I’m new here, and hoping you’ll be easy on me (the bones are brittle, and the brain is getting there)
I have some commercial movies on vhs, fair quality, but lots of flaws.
There are small white spots (dropouts?), black spots (grain? noise?) and just plain dirt (dust & hair curlicues, even what looked like brush bristles! )
Most only last a couple of frames, and the flash of their appearance draws the eye and is very distracting.
At least 50% of the movie is flashing junk, like a kaleidoscope show.
I wonder if they hauled the footage out of the trash can.
I’ve captured to avi, and want to filter to get rid of as much junk as possible and keep the avi format so that I can edit using Pinnacle Studio 9.
I tried “FulciLives” guide for Convolution3d, (wonderfully simple to follow, Thank You FulciLives).
But the video seemed untouched, and audio turned out bad with lots of dropouts, I don’t know what I did wrong.
Right now I am overwhelmed by the dozens of guides and hundreds of posts, it all seems impossibly complicated. I just can’t get my mind around how to use AviSynth or VirtualDubMod, and they seem to be required.
Please, can someone get me pointed in the right direction to get me to a good starting point?
I’d be most grateful for any and all help.
+ Reply to Thread
Results 1 to 13 of 13
-
-
VirtualDub Mod is still my favorite for that type of filtering. Sorry, VD and it's variants are kind of the 'Swiss army knife' of AVI editors. It is really worth learning.
There are a lot of guides for using it available in 'Tools' <<<<<<
You probably already know this link, but http://neuron2.net/ has a lot of filters there for old movie problems. You have to dig around a little, though, to find just what you need. And experiment a little.
What kind of AVI is the file in? DV AVI is one of the best for editing. Video filtering shouldn't affect the audio. I usually do my video filtering (And editing, if needed), then save it out as DV again. I save the audio out as a WAV and put that into the freeware Audacity to filter and clean up the audio. Then you can either put the audio back in or if your encoder or editor can accept the WAV, pass it on.
You might wait a little on AVISynth. It is very powerful and fast, but a little hard to learn for beginners.
There are a lot of tricks with VD. You can save your settings and come back at a later date and apply them again. This makes it easier for experimenting when you find just the right settings. It sounds like you could use a 'spot' filter and maybe a 'antiflicker' filter. -
This is the mere basics as far as what these instructions do, but hopefully they will get your feet wet just enough for you to get by until you're ready to enhance your knowledge on the subject.
It assumes you will have interlaced DV-AVI as your source, but shouldn't be too hard to adapt to any source.
************************************************** **********
1. (If required) Install Panasonic DV codec from here:
http://users.tpg.com.au/mtam/install_panvfwdv.htm
Follow the instructions on that page to the letter.
2. (If required) Install convolution3d from here:
If using YUY2 sources:
http://www.avisynth.org/warpenterprises/files/convolution3d_25_dll_20030422.zip
If using YV12 sources:
http://www.avisynth.org/warpenterprises/files/convolution3dyv12_25_dll_20030329.zip
By "install" all I mean is extract the .DLL file from the Zip file and into your windows\system32 directory.
3. Open virtualdubmod. Go to Tools > Script Editor, and copy the text below into the script editor.
Code:LoadPlugin("Convolution3d.dll") avisource("c:\capture.avi") SeparateFields() odd=SelectOdd.Convolution3D (0, 6, 10, 6, 8, 2.8, 0) evn=SelectEven.Convolution3D (0, 6, 10, 6, 8, 2.8, 0) Interleave(evn,odd) Weave()
4. Go to Audio > Direct Stream Copy. This will copy the audio as 1536kbps PCM WAV, which is what AVISynth will decompress it to. Go to Video > Full Processing Mode. Go to Video > Compression, and choose the Panasonic DV Codec. Go To File > Save as AVI, and save out to a new AVI. The new AVI will be DV-AVI at around 13.2GB per hour, so make sure you have plenty of hard drive space.
************************************************** **********
That should get you started. All this does is applies the convolution3D noise removal filter to your AVI and outputs back as DV-AVI for you to use in Studio. I'll skimp on the areas to go wrong as I don't know a lot about your source. I''m sure there'll be a couple of script tweaks required but I'll wait until you get errors before suggesting how to get around them.If in doubt, Google it. -
Hi redwudz, thank you for responding.
I captured using "DV Encoder", it's the default DV codec when you capture in Pinnacle Studio.
So my cap is DV Avi, about 23 GB for a 2-hour movie.
Somehow I got it into my head that VirtualDubMod only created mpeg files, was that incorrect?
If so, I guess what I need is a bone-simple procedure for using VDM to create dv / avi files.
There are so many guides listed here, and I sure could use a recommendation for a basic how-to.
Thank you for the link to neuron.net, I didn't know that site before, and will check it out. -
Originally Posted by grannyGeek
That makes things a boatload easier.
Originally Posted by grannyGeek
Originally Posted by grannyGeekIf in doubt, Google it. -
jimmalenko, thanks so much for the instructions!
These tell me that I did not choose the correct format or codec when I tried this this the first time.
I already have a Panasonic DV codec installed, I think I'll download and install the one you list just-in-case it's a different version.
I will run through this procedure tonight, and see how things turn out.
Thank you again. -
jimmalenko's instructions will work fine. I would definitely DL the Panasonic DV codec or similar and install that if you don't have it. VDM will only basically save in AVI format, but you can use DV format for the output and have the essentially same file type out of it. It should show up in 'Compression' for you to select.
But you would have another 23G file.
If you wanted to edit in VDM, then you could frameserve the output to a MPEG-2 encoder like TMPGEnc and not create that second 23G file.
But for the moment, let's try something simpler. Put the DV file into VDM, then chop out a representative part, about 15 minutes long. Save it with a new file name so you don't damage your original file.
Load that into VDM to experiment with filters. This will be much quicker to work with than the whole file. When you find the right settings, save them in 'Save Processing settings'. After you set up your filters, use 'Preview filtered' to see what it looks like in motion. Then if you have the filtering set exactly the way you want, you can put in the whole video and apply the filters with the saved processing setting. This may take a while. AVISynth makes the filtering much faster, but you have to use the scripts like the one shown above.
This is the way I do it with large videos, it saves a lot of time and drive space.Getting the filters right is the hardest part.
-
redwudz, great suggestions, thank you.
I am archiving this thread to my usb drive.
You and jimmalenko have definitely gone to the top of my "hero list".
I'll post results tomorrow. -
Well, now, things are getting really interesting.
I have 3 new questions.
Q - 1. Can you recommend a different codec to process my capture as DV Avi?
The Panasonic codec is causing a conflict with Convolution3D.
If this is a known issue, just skip the next 3 paragraphs.
==========
I had to download the panasonic codec from the mirror site listed on the tools page
(no internet at home, work computer web filter blocks me from accessing users.tpg.com as a “personal” site. Shhhh, don’t tell on me….)
I installed from the INF file.
When I loaded the AVS script, I got error message Line 4 Convolution only supports YUY2.
I installed the other Convolution3d version, created a new script for that version, and got same type of error message “Convolution only supports YV12
Simply removing the panasonic codec in Control Panel did NOT fix the problem, I had to do a system restore to yesterday’s settings.
I ran thru this cycle 3 times to confirm the problem --- install panasonic codec, run script, get error message “does not support”, un-install codec, run script, get error message, system restore, run script, NO error message.
=============
So, I installed MainConcept codec (demo version that watermarks output).
Ran the script with Convolution3D, and processed with MainConcept.
But I can’t have that watermark, so please, can you recommend a different codec?
These are what VDM lists as available for use, and I must add that my Pinnacle Studio causes a known conflict with PicVideo MJPEG codec that causes DEP error in Windows Explorer, so I can't use PicVideo.
Uncompressed RGB
Cinepak
Indeo
Intel Indeo R3.2
Itel Inde 4.5
Intel IYUV
LEAD MCMP/MJPEG
MainConcept DV
Microsoft H.261
Microsoft H.263
Microsoft RLE
Microsoft Video 1
PCLEPIMI 32-bit compressor
PicVideo MJPEG
Q – 2. I would like to find an alternate download source to get RemoveDirt.dll.
Work computer server is blocking me from http://www.removedirt.de.tf/.
I don’t think Convolution is designed for what I want to clean up, and I thought RemoveDirt might work for me. But I can’t get hold of it.
Q – 3. Jimmalenko, in VDM, I do not see any “AUDIO” menu choice.
4. Go to Audio > Direct Stream Copy. This will copy the audio as 1536kbps PCM WAV,
I'm looking forward to hearing from you, my apologies for being a pain. -
thanks, jimmalenko.
Tonight I will install Panasonic codec again, and add the code to convert to YUV2.
I'll see how that goes, and post back.
Strangely enough, I didn't make any selections for the audio stream, but the final output was fine even so.
I do appreciate your patience -
Hooray! I successfully ran my sample with Convolution3D.
It took care of the graininess very nicely,.
< ok, this is old news to you guys, but I’m as buzzed as a kid with a new tricycle! >
The noise / grain in this clip apparently tends into the red tones, and when finished, the overall tint of the sky changed away from the original purple effect, and now shows as blue / gray tones.
The down side is skin tones now look a wee bit greenish.
But I will do a test burn to dvd-rw to see how they look on tv.
But, as I feared, Convolution couldn’t handle the big industrial-strength chunks.
Could this be adjusted by changing the script parameters?
Or, are there VirtualDub filters out there that might work for this?
I didn’t yet find any VirtualDub filters designed for this, so I downloaded DeSpot and RemoveDirt from the AviSynth site.
Can you point me toward some sample scripts for either or both?
< I need “cookbook recipes”, I don’t have a clue how any of their parameters work >
I’m trying to attach some sections of screenshots so you can see what I’m working with.
Screenshots were taken at native resolution 720 x 480 using ImageGrab (wonderful app!)
I really appreciate all your help, I hope I’m not wearing out my welcome.
clip%20compare%202.jpg
clip%20compare%201.jpg
Similar Threads
-
How to embed subtitles into mkv STEP BY STEP using MKVMERGE and MKVEXTRACT
By lelouch283 in forum SubtitleReplies: 30Last Post: 23rd Jun 2017, 14:04 -
MP4 X.264 to AVI XviD - Hard Sub with ssa subtitles (1 step or 2 step?)
By ssray23 in forum SubtitleReplies: 4Last Post: 29th Aug 2010, 19:35 -
Need DVD Rip Step-By-Step Instructions For Best Quality
By bht42 in forum DVD RippingReplies: 4Last Post: 26th Jul 2010, 09:28 -
The Complete Step By Step DVD Ripping/Burning Primer for Mac...?
By oliasdoug in forum MacReplies: 0Last Post: 10th Nov 2009, 07:04 -
Newbie, advise for easy 1 step backup of dvds
By mbogda in forum Newbie / General discussionsReplies: 5Last Post: 4th Sep 2008, 13:18