I'm waiting for a Hardware & Software upgrade as well, i want an iMac, just don't need one desperately right now. But i didn't get one thing: Does the iMac get Santa Rosa first and later this year the Penryn chips? Or is Penryn desktop/laptop only and the iMac sticks with Santa Rosa for a while?
imac. 20" or 24". not urgent so i'll wait until june at least although it would be nice the sooner the better.
might be interested in a widescreen ipod although i think it would have to have something else going from it since my 60gig video is fine and dandy still
I have been a PC user since I started using a computer. I now hate Microsoft and am going to buy myself a octo-core Mac Pro. Cant wait. Hope they feature some new graphic cards at NAB though, been waiting for ages like the rest of you.
I have been a PC user since I started using a computer. I now hate Microsoft and am going to buy myself a octo-core Mac Pro. Cant wait. Hope they feature some new graphic cards at NAB though, been waiting for ages like the rest of you.
Holding out for a second gen 8 core mac pro, blue ray would be nice, and i'm not sure how long i can sit on my money for that 30" monitor i've ben having at least one eye on for years. having seen it drop from 5 to 2 grand should be long enough, but who knows about that resolution and led backlit rumour...
also hoping for an ultraprotable macbook or pro to replace my newton... ;-)
I have been a PC user since I started using a computer. I now hate Microsoft and am going to buy myself a octo-core Mac Pro. Cant wait. Hope they feature some new graphic cards at NAB though, been waiting for ages like the rest of you.
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
Forgive my igornace but would the addition of Penryn make such a large difference to something like the MacPro over the current 8 core setup. As I said, I have waited till the octo came out. It is here no much has changed. I intend to buy but am not fussed over a few months. Sorry for the noob question. Thanks in advance.
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
Need a laptop bad for the end of May when I will be travelling. In a perfect world, both the Macbook would be updated, and Leopard would be released by then.
<crossing fingers>
If just the Macbook is updated I'll still buy and bite the bullet on Leopard.
Or if Leopard is released but the macbook isn't updated, I'll buy and bite the bullet on the book.
OPenGL 2.1, and Leopard's LLVM info (Low Level Virtual Machine.)
Quote:
Originally Posted by wikipedia
Leopard's OpenGL stack has been updated to version 2.1, and will use LLVM to increase its vertex processing speed. Apple has been working to get LLVM integrated into GCC; Usage of LLVM in other parts of the OS has not been announced.
Quote:
Originally Posted by Chris Lattner sabre at nondot.org
I just got official okay to mention this in public. This was previously
announced at Apple's WWDC conference last week.]
For those who are interested, Apple announced that they are using the LLVM
optimizer and JIT within their Mac OS 10.5 'Leopard' OpenGL stack (which
was distributed in beta form to WWDC attendees).
LLVM is used in two different ways, at runtime:
1. Runtime code specialization within the fixed-function vertex-processing
pipeline. Basically, the OpenGL pipeline has many parameters (is fog
enabled? do vertices have texture info? etc) which rarely change:
executing the fully branchy code swamps the branch predictors and
performs poorly. To solve this, the code is precompiled to LLVM .bc
form, from which specializations of the code are made, optimized,
and JIT compiled as they are needed at runtime.
2. OpenGL vertex shaders are small programs written using a family of
programming langauges with highly domain-specific features (e.g. dot
product, texture lookup, etc). At runtime, the OpenGL stack translates
vertex programs into LLVM form, runs LLVM optimizer passes and then JIT
compiles the code.
Both of these approaches make heavy use of manually vectorized code using
SSE/Altivec intrinsics, and they use the LLVM x86-32/x86-64/ppc/ppc64
targets. LLVM replaces existing special purpose JIT compilers built by
the OpenGL team.
LLVM is currently used when hardware support is disabled or when the
current hardware does not support a feature requested by the user app.
This happens most often on low-end graphics chips (e.g. integrated
graphics), but can happen even with the high-end graphics when advanced
capabilities are used.
Like any good compiler, the only impact that LLVM has on the OpenGL stack
is better performance (there are no user-visible knobs). However, if you
sample a program using shark, you will occasionally see LLVM methods in
The way I interpreted what onlooker just quoted is...
The LLVM will govern what graphics effects get used and displayed by only switching on the effects the graphics card can handle, other effects get switched off, so that they aren't running and slowing the applications down for no reason.
Kind of optimising the graphics output to the capability of the machine, right??
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
Quote:
Originally Posted by BenRoethig
Newer, OpenGL 2.1
Quote:
Originally Posted by regan
Well, heres hoping.
Need a laptop bad for the end of May when I will be travelling. In a perfect world, both the Macbook would be updated, and Leopard would be released by then.
<crossing fingers>
If just the Macbook is updated I'll still buy and bite the bullet on Leopard.
Or if Leopard is released but the macbook isn't updated, I'll buy and bite the bullet on the book.
We shall see. :-)
Quote:
Originally Posted by onlooker
OPenGL 2.1, and Leopard's LLVM info (Low Level Virtual Machine.)
Getting Anxious over here.
Quote:
Originally Posted by lfe2211
Onlooker,
Please expand your post "Getting Anxious over here."
Thanks,
That is the expansion. I'm merely posting some documentation I found on items that are occurring in the current discussion.
I'm Anxious to see leopard and even more Anxious to see OpenGL compete in areas that Direct X has been dominating. I didn't realize it was that difficult to understand. Sorry.
I've been eyeballing the new Mac Pro and am holding off until NAB just incase there are new graphics cards, or new monitors revealed. If they don't give us any new GPU's I'll have to drop the cash for a Quadro FX 4500. I'd rather see newer offerings than what Apple offers.
It causes me concern if they are talking about bottle necks. I'd like to hear what you all think about the review.
If you're dealing with an app that requires a lot of memory swapping a la photoshop then the Octo isn't a huge update. If you're doing a lot of encoding or rendering then the Octo shows a marked improvement in speed.
The problem right now is two fold.
1. Apple doesn't have enough intelligence in their threading so that we have core affinity in Tiger. Threads will move to different procs when they should be be locked to a single proc/cache. Leopard will improve this
2. Bensley, the chipset of current Mac Pro, has a snoop filter that only tracks two procs and the shared cache. When you move to a Quad Core system Bensely if inefficient in determining what goes where" beyond two procs. Stoakley/Seaburg and Penryn will fix this because Seaburg the controller will know "what goes where" for 4 procs with an improved snoop filter.
If I needed power right now and I wasn't running heavy CPU bound stuff I'd qo Quad and skip the Penryn/SS architecture and wait for Nehalem to hit in late 2008 early 2009 before upgrading.
Comments
might be interested in a widescreen ipod although i think it would have to have something else going from it since my 60gig video is fine and dandy still
I have been a PC user since I started using a computer. I now hate Microsoft and am going to buy myself a octo-core Mac Pro. Cant wait. Hope they feature some new graphic cards at NAB though, been waiting for ages like the rest of you.
I'm envious. Have fun with your new computer.
Lemon Bon Bon
I was...until I scored a 20" Core Duo iMac for $950 CAD locally
Your sig says that you have a 20" 2Ghz Intel iMac... that's not possible, they start at 2.16 Ghz.
also hoping for an ultraprotable macbook or pro to replace my newton... ;-)
I have been a PC user since I started using a computer. I now hate Microsoft and am going to buy myself a octo-core Mac Pro. Cant wait. Hope they feature some new graphic cards at NAB though, been waiting for ages like the rest of you.
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
Newer, OpenGL 2.1
Need a laptop bad for the end of May when I will be travelling. In a perfect world, both the Macbook would be updated, and Leopard would be released by then.
<crossing fingers>
If just the Macbook is updated I'll still buy and bite the bullet on Leopard.
Or if Leopard is released but the macbook isn't updated, I'll buy and bite the bullet on the book.
We shall see. :-)
Leopard's OpenGL stack has been updated to version 2.1, and will use LLVM to increase its vertex processing speed. Apple has been working to get LLVM integrated into GCC; Usage of LLVM in other parts of the OS has not been announced.
I just got official okay to mention this in public. This was previously
announced at Apple's WWDC conference last week.]
For those who are interested, Apple announced that they are using the LLVM
optimizer and JIT within their Mac OS 10.5 'Leopard' OpenGL stack (which
was distributed in beta form to WWDC attendees).
LLVM is used in two different ways, at runtime:
1. Runtime code specialization within the fixed-function vertex-processing
pipeline. Basically, the OpenGL pipeline has many parameters (is fog
enabled? do vertices have texture info? etc) which rarely change:
executing the fully branchy code swamps the branch predictors and
performs poorly. To solve this, the code is precompiled to LLVM .bc
form, from which specializations of the code are made, optimized,
and JIT compiled as they are needed at runtime.
2. OpenGL vertex shaders are small programs written using a family of
programming langauges with highly domain-specific features (e.g. dot
product, texture lookup, etc). At runtime, the OpenGL stack translates
vertex programs into LLVM form, runs LLVM optimizer passes and then JIT
compiles the code.
Both of these approaches make heavy use of manually vectorized code using
SSE/Altivec intrinsics, and they use the LLVM x86-32/x86-64/ppc/ppc64
targets. LLVM replaces existing special purpose JIT compilers built by
the OpenGL team.
LLVM is currently used when hardware support is disabled or when the
current hardware does not support a feature requested by the user app.
This happens most often on low-end graphics chips (e.g. integrated
graphics), but can happen even with the high-end graphics when advanced
capabilities are used.
Like any good compiler, the only impact that LLVM has on the OpenGL stack
is better performance (there are no user-visible knobs). However, if you
sample a program using shark, you will occasionally see LLVM methods in
the stack traces.
Getting Anxious over here.
The LLVM will govern what graphics effects get used and displayed by only switching on the effects the graphics card can handle, other effects get switched off, so that they aren't running and slowing the applications down for no reason.
Kind of optimising the graphics output to the capability of the machine, right??
Please expand your post "Getting Anxious over here."
Thanks,
If it's true that Leopard will come out with a major new OpenGL version (2.0 I suppose), and if it's further true that Leopard will be released around WWDC in June, I'd say any new high-end graphics card would probably wait for that too.
They're not going to sell even nearly the numbers as for PC hence I doubt that whoever develops the drivers will develop one for OS X 10.4 (OpenGL 1.4) AND one for OS X 10.5 (OpenGL 2.0). They'd probably settle for one version only. And that would likely be the new one.
Newer, OpenGL 2.1
Well, heres hoping.
Need a laptop bad for the end of May when I will be travelling. In a perfect world, both the Macbook would be updated, and Leopard would be released by then.
<crossing fingers>
If just the Macbook is updated I'll still buy and bite the bullet on Leopard.
Or if Leopard is released but the macbook isn't updated, I'll buy and bite the bullet on the book.
We shall see. :-)
OPenGL 2.1, and Leopard's LLVM info (Low Level Virtual Machine.)
Getting Anxious over here.
Onlooker,
Please expand your post "Getting Anxious over here."
Thanks,
That is the expansion. I'm merely posting some documentation I found on items that are occurring in the current discussion.
I'm Anxious to see leopard and even more Anxious to see OpenGL compete in areas that Direct X has been dominating. I didn't realize it was that difficult to understand. Sorry.
Just a few more days...
REVIEW
It causes me concern if they are talking about bottle necks. I'd like to hear what you all think about the review.
I also found this review of the Octo-Mac Pro:
REVIEW
It causes me concern if they are talking about bottle necks. I'd like to hear what you all think about the review.
If you're dealing with an app that requires a lot of memory swapping a la photoshop then the Octo isn't a huge update. If you're doing a lot of encoding or rendering then the Octo shows a marked improvement in speed.
The problem right now is two fold.
1. Apple doesn't have enough intelligence in their threading so that we have core affinity in Tiger. Threads will move to different procs when they should be be locked to a single proc/cache. Leopard will improve this
2. Bensley, the chipset of current Mac Pro, has a snoop filter that only tracks two procs and the shared cache. When you move to a Quad Core system Bensely if inefficient in determining what goes where" beyond two procs. Stoakley/Seaburg and Penryn will fix this because Seaburg the controller will know "what goes where" for 4 procs with an improved snoop filter.
If I needed power right now and I wasn't running heavy CPU bound stuff I'd qo Quad and skip the Penryn/SS architecture and wait for Nehalem to hit in late 2008 early 2009 before upgrading.