Vegas - rendered video blurry
I bought Sony HDR CX405B.CEN camcorder because I wanted to capture family events etc. I am pretty satisfied with picture quality but the problem is when I want to render video. I have many questions but I will try to focus on the main problem :).
Camcorder captures video in 1920x1080 50i. That setting is recommended (but it also can record in 1920x1080 50p). I want to do some basic video trim, add title etc. so I have tried Vegas pro. I like software but I have problem with rendering video out. I want to render it to format which is good for viewing on a TV and on computer and which is easy to store (video and audio in one file). It supposed to be .MP4 file format (or maybe other?). The video size is not the problem because I will not record video very often.
Render settings are 1920x1080 25p. I tried XAVCS, XDCAM (and many more settings which can't remember now) and in every situation, when camcorder is moving there is a little bit of blurriness.
Field order is upper field first, deinterlace method: blend fields. Resample is disabled.
The worst thing of all is that if I used Sony Playmemories rendered video is perfect (it offers only 3 render settings - I choose 1920x1080 25p 14Mbits). Great picture quality, and no annoying motion blur, but Playmemories seems to be unusable to trim video and add some title etc.
I don't know what to do to solve this problem. Do you have some advice to set up Vegas project or to use some other software.
if your source is interlaced (50i) and you render to 25p, the deinterlacing is done, which definitely decreases sharpness or clarity.
You did not mention which version of Vegas you use, but if it's V14, you may choose adaptive deinterlacing, which is quite superior to the default "blend fields". If your Vegas is 13 or below, use 3rd party deinterlacing. Yadif is one I can recommend, but remember there was something Wayne Waag suggested here on the forum, which is even better than Yadif. (Sorry can't remember the thread nor the name of it - Wayne, are you here? ☺ )
Or your very best option is to shoot in progressive (50p), because your cam supports it. I do the same, my target is also 25p all the time, but shoot 50p.
You can easily render 25p from 50p without involving different deinterlacing algorithms, that will give the best possible result.
Thanks for advice. I tried Vegas 14 adaptive deinterlace but didn't got better result, so I used Yadif and the rendered video is much better.
Now, I set up my camcorder to record in 1920x1080 50p (now if I understand, there will not be problem to render it out to 25p?). Is it normal that camcorder uses more battery to perform that recording (I suppose yes)?
What settings do you recommend to render? I used HQ 1920x1080, 25p XDCAM, but it's 35Mbps looks overkill (18 min of video = 3 GB) for my low budget home camcorder, so I am looking for optimal codec.
I read through quickly so forgive me if I missed something, but are we sure this is a render issue and not a playback issue? Have you tried playing back the rendered files in different players (VLC player, Windows Media Player)?
I recommend shooting in progressive mode -25p. When shooting in progressive mode, there is no interlacing in the picture, so there is no need to deinterlace anything.
Here are some good render settings for upload to youtube, vimeo, etc. and a generally very compatible format:
When choosing your render format, select "Sony AVC/MVC", and choose the preset "Internet 1920x 1080-30p". Then, choose to customize the template and do the following:
-On the "Video" tab: Change the Frame Rate to 25.000 (PAL)
-Leave 'field order' set to "None (progressive scan)"
-Bit rate is fine at 16,000,000, but if you want a little more quality, you can boost this. no need to go above 25,000,000
-On 'Project' tab: set Video rendering quality to "Best".
-everything else stays as it was
-Save this as your default rendering template
This will render as an H.264 with .mp4 wrapper.
I can't say for sure, but I wouldn't expect shooting in progressive to use more battery than interlaced.
Ok, thanks for advice, I will try to render as you suggested. In a few next days, I will test progressive recording.
I forgot say that I watched videos on TV, so deinterlace with Yadif gives best results in my case (with blend fields and interpolation there is little bit of blur when camera is moving - it's not a large amount of blur but it is annoying and hard to watch).
I have question about yadif deinterlacer, should I change any settings? What means processing mode, field order (HD-upper, SD lower - what that mean? - Does that depend on camcorder? - How I could know what field is HD, and what is SD?) What mean parity?
You see that I have a lot of questions, but have few hours of recorded material waiting to be rendered :D
Is there any difference between .mp4 and .MP4?
[Domagoj Hasija] "Is there any difference between .mp4 and .MP4?"
just the letters in the filename. In Windows that does not matter ☺
If you are absolutely sure that you won't need slowmotion at any time, than just shoot 25p.
Otherwise if your camera supports 50p, why not shoot 50p? The extra frames may come handy if you need a slowmo ☺
It's up to you anyway which mode you choose.
With 50p compared to 25p means the double of the data the camera has to precess. I'd expect, yes, somewhat more battery drain - but not significantly- and of course you'll need the double memory card for the same time to record.
For example, I have 2 pieces of 64GB SD cards in my HM600, then with my 50p settings that's enough for 604 minutes.
Going down to 25p, the same cards can hold more than 1200 minutes ☺
The render setting depends on your target. If you intend your video to youtube or similar, 15..20Mbps AVC mp4 is excellent for fullHD (I use similar template), like Tim suggested.
If you intend to bluray, choose an appropriate template.
I used yadif with Vegas10 last time, it was quite good. Then in Vegas 14 I looked at the builtin adaptive deinterlacing, and found it similarly good, but did not use it too much. (I don't have any interlaced material since a while).
If you find yadif better, just keep using it until you finish with your interlaced media. Then later when you already shoot progressive, you just won't need it :)
I may be stating the obvious here, but just want to make sure that you are not shooting at too slow of a shutter speed as well. Typical rule of thumb is that the shutter should be around double the frame rate to avoid motion blur. So if you're shooting at 25p, the shutter should be at 1/50th; if you're shooting at 50i -that is still 25 fps but broken into 2 interlaced fields per frame, so you would set shutter to 1/50th. Shooting at 50p -set the shutter to 1/100th. Of course this is just a very general rule of thumb but for simplicity, you might want to stick to that. As Laszlo mentioned, shooting at a higher framerate allows you the ability to do smooth slow motion as well. However, (as he also mentioned) it will require more hard drive space ...or it will not look as clean (because of trying to squeeze double the picture data into less than double the file size). Also, as I mentioned, the faster the framerate, the faster you should have your shutter, and that means you'll need more light to shoot, so your video won't be as clean in low-light situations. If you know you won't be doing any slow motion, just shoot at 25p.
If you shoot progressive (20p, 30p, 50p, etc.), you don't need to worry about upper/lower field first, because you are recording/playing back both upper and lower fields at the same time -one full frame at a time. Interlaced video (50i, 60i, etc.) means that each frame is broken up into 2 fields: odd scanlines & even scanlines. HD video is upper field first, SD is lower field first (usually). Interlacing was started because of hardware shortcomings in the early days of video. One could argue that it's still useful for capturing smoother video, but in my opinion, it's just an old way of capturing video that doesn't look as good as progressive because you are never really seeing a whole frame at any one moment in time. Also, if you grab a still from interlaced video, it'll have horizontal lines through it, or have less clarity (if you use deinterlacing).
Basically, just shoot progressive and don't worry about deinterlacing, upper/lower fields, etc. Of course, if you already have a lot of footage already shot and it's interlaced, you will need to treat it accordingly.
As for your motion blur issue, if you have a screen grab of the issue, I might (emphasis on "might") have a better idea of what is going on. Interlace issues are pretty obvious.
I believe "Parity" typically refers to error-detection. It's just a way to check to make sure there were no errors in the file you just rendered/recorded/etc. It's used in a lot of different applications, but usually takes a little more time to verify that all the bits are as they should be before you consider your video render (or capture) successful.
Hope this helps. If you want to learn more about video, there are lots of good educational sources out there. One I remember seeing is http://www.videouniversity.com, though there are probably better sources out there. You could also sign up for a course on udemy.com or linda.com. Get a public library card and linda.com is free. ☺