toggle

AAPL Stock: 111.78 ( -0.87 )

Printed from http://www.macnn.com

Mac Pro up to 62% faster than quad-G5

updated 07:50 pm EDT, Thu August 10, 2006

Mac Pro real-world results

Announced earlier this week, users are already receiving their Mac Pro systems and offering some "real-world" insight to the performance. Barefeats.com has completed a series of benchmarks to show the performance of the new system. The test system is the 'base' Mac Pro featuring two dual-core Xeon processors running at 2.66GHz with 1GB of RAM priced at $2499 (new), while the test G5 system is a Quad PowerMac G5 system running at 2.5GHz--also with 1GB of RAM with a higher price of $3299 (new). The test covers both Universal and non-Universal tests of Applications including Final Cut Pro 5, Photoshop CS2, and CineBench 9.5. In the non-Universal Binary tests, the G5 was 20-37 percent faster than the Xeon system; however, when tested with Intel-compatible applications, the Mac Pro took a significant lead of between 35 and 62 percent in "real-world" tests and a 15 percent higher Cinebench CPU score. "We are impressed with the Mac Pro's performance. We'll be even more impressed when Adobe's Universal Binary apps appear in 'second quarter of 2007,'" the report said. [Graph included].




by MacNN Staff

POST TOOLS:

TAGS :

toggle

Comments

  1. e2Sync

    Joined: Dec 1969

    0

    linking to sources

    It would be appropriate I think to make that mention of barefeats an actual link.

  1. bedoughty

    Joined: Dec 1969

    0

    A link and context, pleas

    A second vote for an actual link to the test, as well as making sure readers understand that the pretty graph you included is the non-Universal Final Cut test. Otherwise people might assume some really inaccurate conclusions.

  1. jimothy

    Joined: Dec 1969

    0

    Another correction

    "In the non-Universal Binary tests, the G5 was 20-37 percent faster than the quad-G5 system."

    The second "G5" reference should say "Xeon," I presume?

  1. bfalchuk

    Joined: Dec 1969

    0

    another correction

    Actually, the FIRST G5 should be Xeon, not the second. I believe the point is that the Xeon is always faster, but not by as big of a margin. When you go on and read the next bit, it says that the margin increases, which implies that it was faster even with PPC code.

    That's VERY impressive considering the same isn't true for a Core Duo or Core Solo.

  1. mactalent

    Joined: Dec 1969

    0

    Subjective test

    Spec tests are fine but I tried out a Mac Pro Dual 2.66 at the Apple Store and it did not seem any faster than previous towers. Still got spinning beachballs switching between apps. Tried some Photoshop stuff too and it did not blow me away. Maybe 10.4.8 when it comes will help.

  1. bobolicious

    Joined: Dec 1969

    0

    Ooops...

    Quad-G5 up to 28% faster than Mac Pro & has critical issues running some high end software like ArchiCAD 10 in rosetta or in beta universal...

  1. LouZer

    Joined: Dec 1969

    0

    What???

    The new Macs are faster then the old macs????? I'm shocked! Shocked, I say! I would have just assumed that Apple would start releasing new computers that were slower then the previous models....

    And I always find it hillarious how it seems everyone is betting that once Adobe's apps get released, they'll somehow make all the new intel macs magically faster.

    And if you all can't find barefeats.com on your own, well...

  1. Toyin

    Joined: Dec 1969

    0

    Impressed

    I don't think people are surprised that the Quad Xeon is faster, but I'm surprised the the Quad-Xeon is ONLY 30% slower running Rosetta translated applications. I'm hoping that Merom shows the same performance gains. That would mean a 2.3ghz Merom Mac Book Pro will run emulated applications as fast as my 1.5ghz Powerbook.

  1. Terrin

    Joined: Dec 1969

    0

    Actually

    Actually if you go check out the Barefeats test the old G5 beats the Xeon in more then one test.

  1. Hobeaux

    Joined: Dec 1969

    0

    test hobbled

    The machine is being hobbled by the testers lack of memory, not only in total amount, but in quantity of DIMMs filling the cards.

    For best memory performance you need to fill all the slots on that riser card. As that machine only has two slots filled it's only getting half the throughput.

    "There are a couple of things you can do to maximize performance and minimize the cost of additional memory on your Mac Pro, and it starts with the number of FB-DIMMs you configure your system with. The Mac Pro ships with a default configuration of 2 x 512MB FB-DIMMs, unfortunately that means that you're only using two of the four available memory channels, cutting your peak theoretical memory bandwidth in half. You'll want to upgrade to at least four FB-DIMMs so that you can run in quad-channel mode, in the coming weeks we'll be running some tests to figure out exactly how much additional performance you'll gain by doing that and if it's noticeable or not." -Anandtech (http://anandtech.com/mac/showdoc.aspx?i=2811&p=5)

    Also, system performance for both machines increases with the total amount of RAM available. Installing 4GB or 8GB will make a both systems improve substantially, but the differences between them will be more evident.

    Lastly, if he's ordered 3rd-party RAM it's highly likely that the memory will run hotter than the Apple RAM (due to the lack of heat-sinks and using spreaders instead) and thus perform slower and more loudly as the system tries to cool them.

Login Here

Not a member of the MacNN forums? Register now for free.

toggle

Network Headlines

toggle

Most Popular

MacNN Sponsor

Recent Reviews

Dell AD211 Bluetooth speaker

For all of the high-priced, over-engineered Bluetooth speakers in the electronics market, there is still room for mass-market solution ...

VisionTek 128GB USB Pocket SSD

USB flash drives dealt the death blow to both the floppy and Zip drives. While still faster than either of the old removable media, sp ...

Kodak PixPro SL10 Smart Lens Camera

Smartphone imagery still widely varies. Large Megapixel counts don't make for a good image, and the optics in some devices are lackin ...

toggle

Most Commented