A set of banchmarks has proven that under some conditions, WINE can beat Windows XP SP2 at some tests. To be exact, out of 147 tests, WINE beats Windows XP on 67. The WINE version used was 0.9.5, running on a Gentoo Linux system, with some fairly decent hardware.
Why would they run Quake or UT2004 within Wine?
These games have native Linux support, I’m sure performance would be better when running native.
But still, I would never expect this little difference between a Windows OS which is in development for about 20 years now and Wine which is still in Beta fase.
lol. GNU never exits the beta fase. Its either beta or dead and unusable.
Dont get me wrong I love GNU linux, or atleast the illusion that it and all of its parts will do all of the things it claims eventually.
rofl, beta fase give me a brake.
This is just not true.
There is a lot of software past beta fase.
And don’t forget that a lot of “beta” software is often more finished than a lot of “professional quality” software on other platforms.
See, grownups usually spell it “phase”
exactly what grade of elementary school did you drop out of?
rofl Eighth actually.
Well I plan to post the native Linux scores as well. I was interested to see where Wine stacks up against Windows and Linux on OpenGL performance. And the best way I know to test our OpenGL performance is to run the native test and then run the same test in Wine to see the difference.
Tom Wickline
Why would they run Quake or UT2004 within Wine?
Yes, native versions of Quake and UT2004 exist, but the speed at which the Windows versions perform under Wine can be compared against the native ones to give some idea of how fast/slow Wine gaming is compared to native Linux gaming. If Wine proves fast enough with those two, then perhaps it’s fast enough for other games.
I think wine is at a disadvantage here, as it’s main use is no to run games, and wine’s directX support still has a bit to go (for example, at least for me, in 3d Mark 2000 the helicopter test is rendered incorrectly, so it’s not only slower, but cedega renders it correctly).
Some general-purpose benchmarks like multitasking various windows apps, encoding, photoshop, etc would give a better indicator of wine’s performance afaik.
Also, they don’t say if ut was running in directX or openGL mode, that could be a pretty good difference.
Some general-purpose benchmarks like multitasking various windows apps, encoding, photoshop, etc would give a better indicator of wine’s performance afaik.
Well, I tested speed long ago for some customer of mine. With Photoshop. First op on 150MB picture was 14% slower. Second 13%. Thrird 12%. But as soon as Photoshop touched swap, Wine outbeat Windows almost by factor 2x and it even seemed to increase with any io involving op pending.
But to be fair. Speed is not everything. That move of Adobe Collection on Linux was not done for various things.
1. Photoshop had few strange usability glitches
2. Lack of possible HW color calibration. No native support from any HW vendor (yes, I know manual is possible, but I wouldn’t even think about calibrating 15 monitors like that). Although I have in plan (somewhere in future) to test my new calibrator, older one didn’t work under wine, end.
3. Not really working InDesign. Especialy anoying was lack of redraw after menu was opened. On menu hiding, only window backgroud without contents would repaint
4. Illustrator was painfully slower on redraw at various occasions beside the usability glitches
5. Distiller was faster on larger PS files by a long margins, equal on smaller files
Edited 2006-01-19 19:54
Try to use it using codeweaver’s crossover office,
It’s a real pleasure to use photoshop 7 with it…
(I’m a graphic designer and I work with it…)
(I’m a graphic designer and I work with it…)
As I said I don’t need it personally.
Is any hardware calibration now possible or not? As I said customer of mine only waits to make this transition, but calibrating 15 monitors manualy and on a regular scale would be a waste of time.
btw. Photoshop is not the only thing they need. The only app running 100% when I was trying was Distiller (which I can only wish it could be run in some service option without GUI, thing rocked major).
For example Quark can’t be installed, but if you copy Windows installed Quark folder runs without a glitch (that would be as far as our testing went). Problem were InDesign and Illustrator. And the sad fact that customer upgraded to cs2. As I look on CodeWeavers cs2 is not supported.
given that most of wine’s weak scores were found in the graphics/3D realm, it would be interesting to see how Cedega (formerly wineX) preforms.
and while you’re at it, may as well include CrossOver and possibly other versions of windows.
Wine has the current lead on 67 test
there are 2 tests included where win and wine are equally fast. so it only leads in 65 tests
and in 50 of those 65 it has less than 10% advantage.
Wine or XP aborted on 17 test
to be correct it are 18. and wine failed 15 out of those 18.
So your point is … ? 10% or else, it’s faster in a failry large number of cases. It rocks and it’s funny at the same time.
Getting statistics corrected is always a nice thing.
smashIt did a good job pointing out some flaws.
it’s 17 learn how to count! and in 14 of the test Wine only has a 10% or less lag 🙂
Under Wine, a work-related ‘vertical’ application crashed instantly, but it took several minutes/hours under XP.
So, Wine is already beating it there.
(edited to remove some redundancy by removing a duplicate phrase to avoid redundancy)
Edited 2006-01-19 20:01
This story has sod all to do with MorphOS.
I don’t need no stinkin windows. Though I would like wine to progress for the people that rely on it. About the beta fase thing, we shall remember Windows has always been alpha quality not even beta. So wine is better in some tests because wine in it’s beta stage can be more stable than windows on it’s “final” which is alpha stage always. About the person who said that in 50 of the test it wasn’t more than 10% performance increase in wine, man what more would you like I though wine could not beat XP in a single part I’m very impressed of wine now. It proves how GNU progress for good and at a better rate than windows which has been almost beta quality for much more years than Linux. still Linux isn’t as stable as FreeBSD so don’t see me as a Linux fanboy, I’m not.
These are all broad generalizations, and XP still comes out on top. It’s like someone saying “Well, you might be able to beat my car with your 300 horsepower engine, but at least I have cooler rims.”
Thanks for the benchmarks, but i’ll stick to gaming on Windows.
These are all broad generalizations,
Look to be some pretty exact measurements to me, most down to a tenth of a frame.
but i’ll stick to gaming on Windows.
obvious not programming or engineering or you wouldn’t be saying 48% of your car is the rims.
😀
Like that last comment
Results look exactly what I could expect – some syntethical test suites are almost equal (this means that wine developers are using these suites while testing/optimizing wine), some suites fail completely (this means that devs just aren’t used these), working real applications are a bit slower (because these tend to be heavily optimized for windows anyway).
Mathematical tests should theoretically behave almost same – they don’t use API calls extensively. Same for OpenGL/DirectX – both worlds (windows/*nix) do optimize their graphic libraries, NVidia probabaly has almost identical driver code – 3D/gaming tests do not use win generic APIs too much either. Apparently some specific driver or GL functions are implemented differently.
Looking at results – wine (ie Debian) file system is faster than XP one. What FS did they use? How was swap located and configured in XP? The latest affects performance hugely; probably in Debian default configuration FS is faster than unconfigured NTFS – hard to tell though.
What I would want to see – comparison of applications, using lot of different kind API calls (OLE/COM based GUI applications with some media and database capabilities or something alike). Well, this can be difficult to compare – for many such apps win core libraries are needed, this makes differences between wine and win api speed less measurable.
Does anyone know – do old’n’good InstallShield 5 setups (16/32bit) already run on wine? About year ago this was ‘mission impossible’.
default configuration FS is faster than unconfigured NTFS
Right, and you get scored 3 for a bullcrap line like that. You want to talk about FS, fine, talk about FS. Also, I’d know a fine place for that “unconfigured ntfs” to be squeezed into.
Don’t get fooled by these test results by looking only at the greens.
3DMark2000 is a DirectX7 benchmark. DX7 performance is irrelevant in today’s times. Once the testing jumps over to 3DMark2001 (DX8.1), WINE loses terribly.
DronezMark is a fairly low-detail, low-poly game. Linux seems to have the upper-hand when it comes to OpenGL rendering of older stuff (meaning small texture sets, low polys, no shaders, etc.) — but only in the older stuff. Quake 3 and UT2004 once again prove that when doing any higher-detail work, it loses.
PCMark shows us that when WINE wins, it mostly wins by a tiny margin, but when it loses, it loses by a huge margin. To Linux’s credit, it seems to have much better disk I/O performance, as evidenced by virus scanning and file copying tests. I don’t really know what to make of the memory test scores, as I’ve never seen PCMark being used as a memory benchmark. I would recommend that the comparison be supplemented with at least some SiSoft Sandra results, or if you want to go with higher verbosity, RightMark Memory Analyzer.
GL-Excess again shows us that wins are mostly by low margins, but losses are by huge margins.
I don’t really know what to make of this, but my hunch would be that WINE is optimized to do certain kinds of operations very fast, and Linux’s threading/disk I/O help it out considerably, but there is still a lot of work to be done for the most part, because where WINE loses, it loses *badly*. Also don’t forget about the test failures.
In any case, WINE has gotten a lot further in the last 2-3 years. They just need to stabilize it a bit more, and optimize the parts where it lags behind a lot.
In any case, WINE has gotten a lot further in the last 2-3 years. They just need to stabilize it a bit more, and optimize the parts where it lags behind a lot.
(:Funny you said that:) Yes, it got very further, but to stabilize it more. Fsck, I’m waiting for that to happen’ about 8 years. I think it will be sooner that native software starts poping out than Wine starts stabilizing. Progress is great, but goals get wider and wider because their target is not in stand still, which makes it a problem.
Ok, long term joke aside.
From another point of view, I think Wine might actualy get the bigest help from Mono and MS itself. Since MS is moving to .NET and that is where Wine doesn’t need to follow, the one that gets more job here is now Mono instead of Wine which is actualy now following non-moving target for the first time. This would actualy make stabilization possible
.Net is still a moving target. It simply moves in more publicized standardized versions that don’t hit as often…
Until Mono supports windows forms it’s going to be pretty useless for most everyone (other than the asp.net folk). I know they’re working on that.
But when they finish that, and .Net 2.0 compatibility; .Net 3.0 will come out with things like LINQ for them to work on. And developers are typically playing with the latest toys for their main money-maker platform. So, I think that if Microsoft plays its cards right it can keep Mono a full year behind on things that people are shipping. This is a smaller gap than Wine has had, but they also got the benefit of appearing to care about standards .
And I didn’t mention WPF.
Net is still a moving target. It simply moves in more publicized standardized versions that don’t hit as often…
I said the same, I never said .NET is standstill. I just pointed out that it is not WineHQ problem anymore but Mono. So Wine has option to stabilize.
Mono on the other hand can stabilize non-Windows and Mono parts separate from Windows parts.
RightMark Memory Analyzer from: http://cpu.rightmark.org/download.shtml ?
I just downloaded it and ran the Microarchitecture & RAM Performance test in Wine….. Is anyone else here besides POO interested in this benchmark?
Also do you want me to run GL-Excess at a higher res? it will run at higher resolutions btw.. 🙂
And as for DX 8.1 there are three guys working on this in there spare time and to even run the benchmark and get a score is a huge accomplishment on Wine’s part!
Tom
Rightmark Memory Analyzer is a comprehensive memory benchmark that lends insight into your computer’s memory subsystem’s performance. PCMark’s memory bandwidth tests are pretty simple, and like I said, I’ve never seen them used in any real review.
As for GL-Excess … sure, let’s see a higher-res comparison between WinXP and WINE on the same hardware.
DirectX 8.1 runs, yes … but are there many rendering errors?
Thanks, I thought I was the only one impressed with this. So Wine isn’t far superior to a native XP install in all cases. No kidding. It is quite an accomplishment that they have been able to develop a system that can not only run quite a lot of stuff from another operating system whose notorious for their undocumented API calls and developer/hacker unfriendly attitutde, but that they can approach, and occasionally surpass the performance of said system.
Amazing. They deserve a round of applause, not this drivel of ‘ohhh wake me when it is 4x faster playing Call of Duty 2’.
3DMark2000 is a DirectX7 benchmark. DX7 performance is irrelevant in today’s times. Once the testing jumps over to 3DMark2001 (DX8.1), WINE loses terribly. —Linux Is Poo
Thanks for pointing that out. That said I’d disagree that Direct 7 is completely useless. My WinTV-Radio card (just one of many examples I’m sure) uses direct X 7 9 to provide the best rendering and overlays. As does Media Player Classic (now that I think of it) and like I said I’m certain there are many other applications that offer Direct X 7 as a fall back point for api support or hardware compatibility reasons.
I don’t really know what to make of this, but my hunch would be that WINE is optimized to do certain kinds of operations very fast, and Linux’s threading/disk I/O help it out considerably, but there is still a lot of work to be done for the most part, because where WINE loses, it loses *badly*. Also don’t forget about the test failures. —Linux IS Poo
Ehhh.. I think yoiu forget who we’re talking about here– Gamers. The same kinds of people who’ll run out and buy a $400.00 video card just to bump their performance by 10% on WoW or other games. To Gamers the parts where Linux does better is of interest and will likely be used by some of the more knowledgeable geeks to boost their performance. The areas where Linux falls on its face are of interest too, but in that case as an excuse for these same Gamers to dual-boot back into Windows for the better (native?) support offered there.
The nice thing is being able to have any kind of scoreboard at all by which to be able to compare and contrast performance and more importantly a way to improve that performance.
Personally, I’m looking forward to when ReactOS finally hits the 1.0 release because I’d imagine that once things start coming together for ReactOS as a workable (daily driver) type system Wine will begin really improving the performance at a rapid rate! Actiually come to think of it in the next few months and years now that MacOS X has been publically released for X86 hardware, I’d be surprised if the work on DarWine didn’t help in ratcheting up performance and general improvements as more and more people want to run Win32 applications on none-Microsoft Operating Systems.
Thiknk what you want about Linux or Microsoft, I don’t think anyone here disagrees that we’d all be better off if we had a more diverse OS enviroment– even the Windows boosters ought see that, because traditionally whenever Microsoft is in direct competition their inovation and coding skills explode. Should be some interesting times once things start really moving.
–bornagainpenguin
The kind of gamers who will run out and buy the latest hardware will *never* run Linux. Don’t lose your head here. Windows is *the* platform for gamers everywhere, and that is not going to change because Linux might do a few FPS better in a few games.
As for OS X … it hasn’t been publically released for x86 hardware. It has been released along with x86 Mac hardware.
“The kind of gamers who will run out and buy the latest hardware will *never* run Linux. Don’t lose your head here. Windows is *the* platform for gamers everywhere, and that is not going to change because Linux might do a few FPS better in a few games. ”
I know a few gamers who dual boot. If GNU/Linux performs better at some games they’ll certainly take advantage of that.
“As for OS X … it hasn’t been publically released for x86 hardware. It has been released along with x86 Mac hardware.”
What’s your point? The poster was saying that now that Apple uses x86 you can run Wine on it.
http://www.opendarwin.org/projects/darwine/
Since more people will be running Darwin on a x86 more people will likely take interest in Wine. When it works it’s a lot better than VirtualPC or dual booting.
Wine is pretty good with older programs. They don’t have to be that much older, but definitely the newest latest and greatest does not run well. I am using it for Internet Explorer 6, iTunes and Microsoft Office 2000 and for the most part, performance is okay. Not as good as native apps (or when running those apps on Windows), but these programs are definitely quite usable. I plopped down 40 bucks for Crossover Office and for the supported programs, it works great. Unsupported programs are hit or miss, but I think overall Wine does a pretty good job given the task set before it.
i use wine for one purpose only
and that is to install FlashFXP (for implicit SSL)
works great
as regards the test results, who cares ! how about 3d mark 2003 or 2005. That would be way more interesting to me as I use both those apps to benchmark systems.
cheers
anyweb
http://linux-noob.com
3DMark 2003 will run the first test 🙂 2005 & 2006 will install and start up just fine. they need Pixel Shader 2.0 support to run there test and Wine currently only supports version 1.1
When PS 2.0 is implemented ill post the results of those benchmarks as well.
Tom
Wow this is quite awesome.
well, I’ll be honest, I don’t play games much, nor do I care about games, but Office XP runs substantially faster in Wine then it does in Windows, the same is true with Photoshop 7.
I am sure Wine runs slower in many apps, but the ones I use it for are faster with one exception: emule, which starts slower but runs just as fast.
Crossover Office and Cadega have been making waves in the Wine community and much of their development goes straight back into Wine proper, evntually Wine will have near full xp compatability but i can’t say much about vista’s compatability (I smimply don’t know)
TransGaming / Cedega gives nothing back to Wine!
The fine folks at CodeWeavers gives almost everything back to Wine.
“TransGaming / Cedega gives nothing back to Wine!
The fine folks at CodeWeavers gives almost everything back to Wine.”
They are not required to do so. The version of Wine that they started from was not covered under Wine’s current license. If the people at winehq.com can’t manage to do the things that the other two groups have accomplished, it’s hardly the fault of Transgaming or Codeweavers.
Edited 2006-01-20 00:23
“If the people at winehq.com can’t manage to do the things that the other two groups have accomplished…”
Ah, you’re joking right? Winehq has been developing Wine for 12 years, Transgaming takes that work, adds functions to improve DirectX support and the Wine developers “can’t manage” to keep up with that.
I’m glad Wine is LGPLed now.
“Ah, you’re joking right? Winehq has been developing Wine for 12 years, Transgaming takes that work, adds functions to improve DirectX support and the Wine developers “can’t manage” to keep up with that.”
No, actually, I’m not. At the time Transgaming grabbed up a copy of Wine, they were both working with the exact same code, regardless of the fact that the Wine project had been working on it for years prior to that event. Neither had any distinct advantage over the other WRT codebase.
I suspect that the Transgaming folks were simply more knowledgable about DirectX than the people working for the Wine project, hence Cedega is now the superior of the two WRT games. If the Wine project can’t do the same on their own, they should look for more qualified developers.
The Wine project is doing great work, granted, but DirectX seems to be a sore spot for them ATM, and that cannot be blamed on the other two groups for not sharing, when sharing was not required.
Edited 2006-01-20 01:30
You’re forgetting that Wine is a project where most of the developers are volunteers, and they work on what’s important for them. Trasngaming is a company, with resources to spend and focus on one thing: making wine play games. To suggest that Wine should “look for more qualified developers” is ridiculous.
To expect the Wine developers to not only provide millions of lines of code for Transgaming (as they do) but also work faster than full time developers is something a reasonable person would not do.
“You’re forgetting that Wine is a project where most of the developers are volunteers, and they work on what’s important for them.”
I’m forgetting nothing, and your statement only re-enforces my position that it’s not the fault of the other groups for winehq’s comparitive lack of progress WRT DirectX.
Actually Wine is moving quite fine along in regard to DirectX.
It’s a lot better now than it was in earlier days. A lot more DirectX games are running now under Wine, and there is little to gain in using other versions, perhaps apart from CrossOver Office.
“They are not required to do so.”
Correct, and they don’t! and still people go around to different forums posting that they give back to Wine when they DON’T.
Cedega should be considered as proprietary software that got its start on top of Open Source Software.
When CrossOver 6.0 comes out with Gaming support, hence better DX support TG days will be numbered. 🙂
Tom
Although these tests may not be completely representative of neither platform, it is amazing to see that Wine beats XP in almost 50% of the tests. But considering the uber-optimized environment that Gentoo provides, this could be considered a ideal scenario right? Being to lazy to test all that myself, how do you think Wine on for instance Debian would compare?
Or the (sort of) other way around, how does some random Cygwin stuff on XP compare to the real thing GNU/Linux?
Edited 2006-01-19 22:43
Or the (sort of) other way around, how does some random Cygwin stuff on XP compare to the real thing GNU/Linux?
I use Ghemical an advanced GPL molecular modelling package:
http://www.uku.fi/~thassine/ghemical/
It works pretty well even on my old 700MHz PIII Celeron, 128 Mb RAM box with no hardware OpenGL and using MESA instead. It generates and manipulates 3D molecular graphics at a reasonable speed even on this old system (kernel 2.6.7 multimedia optimized).
The University of Iowa has done a port and released a Pre-Configured Cygwin + Ghemical Environment binary:
http://www.uiowa.edu/~ghemical/windows.shtml
which I installed on a 2GHz, 256 Mb Win2000 box with of course hardware OpenGL at work. The graphics though looking good were incredibly slow to manipulate, I don’t know if it was using the hardware OpenGL or not, but it was really an unusable bummer.
On the subject of Wine and chemical software. The International Union of Pure and Applied Chemists (a chemistry standards body) has recently launched InChI, the International Chemical Identifier developed for them at NIST:
http://www.iupac.org/inchi/
They released a command line tool for converting standard molecular structure files to InChI’s with both Win and Linux binaries and the source code for porting to other Unix based systems. They also produced a graphical Windows program which gives you quite a lot of information with what is happening in the conversion. I think they really should have developed it in Java (or Python-Tkinter) to provide an instantly cross platform tool to support the international standard. However they developed a Windows only program called WInChI.
I tried it out under Wine on my Linux box at home and it worked perfectly it did everything the same as running natively under Windows except for the ability to drag and drop molecular structure files onto the program window to load them. So overall I was very pleased with this and gained a new respect for Wine.
Edited 2006-01-20 01:45
Cedega gives more back to wine then you think, they just make it a bit more inconvenient to use their code.
Cedega does allow you to compile from cvs and hence, wine proper has access to all the code but wine has to leave out anything that has to do with directX.
The code in Cedega cannot be used because of license issues.
Besides that, pulling out the directX part makes it pointless to use code from Cedega. And again. You can’t use the code due to license incompatibilities.
wine benches would most likely be very porportional to the linux to windows performance of your particular distro compared to gentoo
debian runs a bit slower then a well configured gentoo box and hence I would expect the wine performance to be proportional to the gentoo to debian performance ratio….. so if debian runs 10-15% slower, i would expect a 10-15% performance loss in wine.
WINE is almost reaching a point where it will be good enough to most people. I was kinda surprised when I began to realize that some Windows shareware/freeware developers were recommending WINE as a solution to run their software on Linux as is the case of ArtRage – http://www.ambientdesign.com/artrage.html – which is a nice paint package that reproduces natural media (aaaaaand works perfectly, as I am rather fond of this app on Linux) and Shareaza – http://wiki.shareaza.com/static/ShareazaLinux – which doesn´t need to be introduced, I guess… 🙂
KaZaa is another well known app that worked perfectly on Linux on top of Wine (or it could be Kazaa Lite, its full-featured-but-non-ad/spyware-crippled cousin, I can´t quite remember…). I used to use it until Apollon hit the streets.
I´ve even seen some people developing keeping Wine compatibility as a goal. The Application Database at WineHQ have valuable information about the status of popular applications and the simple ones “just work”, without requiring tweaks on config files nor anything like that.
Sure, it has room to growth but it is still one of the best tools to help one to maintain his/her transition to OSS operating systems smoother than it could be otherwise.
Transgaming doesn’t have to give anything, since they (as was pointed out) forked before Wine went to LGPL and thus aren’t required to. CodeWeavers (Crossover) on the other hand didn’t fork from the MIT licensed version, and *does* work closely with Wine, and almost everything finds its way back. Transgaming and CodeWeavers shouldn’t be lumped together quite so easily.
Wine doesn’t do half bad on directX apps. One game where it works better than Cedega is Operation Flashpoint. Black screen in Cedega, flickery bushes but otherwise normal graphics in Wine. Well, the sound is another story. Terribly skippy in Wine, just fine in Cedega The visuals anyway are better in Wine, and the sound should improve, hopefully to the point it is at with Deus Ex and Jedi Academy among others.
Anyway, I don’t think games are really Wine’s main focus There’s dual booting for those who really “need” the latest games, or increasingly: consoles. Wine IMHO really exists for the apps, and it is getting better there. I just wish each new version didn’t break an app that used to work (current formerly working now broken app: WinMX).
I salute those working on all three branches. They’ve taken on quite a task. Best of luck to them.
> I just wish each new version didn’t break an app that
> used to work (current formerly working now broken app:
> WinMX).
WinMX runs just fine in Wine 0.9.6.
It’s news that at least wine is not too bad performance wise. But the problem is not there. The real wine test is in outright compatibility with Windows binaries, and on that front, it’s not so good is it? And what will happen when Vista comes out? Are we back to square one?
Wine is one of these red herring that is not only distracting but also a big dead end, since no vendor is going to consider a problem on wine as a problem they have to fix and the Wine team cannot hope to fix all the problems with all Windows apps. Right now, the solution is to use crossover for Office and Cedega for games. Then what ? Another provider for graphical apps ? That’s going to be expensive and not very attractive.
The Linux movement seems only interested in following other people’s rules by re-implementing their wares and running on hostile hardware, rather than clearly setting its own sound and unique set rules that would-be adopters could rely on for a good length of time and defining a hardware platform that any local assembler and willing OEMs could implement.
Sometimes I wish Windows was a good OS.
The real wine test is in outright compatibility with Windows binaries, and on that front, it’s not so good is it?
It is good already, but sure it’s not perfect, which is what you meant I think.
And what will happen when Vista comes out? Are we back to square one?
No, Wine will still be there and still work.
the Wine team cannot hope to fix all the problems with all Windows apps
That’s for sure, even MS can’t do that.
Right now, the solution is to use crossover for Office and Cedega for games. Then what ? Another provider for graphical apps ? That’s going to be expensive and not very attractive.
No, you’re wrong. The thing here is to evaluate your needs. Everybody does not need what you are talking about. Wine is well enough for my needs, and it’s not expensive at all (it’s free actually).
It may not suit your needs, that does not mean it won’t suit needs of other people. You are an agocentric person, that’s the only problem I see here.
The Linux movement seems only interested in following other people’s rules by re-implementing their wares and running on hostile hardware
To add insult to injury, you’re clueless. There’s no Linux movement, perhaps you meant FOSS or Libre movement ? Wine is not head of any of these movements, and your description of these movements is wrong at best. You perfectly described MS behaviour though, that’s funny.
rather than clearly setting its own sound and unique set rules that would-be adopters could rely on for a good length of time and defining a hardware platform that any local assembler and willing OEMs could implement
A bit more clueless …
FOSS already has its own sound and unique rules : GPL, LGPL, community participation, share, …
Would-be (and some misleaded) adopters already rely on them, especially applicances manufacturers.
But the more clueless sentence of your is the one about defining hardware. One of the strong points of Linux and FOSS is that it can run on every hardware platform if you want, no need to define anything.
That’s precisely what a lot of manufacturers are doing, get a clue please !!!
Sometimes I wish Windows was a good OS
Some have been wishing for more than a decade …
Not to brag, but I understood MS was hopeless in 2 years time (I was fed up quite fast actually).
Right now, the solution is to use crossover for Office and Cedega for games. Then what ? Another provider for graphical apps ? That’s going to be expensive and not very attractive.
Actually Crossover runs graphical apps like photoshop quite well. In fact Crossover runs a lot of different types of programs well. I wouldn’t use Cedega for anything other than games but Crossover runs a good mix of software.
The Linux movement seems only interested in following other people’s rules by re-implementing their wares and running on hostile hardware, rather than clearly setting its own sound and unique set rules that would-be adopters could rely on for a good length of time and defining a hardware platform that any local assembler and willing OEMs could implement.
Linux is much more than re-implementing other people’s wares. In fact that is the minority of the software out there that runs on Linux. As for implementing its own hardware…why? The hardware is already there. The hardware support is already there. Sure, it’s a little rough around the edges but most of it works fine. It would be a much bigger task implementing new hardware instead of fixing the problems with current hardware.
Since WINE sits on top of LINUX, you lose half the MSFT “spaghetti code”. I’m surprised WINE didn’t win ALL the benchmarks.
Please inform us how you know about their “Sphaghetti Code”. Do you work for microsoft?
Please, let us know!
I wish he would. I had always envisioned it as being more of a ghoulash but that’s just because of the Hungarian notation.
I did not write the original comment that you replied to, but if the MFC headers are any indication of the design & implementation of Windows I will have to concur.
Far too many macros (minmax anybody?), non standard language extensions, data type redefinitions — often different, etc, etc.
Uhm… the macros are there to make the windows headers as compiler and platform-independent as possible. For instance, if you define MAC, you can use some of those MFC headers for porting your code to a Mac with some shim libraries. That’s part of how they make Office:Mac (though not with MFC, of course). The macro functions are there for efficiency before compilers were so strong on optimizations (compilers probably also had fewer intrinsics at that time too) (not to mention they are not susceptible to weird/random rules that optimizers often apply).
Most of the windows headers can be understood in terms of the fact that they were written before C standards contained all of the features they use. In some ways, they created things that later became standards. Win3.1 and probably many parts of Win95 were probably written in ASM.
I’m not going to post the results of the other test. I have to save something for the next update right?
3DMark 06 partial results in Wine 0.9.6 :
http://wiki.winehq.org/BenchMark-0.9.6
Tom
Under certain conditions you can travel through time !
Under some circumstances — say somebody ties my hands behind my back — you could probably kick my ass…