PAR2 Multithreaded Archive Recovery Performance

Par2 is an application used for reconstructing downloaded archives. It can generate parity data from a given archive and later use it to recover the archive

Chuchusoft took the source code of par2cmdline 0.4 and parallelized it using Intel’s Threading Building Blocks 2.1. The result is a version of par2cmdline that can spawn multiple threads to repair par2 archives. For this test we took a 708MB archive, corrupted nearly 60MB of it, and used the multithreaded par2cmdline to recover it. The scores reported are the repair and recover time in seconds.

Data Recovery - par2cmdline 0.4 Multithreaded

The Athlon II X3 and X4 do very well here thanks to the test being quite parallelized. The dual core options are competitive though.

Sorenson Squeeze: FLV Creation

We're using Sorenson Squeeze to convert regular videos into Flash videos for use on websites.

Sorenson Squeeze Pro 5 - Flash Video Creation

The trend continues given the threaded nature of this test. The dual-core AMD chips are more competitive than usual here, they are significantly faster than the Pentium E6300.

WinRAR - Archive Creation

Our WinRAR test simply takes 300MB of files and compresses them into a single RAR archive using the application's default settings. We're not doing anything exotic here, just looking at the impact of CPU performance on creating an archive:

WinRAR 3.8 Compression - 300MB Archive

3dsmax & Cinebench Performance Gaming Performance
Comments Locked

63 Comments

View All Comments

  • mindless1 - Saturday, January 30, 2010 - link

    Not necessarily, some e-series run any given frequency at a lower voltage, so by the same token whatever a non-e can do undervolted, an e- might do a bit higher speed at same voltage.

    Further, not all boards allow undervolting much, if at all, especially when you're trying to build a very small system. Big system with full mATX or ATX board with full o'c features, and silent fans is something just about any Tom, Dick, and Harry can build with no need to pick an e-series or undervolt a regular CPU.
  • play2learn - Monday, January 25, 2010 - link

    Not everybody have up to date CPU:s. Anand, it would be fun if you sneak in an old processor like pentium 3, 4 or Pentium D next time.
  • Taft12 - Wednesday, January 27, 2010 - link

    Anand did just that back in when Phenom II X2 and Athlon II X2 were released:

    http://www.anandtech.com/cpuchipsets/showdoc.aspx?...">http://www.anandtech.com/cpuchipsets/showdoc.aspx?...

    It was an embarrassment, no self-respecting hardware geek should be using that crap when the cheapest CPU and motherboard on the market today buries that stone-age technology.
  • mindless1 - Saturday, January 30, 2010 - link

    ... but to many geeks, it's the love of tech, not the ePenis you claim by virtue of your system performance.

    I mean really, there is a point where even a geek takes a sensible position about what is appropriate for the work or play they actually do. Would you be amazed to know I still have a Celeron 500MHz fileserver running? The geek in me made it right in the first place including beefed up mobo capacitors, very open heatsink that takes years to get appreciable dust in it, filtered intake/positive pressure, and a few others things faster, but inferior for the purpose, systems have today.
  • mindless1 - Saturday, January 30, 2010 - link

    lol, meant ... DON'T usually have today.
  • play2learn - Thursday, January 28, 2010 - link

    You have a point there. But then there are people like me that's still not satisfied with today's technology and just sit and wait and wait for that ultimate experience to show up. If you want me to buy a new system today (which is not going to happen) I need to be humiliated more than once.
  • ssj4Gogeta - Monday, January 25, 2010 - link

    Won't it skew the graphs (eg. a bar too long to fit or a bar too small to be visible)
  • NJoy - Monday, January 25, 2010 - link

    Sure it would. Last week just out of curiosity I decided to run CinebenchR10 on a Dell Precision with dual Xeons (PD 3Ghz based, 4 threads) against C2D E8400. Guest what? Over 4 minutes for Xeons whilst about 2:20 for C2D. Humiliation...
  • mattscottshea - Monday, January 25, 2010 - link

    These are assumptions made by current naming standards. However, if these were to hold true, it would be quite a coup by AMD as there current unreleased top of the line 975 will run at 3.6GHz with 4 cores at 140W using the C3 stepping.

    AMD Phenom II X6 1075T 3.6GHz 6 cores / 125W (D1 stepping?)
    AMD Phenom II X6 1055T 3.2GHz 6 cores / 95W
    AMD Phenom II X6 1035T 2.9GHz 6 cores / 95W
    AMD Phenom II X4 960T 3.3GHz 4 cores / 95W
  • MrSpadge - Monday, January 25, 2010 - link

    LOL. Naming conventions are there to lure people into buying just when you break the convention (*looking at the greenteam*)

    Anyway, their process has not magically changed - otherwise todays CPUs would be more energy efficient. And a new stepping is not going to turn the world upside down.. maybe a gain of 5% power at similar settings and a slightly higher clock speed at similar power levels.

    And today you're getting 4 cores at 3.4 GHz for 140 / 125W. They'll be lucky and need heavy speed binning to get 3 GHz at 140W for the best 6-core CPU.

Log in

Don't have an account? Sign up now