digitalFAQ.com Forum

digitalFAQ.com Forum (https://www.digitalfaq.com/forum/)
-   Project Planning, Workflows (https://www.digitalfaq.com/forum/video-workflows/)
-   -   Edit/restore videos on capture computer or another? (https://www.digitalfaq.com/forum/video-workflows/9866-edit-restore-videos.html)

Koreth 07-17-2019 02:31 PM

Edit/restore videos on capture computer or another?
 
I'm curious of the workflow many of you here use or recommend. From my lurking and reading, I see that older video cards are most often recommended, which means most of the rest of the components in the computer will often have to be of similar age.

What I find myself wondering, do you guys then perform edits on the same computer you did the capture on? Or is the capture computer dedicated to capture only? And if the capture computer is capture only, how do you move the freshly captured videos off the computer and onto your computer you use for editing/restoration? USB stick? shared network drive?

I ask because it occurs to me that one might be more productive, especially if one had a lot of tapes to capture, to split duties across computers. The capture computer could be capturing a tape, while you're currently working on restoring the one captured previously on another computer. Having a separate editing computer would also let it be modern spec, as it would not need to be tied to compatibility with capture card, and thus could have a lot more horsepower for editing and encoding than the P4-era systems needed for capture.

Am I onto something here? Is this indeed how most of you have your workflows set up, or am I coming up with crazy ideas?

keaton 07-17-2019 04:35 PM

Hello, and welcome! All perfectly good questions. I think it takes everyone some time to develop their workflow. I understand the idea to be efficient, because this can be a very long project. With time, I've mellowed a bit as I've gone through all the steps of video capture and restoration. I have a better understanding of what my process is, and am more patient that this will take time. Above all, I want to do it as right as can as fast as I can. Which means sometimes things will go slowly because I want to do it correctly the first time. I have gone back and redone things as I've learned more. So, you may need a trial period of just going through the whole process and finding out what works best for you and getting some confidence that you have a good plan/workflow in place and that may relax the concerns about whether it can be done and how long you might expect. I've come to a place of calm and confidence with what I'm doing, and I don't assign deadlines or rush anything. I know I've got the pieces in place and it will come with time. It's a marathon, not a race. :)

My setup is an XP SP2 station for capture. It's an older machine, and doesn't need to be great with processing power or hard drive speed. Just enough to capture in real time with no other tasks running. I have another XP SP3 machine with latest Microsoft C++ Runtime libraries with beefier processor, faster rpm hard drive for the more intensive video editing (deinterlacing with QTGMC, for example), and color accurate monitor (important for video editing). I also have a virtual machine of XP SP3 with latest Microsoft C++ Runtime libraries on a linux laptop for less processor intensive tasks or experimenting in avisynth with some short demos before I put the final script to the whole video on my main machine.

There is debate on here as to whether one could use SP3 for capture if they had to. I think the ideas against are that it adds nothing and may inhibit your machine some from the task of capturing in real-time. What I realized is that once you switch to editing/restoring with Avisynth, that I needed to use a lot of plugins which required Microsoft C++ Runtime libraries which I was not allowed to install unless I had SP3. I happen to have the luxury of a separate machine that can have the latest XP on it. If you only had one XP machine to both capture and do video work with, you'd have to use SP3 if you wanted to take advantage of all the great plugins that Avisynth has to offer. I suppose many use Windows 7 machines for Avisynth work as well (I'm XP only for video/audio, otherwise linux for everything non-video/audio). It may depend on whether you would like to monitor the transfer process. If, for example, you want to be sure your video or audio levels are not clipping during capture, then there may be no advantage to trying to multi-task with separate machines. Some cards allow you to do this. Some do not, and so you'd have to preview the tape before capture to ensure of this. Although, when you have a machine chugging on an hour of video at 4 frames per second, then I could see the advantage of a separate capture station. One reason I use the older station for capture is that it has a rather good onboard sound capture chip (given decent praise on this forum anyway), but my XP editing station can hold a better processor.

My main editing XP machine can do QTGMC on a medium setting at about 4 frames per second. Where as if I try that in a virtual machine on my laptop, it would be about 1.5 frames per second. I don't know if that would be significantly different if I wasn't using a virtual machine. But the idea is that some types of video work are very intensive, and require as much processor and I/O power as you can get.

I have an external USB hard drive that I use to transfer video off the capture station. 1 or 2 TB is what I'd have to use, as XP has a 3 TB limit on the file system, I believe. But, that's just a swap drive. I have multiple copies of my lossless captures backed up. I have a non-XP machine with a 4TB drive to store that, as well as an external USB 4TB drive that I can store offsite. Backing up data is quite important. I probably have less than 100 hours of video to archive, so I can store HuffYUV AVI or Lagarith AVI of that within 4TB (roughly 30 to 40 hours per TB). If you had many times more, I would consider not keeping all the original captures. I personally like the idea of having a "digital master" of the tape that I could always go back to if something better came along in the future or I just wanted to take another pass at how to restore something.

Everyone has an opinion on what order to do things in. The most important thing is to stay lossless until your done improving the video. I'm just sharing what I do, and isn't necessarily to be taken as advice. But if you are curious, here's what I've arrived at thus far, and am always open to changes as I learn more. After capture, I edit out any unwanted beginning/middle/ending frames (Direct Stream Copy save in Virtualdub). Switch to editing machine to work in Virtualdub fiters for things like Color/Contrast correction, Camcorder Color Denoise, keeping that in a lossless format. Part of this step includes cropping video of any overscan and other black edges (so they don't interfere with histograms). Some advise to color correct later, as it is true some Avisynth processing could impact your histograms. If that is an issue, the plan isn't set in stone. But my current plan chooses to color correct before losing the 4:2:2 color resolution when converting to 4:2:0 for the next step. Next step is to do cleanup of timing errors, noise, deinterlacing, etc. (the computationally intensive stuff in Avisynth), add the border to pad aspect back to 720x480 and save that to lossless format. Export the audio to wav for any Audacity or SoundForge work then merge that audio back to the video. Lastly, convert the lossless video to the lossy format of choice at a high bitrate so I can watch it from a file server or hard drive.

Having said all of that, when I need to make a video for someone else, I do all of this. But for my own collection, I am more focused on just digitizing the video. I don't like to use fear as a motivator. However, it is true that the hardware and tapes involved are the biggest risks in the equation. So I mostly charge ahead at getting it all digital, before any of those analog pieces fails. I also got here after doing some video and audio work on what's been captured so far. Building comfort with what problems I see when I capture tapes has a plan of attack later when I go to restore it. If I encounter a new problem, I'll step back and see if I have a plan for that kind of defect by trying a different VCR or explore other Avisynth plugins, etc. Or come to accept that it is the best I can do, and if I cannot fix it later that is OK. I find that I can get into a better "groove" by just being in "capture mode". It is true that regular repetition keeps the skills sharper. Once it's all digital, then I can switch to a different "groove" of restoration.

I've arrived at all of this over a few iterations, before really get comfortable with an approach. The approach is always open to more modifications. But they become fewer and smaller with time, because there's so many things to consider, depending on how detail oriented you are. :D

Best wishes and good luck!

lordsmurf 08-01-2019 04:16 AM

We move files via eSATA and USB3. (A 10gb LAN may be faster, but it's just more wires to lay, so we avoid it.)

- Capture systems are old/legacy boxes dedicated to capture, offline, mostly XP and ATI AIW, some 7 with newer USB cards.
- Editing/restore/author system are modern i7, all use Win7.

Man power is a bottleneck more than computers. Sure, 2 is manageable. But when you're tasked with 8, things get complicated quick. (Seriously, that's how many I sometimes maintain at once! :eek:)


All times are GMT -5. The time now is 07:57 AM

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