[Norman Black]

I have doubts that TP1 vs TP2/3/4 has anything to do with it. The Velocity envelope affects all video streams the same. Physical media or generated.

Can you show us the exact details of your use of the Vegas velocity envelope as described in this thread?

[Norman Black]

In both examples you need a mask of the subject you want to do the gradient fill or stroke. How you get that mask is quite variable. The subject could have been green screened in capture or you can rotoscope them. In the case of animation the subject might already have a transparent background. Once you have a subject separated with a transparent background it can be transformed into a solid color mask (black or white).

Once you have a masking video of the subject…

In the first example you can create a gradient fill event on a track and set that track as a composite child to your mask track video. Then you typically choose Multiply (mask) blending mode on the masking track. The gradient will only show in the masked area.

In the second example, the “stroke” looks like a simple solid white fill of the mask. The first example had a gradient fill. That mask has an outline on the mask which can be done simply with the Layer Dimensionality effect. In this example the mask was enlarged to be bigger than the subject. The track order on this would be
Subject
Mask
White fill.

The subject is on top of the white filled mask with outline.

[Norman Black]

You have given no information on the source file properties. The Windows properties panel gives almost no information. If you looked around the web a lot, you would know to give information provided by the free MediaInfo utility.

What I can see in the Windows properties panel is that the source file is 800×600 but you have project settings of 1920×1080 with square pixels. Increasing the resolution is not so much as big a deal as the change in aspect ratio from 4:3 to 16:9. Both are question marks.

[Norman Black]

I use driver 15.7 on my 7950.

[Norman Black]

You never mentioned what encoder you are using. That is a critical piece of information don’t you think!?

It sounds like you were using Mainconcept AVC.

One thing I would add is to not set the max bitrate and target bitrate to the same value. This defeats the whole idea of variable bitrate encoding.

[Norman Black]

[Aaron Star] “Peoples hatred of CLI is amazing.”

Well, FFMPEG is a PITA to use so I understand that, but it can do just about anything and most of it’s encoders are actually better than most commercial options.

One thing ffmpeg has a problem with is RGB to YUV conversions which I found out about when I started using ffmpeg to frameserve from Vegas. It was easier to use Avisynth commands than try and figure out how to make ffmpeg do it right.

I use ffmpeg to frameserve from Vegas for encoding.

Here is an example to x264

@echo off

title encode crf23

cd d:\renders
d:

REM output the AviSynth commands
echo AviSource("server.avi") > server.avs
echo ConvertToYUY2(matrix="rec709") >> server.avs
REM echo ConvertToYUY2(matrix="PC.709") >> server.avs

c:\systools\ffmpeg32\bin\ffmpeg.exe -i server.avs -c:v libx264 -preset medium -profile:v high -me_method umh -crf 23 -colorspace bt709 -color_primaries bt709 -color_trc bt709 -
pix_fmt yuv420p -bufsize 40M -maxrate 40M -c:a libmp3lame -qscale:a 2 -chunk_size 64K output.mp4

pause

del server.avs


I have one that encodes ProRes from Vegas as well.

I have other standalone scripts which combine the mono channels in XDCAM and XAVC MXF files so Handbrake can encode them. Handbrake cannot handle the multiple mono channels of MXF.

Cliptoolz Convert is free and while no longer developed is probably the best utility to easily convert some camera format to edit formats like DNxHD, XDCAM and ProRes. Sadly no XAVC Intra, but that came after development stopped.

[Norman Black]

Vegas will use the full res of your photos when you pan/crop. By default Vegas does conform the source to the output dimensions but that is done after pan/crop in the pipeline chain. Therefore pan/crop sees the full source file dimensions.

I have done some slideshow stuff on 4000×3000 jpegs in a 1440×1080 project.

[Norman Black]

Of course Youtube targets different bitrates for 30p and 60p. 30p has half the data of 60p and thus needs less bitrate.

Just going by the Youtube recommendations they give 60p about 50% more bitrate than 30p. Not twice since there is normally much more interframe redundancy in 60p material than 30p and hence one can get a better compression ration before losing visual quality. However, if the 60p is fast moving with high frequency detail even the extra 50% can still be worse. Simply put, if the 60p does not get at least 50% better interframe compression than 30p then the 60p will normally come out worse. Bitrate must be thought of in bitrate per frame and not an absolute number independent of the frame rate.

[Norman Black]

[Lars Pijnboom] “Will encoding at a lower bitrate -say like 12Mbps- give me better results at Youtube?”

Unlikely. At lower bitrates your encode from Vegas will start to breakdown just as it is breaking down with the Youtube encode.

[Lars Pijnboom] “Also: Is there a better codec than x264 for Youtube?”

Youtube could care less what you upload.

I’ll repeat myself. If the video you are uploading to Youtube looks good but the Youtube result breaks down in places there is nothing you can do. The content of the video is simply not able to be compressed enough to maintain a certain visual quality level at the Youtube encode bitrate.

As a test, just take your video and encode it from Vegas from your 50Mbps in various steps down to 8Mbps. Watch each one. You will see the video break down more and more in quality as you lower the bitrate. Each encode started with the exact same “perfect” source. Exactly where and when the breakdowns occur depends on the video content. Some things compress very well. Typical movie and tv content where you have actors standing around talking compresses extremely well and can look good a very low bitrates. Video game videos very commonly break down in quality. Every week we have at least one thread about Youtube video quality from a game video.

Encoding from Vegas with a “perfect” source will show breakdowns when lowering bitrate to certain levels. The same situation occurs when uploading something to Youtube. Giving Youtube an even more perfect source is not going to help their resulting encode be any better. Their encode is doing the best it can given the source material for the bitrate they encode to.

When you tried a 30p encode did you disable resampling?

[Norman Black]

[Lars Pijnboom] “I’m experiencing heavy pixelation on scenes with high motion and lots of details.”

That is quite common and normal. Youtube/vimeo re-encode your video to their spec and that spec is a pretty low bitrate where things like this can happen if the video overall does not get a high compression ratio.

You seem to be encoding to 50Mbps which I’ll bet looks fine when you play it on your system. If the video you upload looks fine but their result has problems there is nothing you can do except adjust the content of the video where those problems occur. 50Mbps is probably around 4x higher bitrate than typically Youtube uses. They recommend 12Mbps uploads for 1080p60.