CrossOver Support - Community Forums

We currently face a situation, where, with every new version of CrossOver, things that used to work in previous versions stop working on newer versions. To me this is a sign that quality assurance at CodeWeavers fails utterly these days. It should be self-evident to make sure, a new version is only released after making sure, that each and every application known to work with a previous version will also work on the new version, because we are paying for improvements, not decline. This is not what one could expect from a free product, like Wine, but can from a paid product.

I used to be a big fan of CrossOver but this is something really disappointing to me. Also I don't get why they are wasting resources on bringing CrossOver to Android when there are still so much applications and games not working flawlessly on Linux and Mac and the majority of Android devices being touchscreen-based and not running on x86 processors at all. It's just superfluous.

CodeWeavers' commercials say we don't need to ever pay for a windows license again when using CrossOver. But my experience is that around 80% of all current applications do NOT work flawlessly with CrossOver. Unfortunately. And we still have to fiddle around with configuration, as many CrossTies are outdated and there's no way to know what they will really do to a bottle. I remember a time when those were not community-maintained and were only available for officially supported applications, and they used to work 100% of the time.

I hope that some CodeWeavers official may come to read this and probably think about going back to the roots on some of the things mentioned. Maybe you should take development further away from Wine's releases and make a real difference.

I must agree... mostly! But I think these are hard times for Crossover / Wine users, community and developers. Windows 10 sucks and has changed substantially from Windows 7. We have software developers changing many things. Steam seems to be fine, but Blizzard and Ubisoft launchers are hard to put to work. The move to 64 bit and DirectX 11 are a source of trouble. But maybe Codeweavers has made a too much big step developing Crossover for Android? Games I used to play nearly perfectly like Starcraft II and Star Trek Online stopped to work a couple of versions ago. Not that all these regressions are Codeweaver's faults, many are caused because of these changes in software technology being adopted by the developers.

So, I would really appreciate that no regressions were happening but I must understand that wine / Crossover are masterpieces of software, very complex. Nevertheless, please, fix the regressions. Thank you.

Well, the whole thing around Blizzard games is another example for a faulty decission made by CodeWeavers. Indeed there is a version on which not only World of Warcraft, but also StartCraft II and the 32-bit version of Diablo III run almost flawlessly.

It can be found here:
https://www.codeweavers.com/compatibility/crossover/tips/world-of-warcraft/special-world-of-warcraft-build-for-crossover-17

But they decided to not include those changes in the official releases, because "it may break other things". Even though they know, that not including them DOES in fact break many games that used to work before. Besides that, I used to run other applications on that "Special World of Warcraft Build" and had no problems with whatever software I tried. That means: Codeweavers deliberately decided to not fix things they already have the patches for at hand. It's just ridiculous.

Of course Wine is a masterpiece of software. But it seams to me as if the license fees we're paying for crossover are just given to the developers as some sort of donation and I feel that this is not the part CodeWeavers should play, but instead should use them to steer development towards a more professional direction, set clear milestones and goals for further development and make it an even more amazing and professional product. When Transgaming's Cedega was still alive, they managed to push Wine's DirectX emulation capabilities to a level not even reached by Wine today. Unfortunately they never gave those improvements back to the community and because of the Wine project's license change they were not able to keep up with improvements to the overall codebase. But they gave great support and had an ever growing list of supported games, which they made sure would run as long as one's subscription was active and that is what I demand from a professional product.

I hear you.
One setting I would really want is to be able to set a bottle to wine version.
This would solve a lot of issues because the wine they use (3.2) cannot possibly work for everything. I had this problem with installing Fallout New Vegas which crossover install failed but a simple manual wine install where I could choose the version worked painfree.
I think more advanced configuration options are needed for those that need less hand-holding. If we break the app or game then its on the user to deal with it - which is fine by me

I must agree with Bachsau and Devon.

I, too, would love to be able to choose the wine version. I guess they think things would get too complex for newbies, but it would be nice to see some "Advanced Configurations" or something.

Also, maybe it is time to separate things again and have Crossover Games back.

it seems that the Cx Hey-Day was way back in the heady days of win_xp,
and a whole stack of windows software was available for that era,
& getting at least some of it, working with ./wine was a big bonus.

you can still see this today, as lots of software will only run in a win-xp bottle .. ..
sure, you can push quite a bit of that software into a win-7 bottle, where it will work,
but, increasingly amounts of that older software won't work in a later windows / bottle at all.

thus with the increasing trend to Microsoft Store,

• • which you have to get the later software versions from :(
    and even getting that software that was available for win-7 - is getting to be a right PiTA.

crossover games would be nice but wouldn't that mean they would have to split up the focus , people and time spent on writing and testing

also be nice to have more tools to play with things to help run stuff better but also wouldn't that mean it could make it harder to get crossover to run right due to having to rework the whole program and having more to test ( making it slower to come out of testing and more hours spent on testing then adding new stuff)

just my limited thoughts as I'm not a software guy but as my dad always said the more they overthink the plumbing, the easier it is to stop up the drain

I'll be the devil's advocate here...

As a client, I fully understand the positions and points made by all of you. Yet, I have to question the perspective.

I'm convinced that Codeweavers neither has the staff nor the funds (as in research budget) to avoid such difficulties as regressions as has been discussed. There is no doubt that Codeweavers must choose between growth and quality control trying to balanced them as they can. This in no way invalidates the points made before. If anything, I'm only calling for temperance. I'm convinced there are business considerations that we, as clients, are completely unaware of.

Although I fully support your right to voice your opinions, I'm also certain that your opinions (as my own) are not fully informed about all aspects regarding the variations of Crossover, or Codeweavers business needs and decisions.

I think that CodeWeavers could get money from a lot more people if they would include fixes and patches not (yet) available in the upstream Wine distribution. People might not be willing to pay just for the bottle manager, if they can get their apps to work for free, but if the experience was like: "Hey, I tried my distributor's Wine package, but it didn't work. It took me two hours to find a patch and fix the build errors, but I still had to fiddle around with winetricks, while with CrossOver it took me one click, and it was done." That would make a big difference.

On the contrary, my bet is that it is harder to make office apps and games work in the same product. So splitting up again into Crossover and Crossover Games would reduce regressions (IMHO).

Regressions do happen, that is why ther MUST be an "ADVANCED CONFIGURATION FOR EXPERTS ONLY" window with an option to choose wine version among other things. It is nearly impossible to make all games and applications work everytime, in every version with the latest Crossover. The problem is there are too many regressions happening now...

David wrote:

crossover games would be nice but wouldn't that mean they would have
to split up the focus , people and time spent on writing and testing

also be nice to have more tools to play with things to help run
stuff better but also wouldn't that mean it could make it harder to
get crossover to run right due to having to rework the whole program
and having more to test ( making it slower to come out of testing
and more hours spent on testing then adding new stuff)

just my limited thoughts as I'm not a software guy but as my dad
always said [b]the more they overthink the plumbing, the easier it
is to stop up the drain[/b]

I tend to agree. Previously, my primary computing platform was Mac, but I grew increasingly frustrated with a relative lack of software applications development for Mac. Consequently, my primary computing platform is Windows, but I also use Linux, especially for molecular modeling applications.

I am not a gamer, and I have given up trying to get MS Office working flawlessly on Linux using Crossover or Wine. Instead, for Office applications, I use LibreOffice, which runs natively on Linux (as well as Mac and Windows).

Most other Windows applications that I care about do not run on Linux using Crossover or Wine. Indeed, it is even sometimes problematic getting them to run in a Windows VM. With VirtualBox, software that requires OpenGL and.or OpenCL will not run or will run only imperfectly. VMware Workstation or VMware Workstation Player does much better with graphics, but audio is often absent or heavily distorted.

Currently, I find that only a few legacy Windows applications run well on Crossover, and in these cases, I often have better luck with Wine (I use the development branch from WineHQ).

Wine and Crossover are nice ideas, but for the Windows applications that I need to run, they are much too unreliable to be of much use. Therefore, I have all but given up with Wine and Crossover.

When I need to run a Windows application, I either use a PC with Windows installed or a Windows VM running on a Linux PC.

When using a VM, I use Virtualbox if I don't need 3D-accelerated graphics and if I care about audio quality. On the other hand, if I need 3D-accelerated graphics and I don't care about audio quality, I use VMware Workstation or VMware Workstation Player. Better still, if I want the best possible performance from Windows software, I use a real PC running genuine Windows.

Yet, one of the Biggest Issues, that I'm always coming across,, is to do with the Pricing Structure of Crossover:

• there is a perception that a New User has to $59.95 on an Annual Basis - ie: every year.

now, ideally, from a Business Perspective,, the company "codeweavers" would certainly like that idea,
but, I've pointed out, that idea isn't necessarily true - - you could Buy just the One Copy, for your 'windows program'.

• and never, ever have to buy another copy, & whilst that would be true, it's also a bit UN-realistic as well.
  as your Host System would also evolve & you would, eventually, have to upgrade your copy of Crossover,
  and most probably, also your copy of that 'windows program' as well.

that's all to do with the constant evolving of Computers, in General.

• nothing is Ever Static. ..
  and neither is the crossover product - that the codeweavers company, is constantly creating.

There has been a lot of discussion in this thread, and a lot of topics covered. I want to respond to a couple of the concrete issues.

One of the important factors which I think kicked off discussion on this thread initially is support for World of Warcraft in CrossOver 17. WoW has been broken in wine, but working in wine-staging, for quite a while now because it requires a large set of patches which have not been accepted into the upstream repository. There are big technical challenges involved in solving the problems which need to be solved to get WoW running. The staging patchsets touch major areas of Wine and the work is too large to be realistically segregated behind a registry setting or similar. We determined these were not suitable for CrossOver 17 and built a one-off build in hopes of maintaining WoW compatibility. That has worked to a degree and given the constraints I would chose that approach again if I had it to do over. That said, we also want WoW working better, and at this point, we are exploring getting that patchset into the next release of CrossOver. I can't make any guarantees that it will happen, but we are actively exploring that. A beta for CrossOver 18 is not too far off at this point. I know everyone here wants Blizzard games - and other things, frankly - working in CrossOver. We definitely do also. I'll re-state that the business and technical challenges keeping everything running are very real. Of course, when the software you care about isn't working, the fact that the challenges are difficult ones doesn't help you.

A second topic discussed here is whether CrossOver will support multiple different versions of Wine, something like one version per application. On that point, I think it is unlikely we will go down that road. I myself advocated for that once upon a time, but on the whole I think it would be a mistake. We push instead to get working code upstream, so that one version works with all apps. When that is not possible, we generally aim to make targeted hacks as a practical stopgap so we can support things which don't otherwise work. The 'special build' for WoW is a good example of why having multiple wine versions is not really workable ... we can't maintain them over the long run. That's why, in this case, we are exploring getting those patches into CrossOver rather than, for example, trying to maintain that WoW build as a separate entity. Basically, focusing on a single wine version allows us an economy of scale, and that scale allows us the cycles to focus on other heavy lifting like implementing new dlls or keeping up with major host platform changes.

It is useful to have threads like this, where people tell us what their major difficulties with CrossOver are. Most of these things are things we know, but it's still very helpful to have users giving their perspectives on the pain points. So on that, thank you to folks who have taken the time to write in here and discuss. We can't always address every issue - even some of the legitimate ones - but we are at least here, and thinking about these things and working on them.

Thanks, and Cheers,
Josh