toddicus
Nov 3, 06:08 AM
OK to swerve this thread back on topic, what if Apple is planning to unleash a massive multi-core assault and fill that big middle gap in the lineup at the same time?
Here's the theory;
January Macworld Steve unveils the 8 core Mac Pro, no surprises there, shows off the massive power using Leopard demo's etc. Great for Pro's (like Multimedia and myself) but not much use to the average guy. Prices stay the same or even rise slightly, after all, we are talking 8 cores here. Previously you needed to spend $7-8k to get that kind of power. But what if the one more thing was a Kentsfield Mac Pro (using the C2Q6600), a i975 Mb with DDR2 ram, etc, etc . Sloting into that $1400-2000 zone? I dont see this competing with the iMac, esp. since you get a 24" screen with your $2000 iMac. It's just another choice. Use the same case, make it black or something, but you now have
Mac Mini 2 cores
iMac 2 cores + Widescreen display
Mac Prosumer 4 cores + upgradeable
Mac Pro 8 cores for ultimate power.
Sounds good......:)
I'd have to say my opinion is this is very unlikely. Apple has stuck with the four squares of producst, pro, consumer in desktop and portable for years. A sub mac pro without a xeon wouldn't fit into that model. While you could certainly make nice Mac out of a quad-core Core2 extreme I just don't see it happening. I think the only way we'll see conroe/kentsfield in Macs is if they some how got the components needed small enough and cool enough to cram into all sizes of iMacs (if they don't fit in the smallest, they won't go in any, keeps them all the same), and I don't think that will happen.
I never cease to be amazed though, everytime Steve gives a keynote I feel like he announces stuff I just wouldn't have thought of. So, maybe there is a chance, just not sure what they'd call it, or who it'd be targeted at. My gut says it won't happen.
Here's the theory;
January Macworld Steve unveils the 8 core Mac Pro, no surprises there, shows off the massive power using Leopard demo's etc. Great for Pro's (like Multimedia and myself) but not much use to the average guy. Prices stay the same or even rise slightly, after all, we are talking 8 cores here. Previously you needed to spend $7-8k to get that kind of power. But what if the one more thing was a Kentsfield Mac Pro (using the C2Q6600), a i975 Mb with DDR2 ram, etc, etc . Sloting into that $1400-2000 zone? I dont see this competing with the iMac, esp. since you get a 24" screen with your $2000 iMac. It's just another choice. Use the same case, make it black or something, but you now have
Mac Mini 2 cores
iMac 2 cores + Widescreen display
Mac Prosumer 4 cores + upgradeable
Mac Pro 8 cores for ultimate power.
Sounds good......:)
I'd have to say my opinion is this is very unlikely. Apple has stuck with the four squares of producst, pro, consumer in desktop and portable for years. A sub mac pro without a xeon wouldn't fit into that model. While you could certainly make nice Mac out of a quad-core Core2 extreme I just don't see it happening. I think the only way we'll see conroe/kentsfield in Macs is if they some how got the components needed small enough and cool enough to cram into all sizes of iMacs (if they don't fit in the smallest, they won't go in any, keeps them all the same), and I don't think that will happen.
I never cease to be amazed though, everytime Steve gives a keynote I feel like he announces stuff I just wouldn't have thought of. So, maybe there is a chance, just not sure what they'd call it, or who it'd be targeted at. My gut says it won't happen.
jessefoxperry
Sep 12, 04:34 PM
is apple.com not loading for anyone else? it was fine before with the new content, now everything in the middle is missing. /itunes /ipod still work like they should. weird.
GGJstudios
May 3, 05:30 PM
You told the 100% gospel truth. There IS malware for the Mac
Yes, there is malware for the Mac. I don't see anyone in this thread or others claiming that there isn't. ElCidRo's statement implied that there was a prevalent myth that Macs had no malware which is not true, and triggered the negative responses by throwing out the "fanboy" attack. It was very clear that the post was inflammatory in nature.
What IS true is that there are no viruses in the wild that run on Mac OS X, and there hasn't been for the past 10 years, since it was introduced. The handful of trojans that exist are easily avoided/thwarted by a user exercising a reasonable degree of common sense. It has nothing to do with being a "fanboy". It has to do with facts.
Yes, there is malware for the Mac. I don't see anyone in this thread or others claiming that there isn't. ElCidRo's statement implied that there was a prevalent myth that Macs had no malware which is not true, and triggered the negative responses by throwing out the "fanboy" attack. It was very clear that the post was inflammatory in nature.
What IS true is that there are no viruses in the wild that run on Mac OS X, and there hasn't been for the past 10 years, since it was introduced. The handful of trojans that exist are easily avoided/thwarted by a user exercising a reasonable degree of common sense. It has nothing to do with being a "fanboy". It has to do with facts.
Travisimo
Mar 18, 11:10 AM
Meh... I use MyWi occasionally, meaning only once or twice every TWO months.
Now I would spend an extra $5-10 a month if ATT offered tethering with a 5-10 Gigabyte total data cap on both phone and tethering usage. Spending an extra $25+ to be on a capped 2-4GB plan is BuL*Sh&^ if it means that I have to give up my unlimited plan as well as unrestricted 3G via My3G.
This. I wouldn't mind paying a bit more for tethering, but the $20/mo extra or nothing is really unacceptable. For those of us who only tethering sporadically, it's really a waste of money paying $20/mo. If the carriers really want an extra revenue stream from tethering, they should have different options available.
I would easily pay $5-10 more a month for 1GB of tethering data, and for those who want 2+ gigs for tethering, then $20/mo is fine. They really need a lower option.
Now I would spend an extra $5-10 a month if ATT offered tethering with a 5-10 Gigabyte total data cap on both phone and tethering usage. Spending an extra $25+ to be on a capped 2-4GB plan is BuL*Sh&^ if it means that I have to give up my unlimited plan as well as unrestricted 3G via My3G.
This. I wouldn't mind paying a bit more for tethering, but the $20/mo extra or nothing is really unacceptable. For those of us who only tethering sporadically, it's really a waste of money paying $20/mo. If the carriers really want an extra revenue stream from tethering, they should have different options available.
I would easily pay $5-10 more a month for 1GB of tethering data, and for those who want 2+ gigs for tethering, then $20/mo is fine. They really need a lower option.
recursivejon
Mar 20, 02:23 PM
If this is true (transfer of the music without DRM to be added by iTunes), then couldn't anyone with a bit of networking knowledge just pipe the packets into a file when they purchase something from the store using iTunes?
okboy
Apr 8, 11:21 PM
Not really impressed by the whole push into gaming-- gaming is what I use my iPT and iPad for the least. In fact, Game Center is in a folder marked "Undeletable Crap" on both devices, along with address book, FaceTime, calendar, and stocks (on the iPT). Really wish Apple allowed you to delete whatever you wanted-- but of course they know what I want better than I do.
Oh boo hoo, someone get this guy a tissue. How can it bother you that there is some game executive somewhere inside Apple HQ?
Oh boo hoo, someone get this guy a tissue. How can it bother you that there is some game executive somewhere inside Apple HQ?
KnightWRX
Apr 28, 09:56 AM
The hardware components in a server go through much more testing for reliability. They are meant to work 24/7.
Thank you for pointing this out to me. Just know that I am right now working on a C7000 chassis, specifically on a BL870c Blade, even more specifically on a Itanium version VM running on said blade.
I think I know about servers. ;) This doesn't change the fact of what I said : Servers aren't pcs (as in personal computers) and I doubt they are included in this chart. So why are you even bringing them up ?
Thank you for pointing this out to me. Just know that I am right now working on a C7000 chassis, specifically on a BL870c Blade, even more specifically on a Itanium version VM running on said blade.
I think I know about servers. ;) This doesn't change the fact of what I said : Servers aren't pcs (as in personal computers) and I doubt they are included in this chart. So why are you even bringing them up ?
r1ch4rd
Apr 23, 03:46 PM
I've concluded American Atheists who are continually challenged on their beliefs and "surrounded by enemies" are more likely to read into atheism and all it entails, rather like a convert to a religion knows the religion better than people who were born into it. Europe is very secular, compared to the US at least, and thus a lot of people are "born into" atheism/secularism.
Well now, I don't think being an atheist actually entails anything. I certainly don't do anything specific related to it, but I know what you mean :).
I think this is a positive thing that people can be pushed towards science and understanding - even if it is because they are having to constantly defend themselves!
Who knows, perhaps they will find something they are passionate about and push forward science even further. Religious groups don't come accross as particularly progressive, so I guess it's up to "us" ;)
Well now, I don't think being an atheist actually entails anything. I certainly don't do anything specific related to it, but I know what you mean :).
I think this is a positive thing that people can be pushed towards science and understanding - even if it is because they are having to constantly defend themselves!
Who knows, perhaps they will find something they are passionate about and push forward science even further. Religious groups don't come accross as particularly progressive, so I guess it's up to "us" ;)
Multimedia
Oct 26, 07:06 PM
Mac Pro is only true desktop offering from Apple. That's the problem.
Not that many individuals really want that much power.
However, they do intensive enough tasks requiring more power that exceeds what iMac can offer. The price and power ratio of iMac is just not enough.
Apple really needs something between "Pro" and "Consumer".
If iMac offered the ability to work as monitor, I wouldn't be disappointed by this much.
This is getting old already, but what I need is a decent Conroe Desktop with around 1500 USD price tag.I could not agree more. Apple has got to be in final stages of deploying a sub $2k Kentsfield desktop for 2007 or they will be missing one hell of a sales opportunity.
Not that many individuals really want that much power.
However, they do intensive enough tasks requiring more power that exceeds what iMac can offer. The price and power ratio of iMac is just not enough.
Apple really needs something between "Pro" and "Consumer".
If iMac offered the ability to work as monitor, I wouldn't be disappointed by this much.
This is getting old already, but what I need is a decent Conroe Desktop with around 1500 USD price tag.I could not agree more. Apple has got to be in final stages of deploying a sub $2k Kentsfield desktop for 2007 or they will be missing one hell of a sales opportunity.
citizenzen
Apr 24, 10:03 AM
Intelligence has something to do with it.
Liberals and Atheists Smarter? Intelligent People Have Values Novel in Human Evolutionary History, Study Finds
ScienceDaily (Feb. 24, 2010) (http://www.sciencedaily.com/releases/2010/02/100224132655.htm) — More intelligent people are statistically significantly more likely to exhibit social values and religious and political preferences that are novel to the human species in evolutionary history. Specifically, liberalism and atheism, and for men (but not women), preference for sexual exclusivity correlate with higher intelligence, a new study finds.
The study, published in the March 2010 issue of the peer-reviewed scientific journal Social Psychology Quarterly, advances a new theory to explain why people form particular preferences and values. The theory suggests that more intelligent people are more likely than less intelligent people to adopt evolutionarily novel preferences and values, but intelligence does not correlate with preferences and values that are old enough to have been shaped by evolution over millions of years."
"General intelligence, the ability to think and reason, endowed our ancestors with advantages in solving evolutionarily novel problems for which they did not have innate solutions," says Satoshi Kanazawa, an evolutionary psychologist at the London School of Economics and Political Science. "As a result, more intelligent people are more likely to recognize and understand such novel entities and situations than less intelligent people, and some of these entities and situations are preferences, values, and lifestyles."
Data from the National Longitudinal Study of Adolescent Health (Add Health) support Kanazawa's hypothesis. Young adults who subjectively identify themselves as "very liberal" have an average IQ of 106 during adolescence while those who identify themselves as "very conservative" have an average IQ of 95 during adolescence.
Similarly, religion is a byproduct of humans' tendency to perceive agency and intention as causes of events, to see "the hands of God" at work behind otherwise natural phenomena. "Humans are evolutionarily designed to be paranoid, and they believe in God because they are paranoid," says Kanazawa. This innate bias toward paranoia served humans well when self-preservation and protection of their families and clans depended on extreme vigilance to all potential dangers. "So, more intelligent children are more likely to grow up to go against their natural evolutionary tendency to believe in God, and they become atheists."
I think the last paragraph is a key to why atheists hold out for proof. We've seen time and time again over history where something that has been attributed to the supernatural or a God turned out to be quite natural.
Likewise questions about the origins of the universe, that today seem utterly mysterious and unanswerable, may one day be resolved and explained within the natural confines.
Atheists are loathe to latch on to supernatural conclusions when that camp has been proven wrong time and time and time again.
Liberals and Atheists Smarter? Intelligent People Have Values Novel in Human Evolutionary History, Study Finds
ScienceDaily (Feb. 24, 2010) (http://www.sciencedaily.com/releases/2010/02/100224132655.htm) — More intelligent people are statistically significantly more likely to exhibit social values and religious and political preferences that are novel to the human species in evolutionary history. Specifically, liberalism and atheism, and for men (but not women), preference for sexual exclusivity correlate with higher intelligence, a new study finds.
The study, published in the March 2010 issue of the peer-reviewed scientific journal Social Psychology Quarterly, advances a new theory to explain why people form particular preferences and values. The theory suggests that more intelligent people are more likely than less intelligent people to adopt evolutionarily novel preferences and values, but intelligence does not correlate with preferences and values that are old enough to have been shaped by evolution over millions of years."
"General intelligence, the ability to think and reason, endowed our ancestors with advantages in solving evolutionarily novel problems for which they did not have innate solutions," says Satoshi Kanazawa, an evolutionary psychologist at the London School of Economics and Political Science. "As a result, more intelligent people are more likely to recognize and understand such novel entities and situations than less intelligent people, and some of these entities and situations are preferences, values, and lifestyles."
Data from the National Longitudinal Study of Adolescent Health (Add Health) support Kanazawa's hypothesis. Young adults who subjectively identify themselves as "very liberal" have an average IQ of 106 during adolescence while those who identify themselves as "very conservative" have an average IQ of 95 during adolescence.
Similarly, religion is a byproduct of humans' tendency to perceive agency and intention as causes of events, to see "the hands of God" at work behind otherwise natural phenomena. "Humans are evolutionarily designed to be paranoid, and they believe in God because they are paranoid," says Kanazawa. This innate bias toward paranoia served humans well when self-preservation and protection of their families and clans depended on extreme vigilance to all potential dangers. "So, more intelligent children are more likely to grow up to go against their natural evolutionary tendency to believe in God, and they become atheists."
I think the last paragraph is a key to why atheists hold out for proof. We've seen time and time again over history where something that has been attributed to the supernatural or a God turned out to be quite natural.
Likewise questions about the origins of the universe, that today seem utterly mysterious and unanswerable, may one day be resolved and explained within the natural confines.
Atheists are loathe to latch on to supernatural conclusions when that camp has been proven wrong time and time and time again.
LegendKillerUK
Mar 18, 08:22 AM
You agreed to a contract. Live with it or pay to get out of it, couldn't be simpler.
Apple OC
Apr 24, 06:15 PM
just what we need in the world ... a McPeace treaty:cool:
aegisdesign
Oct 26, 05:11 AM
JUST IMAGINE A COMPUTER IN WHICH EACH PIXEL IS CONTROLLED BY A SINGLE PROCESSOR.
I've used one. Back in the 1980s, beginning of the 90s. The low end model had 1024 processors and the high end model 4096 processors. It was a pig to program. When drawing on the screen you split the task at hand up into many parallel threads each drawing a part of the screen. Not quite 1 CPU per pixel but you get the idea.
I've used one. Back in the 1980s, beginning of the 90s. The low end model had 1024 processors and the high end model 4096 processors. It was a pig to program. When drawing on the screen you split the task at hand up into many parallel threads each drawing a part of the screen. Not quite 1 CPU per pixel but you get the idea.
larrybeo
Mar 18, 09:43 AM
People who complain that your service provider is going to make you follow the ru:eek:les unnerve me with their uncanny ability to disregard all that stands to reason with the sustainability of your "toys." They are like little sissies on the playground crying after a Barbie Doll has been taken from them. Those people should man up and start paying for the footprint they leave on the network.
Bill McEnaney
Apr 23, 03:42 PM
The word translated "day" can mean various lengths of time, not just a 24-hour period.
I think ancient Jews thought each day began at dawn and ended at sunset. If I'm right, they would have thought summer days were longer than winter ones.
Our Lord died died on Good Friday and rose on Easter, but does anyone know exactly what time he rose? The Bible says he rose on the third day. Say he died at 3:00 PM. on Friday. Then a 24-hour day from his death would end at 3:00 PM on Saturday, and another 24-hour day would end at 3:00 PM on Sunday. That's only two 24-hour days. We say it's daytime when the sun is shining and that it's nighttime when it's dark outdoors. Even we talk as though the word "day" stands sometimes stands for less than 24 hours.
Many of the Bible's atheistic critics oversimplify because the ignore the Bible's literary genres, the meanings of ancient expressions, cultural details, and other details. If you say something, your sentence, the string of words, differs from what it means. That's why you can translate a sentence from one language to another. When you translate a sentence from English to French, the French sentence needs to mean what the English one means, or there's something wrong with the translation.
To know what, say, Genesis 1:1 means, you need to know what its author meant by the words it consists of. If you impose a 21st-century meaning on a sentence that meant something else when the author wrote it, you're misinterpreting what he said.
You and I see three colored objects and three people. You tell me, "Bill, Green is the third one from the left." You're talking about the third person, a man named "Joe Green," when I think you're talking about the third colored object. Green is the color of the third object from the left. The word "Green" is the last name of the man who's third from the left. The proposition "Green is the third one from the left" is true in both cases, but the string of words means one thing when you talk about the man. It means something else when you're talking about the green object. To find out which truth you're telling me, I need you to tell me that you're talking about the colored object.
You wake at 7:00 AM on Friday. The next calendar day begins at midnight, but there's only 17 hours between 7 AM and midnight. Truth is objective, but the meanings of words, phrases, and sentences depend on context. So do the referents of the words, the people, places, or things that words, phrases, and sentences denote.
I think ancient Jews thought each day began at dawn and ended at sunset. If I'm right, they would have thought summer days were longer than winter ones.
Our Lord died died on Good Friday and rose on Easter, but does anyone know exactly what time he rose? The Bible says he rose on the third day. Say he died at 3:00 PM. on Friday. Then a 24-hour day from his death would end at 3:00 PM on Saturday, and another 24-hour day would end at 3:00 PM on Sunday. That's only two 24-hour days. We say it's daytime when the sun is shining and that it's nighttime when it's dark outdoors. Even we talk as though the word "day" stands sometimes stands for less than 24 hours.
Many of the Bible's atheistic critics oversimplify because the ignore the Bible's literary genres, the meanings of ancient expressions, cultural details, and other details. If you say something, your sentence, the string of words, differs from what it means. That's why you can translate a sentence from one language to another. When you translate a sentence from English to French, the French sentence needs to mean what the English one means, or there's something wrong with the translation.
To know what, say, Genesis 1:1 means, you need to know what its author meant by the words it consists of. If you impose a 21st-century meaning on a sentence that meant something else when the author wrote it, you're misinterpreting what he said.
You and I see three colored objects and three people. You tell me, "Bill, Green is the third one from the left." You're talking about the third person, a man named "Joe Green," when I think you're talking about the third colored object. Green is the color of the third object from the left. The word "Green" is the last name of the man who's third from the left. The proposition "Green is the third one from the left" is true in both cases, but the string of words means one thing when you talk about the man. It means something else when you're talking about the green object. To find out which truth you're telling me, I need you to tell me that you're talking about the colored object.
You wake at 7:00 AM on Friday. The next calendar day begins at midnight, but there's only 17 hours between 7 AM and midnight. Truth is objective, but the meanings of words, phrases, and sentences depend on context. So do the referents of the words, the people, places, or things that words, phrases, and sentences denote.
Edge100
Apr 15, 11:52 AM
I feel sad at how many of you are totally distorting the message of Christ. The real blame goes on those who use his name to sully his very purpose. Those false Christians make me sick.
I agree. People should focus exclusively on the New Testament, where hateful behaviour is unequivocally denounced.
Take slavery, for instance. If ever there was a hateful action that we call all be united against, it's slavery. Good thing the NT takes a firm stance against slavery...
ohh.... (http://skepticsannotatedbible.com/1pet/2.html#18)
damn... (http://skepticsannotatedbible.com/1tim/6.html#1)
it doesn't... (http://skepticsannotatedbible.com/col/3.html#22)
We should also commend Jesus himself, for taking such a firm stance against the horrors of the Old Testament...
wait... (http://bible.cc/matthew/5-18.htm)
that's not correct... (http://bible.cc/luke/16-17.htm)
I agree. People should focus exclusively on the New Testament, where hateful behaviour is unequivocally denounced.
Take slavery, for instance. If ever there was a hateful action that we call all be united against, it's slavery. Good thing the NT takes a firm stance against slavery...
ohh.... (http://skepticsannotatedbible.com/1pet/2.html#18)
damn... (http://skepticsannotatedbible.com/1tim/6.html#1)
it doesn't... (http://skepticsannotatedbible.com/col/3.html#22)
We should also commend Jesus himself, for taking such a firm stance against the horrors of the Old Testament...
wait... (http://bible.cc/matthew/5-18.htm)
that's not correct... (http://bible.cc/luke/16-17.htm)
alexf
Aug 29, 12:17 PM
I could not care any less.
Although, I do know of one thing Apple does that hurts the environment. They make me drive 3 hours to get to the closest Apple Store and 3 hours to get back home plus sitting in all the traffic in Atlanta. However, I drive a Nissan Armada (of course it has a V8) so I'm not too worried about gas consumption. ;)
Yep, just another wasteful American. Same sad story.
Although, I do know of one thing Apple does that hurts the environment. They make me drive 3 hours to get to the closest Apple Store and 3 hours to get back home plus sitting in all the traffic in Atlanta. However, I drive a Nissan Armada (of course it has a V8) so I'm not too worried about gas consumption. ;)
Yep, just another wasteful American. Same sad story.
Analog Kid
Oct 26, 01:42 AM
Do either IBM or Motorola have a quad-core chip on the horizon?
How many cores in a Cell? Nine, depending on how you count...
How many cores in a Cell? Nine, depending on how you count...
KnightWRX
May 2, 03:35 PM
It can't affect the user's account if the user doesn't proceed with the installation. If the installer is closed without proceeding, nothing is affected.
You're not quite understanding what I'm saying or the situation here. Safari auto-downloads a zip file, runs it through Archive Utility which extracts something and then runs it.
It happens to be an installer this time. What if next time it's a malicious piece of code ? Why did it auto-execute, under what conditions and could these conditions be used to execute something other than an installer ?
Think a bit beyond the current situation. The malware authors do.
It also scans for Mac malware.
ie, not viruses. ClamAV's original intent was Linux e-mail servers and while it may have morphed into more, it's existence is not the proof of Mac viruses.
You're not quite understanding what I'm saying or the situation here. Safari auto-downloads a zip file, runs it through Archive Utility which extracts something and then runs it.
It happens to be an installer this time. What if next time it's a malicious piece of code ? Why did it auto-execute, under what conditions and could these conditions be used to execute something other than an installer ?
Think a bit beyond the current situation. The malware authors do.
It also scans for Mac malware.
ie, not viruses. ClamAV's original intent was Linux e-mail servers and while it may have morphed into more, it's existence is not the proof of Mac viruses.
happydude
Oct 7, 02:24 PM
Of course Android might surpass the iPhone. The iPhone is limited to 1 device whereas the Android is spanned over many more devices and will continue to branch out.
StudioGuy
Sep 26, 10:45 AM
I originally thought that this would make a nice "best" model for the Mac Pro, but the 5160 is surely a great chip. Given the slower clock speed (although not always a good indicator) and more cores, this might be a great chip for a Server, like an updated XServe.
Not sure if the software guys are going to catch up enough in multithreading to make good use of 8 cores, but several folks on an XServe would appreciate it.
Not sure if the software guys are going to catch up enough in multithreading to make good use of 8 cores, but several folks on an XServe would appreciate it.
Elfear
Oct 31, 11:00 PM
Sorry for the noob question, but does anyone know how well Maya 7 will scale with 8 cores? My buddy is debating whether to buy a single Kentsfield or step up to dual Clovertons. He has a freelance business in which he uses Maya 7 quite a bit. Thanks.
javajedi
Oct 10, 10:28 PM
Originally posted by ddtlm
javajedi:
Yes, the JVM is the deciding factor here. If the Java takes that damn long on a G4 but goes fast on a P4, can can rest assured that the JVM Apple is distributing sucks compared to whatever one the x86 machines are using.
There is no way in heck that the performance delta can be so large without a large difference in quality of JVM. G4's may be slower, but they are not as slow as those number indicate.
Like I've been saying, when you start to see 5x leads by the PCs you need to start asking questions about the fairness of the benchmark. The G4 is better than 1/5 the speed. There are very few things were a P4 can get better performance per clock than a G4.
BTW:
Your G3 results as bizzarre as well, because of the contrast between them and the G4 results. Do not take it as proof one way or the other of the G3 or other IBM chips being superior to the G4. What we have here are raw numbers that defy a simple explanations. We should ask why these numbers are popping up, rather than running off with them as if they were uttered by a great voice in the sky or somthing.
I should note that the 90 second and 72 second results I just recently posted are from my cocoa implementation, not java.. so the jvm is out of the picture now on the mac.
Do not take it as proof one way or the other of the G3 or other IBM chips being superior to the G4.
Don't worry, I don't make assumptions like that. And no, I don�t think this does defy simple explanations. I will say that, what we are starting to see here is evidence that the scalar units (integer and fpu) in the IBM 750FX (G3) are more efficient than those in the Motorola G4.
If this is true, then my program hit it right on the nail. Also if this is true, it means there exist theoretical situations when using non altivec code that it would be faster on one of these newer G3 chips.
Also what alex said about how tedious it was to make altivec code, I would agree there is some truth to this. When you vectorize code (either for the P4 or G4), if you don't watch your p's and q's you can actually slow *down* your code. Just because you use the nice and special vector registers on these G4 and Pentium 4 processors does not mean you gain 5 times the speed. You literally have to take your methods back to the drawing board. You will only get peak performance out of pipelined, fully vectorized code.
None the less, scalar operations on both G3/G4 are miserable compared to x86. The JVM is no longer the deciding factor in the performance delta. It's out of the equation on the Mac since the benchmark is now a 100% native cocoa application with c code and no longer java. Mean while on the x86, the benchmark remains java.
70-ish seconds navtive on a G3
90-ish seconds on a native on a G4
5.9-6-ish seconds running under JVM 1.4.1 on a P4
javajedi:
Yes, the JVM is the deciding factor here. If the Java takes that damn long on a G4 but goes fast on a P4, can can rest assured that the JVM Apple is distributing sucks compared to whatever one the x86 machines are using.
There is no way in heck that the performance delta can be so large without a large difference in quality of JVM. G4's may be slower, but they are not as slow as those number indicate.
Like I've been saying, when you start to see 5x leads by the PCs you need to start asking questions about the fairness of the benchmark. The G4 is better than 1/5 the speed. There are very few things were a P4 can get better performance per clock than a G4.
BTW:
Your G3 results as bizzarre as well, because of the contrast between them and the G4 results. Do not take it as proof one way or the other of the G3 or other IBM chips being superior to the G4. What we have here are raw numbers that defy a simple explanations. We should ask why these numbers are popping up, rather than running off with them as if they were uttered by a great voice in the sky or somthing.
I should note that the 90 second and 72 second results I just recently posted are from my cocoa implementation, not java.. so the jvm is out of the picture now on the mac.
Do not take it as proof one way or the other of the G3 or other IBM chips being superior to the G4.
Don't worry, I don't make assumptions like that. And no, I don�t think this does defy simple explanations. I will say that, what we are starting to see here is evidence that the scalar units (integer and fpu) in the IBM 750FX (G3) are more efficient than those in the Motorola G4.
If this is true, then my program hit it right on the nail. Also if this is true, it means there exist theoretical situations when using non altivec code that it would be faster on one of these newer G3 chips.
Also what alex said about how tedious it was to make altivec code, I would agree there is some truth to this. When you vectorize code (either for the P4 or G4), if you don't watch your p's and q's you can actually slow *down* your code. Just because you use the nice and special vector registers on these G4 and Pentium 4 processors does not mean you gain 5 times the speed. You literally have to take your methods back to the drawing board. You will only get peak performance out of pipelined, fully vectorized code.
None the less, scalar operations on both G3/G4 are miserable compared to x86. The JVM is no longer the deciding factor in the performance delta. It's out of the equation on the Mac since the benchmark is now a 100% native cocoa application with c code and no longer java. Mean while on the x86, the benchmark remains java.
70-ish seconds navtive on a G3
90-ish seconds on a native on a G4
5.9-6-ish seconds running under JVM 1.4.1 on a P4
javajedi
Oct 11, 12:26 PM
What you are saying makes a lot of sense. Now that I think about, I too recall reading this somewhere.
Now that we know the real truth about the "better standard FPU", I thought it was time to shed some light on non vectorized G4 integer processing.
It still does 200,000,000 calculations, but this time I'm multiplying ints.
Motorola 7455 G4@800Mhz: 9 seconds (Native)
IBM 750FX G3@700Mhz: 7 seconds (Native)
Intel P4@2600Mhz 2 seconds (Java)
PowerPC 7455 integer processing is consierabley better than floating point (obviously less work doing ints), but still less per cycle than the Pentium 4.
Very intresting the G4 looses both floating point and integer to the IBM chip, at a 100MHz clock disadvantage.
I'm still waiting to see that "better standard FPU" in the G4. It seems the G4 is absolutely useless unless you are fortunate to have vectorized (AltiVec) code.
Now that we know the real truth about the "better standard FPU", I thought it was time to shed some light on non vectorized G4 integer processing.
It still does 200,000,000 calculations, but this time I'm multiplying ints.
Motorola 7455 G4@800Mhz: 9 seconds (Native)
IBM 750FX G3@700Mhz: 7 seconds (Native)
Intel P4@2600Mhz 2 seconds (Java)
PowerPC 7455 integer processing is consierabley better than floating point (obviously less work doing ints), but still less per cycle than the Pentium 4.
Very intresting the G4 looses both floating point and integer to the IBM chip, at a 100MHz clock disadvantage.
I'm still waiting to see that "better standard FPU" in the G4. It seems the G4 is absolutely useless unless you are fortunate to have vectorized (AltiVec) code.
No comments:
Post a Comment