frameserve from Vegas to HCEncGUI?

[ecc]

is there anyway to frameserve a Vegas project to HCEncGUI? Debugmode FrameServer outputs an avi file that you can open in encoders like CCE and works fine (as long as you keep Sony Vegas open) but HCEncGUI needs an avs or d2v project. I tried making an avs with the FrameServer avi as an avisource but HCEncGUI wouldn't open it.

Are there any other frameserver programs?

[guns1inger]

That is the only option for Vegas AFAIK. Have you tried opening the signpost file using DirectShowSource() instead ?

What happens when you open the avs file in virtualdub ?

[ecc]

Whoops. It does work. Just misspelled the input file name in the script. Works fine although HCEncGUI tells you to add converttoyv12() even though I believe YV12 (or is it YV2) was selected in the FrameServer options. As with CCE, frameserving from Vegas seems slower than using an avs although maybe its because I used a filter in Vegas.

[guns1inger]

Also be aware that if you are doing multi-pass rendering you will be applying the filters on each pass. I have found that if I have applied a number of filters it is often faster to do a single pass render to a lossless format such as lagarith, then encode this file to mpeg-2 using HCEnc. A real life example - 11 hours per pass (heavy noise reduction with Neat video, colour correction and a few other bits and pieces) x 2 passes - over 22 hours in total, versus a single pass to lagarith (11 hours), plus 2 hours to encode to mpeg-2 (2 passes) - 13 hours in total. If you don't have the disc space (the lossless file was 60GB), then consider DV as an intermediate stage - Sony's DV codec is very good.

Sometimes frameserving isn't the answer.

[ecc]

Also be aware that if you are doing multi-pass rendering you will be applying the filters on each pass. I have found that if I have applied a number of filters it is often faster to do a single pass render to a lossless format such as lagarith, then encode this file to mpeg-2 using HCEnc. A real life example - 11 hours per pass (heavy noise reduction with Neat video, colour correction and a few other bits and pieces) x 2 passes - over 22 hours in total, versus a single pass to lagarith (11 hours), plus 2 hours to encode to mpeg-2 (2 passes) - 13 hours in total. If you don't have the disc space (the lossless file was 60GB), then consider DV as an intermediate stage - Sony's DV codec is very good.

Sometimes frameserving isn't the answer.

It was a 9 minute 55 second film (once I clipped the five seconds of color bars). Frameserving to HCEncGUI of the AVI with some color filtering took 1 hour 47 mintues and 23 seconds. Is that fast or slow considering the length of the film?

[guns1inger]

It seems slow, but then your computer specs indicate a pretty old and under-powered system, so the numbers may well be right. 2-pass encoding ?

[ecc]

Yes, two pass encoding and an underpowered system.

Is a statistical pass necessary when your average bitrate and maximum bitrate are the same?

Since this is the video encoding forum, I'll ask here instead of creating a new topic: what determines the allocated memory in HCEncGUI? I see that sometimes in the status bar during encoding (I think it was last set at 56 MB). Will adjusting that make encoding go faster? I seem to remember that HCEncGUI used to encode faster with avs sources of d2v projects.

[guns1inger]

If you set your average and your max to be the same you might as well do a Constant Bitrate (CBR) encode instead. One pass = half the time.

You cannot compare the performance of Vegas through an external frameserver to taking AVS projects from D2V sources. It is apples and oranges as far as conversion goes. You would probably find that the whole workflow would be faster if you did only your editing in Vegas, output as a DV avi file, then complete your other processing in avisynth and out to HCEnc. This is especially true if your input source is DV to begin with, less so if it something like mpeg-2.

[ecc]

You cannot compare the performance of Vegas through an external frameserver to taking AVS projects from D2V sources. It is apples and oranges as far as conversion goes. You would probably find that the whole workflow would be faster if you did only your editing in Vegas, output as a DV avi file, then complete your other processing in avisynth and out to HCEnc. This is especially true if your input source is DV to begin with, less so if it something like mpeg-2.

The input was DV 29.97 but it was dropped into a NTSC DV 24P progressive project (as per a tutorial I was trying out). Wouldn't a DV NTSC output add pulldown (as opposed to the pulldown flag)?

[minidv2dvd]

this does beg the question of why would you frameserve to henc at all. the included vegas mainconcept mpg encoder is superior to henc. i could see a video freak maybe frameserving to cce sp2 but even that would be pushing it, considering the amount of time wasted.

[ecc]

I like HCEnc. Simple as that.

[guns1inger]

You cannot compare the performance of Vegas through an external frameserver to taking AVS projects from D2V sources. It is apples and oranges as far as conversion goes. You would probably find that the whole workflow would be faster if you did only your editing in Vegas, output as a DV avi file, then complete your other processing in avisynth and out to HCEnc. This is especially true if your input source is DV to begin with, less so if it something like mpeg-2.

The input was DV 29.97 but it was dropped into a NTSC DV 24P progressive project (as per a tutorial I was trying out). Wouldn't a DV NTSC output add pulldown (as opposed to the pulldown flag)?

Or use a lossless intermediate, such a lagarith.

this does beg the question of why would you frameserve to henc at all. the included vegas mainconcept mpg encoder is superior to henc

Arguable, and dependent on the target outout. HCEnc excels at lower bitrate rendering, beating CCE for quality and lack of macro-blocking. Blow for blow I think you would find HCEnc is pretty damn close to Mainconcept.

[minidv2dvd]

i've used everything from bbmpeg to cce-sp2. but to each his own.

[JasonK]

this does beg the question of why would you frameserve to henc at all. the included vegas mainconcept mpg encoder is superior to henc.

I was just using Sony DVD architect 4, which has the mainconcept mpeg2 encoder. The results are too soft. DVD architect doesn't give you much control for the mpeg render. I fit to disk with the optimize feature. I compared to TMPGEnc using the same bitrate, and TMPGenc's output looks almost like the source, not soft like the mainconcept mpeg2 file. I read others state that the mainconcept encoder gives a soft look also. So I'm going to stick with using TMPGenc and import those files into the authoring app.