Mac OS X Quicktime, OpenGL Enhancements Coming?

[J Radical]

Did you take a look at the WWDC 06 video of OpenGL demonstration? It was done by ATI, and it was amazing. It had rain that looked real, buildings that looked real, items inside buildings that looked real, well... I think you get the picture.

Perhaps it isn't the engine DirectX vs. OpenGL, but instead the game developers?

I can't find the video you're talking about unfortunately- google tells me that you have to be a member of Apple Developer Connection to access it. So I'll take your word for it that it looks good, but I refuse to believe that it looks nearly as good as what Crysis is doing on DX10.

Ultimately you are right, it is all about the developers- but developers can only do so much with the hardware. The truth is that compared to PCs, Macs do not have competitive graphics cards. Since switching to intel the hardware of Macs and PCs are identical, there is no megahertz myth or anything that can make up the difference between the x1900 and the newer cards.

As for DX10, it is the favored tool for developers and so has a tight grip on a gaming market where good development tools help offset spiraling costs. Making games mac compatible doesn't make a great deal of economic sense unless you are using opengl in the first place- something most studios do not do. I'm all for openGL becoming more competitive with DX10 but it's equally important to ensure the hardware is up to the job.

 

[guzhogi]

You do know that the development time for a new version of OS X is now around 2 years, right?

I agree. Even if it's 1 year between new versions, and we have OS X 10.6-10.9 and then 11, that's 5 years. By that time, pretty much everyone will have an Intel Mac. I can see Apple pulling the plug on PowerPC and only use Intel by whatever comes after 10.5 (10.6?). We'll just have to see.

Also, I think Apple would pull out the G4 & G5 at the same time. Pulling the G4 out before the G5 is stupid because the only Macs that used the G5 were the iMac, Power Mac G5 and xServe. The Mini and both laptops used only G4s before switching to Intels. So if Apple doesn't include support for G4s but includes support for G5s, a lot of people will get POed.

 

[Alloye]

Believe it or not, but XCode just does not work for big software projects. Even the new one that will be out with leopard lacks features necessary for big software development. Most open source projects use XCode, but most commercial software relies on more elaborate systems. I bet a lot of people are actually using visual studio on PCs and have some sorts of plugins to compile for OSX...

As someone who works with VS 2005, Xcode 2.4.1, and the latest Xcode 3.0 preview on a daily basis, I must say that I completely disagree with this statement. IMHO, Xcode is actually much better than VS at managing large projects and also offers a far more advanced build system. VS, on the other hand, is more friendly for newbie developers (just double-click and code) and offers things like better code completion and a nicer debugger. (Without going into detail, let's just say Xcode 3.0 is closing the gap in this area as well.)

FYI, I worked as an SDE for Microsoft for just over five years (four in Office, one in Windows). None of the developers I encountered on either of those teams used VS for anything except the debugger. Most simply used some form of programmer's editor (Source Insight was popular) to edit code. Both products are built with rather complex build systems written in Perl and are managed via a custom version of Perforce called Source Depot.

As for building OS X code on a PC running VS, that's just a fantasy. The toolchains aren't even remotely similar.

 

[nuckinfutz]

I can't find the video you're talking about unfortunately- google tells me that you have to be a member of Apple Developer Connection to access it. So I'll take your word for it that it looks good, but I refuse to believe that it looks nearly as good as what Crysis is doing on DX10.

Ultimately you are right, it is all about the developers- but developers can only do so much with the hardware. The truth is that compared to PCs, Macs do not have competitive graphics cards. Since switching to intel the hardware of Macs and PCs are identical, there is no megahertz myth or anything that can make up the difference between the x1900 and the newer cards.

As for DX10, it is the favored tool for developers and so has a tight grip on a gaming market where good development tools help offset spiraling costs. Making games mac compatible doesn't make a great deal of economic sense unless you are using opengl in the first place- something most studios do not do. I'm all for openGL becoming more competitive with DX10 but it's equally important to ensure the hardware is up to the job.

It does. That ATI modified the scene right on the fly and then showed how it was textured (which was very efficient). Apple isn't claiming a Megahertz myth. They close gap instantly the day they announce G80 based product or ATI's top GPU parts.

DX10 if favored for games. OpenGL forms the graphics enging and backing of Apple's whole OS. DX10 isn't important in any other avenue than PC gaming and Bootcamp will handle it just fine.

In Leopard the GPU will have a constant thread for handling UI rendering whilst still processing other tasks. We've already seen demos where this increases WoW fps.

It's pretty clear why Apple hasn't announced new GPU. Do people really expect them to deliver OpenGL 1.4 drivers for Tiger for the newest parts when Tiger is due to be replaced in two months with Leopard? That's silly...Apple will deliver OpengL 2.1 drivers for the newest ATI and Nvidia parts with a vast improvement in shader language features. June at the latest.

I doubt there's much in DX10 that cannot be done in OpenGL 2.1 graphicswise.

 

[shawnce]

QuickTime under went a major under the hood rework for 10.4 to use a more modern pipeline centered on OpenGL, Core Video and Core Audio. The descriptions given really don't appear to point to much beyond what has already been done... not that more rework and improvements cannot still be done in this space.

...and yes stuff is improving in this space and OpenGL for 10.5 but that is all that can be said at this time.

 

[shawnce]

As someone who works with VS 2005, Xcode 2.4.1, and the latest Xcode 3.0 preview on a daily basis, I must say that I completely disagree with this statement. IMHO, Xcode is actually much better than VS at managing large projects and also offers a far more advanced build system. VS, on the other hand, is more friendly for newbie developers (just double-click and code) and offers things like better code completion and a nicer debugger. (Without going into detail, let's just say Xcode 3.0 is closing the gap in this area as well.)
...
As for building OS X code on a PC running VS, that's just a fantasy. The toolchains aren't even remotely similar.

I want to second this... I work on a family of products that are made from a large mostly shared code base (foundational libraries and end customer applications). On Mac OS X we use Xcode 2.4.1 to build all of the various libraries and applications (in some cases use Xcode to drive some make based builds). In general the Xcode project based build system is easier to maintain and more flexible then the VS based one used on the Windows side of things (they actually purchased some 3rd party products to extend VS to allow give them some missing functionality).

Xcode works well if you don't fight it but instead leverage the paradigm it is designed around (this is true for any tool chain). I say that because I see lots of folks trying to make Xcode work like CodeWarrior or VS instead of really understanding and utilizing Xcode ... then they bitch about "how bad it is" when in fact it is primarily their lack of understanding of Xcode.

Oh... and thanks to Xcode's built-in multi-threaded build system we can build the Mac versions of our products in much less time then Windows versions... even when compiling for twice as many processor architectures (PowerPC and Intel). Nice to compile four files at once without having to lift a finger configuration / tool chain wise.

Yeah I would like better auto-completion, editor and source debugger in Xcode but .... hey is that Xcode 3.0 I see? Oh and we are really excited about IB 3.0... going to let us build better native GUI in less time then Window's team we have here.. suckers

 

[Rod Rod]

10.9 the pulling out of G5 support.
11.0 no G5 support.

What about 10.10, 10.11, 10.12, 10.13? These aren't decimals.

 

[commander.data]

It's pretty clear why Apple hasn't announced new GPU. Do people really expect them to deliver OpenGL 1.4 drivers for Tiger for the newest parts when Tiger is due to be replaced in two months with Leopard? That's silly...Apple will deliver OpengL 2.1 drivers for the newest ATI and Nvidia parts with a vast improvement in shader language features. June at the latest.

I doubt there's much in DX10 that cannot be done in OpenGL 2.1 graphicswise.

Is it harder to program in OpenGL than DirectX? I would guess DirectX is easier otherwise I would think developers would just use OpenGL since it's more portable. I suppose it may be that part of it may be that ATI's GPUs always seem to be weaker at OpenGL than nVidia.

In terms of graphics card selection I can understand why Apple doesn't introduce higher end cards. If Apple has to code their own drivers for each GPU it would make sense that they would choose specific GPUs for each market like a lower-high end X1900XT which is still pretty fast, while has a larger market that than a top-of-the-line model. For notebooks, there really was no reason for Apple to update since the MR X1700 and Go 7700 aren't much faster than the MR X1600 and anything higher end wouldn't fit in the 15.4" MBP's thermal envelop. The 17" MBP could use a better GPU, but I guess they just couldn't justify a separate supply chain. I wonder if the Intel transition makes it easy to port drivers over from Windows since that could help Apple introduce a wider GPU selection.

Is Tiger still using OpenGL 1.4? Tiger was released in 2005 and by that time OpenGL 2.0 had already been around for more than half a year. I'm not sure what exactly is the difference between all the versions, but I would think Tiger should have launched with at least OpenGl or be updated to OpenGL 2.0 by now. I wonder if using older versions of OpenGL prevent use of some modern GPU features or limit the card's performance.

 

[mdriftmeyer]

Real simple. Both QuickTime and OpenGL will have core APIs written for Cocoa.

 

[zolotroph]

OpenGL is a graphic library that to my knowledge was developed for Unix/Linux. Since OSX is a variant of that it can also run it. It is mainly used for 3D games. Making it faster makes Mac's a better gaming choice.

Actually, OpenGL is a graphics API initially developed by Silicon Graphics, Inc. It is primarily used for professional 3D graphics applications such as modeling, animation, CAD and visualization software. OpenGL is truly cross-platform, running on all major flavors of UNIX, Linux, Mac OS and Windows.

Virtually all current professional graphics applications use OpenGL, even on Windows. Professional OpenGL graphics cards like ATI's FireGL and nVidia's Quadro series cost much more than their non-professional gaming card equivalents because of the added cost of application-certified OpenGL drivers and support for professional applications.

 

[McScooby]

Just register with a free a/c,

I can't find the video you're talking about unfortunately- google tells me that you have to be a member of Apple Developer Connection to access it. So I'll take your word for it that it looks good, but I refuse to believe that it looks nearly as good as what Crysis is doing on DX10.

Ultimately you are right, it is all about the developers- but developers can only do so much with the hardware. The truth is that compared to PCs, Macs do not have competitive graphics cards. Since switching to intel the hardware of Macs and PCs are identical, there is no megahertz myth or anything that can make up the difference between the x1900 and the newer cards.

As for DX10, it is the favored tool for developers and so has a tight grip on a gaming market where good development tools help offset spiraling costs. Making games mac compatible doesn't make a great deal of economic sense unless you are using opengl in the first place- something most studios do not do. I'm all for openGL becoming more competitive with DX10 but it's equally important to ensure the hardware is up to the job.

as a student dev, which basically anyone with an active e/m a/c can do and watch via itunes!

 

[nuckinfutz]

Is it harder to program in OpenGL than DirectX? I would guess DirectX is easier otherwise I would think developers would just use OpenGL since it's more portable. I suppose it may be that part of it may be that ATI's GPUs always seem to be weaker at OpenGL than nVidia.

In terms of graphics card selection I can understand why Apple doesn't introduce higher end cards. If Apple has to code their own drivers for each GPU it would make sense that they would choose specific GPUs for each market like a lower-high end X1900XT which is still pretty fast, while has a larger market that than a top-of-the-line model. For notebooks, there really was no reason for Apple to update since the MR X1700 and Go 7700 aren't much faster than the MR X1600 and anything higher end wouldn't fit in the 15.4" MBP's thermal envelop. The 17" MBP could use a better GPU, but I guess they just couldn't justify a separate supply chain. I wonder if the Intel transition makes it easy to port drivers over from Windows since that could help Apple introduce a wider GPU selection.

Is Tiger still using OpenGL 1.4? Tiger was released in 2005 and by that time OpenGL 2.0 had already been around for more than half a year. I'm not sure what exactly is the difference between all the versions, but I would think Tiger should have launched with at least OpenGl or be updated to OpenGL 2.0 by now. I wonder if using older versions of OpenGL prevent use of some modern GPU features or limit the card's performance.

I'm sure that Microsoft probably has easier programming tools for DX9. That's their coupe de grace with most tools. I haven't read much about the programming ease/difficulty of OpenGL 2.1. I do know that many of the ARB members want to promote a "Lean and Mean" subset for rapid advancements (in particular AMD/ATI and Nvidia).

Yeah I was bummed that Tiger didn't have OpenGL 2.0 but hindsight being 20/20 I understand why. 1.4 doesn't have the GLSL shaders and they weren't prepared to have Tiger support GLSL and OpenGL 2.0 when Tiger was being developed. Rather than upgrade Tiger they just pushed OpenGL 2.x to Leopard and tied the development in with QuartzGL (formerly Quartz 2D Extreme) and the multithreading in Leopard. I guess the question is will Apple implement upcoming OpenGL revisions "Longs Peak" and "Mt Evans" ,due this year, into Leopard or will they again bypass the extensions and deliver what will be known as OpenGL 3.0 in 10.6 in 2009?

http://www.opengl.org/pipeline/article/vol002_1/

The main task for the ARB is to deliver two new OpenGL releases in 2007. The first one, code named OpenGL "Longs Peak" (the actual releases will have version numbers), is slated to be released in summer 2007. The second one, code named OpenGL "Mt. Evans", is targeted for an October 2007 release. Why code names? We want to give the ARB's marketing group a chance to think through what the right names would be for these releases. Too many suggestions have already been made, including OpenGL 2.2, OpenGL 3.0, OpenGL 3.1 and even OpenGL 4.0. This is not the time yet to pin down the version number, and therefore we'll be using code names.

OpenGL Longs Peak will be a significant departure for us. While there will still be backwards API compatibility, the new "Lean and Mean" profile, and a substantial refactoring in terms of the new object model, make it in many ways an entirely new API design. This is an ambitious task and requires a high degree of commitment from the ARB members. We are already seeing some welcome participation from Khronos members who were not members of the old ARB, and hope to see much more.
While OpenGL Longs Peak will be implementable on current and last generation hardware, OpenGL Mt. Evans will only be implementable on the newest hardware available. The OpenGL Mt. Evans release will be a continuation of OpenGL Longs Peak, with a lot of new functionality added. Some of the highlights are: geometry shading, a more central role for buffer objects, and a full integer pipeline accessible via the OpenGL Shading Language.

Why two new OpenGL releases do you ask? This split in two makes it easy for ISVs to develop a new title spanning a wide range of graphics hardware. By coding their core rendering engine to OpenGL Longs Peak, both older and the newest hardware will be covered. By then coding incrementally to the OpenGL Mt. Evans API, the newest hardware can be exploited to its maximum potential.

My emphasis added. Sounds interesting. I'd like for Apple to continuously evolve Leopard adding in support for the extensions as they can test them but know knows? I know this..we need not fear DirectX10. Longs Peak will deliver a majority of what DX10 offers with the added benefit of backwards compatibility to existing Unix, XP,Linux and more. Apple likely won't upgrade Tiger but Leopard's a possibility.

 

[zolotroph]

Also Direct X > OpenGl

This is not even remotely true. Suggested reading:

http://en.wikipedia.org/wiki/Comparison_of_Direct3D_and_OpenGL

Excerpt:

"With the exception of Windows and the Xbox, all operating systems that allow for hardware-accelerated 3D graphics have chosen OpenGL as their primary graphics API."

"The principal reason for Direct3D's dominance in the gaming industry is historical."

"Making games that use OpenGL while using the non-Direct3D portion of the DirectX API is no more difficult than making a game using all of the DirectX API. The decision to use Direct3D over OpenGL was made from simple pragmatism: in those [early] days, OpenGL implementations were difficult to work with."

"Later versions of OpenGL would rarely add functionality that wasn't actually widely available in hardware. As such, the issue has, for the most part, become a non-issue."

As someone pointed out earlier, DirectX is only usable now because of the technology Microsoft siphoned from SGI.

 

[zolotroph]

This may make the whole DirectX vs. OpenGL argument a moot point:

TransGaming, nVidia teaming up to bring high-end games to the Mac

http://apple.qj.net/TransGaming-nVidia-teaming-up-to-bring-high-end-games-to-the-Mac/pg/49/aid/82038

 

[nuckinfutz]

This may make the whole DirectX vs. OpenGL argument a moot point:

TransGaming, nVidia teaming up to bring high-end games to the Mac

http://apple.qj.net/TransGaming-nVidia-teaming-up-to-bring-high-end-games-to-the-Mac/pg/49/aid/82038

Hmmm that's pretty cool. I think j.radical is just getting a wee bit too excited about gaming. If you hang out in gaming areas DX10 is going to seem like God returning. But if you expand your scope and look at the 3D overall you see the OpenGL not only has maintained their ground but in 2007 will really extend the API via "Long Peaks" and "Mt Evans" Siggraph 2007 should be fun.

Saying DX10 is greater than OpenGL is simply foolish. You must narrow the context for such a statement to be made. For if someone is using a Mac, Linux or Unix based OS DX10 is a non starter. Hell I'm not sure DX10 will run on XP for the forseeable future.

 

[commander.data]

Yeah I was bummed that Tiger didn't have OpenGL 2.0 but hindsight being 20/20 I understand why. 1.4 doesn't have the GLSL shaders and they weren't prepared to have Tiger support GLSL and OpenGL 2.0 when Tiger was being developed. Rather than upgrade Tiger they just pushed OpenGL 2.x to Leopard and tied the development in with QuartzGL (formerly Quartz 2D Extreme) and the multithreading in Leopard. I guess the question is will Apple implement upcoming OpenGL revisions "Longs Peak" and "Mt Evans" ,due this year, into Leopard or will they again bypass the extensions and deliver what will be known as OpenGL 3.0 in 10.6 in 2009?

http://www.opengl.org/pipeline/article/vol002_1/

My emphasis added. Sounds interesting. I'd like for Apple to continuously evolve Leopard adding in support for the extensions as they can test them but know knows? I know this..we need not fear DirectX10. Longs Peak will deliver a majority of what DX10 offers with the added benefit of backwards compatibility to existing Unix, XP,Linux and more. Apple likely won't upgrade Tiger but Leopard's a possibility.

I guess I can understand why Apple didn't integrate OpenGL 2.0 into Tiger since I guess GLSL is probably a pretty big change. I wonder why they didn't include OpenGL 1.5 though? I guess it wasn't really a major change from OpenGL 1.4 anyways.

What you linked about Long Peaks is very interesting. I'm hoping that when ThinkSecret said the OpenGL API has been overhauled, maybe they were meaning Long Peaks support, since Long Peak definitely seems like a major overhaul. It would be a shame if Apple just missed it just like they just missed OpenGL 2.0. I suppose they other performance improvement could be making the OpenGL API 64-bit. I think having multitheaded and 64-bit games for Leopard would definitely boost performance. Of course porting 32-bit games to 64-bit in addition to porting Windows to Mac might not be the easiest thing.

 

[ChildOL]

FIX FLASH FIX FLASH OMG FIX FLASH PLEASE!

Flash has been a long time hog on the Mac and I think it is about time to fix the issue wth the extreme sluggisness flash on the mac compared to Windows flash! The internet is becoming more and more flash based, expecially video like youtube, google video, yahoo video, etc. etc.

I use a $3,000 Dual 533Mhz Processor PowerMac G4 system with 1.25GB of RAM as my main computer and everything is very fast EXCEPT FLASH!

I hope this OpenGL upgrade will greatly improve flash performance.

Those of you who think maybe in the next version or two of OSX that Apple may drop support for PowerPC/PPC don't seem to realize that over 90% of Macs out there are PowerPC/PPC based, so why in the world would Apple drop support for almost all of their installed base?! It will be many years before PPC is totally dropped, maybe 7-10 years IF they do. The G3 could be dropped soon BUT the G4 and G5 both have a vector processing unit so they may be dropped at the same time, and that would be many years from now, IF they drop it at all.

 

[commander.data]

Those of you who think maybe in the next version or two of OSX that Apple may drop support for PowerPC/PPC don't seem to realize that over 90% of Macs out there are PowerPC/PPC based, so why in the world would Apple drop support for almost all of their installed base?! It will be many years before PPC is totally dropped, maybe 7-10 years IF they do. The G3 could be dropped soon BUT the G4 and G5 both have a vector processing unit so they may be dropped at the same time, and that would be many years from now, IF they drop it at all.

I agree that there is no way that Apple would drop PPC in Leopard. It just makes no sense considering Apple's history of good legacy support. As others have said 10.5 will likely drop the G3. I don't see the G4 and G5 being dropped at the same time though since the G5 has the triple marketing goodness of high-bandwidth FSBs, 4 core support, and 64-bit support to differentiate it from the G4, with the last two probably being the most important. Since 10.5 looks to be the first full 64-bit OS from Apple, it can be viewed as the introduction of broad scale 64-bit support, and so by 10.6 (probably 2009-2010) 64-bit will have become mainstay and the G4 will be dropped. The G5 can then hang-on until 10.7 when it will be dropped, which by the time it's released will be early next decade (probably 2011+) and half a decade after the Intel switch, so assuming Apple doesn't switch back to PPC, it'll be safe to assume that the loss of PPC support won't be a huge hardship to the remaining users.

 

[ChildOL]

I agree that there is no way that Apple would drop PPC in Leopard. It just makes no sense considering Apple's history of good legacy support. As others have said 10.5 will likely drop the G3. I don't see the G4 and G5 being dropped at the same time though since the G5 has the triple marketing goodness of high-bandwidth FSBs, 4 core support, and 64-bit support to differentiate it from the G4, with the last two probably being the most important. Since 10.5 looks to be the first full 64-bit OS from Apple, it can be viewed as the introduction of broad scale 64-bit support, and so by 10.6 (probably 2009-2010) 64-bit will have become mainstay and the G4 will be dropped. The G5 can then hang-on until 10.7 when it will be dropped, which by the time it's released will be early next decade (probably 2011+) and half a decade after the Intel switch, so assuming Apple doesn't switch back to PPC, it'll be safe to assume that the loss of PPC support won't be a huge hardship to the remaining users.

I just can't see Apple dropping the PPC support until the next major OS i.e. 11.0

The reason I said G4 and G5 may be dropped together is because the G4's high installed base and the G5's power and "newness". I am still confused as to why people keep saying by OS X 10.7 which is way too soon to drop PPC support. Like I said almost all the people owning Macs are on a PowerPC based Mac. Maybe it's just the Intel Mac noobies thinking they are the only ones using Macs and thinking that most macs or all macs are Intel based, IDK.

If Apple drops PPC support in less than 5 years it would be very bad for their user base, UNLESS there is some mass exodus of people switching to the Mac getting Intel Macs. For some reason There are MANY people who still believe the PowerPC is the "better" processor being RISC based etc. and would rather stick with a "rocket" G4 or G5.

And what about fixing the slow flash problems? *sigh*

 

[trainguy77]

I use a $3,000 Dual 533Mhz Processor PowerMac G4 system with 1.25GB of RAM as my main computer and everything is very fast EXCEPT FLASH!

You do know its not worth $3000 anymore....not even close.

 

[RedTomato]

FIX FLASH FIX FLASH OMG FIX FLASH PLEASE!

I use a $3,000 Dual 533Mhz Processor PowerMac G4 system with 1.25GB of RAM as my main computer and everything is very fast EXCEPT FLASH!

You do know its not worth $3000 anymore....not even close.

I have a 350mhz G3 with 1GB RAM. For flash purposes, it's the same as yours. (showing flash won't use both cores or probably the altivec either). It was also sold as a supercomputer or something like that when it came out.

I brought it for £50 last year.

And yes, flash, YouTube, GoogleVideo etc are like watching slideshows. It does ok as a server and for minor wordprocessing / small surfing / DVD watching.

 

[eightd]

when does quicktime and opengl improvements NOT come with each system upgrade? ts gotta be more specific then "improvements"

 

[milatchi]

I thought that QuickTime 7 was suppose to be the major removal of legacy code? Oh Apple always cleaning out legacy code. Well if it helps QT, I'm for it.

As for OpenGL: I don't think it's so much an issue of being able to compete speed wise and performance wise. I think it's a matter of weening and/or luring Windows developers from DirectX

 

[diamond.g]

DirectX Wrapper

It seems like there may be some hope after all. Slashdot is reporting that there is a wrapper for DX10 to OpenGL. That could be quite a boon if Apple manages to get OpenGL running super quick on the Mac's.

 

[Fukui]

Most people use Code Warrior. It's old, but so is lots of the code and it can be a pain in the ass to migrate everything. Wolfram Research for instance does not use Xcode, even if Apple showed on a keynote how easy it is to click the little checkbox and recompile for Intel.

The big downside of Xcode is that you can't use it if you want to make an application that runs on OSX and Windows. No one wants to have competely different environments for the same product.

With plug-ins like for C# (Mono) or Objective-C (GNUStep) and pure Java, you can do cross-platform work from Xcode. While not mature, the first two are examples that while apple may not directly support cross-platform work initially from xcode outside of Java, its possible to extend it. GCC does compile for several architectures...