digitalFAQ.com Forums [Archives]

digitalFAQ.com Forums [Archives] (http://www.digitalfaq.com/archives/)
-   Computers (http://www.digitalfaq.com/archives/computers/)
-   -   Video Screen Capture (http://www.digitalfaq.com/archives/computers/14346-video-screen-capture.html)

Boulder 12-21-2006 04:35 PM

You need MaskTools v1.5.8.

nicksteel 12-22-2006 06:15 AM

Quote:

Originally Posted by Boulder
You need MaskTools v1.5.8.

Thanks, had wrong version. Still testing.

I would still like to get more detail in the encode. Being a very high quality screensaver, Marine Aquarium really has a lot of detail. I realize that the encoded 720x480 will lose a lot from the 1280x1024 capture, but want to get it as sharp and detailed as possible. Using seesaw seems to help a little, but the final still looks more blurred than I would like. Spline36Resize() doesn't seem to have any effect.

Thanks again. :D

Boulder 12-22-2006 06:32 AM

Increase the amount of sharpening in SeeSaw, there's not much else you can do.

nicksteel 12-29-2006 07:39 PM

Quote:

Originally Posted by Boulder
And another point : if you encode to MPEG2, ditch TMPGEnc and use HC instead. You can avoid that one (actually two) extra colorspace conversion and HC provides better quality anyway.

Am working with HC now. How is best way to handle avi audio with my concatenated a++b++c,etc files?

Boulder 12-30-2006 03:24 AM

Open the complete script in VDub and save the audio as WAV.

nicksteel 12-30-2006 03:57 AM

Thanks. Haven't worked with avi's for a long time. I have found that Tmpgenc nor HC like too many concatenated files. Either can handle a dozen or so, but bomb out with too many. I wish Fraps would allow larger than 4GB files! But the screensavers to DVD have turned out very good - even as anamophic widescreen on my new Christmas 42" plasma! :D

Boulder 12-30-2006 04:19 AM

It could be that it's rather an Avisynth issue because the encoder should see the avs script as one video file even if multiple video files are loaded in it. You could try using tritical's latest v2.5.7 build, download here : http://www.missouri.edu/~kes25c/Avisynth_251106.exe . They are very stable and need less memory.

Dialhot 12-30-2006 04:52 AM

Yes, surely a memory problem. I already did an encode with 99 pieces of videos merged with "+" in a single script. But it claimed 2GB in memory to work.

(note : all pieces were loaded without audio).

nicksteel 12-31-2006 12:15 PM

Quote:

Originally Posted by Boulder
I don't like to use a lot of time predicting a correct quant so I'm using 2-pass as the rate control is very good. AutoGOP is useful especially with high-motion video. Most of the time the GOP is the maximum length even with autoGOP.

Three questions:

:?: HC with AutoGOP shows GOP 15, even with "Make DVD Compliant". Shouldn't this be 18 for NTSC?

:?: What criteria should I use in choosing between HC 1-pass and 2-pass?

:?: What is difference between "++" and "+"?

Dialhot 12-31-2006 12:29 PM

Quote:

Originally Posted by nicksteel
:?: HC with AutoGOP shows GOP 15, even with "Make DVD Compliant". Shouldn't this be 18 for NTSC?

18 is the max allowed, but 15 is as compliant as 18 :).
Hank probably not adapt the AutoGOP computing to the nature (PAL/NTSC) of the encoding.

Quote:

:?: What criteria should I use in choosing between HC 1-pass and 2-pass?
The time you want to spend on the prediction, the subjective importance for you of having a disc 100% full and more other the acceptance you have about the quality obtained. I did not compare 2-pass and 1-pass in HC so I trust Boulder, but generally encoders have difficulty to obtain the same quality in both mode, the Quality based being the better.

Quote:

:?: What is difference between "++" and "+"?
RTFM :)
(if your encode don't have audio, the two are the same).

Boulder 12-31-2006 12:30 PM

I'm not sure how the gop length goes with NTSC stuff. If you do a pulldown after (or during) the encode, you should probably use 12 frames. If not, 18 is the way to go. 15 is the maximum for PAL. EDIT: if you don't use pulldown, enable autogop, otherwise it might be safest to use 12-2 gops although autogop should work as well. Some people have complained that it doesn't always work but I don't know about that as I don't deal with NTSC encodes.

What do you mean by criteria? Whenever I don't need to worry about the filesize, I use the constant quant mode at q2, vbv check enabled. Whenever I need a specific filesize, I do a 2-pass encode.

See the difference between + and ++ in the Avisynth docs, AlignedSplice and UnAlignedSplice are the same things.

nicksteel 12-31-2006 01:09 PM

Thanks, guys. You always have the answers.

nicksteel 01-11-2007 05:38 AM

After updating to Fraps 2.8.2, having more memory problems than ever. There's problems loading the Fraps codec into memory during processing with HCenc, Tmpgenc or VirtualDub. I'm in contact with Fraps trying to get this corrected.

My only "work arround" seems to be to load the avi into Vdub and save as avi with another (non-Fraps) codec, then using this file to encode.

:?: Is (uncompressed RGB) the best Vdub compression choice for this?

I don't care about the file size, as I'm encoding to dvd, just want to change from Fraps codec and keep maximum quality.

Dialhot 01-11-2007 05:48 AM

Quote:

Originally Posted by nicksteel
I don't care about the file size, as I'm encoding to dvd, just want to change from Fraps codec and keep maximum quality.

*If you don't care about size, then uncomrpessed is always the best choice. Now I wonder about the RGB. May be a YUY2 codec would be better. I would take huffman because I used to it, it is a lossless codec that supports both RGB and YUY2.

nicksteel 01-11-2007 05:57 AM

Huffyuv v2.1.1 :?:

Also, how is best way to trim this thing for anamorphic? (16x9)

[img=http://img151.imageshack.us/img151/508/fitcdus1.th.jpg]

Dialhot 01-11-2007 07:51 AM

Quote:

Originally Posted by nicksteel
Huffyuv v2.1.1 :?:

Yes :o
Quote:

Also, how is best way to trim this thing for anamorphic? (16x9)
Turning a 4:3 into 16:9 ? I always considered that there is no "best way". For me there is no way at all.

nicksteel 01-11-2007 09:43 AM

The aspect ratio of the screensaver is 16:9, but Fraps only allows full screen capture. My LCD monitor is 4:3, so I capture top and bottom borders. I was trying to figure out how to trim the top and bottom of the capture and fix to automatically start as 16:9 widescreen on my widescreen HDTV without encoding large borders and "zooming" on the TV.

In Tmpgenc clip, I show 92 pixels top and bottom.

Dialhot 01-11-2007 10:14 AM

[quote="nicksteel"]The aspect ratio of the screensaver is 16:9, but Fraps only allows full screen capture. My LCD monitor is 4:3, so I capture top and bottom borders. I was trying to figure out how to trim the top and bottom of the capture and fix to automatically start as 16:9 widescreen on my widescreen HDTV without encoding large borders and "zooming" on the TV.

Then you should use Paranoia insteed of FitCD. It has a "autocrop" feature to detect the black borders and trim them.

Quote:

In Tmpgenc clip, I show 92 pixels top and bottom.
... but as you know the border size, just put these values in FitCD (the 2 fields) above and under the rectangle representing your source on the left
+ check "anamorphic" for the destination on the right.


All times are GMT -5. The time now is 09:59 AM  —  vBulletin © Jelsoft Enterprises Ltd

Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.