x264 Benchmark

[Chai]

It's an ES chip, which probably means that the multiplier was changed.

[Adrian Wong]

Originally Posted by hkazemi

I found an inconsistency in the data table for an entry labeled Core 2 Quad Q9550 ES. The problem has to do with the listed multiplier. It is listed as 8x333 for 2.66 ghz.

The problem is that isn't the right multiplier for a Q9550; it is correct for a Q9450 at stock speeds.

Q9450 is 8.0 x 333 for 2.66 ghz
Q9550 is 8.5 x 333 for 2.83 ghz

This can be seen various places on the web, including the tables here:
List of future Intel Core 2 microprocessors - Wikipedia, the free encyclopedia

Both the stock speed table and the Intel OC table show this entry.

I think most likely the chip is a Q9450 not a Q9550.

Yeah, our Desktop CPU Comparison Guide also lists the Q9550 as 8.5 x 333.

However, all Core 2 processors have their lower multipliers unlocked. So, he may be running the Q9550 at 8 x 333. That would make it equivalent to the Q9450.

[hkazemi]

Originally Posted by Adrian Wong

Yeah, our Desktop CPU Comparison Guide also lists the Q9550 as 8.5 x 333.

However, all Core 2 processors have their lower multipliers unlocked. So, he may be running the Q9550 at 8 x 333. That would make it equivalent to the Q9450.

Well, the coloring in the table uses black for processors at their stock speed settings...if it's a Q9550 with a non-stock multiplier it should be shown in red.

If it's really a Q9450, then the chip is mislabeled as a Q9550 in the table. Possibly a typo.

In short, the combination of the following means something is wrong:
1.) black text indicates a stock speed setting
2.) presence on the combined Intel/AMD table indicates a stock speed setting
3.) a chip labeled as Q9550 has a stock multi of 8.5, not 8, as shown in the table

Either the chip is mislabeled in the table, or the table row should be colored red and then the chip should not be listed in the combined stock speeds table.

[hkazemi]

CDRInfo used the graysky x264 benchmark in an Oct 15, 2007 review of X38 chipset motherboards.

Using an Intel E6600 running on 3 different X38 chipset boards (1st pass, 2nd pass):
Asus Blitz Formula 71.47, 17.21
Asus Maximus Formula 71.34, 17.22
Gigabyte GA-X38-DQ6 71.30, 17.19

Intel X38 Motherboard Roundup
Intel X38 Motherboard Roundup

---

One of my encoding benchmark irritants is when a site only lists the encoding times and not the encoding FPS, and also does not give much info about the encoded source (source material, length of source).

Knowing encoding FPS is more useful to me so I can estimate how long other encode jobs are likely to take for content of the same resolution, at any source content framerate.

[kpfong]

Pentium 4 Northwood
14x 200MHz
i865P
2.5-3-3-8@200MHz
XP Home SP2

---------- RUN1PASS1.LOG
encoded 1749 frames, 26.42 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 26.40 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 26.20 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 26.46 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 26.42 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 6.26 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 6.27 fps, 1826.26 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 6.24 fps, 1826.38 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 6.21 fps, 1826.38 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 6.21 fps, 1826.38 kb/s

[hkazemi]

Hello,

I ran a few tests on an overclocked Intel Core2/Pentium Dual-Core E2180 @ 3.2 ghz (10x320) (stock speed 2 ghz, 10x200) on a Gigabyte GA-G33M-DS2R motherboard. The machine is running Ubuntu 7.10 (Gutsy) with VMware Server 1.0.4, installed from the 'ubuntu gutsy partner' repository. Windows XP SP2 was running in a virtual machine on VMware.

Here are results of the x264 benchmark running on WinXP SP2under the VMware Server 1.0.4 virtual machine on Ubuntu on an otherwise idle Intel E2180 @ 3.2 ghz:

First, performance with the VMware virtual machine set to a single processor. VMware Tools not installed:

---------- RUN1PASS1.LOG
encoded 1749 frames, 41.67 fps, 1854.10 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 42.65 fps, 1854.10 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 42.69 fps, 1854.10 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 41.68 fps, 1854.10 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 41.79 fps, 1854.10 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 10.42 fps, 1825.89 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 10.38 fps, 1825.89 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 10.32 fps, 1825.89 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 10.40 fps, 1825.89 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 10.02 fps, 1825.89 kb/s


Second, performance with the VMware virtual machine set to a single processor. VMware Tools for Windows installed on the copy of WinXP SP2.

---------- RUN1PASS1.LOG
encoded 1749 frames, 42.64 fps, 1854.10 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 44.65 fps, 1854.10 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 44.31 fps, 1854.10 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 44.68 fps, 1854.10 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 44.69 fps, 1854.10 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 10.96 fps, 1825.89 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 10.92 fps, 1825.89 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 10.99 fps, 1825.89 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 11.04 fps, 1825.89 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 10.92 fps, 1825.89 kb/s


Third, performance with the VMware virtual machine set to two processors, WinXP HAL changed to support two processors. VMware Tools for Windows installed on the copy of WinXP SP2.

---------- RUN1PASS1.LOG
encoded 1749 frames, 73.55 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 76.36 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 76.98 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 77.14 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 76.36 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 19.00 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 19.20 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 19.14 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 19.06 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 18.53 fps, 1826.37 kb/s

My observations are:
1.) installing VMware Tools slightly helped performance ...I measured a boost of about 5.7%
2.) activating the second processor in the VMware virtual machine gave me a performance boost of 74% over the single processor configuration
3.) using the 2nd pass score of 20.93 fps from the E2160 @ 9x356=3.2ghz that is shown in the published benchmark results table, VMware may have an overhead of about 9.4% (20.93 fps/19.13 fps). This matches closely with a published stat for another VMware product: "for VMware ESX Server, we measured CPU overhead to be less than 10%." seen on http:// blogs.vmware.com/performance/2007/11/ten-reasons-why.html
4.) I don't think there is any measurable overheard on the single processor stats, as VMware could shift its own overhead to the otherwise unused processor.

I did not run a test with a 2 processor virtual machine without VMware Tools.

I do not have any tests of this processor at stock speeds, but they are likely very similar to the E2140 and E2160 results.

I would like to run a test using Avisynth and x264 under WINE to see how that fares compared to VMware.

[Ancient_1]

I just ran this on my new 8400 and decided to post the results sine I didn't see any Wolfdale results yet.

CPU 8400 @ 4197 (466x9)
DFI LP LT P35
A-Data 6400 4-4-4-12 @ 466 5-5-5-14 (2x1g

---------- RUN1PASS1.LOG
encoded 1749 frames, 120.88 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 121.14 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 121.14 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 121.15 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 121.01 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 30.92 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 30.95 fps, 1826.38 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 30.89 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 30.93 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 30.94 fps, 1826.38 kb/s

My old system

930 @ 3702 (247x15)
Asus P5P800SE
OCZ 4400 @ 247 3-3-3


---------- RUN1PASS1.LOG
encoded 1749 frames, 51.25 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 50.97 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 50.95 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 51.39 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 51.32 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 13.00 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 12.90 fps, 1826.22 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 12.90 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 13.00 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 12.99 fps, 1826.38 kb/s

Quite an improvement for me (mostly do 1080i or 720p HDTV to xvid conversion).

[kpfong]

Core 2 Duo E8200
334MHz x 8
x38
5-5-5-18@400.8
Win Vista Ult 64
---------- RUN1PASS1.LOG
encoded 1749 frames, 77.16 fps, 1850.89 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 79.23 fps, 1850.89 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 78.90 fps, 1850.89 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 78.73 fps, 1850.89 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 78.95 fps, 1850.89 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 19.89 fps, 1826.37 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 19.91 fps, 1826.37 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 19.85 fps, 1826.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 19.91 fps, 1826.37 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 19.93 fps, 1826.37 kb/s

[i43]

--------------------------------------------

---------- RUN1PASS1.LOG
encoded 1749 frames, 166.08 fps, 1850.94 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 167.58 fps, 1850.94 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 167.56 fps, 1850.94 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 167.56 fps, 1850.94 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 166.83 fps, 1850.94 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 44.88 fps, 1829.12 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 44.88 fps, 1829.31 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 44.90 fps, 1829.37 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 45.01 fps, 1829.18 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 44.92 fps, 1829.26 kb/s

-------------------------------------------------

Hi
just run Your benchmark on QX9650 / @ Stock speeds
sorry just had ''hash'' url , couldn't attach screen due to post amount ... (just replace ' * ' to ' t ' ) manually,
sorry again

h*tp://img145.imageshack.us/my.php?image=core2extrmeqx96509x333xyr7.png
h*tp://img145.imageshack.us/img145/2357/core2extrmeqx96509x333xyr7.png

-----
other system inf

CPU: QX9650 / TRUE120
RAM: G.Skill F3-10600CL8D-2GBH (DDR3-1333) PC3-10600 1024Mb x2 CL-8-8-8-21 1.65v
MBO: Maximuz Extreme
GFX: Asus 8800 Ultra / Stock Air
PSU: BQ! 1KW

[EvenOlder]

Processors: Two Intel E5410 2.33gHz Xeons (Harpertown) [8 total cores]
CPU Multiplier x FSB: 7x333.34
Chipset: Intel 5100 (San Clemente) [Tyan 5100X Board]
Memory: 5-5-5-13 @ 333mHz [DDR2 ECC Registered 4 gigs]
OS: XP Pro SP2
This is the cheap [$1500 for the box] way of getting 8 cores


---------- RUN1PASS1.LOG
encoded 1749 frames, 144.81 fps, 1849.61 kb/s

---------- RUN2PASS1.LOG
encoded 1749 frames, 146.89 fps, 1849.61 kb/s

---------- RUN3PASS1.LOG
encoded 1749 frames, 144.81 fps, 1849.61 kb/s

---------- RUN4PASS1.LOG
encoded 1749 frames, 147.47 fps, 1849.61 kb/s

---------- RUN5PASS1.LOG
encoded 1749 frames, 144.63 fps, 1849.61 kb/s

---------- RUN1PASS2.LOG
encoded 1749 frames, 64.44 fps, 1834.86 kb/s

---------- RUN2PASS2.LOG
encoded 1749 frames, 64.63 fps, 1834.86 kb/s

---------- RUN3PASS2.LOG
encoded 1749 frames, 64.70 fps, 1834.86 kb/s

---------- RUN4PASS2.LOG
encoded 1749 frames, 64.70 fps, 1834.86 kb/s

---------- RUN5PASS2.LOG
encoded 1749 frames, 64.81 fps, 1834.91 kb/s

1st Pass Average: 145.72
2nd Pass Average: 64.66

1 pass time: 12s + 27s = 39s

This is the most informative test I've seen on the net in the past three years. THANK YOU!!

steve