---------- Original message ---------- Subject: Re: Which is a faster option? Date: Donnerstag, 10. September 2009N From: "ah...clem" <[email protected]> To: "G-Group" <[email protected]>
> On Sep 9, 2:34 pm, Len Gerstel <[email protected]> wrote: > > 1) The apps are not dual processor aware. Most apps are. Even VLC > > player is and will split the load across the 2 processors of a dual. > > > > 2) You are not doing anything with these systems while the apps are > > running. Since they do not appear to be dual processor aware, they > > are running at about 100% utilization on one processor, and the other > > is sitting near 0% I bet. > > actually, you lose. where do i go to collect? ;o) i wouldn't > presume to guess whether the apps are dual processor aware or if the > OS distributes the work, but if i view the activity monitor on the DP > machine while a single job is running, first one processor is 75% > maxed out while the other is near 25% activity, then it flip-flops and > the other is 75% maxed out while the first is near 25% activity. the > frequency of the flip-flops is about twice per second. if i run two > jobs simultaneously, both processors stay 100% maxed out > continuously. and as you might expect, on the SP machine, the > activity monitor shows the one CPU 100% maxed out continuously. > > differences in viewing video, as you describe, is (this time i'll > wager) more a function of the video card processor speed and memory, > not the main CPU processor speed. it really surprises me how often > folks need reminded that a chain is only as strong (fast) as it > weakest (slowest) link. why are CPUs are like quarterbacks . . . ? > they either get all the credit or all the blame, and usually deserve > neither. Like coding a video, the decoding of it is also mostly CPU work at first, the decoded data gets then distributed to the graphics card via, as you say rightly, the memory over the system bus over to the graphics card. But that's normally not the bottleneck. The conventional PCI bus can transfer 133 MB/s of data. PC100 SDRAM has a peak transfer rate of 800 MB/s. According to http://www.blackmagic-design.com/de/support/detail.asp?techID=30 the the most out of a HD video (1920x1080, 10-bit, 60fps) would require 237 MB/s. Thus, for transferring decodeded HD material, you need a faster system bus. So, yes, it's not always the CPU. But I bet that a PowerPC below 1 GHz is not even able to decode a HD video with this resolution out of an MPEG-4/AVC or H.264 stream. Cheers, Andreas. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are a member of G-Group, a group for those using G3, G4, and G5 desktop Macs - with a particular focus on Power Macs. The list FAQ is at http://lowendmac.com/lists/g-list.shtml and our netiquette guide is at http://www.lowendmac.com/lists/netiquette.shtml To post to this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/g3-5-list -~----------~----~----~----~------~----~------~--~---
