I did a search and couldn't find anything worthwhile. Question: What's with the 800MHZ FSB that Intel seems to have? Granted, it's not slow, it's just that no matter if it's a 1.6G or 3.6G system, it never seems to get past that. I'm not trying to start a "Thing" here, but would really like some info.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Power Mac G5 Quad Benchmarks
- Thread starter MacRumors
- Start date
- Sort by reaction score
It's actually what they term "quad-pumped" at 200 MHz FSB.woolfgang said:
http://en.wikipedia.org/wiki/Quad-pumped
Their P4 Extreme Editions are at 1066MHz, quad-pumped.
AidenShaw said:
Thanks for clearing that up. I was going to if you hadn't. 1 thread/core, or 1 process/core, or more of course via scheduling, but .5 threads/core doesn't make any sense.
Where have I been?
It's very interesting to find out what benchmarks show up as or even
what they vary between versions. My G5 2.5 dual produces an utterly
absurd 312 rating in an old version of Xbench I have lying around but
produces 126 or so on the latest. I also find I get better results when
the network cable is UNPLUGGED when running Xbench scans.
And here's something I found very interesting. I used ATIccelerator II
to overclock my 9600 XT video card in my G5 2.5 dual and found that
with all the new Quartz Extreme features it had an effect on the outcome
of the tests.
The 9600 XT comes shipped with the Mac is clocked to 400 mhz with
the memory somewhere in the low 300 range whereas the same card
on the PC is clocked to 500 mhz with memory in the 320 range.
PC overclockers have determined this card can easily be pushed 25%
beyond original rated clock (original being 500) but don't push the
memory beyond 330 or you could have a dead card on your hands.
So I decided to overclock the card and see how it effected some
game play and general use. A slight improvement. Original ratings
were 341, 970 and 1619 for the OpenGL shading portions of the
test. I set the card to 525 mhz clock and 322 mhz ram speed.
Check this:
CINEBENCH 2003 v1
****************************************************
Tester : AfterImages
Processor : G5 2.5 Dual
MHz : 2.5 GHZ
Number of CPUs : 2
Operating System : 10.4.3
Graphics Card : ATI Radeon 9600 XT
Resolution : 1280x1024
Color Depth : 32-Bit
****************************************************
Rendering (Single CPU): 360 CB-CPU
Rendering (Multiple CPU): 657 CB-CPU
Multiprocessor Speedup: 1.75
Shading (CINEMA 4D) : 341 CB-GFX
Shading (OpenGL Software Lighting) : 1000 CB-GFX
Shading (OpenGL Hardware Lighting) : 1795 CB-GFX
OpenGL Speedup: 5.30
****************************************************
Interesting. A bit of extra speed but one suspects Apple clocked the
cards to 400 mhz so as to not necessitate a fan on the cards thus
keeping it quieter than a fanned card. I've since brought it back to
the original clock speed but when playing games or the like I'll bring
it back up there.
Food for thought. /Re-enter lurk mode.
It's very interesting to find out what benchmarks show up as or even
what they vary between versions. My G5 2.5 dual produces an utterly
absurd 312 rating in an old version of Xbench I have lying around but
produces 126 or so on the latest. I also find I get better results when
the network cable is UNPLUGGED when running Xbench scans.
And here's something I found very interesting. I used ATIccelerator II
to overclock my 9600 XT video card in my G5 2.5 dual and found that
with all the new Quartz Extreme features it had an effect on the outcome
of the tests.
The 9600 XT comes shipped with the Mac is clocked to 400 mhz with
the memory somewhere in the low 300 range whereas the same card
on the PC is clocked to 500 mhz with memory in the 320 range.
PC overclockers have determined this card can easily be pushed 25%
beyond original rated clock (original being 500) but don't push the
memory beyond 330 or you could have a dead card on your hands.
So I decided to overclock the card and see how it effected some
game play and general use. A slight improvement. Original ratings
were 341, 970 and 1619 for the OpenGL shading portions of the
test. I set the card to 525 mhz clock and 322 mhz ram speed.
Check this:
CINEBENCH 2003 v1
****************************************************
Tester : AfterImages
Processor : G5 2.5 Dual
MHz : 2.5 GHZ
Number of CPUs : 2
Operating System : 10.4.3
Graphics Card : ATI Radeon 9600 XT
Resolution : 1280x1024
Color Depth : 32-Bit
****************************************************
Rendering (Single CPU): 360 CB-CPU
Rendering (Multiple CPU): 657 CB-CPU
Multiprocessor Speedup: 1.75
Shading (CINEMA 4D) : 341 CB-GFX
Shading (OpenGL Software Lighting) : 1000 CB-GFX
Shading (OpenGL Hardware Lighting) : 1795 CB-GFX
OpenGL Speedup: 5.30
****************************************************
Interesting. A bit of extra speed but one suspects Apple clocked the
cards to 400 mhz so as to not necessitate a fan on the cards thus
keeping it quieter than a fanned card. I've since brought it back to
the original clock speed but when playing games or the like I'll bring
it back up there.
Food for thought. /Re-enter lurk mode.
nick007 said:
Took 2:35 min on my Mac Mini 1.42 with 512 mb ram.. i'll re-run the test and confirm
Edit: This time it took 2:32... I love this li'l pest
oober_freak said:Took 2:35 min on my Mac Mini 1.42 with 512 mb ram.. i'll re-run the test and confirm
Edit: This time it took 2:32... I love this li'l pest
it took me 7:26 on my iMac G3
Driver Heaven Photoshop tests
Has anyone had a chance to run the Driver Heaven Photoshop test suite yet?
Instructions here: http://www.driverheaven.net/photoshop/
Mac download here: http://www.driverheaven.net/photoshop/dhpsbench2.zip
Results for comparison here (there's a few PowerMacs in there): http://www.driverheavendownloads.net/photoshop/results.php
Cheers, Joe
Has anyone had a chance to run the Driver Heaven Photoshop test suite yet?
Instructions here: http://www.driverheaven.net/photoshop/
Mac download here: http://www.driverheaven.net/photoshop/dhpsbench2.zip
Results for comparison here (there's a few PowerMacs in there): http://www.driverheavendownloads.net/photoshop/results.php
Cheers, Joe
new quad benchmarks
>> 11-9-05
http://barefeats.com/quad01.html
Compares quad to dual core 2.5 i believe.... enjoy
Macrumors said:
One of our Wales UK-based members has reported receiving his Power Mac G5 Quad earlier in the week, and has run a preliminary set of Xbench benchmarks on it. While full details are in the forum thread, the overall score for the machine with 3 GB RAM running Mac OS X 10.4.2 weighs in at 151.86.
>> 11-9-05
http://barefeats.com/quad01.html
Compares quad to dual core 2.5 i believe.... enjoy
Yeah. My old 800 MHz G4 PowerBook gets a 72 for rendering, and IIRC my dual 2.0 G5 gets something around 550-600...so your 1016 looks pretty good, considering the overhead of multithreading.Pie said:
jiggie2g said:Your point , Do u think most system builders paid $300 for XP Pro. I've seen more cracked XP Pro Discs passed around then AOL sign up discs. The Pirates are going to Increase OSX's market share more than anyone in the coming years.
Arrrr Matey there be cracked Tigers in the Bay (Hint Hint).
It wasn't hard to crack the security for XP because it was a halfhearted attempt.
Tiger depends upon much secure technology. The software code is tied to the hardware. It's not likely that it will be so easily cracked. Even if it is, the technology assures that no programs will work.
Having the OS with nothing to run on it will be a lot of fun.
Not sure if these have been posted here, but this guy has done a comparison, represented in graphical form, of the quad core 2.5 and the single core dual 2.5.
http://www.barefeats.com/quad01.html
Also, he posted test results for the Dual-Core G5/2.0GHz versus Single-Core
Dual G5/2.0GHz Power Mac:
http://www.barefeats.com/dc20.html
What awesome machines!
YOJ
http://www.barefeats.com/quad01.html
Also, he posted test results for the Dual-Core G5/2.0GHz versus Single-Core
Dual G5/2.0GHz Power Mac:
http://www.barefeats.com/dc20.html
What awesome machines!
YOJ
How does the Quad compare to this:
http://www.torrentspy.com/article.asp?id=3805
The US National Nuclear Security Administration (NNSA) has unveiled two new IBM supercomputers at Lawrence Livermore National Laboratory (LLNL), which are capable of the fastest processing speeds yet achieved.
The 65,536 processor BlueGene/L supercomputer has performed a record 280.6 trillion operations per second on the industry standard LINPACK benchmark - software used to rank the speeds of the worlds fastest machines.
Purple, the other half of the most powerful supercomputing twosome, is a machine capable of 100 trillion operations per second as it conducts simulations of a complete nuclear weapons performance.
Together, the Purple and BlueGene/L systems will perform half a petaflop, or half of a quadrillion (1,000,000,000,000,000) operations per second.
NNSA administrator Linton F Brooks said the machines will be used to run three-dimensional codes at lightning-fast speeds for nuclear weapons analysis. This was formerly accomplished by underground nuclear testing.
The unprecedented computing power of these two supercomputers is more critical than ever to meet the time-urgent issues related to maintaining our nations aging nuclear stockpile without testing, Brooks said.
Purple represents the culmination of a 10-year effort to create a new class of supercomputers. BlueGene/L points the way to the future and the computing power we will need to improve our ability to predict the behaviour of the stockpile as it continues to age. This has reestablished American computing preeminence.
In a recent demonstration of its work capability, BlueGene/L ran a record-setting materials science application at 101.5 teraflops sustained over seven hours on the machines 131,072 processors, running an application of importance to NNSAs effort to ensure the safety and reliability of the nations nuclear deterrent. A teraflop is one trillion computer operations per second.
Both machines were developed through NNSAs Advanced Simulation and Computing (ASC) program and join a series of other supercomputers at Sandia and Los Alamos national laboratories that are dedicated to NNSAs Stockpile Stewardship effort to maintain the nations nuclear deterrent through science-based computation, theory and experiment.
http://www.torrentspy.com/article.asp?id=3805
The US National Nuclear Security Administration (NNSA) has unveiled two new IBM supercomputers at Lawrence Livermore National Laboratory (LLNL), which are capable of the fastest processing speeds yet achieved.
The 65,536 processor BlueGene/L supercomputer has performed a record 280.6 trillion operations per second on the industry standard LINPACK benchmark - software used to rank the speeds of the worlds fastest machines.
Purple, the other half of the most powerful supercomputing twosome, is a machine capable of 100 trillion operations per second as it conducts simulations of a complete nuclear weapons performance.
Together, the Purple and BlueGene/L systems will perform half a petaflop, or half of a quadrillion (1,000,000,000,000,000) operations per second.
NNSA administrator Linton F Brooks said the machines will be used to run three-dimensional codes at lightning-fast speeds for nuclear weapons analysis. This was formerly accomplished by underground nuclear testing.
The unprecedented computing power of these two supercomputers is more critical than ever to meet the time-urgent issues related to maintaining our nations aging nuclear stockpile without testing, Brooks said.
Purple represents the culmination of a 10-year effort to create a new class of supercomputers. BlueGene/L points the way to the future and the computing power we will need to improve our ability to predict the behaviour of the stockpile as it continues to age. This has reestablished American computing preeminence.
In a recent demonstration of its work capability, BlueGene/L ran a record-setting materials science application at 101.5 teraflops sustained over seven hours on the machines 131,072 processors, running an application of importance to NNSAs effort to ensure the safety and reliability of the nations nuclear deterrent. A teraflop is one trillion computer operations per second.
Both machines were developed through NNSAs Advanced Simulation and Computing (ASC) program and join a series of other supercomputers at Sandia and Los Alamos national laboratories that are dedicated to NNSAs Stockpile Stewardship effort to maintain the nations nuclear deterrent through science-based computation, theory and experiment.
PowerBook G4 867 12" - 150 is FAST!
Results 20.98
System Info
Xbench Version 1.2
System Version 10.4.3 (8F46)
Physical RAM 640 MB
Model PowerBook6,1
Processor PowerPC G4 @ 867 MHz
Version 7455 (Apollo) v3.3
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K @ 534 MHz
Bus Frequency 134 MHz
Video Card GeForce4 MX
Drive Type SAMSUNG MP0402H
CPU Test 27.47
GCD Loop 51.60 2.72 Mops/sec
Floating Point Basic 18.04 428.72 Mflop/sec
vecLib FFT 47.20 1.56 Gflop/sec
Floating Point Library 20.15 3.51 Mops/sec
Thread Test 31.44
Computation 30.47 617.26 Kops/sec, 4 threads
Lock Contention 32.47 1.40 Mlocks/sec, 4 threads
Memory Test 18.84
System 18.58
Allocate 74.53 273.68 Kalloc/sec
Fill 13.09 636.67 MB/sec
Copy 13.95 288.16 MB/sec
Stream 19.10
Copy 19.02 392.76 MB/sec [altivec]
Scale 18.48 381.84 MB/sec [altivec]
Add 19.52 415.85 MB/sec [altivec]
Triad 19.44 415.77 MB/sec [altivec]
Quartz Graphics Test 28.19
Line 31.75 2.11 Klines/sec [50% alpha]
Rectangle 27.89 8.33 Krects/sec [50% alpha]
Circle 28.19 2.30 Kcircles/sec [50% alpha]
Bezier 33.43 843.20 beziers/sec [50% alpha]
Text 22.40 1.40 Kchars/sec
OpenGL Graphics Test 46.23
Spinning Squares 46.23 58.65 frames/sec
User Interface Test 9.68
Elements 9.68 44.42 refresh/sec
Disk Test 19.29
Sequential 33.97
Uncached Write 24.18 14.84 MB/sec [4K blocks]
Uncached Write 34.97 19.79 MB/sec [256K blocks]
Uncached Read 44.23 12.95 MB/sec [4K blocks]
Uncached Read 39.70 19.95 MB/sec [256K blocks]
Random 13.47
Uncached Write 4.22 0.45 MB/sec [4K blocks]
Uncached Write 41.54 13.30 MB/sec [256K blocks]
Uncached Read 50.52 0.36 MB/sec [4K blocks]
Uncached Read 62.68 11.63 MB/sec [256K blocks]
Results 20.98
System Info
Xbench Version 1.2
System Version 10.4.3 (8F46)
Physical RAM 640 MB
Model PowerBook6,1
Processor PowerPC G4 @ 867 MHz
Version 7455 (Apollo) v3.3
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K @ 534 MHz
Bus Frequency 134 MHz
Video Card GeForce4 MX
Drive Type SAMSUNG MP0402H
CPU Test 27.47
GCD Loop 51.60 2.72 Mops/sec
Floating Point Basic 18.04 428.72 Mflop/sec
vecLib FFT 47.20 1.56 Gflop/sec
Floating Point Library 20.15 3.51 Mops/sec
Thread Test 31.44
Computation 30.47 617.26 Kops/sec, 4 threads
Lock Contention 32.47 1.40 Mlocks/sec, 4 threads
Memory Test 18.84
System 18.58
Allocate 74.53 273.68 Kalloc/sec
Fill 13.09 636.67 MB/sec
Copy 13.95 288.16 MB/sec
Stream 19.10
Copy 19.02 392.76 MB/sec [altivec]
Scale 18.48 381.84 MB/sec [altivec]
Add 19.52 415.85 MB/sec [altivec]
Triad 19.44 415.77 MB/sec [altivec]
Quartz Graphics Test 28.19
Line 31.75 2.11 Klines/sec [50% alpha]
Rectangle 27.89 8.33 Krects/sec [50% alpha]
Circle 28.19 2.30 Kcircles/sec [50% alpha]
Bezier 33.43 843.20 beziers/sec [50% alpha]
Text 22.40 1.40 Kchars/sec
OpenGL Graphics Test 46.23
Spinning Squares 46.23 58.65 frames/sec
User Interface Test 9.68
Elements 9.68 44.42 refresh/sec
Disk Test 19.29
Sequential 33.97
Uncached Write 24.18 14.84 MB/sec [4K blocks]
Uncached Write 34.97 19.79 MB/sec [256K blocks]
Uncached Read 44.23 12.95 MB/sec [4K blocks]
Uncached Read 39.70 19.95 MB/sec [256K blocks]
Random 13.47
Uncached Write 4.22 0.45 MB/sec [4K blocks]
Uncached Write 41.54 13.30 MB/sec [256K blocks]
Uncached Read 50.52 0.36 MB/sec [4K blocks]
Uncached Read 62.68 11.63 MB/sec [256K blocks]
iBook 1.33Ghz w/ 1Gig RAM (New Model)
Returned the following results
Results 37.97
System Info
Xbench Version 1.2
System Version 10.4.3 (8F46)
Physical RAM 1024 MB
Model PowerBook6,7
Processor PowerPC G4 @ 1.33 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.33 GHz
Bus Frequency 134 MHz
Video Card ATY,M12
Drive Type TOSHIBA MK6025GAS
CPU Test 51.44
GCD Loop 110.60 5.83 Mops/sec
Floating Point Basic 31.39 745.74 Mflop/sec
vecLib FFT 77.14 2.54 Gflop/sec
Floating Point Library 41.85 7.29 Mops/sec
Thread Test 57.72
Computation 55.16 1.12 Mops/sec, 4 threads
Lock Contention 60.54 2.60 Mlocks/sec, 4 threads
Memory Test 23.83
System 27.64
Allocate 101.94 374.37 Kalloc/sec
Fill 27.84 1353.52 MB/sec
Copy 15.92 328.80 MB/sec
Stream 20.94
Copy 21.14 436.71 MB/sec [altivec]
Scale 20.82 430.03 MB/sec [altivec]
Add 20.87 444.59 MB/sec [altivec]
Triad 20.94 447.90 MB/sec [altivec]
Quartz Graphics Test 57.41
Line 42.52 2.83 Klines/sec [50% alpha]
Rectangle 50.89 15.19 Krects/sec [50% alpha]
Circle 50.56 4.12 Kcircles/sec [50% alpha]
Bezier 66.65 1.68 Kbeziers/sec [50% alpha]
Text 109.38 6.84 Kchars/sec
OpenGL Graphics Test 69.25
Spinning Squares 69.25 87.85 frames/sec
User Interface Test 30.20
Elements 30.20 138.62 refresh/sec
Disk Test 24.58
Sequential 43.64
Uncached Write 36.54 22.43 MB/sec [4K blocks]
Uncached Write 38.02 21.51 MB/sec [256K blocks]
Uncached Read 71.89 21.04 MB/sec [4K blocks]
Uncached Read 41.52 20.87 MB/sec [256K blocks]
Random 17.11
Uncached Write 5.92 0.63 MB/sec [4K blocks]
Uncached Write 33.77 10.81 MB/sec [256K blocks]
Uncached Read 51.97 0.37 MB/sec [4K blocks]
Uncached Read 62.44 11.59 MB/sec [256K blocks]
Returned the following results
Results 37.97
System Info
Xbench Version 1.2
System Version 10.4.3 (8F46)
Physical RAM 1024 MB
Model PowerBook6,7
Processor PowerPC G4 @ 1.33 GHz
L1 Cache 32K (instruction), 32K (data)
L2 Cache 512K @ 1.33 GHz
Bus Frequency 134 MHz
Video Card ATY,M12
Drive Type TOSHIBA MK6025GAS
CPU Test 51.44
GCD Loop 110.60 5.83 Mops/sec
Floating Point Basic 31.39 745.74 Mflop/sec
vecLib FFT 77.14 2.54 Gflop/sec
Floating Point Library 41.85 7.29 Mops/sec
Thread Test 57.72
Computation 55.16 1.12 Mops/sec, 4 threads
Lock Contention 60.54 2.60 Mlocks/sec, 4 threads
Memory Test 23.83
System 27.64
Allocate 101.94 374.37 Kalloc/sec
Fill 27.84 1353.52 MB/sec
Copy 15.92 328.80 MB/sec
Stream 20.94
Copy 21.14 436.71 MB/sec [altivec]
Scale 20.82 430.03 MB/sec [altivec]
Add 20.87 444.59 MB/sec [altivec]
Triad 20.94 447.90 MB/sec [altivec]
Quartz Graphics Test 57.41
Line 42.52 2.83 Klines/sec [50% alpha]
Rectangle 50.89 15.19 Krects/sec [50% alpha]
Circle 50.56 4.12 Kcircles/sec [50% alpha]
Bezier 66.65 1.68 Kbeziers/sec [50% alpha]
Text 109.38 6.84 Kchars/sec
OpenGL Graphics Test 69.25
Spinning Squares 69.25 87.85 frames/sec
User Interface Test 30.20
Elements 30.20 138.62 refresh/sec
Disk Test 24.58
Sequential 43.64
Uncached Write 36.54 22.43 MB/sec [4K blocks]
Uncached Write 38.02 21.51 MB/sec [256K blocks]
Uncached Read 71.89 21.04 MB/sec [4K blocks]
Uncached Read 41.52 20.87 MB/sec [256K blocks]
Random 17.11
Uncached Write 5.92 0.63 MB/sec [4K blocks]
Uncached Write 33.77 10.81 MB/sec [256K blocks]
Uncached Read 51.97 0.37 MB/sec [4K blocks]
Uncached Read 62.44 11.59 MB/sec [256K blocks]
AidenShaw said:
strange that link has at the bottom 16GB of ram with just 6 slots; how on Earth are the configuration for the RAM chips?!!
16GB, DDR2 SDRAM Memory, 400MHz, ECC (6 DIMMS) [add $23,300]Prom1 said:
Easy - two 4 GiB DIMMs, and 4 2 GiB DIMMs.... (note that 12 GiB is only $9.2K)
Look Who Got Up On The Wrong Side Of Bed Today
I don't know what you're smoking, but from where I sit I see nothing but faster PowerMacs etc coming down the pike. Nothing is being thrown out any doors that I can see.rlwimi said:
melgross said:
If a man makes it another can break it , this is fact. There are ways to fool software to believe you are running it's certified hardware. all of a sudden because it's made by apple it won't be hacked, cracked , and distributed all over the web. News Flash buddy I can download Tiger 10.4.3 right now for mac or PC and both will work just as good as your copy. You continue to believe the crap Jobs feeds you. and i will be here to gloat this time next year.