Lt Major Burns
"Dicky" Charlteston-Burns
i mean is it like passing through the gates of heaven? or what? do you even notice a difference? does it just make it faster, or is it cool effects and stuff?
What difference does Quartz 2D Extreme make if you turn it on?
- Thread starter Lt Major Burns
- Start date
brianleahy
Colonel Panic
If you turn it on, does it stay on after a restart, or does it have to be switched back on each time?
Lt Major Burns
"Dicky" Charlteston-Burns
switched on each time. is there a walk through on the forum?
Lt Major Burns
"Dicky" Charlteston-Burns
how do i install the developer tools? it doesn't seem to be on the DVD. wait - i didn't have to be a tiger developer did i? because i wasn't one.
brianleahy
Colonel Panic
XCode = Developer Tools
Look for the XCode folder on the DVD.
Look for the XCode folder on the DVD.
Lt Major Burns
"Dicky" Charlteston-Burns
got it. been playing with it. not very EXTREME! is it? it set windows resizing to be ever so slightly faster. and you can turn it on so it's incredibly buggy, but then turn the "make it buggy" feature off again. and you can resize the GUI. but none of the "graphic" bits in it are scalable. so, again kinda useless. all it seems to be is a bugfix for the slow window resizing. i'd hardly call it...
**EXTREME!!**
**EXTREME!!**
It will only be active in application you launch, after you launch Quartz Debug and enable Q2DE, so you won't notice any difference in the app's you already had running. Try quitting them and launching them again.
If you don't notice a real change, then chances are that you don't have a fast enough graphics card. One quick way to find out: switch to dashboard and add a widget. Do you get a "ripple effect" -- where it looks like your screen was a water surface that you just dropped the widget on -- when you activate it? If you do, your graphics card is fast enough to give you Q2DE, if you don't, it isn't.
If you don't notice a real change, then chances are that you don't have a fast enough graphics card. One quick way to find out: switch to dashboard and add a widget. Do you get a "ripple effect" -- where it looks like your screen was a water surface that you just dropped the widget on -- when you activate it? If you do, your graphics card is fast enough to give you Q2DE, if you don't, it isn't.
Lt Major Burns
"Dicky" Charlteston-Burns
oh i have a 128MB radeon 9600. the ripple is very sexy. (if useless?)
and everytime i changed something in Q2dx i made sure i quite safari or relaunched the finder.
is there something i'm not seeing that other people are? or is window resizing an extreme sport in california? are there any cool effects i can play with in q2dx?
and everytime i changed something in Q2dx i made sure i quite safari or relaunched the finder.
is there something i'm not seeing that other people are? or is window resizing an extreme sport in california? are there any cool effects i can play with in q2dx?
scruffy
Notorious Olive Counter
There's a very in depth technical article on Tiger, on arstechnica.com (very well written and worth a read - http://arstechnica.com/reviews/os/macosx-10.4.ars). There is a thorough examination of Quartz 2DExtreme at http://arstechnica.com/reviews/os/macosx-10.4.ars/14
Maybe try a real torture test something like this: Open about 50 terminal windows, all set to about 99% transparent, and then layer them all over top of a playing Quicktime movie. Then grab the topmost terminal window and resize it like crazy, wiggle it around... See how the movie does with and without Quartz 2D Extreme
Maybe try a real torture test something like this: Open about 50 terminal windows, all set to about 99% transparent, and then layer them all over top of a playing Quicktime movie. Then grab the topmost terminal window and resize it like crazy, wiggle it around... See how the movie does with and without Quartz 2D Extreme
Hmm... Guess my No, no and no wasn't explicit enough...
1.) There are no "cool effects" to watch in Q2DE (Quartz 2D Extreme). None.
2.) The performance increase, in real life, isn't very noticeable on my hardware. (See sig.) You might see the most difference with a very old AGP G4 PowerMac and a very new graphics card with lots of VRAM. But either way: It _is_ faster and QuickDraw (which it replaces) will be totally turned off one day. Right now, Apple has decided to _not_ enable it, and I guess that's for a reason. I don't see any bugs, but if you're just looking for effects etc., you might as well forget about it.
1.) There are no "cool effects" to watch in Q2DE (Quartz 2D Extreme). None.
2.) The performance increase, in real life, isn't very noticeable on my hardware. (See sig.) You might see the most difference with a very old AGP G4 PowerMac and a very new graphics card with lots of VRAM. But either way: It _is_ faster and QuickDraw (which it replaces) will be totally turned off one day. Right now, Apple has decided to _not_ enable it, and I guess that's for a reason. I don't see any bugs, but if you're just looking for effects etc., you might as well forget about it.
Lt Major Burns
"Dicky" Charlteston-Burns
Ha Ha! yeah you're right (repped). it's just.... why call it EXTREME? it's Quartz 2D GL., like quartz extreme is Quartz Compositor GL. It's not Wacky, Crazy, Zany or Extreme...
tried the translucsent terminal test with the 1920x1080 batman h.264 movie. silly. totally smooth even when shift-minimizing-to-dock. anything else worth playing with? like Q2DX?
tried the translucsent terminal test with the 1920x1080 batman h.264 movie. silly. totally smooth even when shift-minimizing-to-dock. anything else worth playing with? like Q2DX?
jzdziarski
Registered
I don't think you're quite using it correctly. Problem is, once you enable it, none of the existing applications recognize it until you restart...and since the system is already up, well... it just doesn't look impressive. But check out these benchmarks I've included. They show a significant improvement of all related components after booting into a Q2DE-enabled system. To enable it permanently, turn it on in Quartz Debug and then kill -9 Quartz Debug (or 'Force Quit' the application). Then reboot. I think you'll see a significant difference when the OS is booted into Q2DE as I have. It sped up all Quartz components to nearly twice as fast, and even sped up user interface components.
These are my latest benches, much more accurate than my previous ones, as I didn't realize you needed to reboot into Q2DE at first.
No Q2DE
Results 137.15
System Info
Xbench Version 1.1.3
System Version 10.4 (8A428)
Physical RAM 1024 MB
Model PowerBook5,7
Processor PowerPC G4 @ 1.67 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.67 GHz
Bus Frequency 167 MHz
Video Card ATY,RV360M11
Drive Type ST9100823A
CPU Test 183.30
GCD Loop 185.65 7.25 Mops/sec
Floating Point Basic 193.26 698.88 Mflop/sec
AltiVec Basic 193.06 5.61 Gflop/sec
vecLib FFT 189.47 2.94 Gflop/sec
Floating Point Library 159.79 6.40 Mops/sec
Thread Test 138.19
Computation 99.38 1.34 Mops/sec, 4 threads
Lock Contention 226.70 2.85 Mlocks/sec, 4 threads
Memory Test 119.49
System 137.25
Allocate 800.03 521.86 Kalloc/sec
Fill 192.63 1533.36 MB/sec
Copy 64.87 324.33 MB/sec
Stream 105.80
Copy 101.58 742.53 MB/sec [altivec]
Scale 103.31 762.41 MB/sec [altivec]
Add 111.42 713.06 MB/sec [altivec]
Triad 107.43 656.38 MB/sec [altivec]
Quartz Graphics Test 152.12
Line 123.96 3.16 Klines/sec [50% alpha]
Rectangle 129.59 9.12 Krects/sec [50% alpha]
Circle 146.41 3.37 Kcircles/sec [50% alpha]
Bezier 176.93 1.92 Kbeziers/sec [50% alpha]
Text 217.23 3.54 Kchars/sec
OpenGL Graphics Test 143.34
Spinning Squares 143.34 100.31 frames/sec
User Interface Test 246.05
Elements 246.05 79.14 refresh/sec
Disk Test 80.89
Sequential 99.89
Uncached Write 101.02 42.11 MB/sec [4K blocks]
Uncached Write 80.50 32.96 MB/sec [256K blocks]
Uncached Read 158.57 25.10 MB/sec [4K blocks]
Uncached Read 87.62 35.40 MB/sec [256K blocks]
Random 67.96
Uncached Write 49.85 0.75 MB/sec [4K blocks]
Uncached Write 71.75 16.18 MB/sec [256K blocks]
Uncached Read 76.31 0.50 MB/sec [4K blocks]
Uncached Read 85.03 17.50 MB/sec [256K blocks]
Q2DE Enabled
Results 144.44
System Info
Xbench Version 1.1.3
System Version 10.4 (8A428)
Physical RAM 1024 MB
Model PowerBook5,7
Processor PowerPC G4 @ 1.67 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.67 GHz
Bus Frequency 167 MHz
Video Card ATY,RV360M11
Drive Type ST9100823A
CPU Test 191.32
GCD Loop 183.60 7.17 Mops/sec
Floating Point Basic 192.39 695.76 Mflop/sec
AltiVec Basic 190.21 5.53 Gflop/sec
vecLib FFT 196.42 3.05 Gflop/sec
Floating Point Library 194.52 7.79 Mops/sec
Thread Test 137.51
Computation 98.45 1.33 Mops/sec, 4 threads
Lock Contention 227.96 2.86 Mlocks/sec, 4 threads
Memory Test 121.52
System 143.66
Allocate 696.35 454.23 Kalloc/sec
Fill 214.83 1710.02 MB/sec
Copy 67.60 338.01 MB/sec
Stream 105.29
Copy 101.59 742.63 MB/sec [altivec]
Scale 102.40 755.73 MB/sec [altivec]
Add 110.71 708.52 MB/sec [altivec]
Triad 106.98 653.66 MB/sec [altivec]
Quartz Graphics Test 218.84
Line 213.78 5.44 Klines/sec [50% alpha]
Rectangle 241.87 17.02 Krects/sec [50% alpha]
Circle 287.30 6.62 Kcircles/sec [50% alpha]
Bezier 172.86 1.88 Kbeziers/sec [50% alpha]
Text 209.62 3.42 Kchars/sec
OpenGL Graphics Test 146.30
Spinning Squares 146.30 102.38 frames/sec
User Interface Test 260.63
Elements 260.63 83.83 refresh/sec
Disk Test 80.04
Sequential 98.52
Uncached Write 102.43 42.70 MB/sec [4K blocks]
Uncached Write 79.85 32.70 MB/sec [256K blocks]
Uncached Read 146.66 23.22 MB/sec [4K blocks]
Uncached Read 87.00 35.15 MB/sec [256K blocks]
Random 67.39
Uncached Write 49.62 0.74 MB/sec [4K blocks]
Uncached Write 68.92 15.54 MB/sec [256K blocks]
Uncached Read 76.35 0.50 MB/sec [4K blocks]
Uncached Read 86.25 17.75 MB/sec [256K blocks]
These are my latest benches, much more accurate than my previous ones, as I didn't realize you needed to reboot into Q2DE at first.
No Q2DE
Results 137.15
System Info
Xbench Version 1.1.3
System Version 10.4 (8A428)
Physical RAM 1024 MB
Model PowerBook5,7
Processor PowerPC G4 @ 1.67 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.67 GHz
Bus Frequency 167 MHz
Video Card ATY,RV360M11
Drive Type ST9100823A
CPU Test 183.30
GCD Loop 185.65 7.25 Mops/sec
Floating Point Basic 193.26 698.88 Mflop/sec
AltiVec Basic 193.06 5.61 Gflop/sec
vecLib FFT 189.47 2.94 Gflop/sec
Floating Point Library 159.79 6.40 Mops/sec
Thread Test 138.19
Computation 99.38 1.34 Mops/sec, 4 threads
Lock Contention 226.70 2.85 Mlocks/sec, 4 threads
Memory Test 119.49
System 137.25
Allocate 800.03 521.86 Kalloc/sec
Fill 192.63 1533.36 MB/sec
Copy 64.87 324.33 MB/sec
Stream 105.80
Copy 101.58 742.53 MB/sec [altivec]
Scale 103.31 762.41 MB/sec [altivec]
Add 111.42 713.06 MB/sec [altivec]
Triad 107.43 656.38 MB/sec [altivec]
Quartz Graphics Test 152.12
Line 123.96 3.16 Klines/sec [50% alpha]
Rectangle 129.59 9.12 Krects/sec [50% alpha]
Circle 146.41 3.37 Kcircles/sec [50% alpha]
Bezier 176.93 1.92 Kbeziers/sec [50% alpha]
Text 217.23 3.54 Kchars/sec
OpenGL Graphics Test 143.34
Spinning Squares 143.34 100.31 frames/sec
User Interface Test 246.05
Elements 246.05 79.14 refresh/sec
Disk Test 80.89
Sequential 99.89
Uncached Write 101.02 42.11 MB/sec [4K blocks]
Uncached Write 80.50 32.96 MB/sec [256K blocks]
Uncached Read 158.57 25.10 MB/sec [4K blocks]
Uncached Read 87.62 35.40 MB/sec [256K blocks]
Random 67.96
Uncached Write 49.85 0.75 MB/sec [4K blocks]
Uncached Write 71.75 16.18 MB/sec [256K blocks]
Uncached Read 76.31 0.50 MB/sec [4K blocks]
Uncached Read 85.03 17.50 MB/sec [256K blocks]
Q2DE Enabled
Results 144.44
System Info
Xbench Version 1.1.3
System Version 10.4 (8A428)
Physical RAM 1024 MB
Model PowerBook5,7
Processor PowerPC G4 @ 1.67 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.67 GHz
Bus Frequency 167 MHz
Video Card ATY,RV360M11
Drive Type ST9100823A
CPU Test 191.32
GCD Loop 183.60 7.17 Mops/sec
Floating Point Basic 192.39 695.76 Mflop/sec
AltiVec Basic 190.21 5.53 Gflop/sec
vecLib FFT 196.42 3.05 Gflop/sec
Floating Point Library 194.52 7.79 Mops/sec
Thread Test 137.51
Computation 98.45 1.33 Mops/sec, 4 threads
Lock Contention 227.96 2.86 Mlocks/sec, 4 threads
Memory Test 121.52
System 143.66
Allocate 696.35 454.23 Kalloc/sec
Fill 214.83 1710.02 MB/sec
Copy 67.60 338.01 MB/sec
Stream 105.29
Copy 101.59 742.63 MB/sec [altivec]
Scale 102.40 755.73 MB/sec [altivec]
Add 110.71 708.52 MB/sec [altivec]
Triad 106.98 653.66 MB/sec [altivec]
Quartz Graphics Test 218.84
Line 213.78 5.44 Klines/sec [50% alpha]
Rectangle 241.87 17.02 Krects/sec [50% alpha]
Circle 287.30 6.62 Kcircles/sec [50% alpha]
Bezier 172.86 1.88 Kbeziers/sec [50% alpha]
Text 209.62 3.42 Kchars/sec
OpenGL Graphics Test 146.30
Spinning Squares 146.30 102.38 frames/sec
User Interface Test 260.63
Elements 260.63 83.83 refresh/sec
Disk Test 80.04
Sequential 98.52
Uncached Write 102.43 42.70 MB/sec [4K blocks]
Uncached Write 79.85 32.70 MB/sec [256K blocks]
Uncached Read 146.66 23.22 MB/sec [4K blocks]
Uncached Read 87.00 35.15 MB/sec [256K blocks]
Random 67.39
Uncached Write 49.62 0.74 MB/sec [4K blocks]
Uncached Write 68.92 15.54 MB/sec [256K blocks]
Uncached Read 76.35 0.50 MB/sec [4K blocks]
Uncached Read 86.25 17.75 MB/sec [256K blocks]
I hope you didn't mean me, jzdziarski. I'm certainly aware of it not being in effect for apps that are started before activating it. Mine's active now even through reboots, so that's not the case here. Benches are nice n' all, but the question was "noticeable", and I think it's not so much noticeable. Real life use, I mean. Benchmarks ain't real life use for me.
JPigford
I'm awesome...seriously..
Maybe I'm missing something (I'm sure of it actually)...but does Quartz 2D Extreme have any benefit to anybody other than developers? For instance, I do a ridiculous amount of design and video editing...does Quartz 2D Extreme help me? Or does it even have anything to do with that?
Viro
Registered
JPigford said:
The guys at Arstechnica have written about Quartz Extreme 2D and explained it far better than I ever could. They even include some nice diagrams.
http://arstechnica.com/reviews/os/macosx-10.4.ars/14
Basically, Quartz Extreme 2D moves a lot more of the graphics operations onto the GPU, freeing up the CPU to do other things. Also this means that the window manager will be more responsive when there are many windows open. It should benefit the normal user too, especially those who have loads of windows open.