The new G5's... Give it up!!!

[archiea]

OK, there's gotta be a few early adopters out there who fearlessly plopped down over 3 grand for the new G5.

You know who you are!!!!

Snap out of your "out of Box" experience!!!!

Sober up from sniffin' the factory fresh silicone in those first few hours of burn.

Don' t even waste time callin' Newtek to transfer the License key...

Just fire up the discovery edition, bee-yotch, and get those watermarked benchmark renders going.....

Why are you still reading this... get going!!!!!

Hit the fawking F9 key why don't ya!!!!

:D

Oh and thanks!

[denny]

I would think most would be getting the Dual 2 Gig, which is not yet here. But I could be wrong...

[NigelH]

Even the single proc. models will take a day or two for delivery. Give 'em time. :-)

[Beamtracer]

I think that Arthur wrote this thread just to make fun of those who are drooling over the new G5.

[ingo]

Huh, maybe he just realized that its not the computer that makes the pictures, so a G5 wouldn't help him much :)

Too much coffee can sometimes be a problem......

[Darth Mole]

Come on, people - I desperately need some justification for dropping a huge wad on my as-yet-undelivered G5.

I wanna see some serious speed increases over my old dual 800... :-)

[Johnny]

a cross-platform, Cinema-using friend at work saw some post on a message board that those who had G5s in their hands were getting 33% better performance than corresponding Pentiums.

that sounds encouraging.

J

[archiea]

Nawh, I'm not making fun of people drooling over the G5...

I just want some posts!!!!! We've waited and speculated for so long....

[Darth Mole]

33% over corresponding Pentiums, eh? Sounds... er... I dunno.

Come on - ask 'em to do some LW benchmarks!

[Johnny]

Originally posted by Darth Mole
33% over corresponding Pentiums, eh? Sounds... er... I dunno.

Come on - ask 'em to do some LW benchmarks!

more specifically: the Dual 2GHZ was 33% faster than some Dual 3Ghz Pentium setup.

AND at running software not optimized for the G5.

J

[Darth Mole]

Aw, yeah - talk dirty to me... Come on G5...

[Triple G]

A dual 2Ghz G5 outperforming a dual 3Ghz Pentium? I'm assuming that would be a dual 3Ghz Xeon, no? (Since P4 chips are not able to be DP-configured, AFAIK.) Anyway, that definitely sounds like good news for Mac users. And um...no offense Darth Mole, but man...you really need to get a girlfriend or something. :D :p

[Ade]

Originally posted by Darth Mole
Aw, yeah - talk dirty to me... Come on G5...

LOL..

[Darth Mole]

Hey, TripleG - I HAVE a girlfriend, and a fine and sexy woman she is. But she ain't cost me two-and-a-half grand and kept me waiting for two months...

(And besides, her benchmarks are terrible. I showed her the Radiosity_Things.lws file like weeks ago - so far, nuthin')

[Ade]

Originally posted by Darth Mole
Hey, TripleG - I HAVE a girlfriend, and a fine and sexy woman she is. But she ain't cost me two-and-a-half grand and kept me waiting for two months...

(And besides, her benchmarks are terrible. I showed her the Radiosity_Things.lws file like weeks ago - so far, nuthin')

Some pics please.....:D

[Darth Mole]

Hey, Ade - maybe YOU need to get a girlfriend :D

[Triple G]

Originally posted by Darth Mole
(And besides, her benchmarks are terrible. I showed her the Radiosity_Things.lws file like weeks ago - so far, nuthin')

LOL! :D

[tumblemonster]

I'm waiting for the second run G5's, which, if Apple history holds true, will be speedbumped and released in January.

I'm not waiting because I think there'll be problems. I'm waiting because I'm broke!!!

-tm

[wapangy]

I've got a 1.8 Gig ordered that should be here hopefully soon. I'll post my benchmarks and such for LW when I get it.

The bad part is that LW is not optimized for the G5 and I doubt that NewTek is working on that. I emailed them asking if they were going to update LW for the G5 and this is the reponse I got:

Thank you for your interest in NewTek products. Please remember to visit our website at www.newtek.com for product updates and announcements.
-Gil
NewTek Technical Support


Wow, thanks a bunch....

[DaveW]

I know most people around here don't believe it, but from what I understand LW is fairly well optimized for the G4 (not to say it couldn't be better), and I don't think it would be unreasonable to assume that at least the same level of optimization would be done for the G5.

[luka]

would it be unreasonable to ask just what or how will lw8 be optimised?
I remember the Corona.p in the LW os9 was Altivec enhanced. So maybe we could get a list of Altivec enhanced features.

[Ed M.]

DaveW. : Who told you that?

--
Ed

[DaveW]

There was a review of G4's vs. Pentium 4 vs. AthlonXP, and the G4 did quite well in the Lightwave test. It was a dual 1.25ghz G4 vs. 3ghz Pentium 4, I forget what the Athlon was. According to their tests, the G4 rendered almost as fast as the P4, compared to the Adobe Premiere tests (and they mentioned a After Effects test) where the G4 performed closer to a 2.2 or 2.4 ghz P4. This was from a pretty PC-centric site (Aces hardware I think) and they were suprised that in Lightwave the G4 actually held up very well against the P4 and was quite a bit faster than the AthlonXP. They only tested the raytracing scene though, but at Blanos the G4 does quite well compared to x86 cpu's. Like I said though, this doesn't mean that there isn't plenty of room for improvement on Mac LW, just that it doesn't appear to be as bad as some people say. NewTek apparently did a better job optimizing for G4 than Adobe did with Premiere and After Effects. Even Photoshop feels more sluggish than it should on my 1ghz G4 compared to my 1.3ghz Athlon T-Bird.

I think the biggest problem is that you can't make full use of dual processors for most things in LW except rendering. So even though it renders about as fast as a P4, it doesn't feel as snappy when you model or animate since you're cutting your speed in half by only using one processor.

[NigelH]

Isn't it the same on the PC side? When I freelanced at a studio that used DP (2Ghz) Athlons, rendering was much faster than using my DP (533mhz) G4 at home, but modeling and animating didn't feel any snappier. I believe they used the same GeForce4Ti cards that I have (more or less), so all things being equal, I get the impression that Modeler and Layout are just not MP aware in general.

[DaveW]

Yeah, it is the same on the PC side. What I meant though, was that it takes two G4's to compare to one P4. So a dual 1.42ghz G4 system might render about as fast as a 2.8ghz or 3ghz P4, when you're modeling/animating it feels more like a 1.4ghz P4. That's why LW feels so slow on a Mac compared to a PC. On top of that, even with Jaguar OSX still isn't quite as responsive as Windows, but it's very close. I've heard Panther will finally solve that problem. And when the G5's get here then they should at least be on par with a 3ghz P4/Xeon as the Renderman benchmarks from Siggraph and the NASA tests indicate. And I'm sure anything that can benefit heavily from Altivec will be much faster on the G5. Isn't Apple or IBM working on a PPC compiler that will auto-vectorize code like Intel's compiler?

[Ed M.]

Isn't Apple or IBM working on a PPC compiler that will auto-vectorize code like Intel's compiler?

Auto-vectorization is not the answer. It's the answer for a lot of lazy programmers, but the only way to get the BEST possible performance out of the G4 (or G5) is to hand-tune the code. Well, the vector code anyway.

Read the comments by MrNSX, hohold and BadAndy here:

http://arstechnica.infopop.net/OpenTopic/page?q=Y&a=tpc&s=50009562&f=8300945231&m=9080959175&p=9

--
Ed

[Ed M.]

But to answer your question.. Yes Apple and IBM are working on bringing autovectorization to gcc.

--
Ed

[panini]

More proof that Mac users are delusional.

Quote: "and they were suprised that in Lightwave the G4 actually held up very well against the P4 and was quite a bit faster than the AthlonXP."

Really.

I just went to Blanos site and in half of those tests top scores for PCs are 2-3 times faster than the fastest G4s

In the other half the difference is from 50% - 100%

If that is considered holding up well, then my name is Brad Peebler. The Lightwave test somebody here did on dual 2Ghz G5 was about 60% slower than top PC scores. What a load of crap in shiny gay boxes those Apples are.

[Ed M.]

TROLL ALERT! TROLL ALERT!

PANINI IS A TROLL!

JUST WARNING YOU GUYS WHO DON'T ALREADY KNOW THAT.

--
Ed

[panini]

Here you go . Brand new G5s shipped and tested:

Cinebench rendering:

single 1.6 ghz G5 takes 160 seconds

dual 1.42 ghz G4 takes 107 seconds

Dual 2.4 ghz xeon takes only 51 seconds.



Maxon reports that 1.8 ghz G5 scores 188 on cinebench ( overall ).
Once software is optimized it can go up to 240 or so.

3 Ghz Pentium 4 already scores 360, in other words twice as good as current 1.8ghz G5.

By the way trolls are those who talk nonsense without any facts whatsoever. I always give you numbers, you Ed never give anything other than mysterious doom and gloom predictions for AMD.

Somebody please ban me from these boards. It's like a train wreck, I can't resist coming back ( after all it's only a click away from the PC forum )

[js33]

Hehehe,

It's just so fun to come back everyday to see what Ed has to say.:D

Cheers,
JS

[Ed M.]

That's funny, js, I say the same thing to myself abot you ;)

--
Ed

[DaveW]

Panini, it was a very pc-centric site, not a Mac biased one at all. In fact, the first page of the review was a bunch of anti-Mac spew about Apples advertising practices. And they were comparing single-processor x86 systems to dual-processor G4 systems. If you go back to Blanos' site and look at just the single processor x86 systems you wil see that the dual G4's hold up pretty well, better on some scenes than others of course. I looked it up for you, it was at Ace's Hardware: http://www.aceshardware.com/read.jsp?id=50000333

The last page is the one with the LW benchmark and conclusion:
http://www.aceshardware.com/read.jsp?id=50000336

The dual 1.25ghz G4 rendered the raytrace scene only 4 seconds slower than the P4 3ghz with HT off, and 33 seconds faster than the AthlonXP 2800+.

[Ed M.]

CineBench is being discussed here:

http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=48409524&m=5790920455&r=4820954185#4820954185

Other interesting threads discussing G5 performance can be found on ArsTech at the following links:

http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=9600994185&r=1770974185#1770974185

http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=1440956085&r=4520947085#4520947085

http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=8790912185&r=2750982185#2750982185

I find hobold's posts particularly interesting. Well worth the read.

--
Ed

[Ade]

Does Panini actually use Lightwave?

[js33]

Originally posted by Ade
Does Panini actually use Lightwave?

I don't know? But I know that Ed M. doesn't use Lightwave. Come Ed we know you really do use it or you wouldn't be here and be so passionate about getting Lightwave to run well on the Mac. So let's see some of your work Ed. You can see some of my older stuff on my website. :D

Cheers,
JS

[toby]

Come on JS, Ed has made the very logical point ( several times ) that whether he uses LW or a Mac or not has no bearing on the validity of his arguments. No need to bring that up again.

He seems to me like someone who's passionate about technology and processors, and if he doesn't use the products that he's arguing for, couldn't that imply that his arguments are that much more credible? He backs up anything he has to say, and he doesn't go to the PC forum to tell them that their systems suck.

Panini, on the other hand, spews obviously opinionated garbage and either doesn't even try to back it up or embarrasses himself in the attempt. He's obviously an insecure, attention-starved drama queen who has to make drama if there's none around. Add Macho jerk to that after his last post.

<sarcasm>
You found another test where a top-of-the-line P4 beats a bottom-line Mac!! Good for you Panini!! We'll ignore all the ones that show the G5 to be faster, because Macs are 'gay', and we want to be 'cool' like you.
</sarcasm>

[js33]

Hi Toby,

I disagree. How can Ed have any real insight to the problems if he doesn't use the equipment with the problems. All he can do is say what others have said. Plus he spouts off every chance he gets with why Windows is so bad. If it is so bad how come it runs Lightwave like a champ? He just strikes me as someone with a chip on his shoulder that has nothing better to do than take one side of an argument and dig up any person or quote that helps to further his cause. I just can't take him seriously in here because he doesn't use Lightwave and has no stake in it's future. I find the comments and opinions of real users far more valuable...such as yours, Beams, tallscot and eblu.

Cheers,
JS

[Ed M.]

I disagree.

That's the point. Your arguments are filled with nothing but objections (that would make it a fallacy). Your attempt at debunking anything I say based whether I use something or not is severely overstated and you sound like a whiner. Perhaps you can think of me as a research journalist. I also ask a lot of questions and the *Mac* users of this forum seem to appreciate the information that I bring to the discussions. These people see the wholes throughout your arguments. You bring no *real* substance to these discussions other than "your personal experience" Which by the way is another fallacy; It's called only providing Anecdotal evidence (AKA: confabulation) which is: using stories, opinions and personal experiences as evidence in support of *your* arguments. So, your only purpose here must be to annoy and enflame Mac-Lightwavers and rain on their parade. That's being a killjoy. Furthermore, bringing all these Ad hominem attacks against me isn't going to change a thing. As Toby said. It doesn't change a thing or the validity of the information I provide for the Mac-Lightwavers.

How can Ed have any real insight to the problems if he doesn't use the equipment with the problems. All he can do is say what others have said.

See the explanations above to see just how silly this sounds. I needn't be a medical doctor or surgeon to know that if I ingest rat poison and forgo seeking medical attention that it's likely I'll expire in a heap. Same goes if I were to slice my own wrists. You use these arguments to divert people's attention from the real issue at hand. This is a well-known tactic of a troll.

Plus he spouts off every chance he gets with why Windows is so bad.

Wrong. I don't complain half as much about Windows as I do about Mac-Lightwavers getting the shaft from NewTek. Mac-users reading these forums and posts know this. Besides, I don't have to spout. Windows speaks for itself.

If it is so bad how come it runs Lightwave like a champ?

Now that is *really* silly. It's also a loaded question. Did everyone read that question? Here's one for you js... How come a diesel engine runs so well on diesel fuel? Or the mayonnaise jar lid fits better on the mayonnaise jar from which it was removed than it does on a beer bottle? Jeez, your really reaching.

He just strikes me as someone with a chip on his shoulder that has nothing better to do than take one side of an argument and dig up any person or quote that helps to further his cause.

Lets stop with the Ad hominems already. If you have hard evidence or quotes from experts to bring to the table then present it to the Mac Lightwave users, but stop using vague, glittering generalities. You commit enough accidents when you *only* choose to look at benchmarks and information that *might* show the G5 in a sour light and use that information as the basis of the argument that all Macs suck at everything. You are making generalizations when in fact all these tests are the *exception* due to the information, knowledge, quotes and other reasons that I (as well as others) put forth. This is an accident of logic and it's a prime example of the fallacy of exclusion (AKA: card-stacking).

I just can't take him seriously in here because he doesn't use Lightwave and has no stake in it's future.I find the comments and opinions of real users far more valuable...such as yours, Beams, tallscot and eblu.

Js, people are onto you. You are attempting to form extremely weak arguments and yell as loud and as often as you can in hopes that you might convince someone that those arguments have any merit. You are attempting to sway the jury, yet there is no trial. You enjoy poisoning the well.

Personally I'm surprised the Mac-Lightwavers keep you around on their forums. You bring nothing interesting or terribly though-provoking to any of the discussions other than links and experiences that are meant to take the wind from their sails or enflame. I'm not aware of any of them wandering around the PC forums doing that. So, again, what is your purpose for contributing to these *Mac-LW* forums? What do you hope to accomplish? What interesting, thought-provoking information have you contributed? What claims were you able to back up with other information collected from a wide array of experts presenting different angles of view and varying perspectives?

Let's cut it out already js.. If you don't have anything constructive or interesting to add to these discussions other than posting bogus benchmarks and posting stuff that will only enflame the Mac group then please remain silent or perhaps you should wander around forums where more people think like yourself.

I will say nothing more regarding any diversions js tries to lure us toward.

OK, back on topic.

Has anyone checked out the links I provided? Any thoughts?

--
Ed

[panini]

Nooo, noo, no, no. Stop right there.

Lets not complicate thing. This is after all very simple.

No need to look up single this vs. dual that.

On Blanos just find the best scores for the PC platform and the best scores for the Mac platform.
Then compare. That's all you need to do.


You'll see that in about 1/2 of those PC scores are 2-3 times faster.
and the other half PCs are 50% or so to 100% faster.

And Cinebench so far shows that new G5s is about as fast as a year old Intel chips.
Spec scores show that AMD and intel top of the line chips are about 30-50% faster than top G5 ( a faster Opteron has just been released so that one should be another 20% faster than G5

It's pretty simple.

[toby]

"Did everyone read that question? Here's one for you js... How come a diesel engine runs so well on diesel fuel?"

Yes, I caught that question - good analogy. Another one - Photoshop works better on Macs, so something is wrong with Windows.

Ed M. and JS, you're both pretty cool except to each other - obviously clashing personnalities - can we please kick it down a notch

panini -
blah, blah blah blah, blah, blah blah!
You ignore everything that you don't agree with, so we'll be ignoring you.

[DaveW]

I never said or implied that the fastest G4 was as fast as the fastest PC. My original argument had nothing to do with Mac vs. PC, it was simply that NewTek has done a decent job of optimizing LW (or the renderer at least) for the G4.

And since you brought up the Opteron:
http://www.tomshardware.com/cpu/20030422/opteron-23.html#3drendering

[js33]

Originally posted by Ed M.
That's the point. Your arguments are filled with nothing but objections (that would make it a fallacy). Your attempt at debunking anything I say based whether I use something or not is severely overstated and you sound like a whiner. Perhaps you can think of me as a research journalist. I also ask a lot of questions and the *Mac* users of this forum seem to appreciate the information that I bring to the discussions. These people see the wholes throughout your arguments. You bring no *real* substance to these discussions other than "your personal experience" Which by the way is another fallacy; It's called only providing Anecdotal evidence (AKA: confabulation) which is: using stories, opinions and personal experiences as evidence in support of *your* arguments. So, your only purpose here must be to annoy and enflame Mac-Lightwavers and rain on their parade. That's being a killjoy. Furthermore, bringing all these Ad hominem attacks against me isn't going to change a thing. As Toby said. It doesn't change a thing or the validity of the information I provide for the Mac-Lightwavers.

<snipped to save space>


--
Ed


Ed,

I have nothing against you personally. But it's like this.
Someone asks you how the car handles and you can't really say because you never drove it. You can say "This guy said it handled pretty good but was kinda sluggish in the turns" or "This other guy said the brakes seemed to grab to hard" etc...

You can only bring a 3rd party perspective into things because you don't have any real experience with the program.

I have used Lightwave for about 8 years now so I know a little bit more about it than you do.

I also own Lightwave for the PC and the Mac and prefer to use it on the PC because I can work faster with it.

I bought an iMac about a year ago because I used to use Macs at a big printing company back in the early 90's and I liked it then.

Then I saw OSX and I thought it was pretty cool. So I bought it and was kinda disappointed in the performance. So then I went out and bought a 2.5 Ghz P4 PC for about half the price and have been using that ever since.

So since I own both it's hard not to compare them against each other. I really wanted the G5 to be a screamer but so far that's not turning out to be the case but I am still withholding final judgement.

Cheers,
JS

[Ed M.]

I really wanted the G5 to be a screamer but so far that's not turning out to be the case but I am still withholding final judgment.

As you should.

The G5s are out and the key apps that you want to compare on it need some tweaking. However, we could say the same about the Opteron, no?

Anyway, It's obvious and well-documented that these 'gotchas' exist and that it's highly likely that they are evident in the code of the slower apps. It's as simple as that. Once things are ironed out and developers take advantage of all the tools Apple has given them then things should run like a charm. Adobe did it and others are going to be releasing optimizations shortly.

--
Ed

[panini]

Well, I have pointed out too that New Tek isn't to blame for a crappy Mac platform.

What do you expect them to do? They can't make Macs faster.

Some people actually mentioned Cinema4D as a good example because they are coming out with an optimized version for G5.

Well, according to Maxon's own tests, even this optimized version will be slower than on top of the line Pentiums.

It's like this:

Adobe After effects: Pcs are generally twice as fast in most tests using AE

Lightwave: again, about twice as fast

Flash app performs better on a Mac, Flash player when viewing
flash movies runs significantly faster on PCs

Cinema 4D, again faster on a PC

Photoshop, unless you are talking about those bogus, paid for filters Apple uses to "prove" that mghz don't matter, even Photoshop is faster on a PC.

Doesn't this tell you people that it's not Macromedia, Adobe, New Tek and whoever to blame, BUT THAT IT IS APPLE AND THEIR SLOW CRAPPY MACHINES.

Also Maya ( the top of the line version ), Digital Fusion and Vegas do not even exist on a MAC. I mean, what the hell , All Macs seem to be good for are kindergartens and stoned teenage girls who like all the fruity iMac colors ( hey, Apple used that stoned chick in a commercial, don't blame me )

[Ed M.]

Wrong again. Check your facts. :p

--
Ed

[toby]

Did you guys hear something? - like a cat in a blender?

i meant to go before you Ed

[Ed M.]

Was that what that was? :D

[Ed M.]

Toby, did you check out the CineBench discussion?

--
Ed

[toby]

I will in a few minutes -
but panini is so full of stupid comments like
" Photoshop, unless you are talking about those bogus, paid for filters Apple uses to "prove" that mghz don't matter, even Photoshop is faster on a PC."
it hardly seems worth proving him wrong, because he just ignores facts and goes on to the next concocted accusation.

[Ed M.]

We shouldn't be loking to prove anyone wrong. When these questions arise, I find a list or forum where the developers and experts are actively discussing it. See the links and notes I posted a little ways back ;)

--
Ed

[tumblemonster]

Panini -

I think it comes down to user experience. I support a small office of PC users. Based on the number of major complaints and difficulties they experience, I will always buy Macs.

So what if some benchmarks show some PCs are a little faster. I still get more work done, in a more timely manner, on my Mac.

-tm

[Darth Mole]

And surely that's the key thing? yes, I need a machine to render quickly, but if I ONLY needed that, I'd buy a whole raft of cheap PCs and network them together.

However, I can only afford one machine. I need it to be a master of all trades - which also incldues video and DTP. And when the only real alternative is WinXP, then I'm happy to pay a premium for Apple products, for the ease-of-use, ease of installation, hardware integration - and Panther in a few week's time :)

[Ge4-ce]

Hey guys...

Check out the blamos website!!!

G5 lightwave benchies..

PC has been beaten!!

now Panini and other PC guys.. come out and play..

[js33]

Removed due to fake benchmarks

[Ge4-ce]

Well,.. I'm a misleaded myselves.. I know the guy that posted the benchmarks.. and contacted him

seems that they are fake.. He actually was trying to prove a point! These benchmarks at Blanos are just not right!! everybody can enter whatever they want! he said he just looked up the best scores and just made them a little bit better for the G5.. remember.. the G5 duals aren't shipping yet so this can't be true.. wat he just did is undermine the entire credibility of that Blanos benchmark site. You need an undependable organisation to mesure benchmarks.. So the arguments of all the pc guys based on blanos' site are just nonsense.. the guy posting these benchies proved that now!

an undependable test from newtek has to be done!

[js33]

OOPS!

I thought it seemed a little to soon for the Dual G5s. But really these are the kind of benchmarks I'm expecting from the G5.

Perhaps the PC people are more honest about their benchmarks.

If that guy faked them he just slapped every Mac user in the face and did a big disservice to the entire Mac community.

Shame on him.

I do think Blanos should make people send in their screens of a finished render to show the time but those could be faked as well.

Also some of the benchmarks only take a few seconds so they don't really prove anything anyway.

Man, Beam is probably turning red about now.

But seriously aren't those the kind of numbers we should expect out of the G5? I will still be the first to congratulate the Mac if it does have good numbers.

Now waiting for REAL benchmarks........

Cheers,
JS

[Ge4-ce]

Well.. I was really pissed when I heared he faked the benchmarks.. But somewhere he has a point on the credibility of these benchmarks..

BTW who says! that pc-people are more honest in these things? that's personal.. If one guy want's to have the fastest machine to prove his point,.. he maybe fakes results! I do not approve this kind of stuff, but It does put these blanos benchmarks in another perspective!

[Ge4-ce]

also.. the guy who posted the benchies is a big Mac-fan and wanted to undermine panini's idiot arguments..

[js33]

Originally posted by Ge4-ce
also.. the guy who posted the benchies is a big Mac-fan and wanted to undermine panini's idiot arguments..

Well he's not too smart because he just caused a lot more harm than good. He just trashed the crediblity of all Mac benchmarks even more than they were already.

I'm sure as people get them we will see real numbers soon.

Cheers,
JS

[Ge4-ce]

indeed.. the world still has stupid people..

[Chris]

Hey gang...

My benchmark site was created to assist professionals and serious hobbiests determine how fast they could expect their computers to render standard LightWave benchmark scenes. I also know that it helps many people make purchasing decisions when they are looking at new hardware. It was never intended to be a comprehensive hardware benchmarking resource. If you are looking for that type of information, there are plenty of hardware review sites that purport to do a battery of tests and present the results with all kinds of nifty charts and graphs.

As mentioned in this thread, I could require that people submit screen shots with their scores to somehow prove that they aren't lying. This would not help at all, and I don't want to do this for a number of reasons: It is more work for the submitter, and therefore less people would contribute. It would be additional work for me to verify and confirm, so new data would take longer to appear on the site. Finally, most of the people using visiting my site are graphic artists - how difficult do you really think it would be to fake a screen shot?

There is no good way for me (or anyone) to tell if people are lying. My benchmark site is based on trust. For whatever reason, some people feel compelled to post bogus times. I don't understand why - it's not like I'm giving out prizes for the fastest computer or anything. If people want to post fake times to somehow justify their hardware preference, that's fine. I will remove the posts as I can. As more times are posted for any given class of hardware, the bogus times tend to become visible and they are removed.

Thanks to everyone who notified me about the lastest bogus posting. It will be removed shortly.

- Chris

[turbo]

I just can't believe this always has to turn into a mac vs pc debate.. who cares! You want pc, go pc, you want mac, go mac.

But, I for one, have always and will always buy mac. Productivity is high, smooth, happy, instinctive and I have a real affinity with these machines. No problems, no need for tech support, no swearing at winblows..

My 5th mac arrives in 22 days. I am confident I will have nothing but raves to report. 3 weeks! w00t! can't wait! :p

Dual G5 2 GHZ
256 MEG nVidia (pulling the ATI 9600)
160 GIG Hard Drive
2 GIGs RAM
20" Apple Cinema Display
..and whatever I can migrate from my current DP 500

Then, later I will upgrade my LW to 8.
Forever happy on my macs. :D :D :cool: :cool: :D :D

[js33]

Originally posted by Chris
Hey gang...

My benchmark site was created to assist professionals and serious hobbiests determine how fast they could expect their computers to render standard LightWave benchmark scenes. I also know that it helps many people make purchasing decisions when they are looking at new hardware. It was never intended to be a comprehensive hardware benchmarking resource. If you are looking for that type of information, there are plenty of hardware review sites that purport to do a battery of tests and present the results with all kinds of nifty charts and graphs.

As mentioned in this thread, I could require that people submit screen shots with their scores to somehow prove that they aren't lying. This would not help at all, and I don't want to do this for a number of reasons: It is more work for the submitter, and therefore less people would contribute. It would be additional work for me to verify and confirm, so new data would take longer to appear on the site. Finally, most of the people using visiting my site are graphic artists - how difficult do you really think it would be to fake a screen shot?

There is no good way for me (or anyone) to tell if people are lying. My benchmark site is based on trust. For whatever reason, some people feel compelled to post bogus times. I don't understand why - it's not like I'm giving out prizes for the fastest computer or anything. If people want to post fake times to somehow justify their hardware preference, that's fine. I will remove the posts as I can. As more times are posted for any given class of hardware, the bogus times tend to become visible and they are removed.

Thanks to everyone who notified me about the lastest bogus posting. It will be removed shortly.

- Chris

Thanks for being on top of it Chris.
It is sad that someone so wanted the Mac to beat everything that he was compelled to post fake benchmarks.
I'm sure the real proper benchmarks will be posted soon enough and Chris will know the right ones from the fake ones.

Cheers,
JS

[Ed M.]

It is sad that someone so wanted the Mac to beat everything that he was compelled to post fake benchmarks.

I THOUGHT HIS INTENTION WAS TO SHOW HOW EASY IT WAS TO MANIPULATE PEOPLE'S PERCEPTION BY BENDING THE TRUTH WHILE AT THE SAME TIME SHOWING THE WORLD HOW EASY IT IS TO POST BENCHMARKS THAT AREN'T VERIFIED, YET TAKEN TO BE TRUE. WASN'T THAT THE PURPOSE BEHIND IT?

--
Ed

[panini]

Here you go.

Very detailed testing compares the brand new 1.6 ghz G5 to older g4s and a very, very old AMD Athlon running at 1.67 ghz

Athlon wins pretty convincingly.

It would be scary to see what kind of a$$whooping one of those 3ghz + chips would have unleashed upon G5.

http://www.theandyzone.com/Computer/shootout.html


I'm not even going to comment on those fake Blanos G5 benchmarks other than to say: "What else do you expect from Apple fans , they learned from the worst liar of them all, Steve Jobs"

[Ed M.]

Hey, someone here something?

When is NewTek going to take care of the bats in their attic?

--
Ed

[Beamtracer]

Originally posted by panini
What else do you expect from Apple fans
As I said... there's absolutely nothing going on in the Windows forum to keep him entertained.

[toby]

Originally posted by Beamtracer
As I said... there's absolutely nothing going on in the Windows forum to keep him entertained.

Evidenced by the fact that he's NEVER posted there.

[js33]

Originally posted by Ed M.
I THOUGHT HIS INTENTION WAS TO SHOW HOW EASY IT WAS TO MANIPULATE PEOPLE'S PERCEPTION BY BENDING THE TRUTH WHILE AT THE SAME TIME SHOWING THE WORLD HOW EASY IT IS TO POST BENCHMARKS THAT AREN'T VERIFIED, YET TAKEN TO BE TRUE. WASN'T THAT THE PURPOSE BEHIND IT?

--
Ed

I think he just couldn't stand not seeing the Mac in the top ten for the last 4 years and he finally snapped. What ever his intention was it sure backfired on him. It just makes Mac benchmarks even less credible than they already were. Anyway Chris is on top of it and already removed the bogus benchmarks. As Chris stated the benchmark site is based on trust and that when you have several benchmarks for the same system it is easy to spot the fake ones. But since there was only one set of benchmarks posted by one Mac user of a machine that isn't even shipping just yet it was pretty easy to determine they were faked.

I actually hoped they were real. Oh well time will tell what the real benchmarks will be.

Cheers,
JS

[toby]

"I think he just couldn't stand not seeing the Mac in the top ten for the last 4 years and he finally snapped"

As GeForce said, "He actually was trying to prove a point", and he does know the person.

"Perhaps the PC people are more honest about their benchmarks"

Please take your insults somewhere else.

"makes Mac benchmarks even less credible than they already were"

What made them less credible to begin with? That they were faster than pc's? That's all been backed up, but you still think they're suspect.

You're just too biased for me JS.

[Chris]

Originally posted by toby
"I think he just couldn't stand not seeing the Mac in the top ten for the last 4 years and he finally snapped"

As GeForce said, "He actually was trying to prove a point", and he does know the person.


Well, according to my logs, he actually IS the person. There is no 'friend.'

- Chris

[toby]

Originally posted by Ge4-ce
indeed.. the world still has stupid people..

weird!

[js33]

Originally posted by Chris
Well, according to my logs, he actually IS the person. There is no 'friend.'

- Chris

So Ge4ce was the one that posted the fake G5 benchmarks!

Man this is getting weird.


Originally posted by Ge4-ce
Hey guys...
Check out the blamos website!!!
G5 lightwave benchies..
PC has been beaten!!
now Panini and other PC guys.. come out and play..

So you figured you'd post some really good "fake" benchmarks to make a point?


Originally posted by Ge4-ce
Well,.. I'm a misleaded myselves.. I know the guy that posted the benchmarks.. and contacted him

seems that they are fake.. He actually was trying to prove a point! These benchmarks at Blanos are just not right!! everybody can enter whatever they want! he said he just looked up the best scores and just made them a little bit better for the G5.. remember.. the G5 duals aren't shipping yet so this can't be true.. wat he just did is undermine the entire credibility of that Blanos benchmark site. You need an undependable organisation to mesure benchmarks.. So the arguments of all the pc guys based on blanos' site are just nonsense.. the guy posting these benchies proved that now!

an undependable test from newtek has to be done!

Then you felt guilty about it so you tried to find a way out.


Originally posted by Ge4-ce
Well.. I was really pissed when I heared he faked the benchmarks.. But somewhere he has a point on the credibility of these benchmarks..

BTW who says! that pc-people are more honest in these things? that's personal.. If one guy want's to have the fastest machine to prove his point,.. he maybe fakes results! I do not approve this kind of stuff, but It does put these blanos benchmarks in another perspective!

Well apparently PC people ARE more honest about it!
You just proved that bigtime!

Originally posted by Ge4-ce
also.. the guy who posted the benchies is a big Mac-fan and wanted to undermine panini's idiot arguments..

Well I think you went about it in a very bad way.


Originally posted by Ge4-ce
indeed.. the world still has stupid people..

Indeed.

Cheers,
JS

[js33]

Originally posted by toby
"I think he just couldn't stand not seeing the Mac in the top ten for the last 4 years and he finally snapped"

As GeForce said, "He actually was trying to prove a point", and he does know the person.

"Perhaps the PC people are more honest about their benchmarks"

Please take your insults somewhere else.

"makes Mac benchmarks even less credible than they already were"

What made them less credible to begin with? That they were faster than pc's? That's all been backed up, but you still think they're suspect.

You're just too biased for me JS.

No not biased. Just honest. See above.
Still waiting for the real benchmarks.

Cheers,
JS

[Ed M.]

Ya know js.. some people might consider *other* benchmarks to be the *real* ones (for them). I suggest you change your comments to the following:

"I'm waiting for the G5-optimized plug-ins to be made available before I render final judgment because for *me* Lightwave is the only benchmark that I'm interested in, since I depend on that app so heavilly"


--
Ed

[js33]

Originally posted by Ed M.
Ya know js.. some people might consider *other* benchmarks to be the *real* ones (for them). I suggest you change your comments to the following:

"I'm waiting for the G5-optimized plug-ins to be made available before I render final judgment because for *me* Lightwave is the only benchmark that I'm interested in, since I depend on that app so heavilly"


--
Ed

Ed,

I've already said that Lightwave benchmarks are the ones I'm waiting for. :D

I'll let Ge4ce's fiasco go as I think he did a lot more to harm the Mac than he did to harm Chris' benchmark site.

After all the site is based on trust. Something that Ge4ce is no doubt doing a lot of soul seaching for right now.

Ge4ce I forgive you. But let's let the real benchmarks speak for themselves.

Cheers,
JS

[Beamtracer]

Wow! So Ge4ce is the mystery benchmark faker.

I think this goes down as being the most bizarre thread I've seen yet on the Mac forum.

So... where is Ge4ce?
What has he got to say?
Will he post back here ever again?
Will he devise a new alias so nobody knows when he's returned?

[Beamtracer]

Hello Ge4ce!
Where are you?
Come out, come out, wherever you are!

[js33]

Originally posted by Beamtracer
Wow! So Ge4ce is the mystery benchmark faker.

I think this goes down as being the most bizarre thread I've seen yet on the Mac forum.

So... where is Ge4ce?
What has he got to say?
Will he post back here ever again?
Will he devise a new alias so nobody knows when he's returned?

Yes Beam I agree.
This is better drama than you see in most movies today.

I think he at least owes you an apology.

I think he'll probably lay low for awhile or come back as somebody else.

Again ge4ce I forgive you but I think Beam wants to string you up to the nearest tree.

Cheers,
JS

[toby]

Originally posted by js33
No not biased. Just honest. See above.
Still waiting for the real benchmarks.

Cheers,
JS

That's ridiculous.

Somebody posting a false benchmark does not make your slanted opinion any less slanted, or PC people any more honest than Mac people.

You're also assuming that the false benchmark was not meant to make a point, and riding your moral high-horse based on your assumption that he "snapped". Notice that Ge4ce was the one to blow the whistle.

[js33]

Originally posted by toby
That's ridiculous.

Somebody posting a false benchmark does not make your slanted opinion any less slanted, or PC people any more honest than Mac people.

You're also assuming that the false benchmark was not meant to make a point, and riding your moral high-horse based on your assumption that he "snapped". Notice that Ge4ce was the one to blow the whistle.

Toby,

I think you need to reread the thread. Ge4ce was the one that posted the fake G5 benchmarks in the first place.

Cheers,
JS

[toby]

Originally posted by js33
Toby,

I think you need to reread the thread. Ge4ce was the one that posted the fake G5 benchmarks in the first place.

Cheers,
JS

:rolleyes: Yes, I know, that's the point.
Don't you get it?

[js33]

Originally posted by toby
:rolleyes: Yes, I know, that's the point.
Don't you get it?

So your saying he lied to make a point... and that's a good thing?

No I guess I don't get it.

Cheers,
JS

[toby]

Originally posted by js33
So your saying he lied to make a point... and that's a good thing?

No I guess I don't get it.

Cheers,
JS
I'm not saying what he did. What I'm saying is that you are assuming the worst intentions, then using it to portray yourself and PC people as 'above' Mac people.

Judging a persons honesty by what computer they use is absurd. You should be ashamed of yourself as well.

[js33]

Originally posted by toby
I'm not saying what he did. What I'm saying is that you are assuming the worst intentions, then using it to portray yourself and PC people as 'above' Mac people.

Judging a persons honesty by what computer they use is absurd. You should be ashamed of yourself as well.

Well I obviously never meant that each and every person who uses a Mac is a liar. That's ridiculous of you to think that.

But when he did what he did he spoke for the community at large.

The last thing the Mac community needs right now is someone posting fake benchmarks anywhere.

Cheers,
JS

[turbo]

..*runs to unsubsribe from this thread

[archiea]

So far, the G5 seems a dissapointment....

Compared to the 1.6ghz Anthlon, it was the same if not a bit slower..

1.6ghz on the PC is at least a year and a half old..

I got a 2.4 ghz system for $1600 last feb with built in Firewire, USB 2.0, Optical audio out, built in 5.1 surround, ATI 9700, DVD player, CD burner, Multi-media card reader, oh the list goes on....

Point is, if you are going to compare the Mac to a PC, you;re are going to lose.

Buy a mac because you want a mac. Be happy that a new processor stragedy is going to follow. enjoy the iApps, Dotmac, Safari (the font enlarger is almost worth the price of a mac!), OS-X. Enjoy the fact that for the most part you can run LW on it (with an Nvidia card).

If you want a hardcore LW box, mac just isn't it. Its a great proprietary set up designed for a niche market. Its like the audis or VW of car makers. Do you see Beetle and Audi TT owners drag racing against Mustang Cobras or Corvettes? Do you see them bitching about being able to? No.

And if you believed Steve job's crap, then you deserve whatever dissappointment is coming to you...

I'll get a G5, but not in september, and probably not until after panther, when a 64bit OS will surface. THATS when you get a G5....

I started this thread to get a comparison to the G4, not a PC. Like you need brain to figure that a PC will beat out a mac.

[Ge4-ce]

Ok.. so this is the moment in this soap where the bad guy confesses..

I really did a bad thing.. so I DO appologize for my behaviour. In the beginning, I really thought I was proving a point. (actually telling that individual benchmark reports aren't to be trusted) but things got out of controle... and to be honest.. I checked Chris's site regular by myself and once posted benchies of my own (wich were acurate!)

So.. indeed I reported my own mistake at chris..

So what do they do with the bad guy in a soap?

You know what would make this soap actually be a soap? If I would actually be a PC user converted by Bill Gates and set up to harm the Mac-community. It wasn't even my fault because the brainwashing came from the noise from my dual xeon fans!!

May the Mac be with you... Always..

I propose a vote.. (pc-people excluded because I DO NOT Care what they think about me)

Me in or out this forum..

[js33]

Well I said I forgive you. I think you will think it through a bit more next time though. ;)

I have to admit that at first for a brief second I thought they were real benchmarks as Beam did as well. He was ready to shout it from the rooftops so I felt I better tell him quick before he embarrased himself.

Now on to the real LW benchmarks.

Cheers,
JS

[Beamtracer]

Ge4-ce! You've created the most drama-filled thread there has ever been.

I thought you would be in hiding. Maybe even on the run!

Originally posted by Ge4-ce
I really did a bad thing.. so I DO appologize for my behaviour

Your confession is reminiscent of evangelist Jimmy Swaggart! I can imagine you standing there with tears gushing from your eyes!

..sniff...sniff... oh no, now I'm getting emotional.

Originally posted by Ge4-ce
Me in or out this forum..
Oh Ge4-ce, please stay in the Mac community. Work for good, not evil this time.

[mlinde]

So I'm requesting that for Lightwave users, we post (or list) benchmarks that are Mac - to - Mac benchmarks, for systems and such. I use XBench (and just in the last hour or so) which is a Mac OS ONLY benchmarking application, and it will test for all systems, including the newly released G5s.

Feel free to add to the benchmarks, mine is posted as "Orson Welles"

http://www.xbench.com/

[skippy]

anyone alarmed at the tepid xbench results turned in by the G5?

s

[Ed M.]

Xbench has it's known issues with the G5 (supposedly addressed in this version). It was discussed on Ars and elsewhere. The issues were well known and Chris Cox and a couple of Apple performance engineers had looked at the code back in March. It uses a lot of instructions that are *don'ts* for the G5.

on another note, Chris says that his STREAM code on the G5 does over 2400 MB/s and that is without AltiVec.

Their STREAM code uses dcbz, and uses vec_dst very badly. It needs to be corrected for the G5 in order for it to show accurate results.

From a poster named 'gcc' over at Ars:

vecLib needs to be thrown out on this test or rewritten to give consistent results. There's obviously something wrong with the implementation in this test. If we are seeing caching influence the numbers between runs then a much, much longer period of testing should be used to stabilize these effects.

From the asm profiles I've looked at on Xbench most of the tests are far too trivial to make any assessment about performance on real world apps.

From Ashby (From Ars):

I'll reserve any further judgments until I see Photoshop scores (preferably running in 10.3). Based on the scoring weirdness with my Powerbook earlier in this thread, this Xbench app seems fairly unreliable. Not sure I trust the quality of the code.

Notice that the iBook 700 has almost the same GCD recursion score as the G5 1600???

WTF?

68 vs. 73 despite G3 vs. G5 and more than 2x the frequency...

Clearly this test is fubar.

Zapchud:
Quite obviously

This test must be one of the most unreliable ones I've ever seen. Not only does the scores jump up and down depending on how good Xbench's hair day is, but the way it compares different CPU-architectures really makes it the most irrelevant synthetic benchmark to date.

Hell, even Let1kWindowsBloom is a better benchmark, at least, it's benching a real world operation :P

Bottom line is that XBench not only needs a lot more work, but it's really no longer of any use since its results are so incinsistent and *obviousaly* flawed. Even on the same machine runing the test several times.

http://arstechnica.infopop.net/OpenTopic/page?a=srch&s=50009562&findw=Xbench

--
Ed

[Ge4-ce]

Do you mean the G5 single Processor 1.6 Ghz?

I think that the Mac community (besides the suffer from me.. :rolleyes: ) suffers a lot from the Multi vs single processor issue.. What I'm trying to say is: Apple should not release any powermac's without dual processor!

When motorola and Apple some one or two years ago noticed that they did not gonna made it with their G4 Vs pc-world.. they pulled the alarm bell.. Yikes! We need dual to compete with the rest. And actually they saved their *** by doing that. Now we cannot expect that single 1.6 Ghz systems will winn from dual xeons or 3 Ghz P4! you allways have to mesure up the high-end VS high-end. Meaning the dual 2 ghz G5 against something else. Since it is not there yet (again being sarcastic :rolleyes: ) those early results from the 1.6 and 1.8 machines are not that important! you should compare those against the dual 1.25 G4 and the single 1Ghz G4 or meaning: the lowest and midrange model of the previous G4 models..

then also the Apps optimising is a big issue where good code can perform better performance. (Premiere vs FCP, Combustion vs AfterFX,...)

[js33]

Ed,

Are they any GOOD benchmarking tools on the Mac?
If so what are they? I want to try one on my iMac.

Cheers,
JS

[Ed M.]

Xbench was semi-reliable until they hosed it trying to *correct* it for G5.

Here is something that you should try... Download the older *pre-G5* version and then the newest release and look at the results. They should be identical. They won't be. The test is broken.

--
Ed

[js33]

I know Xbench is broken so what is the alternative? I imagine Apple should have some benchmarking tools?

Cheers,
JS

[mlinde]

Originally posted by Ed M.
Here is something that you should try... Download the older *pre-G5* version and then the newest release and look at the results. They should be identical. They won't be. The test is broken.

Ok, I'll bite :)

I don't have a G5 machine, are you saying that if I were to download an older version of XBench on my G4, and run the tests, I would get significantly different results? Or are you saying that the new XBench (released last night) doesn't correct the known G5 issue, rather it introduces new issues?

[Ed M.]

Probably both. Have you seen the results posted over at Ars? If not, then it might be worth checking out.

--
Ed

[mlinde]

Originally posted by Ed M.
Probably both. Have you seen the results posted over at Ars? If not, then it might be worth checking out.

I'm not disputing results from Ars Technica, I'm wondering about how current the discussion of XBench issues is, since the new version was just released last night. I understand that there were issues with XBench, and there may still be issues with it. What I'm questioning is if anyone has tested the brand new version (as I keep saying, released LAST NIGHT) to verify the same issues. Show me a thread with new data from today about XBench inaccuracies, please.

[Johnny]

Originally posted by Ed M.
Probably both. Have you seen the results posted over at Ars? If not, then it might be worth checking out.

--
Ed


where's the link to them? I tried a search: zip. Looked through some likely columns: zip.

J

[Jimzip]

Hey, how about posting a scene with different versions.
Eg. "MyScene_01" has no raytracing,
"MyScene_10" has every raytracing option, Radiosity and caustics on.
Then, we all download the scenes, render them without tweaking the settings at all, and post a list of results.
This would have to be a trust experiment. Some people would have to be honest for once, but we'd get a true indication of where all our computers are at. I think it would be fun and interesting. Anyone else agree?


-------------------------------------------------------------------------------

Eg. of list:
My com: Dual 1.25 GHz PowerMac G4
1.5GB RAM........... 64MB ATi Graphics card.
MyScene_01:................................ 2:53
MyScene_02:................................ 4:23
etc.

-------------------------------------------------------------------------------


regards,
........ Jimzip.
Let me know, I'll make some scenes and get them up here.

[toby]

are we allowed to change the # of threads

[js33]

Why not just use the benchmarks scenes that come with Lightwave? Some of them are silly like textures because they only take a few seconds but the radiosity and raytracing scenes are useful.

Cheers,
JS