Benchmark your computer with Handbrake

[videogames101]

Try the latest Nightly build, I gained almost 5% on ivy i5.

The 4.2GHz battle

[B]
  Name           FPS         CPU       CPU MHz  Ram MHz  Timings      OS / CPU Capabilities[/B]
1 [URL="http://forums.anandtech.com/showpost.php?p=35215306&postcount=149"]YBS1[/URL]           354.352325  i7-3930K  4200       1866   -            MMX2 SSE2Fast SSSE3 SSE4.2 AVX                       
2 [URL="http://forums.anandtech.com/showpost.php?p=35202792&postcount=55"]SKORPI0[/URL]        324.455719  i7-3930K  4200          -   -            MMX2 SSE2Fast SSSE3 SSE4.2 AVX                       
3 [URL="http://forums.anandtech.com/showpost.php?p=35214137&postcount=142"]Makaveli[/URL]       319.169128  i7-970    4200       1600   7-8-7-20-1T  MMX2 SSE2Fast SSSE3 SSE4.2                           
4 [URL="http://forums.anandtech.com/showpost.php?p=35205269&postcount=78"]MrPickins[/URL]      253.857376  i5-4670K  4200       1600   -            MMX2 SSE2Fast SSSE3 SSE4.2 AVX AVX2 FMA3 LZCNT BMI2  
5 [URL="http://forums.anandtech.com/showpost.php?p=35213666&postcount=137"]videogames101[/URL]  221.463760  i5-3570K  4200       1600   9-9-9-24-1T  MMX2 SSE2Fast SSSE3 SSE4.2 AVX                       
6 [URL="http://forums.anandtech.com/showpost.php?p=35200979&postcount=16"]Edgemeal[/URL]       219.427078  i5-3570K  4200       1600   9-9-9-28-2T  MMX2 SSE2Fast SSSE3 SSE4.2 AVX                       
7 [URL="http://forums.anandtech.com/showpost.php?p=35213976&postcount=139"]gbeirn[/URL]         216.190186  i7-960    4200       1600   9-9-9-24-2T  MMX2 SSE2Fast SSSE3 SSE4.2                           
8 [URL="http://forums.anandtech.com/showpost.php?p=35201346&postcount=23"]nyker96[/URL]        194.348465  i5-2500K  4200       1600   -            MMX2 SSE2Fast SSSE3 SSE4.2 AVX

HandBreak Score Chart updated here (for now).

memory timings seem to have net me 2 fps on you edgemeal haha

 

[CHADBOGA]

Using the latest nightly build, it boosted my score by about 16 frames per second.

i5-3570K @ 4.02GHz, TURBO OFF, DDR3-1600 @ 8-8-8-24-1T

HandBrake svn5622 (Nightly Build) - 64bit Version
OS: Microsoft Windows NT 6.1.7601 Service Pack 1
CPU: Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz (actually at 4.02Ghz)
Ram: 8147 MB, Screen: 2560x1600
.
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX
.
[16:38:29] work: average encoding speed for job is 238.412048 fps

 

[monstercameron]

memory timings seem to have net me 2 fps on you edgemeal haha

couldnt that be operating environment or some other variable, memory timings seem so small to make such a difference...

 

[CHADBOGA]

couldnt that be operating environment or some other variable, memory timings seem so small to make such a difference...

Dunno about everyone else, but when I ran Handbrake, it was after a reboot and nothing else was running except for my Steam connection(doing nothing after establishing itself) and MSE in the background, also lazing about.

 

[Edgemeal]

Using the latest nightly build, it boosted my score by about 16 frames per second.

i5-3570K @ 4.02GHz, TURBO OFF, DDR3-1600 @ 8-8-8-24-1T

HandBrake svn5622 (Nightly Build) - 64bit Version
OS: Microsoft Windows NT 6.1.7601 Service Pack 1
CPU: Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz (actually at 4.02Ghz)
Ram: 8147 MB, Screen: 2560x1600
.
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX
.
[16:38:29] work: average encoding speed for job is 238.412048 fps

238.412048 fps! You sure you didn't swap out your ivy with a haswell-e in your sleep? LOL!

You're getting more FPS then others running 200 and even 600 MHz faster?
Communism, 3570K @ 4.6GHz, svn5622(Nightly Build)-64bit, 229.081482 fps
cmdrdredd, 3570K @ 4.5GHz, unknown build,............... 236.807419 fps

You're making us all look bad Whats your secret?

EDIT...
You have power management disabled?
EIST?, C1?, C3?, C6?

 

[CHADBOGA]

238.412048 fps! You sure you didn't swap out your ivy with a haswell-e in your sleep? LOL!

You're getting more FPS then others running 200 and even 600 MHz faster?
Communism, 3570K @ 4.6GHz, svn5622(Nightly Build)-64bit, 229.081482 fps
cmdrdredd, 3570K @ 4.5GHz, unknown build,............... 236.807419 fps

You're making us all look bad Whats your secret?

EDIT...
You have power management disabled?
EIST?, C1?, C3?, C6?

In my bios, on my Asus motherboard, I have chosen the Balanced option, if memory serves me right and I have simply enabled the "Auto" option on pretty much everything..

I also have a 240gig Intel 520 SSD drive and according to CPU-Z, my RAM memory timings are pretty good, especially the 1T setting.

Are other people rebooting before running the benchmark?

When I first ran it, with heaps of Internet Tabs open and not having rebooted for a week or so, my score was considerably lower.

 

[cmdrdredd]

In my bios, on my Asus motherboard, I have chosen the Balanced option, if memory serves me right and I have simply enabled the "Auto" option on pretty much everything..

I also have a 240gig Intel 520 SSD drive and according to CPU-Z, my RAM memory timings are pretty good, especially the 1T setting.

Are other people rebooting before running the benchmark?

When I first ran it, with heaps of Internet Tabs open and not having rebooted for a week or so, my score was considerably lower.

Restarting should not affect the score that heavily. Not when it is a difference of 500mhz and memory running 500mhz faster as well. I also run off an ssd.

From run to run things can change a bit...margin of error and all. I saw 240fps once.

 

[Deders]

I've noticed some mention of OpenCL, is there a way to configure handbrake to use this?

Edit: just got an extra 3.3fps by reading and writing to my SSD but it was only registering around 3MB/s anyway.

I'm a little confused by this:

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!
x264 [info]: using SAR=43127/43200
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2
x264 [info]: profile Main, level 3.0
[13:06:30] reader: done. 1 scr changes
[13:06:30] work: average encoding speed for job is 153.330292 fps

What is PSY and PSNR?

 

[YBS1]

I've noticed some mention of OpenCL, is there a way to configure handbrake to use this?

In some versions there is an option, it was actually slightly slower for me when I tested it so I went back to using the normal version.

 

[Deders]

Ahh, now does anyone know anything about this:

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!

Edit: Just tried putting --tune spnr into the command line, it gave me much faster results with a 5MB smaller filesize but i still get the warning

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!
x264 [info]: using SAR=43127/43200
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2
x264 [info]: profile Main, level 3.0
[13:42:32] reader: done. 1 scr changes
[13:42:32] work: average encoding speed for job is 174.684723 fps


Just found some info here:

psy-rd

Default: 1.0:0.0
The first number is the strength of Psy-RDO to use (requires subme>=6 to activate). The second number is the strength of Psy-Trellis (requires trellis>=1 to activate). Note that Trellis is still considered 'experimental', and almost certainly is a Bad Thing for at least cartoons.
See this thread on doom9 for an explanation of psy-rd.
Life is short, and this article saved valubale time on this Earth.

I'm still pretty clueless as to how to disable it, I might set the setting to animation and see what happens....

Edit 2: now i'm back to around 153FPS, abouth 1MB larger filesize than the last one and still the same message saying I should tune pnsr

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!
x264 [info]: using SAR=43127/43200
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2
x264 [info]: profile Main, level 3.0
[13:48:54] reader: done. 1 scr changes
[13:48:55] work: average encoding speed for job is 152.150421 fps

 

[Edgemeal]

Edit 2: now i'm back to around 153FPS, abouth 1MB larger filesize than the last one and still the same message saying I should tune pnsr

Whats that have to do with this thread?
The OP asks to benchmark your PC using a MOV file with HandBreak, and to use the "Android" preset setting, and post your FPS result, move along, thank you.

Stop tweaking, were all getting basically the same log results because we're all using the "Andriod" preset.

 

[t-ray]

A bit disappointed in my results:

Handbrake Version: 0.9.9 (2013051800)
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX AVX2 FMA3 LZCNT BMI1 BMI2
[08:01:50] work: average encoding speed for job is 188.531113 fps

Operating System: Mint 15 x64 Cinnamon, Kernel 3.9
CPU: i7-4770 (non-k)
Mobo: MSI H87M-G43 (h87) (stock settings)
Ram: 16gb Kingston HyperX 1600 (stock settings)
Disk: Samsung 840 Pro 256

Running a build of Handbrake from mid-May.

It's my understanding that Handbrake doesn't support QuickSync on Linux.

 

[Deders]

Whats that have to do with this thread?
The OP asks to benchmark your PC using a MOV file with HandBreak, and to use the "Android" preset setting, and post your FPS result, move along, thank you.

Stop tweaking, were all getting basically the same log results because we're all using the "Andriod" preset.

So no one else is worried they are using the wrong settings for benchmarking like the warning says?

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!

Is everyone else getting the same warning?

 

[t-ray]

The first time I ran it, I had a VirtualBox session open running Windows 8. I shut everything down, and my average improved:

[08:21:04] work: average encoding speed for job is 200.940979 fps

@Deders - I see the same warnings in my logs as well.

 

[24601]

238.412048 fps! You sure you didn't swap out your ivy with a haswell-e in your sleep? LOL!

You're getting more FPS then others running 200 and even 600 MHz faster?
Communism, 3570K @ 4.6GHz, svn5622(Nightly Build)-64bit, 229.081482 fps
cmdrdredd, 3570K @ 4.5GHz, unknown build,............... 236.807419 fps

You're making us all look bad Whats your secret?

EDIT...
You have power management disabled?
EIST?, C1?, C3?, C6?

I just kinda ran the benchmark while I was litecoin mining, torrenting, and browsing the web.

Guess that effected the benchmark alot >.<

EDIT: Did it again at above normal priority with same scenario as before,
[08:36:30] work: average encoding speed for job is 229.024063 fps
High Priority:
[08:40:16] work: average encoding speed for job is 231.833801 fps

Yeah, guess I need to bench it later without all my stuff running.

EDIT: Ok, found the problem. When I look at process explorer when handbrake is running, I see that firefox is taking 2-5% of the CPU time. The rest of the programs are probably taking some as well.

 

[cmdrdredd]

I am gonna go out on a limb and say this benchmark is unreliable without strict controlled conditions.

 

[Edgemeal]

I am gonna go out on a limb and say this benchmark is unreliable without strict controlled conditions.

Handbrake is not a benchmark!

 

[Deders]

I am gonna go out on a limb and say this benchmark is unreliable without strict controlled conditions.

Handbrake is not a benchmark!

I'm inclined to agree with you both, even though the messages I've been getting do imply that it can be used as a benchmark.

Anywho, after discovering my memory timings were set slightly looser, (probably after I tried to install 8GB corsair vengence before sending them back the next day) I've re-tightened them and these are my official results, no software tweaking. Just loaded the file, selected the preset, pointed it to my SSD and clicked start:

x264 [warning]: --psnr used with psy on: results will be invalid!
x264 [warning]: --tune psnr should be used if attempting to benchmark psnr!
x264 [info]: using SAR=43127/43200
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2
x264 [info]: profile Main, level 3.0
[15:17:31] reader: done. 1 scr changes
[15:17:32] work: average encoding speed for job is 153.211517 fps
[15:17:32] sync: got 14315 frames, 14315 expected
[15:17:32] render: 14315 frames output, 0 dropped and 0 duped for CFR/PFR
[15:17:32] render: lost time: 0 (0 frames)
[15:17:32] render: gained time: 0 (0 frames) (0 not accounted for)
[15:17:33] h264-decoder done: 14315 frames, 0 decoder errors, 0 drops

 

[cmdrdredd]

Handbrake is not a benchmark!

Tell that to review sites who use it as such for system and CPU reviews. Plus you know the whole thread is saying "use this to bench your system and post results".

Before you roll your eyes maybe you should actually think. Don't act like only you know what handbreak is. I have been using it for years to convert videos for my phone and ps3.

 

[Edgemeal]

Tell that to review sites who use it as such for system and CPU reviews. Plus you know the whole thread is saying "use this to bench your system and post results".

Before you roll your eyes maybe you should actually think. Don't act like only you know what handbreak is. I have been using it for years to convert videos for my phone and ps3.

Sorry if I came off that way, didn't mean any disrespect to anyone. Carry on.

 

[Saffron]

Something isn't right here. My sandy bridge mobile quad i7 and your haswell is even on performance. I have a lower clock speed than yours and don't have AVX2. What gives?

I honestly didn't think anything of it until you mentioned something. So I ran Handbrake two more times and this is what I got:

Run#2:
[08:05:22] starting job
[08:06:10] work: average encoding speed for job is 304.600189 fps

Run#3:
[08:07:27] starting job
[08:08:14] work: average encoding speed for job is 305.948273 fps

Almost double the performance as my first run. I might have had my notebook unplugged without realizing it or something.

 

[BallaTheFeared]

That doesn't look right either, you probably used the normal preset.

 

[Makaveli]

Tell that to review sites who use it as such for system and CPU reviews. Plus you know the whole thread is saying "use this to bench your system and post results".

Before you roll your eyes maybe you should actually think. Don't act like only you know what handbreak is. I have been using it for years to convert videos for my phone and ps3.

what is the problem with using handbrake this way?

As long as people post what build they are using and use the andriod preset the numbers should be valid.

 

[cmdrdredd]

what is the problem with using handbrake this way?

As long as people post what build they are using and use the andriod preset the numbers should be valid.

Because it varies too much and there is no way a slower system should show higher performance.

 

[Makaveli]

Because it varies too much and there is no way a slower system should show higher performance.

I agree that the numbers do vary but isn't a margin of error to be expected in any benchmark.

As for the slower system showing numbers higher than higher end systems you are right there has to be something done wrong in the testing.

It would be abit of work but I think a graph for the nightly numbers and a different one for the standard build would show more consistency. Also people making sure to run this benchmark after a reboot may help.

Edgemeal i'm not expecting you do to this as you've already done quite abit.

Obviously if you were presenting these numbers to your boss to show why you need to upgrade workstations you would have a more controlled testing environment. However for our general forum I don't see an issue as long as you understand the testing conditions.