Apple launches the PowerMac G5

[Agathon]

Originally posted by Asher

First line, Agathon. I realize it's hard to miss, what in big font, bright red, and all caps.

That isn't an ad hominem argument. That is a request. An ad hominem argument uses an attack on a person as a premise in an argument.

For example:

We shouldn't believe Asher because he's such an *******.

Previous dishonesties? Examples, please.

Oh, the time you had a go at me and several other people for not understanding Fitts' Law and then posted a link to a site which you said had it right. A site which agreed with what I said. The obvious conclusion being that either you didn't read it or you didn't understand it.

And then about a day later you did it again, by posting a link to a site that you claimed showed XP was a far better OS than X. Again, the site said just the opposite.

After that, I came to the conclusion that you'd post anything to get your way, so I stopped arguing.

I most certainly am not clueless. I know what I'm talking about here, it's the Apple dimwits in this thread who can't do anything to support a case and end up changing their arguments.

One doesn't have to be a computer genius to see what's wrong with your arguments: that is patently obvious to any observer. What's happened here is that you just bought a load of FUD because it suited you.

First we're blaming them not using dual CPUs in one of the benchmarks (duh! It didn't support it), then you ignore the fact that Apple rigged the PCs to perform spectaularly crappy

Apple responded to this today (some of what they said just repeated what's been said on various forums). One of the arguments levelled against them was that they failed to use an Intel optimized compiler in one of the tests. That's true and they admitted this. They also pointed out that they didn't use a G5 optimized compiler on the Mac. They also pointed out that they used Linux instead of Windows in the same test because it gave the PC a better showing. In fact if you had seen the webcast you would have seen that Jobs admitted that the G5 couldn't quite match the PC in certain tests, but argued that it clearly outperformed it in others and that the radical difference in these other areas and coming only slightly short in the former, showed the G5 as the superior computer.

Now if they were going to cook the books, you would think they would do it in all the tests. It's simply ridiculous, unless you are some maniacal conspiracy theorist, to argue that they would do this.

then you insist we don't know what the PC CPUs are going to be like in a few months

I think you have me confused with Cloud9; I never said that. What I said is that the only worthwhile test would be getting hold of a G5 when it comes out and then comparing it using real world applications. These speed tests strike me as a bit silly. It's like when car maniacs test engines for all sorts of obscure things - the results might be interesting, but actually driving the car is what counts.

then Agathon says he thinks the "fastest CPU" comment is silly anyway.

Again, this is something I never said. What I said was that the "fastest desktop" label is silly because it is just a semantic matter as to whether one calls a given computer a desktop or a workstation. That is really a verbal matter, so not a serious topic for debate.

You're both clueless. Not only about computers in general, but consistency and debate tactics.

Well, since I've just shown you radically misquoting me twice, I think I'll leave that comment hanging.

The G5 is the PowerPC 970. You didn't figure that one out yet?

Are you ****ing thick or something? Cloud9 has repeatedly pointed out that the official benchmarks from Apple using the system that was introduced this week are, as far as we know, the Veritest ones. Has someone else been testing the production G5 models (the actual models that Apple is selling)? If we want some other independent study using the computers that Apple is actually selling, we will have to wait. Presumably the honest thing to do would be to wait.

The official SPEC benchmarks are, for right now, dual 1.8GHz G5s/PowerPC 970s. Since 2GHz is ~10% faster, the SPEC benchmarks can only be ~10% higher, which still places it well under a P4 3.0GHz and 3.2GHz.

Get it?

Um, OK. Official SPEC benchmarks for a computer that Apple doesn't sell. I'm sure they would be interested to know that they sell a dual 1.8 GHz G5. I certainly can't find one on their site.

So your argument is that a computer that Apple doesn't sell is slower than some other computer.

[Agathon]

Originally posted by Asher

It's an option in XP. Don't like it?

Right click on the taskbar, go to properties, un-check "Group similar taskbar buttons".

If that's your big complaint about the taskbar, it is frankly hilarious.

I'm not complaining about that feature. I think it is useful. That's why OS X has a similar feature.

You complained that OS X did not have an easy way to switch between open windows, since it didn't have a list of all open windows al la the Windows taskbar.

I have never quite seen the point of this criticism as you can clearly see the Windows you have open on the Macintosh desktop (more so on the new macs which have wide screens). The way to select the one you want is to click on it. The Macintosh doesn't maximze windows to take up the whole screen, because this would prevent you from easily clicking and dragging pictures or text (or links or email addresses or whatever) to the desktop (something I'm always doing). Hence there is no problem finding the one you want.

The problem arises when you have more than about five or six windows open in various different apps. Of course, if you are a real moron and can't see properly selecting between three windows on the desktop might be a bit of a challenge. Well it isn't for me. Nor is selecting between minimized windows, since I can see what they are (I always know where Apolyton is, because I can see a thumbnail of it in the Dock).

Anyway, all you do in a case of overcrowding is select an application's dock icon and then select from the list of open windows. This is exactly what Windows XP does when it groups windows in the taskbar. The Macintosh way is better since it combines the Quicklaunch, Window selection and Tray functionality into one single icon (and no, there isn't any trouble in seeing which apps are open - you can change it to be more obvious if you want).

What neither XP or Jaguar has is a method for selecting between windows qua windows, rather than qua app, when you have a lot of open windows. Expose fixes that.

Now you might object that XP does have a way of doing this - and it does - it just isn't very useful. The taskbar, since it has the Start Menu, Quicklaunch icons, open windows and System Tray items in it, tends to fill up quite quickly. I got four windows in the other day, when I was playing with XP, before it started to group them (and thus became no better than the Dock).

Now you could object that you can increase the depth of the taskbar. My old roommate used to do this. My problems with doing this are that the taskbar immediately becomes huge and hideous (as if the Start Menu wasn't a screen hog already) and that it starts listing open windows from the top; thus leaving the prime real estate at the edges of the screen empty.

Frankly I don't see the point to having bunch of windows open.

When you have six versions of a document, like I did in the previous picture, and you are trying to collate them, it comes in useful. Same if you have a dozen pictures open in Graphic Converter.

And I'd never, ever, use an Expose feature where it zooms everything out so you can see all your windows and select one. I'm not one of those artsy think-in-pictures kinda guy, I'm more procedural than that.

This is why you like Windows and why I don't. The problem is, most people don't operate like you do. I prefer visual cues. I think this is one reason why arty people like macs, although that's debatable.

If this argument is going to boil down to this particular difference, then there really isn't an argument about which GUI is better, just which is better for which people.

So rather than thinking "where did my Apolyton window go???" and hitting some button to cascade all my windows, then finding it...I click the button named "Apolyton..."

All I do is click on the Safari tab or on the minimized window. I assure you, it's no trouble.

Now stop trying to force your "procedural" peculiarities on the rest of us.

[Asher]

Originally posted by Agathon
That isn't an ad hominem argument.

I don't know -- I think calling me thick-headed and not addressing any points is the basis of both your and his' arguments...

Oh, the time you had a go at me and several other people for not understanding Fitts' Law and then posted a link to a site which you said had it right. A site which agreed with what I said. The obvious conclusion being that either you didn't read it or you didn't understand it.

That site certainly did not agree with what you said. You still do not comprehend that Fitts' law is a mathematical equation, which is used by interface designers to determine a more functional interface. There is no such thing as "adhering" to Fitts' law, and because OS X has weird buttons and windows that default to non-maximized, Fitts' law states response times will be larger because of that.

I'm completely and utterly speechless that you cannot comprehend something so simple...

Go ask your CS grad student roommate/friend/whatever what Fitts' law is.

And then about a day later you did it again, by posting a link to a site that you claimed showed XP was a far better OS than X. Again, the site said just the opposite.

Nonsense. This is a red herring -- it was you that linked to the site, and I said it was obviously written by some Mac-user but agreed on some points. You seem to take it as black/white -- you either agree with it all or disagree, or you're lying.

After that, I came to the conclusion that you'd post anything to get your way, so I stopped arguing.

That's interesting -- both examples you gave were bull**** and the only one to blame for them are your own ignorance and apparent lack of intellectual capacity. You still refuse to understand what Fitts' law is.

Go to the UofT bookstore, pick up an interface design book, look in the index for Fitts' law and see for yourself. Fitts' law says absolutely nothing about edges of the screen, etc. -- regardless what you say it does.

One doesn't have to be a computer genius to see what's wrong with your arguments: that is patently obvious to any observer. What's happened here is that you just bought a load of FUD because it suited you.

No, one doesn't have to be a computer genius to understand the arguments. Apparently, one has to not be a philosopher though. Perhaps it's just too much thinking for you?

My arguments are based on:
1) Facts, real-world benchmarks
2) Facts, looking through Veritest's configuration disclosure to determine how the crippled the Pentium 4
3) Facts, how Apple blatantly ignored AMD's chips, even though the Opteron scores well above 1000 with the GCC compiler that Apple's only got 800 on, then they jumped around claiming the first (lie) and fastest (lie) 64-bit desktop processor.

So no, one doesn't have to be a computer genius to understand these arguments. One just has to have a brain, however small it is.

Apple responded to this today (some of what they said just repeated what's been said on various forums). One of the arguments levelled against them was that they failed to use an Intel optimized compiler in one of the tests. That's true and they admitted this. They also pointed out that they didn't use a G5 optimized compiler on the Mac.

This is another LIE. www.veritest.com, download the ~1MB Veritest disclosure with all the details you want of the test. It's 28 pages: http://www.veritest.com/clients/repo...erformance.pdf

Not only did Apple use their own modified version of GCC, they applied some hacks to the CPU registers and memory prefeteres that result in optimal SPEC performance.

They also pointed out that they used Linux instead of Windows in the same test because it gave the PC a better showing.

This is also a LIE. It's well known that only the 2.5 Linux kernel behaves properly with HyperThreading, and Apple used 2.4. Windows is the fastest on the Pentium 4 -- why do you think Intel posts Windows scores to spec.org instead of Linux?

Well, since I've just shown you radically misquoting me twice, I think I'll leave that comment hanging.

Hello, welcome to this thread -- did you notice the top of the post which said I was replying to Cloud9?

I'll leave this comment hanging as a tribute to your intellectual capacity.

Are you ****ing thick or something? Cloud9 has repeatedly pointed out that the official benchmarks from Apple using the system that was introduced this week are, as far as we know, the Veritest ones.

Apple has never submitted SPEC results. It's always the manufacturer of the CPU, IBM or Motorola, that do it. IBM has published the SPEC results for the PowerPC 970 already, you guys choose to plug your ears and ignore them.

Um, OK. Official SPEC benchmarks for a computer that Apple doesn't sell. I'm sure they would be interested to know that they sell a dual 1.8 GHz G5. I certainly can't find one on their site.

That doesn't matter. It's 1.8GHz because that was supposed to be the top end, they squeezed 2.0GHz out of it later. Still, 10% clockspeed = at most 10% performance increase. Simple math.

So your argument is that a computer that Apple doesn't sell is slower than some other computer.

My argument is simple -- the 2.0GHz cannot be more than 10% faster than a 1.8GHz. Is this math over your head, or what?

What neither XP or Jaguar has is a method for selecting between windows qua windows, rather than qua app, when you have a lot of open windows. Expose fixes that.

Now you might object that XP does have a way of doing this - and it does - it just isn't very useful. The taskbar, since it has the Start Menu, Quicklaunch icons, open windows and System Tray items in it, tends to fill up quite quickly. I got four windows in the other day, when I was playing with XP, before it started to group them (and thus became no better than the Dock).

I'm sick of drilling this into your head, so this is the final time -- pay attention.
The Taskbar shows all windows. It does not show all apps.

If you object to it grouping after only 4 windows per app, then I suggest you use the free TweakUI to increase that limit. Or you can disable it.

Either way, it is entirely incorrect to say Windows doesn't have a way to list windows as windows, because it's done that since Windows 95.

[Agathon]

I started reading until I got to your pathetic excuses for your past dishonesty. Then...

[Asher]

Just because I'm going to put and end to all of this, I'm going to quote the relevant parts of Veritest's disclosure:

Page 5, how they tweaked the Mac beforehand for optimal SPEC performance:

Installed th eTachyon development environment version 6K452. This provides the appropriate development tools for generating the SPEC binaries and installs Apple’s version of the GCC compiler ( version 3.3 build 1379 ) on the test system
...
Using the “Reggie” tool available from CHUD, modify CPU registers to enable memory Read By-pass. As Read requests are speculatively sent to the memory controller, this eliminates the need to wait for the snoop response required in a multiprocessor configuration thus reducing the time required for a read request.
...
Used the command “hwprefetch -8” to enable the maximum of eight hardware pre-fetch streams and disable software-based pre-fetching.
...
Installed a high performance, single threaded malloc library. This library implementation is geared for speed rather than memory efficiency and is single-threaded which makes it unsuitable for many uses.

That last one is particularly interesting. Veritest put it diplomatically, "unsuitable for many users" and "geared for speed". That's because that MALLOC (memory allocation) library only works with SPEC benchmarks. It's intentionally designed for them, and is not a real-world scenerio.

Page 7, how they crippled the PC:

Installed a retail, shrink wrapped version of Red Hat Linux 9.0 Professional by selecting the standard “Workstation” option during the installation process.

(read: installed the bloatiest Linux version available, use all defaults...)

Downloaded GCC version 3.3 ( gcc-3.3.tar.gz ) from http://gcc.gnu.org.

(read: downloaded the generic GCC 3.3, which is notorious for horrible Pentium 4 performance)

Installed the NAGWare Fortran 95 compiler using default options by following the instructions provided with the software.

(read: downloaded the Fortran compiler and installed with the default options, which means they're built for a Pentium II/III core.

Use the system setup utility to disable hyperthreading in the system.

(read: We disabled the HyperThreading technology that gives the P4 a ~20% performance boost for no apparent reason)

Page 25 is interesting, and explains some compiler options used for the G5. Particularly of interest:

-fast: This flag is used with C and C++ and specifically targeted to the G5 and enables G5 specific instruction usage, tuning and 64 bit arithmetic. In addition to enabling the -O3 optimization level, it also enables the use of C99 aliasing rules and relaxed IEEE math

Oh boy! This one's great -- specially tuned (optimized) G5 code thanks to IBM. What's more, is the "relaxed IEEE math". What they mean here is they're using AltiVec instead of the FPUs because it's so much faster. Unfortunately, AltiVec can only handle 32-bit numbers. So by "relaxed IEEE math", they mean "far less precise". The Pentium 4 is forced to do all their FPU calculations at 80-bit precision, and Apple's doing them at 32-bit.

-lstmalloc: Link fast malloc libraries. The implementation is geared for speed rather than memory efficiency and is single-threaded which makes it unsuitable for many uses. Special provisions are made for very small allocations (less than 4)

Again, this just enables the SPEC cheat library.

And Page 27 is interesting, showing the compiler flags they use for the Pentium 4:

-mfpmath=sse

SSE, not SSE2. SSE only does integers, SSE2 does floating point. The Pentium 4's FPU is rather weak on its own, it's designed to use SSE2 for optimal speed. Apple disabled that.

-fbranch-probabilities: After running a program compiled with –fprofile-arcs you can compile it a second time using -fbranch-probabilities, to improve optimizations based on the number of times each branch was taken. When the program compiled with -fprofile-arcs exits it saves arc execution counts to a file called sourcename.da for each source file The information in this data file is very dependent on the structure of the generated code, so you must use the same source code and the same optimization options for both

A useless diagnosis flag. It exists so if you compile it a second time, it can optimize it better, but for the first time you run it it adds additional overhead (slows it down). They only ran it once.

--------

So there you have it. Straight from the horses' ass on how they rigged the tests to show the G5 as fast. Normally people don't care much, but Apple has claimed repeatedly that their G5 is the "first" 64-bit desktop (which is a blatant lie), that their G5 is the fastest personal computer (which is a blatant lie), and they based all of this off of rigged benchmarks.

Apple has always done this -- this is just the first time some third party was contracted out to do it and obligated by law to disclose all the details.

I guarantee you Apple's doing the same things with Photoshop. Why wouldn't they? They've shown they have no integrity when it comes to benchmarks.

So there you have it. You can all suck it down now, or you can continue trying to weasel out of this and excuse it. It'll be interesting to see how you deal with all the evidence right from the people who benchmarked it...I'm betting you're all just gonna let the thread drop.

[monolith94]

Honest to god, I don't know which one of you is lamer... I'm just glad I abandoned the thread pages ago...

[Imran Siddiqui]

Yea, I'm with mono... just get the computer you like and save the **** measuring .

[Asher]

It's not about PC vs Mac, it's about Apple's pathetic and devious marketing.

It's, by far, the worst out of any technology company.

It's so disgusting that I would refuse to buy one of their machines if it was the fastest in the world...

[Cruddy]

Nah. I challenged Asher to admit the G5 was best for running Mac apps.

Of course, such an admission is beneath him.

[Asher]

I couldn't tell if you were making a bad joke or you were mentally disabled, and I figured either way it wasn't worth my time.

[monolith94]

"It's so disgusting that I would refuse to buy one of their machines if it was the fastest in the world..."

But if it was the fastest machine in the world, then wouldn't that make their marketing less disgusting???

[Asher]

No, because the past is the past. I was hoping they'd grow up and be real for once.

Apple reminds me of this girl in my jr. high that was so desperate for attention that she chronically lied. Cut out a random picture from the obituaries in the newspaper of a 29 year old man and cried, saying her boyfriend died.

[Cruddy]

Originally posted by Asher
I couldn't tell if you were making a bad joke or you were mentally disabled, and I figured either way it wasn't worth my time.

Aha! Asher refuses to admit that Apple's G5 is somehow superior to any Intel or AMD machine in at least one respect.

What's the matter Asher? Embarrassed by the fact that some people actually choose hardware because they want to run a certain software suite?

No joke. No mental disability (I've been checked by 3 different analysts and they all agree I'm flat normal).

[Asher]

Originally posted by Cruddy
What's the matter Asher? Embarrassed by the fact that some people actually choose hardware because they want to run a certain software suite?

I'm embarrassed for you right now. Truly...

No joke. No mental disability (I've been checked by 3 different analysts and they all agree I'm flat normal).

No, you certainly do have a mental disability if you thought this was profound or something. Probably all the drugs you've done.

I admit it -- the G5 is the best processor to run Mac software on. I also admit that the earth is round, Fez is gay, and streaking at a police station will get you busted.

[Cruddy]

Originally posted by Asher

I'm embarrassed for you right now. Truly...

No, you certainly do have a mental disability if you thought this was profound or something. Probably all the drugs you've done.

Actually that's why I thought I'd get checked out, and was somewhat relieved to find no permanet damage had been done.

As for being profound - those who choose cool hardware without even thinking what softwre they will run on it have their decisions arse about face. The fact is Asher, people who have invested in a large Mac suite will opt for G5 and not your beloved combinations - and that's the part that really hurts you.

So what if Apple were economical with the truth? All manufacturers do this.

Originally posted by Asher
I admit it -- the G5 is the best processor to run Mac software on. I also admit that the earth is round, Fez is gay, and streaking at a police station will get you busted.

Earth is a spheroid, not round as such. As for streaking, it depends what country you're in. Some places public nudity is OK.