Welcome to Part 4 of our series of articles on how to build a solo defense turtle account designed to take solo T3-4 marches. Part 1 was an introduction, Part 2 discussed the fundamental core of research for turtle accounts, and in part 3 of this series, we briefly gave some guidance on which buildings are and are not useful to keep in your urban buildings (inside your walls). Now we move to discussion of the three buildings of which you are able to build multiple copies, hospitals, barracks and villas, as well as the rural buildings, the farms / quarries / mines / logging camps.
Alot of the discussion in this article will be based around hospitals. The main reason for this is that they are absolutely critical for our ability to fight as a turtle. They are the core of the second asymmetry outlined in Part 1 of this guide:
You have hospitals, and the attacker doesn’t. Their troops die, yours don’t have to!
But before we get to a discussion of hospitals, lets first quickly discuss the rural buildings outside your city walls.
A turtle should only have 1 of each building of farm / quarry / mine / logging camp. That equates to 4 lots of 22k power each at sh18, or 4 lots of 74k power at sh21. Adding another 21 lots of rural buildings adds up to an additional 1.5M of power at sh21 that does not help your ability to be a lean mean fighting machine.
I know I shouldn’t need to say this here, but I am going to anyway, as I am constantly surprised by how many people are unaware of it. YOUR TROOPS DO NOT NEED FOOD TO SURVIVE. If you have a blinking red zero on food, they still march, attack, and more importantly, defend just as well as if they have food. If you want to have a small turtle army of around 500k T3, then it is possible to stay positive (or marginally negative) on food, and this can help when trying to have enough food to instant heal troops during battle, but if you haven’t already, learn to love the blinking red zero of negative food. Food just becomes another resource to be used for training, researching or building. I am convinced that there is an economist helping to design this game. When an economist gives you a free lunch, you should darn well take it. The way I think about it, MZ’s magical food elves deliver my troops millions of pieces of food per hour each and every hour, 24 hours a day, and I don’t have to do a thing. Produce food on a farm account, using hero gear that is dedicated to food (and silver) production. You will be much better off in the long run.
If you already have set up hyperfarming on your main, don’t despair, as they are the lowest power buildings in your build. It is completely up to you what you do with your farms / quarries / mines / logging camps. Personally I would slowly bump up a farm account specialising in that resource, and then when it is close to producing enough, then deconstruct all the excess buildings on the main turtle account, to try and save 500k-1.5M power that doesn’t directly contribute to fighting ability of your turtle.
|Fully upgraded building power
|75k||Farm, Mine, Quarry, Logging Camp|
|149k||HOSPITAL, Storehouse, Marketplace|
|187k||VILLA, Forge, Gymnos, Embassy|
|523k||Academy, Hall of War, Gold Mine|
|10 Million||Treasury Level 7|
This is an interesting way of looking at the relative worth of various buildings. MZ clearly values their money buildings (Treasury and Gold mine) more highly that just about anything else, closely followed by the other Money buildings, the T4 triumvirate of Prison, Altar and Hall of War. From a turtle perspective, and for the purposes of this article, we have highlighted the hospitals, barracks and villas in bold. And it is clear that barracks are twice the total power of hospitals. So if you train your troops, then switch out your villas and barracks for hospitals, you are increasing your effectiveness for taking hits, as well as slightly lowering your power (and hence improving your attractiveness to being hit). Double win!
Another interesting comparison is the total building power compared between Sh18 and Sh21. If you upgrade all your buildings at level 21, and have a copy of each of the single buildings, with an even spread of 5 hospitals, 5 barracks and 5 villas, and 1 each of the rural buildings, then your total building power is 7.34M. The equivalent setup at sh18 with all buildings at level 18 is 2.17M building power. If you torch all but the battle focused single buildings of watchtower, embassy and prison, and replace them with hospitals, then the sh21 version drops from 7.34M building power to 6.08M building power. The sh18 version drops from 2.17M to 1.79M building power. This is definitely a consideration when making a decision on which stronghold level to choose as your final destination as a turtle.
Because hospitals are the core part of asymmetry 2 (you have hospitals, your attacker doesn’t), this is the most important factor that should be taken into consideration when building a turtle.
In part 2 of this series our variants had some initial guidance on hospital bed capacity. Sh18 was 200-600 beds, and sh21 was 280-600k beds as a broad guideline. This was based upon a minimum of 7-8 fully upgraded hospitals for each level, and a maximum designed to capture 600k T1s hit by a stacked single troop attack with the best available hero gear and 375k T3-4s.
These decisions are driven by how many hospital buildings you will have, and how many beds each level holds.
|Hospital level||Bed capacity per hospital building|
So, if you have 8 hospitals, at sh18 that equates to 200,000 beds. At hospital level 21, that equates to 320,000 hospital beds. And the number of beds you will need depends upon your research level, your hero gear strength (particularly troop health), the number and level of your troops (Tier 1 vs 2 vs 3), and the troop attack stat of the incoming attacker’s hero.
As a turtle when taking hits, you want really good hospital bed capacity. The great thing about hospital beds is that they are an investment with a very good payoff over time. They cost a lot of resources initially, particularly as you get higher up, but they are relatively cheaper to build in terms of time cost. As a quick comparison, a level 21 hospital requires 17M stone as its largest resource, and takes 10 days to build before boosts. A barracks costs 8M stone as its largest resource, but takes 46 days to build before boosts. Most alliances find that speedups / time become their constraint, more so than resources, so in that setting, you should be able to get the resources you need to build more hospital beds, and speed-ups are less of an issue than for barracks.
The below table show the resource cost, time to build, power gain, and boosts from going from level 20 to level 21 of one each of hospitals, barracks and villas.
These three cells are highlighted in green for a reason. They are all very beneficial aspects of building a hospital. Faster build time (speedups are usually the main constraint), lower power gain per building, and a very important 5% health boost for level 21 hospitals (more on that later). But they are all secondary reasons. The main reason we want hospitals, is to save our troops when we get hit.
In the early part of the game, usually it is best to keep your troop numbers within your hospital bed capacity. In this instance, it is not uncommon to find 150k T3 troops fitting within 150-180k hospital bed capacity. This is great in the sense that you can fake rally 150k troops, and if they are at home when you are attacked, none die. But I can almost guarantee that if you have ‘only’ 150k T3 troops, when it comes to KvK, you will have all 150k of them hospitalized in one hit of a 375k march with a strong hero. If they hit you with 250k strat T4 and 125k T2 meat shield with a good hero, then you will lose on points every time. You will give up 3M points (150k x 20 per T3 hospitalised) whilst only killing 50-100k T2 (or less, depending upon hero). At maximum of 100k T2 killed, that gets you 500k points.
So, at some point, if you want to become an effective turtle and regularly take big hits, most people will need to move beyond their hospital bed capacity. There are of course exceptions out there, people with 500k T3, 100k T2 and a 600k hospital bed capacity, but it is damn hard work getting to 600k hospital beds. At sh18, that is 20 level 18 hospitals, and at sh21 it is 12 level 21 hospitals and one level 16 hospital. Here is an example of a turtle with 21 level 18 hospitals of 25k each, for a total of 525k hospital beds (thanks NoQ).
This particular turtle has chosen 21 level 18 hospitals, as well as 1 barracks, and 0 villas. They also have 1 academy, 1 forge and 1 embassy. The maximum number of hospitals you can have is 24 out of your 25 available slots. It is not possible to have a 25th hospital, as a hospital requires a barracks to build the hospital, so if you want to have 24 hospitals, you have a single building that either has no prerequisite, or only requires a rural building like a mine or quarry. As such, you can have a barracks (requires quarry), embassy (requires quarry), forge (requires mine), villa (requires logging camp and farm), or academy, marketplace, or storehouse, each of which only requires your stronghold. Importantly, you can’t have a watchtower as your 25th building, as it requires a barracks, and you also can’t have a gymnos (requires villa), prison (requires gymnos), altar (requires prison) or hall of war (requires embassy).
One of the meanest turtle setups I have seen was sh21, with 920k hospital beds (23 level 21 hospitals), a level 21 watchtower and a level 21 embassy. They constructed their 23 hospitals alongside their watchtower, and then deconstructed the final barracks and replaced it with a level 21 embassy. But that was in a turtle that had just over a million troops, so their risk was limited to the amount of troops above their hospital bed capacity. They were effectively immortal with 920k T2-3. If you have 23 hospitals (920k beds) and 5 Million troops, then it is psychologically very hard to recover from being zeroed, particularly if your hospitals are filled with T1s and all your T3 are dead.
If you are going to build more than 600k hospital beds, make sure you have a good match between troop type, troop numbers and hospital bed capacity. We are going to discuss sustainability of your build in more detail in a future guide, but ultimately it comes down to how important is it to you that you can’t be hurt by a momentary lapse that means that a full 2.375M T4 rally zeroes you in one hit.
Each villa adds 1% training speed for each level, and 5% for level 21. So that means that each level 21 villa adds 25% to your training time %. If you have 8 level 21 villas, that is 200% increment on top of your training boosts from completing training speed research (3 separate lots), setting your hero tree and wearing hero gear that boosts training time. There is a one-off 20% training time boost in the Combat research tree, after which you can research Training II in the Hero tree and Training III in the restorative tree. Training speed II MAY be worth considering, as it requires Hero Mastery level 4, and is pretty close to the Trap research that you have considered previously. Training III is NOT worth it for a turtle, as it is very far down the tree and most of the payoff is in the level 10 research.
But with villas, there is a diminishing return. Adding more and more villas will increase training % boost, but each one added is less valuable than the prior one. Setting your hero tree to max Troop training 1 and troop training II adds 122% onto your research training boost. Adding villas on top of that further increases your boost training %. But most people find that setting their hero tree on top of their research will REDUCE training time by around 60%. You can check this by clicking on a barracks, then More Information, then scrolling down to the bottom, and it will show you both your boost and your training time reduction. Some big T4 players with maxed research and the best permanent hero gear can get their training time REDUCTION to about 80%. So a massive load of resources and time goes into getting their training time reduction from 60 up to 70% and then even more in getting it from 70 to 80%. The sweet spot is around 70% time reduction (234% training boost), which I find is usually around 4 upgraded villas at level 18. Getting to 75% time reduction (300% training boost), requires an additional 66% boost, which is another 3-4 level 18 villas, or another 2-3 level 21 villas. Getting to 80% time reduction (400% training boost), requires another 4 level 21 villas, so to go from 70 to 80% time reduction requires an additional 6 full level 21 villas and a level 15 villa, or at least one piece of hero gear with a 15-20% training boost.
The number of villas you need will ultimately be determined by the number of troops you want to train. A turtle should not look to hyperproduce silver. Do not be tempted to have lots of villas for silver production on a turtle. It makes you vulnerable to being zeroed if you ever drop your shield and you are wearing silver production hero gear. Again, leave silver production to a farm setup where it does not matter if the hero tree is permanently set on silver and food production, and the hero gear likewise boosts silver production.
Maximum barracks capacity for a turtle is 5,000 at level 21, but only 2,500 at level 20, ie it doubles at its last upgrade from 20 to 21. It can go up to 6,000 if you complete the training capacity research in the Restorative research tree, but that is not worth it for a turtle. You can have multiple barracks, and there is a complex interaction between villas and barracks that determines your training rate, as discussed above.
There are two general ways to approach barracks, the regular consistent trainer, and the spender who speed trains troops to maximize returns from inferno and KvK training events. The first path is usually taken by people who have the patience to train a single batch of troops each and every day, whether they be T1, T2 or T3, and have limited speed-ups or gold to use for speed-training. The usual approach here is to have 1-3 barracks, but then have lots of villas to make the training as efficient as possible in terms of their main constraint, time. The other approach aims to get to a stable point where you can train a defined number of troops per defined period of time, usually 8 hours, 15 hours, 24 hours or 3 days, as these are the common speed-ups used for speed-training troops during events. And it will differ quite markedly by what tier of troop you are training, because of the differing training times of the different tiers of troops.
If you have 5 level 18 barracks, then your training capacity is 10,000. Move them all to level 21 and all of a sudden you have 25,000 training capacity per batch. And the number of barracks you will have will also be determined by the tier of troops you are going to train.
In the below table, we are going to illustrate a variety of different scenarios for the number of barracks for T1, T2 and T3 troops, for the number of barracks required to train troops in 24-hour or 3-day batches. Note that the assumption of 5,000 T3 per 24-hour period is based upon a roughly 70% training time reduction. This table can be useful for both those who use speed-ups for training event purposes, as well as those who train a batch of troops every day.
The colour scheme is 0-5 barracks colored as green, 6-10 as orange, and >10 as red. Note that if you want to mass produce T1 troops with a 3-day speed-up, it can only really be done as an Sh21. The problem with that is that if you have mass-produced T1s, then you will definitely need 600k hospital beds, which is 15 level 21 hospitals. 12 barracks and 15 hospitals does not fit into 25 available urban buildings, so you will have to mass-produce T1, then convert the barracks (and villas) across to hospitals.
The above table is based upon a roughly 70% training time reduction. This is a midpoint between players with minimal villas and research (around 60% training time reduction) and those with lots of villas, all the available training and great hero gear for training (around 80% training time reduction). If you want a more personalised view of training time, it is highly recommended to visit the Troop Training calculator, which has some really nice updates as outlined in this post and accompanying video walkthrough.
If your training is faster because you have more villas or great hero gear that boosts training speed, ie a larger training time reduction, then you will need MORE barracks capacity. So if you have 75% training time reduction, it may increase the number of barracks required by 1-2 compared to the above table, depending upon the situation.
At right we have a screenshot of an sh18 turtle that is setup to speed-train T2. As per the table above, it has 15 level 18 barracks for a total training capacity of 30k, along with 4 level 18 villas, giving it a training time reduction of around 70%. It trains just under 30k T2 per 3-day speedup, and is aiming for 2-3M T2, so will require around 60-100 lots of 3-day speedups, ie 1 gold pack’s worth of 3-day speedups. It will then need to deconstruct some (or all) of those 15 barracks and replace them with hospitals. The same result can be achieved by the consistent trainer who trains a batch of T2 every day, but at 10K T2 per day, it will take 200-300 days to train a full 2-3M T2 army. As to why someone would want to train T2 exclusively, that will be discussed in the next article on troop types.
Most people train T3 almost exclusively once they hit academy level 15 and unlock T3 troops, and the default setup that a lot of accounts have is a roughly even split of 4-6 barracks and hospitals and villas. In this situation, 4-5 barracks gets up to 10,000 T3’s trained at level 18, which for most people takes around 2 days. So you can quite happily train 5000 or so T3 per day and in a couple of months get to 300k T3 or so, assuming you can not lose any in the mean time. And you can sequentially build more and more hospital bed capacity to match your troop number increase up to around 300k beds. So the organic way of growing without spending and using large numbers of 24 hour and 3-day speedups, is to keep training T3 exclusively in moderate sized batches every day or two, whilst progressively upgrading hospital bed capacity. I have seen people get to 600k T3 and 600k beds after 3-4 months of steady work, farming, training and building. This is the main reason that people who choose to train a batch each and every day (and never miss a day), tend to focus the core of their build on T3 troops, as they can build a very decent army in 3 months, whilst it can take alot longer to finish building a T2 (or T1!!) army by training a single batch each day.
So, after 2 long articles discussing buildings, I can hear you asking: what is the ideal composition of your urban buildings? Sorry, but there is no ideal composition. I can give you some guidance, but you ultimately have to make the decisions for your account. Minimize the single buildings, eliminating ones you simply don’t need. Torch your storehouse, torch your marketplace, torch your hall of war once you stop fake rallying, torch your altar (or never construct one), torch your academy once you have finished your turtle combat research, torch your gymnos once you are hero level 50, torch your forge once you have crafted the permanent hero gear that you want. Leave a watchtower, an embassy and a prison, as these all help you fight. These are the core features of turtle builds that should be similar because the underlying principle of what buildings contribute to fighting and which ones don’t contribute to fighting are the same for all turtles. Oh, and stick to only 4 rural buildings outside your walls, one of each type.
Beyond that, there is no straightforward guidance, because the composition of hospitals barracks and villas will vary quite markedly depending upon how you decide to structure your army. I can guarantee that the early composition of hospitals barracks and villas will be quite different for someone looking to build a 600-700k T3 army than it will for a turtle looking to train 8M Tier 1 troops.
As will be discussed in the next article, if you have exclusively trained T3 troops, you will not need 600k hospital beds to take a solo hit. 600k hospital beds can help make your T3 account ‘immortal’, and they can help if you are taking multiple solo hits at once, but they are not absolutely necessary for a single solo hit with current permanent hero gear. However if you have any T1 meat shield, then you may need up to 600k hospital beds. So ultimately, your hospital barracks villas composition is driven by your decision on how many and what tier of troops you are going to train.
The next article will FINALLY get you to a topic you are really interested in: which troops to train.
And most importantly, it will be a discussion of how best to leverage the most important defensive asymmetry: your attacker has a limited attack size with a maximum allowed march size of 375k (using a 50% march size boost), whilst you can have an UNLIMITED number of troops defending your stronghold.
We will discuss a number of different ways you can structure your defensive turtle army, and your decisions on what kind of army you want to train will help a lot with the decision on how many hospital beds and barracks and villas you need to reach your goals.
The final part of this series will give you an executive summary of all of these research, building, training elements, and try and provide some practical guidance on what we recommend. But coming back to one of the key points from Part 1 of this series, you need to make some tough decisions along the way. This guide will not tell you what to do, as there are so many different ways to build a turtle account that successfully takes solo T4 marches and gets regular Kill Event gold, but hopefully will provide some guidance as to ways the various elements can be tied together in a way that fits in with how you want to play the game.
I hope you are looking forward to the next part, the discussion of how to explore options for the composition of your lean mean fighting machine. So far we have deliberately tried to minimise power gain to places where it really is necessary. The main reason for this is to allow as much space for troop power to fit inside a defined power envelope of 30-40M total power. It’s time to go shopping and see what we can buy for 15-25M troop power.
- Guide to Turtle Accounts in Game of War – Introduction (Part 1)
- Guide to Turtle accounts in Game of War – Research (Part 2)
- Guide to Turtle accounts in Game of War – Buildings (Part 3)
- Guide to Turtle accounts in Game of War – Hospital Barracks Villas (Part 4)– You are Here
- Guide to Turtle accounts in Game of War – Troops (Part 5)
- Guide to Turtle accounts in Game of War – Armies (Part 6)
- Guide to Turtle accounts in Game of War – T4 follow-up (Part 7)
- Guide to Turtle accounts in Game of War – Strategy (Part 8)
- Guide to Turtle accounts in Game of War – Wall Traps (Part 9)