Comparison
BrightIntosh or BrighterMac: what did we learn from it and what did we change?
BrightIntosh is free software, and it is where we learned how this is done at all. We built our own engine around a clip ceiling, made the app reach full brightness roughly three times faster, and made it say out loud which hardware we have actually measured. BrightIntosh costs $1.99 once — a small fraction of either of our prices — and has been shipping for years. Both of us let you try before you pay: three days there, 3 at full strength here.
Where this app comes from. BrightIntosh being open is what let us find out that any of this was possible. The code we ship is our own, but the understanding is not something we arrived at alone, and we are not going to compare ourselves with it while pretending otherwise.
What we learned from it
The 256-entry gamma table. macOS will report success for a gamma table written at the device's full 1,024-entry capacity and then clip every entry to maximum. BrightIntosh writes 256. We found out the hard way that this was not an arbitrary choice, and we write 256 too.
The trigger window. Holding a tiny HDR window open so that macOS grants the display its extended range is the mechanism the whole category is built on, and BrightIntosh's implementation is where we learned it.
The idea of an adaptive factor. BrightIntosh varies its multiplier with the display's reported EDR rather than fixing it. On our test machine we measured its factor at 1.36875 with an EDR of 8.0, matching its published formula to the decimal. That idea is theirs. What we do with it is not.
What we changed
The factor is capped by where the curve would clip, not only by EDR. BrightIntosh's factor responds to headroom but not to the shape of the resulting curve, and at full boost we measured 276 of 1,024 gamma entries — 27.0 % — pinned to maximum. Ours computes a ceiling from the headroom and the curve's own peak and takes the lower of that and the design point, so it cannot be driven past the clip point. We have not yet re-measured the current engine on the bench and so we publish no percentage for ourselves; we will when we have.
It gets bright faster. 2.28 s from launch to full boost against 6.5 s. That difference is visible every time you open your laptop.
It refuses to run where macOS is broken. The M5 Pro and M5 Max gamma defect in macOS 26.3 is blocked by model. Every other supported model is labelled measured or merely supported, because exactly one of them has been on our bench.
It lets you see the effect. A split-screen check shows boosted and unboosted halves at once, so the claim on this page is testable on your own machine.
The battery guard has a threshold. Seven steps between 10 % and 50 %, rather than a single behaviour.
Where BrightIntosh is the better buy
It costs $1.99, paid once — and that is roughly a twenty-fifth of our lifetime licence, or a fifth of one year of ours. It has years of App Store reviews from people who are not us. And it is free software you can read and build yourself, today, without paying anyone — ours is not, and that is a real advantage on their side.
We also ask for money differently. BrightIntosh is one payment and so is every other app in this category; we are the only one with a yearly licence, because we treat keeping the mechanism alive across macOS versions as the actual work. Our reasoning is on the pricing section, and if it does not convince you, $1.99 once is a very reasonable answer.
What is no longer a difference: trying it first. This page used to list BrightIntosh's three-day trial as something we did not have. We do: 3 days of the whole app at full strength, starting the moment you first open it, with no account and no card — and after them the app keeps boosting part of your screen, and that part gets smaller week by week instead of stopping. Neither of us asks you to pay to find out.
If price is the deciding factor, buy BrightIntosh. We are charging more because we think the engine and the honesty are worth it, and that is a claim you are entitled to disagree with.
The table
| BrighterMac | BrightIntosh | |
|---|---|---|
| Boost factor 1 | Capped by your panel’s headroom | Adaptive, fixed ceiling |
| Highlight clipping at full boost 2 | Capped by construction | 27.0 % of the curve |
| Time from launch to full boost 3 | 2.28 s | 6.5 s |
| Private Apple frameworks 4 | None | None |
| Restores your display bit-for-bit on quit 5 | Yes | Yes |
| Refuses to run where macOS is broken 6 | Yes, by model | not measured |
| Tells you when it could not verify the effect | Yes | No |
| Built-in visual check that the boost is real 7 | Yes | No |
| Battery threshold you choose | 10–50 %, seven steps | No |
| Scheduling, sunset, per-app rules 8 | No | Partly |
| Shortcuts, CLI, URL scheme 9 | No | No |
| Backs off automatically when headroom drops 10 | Yes, within 0.5 s | No |
| What it costs 11 | 9,99 € a year, or 49 € once | 1.99 USD once |
| Try it before paying 12 | 3 days at full strength, then less area | 3 days, whole app |
| How the licence is checked 13 | A signed document the app verifies itself | No licence check outside the App Store |
| Where the verdict is kept afterwards 14 | A signed file, re-verified on every launch | The App Store receipt |
| How long it works with no network at all 15 | Up to 180 days, and we publish the number | Indefinitely outside the App Store |
| Source code you can read | No | Yes, GPL-3.0 |
- Ours computes a ceiling from the headroom the display reports and takes the lower of that and the design point, so the curve cannot be driven past the panel. BrightIntosh varies its factor with EDR but not with where the curve would clip.
- Measured by counting entries pinned to maximum in a 1024-entry gamma table. Our figure is a property of how the ceiling is computed; the current engine has not yet been re-measured on the bench, and we will publish the number when it has.
- Measured on the same MacBook Pro 16" M1 Max, mains power, at the versions listed below.
- Observed from the shipped binaries’ link tables. Private frameworks are rejected by the Mac App Store and can break on any macOS update.
- All three do, and so does macOS on its own within 19 ms of a force-quit. We list it because people ask, not because it separates anyone.
- The M5 Pro and M5 Max defect. We block those models rather than sell into a failure.
- A split-screen button that shows boosted and unboosted halves side by side.
- A real gap on our side. Vivid is ahead here and we are not going to hide it.
- Also a gap. If you automate your Mac, Vivid does more today.
- Ours re-reads the display’s reported headroom twice a second while the boost is on and recomputes the factor when it falls, measured at 0.364–0.585 s to react. It was measured with a simulated headroom value on a Mac without an XDR panel; that real HDR playback lowers the headroom in the first place is the part we have not yet confirmed on the target machine.
- The two figures on the right are the vendors’ own published prices, not our measurements, and both are one-time payments: nobody else in this category sells a subscription. Ours is the only one with a yearly option, and the honest reason is on the pricing section of the front page — the mechanism depends on macOS behaviour that Apple has already broken once. If you would rather pay once, our lifetime licence is the comparable number, and it is higher than both.
- Ours starts by itself the first time you open the download — no account, no card, no key — and after the first days it keeps boosting a part of the screen that shrinks over time rather than stopping. The Mac App Store edition has no trial at all, because a paid App Store app is bought before it is downloaded. The other two figures are the vendors’ own published terms rather than our measurements: Vivid’s free Splitscreen Mode has no time limit, and BrightIntosh publishes a three-day trial.
- Ours is an Ed25519 signature, so a forged key cannot be produced without our private key, and a forged reply cannot be produced by anyone who controls the network. Vivid 2.18.1, read from the binary the vendor ships: the app calls Gumroad’s licence endpoint and the reply is unsigned JSON, with no cryptographic symbols anywhere in the licence path. BrightIntosh publishes its source under GPL-3.0; the entitlement check is compiled into the App Store build only. Anyone can repeat both checks on the shipped apps.
- Observed in Vivid 2.18.1’s own strings (`userHasValidLicense`). We state where the value lives and draw no conclusion beyond that. Ours cannot be edited into a different answer, because changing it invalidates the signature.
- All of Vivid, BetterDisplay and Lunar require a network to activate; none of the three publishes how long their apps keep working without one, which is why this row says "not documented" rather than a figure we would have to guess. Ours is written into the terms, and the app tells you before anything stops.
Measured 20 September 2026 on a MacBook Pro 16" M1 Max (MacBookPro18,2),
macOS 26, mains power. Versions: BrighterMac 0.1.0, Vivid 2.18.1 (released 11 March 2026),
BrightIntosh 6.0.10. Everything above comes either
from our own measurements of the shipped apps or, where the footnote says so, from the
shipped binaries and the vendors' own published terms — never from anyone's marketing.
Decide on your own screen.
3 days at full strength, free, then 9,99 € a year or 49 € once. Check that your Mac is one we support — and one macOS has not broken — before you download anything from anyone.