Highball › Compare › Highball vs Pixel Port: Open Source vs Closed Beta 2026

Highball vs Pixel Port

Pixel Port is the closest thing to a mirror of this project. One developer, Apple Silicon only, one-click Steam games, compatibility tiers that admit when a game does not work. It launched on 28 June 2026, about two months before Highball, so it is a convergent answer to the same problem rather than a copy. Both projects were built because Whisky stopped.

The differences that matter are structural, not cosmetic, and a few of them go their way.

The differences at a glance

What Pixel Port does better

The first run is simpler. There are no bottles, no renderer picker, no decisions to get wrong. MacHow2's review scored it 75/100 and described setup as easier than CrossOver, Whisky, UTM or Parallels, with gameplay "no different to using Crossover". It is the only independent review either of us has.

They have verified more games by hand. Their stats page lists 58 titles confirmed on a real Mac. Highball has 5. That gap is real and not in our favour: a hand-checked verdict beats any estimate, and they have more of them.

They also ship two things Highball does not have: an in-app compatibility report for your specific Mac, and a command palette. And because their per-game fixes come from their server, they can repair a broken game for everyone without an app update. Highball fixes arrive as releases.

They are also much easier to find. Their search presence in this niche is the best of anyone's; if you got here by searching, you probably saw their page first.

What Highball does differently

Free, and structurally unable to stop being free. Pixel Port has published no prices. MacHow2's review says not to expect it to stay free forever, which is the reviewer's read rather than a Pixel Port statement. Highball is GPL-3: there is no mechanism by which the version you have today becomes paid, and if this project ends the code stays forkable.

No account. Their DMG downloads without a gate, but you sign in with Apple or Google and connect Steam before you play anything. Highball asks for nothing and sends nothing. To be precise: their docs describe a Distributed Validation Network where "every install and launch produces a signal", while their homepage says analytics are off by default and compatibility telemetry is opt-in. Those two statements do not obviously agree.

More than Steam. Steam is one click in both. Highball also does Epic through the open-source Legendary client, has a Battle.net recipe, and will run an ordinary Windows program that is not a game at all. Pixel Port is Steam-only, with no Epic, GOG or Battle.net support and no way to point it at your own .exe.

Three renderers instead of one. Pixel Port's runtime pairs a modern Wine build with DXMT, and their release notes add a DXVK route for older DirectX 8 and 9 titles. There is no D3DMetal path, so games that only behave under Apple's translation layer are out of reach by design. Highball lets you switch DXVK, DXMT and D3DMetal per bottle when a game misbehaves.

The honest counterweight: the Highball database currently has zero games with two dated renderer verdicts. We ship the switch; the evidence for which renderer wins per title is still being collected. How the data works.

Recipes as data. Highball's curated entries and recipes are CC0, the ProtonDB-derived predictions are ODbL, the anti-cheat data keeps AreWeAntiCheatYet's MIT notice, and all of it is plain JSON on GitHub. Pixel Port's equivalent lives behind their API. If they stop, it goes with them. If Highball stops, the data is still there and somebody else can use it.

The catalogue numbers, both of them

Pixel Port advertises 11,046 catalogue entries. Their own stats page puts 58 of those in the verified tier. MacHow2 counted 172 games actually playable when they reviewed it in August. All three are real and measure different things: the big one is mostly automated estimates, the small ones are what somebody ran. Their site does disclose the tiers, so this is a marketing-surface problem rather than dishonesty.

Highball has the same shape at a smaller scale: 101 curated games, 5 verified by hand on real hardware, 39 marked blocked by kernel anti-cheat, and 12,491 ProtonDB-derived predictions that are deliberately kept out of search until a human confirms them. Judge either project by its verified column. Ours is here.

Anti-cheat, where neither of us wins

Kernel anti-cheat loads a Windows kernel driver, and there is no Windows kernel under Wine on macOS. That is architecture, not a missing feature, and it hits both projects identically. Userspace anti-cheat varies per title: Helldivers 2 and Elden Ring's single-player both run under CrossOver despite carrying it, so the answer has to be per game rather than per vendor. Both projects mark blocked titles before you download 80 GB, which is correct and not a difference. The full explanation.

Which should you use

Use Pixel Port if your library is entirely on Steam, you want the shortest possible path from download to playing, an Apple or Google sign-in does not bother you, and you are comfortable with a closed beta whose pricing has not been announced. Their simplicity lead is real.

Use Highball if you buy games on Epic or Battle.net, you want to run Windows software that is not a Steam game, you want the renderer switch when something black-screens, you would rather not have an account, or you want a tool that cannot be paywalled or switched off later.

Do not pick Highball if you want the larger hand-verified list today, because that is theirs, or if you want per-Mac verdicts inside the app, or server-side fixes that arrive without an update. And if your game is kernel anti-cheat, neither of these is the answer, and no future version will be.

Both are free notarized downloads, so trying both costs an evening. Highball builds on Wine, DXMT, DXVK, D3DMetal and Legendary, and so, in large part, does Pixel Port. Credit where it is due.

Download Highball Apple Silicon, macOS 14 or newer. Install notes.