G5 optimised app's

[wizlon]

So far several app's have been or are being optimised, Photoshop and Cinema 4D being two of them.

Will anyone from Newtek confirm that a G5 optimsed version of Lightwave is underway, seems like the perfect time to do it as they have to compile version 8 anyway.?


Maxon's quote:

"OK, some news directly from the MAXON development lab:

Of course all the following numbers are not final, no promise at all !!!!!!!

This is based on the information we have right, now, there is still a of of work to do and we still have to wait for a new compiler...

With the current CineBench a single G5 1.8GHz scores at about 188, the optimized version will maybe score at about 238...

A hypothetical single G5 2.0GHz could score at about 210 on the old CB, optimized could be 265...

A dual G5 2.0 could maybe score at about 480 with the optimized version of CB....

Depending on the new compilers and our findings (thanks a lot to Apple for being extremely helpful and cooperative) we might even crack the 500 score for the dual G5 2GHz...

Again, no promise and of course no release date


Cheers, Richard"

[panini]

If I'm not mistaken a single Pentium running at 3Ghz scores 360
on cinebench. That would be only twice as fast as that G5.

I guess Pentiums aren't all they are cracked up to be. I expected at least 3 times faster.

Here are some day old tests for you:

Cinebench rendering:

Brand spanking new 1.6 ghz G5 takes 160 seconds

old dual 1.42 Ghz G4 takes 107 seconds

dual Xeons running at 2.4 Ghz ( a year old chips ) take only 51 second.

[Lynx3d]

My slowly aging Dual Athlon 1800+ (that's 1,53GHz) has a Cinebench 2003 rendering score of 355

And yea a P4 3GHz with Hyptherthreading scores just above 360...

(if someone understands german: http://www.cgworld.de/thread.php?threadid=2001&sid=&hilight=cinebench )

Ther german site techChannel (http://www.tecchannel.de/hardware/1164/28.html) has some benches too, dual Xeon 3,06GHz scores 627 and Dual Opteron 244 (think that's 1.8GHz?) reaches 475 (that is 32bit, not optimized either, the MPs score higher actually due to higher clock speeds)

Just to give some overview...

[Beamtracer]

Welcome back Panini (our resident Microsoft promotional attaché.)

I hope you enjoy yourself while you're here making fun of the Mac. I guess nothing much is happening on the Windows PC forum to keep you entertained.

[luka]

I have spoken to an Apple SE and this is his quote to me. "Also, in most cases it is necessary for developers to recompile their code with the GCC 3.3 compiler and the G5 processor profile in order
to get the best performance. I have not heard if NewTek have committed to doing this as yet."

So I would like Newtek to respond to this and let us now their G5 optimisation plans.:(

[Zarathustra]

Forward that directly to Netwek and/or Chuck. You can't rely on them reading these forums.

[Beamtracer]

Originally posted by luka
So I would like Newtek to respond to this and let us now their G5 optimisation plans.:( I get the impression that Newtek is rushing to get a workable Lightwave8 out the door, with the intention of doing optimizations at a later incremental revision.

I guess when it is released we'll find out pretty quickly how it runs on the G5. It might be hard to separate what speed increases are a result of optimized code, and what are purely a result of the G5 being faster.

[Lynx3d]

If newtek actually uses GCC for PowerPC platform...donno.

Thought GCC was not really the best for PowerPC, some things "fixed" by changing values that actually don't reflect the hardware.
Doesn't mean they will really get the fastest LW if they use the first "G5 optimized" GCC.

I read that on the x86 side Newtek used MSVC compiler and later (since 7.0b i guess) Intel compiler, and IMO it definitely has performance reasons...

[mlinde]

Originally posted by Beamtracer
I get the impression that Newtek is rushing to get a workable Lightwave8 out the door, with the intention of doing optimizations at a later incremental revision.

Not to accuse you of NT bashing beam, but why do you have the "impression that Newtek is rushing to get a workable Lightwave8 out the door" when the only release announcements were Q4 2003?

Q4 2003 goes on for a while, and it hasn't even started yet. Unless you have contact with a beta tester who's telling you the software is buggy as hell (and they shouldn't be, due to non-disclosure), you are just blowing smoke, and that's not like you, so where are you coming from?

[luka]

Originally posted by Zarathustra
Forward that directly to Netwek and/or Chuck. You can't rely on them reading these forums.

Yes that's a good idea I'm sure I had Chucks email here somewhere?

Yes I have emailed Chuck the message. As soon as I get a response I shall post it here.

[Beamtracer]

Originally posted by mlinde
Unless you have contact with a beta tester who's telling you the software is buggy as hell (and they shouldn't be, due to non-disclosure), you are just blowing smoke, and that's not like you, so where are you coming from?

Michael,

Where did I say that the next Lightwave is "as buggy as hell"? I didn't make any inference about buggyness one way or another. Where was I indulging in Newtek bashing? Why am I "blowing smoke"?

Sure Newtek would be rushing to get Lightwave 8 "out the door". Why wouldn't they? That doesn't mean it'll be shoddy. I'm sure Newtek is working very hard to make sure it's stable. Platform optimizations will come with later incremental updates.


Originally posted by Chuck

There's a sense in the forward-looking discussions of Marketing and Development that after the initial generation of [8], we want to look at taking better advantage of each platform. Right now, though, the team is just concentrating on getting the first iteration of LW [8] out the door, so concrete discussions and planning in that regard will come later.

[mlinde]

Sorry beam, read a bit much into it, didn't I?:eek: