# Master Book in Minecraft: Why Merging All Books First Costs More

_Published October 23, 2026_

Why combining every enchantment into one book before applying it usually costs more XP and leaves more prior work than the cheapest order, with solver-checked numbers and the cases where it is fine.

The "master book" is one of the most repeated tips in Minecraft enchanting: merge every enchantment you want into **one big book first**, then put that book on your item **once**, so the item stays "clean". It sounds tidy, and it is not a crazy idea, but as a general rule it is a myth. It usually costs **more XP levels than necessary**, it can cost so much in its last step that the anvil refuses it, and the finished item still ends up with a **high prior work count**. This guide shows why, with numbers from AnvilCalc's own solver, and also where the master book is perfectly fine.

## What the master book claims

The plan has two phases. First you build the book: take two single-enchantment books, merge them, merge the result with a third, and keep going until one book holds everything (this is usually a **linear chain**, one book into the next). Then you apply that book to the item in a single anvil operation. The appeal is real: the item only goes through the anvil once, so you would expect it to carry almost no prior work, and the book is "just a book".

The catch is that the prior work does not disappear, it moves onto the book, and the book carries it into the item.

## The mechanics that break it

Everything below follows from the Minecraft Wiki's [anvil mechanics](https://minecraft.wiki/w/Anvil_mechanics) page and from the same rules the calculator uses (checked on 2026-10-23).

- **Every merge adds one anvil use to the book.** A book that has been through an anvil k times carries prior work k, and its next operation pays a penalty of 2^k − 1 levels (0, 1, 3, 7, 15, 31, ...). In a linear chain of n books, the finished book has prior work n − 1.
- **The result takes the higher of the two sides, plus one.** The wiki's example: combining two items with two prior work penalties each gives a result with three. The penalties of **both** sides are added to the cost, but only the higher count carries on. We tested this in the solver for both Java Edition 1.21.11 and Bedrock Edition 1.19.1: a book with prior work 3 plus a book with prior work 1 gives 4, and two books with 2 each give 3. The wiki states the rule without an edition qualifier, and the calculator applies it in both editions.
- **So the item ends at the book's prior work plus one.** A linear chain of n books leaves the item at prior work n. A perfectly balanced tree of merges does better, ceil(log2 n) for the book and one more for the item, but that is still never lower than the best possible ceil(log2(n + 1)) from the [Minimum Prior Work](/guides/minimum-prior-work) guide, and it is higher for 3, 5, 6 and 7 enchantments.
- **The last step pays the penalty of the whole book.** If the finished book has prior work 3, applying it to the item costs the enchantments' value **plus 7 levels** of penalty, all in one operation.
- **Enchantment values are paid twice.** Every book you merge into the master book pays its enchantment cost at that merge, and the master book pays it again when it is applied to the item. The wiki's boots example shows the effect: pairing books makes you pay some enchantment costs several times, and its plan that avoids part of the pairing is 2 levels cheaper (66 against 68).
- **One expensive final step can hit the cap.** A single operation of 40 levels or more is refused ("Too Expensive!"). Putting everything into the last step is exactly how a master book reaches it (see [Why Does My Anvil Say Too Expensive](/guides/anvil-too-expensive)).

Here is a pickaxe made this way, to see the numbers. Target: diamond pickaxe with Efficiency V, Unbreaking III, Fortune III and Mending, every enchantment on its own fresh book (Java Edition 1.21.11).

| Step | Master-book way (best order of the chain) | Levels |
|---|---|---|
| 1 | Fortune III book + Efficiency V book | 5 |
| 2 | that book + Unbreaking III book (book prior work 1, penalty 1) | 4 |
| 3 | that book + Mending book (book prior work 2, penalty 3) | 5 |
| 4 | pickaxe + the master book (book prior work 3, penalty 7) | 23 |
| | **Total** | **37**, pickaxe at prior work 4 |

The calculator's cheapest order for the same pickaxe costs **23 levels in total**, the most expensive single step is 9, and the pickaxe ends at prior work 3. It does not build a master book: it merges only Efficiency V with Mending on the side, and puts everything else on the pickaxe one book at a time while its prior work is still low.

## The same item, four ways

Each build below starts from a fresh diamond item and gets every enchantment from its own fresh single-enchantment book. We computed four plans with the solver: the master-book chain in its **best** order (every possible order was tried), the **cheapest possible master book** of any merge tree shape (a balanced tree, with every left and right order tried), the calculator's **Min XP** plan and its **Min prior work** plan. Each cell reads: total levels / most expensive single step / prior work of the finished item. The numbers are identical on Java Edition 1.21.11 and Bedrock Edition 1.19.1 for all builds on this page, because no enchantment is added twice.

| Build | Master book, linear chain | Master book, best balanced tree | Calculator, Min XP | Calculator, Min prior work |
|---|---|---|---|---|
| Pickaxe: Efficiency V, Unbreaking III, Mending | 19 / 13 / 3 | 19 / 13 / 3 | 14 / 9 / 2 | 14 / 9 / 2 |
| Pickaxe: those plus Fortune III | 37 / 23 / 4 | 33 / 19 / 3 | 23 / 9 / 3 | 23 / 9 / 3 |
| Sword: Sharpness V, Looting III, Unbreaking III, Mending, Fire Aspect II | 60 / 35 / 5 | 48 / 27 / 4 | 31 / 13 / 3 | 31 / 13 / 3 |
| Sword: those plus Knockback II | blocked | 55 / 29 / 4 | 38 / 14 / 3 | 38 / 14 / 3 |
| Boots: Protection IV, Feather Falling IV, Depth Strider III, Soul Speed III, Thorns III, Unbreaking III, Mending | blocked | blocked | 66 / 23 / 4 | 68 / 22 / 3 |

"Blocked" means that in every possible order and every possible merge shape, at least one anvil step costs 40 levels or more, so no master book can be made without Creative mode. The calculator finishes the same boots in steps of at most 23 levels.

What the table shows:

- **The chain is the worst version.** With five enchantments the sword chain costs nearly twice what the calculator needs (60 against 31), and the item ends at prior work 5, against 3.
- **Even the best master book loses.** Merging the books in the cheapest tree shape still costs 19, 33, 48 and 55 levels in the four builds it can make, against 14, 23, 31 and 38 from the calculator.
- **A master book can only tie, never win.** A master book is one particular plan out of all the plans the solver searches, so the calculator's cheapest order is never more expensive than the best master book, and it is exactly equal in the already-worked pickaxe case below.
- **Min prior work trades a little XP for a lower count.** On the boots it costs 2 levels more (68 against 66, the same two numbers as in the wiki's own boots example) and ends at prior work 3 instead of 4. See [Minimum Prior Work](/guides/minimum-prior-work) for when that is worth it.
- **The order of the chain still matters.** The worst order of the pickaxe chain costs 41 instead of 37, the worst order of the sword chain 64 instead of 60 and the worst order of the three-enchantment pickaxe 22 instead of 19. The wiki's advice is that the costliest enchantment should be in the sacrifice slot as few times as possible, and the left and right slots are not interchangeable: a Soul Speed III book on the left with a Mending book on the right costs 2 levels, and the reverse costs 12.

[Compare this sword in the calculator →](https://anvilcalc.com/#eyJpIjoiZGlhbW9uZF9zd29yZCIsImUiOltdLCJwIjowLCJuIjoiIiwidCI6W1sic2hhcnBuZXNzIiw1XSxbImxvb3RpbmciLDNdLFsidW5icmVha2luZyIsM10sWyJtZW5kaW5nIiwxXSxbImZpcmVfYXNwZWN0IiwyXV19)

## When the master book is fine

A fair comparison has to include the cases where it is not worse. Two of them, from the same solver (Java Edition 1.21.11 and Bedrock Edition 1.19.1, same numbers):

| Build | Master book, linear chain | Master book, best balanced tree | Calculator, Min XP | Calculator, Min prior work |
|---|---|---|---|---|
| Fresh pickaxe: Efficiency V, Unbreaking III | 12 / 9 / 2 | 12 / 9 / 2 | 9 / 5 / 2 | 9 / 5 / 2 |
| Pickaxe that already has prior work 3: add Efficiency V, Unbreaking III, Mending | 26 / 20 / 4 | 26 / 20 / 4 | 26 / 20 / 4 | 26 / 20 / 4 |

- **An item that already carries prior work.** Every operation on the item pays its penalty (a pickaxe at prior work 3 pays 7 extra levels each time), so putting as few operations on the item as possible is the right idea. Here the calculator itself picks the three-book master book, and all four plans cost the same 26 levels. If your item is already heavily worked, a master book is what the calculator tells you to do anyway.
- **Very few books.** With two enchantments the master book costs 3 levels more than adding them one at a time (12 against 9), which is small, and the plan is simple. It is worse, but not by much.
- **You want the item untouched until the end.** Building the book first leaves the item at prior work 0 until the final step. This only matters if you have a reason to keep the item out of the anvil (a rename or repair in the same last step, say); it does not make the result cheaper.

If you do build a master book, build the **cheapest book side**: merge the books in balanced pairs instead of one long chain (book into book, then pair into pair), keep the most expensive enchantment out of the sacrifice slot as often as possible, and check each step against the 39-level cap before you start.

## How to check your own plan

1. Open the [calculator](/), pick your item and enchantments, and read the plan with **Min XP** (the default). The plan names its total, each step and the prior work the finished item ends with.
2. Switch to **Min prior work** to see the plan with the fewest anvil operations on the finished item. Often both modes give the same plan, as in most rows above. See [How to Use the Calculator](/guides/using-the-calculator).
3. Compare with your own master-book plan: add up the levels of each step as the anvil shows them. If the calculator's total is lower, follow its steps; each step says which slot the item or book goes in, as in [How to Use an Anvil](/guides/how-to-use-an-anvil).
4. If you already own some books, add them to your [inventory](/inventory): a book that carries several enchantments at prior work 0, such as one from [fishing](/guides/fishing-enchanted-books), is the legitimate version of the master-book idea, because it never went through a merge.

For the background on the penalty itself see [Enchanting Strategy](/guides/enchanting-strategy), and for pricing the single-enchantment books see [Cheapest Enchanted Books](/guides/cheapest-enchanted-books).

To decide which item to build first, see [Minecraft Enchanting Order](/guides/minecraft-enchanting-order).

## Frequently asked questions

**Is a master book cheaper in Minecraft?**

Not in our tests. For a fresh diamond pickaxe with four enchantments from fresh single-enchantment books, the best order of a linear master book costs 37 levels and the cheapest possible master book 33, while the calculator's cheapest order costs 23. With five enchantments on a sword it is 60 and 48 against 31. A master book is only one of the plans the calculator considers, so the calculator is never more expensive, and the two tie when the item already has a lot of prior work.

**Why is my master book result too expensive?**

Because the last step pays the whole book at once. Every merge raised the book's prior work, so applying it adds that penalty (7 levels for a book with prior work 3, 15 for prior work 4) on top of the value of every enchantment on it. Operations of 40 levels or more are refused as Too Expensive. In our sword example with six enchantments, no order or merge shape of the master book stays under the limit, while the calculator's plan never needs a step above 14 levels.

**Does the order of the books inside a master book matter?**

Yes. The cost of a book merge depends on which book is in the left slot and which is the sacrifice on the right. According to the Minecraft Wiki, a Soul Speed III book on the left with a Mending book on the right costs 2 levels and the reverse costs 12. In our chains, the worst order cost 3 to 4 levels more than the best one (41 against 37 for a four-enchantment pickaxe). Keep the costliest enchantment out of the sacrifice slot as often as possible.

**How much prior work does an item get from a master book?**

The result of a combine takes the higher prior work of the two sides plus one. A linear chain of n books leaves the book at n minus 1, so the item ends at n. A balanced tree gives the book ceil(log2 n) and the item one more. Neither beats the best possible plan, which is ceil(log2(n + 1)) for n enchantments on separate books, so a master book leaves the item at a higher count for 3, 5, 6 and 7 enchantments.

**Can I reduce prior work after the fact?**

Only by throwing the enchantments away too. According to the Minecraft Wiki, a grindstone removes an item's prior work penalty but also all of its enchantments (it keeps the custom name), and repairing an item in a crafting grid removes the penalty, the enchantments and the name. There is no way to remove prior work and keep the enchantments, so the order of combines is the thing to get right.

**When is a master book a good idea?**

When the item already carries a lot of prior work, because every operation on it pays that penalty. For a pickaxe at prior work 3 that needs Efficiency V, Unbreaking III and Mending, the calculator's cheapest plan is the three-book master book itself, 26 levels, the same as every other plan in our test. For a fresh item with only two enchantments it costs about 3 levels more than the direct way. With many enchantments, merge the books in balanced pairs or let the calculator plan the order.

**Does the master book work the same in Bedrock Edition?**

The same prior work rule applies, and every comparison on this page gives identical numbers on Java Edition 1.21.11 and Bedrock Edition 1.19.1, because none of the builds applies an enchantment to an item that already has it. The two editions only differ in how they price a level that does not rise or a conflicting enchantment, which does not come up here.