Being a noob whos having some difficulty knowing what settings are best, could the community take a look at the settings Im using? I seem to be capturing video well with my workflow (ccd-trv85 to iodata gv-usb2 to amarec TV) but just wanted to make sure these are optimal for what Im capturing (almost all video8 but a few hi8 casettes as well).
Should I really use this resolution? 720 x 480? (even though video8 only has 240 lines and hi8 has 400?)
Should I edit any of my "proc amp" settings in the device settings window?
Does anything look wrong?
The sidebar in the first two pictures is my GV-usb2 device settings. Im exporting to Lagarinth codec.
+ Reply to Thread
Results 1 to 30 of 54
Last edited by videon00b; 15th Jul 2017 at 20:12.
Less importantly, heres my settings for when rendering video in Vegas, with the goal of keeping the video lossless. Im converting mostly video8 and a few hi8 tapes.
Next to the first image is the video format (Lagarinth) configure window.
Does this look good? Anything I should change? I know to disable re-sampling, and use YUY2 now. But thats about it lol. Im using 720x480 because thats how many pixels I'm importing it as into Amarectv. Thanks
In your Sony Vegas settings it looks like you have set Vegas to de-interlace as the field settings are set to `progressive/none`
(assuming the initial Vegas project settings and field setting is set to blend or none progressive etc)
Do want Vegas to de-interlace your lossless export?
I find that under the Advanced/Record Setting tab, turning on "Match the start timing of the audio" can improve audio sync, depending on your hardware.
Why are you using Lagarith? It hasn't been updated in nearly a decade. UtVideo is still supported, works with AmarecTV, and may play nicer with Vegas. I don't use Vegas, but in Premiere Pro, I can import a Ut cap, export as Ut, and the export is completely lossless and behaves similar to smart rendering. If Vegas has scopes (don't trust your eyes), do a lossless pathway check with RGB and/or YCbCr parade and vectorscope to ensure all your export settings are correct and hopefully perfectly lossless. For example, restore-chappy is correct in that you have Vegas set to export as progressive, but this could be because Vegas is interpreting the footage improperly. Also, you have the PAR set to square, and NTSC video is anamorphic. Again, all these issues could be because Vegas is interpreting your footage improperly and defaulting to the settings that it chooses for the timeline. In Premiere Pro, the Ut codec that I choose is labeled "UtVideo YUV422 BT.601 VCM". Hopefully this helps.
Last edited by LMotlow; 16th Jul 2017 at 19:28.- My sister Ann's brother
Mediaplayers? LOL. If that's your criteria for choosing post production formats then I have nothing more to say.
Should it be set to upper field first, or lower field first, if I want to turn off de-interlacing? Thanks
1. Im using lagarith just because I heard it was good. Granted those posts may be 7 years old by now. I'll look into UTVideo. Would you reccomend FFV1 format? I dont know if Im going to end up encoding these videos in a visually lossless format for decent storage, or if Ill try and restore some of them (depending on how they look), or if Ill just buy a 4 TB HD for 100 bucks and save them all losslessly. I guess Ill find out when I get through half of these tapes lol.
2. Ill look into Premiere Pro since it sounds like theres less of a chance of me screwing it up. Right now Im more concerned with making sure my amarectv options are OK since thats the most labor-intensive part (capturing the lossless video). However I never had Vegas do any footage interpreting, I just took an NTSC DV template and adjusted it to the best of my knowledge. There isnt a Video8 option in Vegas lol. Do you know how or if Vegas can interpret my footage automatically?
3. Where do you see that PAR is set to square, when it should be NTSC anamorphic? I looked twice and dont even see that option. Thanks.
UtVideo, just worst compression ratio in exchange for much higher encoding speeds. Something which is much more desirable in HD videos that can have more than 4x the pixels. But with OP's SD video, it's really not a problem and Lagarith should be more than able to deal with it in real time. I'm able to encode 100fps in Lagarith for SD content, playback is perfect. Last time I used Premiere Pro, Lararith worked fine. I see that Vegas also works fine with Lagarith. At the end of the day, OP will still have to wait for the tape to play through in realtime which can't be accelerated so might as well use a more efficient codec.
Last edited by KarMa; 17th Jul 2017 at 02:39.
In Vegas you press ALT+ENTER to get the project properties up,There you set the initial settings etc,
The Par setting is the pixel aspect and yours is set to 1.0000 (Square) in Vegas but should be 0.9091 (NTSC DV)
I would set upper field first and set the same in the Vegas export window, One thing to bear in mind is
There is no interlace flag in Lagerith so when playing with a media player you might want to turn on de-interlacing
for a better playback.
What are your final plans for these captures? Is it just archiving them in a lossless format ?
Vegas is definitely not the best option to be using to do this, however if you are going to be viewing your finished products on a modern progressive display you will want to perform some sort of a de-interlace at some stage.
If you want to get a general idea of what the de-interlacing will do, you can open your capture in VLC and turn the real time de-interlacer on. This wont be nearly as good as a real rendered de-interlace but you will get a feel for what it's like.
I can't really give much advice because like restore-choppy, what are your final plans for these caps? Do you plan to just archive them? Do you plan to encode to a cross-platform format that can be easily viewed? Do you plan to do some restoration? All these decisions influence how much TLC you might devote to each video. My advice, just get them capped. You said you have hundreds of tapes and that will be a monstrous chore in itself. Once you're done with that, post back here for advice on next steps as you might have a better idea what you want to do.
To answer question #3, PAR = Pixel Aspect Ratio and in your Vegas screenshot the PAR says 1.0000 which means square pixels. This is completely wrong for NTSC video which has anamorphic pixels i.e. not square. They have a PAR of 10/11 or 0.9091 which means they are taller than they are wide. NTSC video has a Sample Aspect Ratio (SAR) of 720x480 but that includes overscan. Technically, the picture is supposed to reside within 704x480, and on old analog TV sets that was all you would see (modern digital TVs typically show the overscan or the full 720x480 SAR). SAR x PAR = DAR, Display Aspect Ratio. Thus, 704/480 x 10/11 = 4:3. Square pixels in a 4:3 DAR correspond to a SAR of 640x480. I have no idea why Vegas is defaulting to square pixels which is why I said Vegas may be interpreting your video incorrectly. I am not a Vegas user.
I also notice that Vegas does not say what colorspace. This is one of the reasons I like UtVideo because at least in Premiere Pro, I can choose between BT.601 and BT.709. As you can probably see, video is highly technical and someone who doesn't know what they are doing can easily introduce errors in their workflow. This is why I suggested using scopes if Vegas has them. They will help you catch errors much easier than you could possibly do so with your eyes.
My final plan, as of right now, is to buy a 4 TB hard drive or two, and archive them all losslessly. I want to do this before these tapes from the 80s suffer any more degradation.
From there, I will edit clips to DVDs to send off to a few family members, as well as play with some restoration work. But I will only do that after I finish digitally cataloguing these files to a fresh Hard Drive. So can you please look at the AmaRecTV settings I posted above, I wont be playing around in Vegas for a month or so, if I do a few tapes a day after work. Thank you
the AmaRecTV settings seem okay,Why dont you post a short sample of the capture so we can see if everything look okay?
You can cut a short section using Virtualdub,
mark your in and out point and save as avi, Make sure you select "direct stream copy" under the " Video" setting.
Have you see this tutorial on capturing? its using Virtualdub as the software to capture but you may find
the section about de-interlacing instructive if you want to make your files progressive for web etc
here is a sample I took from my camcorder to amarecTV, identical to the previous give or take a couple frames or so (no editing in vdub)
I think it looks exactly like the original capture from amarecTV. VirtualDub is a lot easier to do the basics with (like video cutting and keeping the lossless format) than Vegas is. I used that youtube video you linked as a guide to capture some video in VirtualDub and unfortunately it dropped a frame in only 20 seconds of recording, so I think Ill stick with AmaRecTV for capture, which didnt drop even one frame in my initial 10 minute capture test.
Last edited by videon00b; 20th Jul 2017 at 19:38.
While youre here, do you know why my GV USB2 device's proc amp settings sharpness is set so low (as seen in original post)? Is that because Im using Standard definition, so it automatically adjusts for my low video quality? These settings were identical in VirtualDub and AmaRecTV (as they should be since its the hardwares settings not the softwares). Should I turn up the sharpness?
Last edited by KarMa; 20th Jul 2017 at 21:25.
Im hesitant to change the proc amp settings on my device since it seems to work so well out of the box. I did two captures in AmaRecTV, one with procamp sharpness set to default (16), and one at 0. would you mind looking at them and seeing which you think is better? Thank you
Procamp sharpness at 16:
Procamp sharpness set to 0:
I cant tell the difference myself.
One more question: while doing multiple samples of the same video, one sample came out with a few messed up lines. Its only for one frame, and on a small part of the picture. But its not something that can be fixed post-capture.
Would using a Datavideo TBC-1000, rather than my Sony CCD-TRV85's built-in TBC, fix this problem? Its kind of nit picky, but I wouldnt mind spending 100 bucks or so on one if it would give me better captures. Thanks
Thanks for more samples.
Last edited by LMotlow; 26th Jul 2017 at 03:11.- My sister Ann's brother
Here is a recent listing of one that was remarkably found new in the box, the price it went for was actually very good but you get an idea of what they are worth.
TGrantSales is a widely recognized pro shop that rebuilds tbc's and other gear and guarantees the work 100%. It takes skill and patience for this sort of thing and does not come cheap. https://www.tgrantphoto.com/sales/index.php/categories/time-base-correctors
Be veeerrry careful about eBay. Too many horror stories connected with their stuff.- My sister Ann's brother
AvsPmod and see what you mean about the clipping.
My capture device (GV-USB2) manual is in Japanese, so I dont know about what the neutral values are. I wish I did though.
I took a variety of samples messing with the procamp settings. Here is the settings of the best capture I got so far. I turned down contrast by 10 points, and turned off sharpness from default value 16.
Here is a sample of this capture:
AVSPmod says there is no more brightness being clipped, however there is some clipping of the low levels. Histogram picture:
Compared to other samples I just did of messing with the procamp, if I turn down brightness 14 points (while keeping contrast at 105 default), then my lows get clipped significantly more, while I still have clipped high ends (though not as much).
Putting my contrast at 100 and brightness at default 128 gives me clipped high ends too, as well as lows.
How should I go about getting the low ends to not be clipped? Should I maybe turn up the brightness while keeping the contrast lower? Or is there another value I should mess with? I want to get this correct so I can start capturing tapes. Thanks a lot.
Last edited by videon00b; 2nd Aug 2017 at 19:18. Reason: added histogram images
I bought a TBC-1000 for $100 a couple of years ago and it works just fine. It doesn't always help and it's not perfect — it takes the gain down a bit too much — but it doesn't have the dreaded softening problem we hear about. Don't think it would help what I'm seeing in your snapshot, however, which appears to be defect in the tape surface.
I detest those DirectShow brightness and contrast controls compared to the gain and black level controls of pro gear, but oh well. Brightness brings the level of everything up and down while contrast expands or contracts the range from darkest to brightest tones. Your goal is to preserve visual detail at both ends of the range. If something has to give, my taste is to keep bright detail and throw away dark detail if I must. Realize also that you are often working with clipped recordings and there is NO detail to bring out.
For instance, notice that much of the closer horse in your snapshot appears to be a featureless white blob. You need to bring the white level down until it looks like a real horse again. You can do that by decreasing brightness or contrast, but you will have to twiddle both to maintain detail in the shadow areas. If you find that the horse's white blob simply turns gray without revealing any detail, then you have a clipped recording and nothing can be done.
For that sample the brightness might have been set a tiny bit too low and the contrast a bit too high. For these tape sources you often get crushed blacks along edges. Better to keep them that way than to ruin the black levels everywhere else, in my opinion.
As I suppose you know, brightness controls the black levels and contrast the white levels. But the two aren't independent of each other so, as JV Raines says, you adjust them both when you want to achieve some specific effect.
Heres a sample of Brightness at 133, contrast at 94. I think this is the best one yet.
As you can see in the histogram, theres no crushing of brights, and only a tiny line of blacks being crushed. I dont know what that line is, maybe its the extra space on the edges from the conversion of standard resolution to DVD resolution. It shows up on all the samples so I think its insignificant.
This is kind of unnecessary, but heres a sample of the same scene, but with contrast down 2 points.
Here you can see that when the contrast is turned down 2 points, the graph moves further from 255.
Better, but I think I'd lower the contrast more. Some of this is somewhat subjective, I think. It might depend on what other scenes look like as in this one you have a bright sky and a white horse. The picture below shows another way to aid in setting up the proc amp. Red shows black levels below 16 and I don't really see where the green is showing white levels above 235. The black bars will either be cropped away or replaced with 'fresh black", I assume, as well as the head switching noise along the bottom.
It also depends on how the end result is to be viewed as what looks 'normal' on a computer screen might look way too 'bright' on a television. Here's the script used:
AviSource("amarec(20170802-2135) b133 c94.avi")