Will 3DMark_Next support DirectX10.1 [Shader Model 4.1]

Well, HLSL shading style and GLSL syntax aren't very diffirent. Effects are other history.
Btw, glFX and GLSL v2 are comming :p

The different syntax isn’t the only GLSL problem. There are big issues with the way OpenGL consume these shaders. Each driver takes the GLSL source code directly. The first problem is that the compilers of todays driver have some differences in their lexers and parsers. This will make shaders accepted by one implementation and refused with a syntax error by another one. The second problem is that there is no way to say beforehand if a shader will work hardware accelerate on a GPU. This may be OK if you write an application and you control the hardware it would run but in the world of PC gaming this just causes massive headaches.

In ID Software linux and MacOSX are valid options too.

This was more or less a personal decision and not business driven. There were never an official Linux SKU of any id game.

Cieder? I think you wanna say Cedega and Wine?

No Cieder. Cieder is Cedega for the Mac but with a complete different business model.

Well whatever... software developers cannot make apps for only one OS. We need portability. See ID's policy change with the Tech 5 engine... read John's interviews and how he remarks what is critical for a game is portability.

If game developers talks about portability they talk about PC, Xbox, Playstation they don’t talk about different OSes for PCs.

3dmark is a synthetic test to measure the graphics card ( well... and computer ) performace. Due to that, must try to test all the aspects of the graphics performance ( fill rate, calling overhead, CPU balance, blah blah blah )...

If you read the technical PDF that comes with 3dmark you can see a brief explanation of the techniques they are using and why are using those tests... For example, the test with the airplanes + smoke was designed to measure the T&L + particles + fill rate.

Now, based on the mark score that gives you can extrapolate the results to know how fast is your card... but then, you run Doom 3 and see it goes very slow... but how can that be if you got 13k amazing 2006 marks? Perhaps your drivers or card is not well optimized for OGL but is a machine using D3D... So need some other indicator to know the GPU speed ( for example the mentioned Doom3 FPS test ).

Again. 3dmark tests how fast a card/driver can run their implementation of a technique.

Games may use other implementation and techniques. Additional they use different combinations. This make the results of an 3dmark run hardly useable for prediction of other game performances.
 
You need to change lots of things to port D3D9 path to D3D10, the API is very different ( and Vista API too ). Probably you will need to do the same work to port to OpenGL 3... but considering the poor D3D10 portability the OGL path will be better because same code will run on different OS, so is very relative.
Not really. The basic concepts of DirectX remain the same between D3D9 and D3D10 - I ported a simple graphics engine over from D3D9 to D3D10 with little effort/hassle. I couldn't say the same for a port to OpenGL. Oh, and the additions to the Win32 API in Vista are minimal. It's pretty much all the same.

Well whatever... software developers cannot make apps for only one OS. We need portability. See ID's policy change with the Tech 5 engine... read John's interviews and how he remarks what is critical for a game is portability. To develop a game or app only for one OS is a suicide.
That's funny. I seem to remember a statistic where Windows users comprised 90% of the entire desktop computer market. I imagine an even higher percentage of gamers would use Windows (since there are practically no games for Mac/Linux compared to Windows).

So I guess you're right, if you develop a game exclusively for a single OS such as Linux it would be suicide. But if you develop solely for Windows you'd be able to do quite fine. As evidenced by the many successful games companies that do just that and are getting along fine.
 
I would like to talk 3dm20008 team and ask them for Vista and DX10 problems... why do you think 3dm2008 is yet not out? I can figure the response... vista ..ux...
I'm here. ;) Why our next gen 3DMark isn't our yet? Because the next 3DMark simply isn't done yet. Also, we have no plans to produce an OpenGL benchmark on the PC. At least not in the near future.

Cheers,

Nick
 
Nick,

I know you often say ...soon... very soon.... etc... with regard to 3dmark 08

But when are you going to release your first image of 3dmark 08 ? if not a date, a solid time frame, about some movement?
 
Each driver takes the GLSL source code directly... The first problem is that the compilers of todays driver have some differences in their lexers and parsers. This will make shaders accepted by one implementation and refused with a syntax error by another one.... The second problem is that there is no way to say beforehand if a shader will work hardware accelerate on a GPU....
Well, OpenGL ES allows to use precompiled shaders. The decision to use source coded shaders was to optimize better but, as you mention, drivers include different GLSL compiler implementations and sometimes a code works on a card and does not in other. They way to solve that is to test your shader on different cards until it works in all.

The OGL software mode fallback was a pain, I agree with you. Hopefully OGL3 gonna change that ( gonna make things to emit an error if are not supported instead of use the software mode ).

No Cieder. Cieder is Cedega for the Mac but with a complete different business model.
Ahhhh I see. Well I googled it and got no emulator entries hehe, must be new or something.

If game developers talks about portability they talk about PC, Xbox, Playstation they don’t talk about different OSes for PCs.
Well, portability is a general term... it refers to consoles, different OS or even different languages ( for example C is portable because can be used inside C++, ObjetiveC, D, etc ). I could accept a "DX is portable" because runs on Windows and XBox and can be used from C++,VB,C#, etc... but is not native(emulators excluded) multivendor portable because only runs 100% well on Microsoft products.
Btw, I think you can program the PS with OGL too.

Games may use other implementation and techniques. Additional they use different combinations. This make the results of an 3dmark run hardly useable for prediction of other game performances.
Well, to be sincere I don't like synthetic test too much... but they should develop it in a way that could represent the general game performance of a card... because if not they are just a nice thing to your eyes.

Oh, and the additions to the Win32 API in Vista are minimal. It's pretty much all the same
But Vista has all the .NET 3.5 things(Thing many game editors, MDX games and .NET apps count too), new security model, new input and sound system, new driver model, etc...

But if you develop solely for Windows you'd be able to do quite fine.
Well, I would like to know then why no company is making a game exclusively for Windows Vista and DX10.
 
Last edited by a moderator:
Well, OpenGL ES allows to use precompiled shaders. The decision to use source coded shaders was to optimize better but, as you mention, drivers include different GLSL compiler implementations and sometimes a code works on a card and does not in other. They way to solve that is to test your shader on different cards until it works in all.

Very bad solution. Especial as it would not be enough to test against every card you need to test against every driver too. Additional what is about unreleased cards and driver?

Btw, I think you can program the PS with OGL too.

A special OpenGL Es version that use Cg instead of GLSL. But it isn’t recommended to use it.

But Vista has all the .NET 3.5 things(WPF,blah blah), new security model, new input and sound system, new driver model, etc...

Some of these things are transparent for the application and you are not forced to use any of the others.
 
Hey, Nick... can you tell us if 3dm2008 gonna use the GI technique in the ATI paper using cubemap arrays or any other exclusive DX10.1 feature(like MSBRW) pls?
 
Last edited by a moderator:
I agree with JohnC... [...]
there is no point to require DX10 for a game..
You completely missed the point of what he is saying. He was saying that requiring Vista for DX10 is an arbitrary way to tie the improved API to the new OS... whether or not that is true is debatable and besides-the-point. No one is arguing that DX10 isn't significantly superior to DX9 as that is just vacuously true.

Totally disagree. Vista, for me, is one of the worse OS created ever [...] due to bad drivers and app incompatibilities with firewall, SATA, printers, scanners, pen tablets
Wow, I'm getting great flashbacks to how bad XP was when it came, or XP SP1, and yet all of those things I had no problem with and indeed in time have been accepted to be clear improvements over what was there previously :D (Aside: seriously you like to overexaggerate much? Your assertion is really that Vista is inferior to DOS? Lol!) My point is not to get in an argument over whether or not you can find people that don't like Vista - because I don't care: I can find people that love it. People are just whiny and conservative, as they always are. In the end, use what you want but you're not gonna get much sympathy when you whine about DX10/3dmark being Vista-only.

Perhaps, but I could show you some games with almost the same appeareance and DX10 runs almost at the 70% of the DX9 path.
The comparison is totally meaningless unless the games are doing the *exact same thing* on each API. I've written such an app and it was faster on Vista/DX10, sometimes by as much as 10-15%, so you can perhaps understand my skepticism :)

Ask any serious FPS gamer and will tell you the typical "vista sux".
= conservative bias and self-selecting nature of internet posting. I'm a fairly "serious FPS gamer" and I'm quite happy with Vista (x64 even!).

I tested Vista myself and I think is very bad OS.
I tested Vista myself and I think "it" is a good OS, already clearly superior to XP and it will only get better. So I guess we'll have to agree to disagree and I expect you to no longer make any sweeping statements to anyone that don't include my quite-informed positive opinion of Vista ;)

What speed are you talking about? Vista came one year ago and I can just see a very small bunch of DX10-prepared games(AFAIK today I can buy in a store just Crysis, Lost Planet and Bioshock...)... and I still see very common BSoDs on the drivers... SLI is unstable, x64 port versions of application lacks quality or is inexistent... and the Vista incompatible list of apps is absolutely preocupant...
Meh... I've played all of the above (and CoH) in DX10 mode without a single blue-screen, and that's on the terrible x64 OS (the horror!) ;)

Anyways enough of this rediculously pointless debate. DX10 and Vista are both here to stay and unless OpenGL 4.0 or 5.0 is going to come out in the next month I'm not going to get very excited and I doubt that many (windows) game developers will. GL is just falling further and further behind and needs a swift kick in the butt if it wants to get back in the game. Until that happens, and a good percentage of games start to use the significantly improved/changed OpenGL, it has no real place in 3dmark.
 
In the end, use what you want but you're not gonna get much sympathy when you whine about DX10/3dmark being Vista-only.
...
Anyways enough of this rediculously pointless debate
All i'm saying is that Vista is getting hard critics from all the sectors(gamers,developers,press,IHVs,users). If 3dm2008 is DX10-Vista only then won't be too popular(and there is other thread for this I think). You like Vista and I hate it with pure passion. I think we aren't going to change our opinion :p Vista sales will say the last word.

So lets center into DX10.1 and 3dm2008 because we're going completely offtopic :D!
But all this was not completely pointless! Nick appeared so we can bomb him with questions!

Nick, pls if you can tell us more about
DX10.1 and 3dmark 2008 we will be very pleased.
Basically I wanna know:

1. Is 3dm2k8 Vista and DX10-only? Is 3dm2k8 enrolled into the "Designed for Windows Vista" or TWIMTBP program? What paper plays Microsoft, NVIDIA, ATI, Intel, etc on 3dm2k8?
2. Are you going to use DX10.1 or just DX10? If only DX10.0... planned any patch for DX10.1 in the future?
3. Any plans to give an app-postmortem telling us about the pros/cons of Vista and DX10 programming you found?
4. Are you going to use any GI technique like Lightsprint, Geometrics, FantasyLabs or ATI GI demo?
5. Which software are you using to create 3dm2k8?
6. How many people are working on 3dm2k8 currently? Which group structure you use(lead graphics, programmers, etc)? Do you use any team system like Subversion?
7. Are you going to release a x64 version of 3dm2k8? Any patch for SSE4 and Penryn/Phenom?
8. Can you give us some info about estimated finalization date? Any screenshot?
9. What do you think in general about Microsoft, XP/Vista, DX9/10, OpenGL, linux, MacOSX and future platforms like android, Larrabee or virtualized graphics?
10. Any plans to port 3dmark to other platforms?
11. Any plans to include the 3dmark technology in a game like Alan Wake(I think madonion/futuremark and Remedy are related?)
12. Any plans to dev-blog?

thx!
 
Last edited by a moderator:
Lets be honest, DX10.1 brings a lot of new features that don't really matter much if at all, and you can read all about them here. That said, there is one there that will matter a lot, contrary to what MS people say. This magic feature is the multi-sample buffer reads and writes(MSBRW). If you are wondering how you missed that big one in the feature list, well shame on you, read better next time.

What MSBRW does is quite simple, it gives shaders access to depth and info for all samples without having to resolve the whole pixel. Get it now? No? OK, we'll go into a bit more detail. DX10 forced you to compute a pixel for AA (or MSAA) to be functional, and this basically destroyed the underlying samples. The data was gone, and to be honest, there was no need for it to be kept around.

Games like Quake3 would do a lighting pass, then a shader pass, and another lighting followed by shaders and so on until everything was rendered right. This was quite precise but also quite slow. Dog slow.

To optimize around this, a technique called deferred shading took was invented. This does all the lighting passes followed by a single shader pass. If you have five passes, you basically can skip four trips through the shaders. The problem? Because the pixel isn't fully computed, just a pile of AA data, there is no way for it to be read. This is horribly simplified, but I don't want to go into the low level stuff here, go look it up if you really care.

What this meant is that you can't turn on AA if you have deferred rendering unless you do Supersampling which is rendering it at higher reolutions and sampling down. This is unusably slow, so it went out the door, meaning if you were designing a game, you picked speed in the form of deferred shading, or beauty in the form of AA. Most DX10 games will go for speed, meaning the AA hardware will sit more or less idle.

DX10.1 brings the ability to read those sub-samples to the party via MSBRW. To the end user, this means that once DX10.1 hits, you can click the AA button on your shiny new game and have it actually do something. This is hugely important.

The first reaction most people have is that if a game is written for DX10, then new 10.1 features won't do anything, AA awareness needs to be coded in the engine. That would be correct, but we are told it is quite patchable, IE you will probably see upgrades like the famous 'Chuck patch' for Oblivion. Nothing is guaranteed, but there is a very good chance that most engines will have an upgrade available.

In the end, DX10.1 is mostly fluff with an 800-pound gorilla hiding among the short cropped grass. MSBRW will enable AA and deferred shading, so you can have speed and beauty at the same time, not a bad trade-off.

Since NV has not done the usual 'we can do it too' song and dance when they are being beaten about the head and neck by a bullet point feature they don't have, you can be pretty sure they can't do it.

Close looks at the drivers, and more tellingly no PR trumpeting that they will have it out before the release of SP1 almost assuredly means that it will never happen. If you have a G8x or a G9x card, the only feature of DX10.1 you will miss is the important one. µ

http://www.theinquirer.net/gb/inquirer/news/2007/11/16/why-dx10-matters

Very interesting feature!

Edit: I wonder what Nick[FM] has to say, since 3DMark predicts future gaming! "Hopefully assuming :D"

Thanks,

Best regards,

Shtal.
 
Last edited by a moderator:
That's Pro version of Vista, and it's the box version, not OEM.
(It's only fair to Compare XP Pro price to Vista Pro)
Actually that's not true since the distribution of features in Home vs Pro is much more sane in Vista. Note that there is no "Vista Pro", only a "Vista Business" which is a good name since they really did move around the features. For instance I'm quite happy with Vista Home Premium on my laptop whereas I consider XP Home to be stupidly gimped in several places. The only thing I miss from Vista Home is remote desktop (server) but arguably you only need that for a few server-like computers anyways.

And seriously, buy OEM! Retail only makes sense if you're going to transfer the license >2 or 3 times, which most people aren't.

For reference, Vista HP OEM = $112CAN @ NCIX... I consider that more than reasonable for an OS. If you want all of the bells and whistles, Vista Ultimate OEM = $183 which is also quite reasonable IMHO.

Thus I'm not saying that everyone needs to run out and upgrade right now, but if you're getting a new PC or video card anyways, the cost of a new OS isn't really that critical (particularly if you'd have to buy a new XP license anyways in the first case).
 
AndyTX, the price of Visyta in Europe differs a lot from the American continent. I bet that is due to the 600M $ UE fine vs Microsoft, so need to raise price to compensate. Here you can only find Vista "N" versions ( without IE7, MediaPlayer, Outlook, etc... ) due to the judge resolution. I could buy it on an USA shop, but then a special VAT + import tax will be applied raising a lot the final price. On the other hand I could but one in eBay, but if the user already activated the license I could not be able to use it and neither use the tons of WGA-protected updates from Microsoft.

For example, see the price of Vista Home Premium "N" here:
http://www.elcorteingles.es/informatica/producto/producto.asp?referencia=28610737869
Yes, 524$ (around 340€).. that's the real price on the street here.
And about WinXP's price... I don't think is good... the only good price was the Win1.0/WIN95 ones... See this funny video "only 99$" from current Microsoft's director... Ballmer... and the people rushing the shop to get Win95 ( I didn't see people rushing the shops to get Vista... )

http://video.google.com/googleplayer.swf?docId=-2992183880068262304&hl=en
googleplayer.swf
http://www.youtube.com/watch?v=JCMZZgBURls

and see what happened when they shown the new speech recoginition system...
http://www.youtube.com/watch?v=oiG7KFDYkLI

find in youtube a bit and you will find nice videos like this:
http://www.youtube.com/watch?v=-kql8cWqiv8
http://www.youtube.com/watch?v=YO8cAwf-weo
http://www.youtube.com/watch?v=PyUENvhsHi4
hahahah!

I personally won't pay more than 100$ for the OS... and nope, I don't want IE7, neither Outlook, Paint, Defrag, a disk compressor based on stolen Stacker's tech, MS Office or Media Player/Media center, Firewall, Windows Defender... I can use firefox, thunderbird, GIMP, OpenOffice, LutelWall, Avast! and Totem/VLC... so I don't wanna pay extra $$$ for all those stinky apps and DRM included in Windows.

And you could need the retail version, because if you change your motherboard or CPU the license will be invalidated so you would need to spend more money on a new license! And btw, virtualization is explicitly prohibited by the Vista's OEM license!

Btw, look at the news appeared this week:
http://www.microsoft-watch.com/cont...maybe_macintosh.html?kc=MWRSS02129TX1K0000535
http://exo-blog.blogspot.com/2007/11/vista-sp1-performance-dud.html
http://news.softpedia.com/news/Vist...-039-s-Windows-for-Supercomputers-71157.shtml

And let me remind you tons of european institutions are migrating to free OS solutions in their public administrations... and thats due to the exagerated Windows price:
http://www.linuxelectrons.com/news/linux/largest-linux-migration-europe-city-bergen
http://www.theregister.co.uk/2002/10/30/brussels_to_spend_euro_250k/
http://www.news.com/Munichs-Linux-migration-slips-to-2006/2100-7344_3-5850633.html
http://www.linuxinsider.com/story/35108.html
http://www.channelregister.co.uk/2007/10/13/dutch_consumer_association_declares_war/

Why you would pay 500 or even 200$ in Vista when MacOSX costs 100$ and Solaris and linux are free... I personally find Ubuntu a very good, user friendly and easy to use OS(and btw Compiz is as good if not better than Aero). The only real reason to use Windows is because some apps still don't exist in other OS(3dsmax for example)... but you can find a lot of very popular apps in other OS like XSI, Maya, Skype, GIMP, OpenOffice, Java/gcc/OGL/Mono/Apache/Eclipse, ID's games, etc and the driver support for linux starts to be good.

So, I have an better and cheaper idea... start programming for these OS... and that includes a petition to port the old 3dmarks pls!
 
Last edited by a moderator:
I thought slashdot was the internet's flypaper to keep this sort of thing away from the civilised and informed internet? :rolleyes:
So, I have an better and cheaper idea... start programming for these OS... and that includes a petition to port the old 3dmarks pls!
Chicken meet egg. When the other OS's are commercially viable targets then we'll develop for them - they're not even close to critical mass as far as I can tell.

The whole "Open source is free therefore everything is free and cheaper" is unrealistic. Yes, some elements of truth but if you blindly believe that then you really need to get yourself a few years in the profession - there are so many costs beyond the most obvious ones it's not funny!

Jack
 
So, I have an better and cheaper idea... start programming for these OS... and that includes a petition to port the old 3dmarks pls!
So, in other words, you want to shift the cost away from yourself (the consumer) and just instead spend all your time blaming Microsoft for their evil marketshare and demand that every other company drive themselves into bankruptcy by spending large amounts of time and money developing for 0.5% of the market?

Sure.
 
The whole "Open source is free therefore everything is free and cheaper" is unrealistic. Yes, some elements of truth but if you blindly believe that then you really need to get yourself a few years in the profession - there are so many costs beyond the most obvious ones it's not funny!
In Europe, the public administration is subject to very strict cost controls by the UE commision. If they are moving from Windows to other OSs then is because the change cost is less than the Microsoft's solutions(time to click again in the links you appeared to ignore before). You need also to consider in UE there are no software patents, which reduces costs too.
Take a look into http://www.proposicion.org.ar/doc/noticias-web.html.en and see how many important organizations migrated to linux due to low global and well meditated costs.

When the other OS's are commercially viable targets then we'll develop for them
I would like you to see why linux hosting are much much cheaper than IIS-based ones... or why tons of mobile phones uses linux kernels with Java technology ( android, Symbian, MIDP, etc ) and not Windows Mobile... or why Google, Sun, Mozilla grew using this software model... or how Apple is there... or what OS uses IBM's BlueGene and the NASA... or this SGI's workstation... or why we can get Maya, XSI, ID's games and other commercial apps on linux/mac... or why we ported a lot of pipeline in the enterprise from WinXP/WinServer to linux/MacOSX in a domain managed using Apache some years ago and we are very happy...

There is a non-Windows business world out there!

other company drive themselves into bankruptcy by spending large amounts of time and money developing for 0.5% of the market?
0.5% of the market... are you referring to Vista-DX10-only, don't you? That could be the market share of 3dm2008 if is vista-dx10 only... Well, I think Vista got only a 4% of marketshare after one year... very poor results:
http://marketshare.hitslink.com/report.aspx?qprid=2&qpmr=15&qpdt=1&qpct=3&qptimeframe=Y

If you program using Visual Studio( specially using dynamically-linked model which requires the msvcrt80.dll, msvcrt.dll, msvb6.dll, mscoree.dll, MFC shared dlls, WinSxS DLLs, etc... ), you use .NET >1.0, you use MFC, DirectX, Office/Access/ADO technologies.... you will need to do a lot of work to port your app... so yes, the linux port cost will be inmense... BUT if you plan well you application based on standard, portable and open things like gcc, Eclipse/Netbeans, OpenGL, OpenAL, Java, PostgreSQL(or MySQL)... then and only then, changing a few lines of code you can port it better.. and the costs will be amortized by the not so small ( specially in the server sector where Apache is really strong, see http://news.netcraft.com/archives/2007/01/overallc.gif ) linux/mac/sun community.

What Microsoft wants is to force you to use their exclusive/non-portable/non-open tech and make you very dependent on it ( aka completely Windozed )... so my advice is not to use any exclusive and non-portable technology like Win32, VS, .NET/WPF/Vista API, IIS/ADO, Windows Media or DX10. You can substitute it using Java, GTK+, Mono, OpenGL, posix, OpenAL, NetBSD-style sockets, Apache, Theora/Ogg/XVid, etc... or whatever. For example, see how Maya, XSI, GIMP, OpenOffice or firefox were easy ported using that model... but you need to start your app and plan it well from a start.

Think you can ALSO program Windows with all this... so the code changes won't be so dramatic/costly as you thing.
 
Last edited by a moderator:
This is veering heavily off course. We're still talking about gaming, DX10 and 3dmark right? For now the only viable PC gaming platform is Windows, so everything else is irrelevant to this conversation.

Furthermore these sorts of conversations are just meaningless... for every link you find I can find another one that says exactly the opposite. Thus while I respect the windows cost situation in Europe (although it kind of sounds like you brought it upon yourselves legislatively, but that's really another topic!), it really is pretty irrelevant. Similarly business adoption of Vista has no bearing on gaming/DX10. There was a discussion a while ago about the Steam survey results (look it up), and it appeared to me that people were indeed getting DX10-ready setups at a fairly rapid pace; of course it could always be more rapid (it could have been with DX9 too!), but there's no cause for concern.

If you program using Visual Studio [..] you will need to do a lot of work to port your app
That's totally false... all of our stuff at work compiles fine in gcc, icc and msvc with very few OS-specific sections. Furthermore there are plenty of libraries and functions that only work on Linux or Mac; shall I dare to mention the huge number of OpenGL extensions that Apple decided to create for Mac, in many cases *rather* than supporting the "open" extensions that already existed for the same functionality!? And last I checked, .NET is a pretty well-accepted and developing standard with increasingly good support on non-windows OSes.

Don't get the wrong impression: *everyone* wants to lock you in to their stuff, not just Microsoft. You're being naive to think that Microsoft is a big bad monster but - for instance - Apple is some sort of holy saint.

Think you can ALSO program Windows with all this... so the code changes won't be so dramatic/costly as you thing.
In some cases portable libraries and software is a no-brainer. In some cases it is required even if difficult. However for games it is neither required nor often easy as the "open/portable" libraries heavily lag the platform-specific ones in functionality, efficiency and particularly support and documentation. OpenGL is clearly behind DX10 right now and it will take a severe overhaul to fix that (3.0 won't do it). Similarly stuff like gtk is a long way behind WinForms.

Thus while we'd all love to live in a wonderful work of software compatibility and portability, it just isn't feasible in many cases. For games in particular there's really no need right now and thus I'd prefer that the developers spend their time making it work flawlessly on even just the windows platforms, let alone bringing in even *more* platforms! For that reason, as well as the current game market, I don't think that it's very critical that an OpenGL path be included in the next 3dmark.
 
If you program using Visual Studio [..] you will need to do a lot of work to port your app
That's totally false..
You omited the important part hidding the statement with the [..] about WinSxS DLLs and language dependencies.
Yep, you CAN port pure C/C++ easy but... Try to port your Visual Basic application to linux... Try to port your MFC8 app to linux... Try to port your Managed C++ app to Mono ( go talk M.Icaza about MS CRT and you will discover a bad thing)... Emulators don't count... and if you count then try to install the WinSxS DLLs using WineHQ... Oppz you need IExplorer and MSIInstaller, which won't install.

Btw, have you tryed to install VS2005 on Windows98/ME? Ooopz, you cannot ( manifest problems, WinSxS models, recompiled DLLs, etc). So i'm forced to buy a Win2k/XP and also VS2005 to make a .NET 2.0 application.

And remember... IExplorer/Media system/DLL File protection/other DLLs were embedded into the OS so Microsoft could not extract it... and thats why the UE required the "N" Windows versions.

Don't get the wrong impression: *everyone* wants to lock you in to their stuff, not just Microsoft. You're being naive to think that Microsoft is a big bad monster but - for instance - Apple is some sort of holy saint.
Thats true. But well, I can buy the only-and-all-features-included Leopard OS by just 100$ ( which i think is a decent price for an OS... not 300€ llike Vista is selling here ). Well, Solaris or Ubuntu only accepts $$$ for the support but not the software itself. And well... GNU or Open source foundation only want you to participate.

And last I checked, .NET is a pretty well-accepted and developing standard with increasingly good support on non-windows OSes.
That's what Microsoft wants you to think. The .NET source code exists(google "rotor"), but license does not allow to use it really. We have seen some deals with Novell too... but the .NET as exists in Windows is not ported to other OSs. Novell IS porting the parts which can be ( others cannot, by technical, license or patents reasons )... the project is called "Mono". GNU is working also on other port(google dotGNU), but is very immature.

Mono just works decently (where it means... not perfectly...) with .NET 1.0, that's why I put the >1.0 on the statement. Also contains C# support ... and, AFAIK, lacks VB, MC++ or J#(which is a very curious Java microsoft implementation btw... created from the Sun vs Microsoft Java fighting ).
System.Windows.Forms namespace is not yet totally implemented. Also got some problems because uses Wine emulation and a strange mixed Gtk model ( because GDI+ was patented or something like that ).
.NET 2.0 is only partially supported(in fact, has no official support). VistaAPI(WPF/Avalon, WCF,etc) is not at all done.
Also, MonoForge no longer exists to host your ASP.NET pages... and hosting enterprises don't like Mono because uses too many resources.
Monodevelop (the Mono IDE) is still very immature.
The garbage collector on Mono could use some optimizations ( ask the Unity Engine people or Artificial Studios team about Mono implementation ).
So, even when Microsoft tries to tell you .NET is a standard and portable thing, no... is not really... not yet and perhaps never will be... and if you don't trust me run your .NET app in linux as "./mono myBeautyNET20App.exe" and see.

Due to all this, I was recommending Apache (+php, etc) and Java as an alternative to .NET... but if you can use pure C/C++ will be better... thats, perhaps, why you say your C apps compile and run well on some OSs... but for other .NET languages is other history... and please notice VS is more than just C/C++ !.

Furthermore there are plenty of libraries and functions that only work on Linux or Mac
For posix/unix libraries on Windows need to try compilation with MinGW or Cygwin. They could work if don't have very complex dependencies or are too HW-dependent. Usually they give you the source code so you can modify them too(now go and get the source code for Avalon, DirectX, Windows kernel or some file systems/formats... and remember the EU directives vs Microsoft for this ).
For example, you can port zlib, tifflib, pnglib, posix threads, openGL, openAL, theora, ogg/vorbis, XViD, libxml2, iconv, postgreSQL, gtk+, wxWidgets, etc...

On the other hand I can see more Windows-only libraries which cannot be natively-ported to other OSs.. for example Direct3D, DirectSound, DirectInput, Xaudio 1/2, Winsocks(well, just the async part), MSXML, .NET 1/2/3, ADO/Access/SQLServer, Office controls/doc SDK, Windows Media technologies(ASF/WMV/codecs...), MFC, ATL, etc etc, but well... sometimes is better to use DX to OpenGL as you mention... but requires to code a different path specific for Windows... while the OpenGL path can work on different OSs with a few modifications(OGL3 gonna be a special case, I admit it... i'm talking to port OGL 1.3 for Windows to linux, for example )

And, again, the first block libraries can be used in linux, macos, solaris and Windows too, the second ones just work well on Windows.

I dare to mention the huge number of OpenGL extensions that Apple decided to create for Mac
You still can use glx protocol and Mac X-Windows port. Humus can tell you if was very hard to port his OGL framework to Mac, pls.
And let me remember Microsoft left the OpenGL's ARB and wanted to kill OGL in Vista... so I still prefer a few extensions than Microsoft wanted to do!

This is veering heavily off course.
Yep, is going a bit offtopic. Pls, somebody start a Vista sux/nosux/Microsoft 666/angel /damm portability / windozed thread! I just hope this contributes a bit to open some windozed minds and to get 3dmark ports to other OSs.

I don't think that it's very critical that an OpenGL path be included in the next 3dmark.
Well, with the current OpenGL 2.1(or 3.0) status, I could agree.
The question is if 3dm2008 gonna be Vista and DX10 only. Or DX10.1 only ( so we need to thrash our GF8800s). Perhaps with Mount Evans(3.1) they could do nice next-gen effects without having to require Vista :p and could open some portability possibilities... but Nick had to GLSL "discard" this pixel fragment!
Cmon! almost tell us if is DX10.1 or what! and if you have plans to port 3dmark to other OSs!
 
Last edited by a moderator:
Back
Top