Solidity library for advanced fixed-point math that operates with signed 59.18-decimal fixed-point and unsigned 60.18-decimal fixed-point numbers. The name of the number format is due to the integer part having up to 59/60 decimals and the fractional part having up to 18 decimals. The numbers are bound by the minimum and the maximum values permitted by the Solidity types int256 and uint256.
- Operates with signed and unsigned denary fixed-point numbers, with 18 trailing decimals
- Offers advanced math functions like logarithms, exponentials, powers and square roots
- Provides type safety via user defined value types
- Gas efficient, but still user-friendly
- Ergonomic developer experience thanks to using free functions instead of libraries
- Bakes in overflow-safe multiplication and division
- Reverts with custom errors instead of reason strings
- Well-documented with NatSpec comments
- Built and tested with Foundry
I created this because I wanted a fixed-point math library that is at the same time intuitive, efficient and safe. I looked at ABDKMath64x64, which is fast, but it uses binary numbers which are counter-intuitive and non-familiar to humans. Then, I looked at Fixidity, which operates with denary numbers and has wide precision, but is slow and susceptible to phantom overflow. Finally, I looked at Solmate, which checks all the boxes mentioned thus far, but it doesn't offer type safety.
First, run the install step:
forge install --no-commit paulrberg/prb-math@v3
Then, add this to your remappings.txt
file:
prb-math/=lib/prb-math/src/
yarn add @prb/math
# or
npm install @prb/math
PRBMath comes in two flavors:
- SD59x18 (signed)
- UD60x18 (unsigned)
If you don't need negative numbers, there's no point in using the signed flavor. The unsigned flavor is more gas efficient.
It is recommended that you import PRBMath in the global scope, because you will often need multiple symbols:
pragma solidity >=0.8.13;
import "@prb/math/SD59x18.sol";
import "@prb/math/UD60x18.sol";
Note that PRBMath can only be used in Solidity v0.8.13 or above.
// SPDX-License-Identifier: UNLICENSED
pragma solidity >=0.8.13;
import "@prb/math/SD59x18.sol";
contract SignedConsumer {
/// @notice Calculates 5% of the given signed number.
/// @dev Try this with x = 400e18.
function signedPercentage(SD59x18 x) external pure returns (SD59x18 result) {
SD59x18 fivePercent = sd(0.05e18);
result = x.mul(fivePercent);
}
/// @notice Calculates the binary logarithm of the given signed number.
/// @dev Try this with x = 128e18.
function signedLog2(SD59x18 x) external pure returns (SD59x18 result) {
result = log2(x);
}
}
// SPDX-License-Identifier: UNLICENSED
pragma solidity >=0.8.13;
import "@prb/math/UD60x18.sol";
contract UnsignedConsumer {
/// @notice Calculates 5% of the given signed number.
/// @dev Try this with x = 400e18.
function unsignedPercentage(UD60x18 x) external pure returns (UD60x18 result) {
UD60x18 fivePercent = ud(0.05e18);
result = x.mul(fivePercent).div(UNIT); // UNIT = 1e18
}
/// @notice Calculates the binary logarithm of the given signed number.
/// @dev Try this with x = 128e18.
function unsignedLog2(UD60x18 x) external pure returns (UD60x18 result) {
result = log2(x);
}
}
There's significant overlap between the functions available in SD59x18 and UD60x18, so I did not duplicate the functions tables below. If in doubt, refer to the source code, which is well-documented with NatSpec comments.
Name | Description |
---|---|
abs |
Absolute value |
avg |
Arithmetic average |
ceil |
Smallest whole number greater than or equal to x |
div |
Fixed-point division |
exp |
Natural exponential e^x |
exp2 |
Binary exponential 2^x |
floor |
Greatest whole number less than or equal to x |
frac |
Fractional part |
gm |
Geometric mean |
inv |
Inverse 1÷x |
ln |
Natural logarithm ln(x) |
log10 |
Common logarithm log10(x) |
log2 |
Binary logarithm log2(x) |
mul |
Fixed-point multiplication |
pow |
Power function x^y |
powu |
Power function x^y with y simple integer |
sqrt |
Square root |
Name | Description |
---|---|
fromSD59x18 |
Converts an SD59x18 number to a simple integer by dividing it by 1e18 |
fromUD60x18 |
Converts an UD60x18 number to a simple integer by dividing it by 1e18 |
sd |
Wraps a simple integer into SD59x18 |
sd59x18 |
Wraps a simple integer into SD59x18 |
toSD59x18 |
Converts a simple integer to SD59x18 by multiplying it by 1e18 |
toSD59x18 |
Converts a simple integer to UD60x18 by multiplying it by 1e18 |
ud |
Wraps a simple integer into UD60x18 |
ud60x18 |
Wraps a simple integer into UD60x18 |
unwrap |
Unwrap an SD59x18 or UD60x18 number into a simple integer |
wrap |
Wraps a simple integer into either SD59x18 or UD60x18 |
In addition to the mathematical and the conversion functions, PRBMath provides many other helpers for the user-defined value types, such as
add
,eq
, and rshift
. These functions are not part of the core API and are frequently updated, so I invite you to take a look at the source code
to see the full list.
The goal with these helpers is not have to always unwrap and re-wrap variables to perform such basic operations as addition and equality checks.
However, you should note that using these functions instead of the vanilla operators (e.g. +
, ==
, and >>
) will result in a higher gas cost.
// SPDX-License-Identifier: UNLICENSED
pragma solidity >=0.8.13;
import "@prb/math/UD60x18.sol";
function addRshiftEq() pure returns (bool result) {
UD60x18 x = ud(1e18);
UD60x18 y = ud(3e18);
y = y.add(x);
y = y.rshift(2);
result = eq(x, y);
}
PRBMath is shipped with typed assertions that you can use for writing tests with PRBTest, which is based on Foundry. This is useful if, for example, you would like to assert that two SD59x18 or UD60x18 numbers are equal.
pragma solidity >=0.8.13;
import "@prb/math/UD60x18.sol";
import { Assertions as PRBMathAssertions } from "prb-math/test/Assertions.sol";
import { PRBTest } from "@prb/test/PRBTest.sol";
contract MyTest is PRBTest, PRBMathAssertions {
function testAdd() external {
UD60x18 x = ud(1e18);
UD60x18 y = ud(2e18);
UD60x18 z = ud(3e18);
assertEq(x.add(y), z);
}
}
PRBMath is faster than ABDKMath for abs
, exp
, exp2
, gm
, inv
, ln
, log2
, but it is slower than ABDKMath for avg
, div
, mul
, powu
and sqrt
.
The main reason why PRBMath lags behind ABDKMath's mul
and div
functions is that it operates with 256-bit word sizes, and so it has to account for
possible intermediary overflow. ABDKMath, on the other hand, operates with 128-bit word sizes.
Note: I did not find a good way to automatically generate gas reports for PRBMath. See the #134 discussion for more details about this issue.
Gas estimations based on the v2.0.1 and the v3.0.0 releases.
SD59x18 | Min | Max | Avg | UD60x18 | Min | Max | Avg | |
---|---|---|---|---|---|---|---|---|
abs | 68 | 72 | 70 | n/a | n/a | n/a | n/a | |
avg | 95 | 105 | 100 | avg | 57 | 57 | 57 | |
ceil | 82 | 117 | 101 | ceil | 78 | 78 | 78 | |
div | 431 | 483 | 451 | div | 205 | 205 | 205 | |
exp | 38 | 2797 | 2263 | exp | 1874 | 2742 | 2244 | |
exp2 | 63 | 2678 | 2104 | exp2 | 1784 | 2652 | 2156 | |
floor | 82 | 117 | 101 | floor | 43 | 43 | 43 | |
frac | 23 | 23 | 23 | frac | 23 | 23 | 23 | |
gm | 26 | 892 | 690 | gm | 26 | 893 | 691 | |
inv | 40 | 40 | 40 | inv | 40 | 40 | 40 | |
ln | 463 | 7306 | 4724 | ln | 419 | 6902 | 3814 | |
log10 | 104 | 9074 | 4337 | log10 | 503 | 8695 | 4571 | |
log2 | 377 | 7241 | 4243 | log2 | 330 | 6825 | 3426 | |
mul | 455 | 463 | 459 | mul | 219 | 275 | 247 | |
pow | 64 | 11338 | 8518 | pow | 64 | 10637 | 6635 | |
powu | 293 | 24745 | 5681 | powu | 83 | 24535 | 5471 | |
sqrt | 140 | 839 | 716 | sqrt | 114 | 846 | 710 |
Gas estimations based on the v3.0 release of ABDKMath. See my abdk-gas-estimations repo.
Method | Min | Max | Avg |
---|---|---|---|
abs | 88 | 92 | 90 |
avg | 41 | 41 | 41 |
div | 168 | 168 | 168 |
exp | 77 | 3780 | 2687 |
exp2 | 77 | 3600 | 2746 |
gavg | 166 | 875 | 719 |
inv | 157 | 157 | 157 |
ln | 7074 | 7164 | 7126 |
log2 | 6972 | 7062 | 7024 |
mul | 111 | 111 | 111 |
pow | 303 | 4740 | 1792 |
sqrt | 129 | 809 | 699 |
Feel free to dive in! Open an issue, start a discussion or submit a PR.
You will need the following software on your machine:
In addition, familiarity with Solidity is requisite.
Clone this repository including submodules:
$ git clone --recurse-submodules -j8 git@github.com:paulrberg/prb-math.git
Then, inside the project's directory, run this to install the Node.js dependencies:
$ yarn install
Now you can start making changes.
While I set a high bar for code quality and test coverage, you should not assume that this project is completely safe to use. PRBMath has not been audited by a security researcher.
This is experimental software and is provided on an "as is" and "as available" basis. I do not give any warranties and will not be liable for any loss, direct or indirect through continued use of this codebase.
If you discover any bugs or security issues, please report them via Telegram.
- Mikhail Vladimirov for the insights he shared in his Math in Solidity series.
- Remco Bloemen for his work on overflow-safe multiplication and division and for responding to the questions I asked him while developing the library.
- Everyone who contributed a PR to this repository.