0 AD Strategy Discussion

[Sanguivorant]

Seeing how a lot of threads on this forum are about the different strategies you can perform in 0ad, I believe it would be okay to have one thread where we can talk strategy.

Feel free to post all your tips/tricks/build orders here.

To start this off, I do have one thing we can talk about: Batch training. Is it useful in the early game?

In the late game, it obviously has a benefit of allowing you to create more units in a short amount of time.

However, in the early game, having resources is essential. So comparing batch training to individual training of female villagers for example, I think you end up with much more resources if you train the female villagers individually. What do you guys think?

[bouke]

I usually do batch training: for every batch you get time-wise more then 1/5 woman free. With just training in batches of 5. I could calculate the exact result of both (individual or batch). I think it will favour batch training at every time.

[bouke]

After some calculation I can conclude that training woman constantly (so there is enough food to make woman) from 1 cc in batches of 5 is better from around 85 sec instead off producing them individually. In the calculation the gathered good was food from fields (gather speed 0.5) and no walk time to dropplace (which is possible) and no time for building the farm (this should not matter in my opinion, because 5 woman build a farm 5 times as fast as 1 woman).

I made the calculations with a model in "scydynamics". I would share it with you, if a could upload it. So is it possible to upload an ".xml" in the forum?

Edited December 17, 2014 by bouke

[Lion.Kanzen]

After some calculation I can conclude that training woman constantly (so there is enough food to make woman) from 1 cc in batches of 5 is better from around 85 sec instead off producing them individually. In the calculation the gathered good was food from fields (gather speed 0.5) and no walk time to dropplace (which is possible) and no time for building the farm (this should not matter in my opinion, because 5 woman build a farm 5 times as fast as 1 woman).

I made the calculations with a model in "scydynamics". I would share it with you, if a could upload it. So is it possible to upload an ".xml" in the forum?

you can upload it as zip file.

[zzippy]

... because 5 woman build a farm 5 times as fast as 1 woman

Erm, no. The more workers you add, the less it decreases build time (see the tooltips when hovering the hammer symbol of a building in construction).

For a perfect eco its better to use only 1 worker/building (if there is no need to hurry ) ...

Edited December 17, 2014 by zzippy

[niektb]

Can I ask you how you calculated it? My calculations show me that training 5 female woman in batch is less beneficial resource-wise than training them individually:

My calculation stops when both methods have produced 5 women, so with time = 40 sec. Walking time doesn't need to be taken into account here since the first woman has 32 sec to gather which doesn't fill her capacity (16 out of 20 would be filled). I also assumed that the farm was already present.

Adding multiple batches does nothing but stacking these amounts (except possibly introducing walking time).

Edited December 17, 2014 by niektb

[bouke]

First off all the batch training time for woman is 5.8 per woman not 6.4 (29 for a batch of 5 so 29/5=5.8).

Second off all multiple batches does make difference because with batches you have 5 woman in 29 sec, without in 40 sec, so in 1 min, with batches roughly 10 woman without 7.5, so in time you get more woman so more gather power --> more resource.

Erm, no. The more workers you add, the less it decreases build time (see the tooltips when hovering the hammer symbol of a building in construction).

For a perfect eco its better to use only 1 worker/building (if there is no need to hurry ) ...

sorry did not know about this

[niektb]

Wait, the build time of 5 females is 40 sec and the batch modifier is 0.8. 0.8*40=32 not 29.

Sorry, overlooked the time bonus for consecutive batches.

Edited December 17, 2014 by niektb

[bouke]

The game really says 29 seconds for 5 woman from a cc... so the batch modifier is 0.725 for a 5 batch, for a 10 batch 0.6375 (51 sec) and 0.5833 for 15 (70 sec)

I might be a bit blind, but I cannot find the attach button...

Edited December 17, 2014 by bouke

[thamlett]

What I do first is even train ~40 women to farm 8 farms around my CC and perhaps ~50 to chop wood (they first are split between metal and stone). I first even out the starting soldiers to 5, and use the five melee to build etc.

Then I keep on building up the ranged units, garrisoning them until I have an invasion force. That way, if I am attacked during this time, I will already have powerful defenses and the women can just garrison in houses until the invasion is over.

I try to focus on ranged units that only use wood and food, since they are plentiful in most maps.

I do all of the ranged-unit upgrades in the blacksmith, and preferably do the "Will to Fight" right before my invasion. I also research battlefield health - and seriously injured ranged units come back to base to be garrisoned until they are healed (the battlefield health affects garrisoned units, and while they are garrisoned they can't lose health.

@bouke

Under "More Reply Options"

[bouke]

@bouke

Under "More Reply Options"

Thanks a lot!!!

Here my model for calculating the gathering of resource:

It is build in Scydynamics, best link: http://web.usgym.nl/muusse/nlt12.htm then the on line 7 Scydynamics programmatje (sorry for the Dutch). It will download, keep and open. It is running on java. If it is not running you should set the security settings from java as low as possible.

Note: The train_time should be the same as your timestep (so 8 with one by one and 29 with batches of 5).

batch training0ad.xml.zip

Edited December 17, 2014 by bouke

[Sanguivorant]

Assuming no walk times for the sake of simplicity:

Assume gathering 1 food/second.

A female villager takes 10 seconds to be built individually and 38 seconds to be built in a batch.

You start training:

In the first 10 seconds, nothing happens, because female villager is being built.

In 20 seconds, the female villager gathers 10 food.

In 30 seconds, the female villagers gather 20 food.

In 40 seconds, the female villagers gather 30 food.

In 50 seconds, the female villagers gather 40 food.

So in a total of 50 seconds, a total of 100 food is gathered. Though, only four villagers are created.

In batches:

In 10 seconds, nothing happens.

In 20 seconds, nothing happens.

In 30 seconds, nothing happens.

In 38 seconds, five villagers are trained and start gathering in total 5 food per second.

There are 12 seconds left.

So in 50 seconds, the batch gathers a total of 60 food.

So this leads me to believe that at least in the beginning parts of the game, it is more beneficial to train economic units individually instead of in batches.

Edited December 17, 2014 by Sanguivorant

[Peregriino]

I think it's depend of situation and how/when.

For economy in early game the individual are the best like explained above.

For Military have differences, if you under attack maybe the batch make more sense, because one soldier make nothing in some cases. If you make a army for attack, mayve batch are more interesting.

But I try use other way, I make 5 barracks and tranning infantary individually, it's more fast. The problem is a lot of resources for it.

[Sanguivorant]

How about walls. Do any of you use walls?

I play as Iberians a lot, so I always have a layer of my economy protected by walls.

I find that walls are the best on maps with choke points, but other than that I don't think I use them.

I do like to structure my base in a way that creates a natural wall that protects segments of my economy. So I build houses to surround my civic centre so that my farming is somewhat protected.

With the threat of melee cavalry, I like to palisade/wall my towers and surround them in houses so that it is harder for cavalry to take them down.

[Alekusu]

Hey guys,

The thing is you can't consider only the first 40sec. Because yeah you'll get more food in 40 sec but not enough to invest into anything.

I consider that:
- you can produce non-stop batches of 5 villagers
- gathering food is 0.5f/s
- time to create 1 female (1 by 1) = 8s
- time to create 5 female (batch) = 29s

The first table is the results when you create females 1 by 1
The second table is the results when you create females 5 by 5

What you see is that after about 90sec, creating batch is more efficient (more food gathered and more villagers created). And the extra ressource you get by creating units 1 by 1 in these first 90sec are not enought to invest into anything. so 5by5 >>> 1by1.

Please tell me if I made a mistake
Alex

[niektb]

I figured out how the batch buildtime is calculated:

To conclude:

Time is what matters with batch production. you will be able to train units faster (on average). Delay between training two batches nullifies the advantage of batch production. So Training females in batches is beneficial resource-wise when you train multiple consecutive badges directly after each other (at least three).

The aforementioned delay can for example be caused by a lack of resources or the pop cap.

Also individual training requires a little more micro work (since you need to task a female every 8 sec instead of 5 at once per 32 sec)

I think that the early game preference for batch or no batch depends a bit on the playing style. If you build very much within the first minute (with Ptolemies) it might be good to train the first few females individually to maximize early building speed. Overall batch production has the edge though.

Edited December 18, 2014 by niektb

[greenknight32]

The way I look at it, you've got many things to do - create units, construct buildings, scout - and seconds count early in the game. Anything you can do to save a few of those seconds is good. Set the CC to make a batch of units, set the rally point on a resource so they go start work there, then do something else for a few seconds until it's time to do it again. Line up a team to build a batch of houses as soon as you have enough wood. Hurry, hurry, hurry, that's the only way to win.

Getting a tiny bit of a resource a few seconds sooner is not worth the extra micro - if a scout gets killed because you didn't get back to him soon enough, it wastes much more than you gained. Maybe the first few females, like niektb said, but otherwise it's better to streamline your process as much as possible.

[bouke]

I think it's depend of situation and how/when.

For economy in early game the individual are the best like explained above.

With my and Alekusu calculation (witch has exactly the same results), We proofed that batch training is an investment, where you get profit after about 90 sec, I think that is the investment worth, also with the less micro you need to do.

Still then is the question, if it is any good to go from batches of 5 to batches of 10.

[Peregriino]

I think it's just math, in real game don't work this form, because you don't have resources to make houses, and foood in somecases.

@On Topic

I just use walls in geografics acidents like valley between mountains.

[Alekusu]

With my and Alekusu calculation (witch has exactly the same results), We proofed that batch training is an investment, where you get profit after about 90 sec, I think that is the investment worth, also with the less micro you need to do.

Still then is the question, if it is any good to go from batches of 5 to batches of 10.

The thing is the batch this is good if you create batches of villagers non-stop. It would be impossible to create villagers non-stop 10 by 10 at the beginning of the game so I think the best is still 5 by 5

[bouke]

True, but is there a moment to change?

[Sanguivorant]

I guess the best moment to change is once you have stabilized your food income, and manage to gather at least 250 food every 35 seconds or so.

[Alekusu]

True, but is there a moment to change?

Probably when you reach age 2, if you are going eco you can probably create 3 or even 4 batches of 10 villagers non-stop...

But usually online if you do that you'll be killed by someone who plays agressive. For me the best is still 5by5, it's not a big investment and it's easy to manage.

Edited December 18, 2014 by Alekusu

[Peregriino]

And about get Phase 2. How many time is a good for get Phase 2? And make first CC?

I can get Phase II in 8~10 min.And CC in 12~14 min.Playing with Romans. I think it's slow, I want more speed.

[Alekusu]

It really depends on your strategy/the map/if the ennemy rushes or not, so it's difficult to give good timing but reaching age 2 in 8-10min is good.
About expansions it's in my opinion not possible to give any good timing especially if you play multiplayer games because there will be so much action from 12/13min online and your first CC's timing will depend on it.

For me expansion at 12min can be a really risky investment on multiplayer games, if you play against someone that scouts the map this CC could be your death.