Delhi specialist rejoice!

How do you like 6:15 sacred site capturing and 6:50 pro scouts (enemy side deer ferrying ) meanwhile 3d!bee like ghazi raider harassment into a lucifron style semi Fast Castle???

This playstyle and Build ORDER is very sensitive to map layout (needs be an open land map with standard size hunt) ; high APM and consistent macro.


  • The build starts off a cross between the Spanish bros and 3d!bee Delhi builds.
  • 5 on wood and 1 on mill; mill villager will make the mosque next; the 5 wood villagers after getting 50w go to berries
  • FIRST AND 2ND NEWLY CREATED VILS MUST GO TO GOLD; the first vil out makes the mining camp; the 2nd one instantly builds the house as NOT to be housed. The reason the first 2 MUST GO to gold is because that’s the only way you can ONLY have 2 villagers on gold and get 200g in time (2:25-2:30 feudal transitioning).
  • AFTER THE FIRST 2 CREATED VILLAGERS GO TO GOLD; all other villagers will go to berries until you have 11 on berries (note there is a way you can optimize this even more if you have the APM by only sending 8 to berries and all future villagers directly to sheep under your TC).
  • ONCE you reach about the 2:26+ timing you will FORCE ALL FOOD VILLAGERS; keep THREE (yes only THREE) on berries and pull 8 to slam down DOME OF FAITH!!!
  • Next villager created will go to berries to cap berries at 4 villagers top (NOTE: you’ll have to do 1-2 force drop as to NOT idle your TC and keep villager production constant)
  • Que 6 of landmark builders to stragglers and 2 to gold and rally TC to straggler trees.
  • THE MOMENT you HIT FEUDAL; cancel Piety and que sanctity (if you don’t have the APM just cancel piety early).
  • CANCEL WHEELBARROW; YES CANCEL IT. At that same moment you will take 1 of your berry push villagers to make a 2ND mill within influence while making the further berries more accessible. You will reque WHEELBARROW on that 2nd mill.
  • NEXT you need the very next 50w to NOT GET housed; AND you can use LOOM if you’re having idle TC issue to keep your TC working (unique benefit to delhi alone being able to start and stop research and not lose progression).

AT THIS POINT; its basically a standard DOME of FAITH 5+ scholar build while going balls to the walls ghazi raiders. The constant scholar production will make it so sanctity will complete around 6:10-15s (WITH YOUR SCHOLARS ALREADY ON THE SITE…) and a 6:50-7:00ish pro scout; which you’ll use 3-4 scouts to steal enemy 2nd and maybe even first hunt?!

NO WALLS NEEDED! With such an early site capture the enemy has to now decap your sites and catch your raids; WHILE you steal their food!!

A few disclaimers

using 4 villagers vs using 8 villagers to build a NON CHINESE landmark results in a gathering opportunity LOSE of 50 res for the 8vil build compared to the 4 villager build (this assumes a 0.66 avg resources gathered per sec).

HOWEVER; the 8 vil build ideally results 38 seconds faster feudal age up. I say ideally because often if not highly micro multiple villager building anything spend wasted time walking around the blueprint; so 8 vil VS 4 are more likelihood of wasting time walking around idle. STILL at its worse 8 villagers will at least NET 36s faster age up than 4 villagers and at most lose you 61 effective resources that could have been gathered.

YET the comparison isn’t just 36-38s FASTER at a cost of 50-61 less resources gathered; You have to factor in the faster research timings (which is 2 of the very things we’re taking advantage of) and the tempo. So its not improbable that 8 villager build vs 4 can result in a NET GAIN in total resources gathered ALONE; nevermind faster sites and tempo.

Full analysis of 8 vs 4 can practically result in even more INITIAL gathering opportunity loss than 50-61 resources b/c the walk time TO the landmark placement and FROM that landmark TO the next resource for the additional villagers were NOT factored in the above analysis. But say egregiously the total idle landmark walking for the 4 additional villagers totals 16 seconds!!! (16s x 4vil x 0.66res/s = 42 resources) THIS NOT SOMETHING TO IGNORE ( in the worse case scenario). So it is important to aware of where you’re sending your villagers to build relative to getting back to work.

Both against AI and againt real players i can consistently be on an open map sacred sites like Dry Arabia by 6:05-6:20. And except for rush civs like mongols english and french; most factions will NOT have many units, if any on the map to contest this ESPECIALLY if they went greedy for 2nd TC and alike.

Laslty while this build forfeits potential gathering of 60-80 ish resources by the 4min 15s timing (compared to a 4 villager landmark build); capturing the site a solid 60+ seconds sooner than spanish bro TOV build is NOT to be ignored?!!

I personally use this build for a fast castle; bc IFFF you can capture all 3 you’re guaranteed 1min of 3 site gold (375g/min), and in that case you’re forcing your opponent to respond and likely deviate from his original plans; and you’ll have a few units on the map to contest the decap and harrass; so the FAST castle will be unexpected if not scouted… and really how do you scout delhi is going fast caslte vs going for mass ghazi raiders??? Both require 2 billion vils on food. At most you’d have 4 villagers on gold and can get away with less if you can hold the sites longer than 60s.

Long story short

Is losing out on 80 resources (WORSE CASE… typical is more like 62 res) by the 4min 15s mark WORTH capturing sacred sites 60+ seconds sooner than normal spanish bros TOV build?

Next… let’s talk about berries.

According to patch notes delhi an abbasid gather from berries 30% faster than standard civ. Also listed on said patch notes is standard berry gather rate of 0.66 food per sec (i question how many decimal place does the standard rate really go?). So this means delhi and abbasid at min gather berries at a rate of 0.858 food per second and up to +3 carrying capacity.

BUT here are a few PROBLEMS gathering from berries that are NOT as egregious as collecting sheep under the TC.

  1. INITIAL DISTANCE from TC( or straggler tree) to the berry bushes.
  2. The RANDOM spawn orientation of the berry bushes.
  3. Which sides of the mill is “touching” the berries.
  4. How many villagers are on single bush.

Typical travel time from TC to neaerest berry bush is 6 seconds. Since the berry bush orientation spawns randomly you’re at the mercy of the RNG gods. Next the closest 2 sides of the mill that can “touch” berries are the SOUTH and WEST sides of the mill. And finally no more than 5 villagers can fit on one side of a berry bush before the next villager gets push to further away gathering point from the mill. Which leads to the optimal bush orientation is having TIGHT wedge formation where you can fit the mill directly against the bushes on the NE / NW side of the bushes.

Under the assumptions it takes 6 seconds to initially walk to berries; and that you get a nominal berry bush orientation (half good half bad results in 0.5 tiles of separation between bush and mill) AND for EZ math each alternative villager that goes to sheep spends 1 initial second to kill sheep (directly under TC). Turns out it takes the berry gather nearly 60s to CATCH up and surpass the alternative sheep gatherer directly under the TC.

THIS MEANS for a 2min 30s feudal build time the most villagers you can put on berries is 8 villagers. Once you go past 9 villagers on berries you will experience a LOSS in feudal timing (even with forced drop-offs).

I said all this to say, if you follow my OP build AND pull the 8 berry bush gatherers to do the feudal landmark building; you can optimize quick landmark placement and optimize return to work after the landmark is built by sending the majority of the builders back to berries!!! This will help you realize the lesser gathering opportunity loss of 60 resources for having 4 more builders than the usual.

Keep asking the question:
Is losing 60-80 resources at the 4min mark WORTH capturing sacred sites 60-80 seconds sooner than the Spanish bro TOV build?