I did, John, and it all worked fine. However, an interesting development is when I try a longer section than my test (my test piece is about 30 secs), the export stops at approx. 1 min 38 secs and says there's an error and the resulting file may not be editable, so I should disable 'smart render' and export again.

Any ideas on that?

Seamus

@discburn

Hi Seamus

. . . . export stops at approx. 1 min 38 secs . . .

If the export stops that usually means there is an issue with an object on the timeline at that time.

I am correct in assuming that you are not adding any effects?

Is there a change at that time on the timeline - it may be a new object or a cut?

John EB

Hi John,

No, no effect or cut of any kind, not even in the audio.

When the blue export line is moving along, you can see it slow down, struggle, speed up, stop, and then the error dialogue.

I think it may be because the original files are .M2V and I had to create frame tables to be able to use them at all on the timeline.

Seamus

@discburn

Hi Seamus

. . . . When the blue export line is moving along, you can see it slow down, struggle, speed up, stop, and then the error dialogue. . . . .

I have to admit I am stuck on this one - I have never used m2v files - and I would expect VPX to handle the lack of audio in the m2v correctly, as it does in other formats that have a video only stream.

You could try running the m2v file through a video converter to get it into a true mpg file format and the importing that into VPX.

HTH

John EB

Hi

I don't know much about .m2v either but to me as it is an elementary video stream and you have only edited the audio on the timeline and it is smart rendering the video, then:-

The processing is only being carried out on the Audio and being muxed into an MPEG2 program stream ( I did not think this situation would allow Smart render?)

As the video is not being processed ie Smart Rendered, the "Quality" Slider will have no effect as its' function is purely to control the time the encoder spends whilst carrying out its motion search when compressing a frame ("Quality"actually a bit of a misnomer as increasing motion search parameters often results in no increase in quality....could also be called a timewaster!). So yes, all you exports should be the same quality and unchanged from the original.

Peter

@johnebaker

Hi John,

Yes, it must be the m2v not 'really' being mpg that's the problem. I’ll experiment a bit more when I get a chance.

@Scenestealer

Hi Peter,

...Audio and being muxed into an MPEG2 program stream ( I did not think this situation would allow Smart render?)

That's probably true.

the "Quality" Slider will have no effect as its' function is purely to control the time the encoder spends whilst carrying out its motion search when compressing a frame

I always wondered why it defaulted to 10. Who would want a lower 'quality' export? 😕

Seamus

I always wondered why it defaulted to 10. Who would want a lower 'quality' export?

Someone who wanted the encoding to take less time. If the frames' content were very static (a still image say) and without much detail then you could quite likely reduce the slider position, not lose any quality, and speed up the export. Whoever created the preset decided on 10 as giving adequate quality in a reasonable time for averagely complex material. Another way of looking at it is in the Intel Advanced Video Settings where the top is Best, then Better with the lowest Fastest ie not Worst.

@Scenestealer

I'm a '15'/'Best' man myself! 😊

Seamus

@discburn

I, like John EB am a firm believer in sticking to the presets. I have seen worse quality when shifting to 15.

Peter

@Scenestealer

I have seen worse quality when shifting to 15.

Wow, I'm surprised. I would have thought, given the function of the slider, than '15' was the best quality?

Seamus

@discburn

Hi Seamus

. . . . Wow, I'm surprised. I would have thought, given the function of the slider, than '15' was the best quality? . . .

Not really - where is the extra quality coming from - it must be at the expense of something else - usually the video bitrate.

A good explanation is here. I know it is a draft document, however the explanation is the clearest I have ever found.

John EB

@johnebaker

Hi John,

Well, I've done some tests with different types of content...

For the same bitrate:

Quality setting of 5 gives the 'worst' result.

Quality setting of 10 seems to give the most 'accurate' result, ie like the original timeline.

Quality setting of 15 seems to give a 'sharper' look to the edges of objects.

I say 'seems to give', because I find myself looking back at A/B comparison to define the difference.

My brain tells me that '15' should be better, but I guess it isn't really. I'll do a more definitive test on my next project export.

Seamus

@discburn

Hi Seamus

. . . . My brain tells me that '15' should be better, but I guess it isn't really . . . .

That is the major issue in determining quality - without the correct analytical tools to analyse the video every judgement is a perceived one, which can be subconsciously influenced by what 'should be'.

Cheers

John EB

I assess quality by placing the respective tests one above the other in the timeline of a new project and then applying a section to the lower one so that only half of it shows and thereby creates a split screen effect. From there you can view the preview screen at Full Screen or zoom the preview screen further to 150 or 200% using the command in the hamburger menu.