StaxRip

[Magik Mark]

Did you make 1 pass by different version of x265? Could you paste log file from 1 pass and 2 pass?

http://chopapp.com/#e86f177o

[Ma]

@Magik Mark: Thanks for log file. It looks like there is something wrong with first pass. x265 complains about input file.

I've checked your options and it works, but the 'x265_analysis.dat' file is in current working directory -- you specify '--stats' option but not '--analysis-file' option.

My first impression is that it is StaxRip issue with reading/decoding video.

[stax76]

Profile 10 Bit and Color-Range BT2020 for UHD

The QSVEncC help doesn't contain '2020', you can find the mail address of rigaya on his blog.

http://rigaya34589.blog135.fc2.com/blog-category-10.html


Testing build 1.4.0.5 and it appears the keyboard shortcuts in Preview are non-functional. I know I could use the left and right arrow keys to move forward and backward, and I feel sure the up and down arrows moved forward and backward in larger units. In rev 1.4.0.5 none of the keys seems to do anything.


New build is available, in can be considered RC.

I've checked your options and it works, but the 'x265_analysis.dat' file is in current working directory -- you specify '--stats' option but not '--analysis-file' option.

My first impression is that it is StaxRip issue with reading/decoding video.

You mean the x265 command line is incorrect? Can you elaborate? I've not changed something in this area in months.

Since Mark said it only happens sometimes I would guess it has to do with AviSynth multi-threading and there is little I can do, it would probably also happen when the batch file StaxRip generates is run outside of StaxRip.

[NikosD]

Don't forget to add in your first page the addition of AMD H.265 encoder.

[stax76]

Don't forget to add in your first page the addition of AMD H.265 encoder.

I've reset it now, staxrip moves the old video encoder profiles to a backup sub menu.

[NikosD]

I've reset it now, staxrip moves the old video encoder profiles to a backup sub menu.
No, I meant here, the first page of this thread :)

[stax76]

ah, changed that too, thanks.

[Ma]

You mean the x265 command line is incorrect? Can you elaborate? I've not changed something in this area in months.

x265 command line is OK, but a bit danger. If you make multiple encoding at the same time with analysis file, it will be a problem (all encoding use the same file).

If StaxRip works now:
IF '--pass' option THEN
add option '--stats "output_file_name.stats"'
FI

it could/should work like this:
IF '--pass' option THEN
add option '--stats "output_file_name.stats"'
IF '--multi-pass-opt-analysis' OR '--multi-pass-opt-distortion' option THEN
add option '--analysis-file "output_file_name.analysis"'
FI
FI

[Magik Mark]

Thanks Ma for clarifying. I hope stax can incorporate that to the new build

[Magik Mark]

x265 command line is OK, but a bit danger. If you make multiple encoding at the same time with analysis file, it will be a problem (all encoding use the same file).

If StaxRip works now:
IF '--pass' option THEN
add option '--stats "output_file_name.stats"'
FI

it could/should work like this:
IF '--pass' option THEN
add option '--stats "output_file_name.stats"'
IF '--multi-pass-opt-analysis' OR '--multi-pass-opt-distortion' option THEN
add option '--analysis-file "output_file_name.analysis"'
FI
FI

If I'm not mistaken, I think this is also the issue with hybrid since "--multi-pass" CLI has been introduced

[slushay]

please help. staxsrip gives an error at the stage of video encoding. More precisely the error causes the decoder. I tried all the available options.
Error:
Encoding using NVEncC 3.06 x64 failed with exit code: 1 (0x1)
cuvid: Parsed header too large!
failed to init decoder.

script:
C:\Users\Andrey\Downloads\StaxRip-x64\Apps\NVEncC\NVEncC64.exe --cqp 18:20:22 --codec h265 --output-depth 10 --lookahead 32 --qp-init 1 --max-bitrate 30000 --aq --output-buf 128 -i D:\temp\temp.mkv -o D:\temp\temp_temp\temp_out.h265

[stax76]

@Ma

Thanks for the clarification.


@Mark

It will be fixed in the next build.


@slushay

You can report it to rigaya, is mail address is on his blog.

http://rigaya34589.blog135.fc2.com/blog-category-10.html

[slushay]

@slushay

You can report it to rigaya, is mail address is on his blog.

http://rigaya34589.blog135.fc2.com/blog-category-10.html

Thank you for your program. I use myself and recommend to friends.
It is now writing a letter to rigaya.

[Alexander]

Hello Frank,

A few other apps need to be updated :)

x264 v.2762
MediaInfo v.0.7.93
FFmpeg v.3.2.2
DGIndexNV v.2053
Qaac v.2.62
L-Smash-Works (avs) v.924
Java v.121
:thanks:

[NikosD]

ah, changed that too, thanks.

Hello again.

I don't know how hard is it to add in your Media Info tool just one more column.

It's called "Color primaries" and it has a value of BT.2020 when the clip is HDR.
I can see that property in "tree" mode.

[lotnybartek]

NikosD can you tell me what might be the best way to encode 1080i recorded tv channel with many fast motions? Source file is: *.ts, around 6.686 kb/s, 2h long, 6,5GB. I need to produce smaller size file maintaining good quality.

In your opinion, what is the best choice in terms of encoding scheme: LA-VBR Lookahead, VBR only, LC-ICQ (which preset - 17? file is too large), maybe QVBR? Bitrate would be 4.000 kb/s.

I got Intel Core i7-6820HQ.

Barty

[NikosD]

From all of my tests using QSVEnc I had decided that plain ICQ even with no lookahead was totally fine.

You could try --scene-change (but not with ICQ IIRC) and lookahead, but I was happy with plain ICQ.

[lotnybartek]

Thank you. One more thing, source is 1080i, it's 50fps, target is 25fps. Do you think that I can maintain 50fps after encoding? Is it worth it? It'll double the file size Am I right?

[NikosD]

Are you going to convert your interlaced file to progressive ?

Then I think you can't avoid the 50fps to 25fps conversion.

But IIRC, QSVEnc supports interlaced encoding and you could leave it interlaced at 50fps

[lotnybartek]

Yes, 50 fps to 25.

I'm wondering about one thing. If source is 50fps 6700 kb/s, so If I divide fps by 2, Can I also divide bitrate by 2 ? Afterall, I'm encoding half of the source fps - so I should get similar quality.

[stax76]

Hello again.

I don't know how hard is it to add in your Media Info tool just one more column.

It's called "Color primaries" and it has a value of BT.2020 when the clip is HDR.
I can see that property in "tree" mode.

I build a much better tool from scratch!

https://i.imgur.com/rIHzOjj.png

[NikosD]

Yes, 50 fps to 25.

I'm wondering about one thing. If source is 50fps 6700 kb/s, so If I divide fps by 2, Can I also divide bitrate by 2 ? Afterall, I'm encoding half of the source fps - so I should get similar quality.

It's not as simple as that actually.

You aren't encoding half of the source, you deinterlace the source making 1 frame from 2, which is different.

Anyway, you could try half bitrate and see for yourself if you are happy with the result, but don't expect to get exactly the same quality as the source - it's not the same.

I build a much better tool from scratch!

https://i.imgur.com/rIHzOjj.png

I have downloaded latest test build v1.4.0.6 but it's not inside (I think)

Are going to release it as a separate tool or inside StaxRip ?

It looks good, BTW.

[Zetto]

Is there any way to tell nvenc which nvidia videocard to use? If there are more than one present, of course.

[DaMacFunkin]

Hello i have just upgraded to
StaxRip x64 1.4.0.6 unstable test build
I am using Nvenc x265 vbr, the vbr setting seems to be having no effect, i have encoded a movie with --vbr-quality 27 and it is the same size as 21, 23, 25.
Please advise, it worked fine on the test build from January?

[stax76]

Are going to release it as a separate tool or inside StaxRip ?

It's a separate tool that will be included in staxrip.

Is there any way to tell nvenc which nvidia videocard to use? If there are more than one present, of course.

Next build will have a device option. Meanwhile, you can add a custom option (--device <int>).

I am using Nvenc x265 vbr, the vbr setting seems to be having no effect

Search the thread for '--vbr-quality', I think you have to raise the bitrate in the main dialog as it acts as a constraint.

[Zetto]

Next build will have a device option. Meanwhile, you can add a custom option (--device <int>).Thanks for the response! :)

[AMED]

it appears that's HAvsFunc.py needs to be updated to support changes to KNLMeansCL so prefilter 4 can be used with SMDegrain.

https://raw.githubusercontent.com/HomeOfVapourSynthEvolution/havsfunc/master/havsfunc.py

[lotnybartek]

Can I use SW deinterlacing, let's say QTGMC and in the same time HW x264 encoding using QuickSync?

[WIGGER]

Dear all,

I have a question regarding usage of StaxRip with 2 x 6-core (12-threads) processor. I'd like to buy a Working station dedicated only to video encoding with StaxRip/x264 and I'd like to ask for your help/advice. I'm planning on 2 x Intel Xeon X5675 with 24 GB DDR3 ECC and nVidia GTX 480 card in HP Z800.

As far as I read --threads should call 1,5 x number of logical CPU but does it apply to both CPUs or not? In my case it would be 9 with one CPU and 18 with two. I also read that 10 is a recommended maximum for --threads.

So, my problems are two:

1. Is there any danger that 2 CPU could cause improper encoding with StaxRip/x264? I'm going to do a big investment and it would be stupid to end with a failure.
2. Should I expect reduced encoding time and better encoding quality with 2 CPUs or I can keep to my exisitng PC with i5-2500k CPU and to spend the money on beer?

Thank you!

[NikosD]

If you are going to invest on CPU encoding using x264, you should forget Intel.

Intel is dead regarding multicore x264 encoding.

Grab an 8C/16T RyZen R7 processor or wait for Naples which will go up to 32C/64T and even in dual socket configuration.

[WIGGER]

@NikosD - this was not the answer I needed/expected, but you are free to share it. I hope someone else can give more useful reply.

[Atak_Snajpera]

Personally I wouldn't spend any money on such old architecture
http://i.imgsafe.org/d948151608.png

results from x264 FHD Benchmark
40.0 fps - Intel i7-3960X @ 5GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
37.7 fps - Intel i7-3930K @ 4.8GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
35.5 fps - 2 x Intel Xeon X5670 @ 2.93GHz ( 6C / 6T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
33.1 fps - Intel i7-3930K @ 4.4GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
32.9 fps - Intel Xeon E5-2690 @ 2.9GHz ( 8C / 16T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
32.9 fps - Intel Core i7 970 @ 4.5GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
31.4 fps - Intel Core i7 980 @ 4.4GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
29.7 fps - AMD FX-8350 @ 5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
28.1 fps - Intel i7-2600K @ 5.2GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
27.9 fps - Intel i7-2600K @ 5.1GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
27.7 fps - Intel i7-3930K @ 3.2GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
27.7 fps - Intel i7-2600K @ 5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
27.7 fps - 2 x Intel Xeon X5670 @ 2.93GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
26.8 fps - AMD FX-9590 @ 4.73GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
26.8 fps - AMD FX-8320 @ 4.62GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
26.5 fps - AMD FX-8150 @ 5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
26.1 fps - Intel i7-2600K @ 4.9GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
25.8 fps - Intel Core i7 X 980 @ 3.57GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
25.0 fps - Intel i7-2600K @ 4.5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
24.0 fps - AMD FX-8120 @ 4.5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
23.8 fps - Intel i7-2600K @ 4.3GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
23.5 fps - AMD FX-8350 @ 4GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
23.2 fps - Intel Xeon W3670 @ 3.2GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
23.0 fps - Intel Core i7-2600K @ 4.2GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
22.8 fps - Intel Core i7-3820 @ 3.6GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
22.5 fps - Intel Core i7-4770S @ 3.1GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
22.3 fps - 2 x Intel Xeon X5550 @ 2.67GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
22.1 fps - Intel Core i7-2600K @ 4GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
21.7 fps - Intel Core i7-3770K @ 3.5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
21.3 fps - AMD FX-8150 @ 4GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
20.6 fps - Intel Xeon E3-1230 V2 @ 3.3GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
20.6 fps - AMD Phenom II X6 1090T @ 4GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
20.6 fps - 2 x Intel Xeon E5450 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
20.1 fps - Intel Core i7-2700K @ 3.5GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
20.1 fps - Intel Core i5-2500K @ 4.5GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
20.0 fps - Intel Core i7 920 @ 4GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
19.8 fps - AMD Phenom II X6 1090T @ 3.85GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
19.7 fps - Intel Core i5-2500K @ 4.4GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
19.6 fps - Intel Core i7 950 @ 3.68GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
19.4 fps - Intel Core i7-2600K @ 3.4GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
19.2 fps - 2 x Intel Xeon X5365 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
19.0 fps - AMD Phenom II X6 1100T @ 3.7GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
18.6 fps - Intel Xeon E5645 @ 2.4GHz ( 6C / 12T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
18.5 fps - AMD Phenom II X6 1090T @ 3.6GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
18.2 fps - 2 x Intel Xeon E5430 @ 2.67GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
18.0 fps - AMD Phenom II X6 1055T @ 3.5GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
17.9 fps - AMD FX-6100 @ 4.4Ghz ( 3C / 6T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
17.8 fps - AMD FX-8120 @ 3.1GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
16.9 fps - AMD FX-6100 @ 4.2GHz ( 3C / 6T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
16.8 fps - AMD Phenom II X6 1100T @ 3.3GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
16.4 fps - AMD Phenom II X6 1090T @ 3.2GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
16.3 fps - 2 x Intel Xeon E5506 @ 2.13GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
16.0 fps - Intel Core i5 750 @ 4GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
15.7 fps - Intel Core i5-2500K @ 3.3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
15.2 fps - Intel Core2 Quad Q9650 @ 4GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
15.0 fps - AMD Phenom II X6 1075T @ 3GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
14.9 fps - Intel Core i5 760 @ 3.5GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
14.7 fps - Intel Core i5-2400 @ 3.1GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
14.7 fps - Intel Core i7 930 @ 2.8GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
14.6 fps - AMD Phenom II X6 1055T @ 2.8GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
14.4 fps - Intel Xeon E31230 @ 3.2GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
14.3 fps - AMD Phenom II X6 1045T @ 2.7GHz ( 6C / 6T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
14.0 fps - Intel Core i7 920 @ 2.67GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
14.0 fps - AMD Phenom II X4 965 @ 4GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
13.6 fps - Intel Core2 Quad Q9650 @ 3.8GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
13.6 fps - AMD FX-4100 @ 5GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
13.4 fps - AMD FX-6100 @ 3.3Ghz ( 3C / 6T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
13.0 fps - Intel Xeon E5450 @ 3.6GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
12.6 fps - AMD A8-3850 APU @ 3.5GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
12.6 fps - Intel Core2 Quad Q6600 @ 3.6GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
12.4 fps - Intel Core2 Extreme X9650 @ 3.33GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
12.3 fps - AMD FX-4100 @ 4.75GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
12.2 fps - Intel Core i7 860 @ 2.8GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
12.0 fps - AMD FX-4100 @ 4.5GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX XOP FMA4 SSEMisalign LZCNT
11.9 fps - Intel Core2 Quad Q6600 @ 3.46GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
11.8 fps - Intel Core2 Quad Q9400 @ 3.2GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
11.7 fps - AMD Phenom II X4 965 @ 3.4GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
11.6 fps - 2 x Intel Xeon X5260 @ 3.33GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
11.2 fps - Intel Core2 Extreme X9650 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
11.1 fps - Intel Core2 Quad Q8300 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
11.0 fps - Intel Core i5 750 @ 2.66GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
10.9 fps - Intel Core i7-2630QM @ 2GHz ( 4C / 8T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
10.9 fps - 2 x Intel Xeon E5310 @ 1.6GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
10.8 fps - Intel Core i5 750 @ 2.66GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
10.5 fps - AMD Phenom II X4 945 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
10.4 fps - Intel Core2 Quad Q6600 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
10.2 fps - Intel Core2 Quad Q8200 @ 2.8GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
10.1 fps - AMD Athlon II X3 455 @ 3.85GHz ( 3C / 3T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
10.0 fps - AMD Athlon II X4 640 @ 3GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
9.5 fps - AMD A6-3400M APU @ 2.7GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
9.2 fps - Intel Core i3-2120 @ 3.3GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
9.2 fps - Intel Core2 Quad Q8300 @ 2.5GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
9.1 fps - AMD Phenom II X4 925 @ 2.8GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
8.7 fps - Intel Core2 Quad Q8200 @ 2.33GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
8.6 fps - Intel Core i3-2100 @ 3.1GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
8.6 fps - AMD Phenom 9950 @ 2.6GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
8.5 fps - Intel Core2 Quad Q6600 @ 2.4GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 Cache64
8.2 fps - AMD Athlon II X3 450 @ 3.2GHz ( 3C / 3T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
8.0 fps - AMD Phenom 9750 @ 2.4GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
7.8 fps - AMD A6-3400M APU @ 2.2GHz ( 4C / 4T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
7.6 fps - Intel Core2 Duo E8200 @ 4GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
7.6 fps - Intel Core i5-2430M @ 2.4GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
7.5 fps - Intel Core i5-2410M @ 2.3GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
7.5 fps - Intel Pentium G2120 @ 3.1GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
7.4 fps - Pentium Dual-Core E5400 @ 4.16GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle Cache64
7.2 fps - Intel Core i3-2120T @ 2.6GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
7.1 fps - AMD Phenom II X3 720 @ 2.8GHz ( 3C / 3T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
6.9 fps - Intel Core i5-3317U @ 1.7GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
6.7 fps - Intel Core2 Duo E8400 @ 3.6GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
6.2 fps - Pentium Dual-Core E6300 @ 3.5GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle Cache64
5.8 fps - AMD A6-3500 APU @ 2.1GHz ( 3C / 3T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
5.6 fps - AMD Phenom II X2 555 @ 3.2GHz ( 2C / 2T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
5.4 fps - Intel Core i3-3217U @ 1.8GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX
5.3 fps - Intel Core i3 M 330 @ 2.13GHz ( 2C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
5.3 fps - AMD Phenom II N830 @ 2.1GHz ( 3C / 3T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
5.2 fps - AMD Athlon II X2 250 @ 3GHz ( 2C / 2T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
5.1 fps - Intel Celeron G530 @ 2.4GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
4.8 fps - Intel Pentium Dual E2220 @ 3GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 Cache64
4.4 fps - AMD A4-3300 APU @ 2.5Ghz ( 2C / 2T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
4.4 fps - Intel Core2 Duo T9300 @ 2.5GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.1 Cache64
3.8 fps - Pentium Dual-Core T4400 @ 2.2GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle Cache64
3.4 fps - Pentium Dual-Core T4200 @ 2GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle Cache64
3.3 fps - AMD Athlon Dual Core 4450e @ 2.4GHz ( 2C / 2T ) MMX2 SSE2Slow SlowCTZ
3.0 fps - AMD Athlon II x2 M300 @ 2Ghz ( 2C / 2T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT
1.7 fps - AMD E-350 @ 1.6GHz ( 2C / 2T ) MMX2 SSE2Fast SSSE3 FastShuffle SSEMisalign LZCNT
1.5 fps - VIA Nano X2 L4050 @ 1.4Ghz ( 2C / 2T ) MMX2 SSE2Fast FastShuffle SSEMisalign LZCNT

Ryzen 7 1700@3.7GHz will be definitely faster than those two old Xeons.

[NikosD]

Personally I wouldn't spend any money on such old architecture
http://i.imgsafe.org/d948151608.png
Nice table, RyZen is missing though (probably because it's too new architecture)

I would like to see SSE2 integer data and AVX2 integer data too which are more useful to x264/x265 than FP.

Is there any link for that table ?

[Atak_Snajpera]

http://forum.pclab.pl/topic/1105978-FlopsCPU-klasyczny-benchmark-z-1992-roku-w-nowej-oprawie/

[NikosD]

What a coincidence !

I have done the same thing, recompiling that old flops.c source code with MS VC++ and ICC compiler but I don't have that much hardware to test it.

Also, IIRC it's a single threaded app.

How did you manage to make it multithreaded ?

[Atak_Snajpera]

How did you manage to make it multithreaded ?
I simply run multiple instances at the same time and then add MFLOPS(4) results.

[WIGGER]

@Atak_Snajpera - thank you so much for this clear reply. I know it is an old stuff, but I can't afford much more money for other equipment.
I see you are highly experienced in x264 encoding so my last request is to tell me what would be the difference between current i5-2500k and the 2 x X5675? Would be there an improvement or no big change is expected? Reason to aks is because of your post above:

15.7 fps - Intel Core i5-2500K @ 3.3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX

which is lower than

35.5 fps - 2 x Intel Xeon X5670 @ 2.93GHz ( 6C / 6T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
(closest to my desired X5675)

[NikosD]

He has already replied to you that a 330$ CPU - Ryzen R7 1700@3.7GHz - is faster than those two xeons.

Why do you insist on that configuration ?

[Atak_Snajpera]

Wigger what is your budget?

[WIGGER]

Well, this one would cost 700 EUR because it is HP Z800 workstation. CPU itself is about 40-50 EUR because is old.

[Atak_Snajpera]

Well, this one would cost 700 EUR because it is HP Z800 workstation. CPU itself is about 40-50 EUR because is old.

Why don't you buy regular desktop CPU like RyZen 1700 (?!).
Example prices in my country

Motherboard ASUS PRIME B350-PLUS = ~100 EURO (with VAT)
RyZen 7 1700 = ~370 EURO (with VAT)
RAM 2x8 GiB DDR4 2400MHz = ~130 EURO (with VAT)

[WIGGER]

Well, I don't know. I just decided that a dual CPU workstation would perform better than a desktop PC.
Prices in Bulgaria are very common to Poland, so I can really do such an upgrade.
Thank you for all your efforts and explanation. Have a nice evening!

[Atak_Snajpera]

15.7 fps - Intel Core i5-2500K @ 3.3GHz ( 4C / 4T ) MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2 AVX

- Ryzen has atleast 1.1x better IPC than SandyBridge.
- RyZen 1700@3.7 GHz has 1.12x higher clock.
- RyZen 1700 has 2x more cores
- RyZen 1700 has SMT so another 1.2x boost

So 15.7 fps * 1.1 * 1.12 * 2 * 1.2 = 46.4 fps (estimated performance)

[WIGGER]

I have no words to thank you for your kind support! I already put proposed configuration in my shopping basket. I hope to order it soon and I'd be glad to share the results with you.

[NikosD]

http://forum.pclab.pl/topic/1105978-FlopsCPU-klasyczny-benchmark-z-1992-roku-w-nowej-oprawie/

I did some comparison tests with my binaries today and found out that mine are a little faster.

I have compiled the original source with ICC 11.1 on August 2010, with MS VC++ 2013 and with ICC 13.1 on August 2014.

It seems that August is my month for such things.

Anyway, my executables from ICC 13.1 are 2.5x bigger than yours (~106KB), but I really don't remember the exact settings in Intel's compiler.

Using Sandybridge Core i5-2400 and AVX instructions and Core i3-4170 and AVX2 instructions my binaries are ~12% faster than yours and I saw that bigger differences existed in Modules 3 and 4.

Of course someone could replace your binaries with his, keeping the excellent GUI as it is, I think.

I will check your page for updates, but if you see a RyZen result, please give me a heads-up in a private message.

It's my last post here regarding this subject, since it's clearly off-topic in StaxRip's thread.

[stax76]

@Alexander

I've updated the apps except for DGDecNV where my license for some reason has stopped working, just press F12 and enter the version you are using.

@Nikos

The new MediaInfo app didn't work out well, the new control (DataGridView) was too slow and I also discarded the idea to store the data so it's fine as it is, I added your requested property and fixed a few issues.

new RC test build:

https://github.com/stax76/staxrip/blob/master/md/test-build.md

[NikosD]

new RC test build:

https://github.com/stax76/staxrip/blob/master/md/test-build.md

Thank you Frank, it works fine.

Don't forget to add in the Requirements section of the first page of this thread, that you need a Polaris card for AMD H.265 encoding.
No previous AMD card can do that.

[easyfab]

- Ryzen has atleast 1.1x better IPC than SandyBridge.
- RyZen 1700@3.7 GHz has 1.12x higher clock.
- RyZen 1700 has 2x more cores
- RyZen 1700 has SMT so another 1.2x boost

So 15.7 fps * 1.1 * 1.12 * 2 * 1.2 = 46.4 fps (estimated performance)

isn't it x4 ? 4T vs 16T ?

[Alexander]

@Alexander

I've updated the apps except for DGDecNV where my license for some reason has stopped working, just press F12 and enter the version you are using.



Thanks, I've seen it in the new test build
My DGDecNV 2053 r170305 works fine

[need4speed]

Hi all, quick question on compressibility test for x265: is it worth it?
If so can I set some parameters or it is supposed to run as is and suggest settings?
Many thanks in advance!

Inviato dal mio GT-N7100 utilizzando Tapatalk