The G5 is competitive in that it matchesor outperforms the AMD Opteron (that frontside bus helps)
BerryTruFax: The HyperTransport bus used in the Opteron is the same thing used in the G5 Macs and XServes. Both AMD and Apple are in the HyperTransport Consortium.
g5 is to g4 as p4 was to p3 better overall IPC, less picky about memory latency, less power, basically a great thing to quad core if you're looking for perf/watt
I refuse to believe that the 28- and 31-stage Pentium 4 pipelines are a better thing than the 10-stage pipeline in the Pentium III, particularly when we're talking about IPC. Do you remember the fuss made about P4 being slower at the same clock speed than the PIII? That's proof it has worse IPC rate.
Neither the P4 or the G5 are lower-power than their predecessors and they fail to provide better performance/watt, in any configuration. This is why the P3 architecture has been adopted into the Pentium M line for low power use and the G4 processors remain the chips used at the core of Apple's iBooks and PowerBooks.
The great thing to do with the Pentium 4 architecture would be to put in on good Strained Silicon and SoI processes to push it above the 4.0GHz clockrate at which it is believed to be a very strong chip.
The differences between the G4 and G5 chips are what happens when you move from a desktop computer chip to a cut-down Big Iron chip (IBM's POWER4, IIRC). The G5's are inherently 64-bit capable in a way that the first three generations (Willamette, Northwood and Prescott) of the Pentium 4 are not, although there exist Prescott-based Pentium 4 processors with Intel's EM64T implementation.
BTW: http://arstechnica.com/ is your friend. Hannibal has done a good job of talking through the history of the Pentium chip family (1 & 2) and the PowerPC family (1 & 2, part 3 hasn't yet arrived) up to the G4's. There's discussion of the IBM POWER5 architecture too, and some commentary on pipelines in processor design (1 & 2). I learned a lot from these, and value their information. But I'm going to stop telling Granny to suck eggs now.
Quite bizarre. Am I right in thinking that you were visiting this seedy side of the Interweb using Internet Explorer? I'm surprised that you don't have something by the MoFo on your computer to put a gap between the OS and your web browsing for safety's sake (I won't pretend that this will make you immune to catching the STIs of the Internet, but it will help).
I think you underestimate the importance of not using Internet Explorer. Microsoft got out of an Anti-Trust suit by tying the core of the computer to IE, but in doing so made IE insecure as Woody Allen having a paranoid attack while all his friends party in the room next door.
the (completely uncalled for) war in Iraq has given extremists a made-to-order training camp And also, the extremists behind the military-industrial complex in Western nations reasons to keep making and selling weaponry -- to both sides.
Sir, I disagree with you. It matters not whether the product is amazing (and I have quite a lot of respect for the software writers at nVidia), but it's not Libre Open Source if we don't have the code. So I will not spend my %currency_units% on that stuff.
I read AMD's press release and think that the attitude of Intel is anticompetitive, but the actual transgressions did not seem wholly worthy of anti-trust litigation. I wouldn't go so far as to say you lines of "if you can't compete, legislate", but I doubt AMD will have success with this.
Tha would be a shame, because being able to buy a notebook computer from Dell with a Turion in it and without the Microsoft Tax would make a nice political message.
First, I don't click adverts. Ever. So I haven't missed them using Adblock, FlashBlock or Firefox's PopUp Blocker. And this means that noone has lost any click-through revenue from me. I will tolerate arguments that the page-impression revenue is impacted, but counter them with the fact that simple page-impression revenue hasn't been much since 1999.
Second, I'm saving myself bandwidth by not downloading unnecessary information from the sites I browse. And saving the Ad companies or the site hosts from paying for unnecessary bandwidth.
Third, DoubleClick want to monitor market trends, in so doing they would rid of the anonymity of my browsing. This would enable them to pitch at me exactly what the PR people want me to hear. This impinges the objectivity of purchasing decisions and intrudes on my Internet experience. If I want to go shopping, then I can go and look at products I want in an online store.
However, the integration of advertorial into what used to be enthusiasts' information sites, as has been criticised among the Tech sites in the last few months, is a greater concern to my shopping habits because unbiased information from research informs my shopping habits. I welcome a prejudice-free online catalogue of stuff to buy, like Froogle, but would like more information from such a catlogue than is presently supplied by Froogle and would miss impartiality from the enthusiast sites should they become advertorial.
I'm surprised you haven't been moderated down, laughed at or chased out of town by the fanboys...
Apple's Power Mac page speaks of four thermal zones and fans. I think that the present G5's have nine fans inside the box (source; highlighted by Google). It's a weird world of DRM that the cooling fan controls which operating system used...
I see. We have a disagreement about what constitutes Mac OS X. Perhaps we're also disagreeing over how lazy Apple are, as in how much of their software they will simply emulate -- the things they can't be bothered to fix.
I read that Apple has had in-house x86 editions for a few years, and I don't believe that their Intel-processor edition of OS X will have any emulated code in it, because of the flexibility of their XCode programming environment and because, were I in their shoes, I wouldn't want performance compromises that come from emulation, no matter how good Rosetta is.
Applications which one uses with OS X will be a different matter and I do understand about Rosetta and what it does for running program binaries compiled for the PowerPC architecture.
I don't think that the 64-bitness of the previous versions of OS X is an issue when Apple are moving the lower-end systems to Intel chips -- almost certainly 32-bit only -- first.
Watch it, or I'll get my outsourced Indian GPL violators to write an iPod-killing Flash interface to Slashdot, beta test it on a Beowolf cluster of Soviet Russians (running Linux, of course), and then get my ancient Korean e-mail users to spam you about it.
I had thought that the success of Windows Server 2003 had something to do with it needing to compete with Linux in the small server space. I understand that they had Windows NT Server, Windows 2000 Server and others available, but had only recently heard that it had been making decent inroads in this territory.
Now I realise that this doesn't make sense: Win2K is rated as a decent piece of software, so its deployment on server systems should be no surprise.
I'd really like to see a de-Balkanization of the Linux universe...
It seems Theo's started doing the bombing. Perhaps after a few years of OSX peacekeepers they will bring the flame-war perpetrators the the Hague to be tried for war-crimes.
Funny, the default mixing of apps and OS in linux distros is what I dis-like the most about linux.
I would like to have a word with the LSB and/or freedesktop.org people about that. It seems essential to desktop or home use of computers that installing function-creating programs, such as office software, messaging clients, mp3 players, art & other creative stuff that's clearly not system software, should be something that users are able to do without needing superuser privileges. I haven't seen a package management tool in a common GNU/Linux distribution which allows packages to be recognised and identified to a single user.
Perhaps this will ignite an argument among Unix people versus non-Unix people (it would really be between multi-user systems people and single-user systems people) over whether all users should gain the functionality of a package installed by one particular user... Perhaps the argument needs to be ignored and a solution found so that installing software as a home user is as easy as installing an.msi in Windows*.
(* I know that you can double-click an rpm in GNOME under FC3 and have it install itself happily, but this doesn't quite play nicely with the yum/apt/etc. tools for package management and system updates. These tools are needed to make sure that installed software remains up-to-date but they don't appear complete to me.)
Could you imagine where we'd be if all that brain work had been done on top of ancient UNIX? Wow and goosebumps time, for me. [linebreak-as-if-pause] Microsoft probably wouldn't exist any more.
I would disagree. The development of the user interface stuff like KDE and GNOME and the addition of the desktop-friendly hardware stuff (hot-plugging USB and firewire, the like) would still have had the slow progress it had in the Linux development. It's also reasonable to claim that sense Windows Server is Microsoft's response to deployments of Linux-based server systems and given that people here are partisan toward the underdog and the free software they (may be/are) involved in creating, MS is still behind. So from both a home-/desktop-use and from a server-use perspective I disagree with your view.
However, I get what you mean about the development of mass-use free software system along the lines of the OpenBSD paradigm (of clean, reliable and secure software), free from the legacy of rpm hell and the rest of that teething-trouble jazz: that would be something special.
Who will use the software, and how will them having the source benefit us?
Good point. Well made.
he should be fired immediately, and replaced by someone whose work is going to benefit the employer who's paying for it
May I slap you with "diversity is good; monoculture bad"? I would have said that having employees capable of independent thought and also capable of asking questions and leading the company to potential new markets and products is a good thing.
The problem with convincing PHBs about revenues from an existing project is one of confidence and reasonable hand-waving. Everyone else here is saying that the project will have to convince the VP that it will bring in revenue and be a profit maker for the company. The key issue is how this is presented to the VP. Speak calmly and boldly, be sure of what you are trying to ask for and what you want to get out of the discussion.
The submitter will have to say that the R&D costs are recouped internally, which helps convince the VP that the project is profitable whatever the income generated. However, this will mean that letting the code go at no monetary cost will give the advantages of the software to competitors.
Then we ask: who will use the software and how will having the source code help them? If the people who end up having the stuff only want turn-key solutions, then the source is of no merit to them. Selling these people the source code will be a business liability: they won't want to pay for the solution and support were they to have the skills to make their own from your sources. If the people who would buy the software would definitely tinker with it, then licensing the source code also may make additional money for the company.
The next decision will be where to put the price on the price-volume curve. The software, released with its source code under something like the MPL 1.0 licence (I believe that's the one where you have to contribute your changes back to the originators if you make any) which could then be sold for an accessible and small amount of money. If you're hoping to help out developing economies, then a low price is essential, and you retain some control of the software.
However, I would suggest that a turn-key edition of your project should be sold for $10k+, incorporating a support contract at a few thousand per year in fees. The source code can be licensed for a negotiable fee according to how rich the people using it are, at low cost for small enterprises and larger cost for the big boys, with the requirement that all source code licences require them to report any changes made. This will bring about associated costs of having a discussion forum, mailing list and online code repository, but perhaps you can do this with existing servers.
Perhaps you might consider limiting people to either the turn-key solution or the source code edition will allow you to make money from both people who want convenience and those who are capable of getting on with their own thing. Market the supported version as them outsourcing the maintenance (be careful about these words!) for the software which they could make themselves from the source code.
Slashdot Mirror
From the GP: ARM just isnt aiming for the same market
I don't think he missed ARM/X-Scale.
I don't know, who says you can't convey sarcasm in written form?
Oh! That wasn't a joke.
The G5 is competitive in that it matchesor outperforms the AMD Opteron (that frontside bus helps)
BerryTruFax: The HyperTransport bus used in the Opteron is the same thing used in the G5 Macs and XServes. Both AMD and Apple are in the HyperTransport Consortium.
g5 is to g4 as p4 was to p3 better overall IPC, less picky about memory latency, less power, basically a great thing to quad core if you're looking for perf/watt
I refuse to believe that the 28- and 31-stage Pentium 4 pipelines are a better thing than the 10-stage pipeline in the Pentium III, particularly when we're talking about IPC. Do you remember the fuss made about P4 being slower at the same clock speed than the PIII? That's proof it has worse IPC rate.
Neither the P4 or the G5 are lower-power than their predecessors and they fail to provide better performance/watt, in any configuration. This is why the P3 architecture has been adopted into the Pentium M line for low power use and the G4 processors remain the chips used at the core of Apple's iBooks and PowerBooks.
The great thing to do with the Pentium 4 architecture would be to put in on good Strained Silicon and SoI processes to push it above the 4.0GHz clockrate at which it is believed to be a very strong chip.
The differences between the G4 and G5 chips are what happens when you move from a desktop computer chip to a cut-down Big Iron chip (IBM's POWER4, IIRC). The G5's are inherently 64-bit capable in a way that the first three generations (Willamette, Northwood and Prescott) of the Pentium 4 are not, although there exist Prescott-based Pentium 4 processors with Intel's EM64T implementation.
BTW: http://arstechnica.com/ is your friend. Hannibal has done a good job of talking through the history of the Pentium chip family (1 & 2) and the PowerPC family (1 & 2, part 3 hasn't yet arrived) up to the G4's. There's discussion of the IBM POWER5 architecture too, and some commentary on pipelines in processor design (1 & 2). I learned a lot from these, and value their information. But I'm going to stop telling Granny to suck eggs now.
Quite bizarre. Am I right in thinking that you were visiting this seedy side of the Interweb using Internet Explorer? I'm surprised that you don't have something by the MoFo on your computer to put a gap between the OS and your web browsing for safety's sake (I won't pretend that this will make you immune to catching the STIs of the Internet, but it will help).
I think you underestimate the importance of not using Internet Explorer. Microsoft got out of an Anti-Trust suit by tying the core of the computer to IE, but in doing so made IE insecure as Woody Allen having a paranoid attack while all his friends party in the room next door.
the (completely uncalled for) war in Iraq has given extremists a made-to-order training camp
And also, the extremists behind the military-industrial complex in Western nations reasons to keep making and selling weaponry -- to both sides.
"Google no longer following 'don't be evil' mantra" headlines?
Sir, I disagree with you. It matters not whether the product is amazing (and I have quite a lot of respect for the software writers at nVidia), but it's not Libre Open Source if we don't have the code. So I will not spend my %currency_units% on that stuff.
I read AMD's press release and think that the attitude of Intel is anticompetitive, but the actual transgressions did not seem wholly worthy of anti-trust litigation. I wouldn't go so far as to say you lines of "if you can't compete, legislate", but I doubt AMD will have success with this.
Tha would be a shame, because being able to buy a notebook computer from Dell with a Turion in it and without the Microsoft Tax would make a nice political message.
Post. Of. The. Day.
I, for one, will welcome our new Control Freak Overlord, but only when your domination plans have finished compiling. :P
Ah. Sorry Mr Bobo. (Or is it Bobobo? Or Bo? Or Miss/Mrs/Ms?) Thanks for being a sport, sport.
:-P
BTW: "your" not 1337 anyway.
First, I don't click adverts. Ever. So I haven't missed them using Adblock, FlashBlock or Firefox's PopUp Blocker. And this means that noone has lost any click-through revenue from me. I will tolerate arguments that the page-impression revenue is impacted, but counter them with the fact that simple page-impression revenue hasn't been much since 1999.
Second, I'm saving myself bandwidth by not downloading unnecessary information from the sites I browse. And saving the Ad companies or the site hosts from paying for unnecessary bandwidth.
Third, DoubleClick want to monitor market trends, in so doing they would rid of the anonymity of my browsing. This would enable them to pitch at me exactly what the PR people want me to hear. This impinges the objectivity of purchasing decisions and intrudes on my Internet experience. If I want to go shopping, then I can go and look at products I want in an online store.
However, the integration of advertorial into what used to be enthusiasts' information sites, as has been criticised among the Tech sites in the last few months, is a greater concern to my shopping habits because unbiased information from research informs my shopping habits. I welcome a prejudice-free online catalogue of stuff to buy, like Froogle, but would like more information from such a catlogue than is presently supplied by Froogle and would miss impartiality from the enthusiast sites should they become advertorial.
I'm surprised you haven't been moderated down, laughed at or chased out of town by the fanboys...
Apple's Power Mac page speaks of four thermal zones and fans. I think that the present G5's have nine fans inside the box (source; highlighted by Google). It's a weird world of DRM that the cooling fan controls which operating system used...
I see. We have a disagreement about what constitutes Mac OS X. Perhaps we're also disagreeing over how lazy Apple are, as in how much of their software they will simply emulate -- the things they can't be bothered to fix.
I read that Apple has had in-house x86 editions for a few years, and I don't believe that their Intel-processor edition of OS X will have any emulated code in it, because of the flexibility of their XCode programming environment and because, were I in their shoes, I wouldn't want performance compromises that come from emulation, no matter how good Rosetta is.
Applications which one uses with OS X will be a different matter and I do understand about Rosetta and what it does for running program binaries compiled for the PowerPC architecture.
I don't think that the 64-bitness of the previous versions of OS X is an issue when Apple are moving the lower-end systems to Intel chips -- almost certainly 32-bit only -- first.
OS X will be running, in part, in emulation on the Wintel
I doubt that the last few years' effort to have OSX run on Intel processors mean that anything in Apple's operating system is left emulated.
Watch it, or I'll get my outsourced Indian GPL violators to write an iPod-killing Flash interface to Slashdot, beta test it on a Beowolf cluster of Soviet Russians (running Linux, of course), and then get my ancient Korean e-mail users to spam you about it.
I had thought that the success of Windows Server 2003 had something to do with it needing to compete with Linux in the small server space. I understand that they had Windows NT Server, Windows 2000 Server and others available, but had only recently heard that it had been making decent inroads in this territory.
Now I realise that this doesn't make sense: Win2K is rated as a decent piece of software, so its deployment on server systems should be no surprise.
Duh. Sorry.
I'd really like to see a de-Balkanization of the Linux universe...
It seems Theo's started doing the bombing. Perhaps after a few years of OSX peacekeepers they will bring the flame-war perpetrators the the Hague to be tried for war-crimes.
It's just a publicity stunt for the upcoming troll.slashdot.org ;-)
Funny, the default mixing of apps and OS in linux distros is what I dis-like the most about linux.
.msi in Windows*.
I would like to have a word with the LSB and/or freedesktop.org people about that. It seems essential to desktop or home use of computers that installing function-creating programs, such as office software, messaging clients, mp3 players, art & other creative stuff that's clearly not system software, should be something that users are able to do without needing superuser privileges. I haven't seen a package management tool in a common GNU/Linux distribution which allows packages to be recognised and identified to a single user.
Perhaps this will ignite an argument among Unix people versus non-Unix people (it would really be between multi-user systems people and single-user systems people) over whether all users should gain the functionality of a package installed by one particular user... Perhaps the argument needs to be ignored and a solution found so that installing software as a home user is as easy as installing an
(* I know that you can double-click an rpm in GNOME under FC3 and have it install itself happily, but this doesn't quite play nicely with the yum/apt/etc. tools for package management and system updates. These tools are needed to make sure that installed software remains up-to-date but they don't appear complete to me.)
Could you imagine where we'd be if all that brain work had been done on top of ancient UNIX? Wow and goosebumps time, for me. [linebreak-as-if-pause] Microsoft probably wouldn't exist any more.
I would disagree. The development of the user interface stuff like KDE and GNOME and the addition of the desktop-friendly hardware stuff (hot-plugging USB and firewire, the like) would still have had the slow progress it had in the Linux development. It's also reasonable to claim that sense Windows Server is Microsoft's response to deployments of Linux-based server systems and given that people here are partisan toward the underdog and the free software they (may be/are) involved in creating, MS is still behind. So from both a home-/desktop-use and from a server-use perspective I disagree with your view.
However, I get what you mean about the development of mass-use free software system along the lines of the OpenBSD paradigm (of clean, reliable and secure software), free from the legacy of rpm hell and the rest of that teething-trouble jazz: that would be something special.
Who will use the software, and how will them having the source benefit us?
:P
Good point. Well made.
he should be fired immediately, and replaced by someone whose work is going to benefit the employer who's paying for it
May I slap you with "diversity is good; monoculture bad"? I would have said that having employees capable of independent thought and also capable of asking questions and leading the company to potential new markets and products is a good thing.
P.S. Remind me never to work for you
The problem with convincing PHBs about revenues from an existing project is one of confidence and reasonable hand-waving. Everyone else here is saying that the project will have to convince the VP that it will bring in revenue and be a profit maker for the company. The key issue is how this is presented to the VP. Speak calmly and boldly, be sure of what you are trying to ask for and what you want to get out of the discussion.
The submitter will have to say that the R&D costs are recouped internally, which helps convince the VP that the project is profitable whatever the income generated. However, this will mean that letting the code go at no monetary cost will give the advantages of the software to competitors.
Then we ask: who will use the software and how will having the source code help them? If the people who end up having the stuff only want turn-key solutions, then the source is of no merit to them. Selling these people the source code will be a business liability: they won't want to pay for the solution and support were they to have the skills to make their own from your sources. If the people who would buy the software would definitely tinker with it, then licensing the source code also may make additional money for the company.
The next decision will be where to put the price on the price-volume curve. The software, released with its source code under something like the MPL 1.0 licence (I believe that's the one where you have to contribute your changes back to the originators if you make any) which could then be sold for an accessible and small amount of money. If you're hoping to help out developing economies, then a low price is essential, and you retain some control of the software.
However, I would suggest that a turn-key edition of your project should be sold for $10k+, incorporating a support contract at a few thousand per year in fees. The source code can be licensed for a negotiable fee according to how rich the people using it are, at low cost for small enterprises and larger cost for the big boys, with the requirement that all source code licences require them to report any changes made. This will bring about associated costs of having a discussion forum, mailing list and online code repository, but perhaps you can do this with existing servers.
Perhaps you might consider limiting people to either the turn-key solution or the source code edition will allow you to make money from both people who want convenience and those who are capable of getting on with their own thing. Market the supported version as them outsourcing the maintenance (be careful about these words!) for the software which they could make themselves from the source code.
In Soviet Russia, the processors acquire you!