• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Media encoding test i7 920@ 4ghz V Hex core AMD

Status
Not open for further replies.
but he could atleast choose one that uses 100% cpu usage.

on the first pass i get 60% cpu usage at best , on the second pass 80-97%

x264 is obviously going to favor less cores if they are faster cores wheres handbrake which he originaly wanted to compare can max out 6 cores easily

heres stock 1055t results
encoded 1442 frames, 64.59 fps, 3898.70 kb/s
encoded 1442 frames, 64.84 fps, 3898.70 kb/s
encoded 1442 frames, 64.02 fps, 3898.70 kb/s
encoded 1442 frames, 64.62 fps, 3898.70 kb/s
encoded 1442 frames, 25.62 fps, 3971.34 kb/s
encoded 1442 frames, 25.56 fps, 3946.44 kb/s
encoded 1442 frames, 26.47 fps, 3968.49 kb/s
encoded 1442 frames, 26.04 fps, 3953.60 kb/s


System Details
--------------
Name AMD K10
Codename
Specification AMD Phenom(tm) II X6 1055T Processor
Core Stepping
manufacturer Gigabyte Technology Co., Ltd.
vendor Gigabyte Technology Co., Ltd.
manufacturer Gigabyte Technology Co., Ltd.
Stock frequency 3300 MHz
Core Speed 2812.5 MHz
FID range 5.0x - 16.5x
but look at my cpu usage graph...
53036006.jpg


handbrake can use multiple cores properly , this can not results are not a good comparision because anyone with a 6 core cpu wouldnt be using an encoder that cant use the cores.

[edit] still faster than the stock 920 someone linked on the benchmark image up above though, this is why i always say reviews are pointless without cpu graphs these days because the results dont display whats really going on
 
Last edited:
It's the version of the h264 encoder they use, the newer one doesn't have that problem and uses full 100% of all the cores.

If we do it we should use the latest x264.exe, doesn't have to be the one in their test package as long as we all use the same version.

Arknor, run the test again but using the latest exe from here:

http://x264.nl/

See what cpu usage you get in comparison.
 
Last edited:
Yeah, they look slightly down.

The 1055T i've used returned an average in the x264 benchmark of 76.79fps and 30.47fps @ 3.2ghz.

gt
 
Just some results for an OC'ed 1055T @ 3.6Ghz

encoded 1442 frames, 78.38 fps, 3898.70 kb/s
encoded 1442 frames, 78.19 fps, 3898.70 kb/s
encoded 1442 frames, 78.52 fps, 3898.70 kb/s
encoded 1442 frames, 78.36 fps, 3898.70 kb/s
encoded 1442 frames, 33.38 fps, 3951.35 kb/s
encoded 1442 frames, 33.36 fps, 3968.84 kb/s
encoded 1442 frames, 33.53 fps, 3972.00 kb/s
encoded 1442 frames, 33.65 fps, 3972.13 kb/s


System Details
--------------
Name AMD K10
Codename
Specification AMD Phenom(tm) II X6 1055T Processor
Core Stepping
manufacturer Gigabyte Technology Co., Ltd.
vendor Gigabyte Technology Co., Ltd.
manufacturer Gigabyte Technology Co., Ltd.
Stock frequency 3300 MHz
Core Speed 3640.0 MHz
FID range 5.0x - 16.5x

All cores around 96-98%

**edit

First pass, CPU is only at 52%
 
Last edited:
It's the version of the h264 encoder they use, the newer one doesn't have that problem and uses full 100% of all the cores.

If we do it we should use the latest x264.exe, doesn't have to be the one in their test package as long as we all use the same version.

Arknor, run the test again but using the latest exe from here:

http://x264.nl/

See what cpu usage you get in comparison.

doesnt work i get failed to load avisynth for every pass if i try to replace x264 with the newer version.

i obviously have avisynth installed properly to do the run i did earlier
 
I'll have a play when I get home, as I use the latest version all the time with no fuss, not tried it with this benchmark though.

But the latest version is much faster on the X6's compared to the old versions.
 
i just got the newer version working i originaly tried the 64bit exe , the 32bit one works.
stock cpu again btw

pass one is 40fps higher with the newer exe while still not beeing anywhere near 100% cpu utilisation.
pass 2 only gained around 1fps with not much better cpu utilisation.

encoded 1442 frames, 103.53 fps, 3909.64 kb/s
encoded 1442 frames, 104.83 fps, 3909.64 kb/s
encoded 1442 frames, 104.51 fps, 3909.64 kb/s
encoded 1442 frames, 102.93 fps, 3909.64 kb/s
encoded 1442 frames, 26.60 fps, 3961.67 kb/s
encoded 1442 frames, 26.95 fps, 3960.81 kb/s
encoded 1442 frames, 26.82 fps, 3961.64 kb/s
encoded 1442 frames, 26.93 fps, 3957.70 kb/s

58600231.jpg


(faster than a stock 920 again though which kills the rumour of intel for encoding because amd are only good at games people keep spouting on these forums)
 
Much more like it :)

Yeah still not 100% utilisation, but deffo a big improvement over the poorly optimised old version.

Everyone should run the test using the latest 32bit x264 version 1666:

http://mirror01.x264.nl/x264/revision1666/x264.exe


Fairer playing field then as at least some optimisation has been done for the X6's.
 
Latest 32bit x264.exe, i cannot get the 64bit working

1055T @ 3.6GHz

encoded 1442 frames, 125.76 fps, 3909.64 kb/s
encoded 1442 frames, 126.81 fps, 3909.64 kb/s
encoded 1442 frames, 126.26 fps, 3909.64 kb/s
encoded 1442 frames, 126.89 fps, 3909.64 kb/s
encoded 1442 frames, 34.38 fps, 3960.98 kb/s
encoded 1442 frames, 34.28 fps, 3960.58 kb/s
encoded 1442 frames, 34.37 fps, 3960.58 kb/s
encoded 1442 frames, 34.29 fps, 3960.50 kb/s


System Details
--------------
Name AMD K10
Codename
Specification AMD Phenom(tm) II X6 1055T Processor
Core Stepping
manufacturer Gigabyte Technology Co., Ltd.
vendor Gigabyte Technology Co., Ltd.
manufacturer Gigabyte Technology Co., Ltd.
Stock frequency 3300 MHz
Core Speed 3639.9 MHz
FID range 5.0x - 16.5x
 
two tests , first with HT off and second with HT on, sorry no 4 ghz here but a 3.5Ghz i7 920.
1
encoded 1442 frames, 89.37 fps, 3900.68 kb/s
encoded 1442 frames, 89.29 fps, 3900.68 kb/s
encoded 1442 frames, 88.62 fps, 3901.21 kb/s
encoded 1442 frames, 88.98 fps, 3901.21 kb/s
encoded 1442 frames, 26.66 fps, 3970.42 kb/s
encoded 1442 frames, 26.74 fps, 3971.61 kb/s
encoded 1442 frames, 26.66 fps, 3971.37 kb/s
encoded 1442 frames, 26.79 fps, 3970.70 kb/s

2
encoded 1442 frames, 78.47 fps, 3899.02 kb/s
encoded 1442 frames, 82.53 fps, 3899.02 kb/s
encoded 1442 frames, 81.80 fps, 3899.02 kb/s
encoded 1442 frames, 80.72 fps, 3899.02 kb/s
encoded 1442 frames, 32.14 fps, 3960.93 kb/s
encoded 1442 frames, 32.47 fps, 3961.14 kb/s
encoded 1442 frames, 32.49 fps, 3967.55 kb/s
encoded 1442 frames, 32.38 fps, 3965.18 kb/s
 
I've got some results but from the old executable:

http://www.vortez.co.uk/contentteller/articles_pages/asus_p6x58d_e_motherboard_review,14.html

i7 920 @ stock:

encoded 1442 frames, 64.81 fps, 3899.02 kb/s
encoded 1442 frames, 66.01 fps, 3899.02 kb/s
encoded 1442 frames, 66.24 fps, 3899.02 kb/s
encoded 1442 frames, 65.44 fps, 3899.02 kb/s
encoded 1442 frames, 25.20 fps, 3965.62 kb/s
encoded 1442 frames, 25.09 fps, 3970.28 kb/s
encoded 1442 frames, 24.99 fps, 3964.69 kb/s
encoded 1442 frames, 25.24 fps, 3971.98 kb/s

i7 920 @ 4ghz:

encoded 1442 frames, 92.78 fps, 3899.02 kb/s
encoded 1442 frames, 92.64 fps, 3899.02 kb/s
encoded 1442 frames, 92.02 fps, 3899.02 kb/s
encoded 1442 frames, 93.46 fps, 3899.02 kb/s
encoded 1442 frames, 36.50 fps, 3971.47 kb/s
encoded 1442 frames, 36.57 fps, 3959.97 kb/s
encoded 1442 frames, 36.73 fps, 3971.55 kb/s
encoded 1442 frames, 36.73 fps, 3971.13 kb/s
 
Last edited:
You can't run the 64bit encoder without using 64bit codecs, and the benchmark util is setup to use the 32bit encoder.
 
Status
Not open for further replies.
Back
Top Bottom