Benchmark your computer @4K with Handbrake 1.1 and H265! NOW WITH ZEN3!

Page 13 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

EvilRick

Member
Dec 4, 2003
187
0
86
Handbrake Version: 1.3.3 (2020061300)
LG New York HDR UHD 4K Demo 1.mkv
Preset: H.265 MKV 2160p60
macgui: QueueCore scanning specifically for title: 1
CPU: Intel(R) Core(TM) i7-3720QM CPU @ 2.60GHz
Intel microarchitecture Ivy Bridge
logical processor count: 8

(MacBook Pro 9,1 Mid-2012)

encoded 1806 frames in 1186.52s (1.52 fps), 11821.39 kb/s, Avg QP:29.08
 

LightningZ71

Golden Member
Mar 10, 2017
1,627
1,898
136
Judging by the one Ice Lake result in the table, I am interested to see how Rocket Lake shakes out when it gets tested. It seems that Ice Lake is second only to Zen2 in fps/core/ghz. Tiger lake should be a bit better, and Rocket lake, sharing a core layout, should achieve some impressive throughput. Can't wait to see Zen3 and the M1 show up.
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
I know it's hard to do but if you can report an average frequency during the run it would really help to try to narrow down an IPC figure for these various cores.
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
Judging by the one Ice Lake result in the table, I am interested to see how Rocket Lake shakes out when it gets tested. It seems that Ice Lake is second only to Zen2 in fps/core/ghz. Tiger lake should be a bit better, and Rocket lake, sharing a core layout, should achieve some impressive throughput. Can't wait to see Zen3 and the M1 show up.

Yes I am curious as well. If I had to guess I don't think Rocket Lake/Tiger Lake will show any significant IPC improvement in this benchmark since RL is a port of Ice Lake to 14nm and the larger caches of Tiger Lake (Willow Cove) probably won't impact performance of this sequential type of operation.
 

bbhaag

Diamond Member
Jul 2, 2011
6,604
1,990
146
Used HB 1.1.0 and the preset in the OP. System is an
i7-6700K@4.5 all core OC
16gb of ram@3200

encoded 1806 frames in 552.10s (3.27 fps), 11820.04 kb/s, Avg QP:29.09

I think that result puts me right in the i5-8400 territory maybe a little faster which is really kinda sad if you think about it. For the price I paid for that Skylake i7 it really didn't age well at all. At least in a bench like this.
 
Last edited:

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Is Handbrake working on M1 yet? Natively, without using the fixed function hardware?
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
So many results submitted with so many different versions. OP says which version to use.

It's really hard if not impossible to get everyone to find and load the same version. Plus the underlying transcoding "engine" doesn't really change much.
 

HutchinsonJC

Senior member
Apr 15, 2007
465
201
126
April 2018 Handbrake 1.1.0 - Library was updated to x265 2.6 (H.265/HEVC video encoding)

Dec 2018 Handbrake 1.2.0 - Library was updated to x265 2.9 (H.265/HEVC video encoding)

Nov 2019 Handbrake 1.3.0 - Library was updated to x265 3.2.1 (H.265/HEVC video encoding)

And that's just me looking for the x265 library specifically. There's a multitude of other libraries updated through that time span.
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
April 2018 Handbrake 1.1.0 - Library was updated to x265 2.6 (H.265/HEVC video encoding)

Dec 2018 Handbrake 1.2.0 - Library was updated to x265 2.9 (H.265/HEVC video encoding)

Nov 2019 Handbrake 1.3.0 - Library was updated to x265 3.2.1 (H.265/HEVC video encoding)

And that's just me looking for the x265 library specifically. There's a multitude of other libraries updated through that time span.

If you are interested test a couple versions with this bench and see if you come up with significant performances differences. If you have time/inclination to help out. I'll keep the chart up-to-date.
 

bbhaag

Diamond Member
Jul 2, 2011
6,604
1,990
146
If you are interested test a couple versions with this bench and see if you come up with significant performances differences. If you have time/inclination to help out. I'll keep the chart up-to-date.
Ok so I was a little bored tonight so I went ahead and tested a couple of different version. Obviously my post above was using 1.1.0 as the OP suggested so here's a recap of that score and then of course the other two versions. System is an i7-6700K@4.5 all core OC and 16gb of 3200 ram 15-17-17-35. I just ran both tests back to back no reboot in between or anything like that.

I don't really see much of a difference between 1.1.0 and 1.3.0 but for some reason there was a little uptick when using 1.3.2 roughly 7 seconds I guess.

Version 1.1.0
encoded 1806 frames in 552.10s (3.27 fps), 11820.04 kb/s, Avg QP:29.09

Version 1.3.0
encoded 1806 frames in 550.96s (3.28 fps), 11820.04 kb/s, Avg QP:29.09

Version 1.3.2
encoded 1806 frames in 545.06s (3.31 fps), 11820.04 kb/s, Avg QP:29.09
 

EvilRick

Member
Dec 4, 2003
187
0
86
So many results submitted with so many different versions. OP says which version to use.

Handbrake Version: 1.1.0 (2018040700)
LG New York HDR UHD 4K Demo 1.mkv
Preset: H.265 MKV 2160p60
macgui: QueueCore scanning specifically for title: 1
CPU: Intel(R) Core(TM) i7-3720QM CPU @ 2.60GHz
Intel microarchitecture Ivy Bridge
logical processor count: 8

encoded 1806 frames in 1142.41s (1.58 fps), 11821.39 kb/s, Avg QP:29.08
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
Ok so I was a little bored tonight so I went ahead and tested a couple of different version. Obviously my post above was using 1.1.0 as the OP suggested so here's a recap of that score and then of course the other two versions. System is an i7-6700K@4.5 all core OC and 16gb of 3200 ram 15-17-17-35. I just ran both tests back to back no reboot in between or anything like that.

I don't really see much of a difference between 1.1.0 and 1.3.0 but for some reason there was a little uptick when using 1.3.2 roughly 7 seconds I guess.

Version 1.1.0
encoded 1806 frames in 552.10s (3.27 fps), 11820.04 kb/s, Avg QP:29.09

Version 1.3.0
encoded 1806 frames in 550.96s (3.28 fps), 11820.04 kb/s, Avg QP:29.09

Version 1.3.2
encoded 1806 frames in 545.06s (3.31 fps), 11820.04 kb/s, Avg QP:29.09

Nice work. 1% or less variation among those versions.
 

EvilRick

Member
Dec 4, 2003
187
0
86
So many results submitted with so many different versions. OP says which version to use.

HandBrake 1.1.0 (2018040700) - 64bit
OS: Microsoft Windows NT 10.0.19042.0 - 64bit
CPU: AMD Ryzen 7 2700 Eight-Core Processor
Ram: 16334 MB,
GPU Information:
AMD Radeon RX 5700 XT - 27.20.14501.18003
Screen: 3440x1440

encoded 1806 frames in 391.98s (4.61 fps), 11820.04 kb/s, Avg QP:29.09
 

bbhaag

Diamond Member
Jul 2, 2011
6,604
1,990
146
Nice work. 1% or less variation among those versions.
Thanks, I was thinking the same thing as you as the 1% difference could easily be attributed to slight system differences in between runs on my setup. Anyway, I don't think it's that big a deal if other users report their bench when using a slightly different version of HB.
 

HutchinsonJC

Senior member
Apr 15, 2007
465
201
126
And if no one says anything, the next person will submit results on the next updated version, and the next after that. 1% may not sound like a lot, and it's easy to assume it's just "margin of error", but it's a matter of simple science. You can't do a proper test if your software isn't the same for everyone. I don't think I'm being unfair in pointing it out.
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
And if no one says anything, the next person will submit results on the next updated version, and the next after that. 1% may not sound like a lot, and it's easy to assume it's just "margin of error", but it's a matter of simple science. You can't do a proper test if your software isn't the same for everyone. I don't think I'm being unfair in pointing it out.

You're right 1% isn't a lot.
 

EvilRick

Member
Dec 4, 2003
187
0
86
And if no one says anything, the next person will submit results on the next updated version, and the next after that. 1% may not sound like a lot, and it's easy to assume it's just "margin of error", but it's a matter of simple science. You can't do a proper test if your software isn't the same for everyone. I don't think I'm being unfair in pointing it out.

I agree completely. I wasn't paying attention to the version I downloaded. I DDG'd "handbrake benchmark" and this was the top-most link with some data. I also DDG'd "margin of error" and that's some math I don't want to get into a relationship with.
 

EvilRick

Member
Dec 4, 2003
187
0
86
I know it's hard to do but if you can report an average frequency during the run it would really help to try to narrow down an IPC figure for these various cores.

Are you talking about an external reading, e.g. HWiNFO64? I didn't see anything related to avg. freq. within the log file. I'm OC'd @ 4.1 GHz, so it's always 4.1 GHz.
 

bbhaag

Diamond Member
Jul 2, 2011
6,604
1,990
146
And if no one says anything, the next person will submit results on the next updated version, and the next after that. 1% may not sound like a lot, and it's easy to assume it's just "margin of error", but it's a matter of simple science. You can't do a proper test if your software isn't the same for everyone. I don't think I'm being unfair in pointing it out.
Perhaps but we are talking about a 1% increase over the course of roughly two and a half years. And honestly it was less than 1% if you do the math at least on my setup it was.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
Is Handbrake working on M1 yet? Natively, without using the fixed function hardware?

Yes it's working, as of version 1.4. Linus benchmarked it with H264 and H265 and it did fairly well with H264, but with H265 it lost handily against the Intel Mac Mini and was utterly destroyed by the 5600x. With better optimization that gap should be reduced, but I doubt it will ever make it up significantly with H265.

The hardware acceleration made a huge difference in making up that performance.


 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
That's not an accurate way of showing the clock frequency either. Windows task manager "Speed" is actually very inaccurate when it comes to Zen unless you dumb it down by running an allcore oc. As I type this, task manager is showing 4.2ghz and yet no cores are effectively over 1.5ghz. :rolleyes:


HWinfo can give you average CPU frequency during your run.

Moving forward why don't we set some additional test parameters?

1. Use the following version of Handbrake with the built-in h.265 mkv 2160p60 preset
HandBrake-1.3.3-x86_64-Win_GUI.exe
Once this current version is replaced you'll be able to access it from the following link.

2. Report average effective clockspeed using HWinfo - https://www.hwinfo.com/
I'm sure there are better ways to do this using the logging feature but here's how I did it. I downloaded the portable version, which runs from the exe file.

Run the exe and select "sensors only." You only need this window open.
Minimize the sensors window so you can have it and Handbrake visible at the same time. Select the "average effective clockspeed" line so you can spot it easily. It will highlight.
Note the icon for "reset min/max/average values and timer" in the bottom right area of the window, it's the clock icon.

Start the handbrake but don't start/reset the timer until handbrake is actually encoding video. Record the average effective clockspeed right before handbrake finishes the encode. It should settle in pretty quickly after the encode starts and not change during the encode.

3. Report "Total CPU Usage" (also from HWinfo)

4. Report your encoding time and CPU model
 
Last edited:

EvilRick

Member
Dec 4, 2003
187
0
86
Average Effective Clock: 3907 MHz

Total CPU Usage: 95.2 %

encoded 1806 frames in 393.45s (4.59 fps), 11820.04 kb/s, Avg QP:29.09

CPU: Ryzen 7 2700
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
If you're up for it let's try it with a more rigorous approach. Instructions below if you want to post a score.

RankUserCPUHandbrake Version VerifiedTotal CPU Usage (Hwinfo)Average Effective Clockspeed (Hwinfo)secondsfps/core/GHz (fps per available GHz)fpsCPU CoreCores
1EvilRickRyzen 7 2700x 8/161.3.395.2%3.9103920.1474.61Zen+8
2Hulk4770K 4/81.3.398.1%3.6208150.1532.22Haswell4
3Hulk8250u 4/8 (Surface 2 laptop)1.3.398.3%2.40010650.1771.70Kaby Lake R4

1. Use the following version of Handbrake with the built-in h.265 mkv 2160p60 preset
HandBrake-1.3.3-x86_64-Win_GUI.exe

Once this current version is replaced you'll be able to access it from the following link.
HandBrake: Nightly Builds
Nightly builds of HandBrake

2. Report average effective clockspeed using HWinfo - https://www.hwinfo.com/
I'm sure there are better ways to do this using the logging feature but here's how I did it. I downloaded the portable version, which runs from the exe file.

Run the exe and select "sensors only." You only need this window open.
Minimize the sensors window so you can have it and Handbrake visible at the same time. Select the "average effective clockspeed" line so you can spot it easily. It will highlight.
Note the icon for "reset min/max/average values and timer" in the bottom right area of the window, it's the clock icon.

Start the handbrake but don't start/reset the timer until handbrake is actually encoding video. Record the average effective clockspeed right before handbrake finishes the encode. It should settle in pretty quickly after the encode starts and not change during the encode.

3. Report "Total CPU Usage" (also from HWinfo)

4. Report your encoding time and CPU model
 
  • Like
Reactions: lightmanek