Cure for lag? ^_^

The old Video Software Help forum, left visible as an archive.

Postby x_rex30 » Wed Jun 21, 2006 5:58 pm

AtomX wrote:
x_rex30 wrote:I recommend adding the ones needed to make your footage look nicer after you have finished your video, witch is what I did with my last two videos.


This completely defeats the purpose of pre-processing. If you add those filters to your final video, things like added effects, blurs, some transitions, etc. are all going to be filtered too
Oh I missed this here. I think you misunderstand, This isn't after I "Render as"(Vegas) or "Export"(Premiere). I do the adjustments before I render it so it does not filter any effects, filters you used in the editing program.
User avatar
x_rex30
 
Joined: 10 Apr 2001

Postby Aki Ameko » Wed Jun 21, 2006 7:55 pm

AtomX: Yes, I did use the method you perscribed, and it works really well with the exception of the lag. :D

Here's a copy of my AviSynth scripts before I made them into clips:

mpeg2source("C:\blah.d2v")
KernelDeint(order=1, threshold=10)
ADSwapFieldOrder
Crop(8,0,-8,-0)
LanczosResize(640,480)
ConvertToRGB32()
AssumeFPS(29.97)

(The other script is the same thing excluding the field order swap)
Working on my first AMV! Any help is greatly appreciated.
Previously known as Kaisune
User avatar
Aki Ameko
 
Joined: 19 Dec 2004

Postby Aki Ameko » Thu Jun 22, 2006 1:19 pm

I was wondering, could multiple desktops and lots of image and music files cause lag? 'Cause my comp has a lot of that crap on it. *deletes some un-needed pics*
Working on my first AMV! Any help is greatly appreciated.
Previously known as Kaisune
User avatar
Aki Ameko
 
Joined: 19 Dec 2004

Postby Zero1 » Thu Jun 22, 2006 2:03 pm

Kerneldeint and lanczos resize will eat a lot of cpu time. If lag is really a problem it would be best to export to some form of lossless coding. You could use HuffYUV, Lagarith, or if your harddrive is up to it (both speed and size wise) raw YV12 (which for 640x480 is 450KB per frame).

Also you will get better image quality using bilinear resize, it will appear softer than lanczos, but you won't get ringing (haloing, or those light edges). The rule of thumb is bilinear/precise bilinear for downsizing and bicubic/lanczos for upsizing (you shouldn't be upsizing anyway unless necessary, it doesn't improve the quality and takes up more space).

And yeah, even HuffYUV can lag if you have a slow ass hard drive, there is more to lag than just CPU speed. RAM and HD are huge factors in video editing.
User avatar
Zero1
 
Joined: 02 Jan 2004
Location: Sheffield, United Kingdom

Postby x_rex30 » Thu Jun 22, 2006 4:14 pm

Zero1 wrote:Also you will get better image quality using bilinear resize, it will appear softer than lanczos, but you won't get ringing (haloing, or those light edges). The rule of thumb is bilinear/precise bilinear for downsizing and bicubic/lanczos for upsizing (you shouldn't be upsizing anyway unless necessary, it doesn't improve the quality and takes up more space).
Damn, I've always been using lanczos.. :( do you know a page that explains about that.. I hope haven't been butchering the image quality.

AVISynth Manual wrote:Lanczos
LanczosResize is an alternative to BicubicResize with high values of c about 0.6 ... 0.75 which produces quite strong sharpening.
It usually offers better quality (fewer artifacts) and a sharp image.
User avatar
x_rex30
 
Joined: 10 Apr 2001

Postby x_rex30 » Thu Jun 22, 2006 4:17 pm

nvermind :oops:

AVISynth Manual wrote:Bicubic
BicubicResize is similar to BilinearResize, except that instead of a linear filtering function it uses the Mitchell-Netravali two-part cubic.
The parameters b and c can be used to adjust the properties of the cubic, they are sometimes referred to as `blurring' and `ringing,' respectively.

With b = 0 and c = 0.75 the filter is exactly the same as VirtualDub's "precise bicubic," and the results are identical except for the VirtualDub scaling problem mentioned above. The default is b = 1/3 and c = 1/3, which were the values recommended by Mitchell and Netravali as yielding the most visually pleasing results in subjective tests of human beings. Larger values of b and c can produce interesting op-art effects--for example, try b = 0 and c = -5.

If you are magnifying your video, you will get much better-looking results with BicubicResize than with BilinearResize. However, if you are shrinking it, you are probably just as well off, or even better off, with BilinearResize. Although VirtualDub's bicubic filter does produce better-looking images than its bilinear filter, this is mainly because the bicubic filter sharpens the image, not because it samples it better. Sharp images are nice to look at--until you try to compress them, at which point they turn nasty on you very quickly. The BicubicResize default doesn't sharpen nearly as much as VirtualDub's bicubic, but it still sharpens more than the bilinear. If you plan to encode your video at a low bitrate, I wouldn't be at all surprised if BilinearResize yields better quality.
User avatar
x_rex30
 
Joined: 10 Apr 2001

Postby Aki Ameko » Sun Jun 25, 2006 11:13 pm

Zero1 wrote:Kerneldeint and lanczos resize will eat a lot of cpu time. If lag is really a problem it would be best to export to some form of lossless coding. You could use HuffYUV, Lagarith, or if your harddrive is up to it (both speed and size wise) raw YV12 (which for 640x480 is 450KB per frame).

Also you will get better image quality using bilinear resize, it will appear softer than lanczos, but you won't get ringing (haloing, or those light edges). The rule of thumb is bilinear/precise bilinear for downsizing and bicubic/lanczos for upsizing (you shouldn't be upsizing anyway unless necessary, it doesn't improve the quality and takes up more space).

And yeah, even HuffYUV can lag if you have a slow ass hard drive, there is more to lag than just CPU speed. RAM and HD are huge factors in video editing.
So, just use bilinear instead of biubic?

Also, if I change my avisynth code, will it affect all the clips that I've made, including the ones already imported to WMM?
Working on my first AMV! Any help is greatly appreciated.
Previously known as Kaisune
User avatar
Aki Ameko
 
Joined: 19 Dec 2004

Postby Aki Ameko » Sun Jul 09, 2006 12:33 am

Anyone?
Working on my first AMV! Any help is greatly appreciated.
Previously known as Kaisune
User avatar
Aki Ameko
 
Joined: 19 Dec 2004

Postby madbunny » Sun Jul 09, 2006 12:51 am

You have a few options.
Your ram should be fine for what you're doing. Keep your system running as lean as you can so that you're not wasting CPU power on stuff you don't need to have running, like 'quicktime news update'.
Do the obvious by defragging your sytem and cleaning it up.

To speed up your clips:

1) use the bait and switch method. Blazingly fast, and smooth to work with, but a pain in the ass to keep track of.

1.2) use another codec. Something not lossless that runs more smoothly but still looks good.
this one was done in Mjpeg as an example

2) Get a really fast hard drive - and keep it dedicated for editing. If you have an older system you might want to consider getting an SATA controller card instead of using IDE.

3) Are you editing in WMM, or premiere? If it's premiere, try rendering your clips out, and check what codec your preview renders are in.

4) annoying but theoretical speed boosts - Check your motherboards frontside buss and compare it to your ram. If you have an 800mhz fsb and 300mhz ram you can get better stuff.
4.1) overclocking - At the least, you can look into high quality cooling. If your computer is running hot, it'll er... 'de-cycle' so to speak and slow down.
User avatar
madbunny
 
Joined: 17 Jun 2003

Postby Melanchthon » Sun Jul 09, 2006 10:11 am

Aki Ameko wrote:Here's a copy of my AviSynth scripts before I made them into clips:

[snip]

If it's rendered already, then it doesn't matter what the script is. I still don't know what codec you used (you mentioned Huffyuv, but I'm not sure if you are using it). The resolution might be causing some of the lag-- maybe you try could resizing to 512x384?

Changing your script won't affect files that are already rendered.
User avatar
Melanchthon
 
Joined: 02 Sep 2004

Postby Gepetto » Mon Jul 10, 2006 7:29 am

Back to the topic...

why did everyone assume Aki Ameko to be using HuffYUV?

Aki Ameko: If you don't know what codec you used in vdub, it's probably uncompressed video, which is the default setting. That would explain your lagging, because uncompressed video is BIG. Create a HuffYUV or Lagarith file on vdub. Just do what you already did, but before saving the file, click "Compression" on the "video" tab and select a lossless codec (of course, you must have them installed). Then, save normally. The file will be smaller than uncompressed without any quality loss (hence the name "lossless") and the lag should go away, or at least become considerably smaller.
And God spoke unto the Chicken, and He said: "Thou shalt crosseth the road", and the Chicken did cross the road, and there was much rejoicing.
My DeviantART profile
User avatar
Gepetto
Mr. Poopy Pants
 
Joined: 10 Jun 2004
Location: The Tokyo Settlement
Status: Bored to tears

Previous

Return to Video Software Help Archive

Who is online

Users browsing this forum: No registered users and 1 guest