-
Elderly CPU schools Whipper-snapper Quadcore
Trying to get a lot of rendering done for a surprise short deadline freelance gig, and discovered that my old system is way faster than the new box I put together this month…
Old system:
AMD Athlon 4600
2gb mem
XP ProMy old system:
AMD Phenom 9850
4Gb mem
Vista Home Premium x64Opening the same project file in Vegas 8 on both and rendering some hour long clips to WMV takes about an hour on the old system, but 3.5 to 4 hours on the new one. Looking at CPU usage, the old system is using 100% while the new one never goes over 40% usage.
Any idea of what tweaks I need to do? As far as I can see, I’ve got Vegas’ preferences the same between both systems. Disable multi-processor is not checked. I’d blame it on a Vista glitch if my third computer, a Vista x64 laptop, wasn’t using the whole CPU when I tested rendering the same files on it.
—
Elvis Deane!
The Apprentice Magician’s Guide to particleIllusion
Wilbur of Wumbaberry -
7 Replies
-
-
Indeed it is, a dual core Turion. Perhaps those extra two cores in the desktop have low self esteem because they can’t compete with Intel chips.
—
Elvis Deane!
The Apprentice Magician’s Guide to particleIllusion
Wilbur of Wumbaberry -
It is certainly perplexing. The reason the QuadCore is only at 40% is probably because the WMV encoder is only using two threads (one for video and one for audio) and therefore only two of the 4 CPUs. But the strange thing is that the QuadCore is rated at 2.5GHz and the DualCore is rated at 2.4GHz so even at two cores the render time should be the same.
Now I know that AMD always had a funny numbering scheme because their clock speed could not be compared one-for-one with Intel because their pipeline was shorter and therefore they ran faster at the same clock speed. If this is no longer true for their QuadCores it could be that a 2.4 AMD DualCore has faster processors than a 2.5 AMD QuadCore. I’m just guessing here.
~jr
http://www.johnrofrano.com
http://www.vasst.com -
The results are very strange though. Your new system configurations should have left your old PC in the dust.
Maybe if you install XP instead of Vista on your new System
you may see the improvements you were expecting….LOL,who knows.Steve Rhoden
Creative Director
TNX EFFECTS STUDIOS.
sample bits at
http://www.youtube.com/hentys -
The source video is an mp4 and I wonder if that has anything to do with it, as editing some HVX shot MXF clips using Raylight doesn’t have the same problems. Initially the render starts at breakneck speed, the first 10% go way faster than on the old system, but about then they slow to a crawl and start doing 2 or 3 frames per second.
With luck, I’ll never get an mp4 source file again.
—
Elvis Deane!
The Apprentice Magician’s Guide to particleIllusion
Wilbur of Wumbaberry -
Hi,
[Elvis Deane] “The source video is an mp4”
Do you have the exact same codec for mp4 installed on all systems?
Regards
K.L.
-
Maximum PC recently did some head-to-head tests between XP and Vista on otherwise identical hardware. They reported that in every test case the XP was averaging 25% faster than the Vista box. Their test suite includes a Photoshop render and a Premiere render.
They also tested memory and processors on otherwise identical computers. One to two cores was definitely faster in all their tests, four cores was a mixed bag because some of the test applications would only use two cores. (Even so, if you have four cores available, then Windows background processes could have a core to themselves). Memory had a very small difference from DDR2 to DDR3 except in Games, where the difference was substantial.
This matches my experience where I have two identical laptops except that one came with XP installed, and the other that was purchased a few months later had Vista. Everything on the Vista laptop runs slower than the same application on the XP laptop. Everything.
Reply to this Discussion! Login or Sign Up