Is Democratic Reality the Key to Truth?

[dduardo]

This is very exciting news. When OX X 10.5 Leopard comes out I'm definitely going to get a copy. Woot Woot !

There is no way Microsoft can compete. Apple has dehorned Longhorn.

 

[mattmns]

Sounds like great news, I may have to grab a copy too.

 

[dduardo]

Yeah, only if Apple doesn't cripple OS X to only work on their hardware.

[edit] It looks like they are going to restrict OSX to Mac only:

"After Jobs' presentation, Apple Senior Vice President Phil Schiller addressed the issue of running Windows on Macs, saying there are no plans to sell or support Windows on an Intel-based Mac. "That doesn't preclude someone from running it on a Mac. They probably will," he said. "We won't do anything to preclude that."

However, Schiller said the company does not plan to let people run Mac OS X on other computer makers' hardware. "We will not allow running Mac OS X on anything other than an Apple Mac," he said. "

 

[franznietzsche]

Yeah, only if Apple doesn't cripple OS X to only work on their hardware.

[edit] It looks like they are going to restrict OSX to Mac only:

"After Jobs' presentation, Apple Senior Vice President Phil Schiller addressed the issue of running Windows on Macs, saying there are no plans to sell or support Windows on an Intel-based Mac. "That doesn't preclude someone from running it on a Mac. They probably will," he said. "We won't do anything to preclude that."

However, Schiller said the company does not plan to let people run Mac OS X on other computer makers' hardware. "We will not allow running Mac OS X on anything other than an Apple Mac," he said. "

Sadness. Looks like they just lost a fair share of people who might have otherwise bought it (me included)

 

[graphic7]

I just despise this whole movement to x86 in general. What I would've like to have seen is Apple producing both x86 and PowerPC workstations. The more architectures that are available, the more interesting choices a consumer has - we're just locking ourselves into an obsolete piece of crap that should've been killed off years ago.

MIPS is dead. Alpha is dead. PowerPC is in the process of dying (as far as the consumer is concerned). Sparc will soon be dead (again, as far as the consumer is concerned). See a pattern?

 

[jdavel]

MIPS is dead. Alpha is dead. PowerPC is in the process of dying (as far as the consumer is concerned). Sparc will soon be dead (again, as far as the consumer is concerned). See a pattern?

Yeah, increasing compatibility. Hallelujah!

 

[graphic7]

Yeah, increasing compatibility. Hallelujah!

How is this increasing compatibility? Increasing compatibility is not killing other architectures - it is defining more standards to allow interoperability between operating systems and architectures.

You're not increasing compatibility - you're destroying the need for it.

 

[franznietzsche]

How is this increasing compatibility? Increasing compatibility is not killing other architectures - it is defining more standards to allow interoperability between operating systems and architectures.

You're not increasing compatibility - you're destroying the need for it.

Heavily agreed.

 

[jdavel]

graphic7,

For most computer users who are affected by computer incompatibility, the issue is that files created on one machine can't be shared with another machine. The most common problems are with graphics and text formatting getting screwed up. When you work in an organization that uses both x86s and macs, it's a colossal mess. I can't count the number of times a Mac guy has told me that a Word file on a mac is the same as a Word file on an x86. But (unless this has changed in the last year or so) it just isn't true.

As to your point that "compatibility...is defining more standards to allow interoperability between operating systems and architectures" all I can say is, you've had your chance, and it didn't get done. If it had, the other platforms and OSs wouldn't be at the edge of extinction.

The real world moves on; it can't wait around for standards to be defined. A network of identical, mediocre, computers that can communicate seamlessly is far more productive than a hodgepodge of outrageously great computers that can't. That's why Bill Gates is the richest person in world and Steve Jobs isn't.

 

[graphic7]

For most computer users who are affected by computer incompatibility, the issue is that files created on one machine can't be shared with another machine. The most common problems are with graphics and text formatting getting screwed up. When you work in an organization that uses both x86s and macs, it's a colossal mess. I can't count the number of times a Mac guy has told me that a Word file on a mac is the same as a Word file on an x86. But (unless this has changed in the last year or so) it just isn't true.

From an end-user perspective all of this is valid, but irrevelavant to computer science, which is what's important. Just because we can't open a Word file, let's kill a fairly decent, if not superior architecture to x86.

As to your point that "compatibility...is defining more standards to allow interoperability between operating systems and architectures" all I can say is, you've had your chance, and it didn't get done. If it had, the other platforms and OSs wouldn't be at the edge of extinction.

What computer science has come down to is end-users dictating the path of the field. Not all operating systems are meant to be used by a typical user. Take Plan 9 for example - it's a research operating system with only research in mind. It doesn't care what some end-user will think of its interface - it's there for just purely research and implementing interesting concepts. Should we get rid of it, too? (it does have an MS word to Postscript converter, by the way - so please spare it)

Point is, even if someone did come out with an operating system that was the coolest thing since slice bread, it wouldn't be accepted. Why? People don't care about new, exciting things - they're in some sort of mindset where they hear things like: "Linux, x86, Windows" all day long.

[sarcasm]If it ain't Linux or x86 it needs to be killed.[/sarcasm]

To be honest, I don't understand anything that I just quoted you on. Are you saying that interoperability is not a needed among other operating systems? That we should just kill every other OS other than Linux and Windows along with every other architecture other than x86? Is that what you're saying?

The real world moves on; it can't wait around for standards to be defined. A network of identical, mediocre, computers that can communicate seamlessly is far more productive than a hodgepodge of outrageously great computers that can't. That's why Bill Gates is the richest person in world and Steve Jobs isn't.

Again, you're acting like the end-user should dictate the path of computer science. There's more to just judging an OS or an architecture by it's capabilities to an end-user, such as someone in an office environment. If that were true, many of the technologies that are in use today would not be around. Of course, that was in the 70s when those projects were given the chance to thrive, because the mindset of today did not exist then.

It seems as if you're admitting that indeed these extinct architectures are superior to what's generally accepted. Why is this? Why are you so unwilling to adopt or support something that's radically different (possibly better) than what you're using now? It's that mindset that's around nowadays.

I suggest you read this: www.cs.bell-labs.com/who/rob/utah2000.pdf , to get a feel of what I'm saying.

Essentially, we now have one less avenue for possible alternative, since Apple has switched to x86.

 

[graphic7]

Also, while some of these research operating systems and architectures that are around (most of them are on the edge of extinction because of lack of corporate of government support) may not be the most suitable operating systems for a office environment or a home user, each of them offer something unique and special that aren't found anywhere else.

Apple will be switching from an architecture that uses OpenBoot (if you've used it before or have a clue you know what I mean) to an architecture whose "BIOS" hasn't had an upgrade since it was conceived, other than support for booting of CDROMs, USB devices, and using USB keyboard

The technologies that are being phased out are usually superior to their accepted counterparts. There's currently some sort of plague going on in the systems research field - money.

 

[jdavel]

What computer science has come down to is end-users dictating the path of the field...

Well, that's only because "computer science" isn't really science. It's a mix of science and commerce, in other words it's technology. The superior architectures you've mentioned are only physically realized at huge cost. I saw an 8" wafer of the latest generation of Sparks, and there were only about 25 chips on it. Those things are gigantic! Who's going to invest in something like that unless there's a market of "end-users"? And there just isn't. 99% of computer users don't even need the performance of a P4. This technology has just gotten way too far ahead of its market.

Physics has a related problem, not commerce exactly, but money. "The path of the field" of high energy physics is hitting a dead end, because it's gotten too expensive to investigate higher and higher energies. Not enough people care what the energy of the top quark is to have their tax dollars spent trying to find out.

My advice is get used to mediocrity. In a democracy, by definition, that's usually what you get.

 

[exequor]

I don't know but this seems like one of the most careless decisions that apple has ever-made. to me they are just concentrating on making money since dell, compaq, etc. have been selling more systems than them. To me they are forgetting what apple stands for. I was listening to the TWiT podcast and they interviewed someone from apple and he was like "oh this should have been long ago" but I still don't get it, isn't x86 going to be obsolete soon?

if this does work then it will be great because you will have both worlds.

 

[graphic7]

isn't x86 going to be obsolete soon?

It's been obsolete since it's conception. Almost everything you take for granted with the x86 architecture has been added as an afterthought - specifically, the conversion from CISC to RISC (in Intel's on perverted way, of course).

MIPS, Sparc (calling it `Spark' only verifies your lack of knowledge), PowerPC, and Alphas were designed from the ground up to use a boot PROM. I've seen a hack of the boot PROM concept on x86 servers a lot lately, usually they have some sort of services board running a PowerPC or ARM strapped to the motherboard that has it's own networking interface (and possibly a serial port). Sure, this allows you to configure the system remotely, but you still can't do installs via serial or over a network remotely. The processors I listed above have had boot PROM technology since they were conceived. My little SparcStation 5 (a real gem in it's day) had PROM technology in 92', and Sun systems had it way before that - the same with SGI, IBM, Dec, etc. I can have a customer who lives in Kentucky order a few Sun servers, and I can configure them and install Solaris remotely, while I'm sitting in Florida.

This all goes back to what I said earlier. Even if the best architecture or operating system were to arrive nobody would care. If you need anymore evidence remember that DEC is dead, SGI is in the process of dying, Apple has had to switch to x86 to survive, and Sun is still fighting. Honestly, though, I don't expect Sun to continue their UltraSparc line for eternity.

 

[jdavel]

MIPS, Sparc (calling it `Spark' only verifies your lack of knowledge), PowerPC, and Alphas were designed from the ground up to use a boot PROM. I've seen a hack of the boot PROM concept on x86 servers a lot lately, usually they have some sort of services board running a PowerPC or ARM strapped to the motherboard that has it's own networking interface (and possibly a serial port). Sure, this allows you to configure the system remotely, but you still can't do installs via serial or over a network remotely. The processors I listed above have had boot PROM technology since they were conceived. My little SparcStation 5 (a real gem in it's day) had PROM technology in 92', and Sun systems had it way before that - the same with SGI, IBM, Dec, etc. I can have a customer who lives in Kentucky order a few Sun servers, and I can configure them and install Solaris remotely, while I'm sitting in Florida.

Actually it's SPARC not "Sparc". What does that verify about you?

More to the point, the paragraph above makes my case, not yours. Why should 99% of computer users pay the price, chaos and loss of productivity in their work place, so that the 1% who understand what you were talking about can get a technology fix. It's a free country. You want to design unnecessarily high performance processors, get some work stations, hire a few hundred of your ilk, and get to work. Of course if you want actual chips to play with you'll have to get a $3B fab to build them for you, so they won't be cheap. And don't count on anybody buying them, because 99% of us don't want them. That's how it is with free enterprise.

Good luck!

 

[graphic7]

Actually it's SPARC not "Sparc". What does that verify about you?

I have no clue. Feel free to enlighten me, though.

More to the point, the paragraph above makes my case, not yours. Why should 99% of computer users pay the price, chaos and loss of productivity in their work place, so that the 1% who understand what you were talking about can get a technology fix. It's a free country. You want to design unnecessarily high performance processors, get some work stations, hire a few hundred of your ilk, and get to work. Of course if you want actual chips to play with you'll have to get a $3B fab to build them for you, so they won't be cheap. And don't count on anybody buying them, because 99% of us don't want them. That's how it is with free enterprise.

You took the words right out of my mouth. Do you not understand that the information technology industry is boxing itself in? By elliminating all these wonderful technologies and reverting back to piss-poor ones, we're not only not advancing at the rate we once were (mid 90s), but, in fact, we're retrogressing. Sure, consumers can now get their hands on a somewhat powerful platform, but they're going to keep that platform for a long time to come. Are you really willing to use an archaic, much less than perfect set of technologies for years to come just because they're cheap, while killing off what could offer future advances to systems research?

 

[faust9]

Here, Here G7! Hmmm, let's all revert to a processor who's ability to arbitrate conflicts between external devices is circa 1971. Nothing like a good old IRQ conflict when installing a new PCI/AGP/whatever card---these conflicts don't occur on modern processors BTW, just the P4(x86) dinosaurs. Ohhh we have HT. Wow guys you added a pipeline! Great job. So long RISC. Hello 1980. Get the new Apple with a black joystick featuring a big red button and a DB9 connector. Cartridges sold separately.

In the end I think Apple will still make a good machine even if it is based on a crap processor, but I won't be buying a new Apple until the conversion is complete. I was planning on getting a new G5; however, Apple has all but said "we plan on making your $2000 purchase obsolete next year!"

I guess I'll swing by the junkyard(aka any retailer of x86 processors) and pick myself up an AMD to fill the gap until the transformation to the darkside is complete.

 

[franznietzsche]

Actually it's SPARC not "Sparc". What does that verify about you?

It certainly verifies that you're a jackass.

Now, granted graphic7 is incapable of posting without being an ass and attacking anyone he disagrees with unnecessarily, but he is right. For purely economic reasons the elmination of other architectures is bad. For technical reasons its an absolute disaster.

And just so you know, those compatibility issues have nothing to do with the architectures. They have much more to do with the incompetent design of Windows that is intentionally made to be as incompatible as they can legally get away with (since the antitrust judgement forced them to make certain measures of compatibility, and now the EU is finally cracking down on their intentional incompatibility, though not enough).

 

[exequor]

and now the EU is finally cracking down on their intentional incompatibility, though not enough).

I like what the EU did, merely due to the fact that what "the company" is doing is abusive. After installing windows I uninstalled a lot of the software that came with it, in order to put on my preferred software.

 

[Anttech]

It certainly verifies that you're a jackass.

owch..
[rant]
no need for that, You are also very opinionated, and a bit abrasive
[/rant]

 

[franznietzsche]

owch..
[rant]
no need for that, You are also very opinionated, and a bit abrasive
[/rant]

Not going to contest that, but doesn't that go without saying?

 

[mewmew]

I think apple did the right thing. I don't know too much about how chips are designed or work but I work in a lab with both G5's and P4's. I do everything on a G5 because I love os x but the P4's cost half the price as our dual G5's and are faster. Not to mention no matter what apple has done it can't get a G5 into a laptop because of the power and heat. The powerpc just wasn't cutting it and I for one am glad to see it leave.

 

[graphic7]

I think apple did the right thing. I don't know too much about how chips are designed or work but I work in a lab with both G5's and P4's. I do everything on a G5 because I love os x but the P4's cost half the price as our dual G5's and are faster. Not to mention no matter what apple has done it can't get a G5 into a laptop because of the power and heat. The powerpc just wasn't cutting it and I for one am glad to see it leave.

Depends on what you're doing. Doing floating point operations with Altivec is just awesome, not to mention the G5 actually has a sqrt function - pure genius. What's a pity, though, is that all you care about is performance. Even if the G5 lacks there (it doesn't), it's still better than an x86 in every single way, other than cost. You can't expect to get a Ferrari at the cost of a Honda. So, don't expect to get a G5 at the cost of a toy - the x86. It's that simple.

 

[franznietzsche]

I think apple did the right thing. I don't know too much about how chips are designed or work but I work in a lab with both G5's and P4's. I do everything on a G5 because I love os x but the P4's cost half the price as our dual G5's and are faster.

Someone's been praying to the false god of clock speed entirely too much if you think the P4 is faster than the dual G5s. No chance can the P4 match dual G5s in terms of calculations. Of course, software design and utilization of available power will matter as much in the end, but in hardware terms, not a chance.

Not to mention no matter what apple has done it can't get a G5 into a laptop because of the power and heat. The powerpc just wasn't cutting it and I for one am glad to see it leave.

How is it not just cutting it? You don't need a 3Ghz processor for any current PC application at all, no matter what Wintel wants you to believe. Its entirely unnecessary for any current PC purpose.

 

[mewmew]

It's a physics computer lab that I am in every day running Mathematica. Look at the speed of the top of the line apples going against an intel or AMD machine using mathematica and you will see the powerpc plain isn't as fast as top of the line intels and AMD's. I realize I sounded like I was saying the G5's are slow but that's now what I really wanted to say, it is fast, but the P4's are just as fast and are not top of the line. Note, that the mathematica we have doesn't support dual processors so more or less its a 1 processor machine. I am not saying that x86 is better or anything I am just telling you how things are in my lab running a particular important(I think atleast) program.

Also yes, all I care about is performance, a better chip does me no use if I don't have any programs that can take better advantage of the chip over anything else. I love my macs and will continue to use them but as of late the powerpc has been lagging behind slightly for my purposes. http://www2.staff.fh-vorarlberg.ac.at/~ku/karl/timings50.html has some numbers to get an idea of what I mean. The G5 is by no means slow, but I still do not understand why there is any reason to keep it, the extra price for something that has no tangable benefits? Like I said though, all the stuff about how the designs are is over my head so I can't comment on what chip is actually better, I just know that for my purposes x86 is better and the fact that apples laptops are limited with the powerpc is a problem too(for me personally).

 

[faust9]

It's a physics computer lab that I am in every day running Mathematica. Look at the speed of the top of the line apples going against an intel or AMD machine using mathematica and you will see the powerpc plain isn't as fast as top of the line intels and AMD's. I realize I sounded like I was saying the G5's are slow but that's now what I really wanted to say, it is fast, but the P4's are just as fast and are not top of the line. Note, that the mathematica we have doesn't support dual processors so more or less its a 1 processor machine. I am not saying that x86 is better or anything I am just telling you how things are in my lab running a particular important(I think atleast) program.

Also yes, all I care about is performance, a better chip does me no use if I don't have any programs that can take better advantage of the chip over anything else. I love my macs and will continue to use them but as of late the powerpc has been lagging behind slightly for my purposes. http://www2.staff.fh-vorarlberg.ac.at/~ku/karl/timings50.html has some numbers to get an idea of what I mean. The G5 is by no means slow, but I still do not understand why there is any reason to keep it, the extra price for something that has no tangable benefits? Like I said though, all the stuff about how the designs are is over my head so I can't comment on what chip is actually better, I just know that for my purposes x86 is better and the fact that apples laptops are limited with the powerpc is a problem too(for me personally).

Straigh no-nonesence computations are faster on an x86. This has been shown time and again. Every other thing a computer needs to do is done faster on a Mac---ESPECIALLY if the software running on the Mac has been optomized to take advantage of the PPC and not just a recompile. I use Maple and see no difference between the windows version and the Mac version on a G5. I do see a difference between Matlab running under linux and Matlab running under windows on the SAME x86 computer. What does that tell you? It tells me a) windows is a poorly optimized OS when compared to Linux or b) Matlab itself has been optimized to take advantage of the Linux environment more so than the windows environment. In fact there are a few benchmarks showing some windows programs run faster using wine on the same hardware than they do natively under windows... What am I getting at here? Software developers who do recompiles only are doing a disservice to themselves and to their user base. Software developers need to take the time to tune their code for the different systems that the software will run on. It has little to do with the processor and much much more to do with the developers IMHO.

A single benchmark (all calculations were done using Mathematica in your reference so the various tests are all showing the same thing---mathematica's ability to perform calculations on the various combinations of HW/SW) doesn't prove anything. How does Linux PPC do in those tests? Maybe the problem isn't the PPC but rather the mach kernel. Is mathematica the problem or is it PPC? What about other mathematical programs? How did each one fair with each OS and hardware combination? What about other test outside the relm of calculations? How will transitioning PPC to x86 improve OSX performance? How will this transition improve the Mach kernal response? Oh, and what are the relative hardware costs? I'll tell you as a embedded developer that PPC processors are not really more expensive than top of the line x86 chips. When you buy an Apple you buy the name, the quality, the OS, sleek hardware. You pay for all of that---Hmmm how much does a comperable x86 PC cost compared to a G5... I bet you the cost comparison would leave you shocked and amazed especially when you find out feature for feature an Apple costs about the same as a Dell/HP/Gateway/Alien.

As for the laptops, I have NEVER been limited by the G4 inside.


My thoughts.

[edit] It appears the dual G5's still outperform the intel processor based Macs. http://www.thinksecret.com/news/0506intelxbench.html

 

[mewmew]

As for the laptops, I have NEVER been limited by the G4 inside.


My thoughts.

[edit] It appears the dual G5's still outperform the intel processor based Macs. http://www.thinksecret.com/news/0506intelxbench.html

"The benchmarks do not reflect native performance of the 3.6GHz systems, however, but rather provide an indication of how PowerPC-compiled applications will run under Rosetta on Intel-based systems."

I love my ibook, don't get me wrong, but faster is always better. The fact that apple could not get a G5 into a laptop seems like a large drawback to me. Jobs and the apple engineers know a whole lot more about what the future of the G5 held for them than I or many of us I would say and decided to ditch it, I am sure they had ample reasons to do such a major change.

You guys know a whole lot more than me about all of this stuff so I can't argue what you say is true. I just want to provide my opinion as a consumer and computer lab worker at the end result. I love macs, I own macs, I use them every day so I want what's best for them. I am just saying I think that going to x86 will help my small little mac "bubble" I live and work in. I could very well be wrong but we will just have to wait and see.

 

[jdavel]

... I bet you the cost comparison would leave you shocked and amazed especially when you find out feature for feature an Apple costs about the same as a Dell/HP/Gateway/Alien...

I'm not sure what you mean by feature for feature. But for any meaning I can think of, all I can say is, you've got to be kidding!

 

[faust9]

I'm not sure what you mean by feature for feature. But for any meaning I can think of, all I can say is, you've got to be kidding!

It's pretty self explainatory don't you think? Take a feature from one system and compare it to a feature on another system. It doesn't take a lot to do either. Let's take a system from an industry dig-whig---Dell. Now you'd think that volume would result in a significant savings, so let's take a feature for feature look:

The dell(see attachment):
http://www1.us.dell.com/content/products/features.aspx/cto_inspn600m?c=us&cs=19&l=en&s=dhs
1158 for an ugly dell.

The iBook(the middle one):
http://store.apple.com/1-800-MY-APPLE/WebObjects/AppleStore.woa/71606/wo/agfjQwMV2rC43UITNL22ZJz50QQ/0.0.11.1.0.6.23.1.1.1.0.0.0.1.0
1299 for a cool iBook.

When one looks at the features(the dell has an inferior OS BTW) one sees that the price difference is pretty small. Moreover you get excellent service and if you live near an Apple store then you can go pick the brain of the on-staff genius.

 

[Anttech]

the dell has an inferior OS BTW

opinion not factual... With the Apple you are locked into using an apple os, with a dell you can use more or less every other desktop/server OS (with a few excpetions of course)