Problems with rendered bitrate Magix AVC/AAC MP4 MainConcept

john-rappl wrote on 2/19/2020, 11:11 AM

Using VP 17(387), Intel i7-8700K, Intel HD 630, 14 second 4K source video file from Sony NX80. This always happens no matter source format or length with or without any editing. To reproduce just drop any file on the timeline and render...

Using the Magix AVC/AAC MP4 templates if I select MainConcept the rendered file bitrate does not match the selected template. The rendered output appears to match the Max rate not the average!

For example if I select the built-in Internet 1080p 29.97 and change the bitrates to 10,000,000 max and 6,000,000 avg the rendered output mp4 actually is encoded (as reported by MediaInfo) at a bitrate of 10.1Mbps with a maximum bitrate of 10,000 kbps.

If I encode with everything exactly the same except changing the Encode Mode in the template from Mainconcept to Intel QVC the resulting mp4 file matches the template! The resulting file has a Bitrate of 6,000 kbs and a max of 9999 kbs.

FYI: It's not a Mediainfo problem as the file sizes verify the higher bitrate (10Mbps vs 6) with a much larger file! In my about 14 second sample the QSV encoded file is about 10.5 MB while the Mainconcept is 17.6 MB.

I've been using Vegas Pro for years. It seems like Sony VP 13 did not have this problem, several of the VP15 patches including the last one did have this problem. I believe the original and previous VP 17s versions did not have this problem and now it is back again in 387.

I'd like to just use either the QSV or AMD encoding as it is much faster but I see a decrease in rendered image quality that I am not willing to accept.

Comments

fifonik wrote on 2/19/2020, 2:46 PM

When I investigated Magix/Mainconcept AVC issue in the past I rendered using these codec (CPU only) from VP11, 12, 13, 15 and have not noticed this. Bitrate in output was as specified. However, I was using FullHD, not 4K and have not tried VP17.

P.S For quality you should try Voukoder or HappyOtterScripts with x264 encoder. Faster than Magix AVC CPU and better quality. Not as fast as NVEnc/QSV/AMF.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B650P, CPU: AMD Ryzen 9700X, RAM: G'Skill 32 GB DDR5@6000, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

john-rappl wrote on 2/23/2020, 9:29 AM

Does anyone else see this as a problem? I target specific bitrates for my output for different customer requirements and usage. I ask for the required bitrate in the template I'm using and I DON'T get it. It would be ok if it was off by a little (10%) but it's not even close at over 30% off!

I have a sample below but it happens on every Magix AVC render I do no matter the project! I did try it with the Magix HEVC templates and it's much better, I just cannot always deliver H.265 media files.

I ran a couple more test to illustrate. I have a 30 sec 4k/100Mbps clip right out of a Sony NX-80. Drop it on the timeline and render:

Magix AVC/AAC MP4 using the built-in template Internet HD 1080P 29.97 (bitrates are set to 12Mbps avg, 24 Max)

Mainconcept (CPU only): Render Time: 0:30, output MP4 size: 61MB, bitrate - 16 Avg, 24 Max!

QSV: Render Time: 0:25, output MP4 size: 34MB, bitrate - 8 avg, 24 Max!!

AMD VCE: Render Time: 0:09, output MP4 size: 43MB, bitrate - 11.8 (no max specified)

I can make AVG and MAX the same and the output is much closer to what was requested but I feel like I'm limiting the encoder. I could also choose constant but again - limiting the encoders ability to vary the bitrate based on content.

john_dennis wrote on 2/23/2020, 1:02 PM

"Using the Magix AVC/AAC MP4 templates if I select MainConcept the rendered file bitrate does not match the selected template. The rendered output appears to match the Max rate not the average!"

How closely the output file matches the target bit rate is directly related to the complexity of the material you're trying to encode.

Example: Same length video, same default settings for the Magix AVC/AAC Internet Mainconcept Render Template, but different complexity of the source video.

Vegas 17-421

Edit: Updated graphic to include video quality measurements.

john_dennis wrote on 2/23/2020, 1:08 PM

AMD VCE exposes more control options...

john-rappl wrote on 2/23/2020, 4:25 PM

I understand there are options but if you request a bitrate that should be a target. I don't recall having this issue with VP15. I think it was a problem with the first VP17 release but was better in a couple of the patches, just to come back worse in 421. I've used a lot of different Video Editing (pro and consumer) and they all get much closer to template requested bitrates. In addition, the Magix HEVC encoders seem to do a much better job.

 

JN- wrote on 2/23/2020, 7:24 PM

@john-rappl I checked this out using VP16, Nvidia Nvenc encode and Mainconcept, CPU only.

I used a 4K uhd clip rendered to FHD. at max/Avg of 28/20.

The nvenc encode was na/18.6, close enough to 28/20 and the MC CPU encode was close also at 28/21.4.

Mediainfo doesn’t report any max bit rate for the nvenc encode, so I entered n/a above, just the bit rate and the overall bit rate, (i.e. video+audio bit rate), although the encoded file is reported as of type variable.

Mediainfo does report the bit rate and max bit rate for the cpu encoded file, see above.

 

The bit rates above don’t include audio bit rate.

Last changed by JN- on 2/23/2020, 7:35 PM, changed a total of 2 times.

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

VFR2CFR, Variable frame rate to Constant frame rate link to zip here.

Copies Video Converts Audio to AAC, link to zip here.

Convert 2 Lossless, link to ZIP here.

Convert Odd 2 Even (frame size), link to ZIP here

Benchmarking Continued thread + link to zip here

Codec Render Quality tables zip

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

PC ... Corsair case, own build ...

CPU .. i9 9900K, iGpu UHD 630

Memory .. 32GB DDR4

Graphics card .. MSI RTX 2080 ti

Graphics driver .. latest studio

PSU .. Corsair 850i

Mboard .. Asus Z390 Code

 

Laptop… XMG

i9-11900k, iGpu n/a

Memory 64GB DDR4

Graphics card … Laptop RTX 3080

Former user wrote on 2/23/2020, 8:44 PM

Does anyone else see this as a problem? I target specific bitrates for my output for different customer requirements and usage. I ask for the required bitrate in the template I'm using and I DON'T get it. It would be ok if it was off by a little (10%) but it's not even close at over 30% off!

 

That is to be expected, only a 2pass VBR would give the accuracy of average bitrate really being the one you select. Try 2 pass with Magix AVC and other codecs that allow it

fifonik wrote on 2/23/2020, 11:50 PM

I cannot confirm the issue in VP17 421.

I've just put 4K-25p file onto timeline and rendered it using Magix AVC twice.

1. Took template "Internet UHD 2160p 25 fps", changed avg bitrate to 10 000 000, max bitrate to 20 000 000. All other render params are default:

Rendered and got 10 000 000 avg bitrate in resulting file:

Video
ID                                       : 2
Format                                   : AVC
Format/Info                              : Advanced Video Codec
Format profile                           : High@L5.1
Format settings                          : CABAC / 4 Ref Frames
Format settings, CABAC                   : Yes
Format settings, Reference frames        : 4 frames
Format settings, GOP                     : M=4, N=15
Codec ID                                 : avc1
Codec ID/Info                            : Advanced Video Coding
Duration                                 : 1 min 3 s
Bit rate mode                            : Variable
Bit rate                                 : 10.0 Mb/s
Maximum bit rate                         : 20.0 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 25.000 FPS

2. Took template "Internet HD 1080p 25 fps", changed avg bitrate to 10 000 000, max bitrate to 20 000 000. All other render params are default:

Rendered and got 10 000 000 avg bitrate in resulting file:

Video
ID                                       : 2
Format                                   : AVC
Format/Info                              : Advanced Video Codec
Format profile                           : Main@L4
Format settings                          : CABAC / 4 Ref Frames
Format settings, CABAC                   : Yes
Format settings, Reference frames        : 4 frames
Format settings, GOP                     : M=4, N=15
Codec ID                                 : avc1
Codec ID/Info                            : Advanced Video Coding
Duration                                 : 1 min 3 s
Bit rate mode                            : Variable
Bit rate                                 : 10 000 kb/s
Maximum bit rate                         : 20.0 Mb/s
Width                                    : 1 920 pixels
Height                                   : 1 080 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 25.000 FPS

As I already mentioned, I did quite a lot tests with previous VP versions. I compared different encoders on the similar bitrate. Never saw such issue. I could not miss it.

 

Have you created your render templates based on new VP17 render templates or using your old templates that were created in previous version of VP?

Last changed by fifonik on 2/24/2020, 12:01 AM, changed a total of 3 times.

Camcorder: Panasonic X1500 + Panasonic X920 + GoPro Hero 11 Black

Desktop: MB: MSI B650P, CPU: AMD Ryzen 9700X, RAM: G'Skill 32 GB DDR5@6000, Graphics card: MSI RX6600 8GB, SSD: Samsung 970 Evo+ 1TB (NVMe, OS), HDD WD 4TB, HDD Toshiba 4TB, OS: Windows 10 Pro 22H2

NLE: Vegas Pro [Edit] 11, 12, 13, 15, 17, 18, 19, 22

Author of FFMetrics and FFBitrateViewer

john_dennis wrote on 2/24/2020, 12:30 AM

@john-rappl

"I don't recall having this issue with VP15."

I ran my complexity tests on Vegas Pro 15-416 and the conformance to target bit rate was worse in Vegas Pro 15 than my earlier tests with Vegas Pro 17-421.