Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > Hardware - Troubleshooting and Discussion > Mac Desktops > Well, well... The G5 dual IS the fastest desktop computer in the world. (Photoshop)

Well, well... The G5 dual IS the fastest desktop computer in the world. (Photoshop)
Thread Tools
Eug Wanker
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 17, 2003, 08:25 PM
 
Thx to eggman for his work doing the dual 2.0 G5 benches. These are the normalized results of various machines from Ars. (Normalization places equal value on each of 21 subtests in the bench, and compares each to a 1 GHz Athlon running Photoshop 6.)

The dual G5 2.0 simply blows everything away.

The G5 1.8 scores are pretty good too. The 1.6 isn't as good, but I wonder if part of that is due to the memory bus speed and because the guy who ran the test hadn't known about the bus slewing issue. (He didn't set it to Highest Performance.)

Code:
2x 2000 G5 OSX 10.2.7 (G5 plugin) 547 (energy settings highest perf) 2x 3060 Xeon (OC'd 2400) 488 2x 2930 Xeon (OC'd 2400) 471 3200 P4 (800MHz) 427 3000 P4 (800MHz) 405 3495 P4 (OC'd 3.06) 386 3060 P4 XP Pro (533 FSB) 358 HT 2x 2200 Xeon PC 800 RDRAM CPQ Evo 357 HT 2x 1500 G4 (OC'd 1420) 348 2x 1333 G4 DDR OS9.2 (oc'd 1.25) 346 1800 G5 OSX 10.2.7 (G5 plugin) 344 (energy settings highest perf) 2x 1420 G4 OSX 10.2.4 338 2x 2400+Athlon MP 338 2x 1250 G4 OS 9.2.2j 337 3200+Athlon XP 332 1800 Opteron(dual-chnlDDR 333) 332 2x 1333 G4 DDR OSX10.2.2(oc 1.25) 326 1800 OPteron(singl-chnlDDR333) 320 3000+Athlon XP 318 2x 1250 G4 OSX 10.2.5 318 2x 1250 G4 DDR OSX 10.2.1 316 2x 1800 Athlon MP 312 2800+Athlon XP Barton 298 2x 2000 P4 Xeon 286 2x 1200 G4Powerlogix(867MHzG4/QS) 285 upgraded 2x 1533 Athlon MP 285 2x 1533 Athlon MP 283 2530 P4 mobile (OC'd 1400) 282 2700 P4B (OC 2400, 600 MHz FSB)280 2x 1466 Athlon XP 279 1600 G5 OSX 10.2.7w/G5 Plugin 276 *MacNNscores 2666 P4 (DDR 333) 269 2x 1000 G4 DDR 10.2 267 2400+Athlon XP 262 2x 1000 G4 OS9 260 2x 1000 G4 OSX 10.1.5 254 2400+Athlon 252 2400 P4B (800MHz) 251 2400b (sis 648 DDR400) 251 1600 Centrino IBM T40 250 2400 P4 (533MHz bus) 249 2400 P4 B 241 2340 P4 (overclock) 239 1600 Centrino Dell D800 236 2400 P4 234 1800+Athlon XP (1533 MHz) 226 1577 oc'd Athlon XP (Lestat) 221 2x 1000 G4 OSX 10.2.2 (upgraded) 218 ?!(dual 533 logic board) 1548 Athlon XP 214 1670 Athlon XP (2000+) 213 1667 Athlon XP 211 1400 Athlon XP 1600+ xp pro 200 1x 1533 Athlon MP 197 1300 Centrino Sony VAIO Z1A 196 1000 G4 17" Powrbk OSX 10.2.6 196 2000 P4 Xeon 194 1400 Athlon XP 1600+'98SE 191 1000 G4 OSX TiPbk 10.2.2 185 2x 533 G4 OSX 10.1.5 175 2x 533 G4 OS 9.2.2 174 1800 P4 173 1200 AthlonMP 168 1508 Celeron (overclock) 167 1400 PIII Tualatin 160 **? 2x 550 G4 OSX 10.2.3 (OC Cube) 160 **? 2x 500 G4 OSX 152 2x 450 G4 OS9 151 1333 Athlon TBird 147 2x 450 G4 OSX 10.1.5 143 800 G4 Pbook OSX 1MB L3 135 733 G4 (miro7) 134 667 G4 PBk OS9 noL3 127 667 G4 PBk OSX 10.2.3 no L3 125 466 G4 OS9 123 667 G4 OSX TiPBk 10.1.5 noL3 121 866 PIII 114 466 G4 OSX 133 MHz bus 112 550 G4 Powrbk OS9* 104 500 G4 Pbook (OC'd 400) 103 1x 450 G4 OSX 100 MHz bus 101 1000 Athlon TBird (PS6.01) 100 550 G4 Powrbk OSX* 95 933 Transmeta Crusoe Sony 78 700 G3 iBook 74 600 G3 iBook OS 9.2.2j 70 233 PII 30
Thread here.
( Last edited by Eug Wanker; Sep 17, 2003 at 08:45 PM. )
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 17, 2003, 09:53 PM
 
Here are the actual scores:
Code:
Apple PowerMac G5 Dual 2.0 GHz G5 1 GB DRAM � 4*256 MB PC3200 CL=3 1 160 GB Serial ATA hard disk ATI Radeon 9600 Pro OS X 10.2.7 (G5) Photoshop 7.0.1 (w/ G5 Update installed) Energy Savings Settings: High Performance 1 90 Degree Rotate .3 .3 .3 2 9 Degree Rotate 1.3 1.3 1.3 3 .9 Degree Rotate 1.2 1.2 1.2 4 1 Gaussian .3 .4 .3 5 3.7 Gaussian 1.2 1.2 1.2 6 87 Gaussian 1.8 1.8 1.7 7 1 pix USM .5 .5 .5 8 3.7 pix USM 1.4 1.4 1.4 9 10 pix USM 1.7 1.6 1.6 10 Despeckle .4 .4 .4 11 RGB-CMYK 1.7 1.7 1.7 12 60% Reduction .4 .4 .4 13 LensFlare 3.4 3.3 3.3 14 Color Halftone 2.8 2.8 2.8 15 NTSC Color 3.2 3.2 3.2 16 Accent Edge 12.1 12.1 12.2 17 Pointilize 10.7 10.7 10.7 18 Watercolor 26.8 26.7 26.7 19 Polarize Coordinates 2.3 2.2 2.2 20 Radial Blur 20.4 20.4 20.3 21 Lighting 1.7 1.6 1.6
Macs tend to do much better on the first 12 tests than the last 9, compared to Windows boxes. Fortunately for Macs, the first 12 are what are used more often. (Who uses pointilize anyway?)


Indeed, this is why the G5 seems to do so well in Apple's tests:




Here are the actual tests used by Apple (600 MB file):

     
ratlater
Grizzled Veteran
Join Date: Nov 2000
Location: Seattle, WA, USA
Status: Offline
Reply With Quote
Sep 17, 2003, 10:13 PM
 
Save me jebus that's fast! Wow, I'm very impressed. The urge to buy a new dual G5 just tripled.

-matt
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 17, 2003, 11:05 PM
 
I'm thinking a 1.6 GHz 1 MB L2 cache G5 90 nm PowerBook with DDR333 a year from now would be nice.
     
Socially Awkward Solo
Professional Poster
Join Date: Jul 2002
Location: Hanging on the wall at Jabba's Palace
Status: Offline
Reply With Quote
Sep 18, 2003, 12:30 AM
 
Shucks. Stupid filters again that I am convinced are used by nobody except for benchmarking and people with bad design taste.

Good to know anyway. I will get a Dual 2GHZ at rev b.

P.S. anyone caught using a lens flair for any reason should be dragged into the streets and shot.

"Laugh it up, fuzz ball!"
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 12:39 AM
 
Originally posted by Socially Awkward Solo:
Shucks. Stupid filters again that I am convinced are used by nobody except for benchmarking and people with bad design taste.

Good to know anyway. I will get a Dual 2GHZ at rev b.

P.S. anyone caught using a lens flair for any reason should be dragged into the streets and shot.
The first 12 filters in the test are commonly used (and those are what the G5 does best on). The rest are mostly useless though.

P.S. Microsoft used the lens flare for an Amped (Xbox) picture. Microsoft was dragged out into the net and flogged.

     
Socially Awkward Solo
Professional Poster
Join Date: Jul 2002
Location: Hanging on the wall at Jabba's Palace
Status: Offline
Reply With Quote
Sep 18, 2003, 01:59 AM
 
Originally posted by Eug:
The first 12 filters in the test are commonly used (and those are what the G5 does best on). The rest are mostly useless though.

P.S. Microsoft used the lens flare for an Amped (Xbox) picture. Microsoft was dragged out into the net and flogged.
Ya the first 12 are fine but we are really talking less then 10 seconds on ANY system, so yippee.

I remember that lens flare whooha, serves them right

"Laugh it up, fuzz ball!"
     
Shaktai
Mac Elite
Join Date: Jan 2002
Location: Mile High City
Status: Offline
Reply With Quote
Sep 18, 2003, 04:53 AM
 
What is really amazing is that is without either the OS or the test app being optimized. It will be interesting to see how much Panther and optimized real world apps can improve on that.
     
REZN8
Fresh-Faced Recruit
Join Date: Sep 2003
Location: The land of the happy.
Status: Offline
Reply With Quote
Sep 18, 2003, 06:55 AM
 
Those scores look amazing. What I'd like to see though, are test renders done in something like LightWave 3D.
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 07:29 AM
 
So somebody else has run the bench now and confirmed the results. His normalized score is even higher, at 555.

What is really amazing is that is without either the OS or the test app being optimized. It will be interesting to see how much Panther and optimized real world apps can improve on that.
The app has been optimized for the G5, with the G5 plugin. And people are reporting that PSBench doesn't get faster in Panther, at least on a G4. Who knows on a G5 though.

Ya the first 12 are fine but we are really talking less then 10 seconds on ANY system, so yippee.
Yeah, but this is a 50 MB file. Takes a lot longer on a 400 MB file. (Apple uses a 600 MB file for its Photoshop bakeoff.)

However, it does goes to show you that a dual 2.0 is overkill for people who work with such files or who don't work with high-end hardcore number crunching apps, etc. And this is something I've been saying all along. For most (but not all) MacNN types, a single 1.8 is fine, esp. if cost is a concern. A lot of video and 3D stuff WOULD benefit from dual processors though.
( Last edited by Eug; Sep 18, 2003 at 07:47 AM. )
     
michaelb
Mac Elite
Join Date: Oct 2000
Location: Australia
Status: Offline
Reply With Quote
Sep 18, 2003, 08:30 AM
 
Originally posted by Socially Awkward Solo:
anyone caught using a lens flair for any reason should be dragged into the streets and shot.
Originally posted by Eug:
Microsoft used the lens flare for an Amped (Xbox) picture. Microsoft was dragged out into the net and flogged.
What a bunch of friggin' drama queens! The image isn't that great anyway, but perfectly suited to its target audience.
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 08:34 AM
 
Originally posted by michaelb:
What a bunch of friggin' drama queens! The image isn't that great anyway, but perfectly suited to its target audience.
Not quite. Not only was the lens flare fake, the resolution of the guy was much higher than it was supposed to be. Thus, the picture was not representative of the game.

For this reason, it was the gaming-oriented sites and forums that criticized Microsoft on this one.
     
proudestmonkey
Fresh-Faced Recruit
Join Date: Sep 2003
Location: upper midwest
Status: Offline
Reply With Quote
Sep 18, 2003, 09:07 AM
 
I would like to know how much memory each system had when tested. MacWorld's recent review of the new systems was flawed in my opinion because they used too little memory (512MB) on each system. That may be the way the system ships, but if you are going to get a G5, you will get more RAM (from somewhere cheaper than Apple). Testing a G5 with 512MB is like testing a porche using regular gasoline. Tests should be conducted with no less than 2 Gigs of Memory.
I don't just know the market, I am the
market.
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 10:07 AM
 
Originally posted by proudestmonkey:
I would like to know how much memory each system had when tested. MacWorld's recent review of the new systems was flawed in my opinion because they used too little memory (512MB) on each system. That may be the way the system ships, but if you are going to get a G5, you will get more RAM (from somewhere cheaper than Apple). Testing a G5 with 512MB is like testing a porche using regular gasoline. Tests should be conducted with no less than 2 Gigs of Memory.
With the above tests, 512 MB is fine, since only a 50 MB file was used. There is no speed difference going to 1 GB. Indeed, the 547 score was with a machine with 1 GB, and the new 555 score was with a machine with only 512 MB.

With the Macaddict tests however, probably about 768 MB - 1 GB would have been necessary, because they used a 115 MB file. Going from 1 GB to 2 GB would make no difference even in the Macaddict tests. Too bad they used only 512 MB though.
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 10:15 AM
 
Test by test speed comparison posted at Chaosmint.com.
     
Busemann
Mac Elite
Join Date: Feb 2003
Status: Offline
Reply With Quote
Sep 18, 2003, 10:38 AM
 
Originally posted by Eug:
The app has been optimized for the G5, with the G5 plugin. And people are reporting that PSBench doesn't get faster in Panther, at least on a G4. Who knows on a G5 though.
What's amazing though is that Photoshop CS will be an order of magnitude faster than PS7 on the G5.
     
Cadaver
Addicted to MacNN
Join Date: Jan 2003
Location: ~/
Status: Offline
Reply With Quote
Sep 18, 2003, 11:47 AM
 
Originally posted by Eug:

However, it does goes to show you that a dual 2.0 is overkill for people who work with such files or who don't work with high-end hardcore number crunching apps, etc. And this is something I've been saying all along. For most (but not all) MacNN types, a single 1.8 is fine, esp. if cost is a concern. A lot of video and 3D stuff WOULD benefit from dual processors though.
Not to hijack the thread, but Eug - while a Dual 2G may be overkill (for most users) today, it may not be 18 months from now. And unlike my last two computer purchases which I found too slow less than two years later, I plan on the G5 lasting at least 3 years. And who knows, in 18 months, a dual 2G could be Apple's midline machine...
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 18, 2003, 11:51 AM
 
Interesting, the dual G5 2.0 vs P4 3 GHz numbers are very similar to Apple's numbers if you look at only the first 12 tests (which are what Apple emphasized in its bench):

Code:
# Filter P4 3.06 HT G5 Dual 2.0 G5 vs P4 1 Rotate 90 0.2 0.2 1.0X 2 Rotate 9 2.7 1.3 2.1X 3 Rotate .9 2.5 1.2 2.1X 4 Gaussian Blur 0.8 0.4 2.0X 5 Gaussian Blur 2.0 1.2 1.7X 6 Gaussian Blur 2.3 1.7 1.4X 7 Unsharp 50/1/0 0.9 0.6 1.5X 8 Unsharp 50/3/7/0 2.1 1.4 1.5X 9 Unsharp 50/10/5 2.2 1.6 1.4X 10 Despeckle 2.2 0.4 5.5X 11 RGB-CMYK 7.3 1.7 4.3X 12 Reduce Size 0.9 0.4 2.3X 13 Lens Flare 2.5 3.3 0.8X 14 Color Halftone 2.2 3.1 0.7X 15 NTSC Colors 2.4 2.6 0.9X 16 Accented Edges 10.9 11.6 0.9X 17 Pointillize 12.1 10.8 1.1X 18 Water Color 26.4 25.4 1.0X 19 Polar Coordinates 7.0 2.2 3.2X 20 Radial Blur 33.1 20.4 1.6X 21 Lighting Effects 1.9 1.6 1.2X Average 1-12: 26.8/12 = 2.2X Average 13-21: 11.4/9 = 1.3X Average overall: 38.2/21 = 1.8X


Not to hijack the thread, but Eug - while a Dual 2G may be overkill (for most users) today, it may not be 18 months from now. And unlike my last two computer purchases which I found too slow less than two years later, I plan on the G5 lasting at least 3 years. And who knows, in 18 months, a dual 2G could be Apple's midline machine...
Like I said, it really depends on what you do. I think for most Photoshoppers, a single 1.8 with 2 GB RAM or more would probably be OK in 3 years. For hardcore 3D it won't though. In fact, a single 1.8 is not even all that great right now for that.

But if you have the extra $550 right now then great, go for it.
     
The Ginger Rat
Mac Enthusiast
Join Date: May 2001
Location: BC, Canada
Status: Offline
Reply With Quote
Sep 18, 2003, 12:09 PM
 
You know, $550 is a fairly small premium for a huge jump in performance. Having seen these results, I think it becomes clearer that Apple priced its products to steer buyers to the duals.
     
SalBaker
Junior Member
Join Date: Feb 2003
Status: Offline
Reply With Quote
Sep 18, 2003, 12:41 PM
 
Those of us who use Macs for music creation/production and use dozens of huge sampled instruments and many real-time effects plugins are waiting to see if the dual G5 is even enough now!
     
chrisutley
Grizzled Veteran
Join Date: Mar 2003
Status: Offline
Reply With Quote
Sep 18, 2003, 12:47 PM
 
Originally posted by SalBaker:
Those of us who use Macs for music creation/production and use dozens of huge sampled instruments and many real-time effects plugins are waiting to see if the dual G5 is even enough now!
What are your options if it's not?
MacBook and iMac Core 2 Duo 24"
     
SalBaker
Junior Member
Join Date: Feb 2003
Status: Offline
Reply With Quote
Sep 18, 2003, 02:40 PM
 
Originally posted by chrisutley:
What are your options if it's not?
Rev. B
     
chris v
Addicted to MacNN
Join Date: Jan 2001
Location: The Sar Chasm
Status: Offline
Reply With Quote
Sep 18, 2003, 03:18 PM
 
Originally posted by SalBaker:
Those of us who use Macs for music creation/production and use dozens of huge sampled instruments and many real-time effects plugins are waiting to see if the dual G5 is even enough now!
Someone on the AI forums was claiming something like 63 Platinum Verbs in Logic on a single 1.8. I'm thinking a Dual 2.0 is going to be able to handle about all the audio plug-ins you can throw at it. (by comparison, think a Dual 1.0 G4 chokes at about 25)

CV

When a true genius appears in the world you may know him by this sign, that the dunces are all in confederacy against him. -- Jonathan Swift.
     
Severed Hand of Skywalker
Addicted to MacNN
Join Date: Apr 2001
Location: The bottom of Cloud City
Status: Offline
Reply With Quote
Sep 18, 2003, 03:50 PM
 
Originally posted by Eug:
The app has been optimized for the G5, with the G5 plugin. And people are reporting that PSBench doesn't get faster in Panther, at least on a G4. Who knows on a G5 though.
Only some of photoshops actions are optimized with that plugin. They are going to have to have it fully compatible in the next version.

Apparently the G5's are MUCH faster when running 10.3 so I am sure things will get even better.

"Ahhhhhhhhhhhhhhhh"
     
rupis33
Fresh-Faced Recruit
Join Date: Jul 2003
Status: Offline
Reply With Quote
Sep 18, 2003, 06:22 PM
 
Originally posted by chris v:
Someone on the AI forums was claiming something like 63 Platinum Verbs in Logic on a single 1.8. I'm thinking a Dual 2.0 is going to be able to handle about all the audio plug-ins you can throw at it. (by comparison, think a Dual 1.0 G4 chokes at about 25)

CV
I am a audio engineer, waiting to order possible a dual G5 I would love to know about a forum that is geared more for audio discussion. Thanks

Luquasious
     
olePigeon
Clinically Insane
Join Date: Dec 1999
Status: Offline
Reply With Quote
Sep 18, 2003, 06:35 PM
 
My 233 G3 with a 16MB VooDoo 3 is still chugging along for Photoshop. I haven't dared install any version of Photoshop past 5.5 yet, don't know how my Mac will handle it.

I'm mostly using airbrush, lasso, etc. for touchups on pictures. Though it would be nice if I could scroll and not have to wait for the grid to fill in.

Illustrator is as snappy as ever. I've run it on a G3 and a dual 533 G4 and didn't see much of a difference in speed until I start using gradients.

A dual G5 with 16GB of RAM... holy cow. I can't imagine. Well, I can, but I'll probably never get to own one.
"…I contend that we are both atheists. I just believe in one fewer god than
you do. When you understand why you dismiss all the other possible gods,
you will understand why I dismiss yours." - Stephen F. Roberts
     
jim overholser
Fresh-Faced Recruit
Join Date: Jul 2001
Status: Offline
Reply With Quote
Sep 18, 2003, 08:57 PM
 
Emapple logic:sonikmatter

OSX Audio

Unicornation

sound on sound

hope this helps for audio forums, sorry for the thread hijak.
     
pliny
Mac Elite
Join Date: Apr 2001
Location: under about 12 feet of ash from Mt. Vesuvius
Status: Offline
Reply With Quote
Sep 18, 2003, 09:20 PM
 
All those naysayers who said this and that about Apple's claims can kiss Apple's a.s.s..The tests posted here simply confirm what Apple's been saying for months and what the real weird slice of the PC weirdos demographic insisted, in very strange shrills, was lies.

i look in your general direction
     
moki
Ambrosia - el Presidente
Join Date: Sep 2000
Location: Rochester, NY
Status: Offline
Reply With Quote
Sep 19, 2003, 04:46 AM
 
Originally posted by rupis33:
I am a audio engineer, waiting to order possible a dual G5 I would love to know about a forum that is geared more for audio discussion. Thanks

Luquasious
http://www.osxaudio.com/
Andrew Welch / el Presidente / Ambrosia Software, Inc.
     
cacarr1
Mac Enthusiast
Join Date: Sep 2000
Location: Rock Island, IL
Status: Offline
Reply With Quote
Sep 19, 2003, 04:56 AM
 
Wow, I have a pretty slow machine!

I feel so inferior...
Uva uvam vivendo varia fit - Augustus McCrae
     
Eug
Clinically Insane
Join Date: Dec 2000
Location: Caught in a web of deceit.
Status: Offline
Reply With Quote
Sep 19, 2003, 09:02 AM
 
The top score for the dual G5 2.0 is now 555 (somebody else ran the test), and the top score for the dual Xeon 3.06 is now 490. Also, the dual Xeon 3.06 is a stock one, and no longer an overclocked dual 2.4 machine.

Also, the tests were done on the G5 with bus slewing turned on (Energy Saver - Automatic) just for fun. It turns out the top speeds are still the same, but the results from test to test might be more variable... Because the score depends on averages of three runs, it dropped from 547 to 497. Some of that is due to chance, but it does seem that some of the 50 point drop is likely due to the bus slewing itself.

OTOH, Cinebench scores with bus slewing yielded no performance difference.

So in other words, for most tasks it seems that Automatic doesn't make a huge difference in performance, but it can affect it. If it does affect performance then likely it's by single digit percentages.
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 20, 2003, 08:59 AM
 
     
workerbee
Mac Elite
Join Date: Jul 2001
Location: Switzerland
Status: Offline
Reply With Quote
Sep 20, 2003, 01:40 PM
 
FYI, the german tech mag c't tested the dual 2GHz G5 in its latest edition.

Here's a few numbers:
Cinebench 2003 Rendering (more = better)
2x 2HGz G5: 504
1x 1GHz G4: 92
2x Xeon 3.06 GHz: 655
1x Athlon 64FX 2.2 GHz: 310
1x P4 3.2 GHz: 380

Photoshop 7 (less = faster)
2x 2HGz G5: 278 s
1x 1GHz G4: 796 s
2x Xeon 3.06 GHz: 287 s
1x Athlon 64FX 2.2 GHz: 337 s
1x P4 3.2 GHz: 362 s

Mathematica 5 (less = faster)
2x 2HGz G5: 1021 s
1x 1GHz G4: 2023 s
2x Xeon 3.06 GHz: 725 s
1x Athlon 64FX 2.2 GHz: 553 s
1x P4 3.2 GHz: 678 s

FileMaker 5.5 (less = faster)
2x 2HGz G5: 82 s
1x 1GHz G4: 147 s
2x Xeon 3.06 GHz: 70 s
1x Athlon 64FX 2.2 GHz: 46 s
1x P4 3.2 GHz: 62 s

MP3 encoding (less = faster)
2x 2HGz G5: 98 s
1x 1GHz G4: 284 s
2x Xeon 3.06 GHz: 68 s
1x Athlon 64FX 2.2 GHz: 89 s
1x P4 3.2 GHz: 91 s

MPEG-4 transcoding (less = faster)
2x 2HGz G5: 42 s
1x 1GHz G4: 85 s
2x Xeon 3.06 GHz: 35 s
1x Athlon 64FX 2.2 GHz: 41 s
1x P4 3.2 GHz: 39 s

UT 2003 Asbestos flyby 1024x768
2x 2HGz G5: 67 fps
1x 1GHz G4: 33 fps
2x Xeon 3.06 GHz: 197 fps
1x Athlon 64FX 2.2 GHz: 203 fps
1x P4 3.2 GHz: 203 fps

So yes, it's true, we have a new winner... for Photoshop. (Actually, a 2 x Opteron 2 GHz scored 275 in Photoshop 7, but I was too lazy to transcribe all of the tests). The Mathematica result is strange, as a single 1.8 GHz G5 scored 1158 -- is Mathematica 5 not MP-aware?

The 1GHz G4 results make me rethink whether to really buy that new 17" PowerBook. The G5 PowerBook should be just so much faster...
MBP 15" 2.33GHz C2D 3GB 2*23" ACD
     
real
Grizzled Veteran
Join Date: May 2001
Location: Ca
Status: Offline
Reply With Quote
Sep 20, 2003, 01:58 PM
 
Originally posted by Socially Awkward Solo:

P.S. anyone caught using a lens flair for any reason should be dragged into the streets and shot. [/B]

What if they add some grain to the flare would that be ok. and what about a drop shadow.

have a good one Solo.
With some loud music + a friend to chat nearby you can get alot done. - but jezz, I'd avoid it if I had the choice---- If only real people came with Alpha Channels.......:)
AIM:xflaer
deinterlaced.com
     
Immortal K-Mart Employee
Mac Elite
Join Date: Sep 2001
Location: Folding customer returned size 52 underwear.
Status: Offline
Reply With Quote
Sep 20, 2003, 02:06 PM
 
Originally posted by real:
What if they add some grain to the flare would that be ok. and what about a drop shadow.

have a good one Solo.
Drop shadows are acceptable on some things as long as it is very subtle.

{v2.3 Now Jesus free}
Religions are like farts: yours is good, the others always stink.
     
workerbee
Mac Elite
Join Date: Jul 2001
Location: Switzerland
Status: Offline
Reply With Quote
Sep 20, 2003, 02:19 PM
 
Originally posted by Immortal K-Mart Employee:
Drop shadows are acceptable on some things as long as it is very subtle.
Like every window in OS X?
MBP 15" 2.33GHz C2D 3GB 2*23" ACD
     
Immortal K-Mart Employee
Mac Elite
Join Date: Sep 2001
Location: Folding customer returned size 52 underwear.
Status: Offline
Reply With Quote
Sep 20, 2003, 02:43 PM
 
Originally posted by workerbee:
Like every window in OS X?
Yes, I like em.

{v2.3 Now Jesus free}
Religions are like farts: yours is good, the others always stink.
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 20, 2003, 04:51 PM
 
     
RooneyX
Mac Elite
Join Date: Mar 2003
Status: Offline
Reply With Quote
Sep 20, 2003, 06:09 PM
 
Originally posted by workerbee:
FYI, the german tech mag c't tested the dual 2GHz G5 in its latest edition.

Here's a few numbers:
Cinebench 2003 Rendering (more = better)
2x 2HGz G5: 504
1x 1GHz G4: 92
2x Xeon 3.06 GHz: 655
1x Athlon 64FX 2.2 GHz: 310
1x P4 3.2 GHz: 380

Photoshop 7 (less = faster)
2x 2HGz G5: 278 s
1x 1GHz G4: 796 s
2x Xeon 3.06 GHz: 287 s
1x Athlon 64FX 2.2 GHz: 337 s
1x P4 3.2 GHz: 362 s

Mathematica 5 (less = faster)
2x 2HGz G5: 1021 s
1x 1GHz G4: 2023 s
2x Xeon 3.06 GHz: 725 s
1x Athlon 64FX 2.2 GHz: 553 s
1x P4 3.2 GHz: 678 s

FileMaker 5.5 (less = faster)
2x 2HGz G5: 82 s
1x 1GHz G4: 147 s
2x Xeon 3.06 GHz: 70 s
1x Athlon 64FX 2.2 GHz: 46 s
1x P4 3.2 GHz: 62 s

MP3 encoding (less = faster)
2x 2HGz G5: 98 s
1x 1GHz G4: 284 s
2x Xeon 3.06 GHz: 68 s
1x Athlon 64FX 2.2 GHz: 89 s
1x P4 3.2 GHz: 91 s

MPEG-4 transcoding (less = faster)
2x 2HGz G5: 42 s
1x 1GHz G4: 85 s
2x Xeon 3.06 GHz: 35 s
1x Athlon 64FX 2.2 GHz: 41 s
1x P4 3.2 GHz: 39 s

UT 2003 Asbestos flyby 1024x768
2x 2HGz G5: 67 fps
1x 1GHz G4: 33 fps
2x Xeon 3.06 GHz: 197 fps
1x Athlon 64FX 2.2 GHz: 203 fps
1x P4 3.2 GHz: 203 fps

So yes, it's true, we have a new winner... for Photoshop. (Actually, a 2 x Opteron 2 GHz scored 275 in Photoshop 7, but I was too lazy to transcribe all of the tests). The Mathematica result is strange, as a single 1.8 GHz G5 scored 1158 -- is Mathematica 5 not MP-aware?

The 1GHz G4 results make me rethink whether to really buy that new 17" PowerBook. The G5 PowerBook should be just so much faster...
Looks like the Xeon is still the fastest CPU overall even though it runs at 1Ghz faster than the G5. Those Unreal 2003 benchmarks make the G5 look very poor. Was it the graphics card?
     
blackwind
Junior Member
Join Date: Oct 2002
Status: Offline
Reply With Quote
Sep 20, 2003, 10:01 PM
 
How did the G5 score 504 in the Cinebench 2003 raytrace?

The best score we seem to have had for the Dual 2-GHz G5 was 402 with the unoptimized Cinebench 2003.
     
dli537
Fresh-Faced Recruit
Join Date: Mar 2001
Location: Baltimore, md, USA
Status: Offline
Reply With Quote
Sep 20, 2003, 10:02 PM
 
Originally posted by workerbee:
FYI, the german tech mag c't tested the dual 2GHz G5 in its latest edition.

Here's a few numbers:
.....

Mathematica 5 (less = faster)
2x 2HGz G5: 1021 s
1x 1GHz G4: 2023 s
2x Xeon 3.06 GHz: 725 s
1x Athlon 64FX 2.2 GHz: 553 s
1x P4 3.2 GHz: 678 s
.....


So yes, it's true, we have a new winner... for Photoshop. (Actually, a 2 x Opteron 2 GHz scored 275 in Photoshop 7, but I was too lazy to transcribe all of the tests). The Mathematica result is strange, as a single 1.8 GHz G5 scored 1158 -- is Mathematica 5 not MP-aware?

The 1GHz G4 results make me rethink whether to really buy that new 17" PowerBook. The G5 PowerBook should be just so much faster...
Current version of Mathematica 5 is not optimized for G5. They don't have a date for the release yet. Here is the link from mathematica website http://www.wolfram.com/news/g5development.html
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 20, 2003, 10:05 PM
 
Originally posted by blackwind:
How did the G5 score 504 in the Cinebench 2003 raytrace?

The best score we seem to have had for the Dual 2-GHz G5 was 402 with the unoptimized Cinebench 2003.
Is it a different setting of the bench? The dual Xeon score is also unusually high.
Originally posted by dli537:
Current version of Mathematica 5 is not optimized for G5. They don't have a date for the release yet. Here is the link from mathematica website http://www.wolfram.com/news/g5development.html
Yeah, Cinebench isn't optimized yet either.
     
Groovy
Mac Enthusiast
Join Date: Apr 2001
Status: Offline
Reply With Quote
Sep 20, 2003, 10:31 PM
 
Originally posted by RooneyX:
Looks like the Xeon is still the fastest CPU overall even though it runs at 1Ghz faster than the G5. Those Unreal 2003 benchmarks make the G5 look very poor. Was it the graphics card?
huh? How did you get that? What those numbers show is that even a slightly optimized app is fastest on G5.
The photoshop G5 plugin only opts a very small part of the whole and it already shows a huge improvement.
Chris Cox at adobe flat out said the G5 trounces all other boxes and that 10.3 and PS 8 truly kick ass.


UT2003 is big time opt for PC side. For mac not only is it not opt for G5 but it is a port and not as much
time/money is spent tweaking it.


-
     
RooneyX
Mac Elite
Join Date: Mar 2003
Status: Offline
Reply With Quote
Sep 21, 2003, 12:16 AM
 
Originally posted by Groovy:
huh? How did you get that?
-
Erm, because most of the benchmarks were in favor of the Xeon?

The Athlon 64 also does amazingly well for a single CPU model with nothing written for it, no Photoshop plugin or anything.

And the difference in Unreal scores can't be down the optimization only. 130FPS difference from optimization? I don't think so. More like Windows simply handles games better not matter what the game is. It's more likely the maturity of Windows' media and gaming APIs than the CPU. That allows games to milk the hardware more effectively. UT2K3 is pretty well coded for the Mac.
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 21, 2003, 12:48 AM
 
Originally posted by RooneyX:
And the difference in Unreal scores can't be down the optimization only. 130FPS difference from optimization? I don't think so. More like Windows simply handles games better not matter what the game is. It's more likely the maturity of Windows' media and gaming APIs than the CPU. That allows games to milk the hardware more effectively.
Barefeats had faster speeds, by a factor of 2 or more (but still much slower than Intel machines).
UT2K3 is pretty well coded for the Mac.
I'm not so sure. This is not really Mac specific, but the game isn't really even dual aware AFAIK. And it wasn't written as an OpenGL game. It runs thru OpenGL on the Mac, no? (I don't have it.)
     
RooneyX
Mac Elite
Join Date: Mar 2003
Status: Offline
Reply With Quote
Sep 21, 2003, 01:10 PM
 
Originally posted by Eug Wanker:
Barefeats had faster speeds, by a factor of 2 or more (but still much slower than Intel machines). I'm not so sure. This is not really Mac specific, but the game isn't really even dual aware AFAIK. And it wasn't written as an OpenGL game. It runs thru OpenGL on the Mac, no? (I don't have it.)
It's OpenGL. But take a look at the single CPU Athlon score. The CPU runs at 2.2Ghz, the G5 at 2Ghz with a faster bus. I don't know what graphics cards were used but UT2K3 isn't optimised for either CPU and the difference is astounding.

The Athlon 64 looks like the fastest CPU overall. It has a tremendous single CPU score in Photoshop (without any optimised plugin), has the fastest Filemaker score, tremendous mp3 and mpeg 4 scores and the fastest Unreal scores.

I wonder what kind of fan the CPU has compared to the G5.
     
workerbee
Mac Elite
Join Date: Jul 2001
Location: Switzerland
Status: Offline
Reply With Quote
Sep 21, 2003, 02:52 PM
 
Originally posted by blackwind:
How did the G5 score 504 in the Cinebench 2003 raytrace?

The best score we seem to have had for the Dual 2-GHz G5 was 402 with the unoptimized Cinebench 2003.
They have -- and did use -- a preview version of the G5-optimized Cinebench.

F�r Anwendungen, die sowohl in Mac-OS- als auch Windows-Versionen zu haben sind, gibt es kaum vergleichbare Leistungs-Analyse-Werkzeuge von der Stange. Eine l�bliche Ausnahme ist der 3D-Renderer Cinema 4D von Maxon, zu dem es mit dem Cinebench 2003 einen Benchmark in Mac- und PC-Version gibt, der nicht nur bereits das Hyper-Threading aktueller Intel-Prozessoren unterst�tzt, sondern mit einer uns zur Verf�gung stehenden Vorabversion auch bereits f�r den G5 optimiert ist.
MBP 15" 2.33GHz C2D 3GB 2*23" ACD
     
workerbee
Mac Elite
Join Date: Jul 2001
Location: Switzerland
Status: Offline
Reply With Quote
Sep 21, 2003, 02:59 PM
 
Originally posted by RooneyX:
The Athlon 64 looks like the fastest CPU overall.
That is also how the testers at c't summed it up.
The Athlon 64 will be a very very fast CPU, but it is not yet entirely operational, e.g. the RAM specified by AMD apparently is produced by exactly 1 company and does not adhere to standards (yet).
I'd get a Dual G5 anytime over any AMD or Intel box. Especially in a PowerBook
MBP 15" 2.33GHz C2D 3GB 2*23" ACD
     
RooneyX
Mac Elite
Join Date: Mar 2003
Status: Offline
Reply With Quote
Sep 21, 2003, 03:59 PM
 
Originally posted by workerbee:
That is also how the testers at c't summed it up.
The Athlon 64 will be a very very fast CPU, but it is not yet entirely operational, e.g. the RAM specified by AMD apparently is produced by exactly 1 company and does not adhere to standards (yet).
I'd get a Dual G5 anytime over any AMD or Intel box. Especially in a PowerBook
Of course. The A64 doesn't run OSX so we don't get the fastest hardware once more. Still, Macs are up there where they should be. Shame the lowest end models still cost much more than a PC equivalent and have sh-t graphics cards.
     
Eug Wanker  (op)
Posting Junkie
Join Date: Jun 2003
Location: Dangling something in the water… of the Arabian Sea
Status: Offline
Reply With Quote
Sep 22, 2003, 12:23 AM
 
Originally posted by workerbee:
They have -- and did use -- a preview version of the G5-optimized Cinebench.

F�r Anwendungen, die sowohl in Mac-OS- als auch Windows-Versionen zu haben sind, gibt es kaum vergleichbare Leistungs-Analyse-Werkzeuge von der Stange. Eine l�bliche Ausnahme ist der 3D-Renderer Cinema 4D von Maxon, zu dem es mit dem Cinebench 2003 einen Benchmark in Mac- und PC-Version gibt, der nicht nur bereits das Hyper-Threading aktueller Intel-Prozessoren unterst�tzt, sondern mit einer uns zur Verf�gung stehenden Vorabversion auch bereits f�r den G5 optimiert ist.
I don't know if I believe this site. They are also claiming to have run SPEC on the G5 and getting scores with the 2.0 worse than what IBM "estimates" for the 1.8, and less for the int score than what Apple got with GCC (although this latter int bit wouldn't be hugely surprising).

I'm told the site is supposedly well respected, but I will remain a skeptic for now, until I see more detailed info.
( Last edited by Eug Wanker; Sep 22, 2003 at 01:03 AM. )
     
 
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 04:44 AM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,