APPLE MAC PRO or The Allenstein "Quadzilla" Machin

[nikki-k]

1. What you are posting about, running OSX on PC hardware, is illegal according to the Mac OSX EULA. Posting about illegal activities here is against the DUC Terms of Use.
2. OSX was designed to be used on MAC hardware. Nikki-k is right, any results you get using the hacked/illegal version are suspect, as you are not running that software as it was intended to be run. If your results came from a comparable MAC running windows, they might be valid, but until then, it's just bunk. It's not disbelief, it's just not a valid test.

TY Spk.

Dude- Diginerd- go place a hard piece of something unpleasureable in an uncomfy place. Is that childish enough for you?

Just so ALL OF THE DUC KNOWS...I am done here. It is BS little twits like this that take things beyond what they should have been. I said that the implications of what you presented were untrue. I never said you lied about what you had happen on YOUR MACHINE. We had a similar situation here back some time ago, and I am sure most will remember it. In fact, I believe it was 400 Dverbs then too, wasnt it? I recall that situation being what actually happened for the person who reported the results, BUT! Eventually it was found that the implications of what the results imparted were...drum rool please...UNTRUE. Me, Shane, Matt...bunch of us doubted the results were possible. None of us called the guy a liar...but, as I said, the end result was that the claimed result- that his particular config would yield a Dverb test of 400 instances, while OS X would only do 220- was invalid.

So, I am saying that while your test might have produced results that you perceive as providing a conclusion that, "Dual CPU system with 2.4G quad cores and 2G RAM will provide a PT system capable of 400 Dverbs." The FACT is, you have a system that produces this result. Until more unique duplicate (or near duplicate) systems with the same CPU's provide similar data, it is a singularity. A very cool one, a very promising one, and definitely a VERY bright star on the horizon. In addition to that, you also added that the same test on the same machine but with OS X only capable of 220 instances. This is a HUGE discrepency, and as such would lead any person utlizing reasoning involving past experiences and reports to seriously question what is wrong...because the implication is very simple: something is wrong with the OS X side at the very least. You never stated "I am using OS X on a non-Apple machine" in that post. That is misleading to someone visiting the thread for the intention of discovering what was said- which was about Mac Pro vs Quadzilla.

Take offense all you want. I am just sick of people playing this OS vs OS game. Evenif you did not intend it to be such, you made it such by including no other info. You simply bragged about your new CPU's, and then casually added a single sentence that condemned OS X...no disclaimer, no "I got this with OS X, and that does not sound right- anyone with a Mac Pro try these quad cores yet so we can see how they compare?" Why could you not simply present the XP exoeriences you had, and be done with it? It doesnt take much thought to realize, DUH! the quad core powered XP machines you can buy NOW are going to outperform the Mac Pro's RIGHT NOW- the Mac Pro's are still dual core. THAT is my problem with your post. And my initial response was a very polite one. I never said you lied, I never implied it- I simply said the facts presented were untrue. I guess I should have explained what that meant, as with a bit of laziness, it could be construed simply as saying, "Oh, dude, you are such a friggin liar!"

You guys like this on the DUC can have fun. I am so sick of the BS. I have stopped thinking before i post, and I am insulting people, and even with the little constraint I have left, I am still unable to post without making some comment to tick off someone else. "I saw a dog flying today!" becomes, "Dude, why are you getting a bird? Dogs can fly, and they are more fun." Nevermind it was some dimwit that glued wings to a dog and catapulted it across the line of vision of the first idiot in that scenario...

To surmise, yes, I believe you got 400 Dverbs going. Yes, I believe you ran (an illegal, altered copy of) OS X on the machine, and only got 220 Dverbs. But I am making one of my finalposts here one that makes sure some dingbat does not see that and decide that it is law, fact, and then base a (possibly stupid) decision to build their own machine instead of going with a Mac Pro, simply becaue they might feel they are getting ripped off, comparitively, for the same hardware. IMO, I made only one mistake: I should have used "invalid" instead of "untrue." But, then I might have been accused of calling people cripples...

Shanes gone, and now me. Luckily, you guys still have SPK and a host of others. See ya!

[Diginerd]

1. Nope, you're actually quite wrong on that point. Though Apple would love for you to be right...

If you actually READ the EULA as opposed to listening to hearsay it actually says must run on "Apple Labeled Hardware". My Computer has an Apple Label on it, which by legal definition (in the abscence of a more specific definition in the EULA) makes it "Apple Labelled".

So no, I'm not in breach.

Annoying for them, yes, but legally accurate yes. Do they care? Maybe, but again my machine is Apple Labelled, but at the same time I am not trying to "pass it off" as being made by them.

2. Again you are close, but not precisely correct. OS X is based on NeXt which is in turn based on BSD which in turn is designed to run on many platforms. The two issues are drivers, and Kernel. The Darwin Kernel was released under an OpenSource(ish) licence and is free to be tinkered with. 3rd Party drivers are just that. Again no harm no foul. Modern Mac Hardware is basically "PC" hardware, the motherboards are Intel 5000 based, the use standard interconnects, standard graphics, standard CPUs, infact you can even boot windows on therm. I suppose that makes them a PC too.

Hmm wait a minute.. bah, so where were we?

Comparing OSX performance to XP performance on the same hardware.

As I said, I'm happy digging up a Mac Pro to stick the CPUs in but there you have the other issue:-Bootcamp isn't fully developed and your XP experience is likely not to be not up to snuff. It was not designed to run on EFI machines, in the same way that the kernal of OSX isn't really designed to run on BIOS machines.. On the other hand the OPEN SOURCE Darwin kernal runs just fine, and that is what underpins my system. 99.9% of the code came from Apple, the rest came from developers working with the OPEN SOURCE code.

Let's get out of the childsh mudslinging and do some verifiable performance tests, now how fun would that be?

[Diginerd]

Bah. Our posts crossed. Oh well. When open testing and discussion is offered and yet declined I give up, nothing else I can do.

I apologies that my system defies your staus quo, but it is what it is.

It works quite nicely under XP, and does well under Vista too if anyone wants to build one.

Parts list above.

Dropping Clovertowns into a Mac Pro has been reported on the likes of Anandtech some time ago. They are now publically available. The performance of them in a current Mac Pro should be as good if not significantly better than what I have reported. Beware though, the bus speed is slower unless you go for the wickedly expensive parts.

Finally, I'm done too. Time to actually go back to making music.

[nikki-k]

1. Nope, you're actually quite wrong on that point. Though Apple would love for you to be right...

If you actually READ the EULA as opposed to listening to hearsay it actually says must run on "Apple Labeled Hardware". My Computer has an Apple Label on it, which by legal definition (in the abscence of a more specific definition in the EULA) makes it "Apple Labelled".

So no, I'm not in breach.

No, you are intentionally using a perceived "loophole" to do something that is, in fact, illegal. I am sure if you ended up in court against Apple's lawyers, you would be either paying your lawyers ALOT of money (more than the Mac Por would have cost), or theirs.

Annoying for them, yes, but legally accurate yes. Do they care? Maybe, but again my machine is Apple Labelled, but at the same time I am not trying to "pass it off" as being made by them.

Again, you are interpretting law. Or, you are doing as many on the internet do and simply relying on regurgitated info that some wiseass felt was a "loophole."

2. Again you are close, but not precisely correct. OS X is based on NeXt which is in turn based on BSD which in turn is designed to run on many platforms. The two issues are drivers, and Kernal. The Darwin Kernal was released under an OpenSource(ish) licence and is free to be tinkered with. 3rd Party drivers are just that. Again no harm no foul. Modern Mac Hardware is basically "PC" hardware, the motherboards are Intel 5000 based, the use standard interconnects, standard graphics, standard CPUs, infact you can even boot windows on therm. I suppose that makes them a PC too.

Dude, I know all that crap...and any who-ha with an extra 10 minutes can find what you just wrote, verbatim, on dozens of sites. Maybe more. You are circumventing security/protection protocols to run that OS..not the code it is based on, but rather the OS as a whole. Alterations are needed to do so. Hence, it is not a "true" or valid version of OS X that is open to flaws not inherent in the OS X running on real Apple computers. Therefore, any testing on said "Faux S" is just that...a "Faux Test."

Hmm wait a minute.. bah, so where were we?

Comparing OSX performance to XP performance on the same hardware.

You really dont get it, do you?

As I said, I'm happy digging up a Mac Pro to stick the CPUs in but there you have the other issue:-Bootcamp isn't fully developed and your XP experience is likely not to be not up to snuff. It was not designed to run on EFI machines, in the same way that the kernal of OSX isn't really designed to run on BIOS machines.. On the other hand the OPEN SOURCE Darwin kernal runs just fine, and that is what underpins my system. 99.9% of the code came from Apple, the rest came from developers working with the OPEN SOURCE code.

Let's get out of the childsh mudslinging and do some verifiable performance tests, now how fun would that be?

...And then you provide the very thing I said made YOUR test invalid. Done on a Mac Pro, with both OS's, even thought it is legal- it is not valid. And thus, I do not ever provide info in a thread comparing two machines that could mislead someone into thinking one is truly better...

I could give a ship less what the results are...I DO give a ship how they are conducted. This is a thread that I posted in, and presented a point of view. You have fun running your illegal OS X, hacked OS that it is. Enjoy! I have not ever presented any tests comparing OS X on Mac Pro with PT vs a Mac Pro with XP via bootcamp and PT for the simple reason that it is not a valid comparison if one is researching performance for a top performer. Wanna see what XP does on the Mac Pro via Bootcamp? Cool! Have fun! It runs well, and seems to be smooth and stable. Wanna compare how hacked OS X on an XP machine runs vs XP on a Mac Pro via bootcamp? Cool! Valid, possibly interesting.

When the Mac Pro is available with quad core CPU's, then the tests that come of it will be valid. And after several tests are done on different machines, a pretty accurate range of expected performance could be derived. Stick quad cores in the current Mac Pro, and you have a chance of valid tests. But, there is also a chance the chipset firmware/etc on the current Mac Pro's might be incapable of producing the full potential on the quad cores. Based on testing, they should be fine though. And if results prodced were inline with the expected performance curve, it would go one step toward providing a definitive bit of reliability in performance.

All I care about is that anyone with any sense hitting this thread will now have reason to question their decision, and not make a rash one simply because you got 400 dverbs and OS X only got 220. And the ones that dont question...well, natural selection does have it's palce, doesn't it?

[nikki-k]

The performance of them in a current Mac Pro should be as good if not significantly better than what I have reported. Beware though, the bus speed is slower unless you go for the wickedly expensive parts.

Finally, I'm done too. Time to actually go back to making music.

Buss speed is the same if you grab a 2.66 or 3.0 current Mac Pro. However, due to memory architecture, adding two more sticks to the lower riser is highly recommended.

And Diginerd- I am really sorry you got caught in the middle of a week or two of me being an ass to everyone who posted anything I found offensive. If you had simply put a disclaimer with the OS X part, or omitted it entirely, I would have smiled and been happy to see the quad cores are actually proving to be worth it. There was concern that Digi might "clip" us, rendering LE even more limited. IOW, they saw the dual dual cores, and said- uh-oh, time to make something to keep quad cores form being a viable option... I am suprised at the 400 dverb result. I expected about 25-30 less. Especially with only 2G ram. But, like I said- I do not doubt what you have experienced and found with your machine. (BTW- are you using M-Powered? sorry if you posted that in this thread and I angrily passed it over...)

BTW- if anyone wants to have some fun, hit the Dell site, go for a medium business machine, select a Precision 290, and then deck it out with two 2.66G quad cores and 4G ram....I got a quote of ~$6500! LOL!!!

And finally- Diginerd- did you try any of the dual dual core machines with the Woodcrst Xeons? Mac Pro? I am veyr interested in the cooling requirements, and how quiet one can get with regular fans, or maybe a pair of Zalmans. Mac Pro is wicked quiet, and I would expect any XP machine using the same CPU's to be capable fo being equally quiet...just wondering how much more heat the quad cores put off, and thus- how much louder due to extra cooling?

Sorry again- it wasnt anything personal, just me being ticked off when someone seemed to be bashing OS X, or making it less than it is.

[Diginerd]

To sum up this thread:

I think fundamentally we actually agree on at least a few things, the first being question everything..

To be clear my machine was built for xp, and it kicks butt in that mode, but unique it is, and unique it will always be though others may well build similar machines.

Running OS X on THIS machine was an experiment, I certainly wouldn't choose to run with something as unsupported as the configuration I have in a "Production" environment. I also agree that 220 dverbs under OS X on THIS machine is an unanticipated result, and the obvious culprit is the way the OS is setup, but that may not be the case as it behaves as anticipated under the likes of the davec test in both modes. Further investigation is warranted. Wether or not the results of that should be published is a different matter. I personally like information to be out there, but other may wish to suppress it.

re: Perceived or actual Loop hole / inaccurate wording on Apples part, it is until they change it I believe that it is the case that I'm LEGAL. I had a discussion with our very expensive attorney at work about this point for some time. I'm comfortable with it, but I strongly suggest that anyone else looking at this gets their own advice, I am NOT saying that someone else is guaranteed to safe from legal action using this arguement. If the Apple thought police come and string you up don't blame me..

Right at the start of the thread was "which is faster, A OR b", to which my point is neither..C.

I do say I'm disappointed at the tempers being lost around here, although I am still open to further testing under whatever OS people want to play with. This is NOT a platform (OS) flame war, it's a discussion about a unique machine's behavior under two OSes. If someone else builds something similar then we should test against each other, not blindly state that data is incorrect.

Final line which I agree with again, Darwinism is real. If enough people start buying clovertown machines then they will be king. For a while at least. Then the inevitable will happen and something else will come along. Spend your $$ wisely, knowing that what they buy will shortly be worthless.

Peace.

[Diginerd]

We cross posted again.. All cool. Gimmie a sec and I'll answer all your questions..

[Diginerd]

Hi,

Righty ho.

First off, I meant clovertown bus speed. Until you hit the 2.0 ghz parts they're 1066MHz FSB. Accross the board the clovertowns are slower per core than the Woodcrests, it's when you stack the cores to handle many threads as you get in PT the magic happens... check out my screen grab. PT had 30 threads running.. Same under XP.. Should hope so too.. ;-)

Agreed heartily on Memory configuration, dual channel interleaved requires four MATCHED sticks of RAM for optimum performance, this is true on My PC too, it's a feature of the Intel 5000X, and xt chipsests. The Quads are in real danger of starving for RAM access inf you're not setup right, you have 4 cpus competing where before there were 2 (Cores) in Woodcrest.

"I am really sorry you got caught in the middle of a week or two of me being an ass to everyone who posted anything I found offensive."

No problem, accpeted, Sorry you caught me after 6 hours craving a cigarette.. Cranky city this side. Sorry again.

" If you had simply put a disclaimer with the OS X part, or omitted it entirely, I would have smiled and been happy to see the quad cores are actually proving to be worth it."

My bad, oopsy I didn't mean to come off as an OSX slur (I love OSX, been a Mac user for 17 years now too)

"There was concern that Digi might "clip" us, rendering LE even more limited. IOW, they saw the dual dual cores, and said- uh-oh, time to make something to keep quad cores form being a viable option..."

Definately doesn't look like that has happened, would be hard for the Genie to go back in the bottle now eiter without a massive backlash...


"I am suprised at the 400 dverb result. I expected about 25-30 less. Especially with only 2G ram."

I was pretty chuffed too! My ram is properly configured which as we've both pointed out is key to improved performance, typically between 10-15%, which would account for your estimated D-Verb count over 400(!) instances.

"But, like I said- I do not doubt what you have experienced and found with your machine. (BTW- are you using M-Powered? sorry if you posted that in this thread and I angrily passed it over...)"

I am using M=Powered 7.3.1 with a Profire Lightbridge (I don't think I mentioned the interface)

"BTW- if anyone wants to have some fun, hit the Dell site, go for a medium business machine, select a Precision 290, and then deck it out with two 2.66G quad cores and 4G ram....I got a quote of ~$6500! LOL!!! "

Those CPUs are wicked expensive, but Dell is smoking crack. Without my CPUS or drives my parts list was about $1150, but 2.66GHz CPUS list for $1250 each!, the 2.33s which are closer to mine, but with a faster bus are $893 each. The 1.86Ghz parts are $495 each.. None of this is cheap, but still less than Dell. For others reading, unless you know what you are running is highly multi-threaded Clovertowns are not for you. Go Core 2 Duo right now, or AMD later on for rocking single / dual threaded performance that will exceed the multiway cpus. If PT is your primary app then MY preliminary conclusion from MY ONE data pont is that multicores are really promising and there is lots of milage in the current intel roadmap.


"And finally- Diginerd- did you try any of the dual dual core machines with the Woodcrst Xeons? Mac Pro? I am veyr interested in the cooling requirements, and how quiet one can get with regular fans, or maybe a pair of Zalmans. "

We hav a boatload of all kinds of maches at work which we play with daily. The Mac Pros are pretty much the queitest things on the street for their power. My machine is pretty quiet, but not as quiet. I have a passive cooled graphics card, the Dynatech CPU coolers I use (Not stock) are prety quiet to at normal temperatures, and not too bad when things get a little toasty under load. The rear fan in my case is the loudest component, but I haven't seen fit to do anthing about it as the whole comp sits in a machine room.


"Mac Pro is wicked quiet, and I would expect any XP machine using the same CPU's to be capable fo being equally quiet...just wondering how much more heat the quad cores put off, and thus- how much louder due to extra cooling?"

Certainly more heat than the woodcrests, close to 2x (As they are essentially two woodcrests in a can), but with good cooling you should be fine. I couldn't find a zalman LGA 771 cooler on Newegg, which is why I wound up with the dynatecs, but agin it's still reasonably quiet. A far cry from my old MDD 1.25 Dual Wind tunnel, or my old P4 3.06GHz!

"Sorry again- it wasnt anything personal, just me being ticked off when someone seemed to be bashing OS X, or making it less than it is. "

Hell no! No OS X Bashing from this side, sorry it came accross that way. I think it was from an initial misunderstanding. I know we're in the windows section of the board, but with the lines becoming increasingly blurred between "Macs" and "PC"s in terms of hardware it starts to come (as you said) down to what pictures you want to stare at all day.

Given that we're tools driven by ProTools, not Platform Driven by OS now there looks to be pretty even parity between hardware life is better than it ever has been with a great and wider selection than ever before. Now where's my ADC to make my life perfect? Digidesign can you hear us?! ;-)

Cheers mate, it has been an interesting evening of discussion.