Photo for Glenn Fleishman

Blog

Writing

What I Do

Biography

GlennLog

Turning technology from mumbo-jumbo into rich tasty gumbo

� New Apples Versus Comparable Dell | Main | Pro-ReplayTV Goes Forward �

August 15, 2002

Speed the Fruit

Unfortunately, according to Rob Art Morgan who has tested this, the new PowerMacs from Apple that use DDR (double data rate) memory like its Xserve rank-mount unit cannot access the memory any faster than the cheaper and slower SDRAM found in the previous system arch. A controller limits the data rate to 1 Gbps, while DDR could work more than twice as fast. Unfortunately, this makes mincemeat of the architecture, as it bus/memory-bounds 2D and 3D graphics and rendering.

In practical terms, my chart showing the Apple vs. Dell server comparison is now pointless: the new Macs can't compete with the same-price or cheaper Dells because the Macs are just as slow as the old Macs. You can boost the processor speed somewhat (jump to dual 1.25 GHz), but you can't extract nearly enough speed from the memory.

Posted by Glennf at August 15, 2002 7:41 PM

Trackback Pings

TrackBack URL for this entry:

Comments

Apple is very up front with this information. Everything on the system excep the CPU can access the DDR memory at full speed, but due to the way the G4 is currently implemented, the CPU can only talk to the System Controller at bus speed (see an excellent explanation of this at http://www.railheaddesign.com/). Still, with a 167 MHz bus, access to motherboard memory will be faster than ever.

And don't forget that the new Power Macs have 1 MB of level 3 cache per CPU -- and the top-end 1.25 GHz model will have twice that. When the CPU is working with the L1 and L2 cache, memory access is at full CPU speed. When it needs to access the L3 cache, there's a wide bus (128 bits) running at half CPU speed, or 433/500 MHz. It's only when the CPU needs to go to motherboard memory that the 1.33 GBps limitation on memory throughput comes into play.

The rest of the mobo support 2.67 GBps throughput, so things like DMA and virtual memory should be faster than on the old 133 MHz non-DDR machines.

Another issue not mentioned in your Dell comparison -- the Dell's throughput is 1.6 GBps per channel (which I assume means per CPU). That's only 20% better than the Mac (which I believe shares bandwidth with both CPUs, although this is outside my area of expertise).

The Intel Xeon processors have a smaller 512 KB cache per processor, which also means they'll have to access motherboard memory more frequently than a G4 with a 1 MB or 2 MB cache.

Anyhow, it's nonsense to state that your earlier comparison is nonsense just because the G4 CPU can't take full advantage of DDR memory. The G4 still has AltiVec to blow away the Xeon for Photoshop work. And it now uses less costly, readily available DDR memory.

But your comparison would be more fair if you put the 1 GHz model up against the Dell, since that would make prices much closer.

Posted by: Dan Knight at August 17, 2002 4:02 AM

Eh...his tests are fairly specious...I'd like to see some ones that stress the whole system, and look at disk I/O...also, the high end is now the middle, so that does make an ecconomic difference.

I definitely agree that the processor is starved, however. i just don't agree that the BreFeats folks were thorough, nor do I think it tells the whole story.

Posted by: mrmister at August 16, 2002 1:41 PM

May 2008
Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

Recent Entries

Archives


May 2008 | April 2008 | March 2008 | February 2008 | January 2008 | December 2007 | November 2007 | October 2007 | September 2007 | August 2007 | July 2007 | June 2007 | May 2007 | April 2007 | March 2007 | February 2007 | January 2007 | December 2006 | November 2006 | October 2006 | September 2006 | August 2006 | July 2006 | June 2006 | May 2006 | April 2006 | March 2006 | February 2006 | January 2006 | December 2005 | November 2005 | October 2005 | September 2005 | August 2005 | July 2005 | June 2005 | May 2005 | April 2005 | March 2005 | February 2005 | January 2005 | December 2004 | November 2004 | October 2004 | September 2004 | August 2004 | July 2004 | June 2004 | May 2004 | April 2004 | March 2004 | February 2004 | January 2004 | December 2003 | November 2003 | October 2003 | September 2003 | August 2003 | July 2003 | June 2003 | May 2003 | April 2003 | March 2003 | February 2003 | January 2003 | December 2002 | November 2002 | October 2002 | September 2002 | August 2002 | July 2002 | June 2002 | May 2002 | April 2002 | March 2002 | February 2002 | January 2002 | December 2001 | November 2001 | October 2001 |

Powered by Movable Type 3.33