Akastroid vs AutoStakkert

The short answer

Choose Akastroid if you want one program from the capture to the finished planet instead of PIPP, AutoStakkert and RegiStax in a row, if you are on a Mac and tired of running Windows software under Wine, or if you shoot deep sky as well and want one application for both.

AutoStakkert suits you if you are on Windows, you already have PIPP and RegiStax set up, and you want a free stacker that does that one job.

What was measured

Both stacks of the same captures, unsharpened, compared at matched physical frequencies so that neither side is flattered by a finer grid, each normalized against its own coarse-scale content.

• Detail. On a Saturn capture, ours carried 2.6x the power at 0.15 cycles per pixel, 7.3x at 0.20 and 10.8x at 0.25. The limb, which noise cannot flatter, measured 24.67 px against 26.00, which is sharper. On Mars the limb was level at 4.00 px each, with 8-20% more energy at low and mid frequencies on our side.
• Noise, earlier comparison. Binned to camera scale, two captures: Saturn measured 27% cleaner on ours, Mars 79% cleaner on AutoStakkert's. The later comparison of eight captures, with the noise floor measured in the same tile of both stacks, found ours quieter on all eight; that is the current result.
• A retraction. The first version of that noise comparison said "AutoStakkert is 56% cleaner", and it was wrong. Per-pixel sigma on a 3.0x drizzle grid is lower than on a 1.5x one whatever the underlying noise, because neighboring pixels come from the same photons. The number changed when the ruler was fixed. I've left the mistake on the page rather than quietly editing it.

What that adds up to

As stackers, level: the same detail at matched noise across eight captures, and quieter on every one. Not sharper: I claimed that once and withdrew it. It was behind on red and blue for one-shot-color captures (the demosaic, not the stacking) until September 2026, when each colour began to be integrated from its own sensor sites; on the Mars capture every channel now reads 1.03x AutoStakkert's on the unsharpened stack.

Where Akastroid differs is the chain. AutoStakkert stacks and stops; the sharpening is RegiStax, which has been unmaintained since 2011; the centering and cropping is PIPP; and all three are Windows-only. Akastroid does the whole of that in one pass, on a Mac or on Windows. That is a point about the workflow, not about whose wavelets are better.

Three programs or one

The usual planetary chain is PIPP to center and crop, AutoStakkert to grade and stack, RegiStax or AstroSurface to sharpen. Each is good and each is a separate export, a separate folder and a separate set of settings to remember.

Akastroid does the three stages in one pass: frames are graded and located from where the planet is, stacked with multi-point alignment, corrected for atmospheric dispersion and sharpened per scale, with the sharpening strength measured from the stack rather than set with sliders. The same application also stacks deep-sky frames, which none of those three do.

Side by side

Questions

Is Akastroid better than AutoStakkert?

As stackers, they are level. Across eight captures compared on the unsharpened stacks (matched physical frequencies, binned to camera scale, noise floor measured in the same tile of both), Akastroid's integration holds the same detail and is quieter on every one. I previously claimed it was sharper on every capture; that was withdrawn in September 2026 when the rulers behind it proved to carry a contrast confound and a noise floor that subtracted more from whichever image resolved more, which penalised the sharper image. It was behind on red and blue for one-shot-color captures until September 2026, when the stacker was changed to integrate each colour from its own sensor sites instead of demosaicing first; on the Mars capture every channel now reads 1.03x AutoStakkert's on the unsharpened stack. So the reason to choose it is not the stacking, which is level. AutoStakkert stacks and stops; Akastroid also sharpens, corrects dispersion, does deep sky and runs on a Mac.

Does AutoStakkert run on a Mac?

Not natively. It is a Windows program, and Mac users run it under Wine, CrossOver or a virtual machine. That works, and it is one more thing to maintain. RegiStax, PIPP and WinJUPOS are Windows programs too, so the usual planetary chain needs Windows. Akastroid is a native Mac application on Apple Silicon and Intel, as well as Windows.

AutoStakkert is free. Why pay for anything?

Because one application replaces three. Akastroid takes the capture from the raw video to the finished, sharpened planet in one run, where the usual route is PIPP, then AutoStakkert, then RegiStax. It runs natively on a Mac with no Wine, and it does deep sky, the Moon and the Sun as well. The free version has every feature, so you can compare it on your own captures first.

Do I still need PIPP?

Usually not. PIPP's common jobs (centering a drifting planet, cropping to the disc, converting between formats) happen inside Akastroid's planetary path: every frame is located from where the planet is before it is stacked, and the result is cropped to the disc.

Do I still need RegiStax?

Not for wavelets. Akastroid sharpens per scale after stacking, with the strength measured from the stack rather than set by six sliders. Measured on the same Mars capture, the fine detail at 1-2 pixels read level with a RegiStax user's own wavelets (0.0111 against 0.0111) and Akastroid's limb was free of ringing. RegiStax still has controls Akastroid does not expose. The RegiStax comparison says for what the measurement does and does not mean.

The planetary workflow in detail is on the planetary page, and what replaces PIPP, AutoStakkert and RegiStax goes stage by stage. On a Mac, start with what runs on macOS, which includes the free tools that do. Download for Mac and give it the same SER or AVI you stacked in AutoStakkert, then put the two results side by side. Give it the capture rather than the stack, because Akastroid does its own grading, alignment and integration.