PowerPC G4 1000 2 1500 256 Mac OSX 7.5 [572] Textures 4 8 Markus Groeteke Jun 11, 2002
(8 seconds)
AthlonMP 1667 2 1024 256 Win XP 7.0 [b / 543] Textures 8 3 Luca [effector] Feb 11, 2002
(3 seconds)
"yes, Newtek optimized for both for Altivec and SSE2!"
Those are your words. Those are the benchmarks pulled off your site. That is not even half as fast rendering the same frame.
Again the farm economy rules that benchmark. More macines mean better performance. Frame rendering scales linearly with machine count. Which means price performance counts. Which means you can buy 3 dual athlon machines instead of a single G4 tower and not just double the G4s performance, but render frames 6 times as fast for a given cost.
I don't understand how price would be a bogus factor.
On an application with performance that scales linearly with more machines, such as the distributed client or webserving, the limit to performance is total cost and price/performance.
How fast do you want to spend.
There is no point introducing price/performance into a comparison of photoshop filters because the application will not scale across multiple machines.
In that case you simply compare the two machines performance and question whether you are willing to spend an extra $2000 for inferior performance just because you like the look and feel of one machine over another.
There are other factors. If you are limited by rackspace, then you need to know how much performance you can stuff into 1u reguardless of cost. In Apple's case, they can place a dual processor X Serve in 1u. In competition to X Serve, there are companies putting two dual processor Athlons into a single 1u rack.
Another factor would be performance per watt. If you needed to perform ogr calculations drawing the minimum amount of power, the obvious power difference between running two Macs or a single Athlon become clear.
So, you are right, it is good to have choice. It is good to be able to choose something besides proprietary software on proprietary hardware. Viva la x86, Viva la Linux.
It is rated a troll because I regulary cut through the "Steve Jobs reality distortion feild".
No one likes their distorted reality crushed, especially the hardened core of mac zealots that apple has been reduced to.
There are not as many "switchers" as Apple would have us belive. Plenty of "twitchers", people that like OS X but balk at the platform's inflated price and poor absolute performance.
This will get a poor mod also, even though I am just answering your question.
Power PC 7450/7455 G4 1000 rc5 10,525,403 50841 0.4 3
AMD Athlon XP (Palomino) 2100 rc5 6,128,961 42741 0.6 3
Intel Pentium 4 2800 rc5 5,522,975 0 0.0 1
rc5 has definitely been optimized for altivec. It is a credit to the Mac community that when source code is available, they do an excellent job with optomized ports.
Power PC 7450/7455 G4 1000 ogr 10,579,931 338 0.0 2
AMD Athlon XP (Palomino) 2100 ogr 13,476,744 298544 2.2 2
Intel Pentium 4 2200 ogr 7,684,803 0 0.0 1
Knowing that the Mac community goes out of their way to create altivec optimized ports, the G4's performance in ogr is also quite telling.
While it is unfortunate that the Pentium 4 2.8 has not been benchmarked in ogr, the best it could hope for is about 9.7 million, showing off it's crippled integer instruction prosessor.
While these are not the fastest processors from Apple or AMD, if the numbers scale linearly, as they seem to in refrence to older processors, both processor types should see a 25% performance improvement with current clock speeds.
This is a sticky point. At $3,299.00 for a dual G4 1.25 compared to just over $1000 for a complete dual Athlon MP system, the keys processed at a given cost look like this:
26449827 running two rc5 clients on a dual 1.25 G4. 36773766 running six rc5 clients on 3 dual 2100MPs.
The Athlons are 40% faster at a given price in rc5.
26449827 running two ogr clients on a dual 1.25 G4. 80860464 running two ogr clients on 3 dual 2100MPs.
The Athlons are 305% faster at a given price in ogr.
So, there is a case where the G4 does not get crushed in absolute performance. But like other closed/proprietary hardware manufacturers such as sun and sgi, they pose absolutely no competition to open platforms on price/performance.
"Scorecard" shows the G4 and Athlon 2600 dead even at 89% of the fastest processor, The P4 2.8. That is sort of funny, considering ZD bench which represents 68% of the weighted mean of iComp is not available on the Mac.
Here are some additional resources using more modern benchmarks:
Benchmarks using Spec2000 instead of the woefully obsolete Spec92.
Comparison of The Dual G4 to other platforms in after effects rendering
Here is some raw data from the obsolete 16-bit Whetstone which shows the G4 not even able to match a Pentium II clock for clock.
This comparison shows how the G4 compares to a variety of rivals.
This shows that under most conditions, the SDR equiped Dual G4 is faster than than the DDR version. Meaning the 1.25 GHz G4 is crippled with DDR. Which means the numbers from the other benchmark links will not scale linearly. The Dual 1.25 G4 is actualy slower clock for clock than the G4 with SDR.
Also refrence Ars Technica's Seti benchmark showing that a Dual G4 1 Ghz produces work units at the same speed that a Dual PIII 1 Ghz.
Has anyone found any other Benchmark comparisons of the G4 (besides the single photoshop benchmark on Apple's website) that might shed some light on the debate?
So, choose whatever platform you like, but be informed about what you are paying for.
You buy a mac, you buy it because you like OS X, which is far and away the nicest OS out there right now. You do not buy a mac because of performance, as anyone who follows the links can very plainly see.
How much dry elitist apple crap do we have to put up with?
This is a cool mod. Why does BitGeek feel the need to break it down?
He says "slow processor" when the 1800 mhz Athlon has benchmarked on many independent tests as being faster than the fastest dual G4 system Apple has. Bitgeek talks about superior architecture and bandwidth when the most recent macs use DDR just as a marketing ploy. Their performance is actually 3-5% slower than identicly clocked G4s using SDRAM. They cobble fast memory onto their low bandwith processor with a memory translator hub similar to Intel's failed MTH design... All for marketing.
He says "If you can't afford" showing extreme class distain and prejudice.
Absolute crap post.
So, bitgeek, tell us, where are you coming from on this? Why the need to criticize other people's art? Did you gain anything? Does it make you feel good? Are you really that insecure? How invested are you in Apple?
Disclaimer: I sold my Apple shares in July of 2000.
That is easy. I belive he wants a voter registration database for the districts themselves...
Mass mailings are Spam.
Keeping an individual specific private database is Privacy invasion. Can I call up his consulting firm and ask them what they know about me? Can I have myself removed? Would anyone realisticly opt-in to this? no. 99% would opt out.
Walking into Kragen Auto Parts I'm pissed that I can't just pay for my parts and leave. No, they keep a damn database. They know everything they've sold me for the past 5 years. Problem is I've only benn shopping ther for 2 years. The database is cataloged by phone numbers. The jerkoff that had my phone number wrote bad checks all over town.
So not only are there databases everywhere about me, most of them are entirely wrong. Kragen can't delete the database entry until they get paid. I get hasseled just because I want a battery.
Databases are privacy invasive bullshit.
Political database... what the hell is this? 1984?
The good pictures will be from seperation and re-entry.
I still think they should fly the otherwise wasted tank into orbit to be used as parts or huge additions to the space station.
Multiply the space station's internal volume by 10 in just one trip. After all the tank is a high pressure vessel. It should be able to hold 14.5 lb/sq in atmospheric.
Just give the shuttle a bit more power, drop the boosters and park the tank next to ISS.
You won't find much in the way of open source software designed to enable you to create privacy invasive databases or spam your special interest group's message.
You just opened up a vipers nest as far as that question is concerned here on slashdot. Privacy invasion and spam are extreme hot button issues here.
You want to do them both. Go spend the money with microsoft. They love people like you.
They'll get beaten until Sledgehammer is released.. not Claw hammer which will have no x86-64 desktop software support right off the bat, and will have to rely solely on it's pure x86 performance.
Microsoft shafted them on the X-box because Intel paid Microsoft 200 million to use the Pentium III. Nvidia was stuck with an unused AMD integrated chipset for X-box and Nforce was born.
Intel will pay Microsoft to shaft them again. No x86-64 Windows XP for AMD despite AMD testifying on Microsoft's behalf in exchange for anti-trust testimony. AMD made an unenforceable agreement with Microsoft. To enforce it would perjure themselves.
So Intel wins again.
Until Sledgehammer arrives. Sledgehammer is a server/workstation chip and will have full support of the dominant server operating system, Linux. Microsoft must support Sledgehammer or risk losing more of their already weak server market share.
Long after Microsoft has done the work to get Windows running on the Server, Microsoft will incorporate x86-64 support into their desktop OS.
Probably about the same time as they support Hyper threading and SSE3 for Intel.
Create a Highway Area Network. Not only should individual vehicles be smart enough to drive themselves, they should communicate and cooperate with each other to get everyone where they need to go as safely as possible. Just as a train is linked by physical means, vehicles should link wirelessly to achieve the same end - bumper to bumper at 100 mph, but safely.
Exactly what human nature cannot do for itself.
Driving is a repetitive, monotonous task. That is exactly what computers are good at. .
On the last page of Tom's advertizers section is a section entitled "Editorial content sponsorships"
Thus the about face on Rambus and Intel in general.
Tom "will never forget the deceit of Intel" revolving around the memory translator hub recall, the 1.13 Ghz.18 micron PIII recall and their support of Rambus when it was clear that Rambus was trying to assert a monopoly on ram.
A few days after I noticed the "Editorial content sponsorships", he posted the editorial "proving" that Rambus was better than DDR because on a heavily overclocked system, DDR stops scaling.
Tom is a heel. I once admired him. He brought Intel to it's knees by proving the 1.13 processor was factory overclocked.
Now Tom is an Intel/Rambus/Bapco stool.
At least Alex "Sharky" Ross had the moral spine to leave Sharky's when that site became an Intel ad.
Never give up sleep time for study time. During sleep your mind makes the transfer to long term memory. If you are not sleeping 7 hours, you might as well not study at all that day.
And get a good outside accountant. Nothing will go wrong with your business that you cannot fix by delegating, except your in house accountant stealing from you.
So, study, sleep, delegate and don't let the mice play while you are away.
This is the same transistor that IBM was talking about. They are sharing (or stealing) this technology. Electron microscope pictures are of the same transistor.
IBM is a trustworthy computing founding member and is on the steering comittee..
They are pushing DRM
The only major company absent from trusworthy computing is VIA... I can't seem to find Sony on the list either but they have to be pushing a DRM of their own. They are both content producer and hardware manufacturer.
"All the information from the 2000 census is available for free from the United States government," Anderson said. "No one is out in front of their mailbox with a shotgun to prevent direct marketers from distributing targeted print advertising. Yet with television, there is a greater level of concern."
I use A P.O. Box.
Are they somehow going to give me the anonymity of P.O. television?
No.
They will use every means at their disposal to generate income once their "???" doesn't become "Profit!". And that means they are going to sell me out.
My mail does not come to where I live. My Television does.
Slashdot Mirror
PowerPC G4 1000 2 1500 256 Mac OSX 7.5 [572] Textures 4 8 Markus Groeteke Jun 11, 2002
(8 seconds)
AthlonMP 1667 2 1024 256 Win XP 7.0 [b / 543] Textures 8 3 Luca [effector] Feb 11, 2002
(3 seconds)
"yes, Newtek optimized for both for Altivec and SSE2!"
Those are your words. Those are the benchmarks pulled off your site. That is not even half as fast rendering the same frame.
Again the farm economy rules that benchmark. More macines mean better performance. Frame rendering scales linearly with machine count. Which means price performance counts. Which means you can buy 3 dual athlon machines instead of a single G4 tower and not just double the G4s performance, but render frames 6 times as fast for a given cost.
Have fun with quake at 640 x 480... I guess...
I don't understand how price would be a bogus factor.
On an application with performance that scales linearly with more machines, such as the distributed client or webserving, the limit to performance is total cost and price/performance.
How fast do you want to spend.
There is no point introducing price/performance into a comparison of photoshop filters because the application will not scale across multiple machines.
In that case you simply compare the two machines performance and question whether you are willing to spend an extra $2000 for inferior performance just because you like the look and feel of one machine over another.
There are other factors. If you are limited by rackspace, then you need to know how much performance you can stuff into 1u reguardless of cost. In Apple's case, they can place a dual processor X Serve in 1u. In competition to X Serve, there are companies putting two dual processor Athlons into a single 1u rack.
Another factor would be performance per watt. If you needed to perform ogr calculations drawing the minimum amount of power, the obvious power difference between running two Macs or a single Athlon become clear.
So, you are right, it is good to have choice. It is good to be able to choose something besides proprietary software on proprietary hardware. Viva la x86, Viva la Linux.
It is rated a troll because I regulary cut through the "Steve Jobs reality distortion feild".
No one likes their distorted reality crushed, especially the hardened core of mac zealots that apple has been reduced to.
There are not as many "switchers" as Apple would have us belive. Plenty of "twitchers", people that like OS X but balk at the platform's inflated price and poor absolute performance.
This will get a poor mod also, even though I am just answering your question.
CPU Name MHz Contest Speed StdDev StdErr% Count
Power PC 7450/7455 G4 1000 rc5 10,525,403 50841 0.4 3
AMD Athlon XP (Palomino) 2100 rc5 6,128,961 42741 0.6 3
Intel Pentium 4 2800 rc5 5,522,975 0 0.0 1
rc5 has definitely been optimized for altivec. It is a credit to the Mac community that when source code is available, they do an excellent job with optomized ports.
Power PC 7450/7455 G4 1000 ogr 10,579,931 338 0.0 2
AMD Athlon XP (Palomino) 2100 ogr 13,476,744 298544 2.2 2
Intel Pentium 4 2200 ogr 7,684,803 0 0.0 1
Knowing that the Mac community goes out of their way to create altivec optimized ports, the G4's performance in ogr is also quite telling.
While it is unfortunate that the Pentium 4 2.8 has not been benchmarked in ogr, the best it could hope for is about 9.7 million, showing off it's crippled integer instruction prosessor.
While these are not the fastest processors from Apple or AMD, if the numbers scale linearly, as they seem to in refrence to older processors, both processor types should see a 25% performance improvement with current clock speeds.
This is a sticky point. At $3,299.00 for a dual G4 1.25 compared to just over $1000 for a complete dual Athlon MP system, the keys processed at a given cost look like this:
26449827 running two rc5 clients on a dual 1.25 G4.
36773766 running six rc5 clients on 3 dual 2100MPs.
The Athlons are 40% faster at a given price in rc5.
26449827 running two ogr clients on a dual 1.25 G4.
80860464 running two ogr clients on 3 dual 2100MPs.
The Athlons are 305% faster at a given price in ogr.
So, there is a case where the G4 does not get crushed in absolute performance. But like other closed/proprietary hardware manufacturers such as sun and sgi, they pose absolutely no competition to open platforms on price/performance.
Scorecard uses "iComp"
"The iCOMP value is the weighted geometric mean of four benhcmarks;
ZD(Ziff-Davis) Bench - 68% 16-bit Whetstone - 2% SPECint92 - 25% SPECfp92 - 5%"
"Scorecard" shows the G4 and Athlon 2600 dead even at 89% of the fastest processor, The P4 2.8. That is sort of funny, considering ZD bench which represents 68% of the weighted mean of iComp is not available on the Mac.
Here are some additional resources using more modern benchmarks:
Benchmarks using Spec2000 instead of the woefully obsolete Spec92.
Comparison of The Dual G4 to other platforms in after effects rendering
Here is some raw data from the obsolete 16-bit Whetstone which shows the G4 not even able to match a Pentium II clock for clock.
This comparison shows how the G4 compares to a variety of rivals.
This shows that under most conditions, the SDR equiped Dual G4 is faster than than the DDR version. Meaning the 1.25 GHz G4 is crippled with DDR. Which means the numbers from the other benchmark links will not scale linearly. The Dual 1.25 G4 is actualy slower clock for clock than the G4 with SDR.
Also refrence Ars Technica's Seti benchmark showing that a Dual G4 1 Ghz produces work units at the same speed that a Dual PIII 1 Ghz.
Has anyone found any other Benchmark comparisons of the G4 (besides the single photoshop benchmark on Apple's website) that might shed some light on the debate?
So, choose whatever platform you like, but be informed about what you are paying for.
You buy a mac, you buy it because you like OS X, which is far and away the nicest OS out there right now. You do not buy a mac because of performance, as anyone who follows the links can very plainly see.
Insightful? That is an absolute Troll.
How much dry elitist apple crap do we have to put up with?
This is a cool mod. Why does BitGeek feel the need to break it down?
He says "slow processor" when the 1800 mhz Athlon has benchmarked on many independent tests as being faster than the fastest dual G4 system Apple has. Bitgeek talks about superior architecture and bandwidth when the most recent macs use DDR just as a marketing ploy. Their performance is actually 3-5% slower than identicly clocked G4s using SDRAM. They cobble fast memory onto their low bandwith processor with a memory translator hub similar to Intel's failed MTH design... All for marketing.
He says "If you can't afford" showing extreme class distain and prejudice.
Absolute crap post.
So, bitgeek, tell us, where are you coming from on this? Why the need to criticize other people's art? Did you gain anything? Does it make you feel good? Are you really that insecure? How invested are you in Apple?
Disclaimer: I sold my Apple shares in July of 2000.
Perhaps he should have too.
Is a member of the Worcester Area Pilots Association.
Nice little slashcode site.
Definite honorary member of the cool geek society.
S@H sends out so many e-mails, it is considered a spam site by most filters.
They will send you your password. Just disable your filters for the 20 minutes it might take them to send you a password.
I had the same trouble with web based mail provider filtering mail from seti
Look at those dirty rat bastards, Microsoft, bundling Internet Explorer in with the OS for free.
It is no wonder that Netscape couldn't compete.
Look at those dirty rat bastards, Apple, bundling InDesign with Powermacs for free.
It is no wonder that Quark couldn't compete.
The smell... liquid hydrogen and liquid oxygen...
Think about it for a minute.. It will come to you.
A robotic car was driven from california to new york. 99% of the trip, the computer was in control, including navigation.
The human driver had to deal with gas stations only.
With spider web protien for hair.
Beats goats with spider web protien for milk..ug.
Added bonus: you have to shear sheep anyway, except now your sweater is potentially bulletproof.
A zip code refrence is available at:
http://www.house.gov/writerep/
That is easy. I belive he wants a voter registration database for the districts themselves...
Mass mailings are Spam.
Keeping an individual specific private database is Privacy invasion. Can I call up his consulting firm and ask them what they know about me? Can I have myself removed? Would anyone realisticly opt-in to this? no. 99% would opt out.
Walking into Kragen Auto Parts I'm pissed that I can't just pay for my parts and leave. No, they keep a damn database. They know everything they've sold me for the past 5 years. Problem is I've only benn shopping ther for 2 years. The database is cataloged by phone numbers. The jerkoff that had my phone number wrote bad checks all over town.
So not only are there databases everywhere about me, most of them are entirely wrong. Kragen can't delete the database entry until they get paid. I get hasseled just because I want a battery.
Databases are privacy invasive bullshit.
Political database... what the hell is this? 1984?
The good pictures will be from seperation and re-entry.
I still think they should fly the otherwise wasted tank into orbit to be used as parts or huge additions to the space station.
Multiply the space station's internal volume by 10 in just one trip. After all the tank is a high pressure vessel. It should be able to hold 14.5 lb/sq in atmospheric.
Just give the shuttle a bit more power, drop the boosters and park the tank next to ISS.
Would make ISS look like a toy.
"take address information, and create mailings"
You won't find much in the way of open source software designed to enable you to create privacy invasive databases or spam your special interest group's message.
You just opened up a vipers nest as far as that question is concerned here on slashdot. Privacy invasion and spam are extreme hot button issues here.
You want to do them both. Go spend the money with microsoft. They love people like you.
They got beaten today.
Down 7% on Intel's 2 cent per share dividend.
They'll get beaten again tomorrow.
They'll get beaten at Christmas.
They'll get beaten until Sledgehammer is released.. not Claw hammer which will have no x86-64 desktop software support right off the bat, and will have to rely solely on it's pure x86 performance.
Microsoft shafted them on the X-box because Intel paid Microsoft 200 million to use the Pentium III. Nvidia was stuck with an unused AMD integrated chipset for X-box and Nforce was born.
Intel will pay Microsoft to shaft them again. No x86-64 Windows XP for AMD despite AMD testifying on Microsoft's behalf in exchange for anti-trust testimony. AMD made an unenforceable agreement with Microsoft. To enforce it would perjure themselves.
So Intel wins again.
Until Sledgehammer arrives. Sledgehammer is a server/workstation chip and will have full support of the dominant server operating system, Linux. Microsoft must support Sledgehammer or risk losing more of their already weak server market share.
Long after Microsoft has done the work to get Windows running on the Server, Microsoft will incorporate x86-64 support into their desktop OS.
Probably about the same time as they support Hyper threading and SSE3 for Intel.
To keep those little brilliant (bastard) children from wrecking your shiny new Linux lab.
Create a Highway Area Network. Not only should individual vehicles be smart enough to drive themselves, they should communicate and cooperate with each other to get everyone where they need to go as safely as possible. Just as a train is linked by physical means, vehicles should link wirelessly to achieve the same end - bumper to bumper at 100 mph, but safely.
Exactly what human nature cannot do for itself.
Driving is a repetitive, monotonous task. That is exactly what computers are good at.
.
On the last page of Tom's advertizers section is a section entitled "Editorial content sponsorships"
.18 micron PIII recall and their support of Rambus when it was clear that Rambus was trying to assert a monopoly on ram.
Thus the about face on Rambus and Intel in general.
Tom "will never forget the deceit of Intel" revolving around the memory translator hub recall, the 1.13 Ghz
A few days after I noticed the "Editorial content sponsorships", he posted the editorial "proving" that Rambus was better than DDR because on a heavily overclocked system, DDR stops scaling.
Tom is a heel. I once admired him. He brought Intel to it's knees by proving the 1.13 processor was factory overclocked.
Now Tom is an Intel/Rambus/Bapco stool.
At least Alex "Sharky" Ross had the moral spine to leave Sharky's when that site became an Intel ad.
Never give up sleep time for study time. During sleep your mind makes the transfer to long term memory. If you are not sleeping 7 hours, you might as well not study at all that day.
And get a good outside accountant. Nothing will go wrong with your business that you cannot fix by delegating, except your in house accountant stealing from you.
So, study, sleep, delegate and don't let the mice play while you are away.
Only 600 posts to go.
This is the same transistor that IBM was talking about. They are sharing (or stealing) this technology. Electron microscope pictures are of the same transistor.
IBM is a trustworthy computing founding member and is on the steering comittee..
They are pushing DRM
The only major company absent from trusworthy computing is VIA... I can't seem to find Sony on the list either but they have to be pushing a DRM of their own. They are both content producer and hardware manufacturer.
C GATE = ( [epsilon] OX [epsilon] 0/TOX)L GATE + 0.26 fF/m
and
P ACTIVE = CEFFV 2DDf
and
[tau] = CGATEVDD/IDSAT
Oh, of course. I understand much better now.
"All the information from the 2000 census is available for free from the United States government," Anderson said. "No one is out in front of their mailbox with a shotgun to prevent direct marketers from distributing targeted print advertising. Yet with television, there is a greater level of concern."
I use A P.O. Box.
Are they somehow going to give me the anonymity of P.O. television?
No.
They will use every means at their disposal to generate income once their "???" doesn't become "Profit!". And that means they are going to sell me out.
My mail does not come to where I live.
My Television does.