# Comparing Setups

> Put two setups of the same car side by side, parameter by parameter — what changed, what only one side has, and what cannot honestly be compared.

Two setups, side by side, one row per parameter. Use it to see what a remix
actually did, to line up the version that worked against the one that didn't, or
to find out why two setups you have been treating as siblings behave nothing
alike.

> Setups is in **Public Beta** and requires a personal **Plus or Ultra** plan. A Team seat alone does not grant access.

## Opening a Comparison

From your library, click **Compare setups**, select two setups and choose to
compare them. From a setup you already have open, use **Compare with…**. That
picker lists up to your 50 most recently updated setups in the same simulator and
leaves out setups for a different car. Either way you can switch the version
shown on each side afterwards — comparing **v1 against v5 of the same setup** is
a perfectly good question, and often the most useful one.

## The Five Kinds of Row

Most rows are identical and stay quiet. The rest are sorted into categories,
because "different" is not one thing:

| Row | What it means |
|-----|---------------|
| *(untinted)* | Both sides agree |
| **Δ ±n** | Both sides have the parameter and the value moved |
| **only in A** / **only in B** | One side has the parameter and the other does not |
| **≠ unit** | Both sides declare a unit, and they are not the same unit |

Each category carries a **word**, not just a color, so the rows stay readable if
you cannot separate the tints.

**Only in A / only in B is not a rounding error.** Two setups you picked
independently can come from different base files, so a parameter existing on one
side only is routine — and reading it as "unchanged" would be a difference you
have no way to notice.

**≠ unit outranks the values.** iRacing writes display strings that follow your
metric/imperial preference, so the same tire reads `165 kPa` in one export and
`24.0 psi` in another. Braking Lab does not convert — a value converted by eye is
worse than an unconverted one, because it looks authoritative — and it does not
pretend the two numbers are comparable. A unit mismatch is flagged **even when
the two numbers happen to be equal**, since that equality is a coincidence of
scale, and it stays out of the change count.

## Reading Δ: Not Every Number Is a Magnitude

In Le Mans Ultimate a setup value is an **index into the car's option table**,
and the label beside it is what that index resolves to. So subtracting two of
them counts **positions on the dial**, not physical units.

Braking Lab says which it is rather than leaving you to guess. A row that reads

```
CamberSetting     -1.00 deg  →  -0.75 deg     +5 steps
```

moved five clicks, and those five clicks are worth 0.25 deg. Printed bare under a
header that says Δ, `+5` beside two physical labels would be a sentence that
contradicts itself. The word is what makes the number true, so it travels with it.

For iRacing the value **is** the magnitude, and Δ is the plain difference.

## When Braking Lab Refuses to Compare

Some pairs cannot be compared honestly, and a refusal that says why is more use
than a table of noise:

- **Different simulators.** LMU and iRacing do not share a parameter vocabulary, so
  lining up their values would compare names that only look alike.
- **Different cars.** Both car identities are known and they do not match. Putting
  them side by side would align values that have nothing to do with each other.
- **No readable values.** An iRacing `.sto` is encrypted, so personal Braking Lab
  Setups does not accept, store or compare it. Export that setup from the garage
  as `.htm` and upload the export to make it comparable — see
  [iRacing setups](/en/docs/ai-setups/iracing-setups).

And one warning that does not block: if either setup stores no car identity, the
comparison runs **without** the same-car check, and says so. Read those rows
knowing a difference may simply be a different car.

## Related

- [Setups and your sessions](/en/docs/ai-setups/sessions) — which setup you actually drove, and what it did
- [Measuring a setup change](/en/docs/ai-setups/measuring-changes) — what two confirmed runs say on track
- [Remixing setups](/en/docs/ai-setups/remix) — where the second setup usually comes from