VideoHelp Forum




+ Reply to Thread
Page 3 of 3
FirstFirst 1 2 3
Results 61 to 67 of 67
  1. Member
    Join Date
    Mar 2005
    Location
    United States
    Search Comp PM
    I think disk I/O matters a great deal.

    I have two systems, a P4 2.8 with 15K SCSI disks and a P4/3.4 with an older WD Raptor, most other things the same. The former is convincingly faster at, say, cce encodes of large huffyuv files to mpeg2, despite the CPU deficit.
    Quote Quote  
  2. Member
    Join Date
    Jan 2002
    Location
    Not the end of the world!
    Search Comp PM
    Originally Posted by johns0
    I do service calls and i check the pentium 4 temps on 2gig hz and up and they all are over 50c.
    My P4 2.8C @ 3.2 overclock with a stock Intel cooler runs @ 36C at full load with 2 panaflo 28cfm case fans.
    Quote Quote  
  3. Member edDV's Avatar
    Join Date
    Mar 2004
    Location
    Northern California, USA
    Search Comp PM
    The only way to compare AMD vs P4 is from a benchmark using the same software you intend to use.

    AMD tends to have more raw CPU power for the buck.
    Video transcoder/encoders are often optimized to Intel's MMC/SSE2/hyperthreading that tips advantage to Intel for those specific application areas (e.g. long encoding or effects rendering tasks).

    Sometimes AMD wins sometimes P4 wins when normalized for cost.
    Recommends: Kiva.org - Loans that change lives.
    http://www.kiva.org/about
    Quote Quote  
  4. Member edDV's Avatar
    Join Date
    Mar 2004
    Location
    Northern California, USA
    Search Comp PM
    Originally Posted by swiego
    I think disk I/O matters a great deal.

    I have two systems, a P4 2.8 with 15K SCSI disks and a P4/3.4 with an older WD Raptor, most other things the same. The former is convincingly faster at, say, cce encodes of large huffyuv files to mpeg2, despite the CPU deficit.

    I've seen where having separate HDD for OS, capture file to RAM and RAM to TMP file can speed renders. This is mostly limited by disk controller contention and disk access times. CPU power still dominantes. The HDD/RAM system just needs to feed the CPU efficiently
    Recommends: Kiva.org - Loans that change lives.
    http://www.kiva.org/about
    Quote Quote  
  5. Member rhegedus's Avatar
    Join Date
    Sep 2002
    Location
    on the jazz
    Search Comp PM
    Originally Posted by swiego
    I think disk I/O matters a great deal.

    I have two systems, a P4 2.8 with 15K SCSI disks and a P4/3.4 with an older WD Raptor, most other things the same. The former is convincingly faster at, say, cce encodes of large huffyuv files to mpeg2, despite the CPU deficit.
    Surely this has minimal effect - most disk systems (barring floppy) will write faster than a CPU can kick out a video file.

    That said, I have noticed a slight increase in speed if the source and target drives are on separate controllers.
    Regards,

    Rob
    Quote Quote  
  6. Member
    Join Date
    Mar 2005
    Location
    United States
    Search Comp PM
    Sorry, I should have clarified, I didn't mean I/O throughput, I meant I/O efficiency.

    It's a foregone conclusion that high I/O can burn CPU clock cycles, and some storage technologies are more CPU-efficient than others.
    Quote Quote  
  7. Member .Kal-El's Avatar
    Join Date
    Jan 2006
    Location
    United States
    Search Comp PM
    You guys forget that bus speeds and cache come into play. Intel's do run hotter, I have a 3.2 Prescott and it runs hella hot. I also have a AMD Barton 3200XP that runs a little cooler. But back to the bus speeds and cache, Intel chips offer larger cache capacity and front side bus speeds than AMD. Basically a bottle neck after the CPU converts the data.
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!