if enough psych energy is applied can turn to all workers to talents. If running wealth then the resulting golden age will allow +1 energy per square worked and harvested for the future. This can be a very important point. In order to acomplish the same effect for Case 2 normally one normally needs to run FM.

Beyond the advanced mid game approach specialists become ultra important. I define this stage where sky farm orbitals, transcendi, and Hab domes are in play. No single square delivers energy as a high nutrient crawlered square. Each square yields 6 nuts matched with orbitals allowing 6 transcendi at 6 energy each. Thats 36 energy per square. On the ocean with kelp and Aquafarms its 24 per square.

So in summary, My game revolves around a traditional approach via forest and tree farms but keeps an eye to keeping coastal bases. Clean allows build of massive farmer (sic) forces which in turn reclaim the interior of the continent to allow Stage 2/4 implementation as fusion power approaches. Base spacing normally follows the base every three squares approach.

Worrying about PB's is and should be a non issue. PB's are a fact of life and no real defense exists other than making sure you get there first and then upwards to ODP's for real defense. I say this knowing full well that any competant player as soon as PBs are discovered can cash in their crawlers at 10 or so bases and have an instant nuclear force for launch of all 10 the next turn. No amount of base spacing etc. is going to realistically allow you a realistic opportunityt o survive that kind of onslought. Moreover, the more likely scenario of x-choppers raining death on you is the more likely approach as MMI normally comes much sooner in the game. The key to both of these is getting to the techs before the other player.


[This message has been edited by Ogie Oglethorpe (edited January 30, 2001).]

[The diplomat]

wow, you all have given me some really cool strategies. I can't wait to try them in a game.

Blake: your first strategy is not a cheat but it seems like it is a "trick" that would be "unfair" because the AI would not be able to compete. Doesn't it give a player an unfair advantage?

------------------
No permanent enemies, no permanent friends.

[Blake]

Diplomat, I said it was "dubious", and it definetly gives the player an unfair advantage against AI's, ofcourse human players always have an unfair advantage against AI's, because they can think. AI's also get an unfair advantage against Humans, in the form of +2 growth +2 industry -2 drones per base, a big bonus to research, ability to see hidden units without sensors and so on. Sometimes it becomes nessecary to use every unfair advantage at your disposal in order to beat the AI's unfair advantages (especially when fighting the aliens). And the strategy of turning your entire population into specialists every 2nd turn is a big MM pain to perform, because every doctor has to be manually turned into a usful specialist (about 4 clicks per population point), it's such a pain that I have only done it a couple of times (once when playing the Ultimate Builder Challenge, I missed the WP and got desperate). But the AI can also do it's own MM which would be an almost impossible pain for humans to perform (for example re-arranging every bases workers every turn). As for the AI not being able to compete, that's the idea I hate losing to AI's.

[Skanderbeg]

Base spacing has not only implications for the use of PB's, but as well for conventional combat.

I give an example: Faction A with two unarmored chaos-rovers is attacking a base of faction B, garrison one chaos-rover, one handweapon-plasma stell garrison one probe team, no defense perimeter. Bases connected with a road.

1. Bases two squares away (the AI does this often, especilly in jungle and garland crater)
First rover attacks, kills the garrison. Second rover attacks, kills the rover and probe team. Both or at least one rover rolls into base with their remaining 1/3 move points.

2. Bases three squares away.
First rover attacks, kills the garrison. Second rover attacks, kills the rover and the probe team, and rolls into the base.

3. Bases four squares away.
First rover attacks, kills the garrison. Second rover attacks, kills the rover and the probe team. But no move points left to take the base.

4. Bases five squares away.
Rovers have to do a hasty attack, or they have to stay in the open for one turn. With hasty attack, at least one rover would be lost, and the base couldn't be taken. The probe team would buy out the second. Staying in the open would invite a counter attack. The rover of base B would kill one rover, the probe team would buy the other.

This is, of course only an example (but one which I think as a typical situation), but it shows clearly that narrow spaced bases support the attacker, and wide spaced bases support the defender.

"Steelborn, Starborn"

[cbn]

Skanderbeg

While I agree that closely space bases do allow an attacking rover to have at least 1/3 of a movement point left to take a base, your analysis did not really consider the defenders options.

let us assume that the defender is competent enough to have sensors and to know where your force is. If we are considering a force of two rovers the closely spaced bases should stop you cold. The defender should have a number of garrisons from the 2-3 bases within 3 squares that can reinforce the target base before you hit it. After all it only takes 1 turn to move the troops 3 squares . If he can get 3 more units to the threatened base, the attackers best case is to beat all 4 garrisons and and then likely see both his (now damaged) rovers get destroyed.

I am usually a believer in active defence and usually have 1 attack rover for every two cities in "safer" frontiers and 1 per city in the direction of potential enemies (general minimums pre-air power). I have always thought of this as typical prudent defense strategy given the weapon-armour disparity at most times in the game. So, as soon as your force appears, there could be quite a few rovers waiting for you.

Meanwhile the horde of near bases, means that the defender can produce a very large number of new units in the area very quickly. An (upgradeable) scout per base is typical each turn. It can be very tough to take that base when defenders can be replenished each and every turn.


Don't misunderstand my point. I have often rolled the AI using the tactic you outline. The rover taking the base gets its damage repaired and next turn -- move two squares -- attack and use the remaining 1/3 pt to take the base. So the point is valid but it would not influence my base spacing as:

1. Defenders can destroy the road on any given square to slow the attacker (or not road it to begin with).
2. Defender can impede you with formers or crawlers (my bases all generally have crawlers as soon as they are available). Crash upgrades to armor these guys can really blunt an attack.
3. Defensive reinforcements arrive in a timely fashion


I generaly hate to defend since in SMAC the best defense is a good offense. But if forced to defend (ignoring the massive PB strike possibility) I would much prefer to have closer spaced bases in the area of attack. I have a whole bunch of reasons for liking wider base spacing (hate micromanaging lots of bases) but found I had to disagree with the rationale you put forth.


cbn

[Skanderbeg]

cbn

You're right in many points, but my example should only be an example, how base spacing could implicate conventional war. I didn't want to involuve to much units to not complicate it to much.

Of course narrow base spacing gives the defender the ability to reinforce fast. I know it well how difficult it can be to take out the AI's 2-squares-spaced bases in the jungle.

But I wouldn't agree to some of Your arguments:

"Defenders can destroy the road on any given square to slow the attacker (or not road it to begin with)."

That's right. But a defender destroying a road loses his move points and is a sitting duck for the next turn. And destryoing you own roads will slow down Your own counterattack.

"Defensive reinforcements arrive in a timely fashion"

If You space Your bases wider, You only need some hand-weapon/best armor rovers instead of normal garrison. These units come also very handy for capturing bases during the counterattack!.

The thing I am unsure about is using formers and crawlers for defense. I usually do this only if it's the last possibility to defend, but not for my general strategy (although I know Vel recommends very much to do this).
Even armored, they are still easy prey, and if they were killed, You loose the money for armoring. I think it would be better spent for upgrading or rush-building fighting units.
Additinally, I don't upgrade crawlers because that's seems to be cheating to me. But that's my personal opinion.

[cbn]

Skanderbeg

It sounds as if we are more in agreement than not. I did expand your example to consider outside forces since it seems necessary to do so to truly consider the effects of closer spacing on base defense. I do agree totally that a close set of bases can be taken by rovers in exactly the manner you outline but doubt that a human defender would permit it so easily.

I know that things could get really complicated if we consider air power, drop troops, choppers and nerve gas etc ( don't you love the wonderful complexity of this game). Thats why I tried to stick to the reinforceability of the bases in your type of example and point out a couple of areas where we differ.

You said:

That's right. But a defender destroying a road loses his move points and is a sitting duck for the next turn. And destryoing you own roads will slow down Your own counterattack.


My response:

Everything you say is correct. However my defense strategy makes frequent use of the scout unit as a garrison. So upon seeing the rovers, and if I am not already comfortable with my defense in the area ( and assuming I want to hold the base) the strategy would be:

1. Move a scout out in the direction of your rovers and cut the road link. This guy is expendable and very cheap. At the cost of one scout the rovers are now unable to take the base next turn. In fact the most that could be accomplished is

-- Rover one-- travels on road 1 tile- attacks and kills scout-- attempts to move next to base with remaining 2/3 points

-- Rover 2 -- travels on road 1 tile-- moves into unroaded tile-- and attacks at 2/3 strength

2. Bring up reinforcements to the area. We differ in that almost all my rovers are attack-only variety. With 3 square spacing there should be 4-6 bases within 6 squares of the attacked base (so at least 2-3 attack rovers) and at least 2-3 bases within 3 squares (other defenders to bring in). So there should be a formidable force waiting and now, each attacking rover gets only 1 attack before it has to sit out the turn next to the base (assuming not elite)

3. As a general strategy point I generally only road between my bases and not out to the frontier (exception-- there may be a road to any heavily crawlered area since the attacker has to plink through them to get on the road) . So to get on the road network to begin with the attacker has to have taken an outlying base (assuming not a seaborne invasion). In addition, most roads only connect rocky squares with crawlers on them (good ZOC blocks) or lead to areas swarming with formers and crawlers. I am former, crawler and sensor happy so my plan is to spot the enemy well in advance. If they are coming in force I would begin to rush reinforcements to the area and I would definitely start cutting roads. The attacker might cut a swath through some formers and crawlers but there should be a fully formed counterstrike force waiting by the time the attacker gate to the base. Then the attacker is faced with getting only one attack on the base per rover (one whole movement point to get next to the base)while each stationary defenders rover gets two hit back. If the attacker has sufficient might or techical superiority to win against those odds the defender is lost regardless of their base spacing.


I do not disagree that 3 square base spacing will allow a rover rush as you describe. I just do not think that any human defender would leave the roads there to be used by the attacker. The defender can use the internal roads for quick defense while the attacker has to slog tile by tile (scorched earth policy)


I'll have to consider the suggestion for defense rovers as it is a concept I have never fully considered. My typical homeland defense model is almost invariably based on the idea of early warning and attack rovers. My idea is that wherever an enemy appears, there is an attack rover to smack 'em. The concept is that at some point the attacker has to first move within two squares of the base so the defender gets the first attack. My hope is that (pre-air power) my base garrison will only rarely actually have to defend. Your use of a defense rover could enable use of rocky choke points (with bunkers or sensors) as defensible positions apart from base defense -- hmmmm might have to think if I can figure out ways to use that type of unit effectively.


Last point:
Most of the armour used on formers and crawlers is there mainly as worm defense (3r) when working near fungus. The extra cost seems to be worth it so that these units can keep working in the face of worm attacks. I do not like using formers/crawlers as defenders (generally do not like defenders-- period) but if sacrificing a unit or four saves a key base and allows time for a counterstrike then thats the strategy I will follow. Whether or not the formers/ crawlers get a crash upgrade depends on whether there is much chance they can survive the attack. Miriam once attacked with a swarm of laser rovers and the successful defence involved an upgrade of a couple of formers to 3r armour. The formers defended against Miriams first strike while mining a rocky tile. There was a monolith nearby for repairs and once a 3r unit and attack rover showed up that point was defensible until Miriam got impact weapons no matter how many units miriam threw at the point (She lost 14 I think while I essentially ignored the war).


We agree on many things but I remain unable to accept your points as a rationale for rejecting 3 square base spacing (maybe I just don't get it). As I considered the arguments it seems to me that close spacing is an advantage against conventional ground attack. I do not dispute your premise that rovers can take bases in the manner outlined. I do have a problem accepting the conclusion that therefore closer spacing is inadvisable as being less defensible.

Just some thoughts. Refutation encouraged since I tend to learn more when people point out where I am going wrong.

cbn

[Ogie Oglethorpe]

A couple of points to add to CBN's points. Skand your evaluation pros and cons is a good one but consider the following points CBN was stating.

Interior lines - Ability to move reinforcements faster. Outer bases tend to have roads spokingout to them but not leading away from them. Outermost bases equiped with sensors underneath allow for timely understanding of what is coming your way. Any attacker will by necessity need to expend its two movement point (except if they are elite units) in order to poistion themselves for an attack else you'llknow they are coming. If the outer squares are forest only then you actually have two turns to respond.

Reinforcements can come from nearest base three squares distant as well as the other two bases inthe 5 dot diamond approach. All of sudden thats 3 more units all upgradeable.

If a crunch is needed, speeders either best armored or best weapon can be supplied from bases 6 squares distant. Normally this should be more than enough to blunt the attack.

The defense model is a gamble tho'. What you've done is put up a point defense matching your best strength to his attack by using superior interior lines. If the defense fails just like in the game of RISK, a break through at key chokepoints hold dire consequences to the rest of your empire. IMHO tho' the risk is well worth it and ultimately employs the best use of your resources.

One should normally never count on formers and or crawlers to be your defensive forces. In a pinch tho' they can work very effectively. Tranceing a synth former works very nice round fungals and the situation described by CBN regarding mining a rocky road allows either a plasma former or a plasma crawler (ECM it for even better results) a very nice position. If covered under the sphere of sensor influence you've got a baby who can give those speeders their moneys worth.

One last thing. Tighter base spacing allows you many more sites for unit builds. Build Q's are a resource not to be underestimated. So in the case where you have purposely chosen to space your empire in a wider pattern your ability to turn your empire on a dime and start cranking out units (even if they need to be shell units for upgrade) is at a disadvantage vs. the more numerous base approach. If you secure your borders and recognize a threat theoutside bases should be much more reinforceable and with a greater amount of units due to the more numerous build sites and units streaming from them.

Just my ramblings tho'. Keep the ideas coming forward.

Trav, Good points on arty.

In my previous calcs I normally assume sensors under outermost bases thereby giving complete coverage of the squares under the base's sphere of influence in question without them being subject to arty destruction. (By the by all my calcs were done considering 8 squares within base sphere of influence i.e. workable but not the base square itself so base energy mins,pre former time for sensors etc. were not considered)

Sometimes inthe haste of plopping down bases sensors may not be built but wherever possible onthe perimeter of my empire I want to have sensors prebuilt onbase sites. AS a result sensor coverage normally can't be cut.

That being said then in order to cut reinforcement lines an arty unit neads to be brought up adjacent to the base site in order to cut the road leading to it from the other side. In this case the arty units is ripe for the counterattack as it is in the open. It however may be worth it to simply sacrifice the arty unit in order to prevent the reinforcements from arriving next turn. So a very good point.

Og

[This message has been edited by Ogie Oglethorpe (edited February 01, 2001).]

[cbn]

Good points Og and Trav

I think Og explained better most of what I was trying to say and I agree with his points on the difficulty of the artillery getting to the interior lines without sacrificing the artillery unit.


As for sensors, I probably build way too many of them. Almost all base sites get one and I put them out as far as I can as well as early warning detectors (and to get the +25% wherever the battle is joined). It may be a waste of former time in one sense but I look at it as an investment to protect my other terraforming and formers by almost always knowing whats coming.

Trav, you make great points on feint moves and the use of artillery. Both of these methods, used intelligently in the manner you suggest, can mess up any defensive scheme and I think your points are relevant regardless of the defenders base spacing. To try to combat such tactics my general playstyle is to have a very active outreaching defense. The usual process would be that the defender should know that "something" is coming even when the attackers are 6-7 squares out. The idea is to meet the attack at the edge of or even outside the base radius (pre air power). Outlying areas almost always have formers about to prepare base sites or crawler squares so there should be ample notice of any landward invasion.

I vary widely in the use of roads on the frontier but usually don't want any that could be potentially useful to an attacker unless I have a lot of military in the area to assure control of the road. But I have fiddled with building 1 tile roads in 4 directions radiating from the city. This ensures that my rover can get 2 smacks at units 2 squares from the base in that direction (second attack at -33%) or can hit once and run back into the city. With a non-elite rover attack there always MUST be a turn where the attacker sits either next to the base or two squares out.

I have also toyed with a rocky road strategy where there is a road into frontier but it is ZOC blocked by crawlers sitting on and working rocky squares. Depending on the weapon-armour ratio these crawlers can put quite a bit of damage on an attacker. I have only really considered this is situations with rovers available to patrol the road as well.


I have always thought that pre air power the active defender has many advantages. Apart from shorter supply lines and the sensor advantage, the prepared defender can sit and wait for the approach at a prepared site . Sometimes, if enough defensive attackers are marshalled, it is best for the attacker to back off. I have seen many situations where large numbers of ground forces were 3+ squares apart in a fairly open battlefield. the weapon to armour situation was disproportionate so that the reality is that the first side to advance ( into range) would take horrendous losses. Regrettably, given my hope for a smarter AI, the AI always advances into the slaughter. It reminds me of CIV 2 when the AI would bash itself senseless against a stacked mountain fortress.

[Travathian]

No one has mentioned artillery. Please keep this in mind. I can bombard your sensor from two squares away, before moving the rest of my forces in. Thus you can't see how much I am bringing. Better yet, I can blow up one on one side, and feint a different direction. Once I get close enough to another city, I bombard the roads connecting it to city I feinted from. Likewise, once a crawler is destroyed, I can use artillery to destroy terrain enhancements, thus saving my other units for killing your units. The net result being I can detroy more in less time.

This, of course, assumes I'm slashing and burning as I go. Assuming I cause a decent amount of damage (formers/crawlers/terrain) and you repel my attack, you are forced to decide whether or not to concentrate on defense of a potential invasion or rebuilding your infrastructure.

Previous posts (Ogie) shows various terraform times, add to that former and crawler creation times, and you're looking at quite an investment to rebuild. and every turn that goes by is huge losses of energy for you.

Ogie, you should also add into your figures the former time to create sensors. Given the close proximity, you shouldn't need that many. Then again, that also means taking out one, means taking out one for potentially 4 bases. That 25% is huge in the early game.

Just playing Devil's Advocate, I actually use this strategy in some games to great effectiveness.

[Skanderbeg]

cbn and Ogie!

Good points.

I think now that I have estimated a narrow base spacing to much dangerous.

This perhaps depends on my current game, where I (Drones) am beating two factions (Spartans and custom-made warmonger because of their narrow base spacing.
In one extreme case, bases were only two square apart, which allowed me to beat their garrison down with my artillery from the first base and easily capture the second.

I normally don't place my bases in strict pattern, but select places after ressources and strategic positions, normally 4 or 5 squares apart. But when I have no room to expand (starting on small island etc), I space them closer together.

One thing at last to the defense rover. Don't forget that this is a very valuable unit for attacks! It can give your best weapon/nor armor attack rovers and your artillery shelter from counterattacks, and it is the ideal unit to finally take and hold the captured base.
If I have broken the border defense line of an enemy, I often need only one single of this defense rovers and a bunch of needlejets/copters for attack to take out the AI's empire.

[Skanderbeg]

cbn and Ogie!

Good points.

I think now that I have estimated a narrow base spacing to much dangerous.

This perhaps depends on my current game, where I (Drones) am beating two factions (Spartans and custom-made warmonger because of their narrow base spacing.
In one extreme case, bases were only two square apart, which allowed me to beat their garrison down with my artillery from the first base and easily capture the second.

I normally don't place my bases in strict pattern, but select places after ressources and strategic positions, normally 4 or 5 squares apart. But when I have no room to expand (starting on small island etc), I space them closer together.

One thing at last to the defense rover. Don't forget that this is a very valuable unit for attacks! It can give your best weapon/nor armor attack rovers and your artillery shelter from counterattacks, and it is the ideal unit to finally take and hold the captured base.
If I have broken the border defense line of an enemy, I often need only one single of this defense rovers and a bunch of needlejets/copters for attack to take out the AI's empire, because on the roads, they can take a base six squares away.