quite a few of questions

[xenonsniper]

first i'd like to say hi to everyone on the forum. for whatever reason, encoding video really appeals to me and i find it quite fun to do. so let me get started with some questions i've had for a while now:

1)what are the differences between these options?

http://i43.tinypic.com/zx5mv8.jpg

as well as automated 3 pass(think i now what constant quality means lol) i tried googling but i couldn't find anything(mainly because i don't really know what to search for)

2)is there any difference between cropping in DGIndex when before making the D2V file with the cropping filter and Cropping in MeGUI? and if so, which one's better?

3)when the term "Noise" is used, what exactly is meant? i've seen the term used in the Avisynth script creator with the "noise filter".

4)what's the difference between sharp and neutral resize filters? and, in your opinions, which ones are good and which ones suck?

5) what's mpeg2 deblocking and what does it do?

6) is AvsP much better to use than meGUI's avs scipt creator?

7) what's adaptive quantizers and what are the advantages to using them?

8) in field order, what does the "source is anime" button do?

9) what are the advantages of using SAR instead of DAR?

uh...that's all i can think of for now lol.

oh yeah and:

10) are there any advantages to using the SNOW codec, and are there any places i could learn about it's options and when to utilize it?

if you can give me the answers in a response that would be awesome, however linking to threads or FAQs where these questions have been asked before would be awesome too, tbh i really just don't know where to look.

thanks again

-xenonsniper.

[nurbs]

First off, sorry for my bad english.

1) When you encode video in multiple passes the first pass is used to analyse the source so it can be encoded efficiently in the second pass. Often faster settings in the first pass are used (e.g. the "turbo" option in megui). If you don't know what you are doing use the automated 2pass (with turbo) since it will prevent you from screwing things up with different settings between the passes. If you don't care about the final filesize use constant quality. ABR only does one pass, but gives worse quality compared to multipass encodes and you might miss the target bitrate. 3pass can help with really small bitrates or if you miss the target bitrate. I can't properly explain constant quantizer, but you wouldn't want to use it if there is a constant quality option (it's similar).

2) Probably doesn't matter, but to be on the save side crop in MeGUI. I don't know but maybe the avs-creator makes wrong decisions when doing anamorphic encodes or resizeing.

3) Noise is for instance what you get when you have bad analog reception on your TV, or pixlels that are radically different from their neighbours for no good reason (not actual detail). There are some interesting avisynth plugins for denoising, motion compensated and everything to differentiate noise from actual information. Denoising will probably also destroy some of the detail (if it's too strong) and can introduce banding. If your source isn't really noisy you probably don't need to denoise it, unless you want to use a very small bitrate for the encode.

4) IIRC all resizers blur, but some blur more than others. I guess it comes down to personal preferance. I use Spline36Resize for my HD encodes.

5) No idea. Guess you should use that if your source has visible blocks, but there are probably avisynth plugins which can do a better job at deblocking.

6) I think AvsP is great when you write your own scripts. If megui does everything you need then just use that.

7) It gives flat (uniform) blocks in the picture more bitrate. Whithout it for instance when you have a blue sky sometimes you can see blocking there without the picture. Leave it on.

8) You should use that when you use the autodetect for interlacing and feed it an anime source.

9) Just two different ways of specifying the same thing. Some encoders accept only one and some only the other.

10) It isn't developed any more, has no hardware support and some quality issues IIRC. Never used it, but there is a thread with comparisons to other codec somewhere in the alternative video codecs part of the forum. The thread is about 2 years old IIRC.

[xenonsniper]

ah thanks for all of your helpful information

sucks about the snow codec =\

[J_Darnley]

ABR -- Average bitrate: one pass bitrate-based ratecontrol, generally considered to produce the worst quality.
Const. Quantiser -- Does what it says, encodes the video at a constant quantiser but still obeying ip- and pb-ratio, disables AQ. For testing, don't use.
Automated 2pass -- MeGUI sets up both passes for you. Uses bitrate-based ratecontrol but with the information gathered in the first pass, it makes much better descisions in the second pass. Use when you need to use vbv(bitrate restrictions for hardware decoding) or if you need to fill a certain storage format.
Automated 3pass -- Uses a third pass which is usually useless, can help on short clips < ~2 minutes (maybe).
The other 1st, 2nd & 3rd passes -- These set up the individual passes manually.
Const. Quality (which you missed) -- Gives a certain quality but unknown filesize. Good when you aren't using vbv or when you don't want to fill a certain storage format.

5 -- In MeGUI, the "MPEG2 deblocking" specifically means turning on DGDecode's "cpu" option which, as the name suggests, is a deblocker. It also does deringing for some vales set. RTM

6 -- Identical scripts are identical, it doesn't matter what wrote it. Avsp won't do anything for you unlike MeGUI. It is more like a text editor specifically for writing Avisynth scripts so it has a few features to help with that.

[xenonsniper]

Const. Quality (which you missed) -- Gives a certain quality but unknown filesize. Good when you aren't using vbv or when you don't want to fill a certain storage format.

what's vbv?

[nurbs]

vbv is video buffer verifier. If your encode for harware playback your encodes will have to be within certain constraints like (vbv)maximum bitrate, buffer size, macroblocks per second (=framesize & frames per second), number of reference frames, ect. If you use Constant quality the encoder has a harder time keeping vbv constraints which may lead to quality issues. If vbv is violated it will show up in the log (at least if you use x264). If it is a problem depends depends on what you encode.
Example:
Say you encode blu-rays at 1280x720 at crf (constant quality) 22 the majority will end up with an average bitrate of 4 to 6 mbps. You want to be within level 4.0 which means you have a vbv max bitrate of 25 mbps along with other constraints. In that case you are unlikely to have a problem since the video bitrate will hardly ever approach that value and if it does you still have the buffer which gives you headroom (usually same as buffer size) for 1 second IIRC. It will work in most (almost all) cases, but there may be some videos where you run into problems. These can go from inperceptible quality degradation to clearly visible artifacts. In the worst case you violate vbv and then your hardware may not play the file, but as I mentioned this would show up in the log.

If you want to know the limitations of the different levels they are in the wikipedia article about "AVC".

[xenonsniper]

ah i see, i'll check out AVC

thanks.