x264 HD Benchmark

[torrentthief]

NON-overclocked:

Q9300, 7.5x333 (2.5ghz), P35, DDR2, 5-5-5-12 @ 400 MHz, XP Pro SP2

1st pass avg: 54.09 fps
2nd pass avg: 15.03 fps

---------- RUN1PASS1.LOG
encoded 1442 frames, 53.59 fps, 3904.67 kb/s

---------- RUN2PASS1.LOG
encoded 1442 frames, 54.10 fps, 3904.67 kb/s

---------- RUN3PASS1.LOG
encoded 1442 frames, 54.42 fps, 3904.67 kb/s

---------- RUN4PASS1.LOG
encoded 1442 frames, 54.26 fps, 3904.67 kb/s

---------- RUN1PASS2.LOG
encoded 1442 frames, 15.06 fps, 3952.97 kb/s

---------- RUN2PASS2.LOG
encoded 1442 frames, 15.02 fps, 3952.97 kb/s

---------- RUN3PASS2.LOG
encoded 1442 frames, 15.02 fps, 3952.97 kb/s

---------- RUN4PASS2.LOG
encoded 1442 frames, 15.01 fps, 3952.97 kb/s












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

Overclocked benchmark:

Q9300, 7.5x440 (3.3ghz), P35, DDR2, 5-5-5-12 @ 528 MHz, XP Pro SP2

1st pass avg: 70.66fps
2nd pass avg: 19.74fps




---------- RUN1PASS1.LOG
encoded 1442 frames, 70.72 fps, 3905.42 kb/s

---------- RUN2PASS1.LOG
encoded 1442 frames, 70.77 fps, 3905.42 kb/s

---------- RUN3PASS1.LOG
encoded 1442 frames, 70.56 fps, 3905.42 kb/s

---------- RUN4PASS1.LOG
encoded 1442 frames, 70.61 fps, 3905.42 kb/s

---------- RUN1PASS2.LOG
encoded 1442 frames, 19.80 fps, 3952.85 kb/s

---------- RUN2PASS2.LOG
encoded 1442 frames, 19.72 fps, 3952.85 kb/s

---------- RUN3PASS2.LOG
encoded 1442 frames, 19.73 fps, 3952.85 kb/s

---------- RUN4PASS2.LOG
encoded 1442 frames, 19.70 fps, 3952.85 kb/s



Btw, which version of x264 is this? since v793 there have been big speed improvements to core2 cpu's, mine were 13% faster with an overclocked e6400 with build 815 or similar compared to build 793 or similar which i tried 1 month before. You should make a test 1.1 which includes the latest build 819 as it will give users with new cpu's better speed.

Also you should make the test do 1080p encoding or something as i get about the same fps with 6*440 as i do with 7.5*440 but there should be a massive difference, You should consider using say a 80mb 1080p mpeg2/mpeg4 source instead of 57mb 720p mpeg2. 6*440 = 2.64GHZ, 7.5*440= 3.3GHZ, there should be a massive difference yet there was about 1fps difference in the 1st pass and 0.5fps difference in 2nd pass.

[graysky]

Originally Posted by torrentthief

Btw, which version of x264 is this? since v793 there have been big speed improvements to core2 cpu's, mine were 13% faster with an overclocked e6400 with build 815 or similar compared to build 793 or similar which i tried 1 month before. You should make a test 1.1 which includes the latest build 819 as it will give users with new cpu's better speed.

Also you should make the test do 1080p encoding or something as i get about the same fps with 6*430 as i do with 7.5*430 but there should be a massive difference, You should consider using say a 80mb 1080p mpeg2/mpeg4 source instead of 57mb 720p mpeg2. 6*430 = 2.58GHZ, 7.5*430= 3.225GHZ, there should be a massive difference yet there was about 1fps difference in the 1st pass and 0.5fps difference in 2nd pass.

First off thanks for the data set. I'll update tomorrow.

To your questions: I believe it's version 747. The problem with making a benchmark like this is that you can retrospectively make a change to it It'll have to stay this version to keep the data consistent. The nice thing about using the same version is that all the data are comparable (i.e. you can find two similar setups and be pretty confident that computer A is 10 % faster than computer B).

If you wanna manually replace x264.exe in the /test with version 819 and report the results, that'd be cool.

EDIT: I just did the experiment:

My q6600 running @ 8x333, P965 chipset, DDR2 4-4-4-12 @ 667, XP Pro x64
x264 v0.58.747 (original):
FPS1: 55.38
FPS2: 15.06

x264 v0.58.819:
FPS1: 59.31
FPS2: 16.84

Differences: 7.1% and 11.7 %

[torrentthief]

How about you host build 819 x264.exe and get users to submit those results too, add another column to the spreadsheet with "build 819" as title and add a new build every 2 months as this will showcase what new cpu's can do as multithreading and sse3 improvements in x264 each month are large.

here is build 819 sharktooth patched: sendspace.com/file/c1ps6j (368kb rarred)

Overclocked result with build 819:

1st pass avg: 77.41 fps (9.55% faster)
2nd pass avg: 22.25 fps (12.71% faster)

Also you have written Q9540 instead of Q9450 several times in your spreadsheet, could you change them please.

I hope you can add a column for build 819 as people use this spreadsheet to decide what cpu to buy if they do lots of x264 encoding, older cpu's wont get much faster with new builds but new cpu's like my 45nm quad will.


---------- RUN1PASS1.LOG
encoded 1442 frames, 76.40 fps, 3888.26 kb/s

---------- RUN2PASS1.LOG
encoded 1442 frames, 78.08 fps, 3888.91 kb/s

---------- RUN3PASS1.LOG
encoded 1442 frames, 77.49 fps, 3888.97 kb/s

---------- RUN4PASS1.LOG
encoded 1442 frames, 77.68 fps, 3888.67 kb/s

---------- RUN1PASS2.LOG
encoded 1442 frames, 22.27 fps, 3963.26 kb/s

---------- RUN2PASS2.LOG
encoded 1442 frames, 22.25 fps, 3962.94 kb/s

---------- RUN3PASS2.LOG
encoded 1442 frames, 22.24 fps, 3962.98 kb/s

---------- RUN4PASS2.LOG
encoded 1442 frames, 22.24 fps, 3962.61 kb/s

[graysky]

@Torrenttheif - Hmm... lemme think about this. I started a new thread here to discuss this possibility. I like it, but at the same time, it can be a bit confusing to people. Let's all talk about it in that thread.

[torrentthief]

i've update my posts now with new fps for overclocked result for both the old x264 build and 819 as i've managed to overclock my cpu further, if you have already added my data could you please update it, if not add it to the database once you are ready.

[Finguz]

Results for x264.exe v0.58.747
encoded 1442 frames, 64.63 fps, 3905.42 kb/s
encoded 1442 frames, 64.20 fps, 3905.42 kb/s
encoded 1442 frames, 65.51 fps, 3905.42 kb/s
encoded 1442 frames, 64.93 fps, 3905.42 kb/s
encoded 1442 frames, 18.88 fps, 3952.85 kb/s
encoded 1442 frames, 18.96 fps, 3952.85 kb/s
encoded 1442 frames, 18.90 fps, 3952.85 kb/s
encoded 1442 frames, 18.80 fps, 3952.85 kb/s

Results for x264.exe v0.59.819M
encoded 1442 frames, 70.84 fps, 3891.55 kb/s
encoded 1442 frames, 70.58 fps, 3891.55 kb/s
encoded 1442 frames, 71.26 fps, 3891.55 kb/s
encoded 1442 frames, 71.92 fps, 3891.55 kb/s
encoded 1442 frames, 20.98 fps, 3963.43 kb/s
encoded 1442 frames, 21.03 fps, 3963.43 kb/s
encoded 1442 frames, 21.16 fps, 3963.43 kb/s
encoded 1442 frames, 21.14 fps, 3963.43 kb/s


System Details
--------------
Name Intel Core 2 Quad Q6600
Codename Kentsfield
Specification Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
Core Stepping G0
Technology 65 nm
Stock frequency 2400 MHz
Core Speed 3399.9 MHz (8.0 x 425.0 MHz)
FID range 6.0x - 9.0x

Northbridge Intel P35/G33/G31 rev. A2
Southbridge Intel 82801IR (ICH9R) rev. 02

CAS# 4.0
RAS# to CAS# 4
RAS# Precharge 4
Cycle Time (tRAS) 12
Command Rate 2T
Memory Frequency 425.0 MHz (1:1)
Memory Type DDR2
Memory Size 2046 MBytes
Channels Dual (Symmetric)

Windows Version Microsoft Windows Vista (6.0) Ultimate Edition Service Pack 1 (Build 6001)

max VID 1.288 V
Voltage sensor 0 1.33 Volts [0x53] (CPU VCORE)
Number of processors 1
Number of threads 4
Number of threads 4 (max 4)
L2 cache 2 x 4096 KBytes, 16-way set associative, 64-byte line size
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, EM64T
Package Socket 775 LGA (platform ID = 4h)

Temperature sensor 0 38°C (100°F) [0x26] (TMPIN0)
Temperature sensor 1 37°C (98°F) [0x25] (TMPIN1)
Temperature sensor 0 50°C (121°F) [0x32] (core #0)
Temperature sensor 1 51°C (123°F) [0x31] (core #1)
Temperature sensor 2 47°C (116°F) [0x35] (core #2)
Temperature sensor 3 48°C (118°F) [0x34] (core #3)
Temperature sensor 0 50°C (121°F) [0x32] (GPU Core)

[graysky]

@Finguz - thanks for the data!
@torrenttheif - can you download the new version of the benchmark and run it @ your new o/c (7.5x440)?

[Mac Daddy]

Here is a pass with the new test running Vista Ultimate X64 Graysky.

I dropped the CPU O/C back and have started tweaking RAM here with Everest getting almost 9 gb /s on reads ... still tweaking. New test is higher in results and including the system info is sweet hope we get our UAC issue resolved

Results for x264.exe v0.58.747
encoded 1442 frames, 34.66 fps, 3905.08 kb/s
encoded 1442 frames, 34.69 fps, 3905.08 kb/s
encoded 1442 frames, 34.38 fps, 3907.07 kb/s
encoded 1442 frames, 34.66 fps, 3906.78 kb/s
encoded 1442 frames, 8.56 fps, 3942.92 kb/s
encoded 1442 frames, 8.58 fps, 3942.01 kb/s
encoded 1442 frames, 8.53 fps, 3942.38 kb/s
encoded 1442 frames, 8.59 fps, 3942.25 kb/s

Results for x264.exe v0.59.819M
encoded 1442 frames, 38.48 fps, 3889.89 kb/s
encoded 1442 frames, 38.50 fps, 3889.56 kb/s
encoded 1442 frames, 36.61 fps, 3889.54 kb/s
encoded 1442 frames, 38.31 fps, 3889.89 kb/s
encoded 1442 frames, 9.59 fps, 3953.27 kb/s
encoded 1442 frames, 9.57 fps, 3953.25 kb/s
encoded 1442 frames, 9.48 fps, 3953.26 kb/s
encoded 1442 frames, 9.57 fps, 3953.26 kb/s


System Details
--------------
Name Intel Core 2 Duo E6550
Codename Conroe
Specification Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz
Core Stepping G0
Technology 65 nm
Stock frequency 2333 MHz
Core Speed 3009.9 MHz (7.0 x 430.0 MHz)
FID range 6.0x - 7.0x

Northbridge Intel X38 rev. 01
Southbridge Intel 82801IR (ICH9R) rev. 02

CAS# 5.0
RAS# to CAS# 5
RAS# Precharge 5
Cycle Time (tRAS) 18
Command Rate 1T
Memory Frequency 516.0 MHz (5:6)
Memory Type DDR2
Memory Size 4094 MBytes
Channels Dual (Symmetric)

Windows Version Microsoft Windows Vista (6.0) Ultimate Edition Service Pack 1 (Build 6001)

max VID 1.350 V
Voltage sensor 0 1.38 Volts [0x56] (CPU VCORE)
Number of processors 1
Number of threads 2
Number of threads 2 (max 2)
L2 cache 4096 KBytes, 16-way set associative, 64-byte line size
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, EM64T
Package Socket 775 LGA (platform ID = 0h)

Temperature sensor 0 30°C (85°F) [0x1E] (TMPIN0)
Temperature sensor 1 30°C (85°F) [0x1E] (TMPIN1)
Temperature sensor 0 43°C (109°F) [0x39] (core #0)
Temperature sensor 1 45°C (112°F) [0x37] (core #1)
Temperature sensor 0 34°C (93°F) [0x22] (GPU Core)

[graysky]

Originally Posted by Mac Daddy

still tweaking. New test is higher in results and including the system info is sweet hope we get our UAC issue resolved

Thanks for the data dude, will update shortly. I just replaced the rar file with a newer version containing a fix (albeit a little clunky) for that UAC crap. Hopefully, someone else can come-up with a more elegant fix, but this should work for now... you wanna switch your UAC back on and test it on your Vista x64?

[Mac Daddy]

Originally Posted by graysky

Thanks for the data dude, will update shortly. I just replaced the rar file with a newer version containing a fix (albeit a little clunky) for that UAC crap. Hopefully, someone else can come-up with a more elegant fix, but this should work for now... you wanna switch your UAC back on and test it on your Vista x64?

Will do man will run it later today after I get back home with a few more memory tweaks