Jump to content


Member Since 10 Aug 2014
Offline Last Active Private

Posts I've Made

In Topic: Catalina and 32bit

23 June 2019 - 04:37 PM

@Zakk if Ubuntu don't change their mind later and the person is a newbie they might as well move to Pop!_OS they said they won't drop 32bit support, so at least a good supported Debian based distro will still have 32Bit support.

And I agree I don't see a decent macOS solution appearing for a good while at this rate, it's sad considering CodeWeavers were making the most money from macOS ports. I only see it happening faster if Value wants to add official Proton support on macOS

@ovvldc QEMU might be slow and I wouldn't even mind if it covers running DirectX9 and below games at playable rates, Hangover on the other hand I can't even launch the few games/programs I had tested. On Reddit someone responded to me saying they did run a few programs on an arm64 version of Ubuntu but on straight Linux it wouldn't even run those same programs.

In Topic: Catalina and 32bit

22 June 2019 - 02:59 PM

The Linux situation is vastly different then on macOS 32Bit libs being removed is different then outright blocking 32Bit code execution like macOS.

CodeWeavers are working on different avenues but it’s not like we will get any confirmation until they are ready (None Discloser Agrements) , this won’t reflect Winehq.

I personally have been contributing directly to https://github.com/P...nicis-winebuild , qparis added a Hangover environment and builder but hangover is basically useless.

It also already integrated the custom version of clang from ccdavis for all macOS builders but -mwine32 that it provides is not usable at precent for wine, lightly needing patches that we frankly might never see until CodeWeavers are ready to release them.

In Topic: Wineskin Unofficial Update

06 June 2019 - 01:08 AM

View PostRastaFabi, on 05 June 2019 - 08:16 PM, said:

Hi @Gcenx, I just wanted to report, that your wineskin fork apparently is working fine on macOS Catalina.
While I did not test existing wrappers I created a new one with a 64bit engine (4.7). Wineskin proceeded to create the wrapper and I did launch wineconfig.exe, regedit.exe and some more tools. Checking the wine environment I found the expected: syswow64 (the 32bit directory :huh:) is empty, as the 32bit wine process setting it up just did not run.

Yeah since on prefix creation wine couldn't function so
wine wineboot
would have failed but
wine64 wineboot
would have functioned.

I recently make sure to have Wineskin default to using wine64 over wine when available, I had the same results when testing on Mojave with 32bit mode disabled.
Still I'm a little surprised it even worked at all considering all the engines other then then the CX ones are from winehq.

If you replace the wswine.bundle contents with a pure 64Bit compile and rebuild the wrapper (ignoring the error about wine) it will function.
When I have some time I might do a proper set of updates more aimed at Catalina

If you would like a pure64 compile to play with let me know I can thrown one together.

I should explain a little more, even if you compile wine as only 64Bit you will get the exact same results as you experienced here, the syswow64 directory and other 32bit folder will still get created by using “wine64 wineboot”

In Topic: Wineskin Unofficial Update

01 June 2019 - 11:25 PM

View Postovvldc, on 01 June 2019 - 06:33 PM, said:

Fully emulating would be crippling. I can imagine a scenario where JIT recompiling makes sense and doesn't slow down too much when going from 32bit to 64bit as they are roughly the same instruction set, but it will still cost memory and CPU cores.

But if they have a patched clang anyway, they can add this as well. As long as Intel and AMD still support the 32bit instructions anyway ;)

I will properly build hangover tomorrow and give it a try myself lets see if it's usable or not before I go making changes to WineskinLauncher to accommodate it.

But yeah I'm sure clang will be used over hangover for macOS, since hangover was more aimed at ppc arm devices

Well looks like I can skip building it myself for the moment, qparis build it and uploaded a build, I did some tests and it’s honestly not very useful in its precent build configuration.

In Topic: WineStagingGnutls3.X+ & WineStagingVulkan64Bit3.X+

01 June 2019 - 03:40 AM

Something is really off while I tested WS10WineStagingVulkan64Bit4.9 on my system with the dylibs embedded without any issues.

It gives the same issue as Winehq builds if the dylibs are not embedded, I will try to investigate the problem when I have time but it’s strange everything below 4.8/4.9 don’t need embedded dylibs to function correctly

Resolved seems skipping my usual step of using "install_name_tool" to fix the path within wines .so files now breaks them is my guess.
I've uploaded a working version to MEGA.

I was informed that the bug affecting Wine-4.8 defaulting to X11 doesn't mean we can't use macDriver it just needs to be changed and since Wineskin already sets
it shouldn't be an issue for us, so I will build it and add to MEGA