1 Attachment(s)
After reading several board messages about this device and what it 'does' to tapes, I decided to grab one cheap ($75) and see for myself how it performed. On the Mac side of the internet, users are in love with it, while on the other side, users critique its affect on video quality due to compression and chroma subsampling. There haven't been too many clips available, so I wanted to see for myself.
For those unfamiliar, this device is made by Grass Valley and works through analog input processed through an onboard DV encoder. It is extremely Mac friendly and works with a Thunderbolt port after adapters are used. I tried to use it to test with Windows 8.1 with no luck, but I didn't plan on using it with Windows anyways. The largest gripe about this device is the codec itself. DV is almost older than I am and by compression standards very outdated for today's needs. DV is lossy, meaning it uses intra-frame compression to reduce file size, but this compression is not as heavy as MPEG-2 or H.264. The largest gripe about DV is chroma subsampling. A fully lossless image will have no subsampling, noted as 4:4:4, while MPEG-2 and H.264 are noted as 4:2:0, or 25% of pixel color fidelity. DV also subsamples at this rate, however, it does it in a different way. Attachment 4136 4:2:0 gives the illusion of better color fidelity as colors are grouped into clusters rather than stripes. This allows for sharper color transitions and better overall appearance. 4:1:1 worked well with CRT and analog equipment, but in a world of LCD panels it doesn't fit very well. With Mac, the device gave pretty good results for what it is. There were some scenes from a tape with text that displayed mosquito noise, but I notice as I cleaned it up that the filters I would have used on a 4:2:2 sample were pretty much the sample and cleaned up any issues that DV might have introduced. There were no huge macroblocks in my test captures and a test conversion to DVD maintained picture fidelity surprisingly well. I don't claim that this is the best piece of equipment out there, but after having tried it, I feel comfortable recommending it as a secondary option. While there are several pieces of equipment that are better, there are a few factors that come into play here:
That doesn't mean to say that anyone can't do it, but there are people out there that are willing to give up a small difference in quality for a device that is readily available and does what it claims. In summary, the 110 or big brother 300 are not the best pieces of equipment on the block, but they do a decent job at their price point compared to other equipment options that can be purchased new. Rating: 8/10 Let the hating begin :P Chroma subsampling table |
You mention not finding any captures anywhere online, yet did not add any yourself.
Could you do so? That may help others here. :hmm: |
Quote:
I hope you're not saying that RGB as used in analog displays is no longer valid? I'm not sure what you mean by that business about some video color formats not being suitable for CRT's. I might try looking for an LCD that uses a different color system for display, but I have it from reliable sources that LCD's use the same RGB system that other display devices use to display images for humans, but LCD's just don't have the black deail or depth, or image stability that other display types obviously have. Also have it from reliable sources that in order for an LCD to show you a picture that your eyes can actually see, it has to convert those 1's and 0's to light waves (analog), just like the analog light rays that other display devices generate. :) I will never understand why anyone would recommend capturing analog originals to DV. Not only is there this myth out there that DV is "sharper" (meaning what, exactly?), but there's this mysterious and stubborn kind of denial going on which insists that two lossy encodes are as good as one. You can get rid of the ragged, noisy edges and mosquito noise, of course, but what about having not to destroy 30% of the image to get rid of that junk in the first place? OK, end of rant. I accept that analog->DV is faulty but it's not all that bad (meaning that it's not all that good, either), it's extra work when it shouldn't be, it'll never look better than it does the first time around, but it's good enough for those who don't need or don't want better, can't possibly get better anyway for one reason or another, or can 't see that well, or don't care, or just think it's cool. I saw what a "pro" did to my niece's wedding tapes by capturing them to DV. She saw it, too, and cried for a week. Thank heaven she kept the tapes, which took me 3 months to recapture as lossless YUY2 and to clean up for a far superior final product. But I get the point of your post. DV isn't really so hot, which many of us already know. There are those who can live with it and, then, there are those of us who can't. |
Quote:
You have to consider that in 2004, the average TV was a 27" CRT, and also consider that over the years they have gotten bigger, so expect that the TV size for 1994-5's DV birth was smaller. DV was primarily a consumer format with broadcast support, and the average person did not have a properly calibrated state of the art CRT TV, and most likely let their kids play with the settings unless it really threw the picture out of whack. Quote:
Quote:
As per the sharper idea, I actually haven't seen it, but the small prevelence of mosquito noise can have a poor sharpening effect. If you look at the color model attached in my post, you can get an idea that the less-than-great chroma separation could create a faux sharpening effect. Because of its compatibility and friendly out of the box with editing software, I would generally recommend this to someone who needs to stick a VHS clip in a larger project. It won't look horrid, and most people will know it will be VHS and won't expect much anyways. I have gotten to the point that I know my way around VirtualDub and decent with filtering, but to explain that to someone that hasn't peaked at the idea of uncompressed video will take a long time. I bought the Canopus to see what the fuss was (for better or worse), and for $75 I will more than get my money back, but I wanted to share my experience with those here so that anyone looking will have a reference of what to expect, and I will be posting clips shortly to give people an idea. I will also be posting a 4:2:2 clip of the same scene so the difference becomes obvious. I don't support DV, but I don't think its the cardinal sin it has been painted to be. More importantly, it should not be a format considered by a professional for someone's personal tapes. A VHS-to-DV professional would be an oxymoron :). Quote:
Do you have copies of what she received and what your final product looked like? It would go along very well with this thread. |
I agree with your conclusions 100%.
That wedding vid project mentioned was 5 years ago. Two captures and working files were long since archived to a small IDE hard drive now in the owner's possession somewhere in San Diego (a few thousand miles from me). Come to think of it, I really should try to have her retrieve that HDD and transfer the contents to a newer drive or other media. |
premiumcapture, what I would suggest for your test is to use a digital source over analog and even test multiple devices. That way you can compare to the source. I did that previously but my original source was already DV output over RCA. If you have non DV digital source and the right card to output over analog that would be ideal.
Quick tip here about the Canopus, you can use the Canopus to output previews directly to TV from the timeline. |
Quote:
|
Quote:
|
Quote:
I posted these previously, here is screenshot of the DV transferred over firewire (note the images are 720*576 but should be 720*480). Not sure how I managed to do that but it was years ago. Wrong template in the editor so they both went through same process. Also note they are two slightly different frames. http://www.digitalfaq.com/forum/imag...13/07/dv-1.jpg Here is the capture from S-video using a Canopus. http://www.digitalfaq.com/forum/imag.../canopus-1.jpg The criticism here is my source is already DV, if you can use a neutral source such as lossless capture from some other card that can provide output to your capture device then you can really make some comparisons becsue you can compare to the source. |
Just to elaborate suppose I have DV camcorder and output the same recording over s-video to the Canopus, DV camcorder with passthrough and some other DV capture device.
I can take screenshots of each capture and compare to the original source on the DV camcorder and see which one made the most reliable capture. You can do the same thing with DVD player.... The key is you have a digital source you can go back to see what it's supposed to look like. The criticism I heard was the source was DV. If you had a lossless source you can output over analog with some other card now you can make decent comparison of the capabilities of any capture device. |
If the original video is saved as DV, it is the best capture it will ever be, because it wasn't really captured at all.
The review of the 110 and DV are for application towards capturing analog video like VHS and others. The only situation where DV - analog - DV would make sense would be if it was a NTSC <-> PAL conversion, but even then there are other tools. |
Quote:
|
Quote:
The analog inputs on the Canopus were never designed to take input from a DV device. The device itself was meant to be an analog pass-through DV camcorder without the camcorder. Since these are no longer made, yet this process was one of the most popular means of transferring analog mediums, the Canopus still sells. This is also the reason why it will show up as a camcorder in capture/transfer programs. |
Quote:
This is what my source looks like: http://www.digitalfaq.com/forum/imag...13/07/dv-1.jpg How well does the Canopus perform? Well we capture from the camcorder over S-Video which is analog. Here it is, now you can compare it to the original above which is wht it is suposed to look like. Now do I want to see how well a camcorder works as passthough? I can do the same thing. |
1. DV Camcorder >> Firewire to computer >> Screenshot to compare captures
2 DV Camcorder >> S-video >> Canopus >> Screenshot to compare to 1 2 DV Camcorder >> S-video >> Camcorder using passthough >> Screenshot to compare to 1 3 DV Camcorder >> S-video >> <insert any capture device here> >> Screenshot to compare to 1 Get it? |
Understood. Screenshots are useful, but very limited. The problem is finding a DV source with exactly the same material as the analog source, and of equal quality or lack of it.
Over the years this comparison has been going round and round, always with the same conclusions. People like comparing apples to oranges for some reason and come up with all kinds of theories. The idea seems to be that all video is alike as long as it ends up "digital". That's not the way it works. People take DV and copy it to a PC using DV hardware/software designed or that purpose. Then they take analog and capture it to a PC using the same DV hardware/software or some cheapo $25 USB stick and try to compare the two. That horse died a long time ago, but anyone who wants to re-invent the wheel is welcome to try. |
CRT vs. LCD
At least half of more of CRTs were terrible, especially models from the late 90s and early 2000s, with their geometric distortions, bad hue/saturation/tint issues, and blown out IRE. Yes, the newer technology that replaced (LCD) it was often worse -- especially for those first 10 years. Mouse trails (response time), poor black, "shiny" screens, etc. Much of that still persists even now. But there were good CRTs, and there are/were good LCDs. So blanket statements, either way, are silly. My favorite Sharp CRT TV blew out last week. The high voltage board blew out on it. RIP 1995-2014. The color was gorgeous, the blacks really black. Lots of CRTs had/have pathetic IRE and black/grayscale palettes. Don't succumb to revisionist myths! Canopus ADVC. I've discussed this to death for 10+ years now. The Canopus works fine, but there are better ways to transfer video. It's 4:1:1, so the colors are overly compressed. Color compression is even more obvious now on modern 50-60-70" HDTV screens. DV was never intended for conversion, and only Canopus (later bought out by Grass Valley) ever did this. Furthermore, Canopus used BS marketing, both suggesting TBCs were magically not necessary, or inferring to other capture cards had a/v sync issues. Then, as if that wasn't bad enough, the Canopus was actually more costly than better cards! For Mac users, it's one of only a few choices. A Mac is not good choice for a capture system. For Windows users, there are many options spanning almost 15 years now. Be it 2014, or 2004, reviews for this unit mostly remained unchanged. People like it because it "works", but others dislike it because the quality is just not there. Much like CRT vs. LCD, you cannot make blanket statements about the Canopus ADVC-50/55/100/110/300 vs. DVD recorders. At least half of all DVD recorders were pathetic, while others can run circles around the Canopus. It's really a model-by-model issue. (For example, an RCA with the Zoran chipset, circa 2007, makes a Canopus look like complete crap. But your average Panasonic deck makes the ADVC look like a champion.) An ideal capture test? A non-retail SP mode clean VHS tape using either a Panasonic AG-1980P or JVC HR-S9600/9800/9900. If the homemade tape had some chroma shifting (color bleeding), then that would make the errors starker. I have a few tapes like this, but I no longer have an ADVC to test with. I do have the DAC-100. Screenshots actually help most for this kind of test. It's always been my intention to show these tests, but I've never gotten around to it. |
Thanks for the extra input.
Quote:
|
And my elederly Dad loved his old Panasonic 32" SD tube TV and cable without the digital HD converted box because it as easy to use, cable channels of interest to him were in a simple line from 2 to 78 (give or take a few), no need to worry about aspects ratios and resolution getting confused when moving among the channels.
The ADVC110, etc. series are but one tool in the chest that can be used to convert SD analog video sources to a digital format ()inthis case DV). They have their place and strengths and weaknesses. The key is to select an appropriate tool for the job at hand considering what the client/customer/end user wants and expects and cost/time limitations on the job. |
Thanks for the review, but without clips to point to and say "this is what I see here" it's just another big war of words, theories, and varied experiences.
Quote:
Quote:
|
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.