I have made some captures for test from VHS in 352x576 and 720x576 (PAL).
I have tried the rmpal filter in vdub described in some messages in the forum.
The strange thing is that the files are much bigger after filtering?
+ Reply to Thread
Results 1 to 15 of 15
-
-
So you're having problems with interchanging "red" and "green" lines? You could try adjusting the tint setting for you capture card, if it has one. I sometimes have to go as high as 145 (where 128 is default) to get rid of these artifacts. I've made filter similar to rmpal because when capturing PAL-60 stuff i can't get rid of the color errors no matter how much i adjust the tint. I don't really know which of the filters do a better job, but i use mine when i need it because i know exactly how it works and what settings to use. It only does a proper job if you've captured either exactly one field or both (ie. either 288 or 576 lines for PAL, 240/480 for NTSC/PAL-60), though. Dunno if rmpal has the same limitation.
-
I'm pretty sure I know exaclty why your fiels are so much bigger when you save them in VirtualDub.
Check the compression setting in VDub. You will probably find that youa re saving files as UNCOMRPESSED RGB. This creates huuuuuuuuuuuuuge
video files.
The problem here is that VDub uses the older VFX drivers, and chances are your DV codec is a WDM driver (mine is) so you can't use your DV codec to save from VirtualDub (I can't).
Not to worry -- download the free uffyuv codec. Then choose huffyuv compression to save.
This is still 2.5 times larger than DV format AVIs (huffyuv = 30 gigs per hour as opposed to Uncompressed RGB's 60 gigs her hour (!) and DV Type 1 or DV Type 2's 13 gigs per hour) but hey..it'll help. -
If you are simply using VirtualDub to apply a/a few filters, there is not necessarily a need to create a second AVI file.
If you plan on using something like TMPGEnc to compress to MPEG, then you can use VirtualDub to frameserve directly to TMPGEnc (with the filters). This saves you having to create another intermediary AVI file.
In any case, if you are working with the HuffyUV AVI codec, it is/mostly (depending on your settings) lossless so you won't get any generational loss of quality.
Regards.Michael Tam
w: Morsels of Evidence -
vitualis,
If you are simply using VirtualDub to apply a/a few filters, there is not necessarily a need to create a second AVI file.
Do you mean: add the filter when you are capturing or after? -
Generally you can't apply filters during capture as it will be too CPU intensive.
What I means is the following:
1. Capture to AVI
2. Load AVI with VirtualDub
3. Load and set the filters that you want.
4. Set VirtualDub to frameserve the output
5. Load TMPGEnc
6. Have TMPGEnc load the "marker" or "dummy" file
7. TMPGEnc "thinks" that there is a full AVI file.
8. Encode to whatever TMPGEnc settings you want and as TMPGEnc tries to "read" this dummy AVI file, VirtualDub serves the information (hence, "frameserving")
Have a look at the various guides here on frameserving with VirtualDub to TMPGEnc and you will get a better understanding on how to do it.
Regards.Michael Tam
w: Morsels of Evidence -
vitualis:
Thanks for your explanation
But was is the reason for the bigger files what spectroelectro said? -
Possibly/probably.
If you find that VirtualDub makes much larger AVI files than the source, you've probably forgot to set the AVI compression setting (and it defaults to uncompressed RBG).
Regards.Michael Tam
w: Morsels of Evidence -
Yes.
If you don't specifically set the compression codec, it defaults to uncompressed RGB (usually VirtualDub warns you before it does make an uncompressed RGB AVI file though...)
Regards.Michael Tam
w: Morsels of Evidence -
@vitualis
I have tried it again as you said but the files are still 1,6x bigger.
Do you have a explanation for this? -
What sort of filtering did you do?
Did you increase the framesize?
Regards.Michael Tam
w: Morsels of Evidence -
One reason why the output file is larger than the input file could be that you're compressing RGB data instead of YUV, which will yield lower compression ratios. You could try changing the RGB compression setting for HuffYUV to "convert to YUV" if you haven't done so already.
However, compressing a file over and over is not a good idea, even with a lossless codec such as HuffYUV. Assuming that your capture card converts the analog data to YUV, your video will be in this format. When applying filters in Virtualdub the YUV data is converted to RGB, and then if you're converting back to YUV that's another conversion for you. All these conversions are most likely MMX-optimized and therefore inprecise (not much, but you can get some information loss). And even though they're optimized they still take a lot of time.
One of the nice things about AVISynth is that a lot (but not all) of the filters available for it can be applied on - and are specifically made for - YUV data, so there are no colorspace transformations necessary.
Similar Threads
-
How do I make a filter for VDub
By blaze01 in forum Newbie / General discussionsReplies: 4Last Post: 4th Mar 2008, 16:16 -
Applying a filter without recompressing in VDub
By chrisr34000 in forum EditingReplies: 5Last Post: 25th Jan 2008, 17:42 -
paging SatStorm, rmPAL filter crashing VirtualDub
By friendly_jacek in forum RestorationReplies: 1Last Post: 9th Jan 2008, 22:18 -
ACDsee Filter&VDUB
By mv0930 in forum Newbie / General discussionsReplies: 0Last Post: 4th Sep 2007, 12:27 -
Vdub Filter Chain
By christopheramos in forum Video ConversionReplies: 16Last Post: 3rd Sep 2007, 07:57