DVD quality capture with what capture card ?

[dvd_master]

The only capture card I have is a Dazzle Digital Video Creator 1, which is a piece of crap.

If I want to capture VHS or LD, and I will eventually output it to DVD, what is the best capture card for preserving all it's quality ?

The cheapest yet highest quality capture card ??

[Ookami]

Read the FAQ, there are many threads linked about "the best capture card" and about "VHS -> DVD".

[Herske]

Do you have a DVD quality source? (as good as 35 mm film, that is)

For VHS or LD, a DC10+ is overkill. If you're going to capture from LD using an SVIDEO cable, you may get decent quality from a 8x8 tuner.

[Ookami]

> Do you have a DVD quality source? (as good as 35 mm film, that is)

This is a joke, right?

> For VHS or LD, a DC10+ is overkill

Another joke, right?

What is DVD quality? There are so many different looking DVDs out there that I do not know, anymore :) .

As for VHS a DC10 is overkill. That is only funny.

Facts (I'm reapeating myself, I know):

a) DC10 quality is average (according to Hawk who knows more about capturing than 99% of the people I've "met" on the internet)
b) VHS captures CAN be improved, if it wouldn't be so then no one would use image improving filters
c) You can beat 99% of the semi-professional capture cards with the cheap Asus 3800 deluxe+SVHS in+Huffyuv combo

Again, read the threads linked in the FAQ, all this has been discussed before.

Best quality

Cheers,

Mijo.

[Herske]

>This is a joke, right?

No, because I assumed the original poster meant the quality of (most) commercially available DVD's (those who do not have obvious mastering/encoding flaws). A DVD with a VHS source will look like crap.

To put it differently: I would not spend my cash on sdi i/o board unless I had a *great* quality source signal.


>a) DC10 quality is average (according to Hawk who knows more about capturing than 99% of the people I've "met" on the internet)

I don't have the slightest clue who is this "Hawk".

Why exactly do you think that DC10+ is not enough for VHS signal? Resolution? Bitrate? Y/C improper separation? Bad dynamic range? MJPEG artifacts at 6 MB/s? Elaborate.


>b) VHS captures CAN be improved, if it wouldn't be so then no one would use image improving filters

True, vhs captures can be perceptually improved but how exactly is this related to capture cards?

[djtrixx]

I have tried many capture cards, primarily from ati, starting with the raedon 64 VIVO on up, and they all suck ass! I have about 5000 music videos on laserdisc and vhs, and needed a surefire way to copy them onto DVD, without dropped frames or worrying about macrovision. I finally spent the money and bought the Pinnacle Pro-One from Newegg on sale for $670.00. I know it sounds expensive, however the quality is more than worth it. Plus it comes bundled with adobe premeire 6.0, illustrator, Pinnacle impression DVD. The "breakout box" as they call it is actually a DV bridge, which is nice for all you cam corder freaks. The only downfall is the size of the card - if you only have a mid size tower, make sure there's nothing in the way of the PCI slot because the card is so long it runs from the back to the front of the box.

Using this card, I capture at 720x480 for 2 hours with 0 dropped frames. No screwing around, no tweaking, no worrying about not touching the mouse. That alone is worth the money. Plus, NO MACROVISION! That's right, I copied about 20 vhs tapes already and didn't have any of the macrovision glitches that every other card has.

What's the best card? Pinnacle Pro-One. Definitely.

[thorazine]

Well, I don't know if a DC10 is overkill for VHS but a Pro One sure is, IMHO :)
You are probably right that a DV capture card is better than those Philips or Conexant low-end chips but you probably will get the same results with a Canopus or Dazzle DV bridge and a cheap firewire card with no software bundle (and that's about half the money of a Pro One).
OTOH I won't believe you can get the quality of a typical major DVD movie starting from a VHS source no matter how good the capture card is and how much filtering you apply until I see proof.

[Ookami]

First, I want to apologize for my quite harsh, original posting.

>>This is a joke, right?

> No, because I assumed the original poster meant the quality of (most) commercially available DVD's (those who do not have obvious mastering/encoding flaws). A DVD with a VHS source will look like crap.

I agree if you don't use filtering to improve the quality. And a DVD with VHS will look like crap if you don't put much time and work in it. It can look good if you put much work in it. But, I still don't believe that you can get DVD quality from a VHS with the methods we use here... I've wrote quite some times near-DVD quality, and I stand by that (I've done that myself).

> To put it differently: I would not spend my cash on sdi i/o board unless I had a *great* quality source signal.

Your standpoint is clear.

>>a) DC10 quality is average (according to Hawk who knows more about capturing than 99% of the people I've "met" on the internet)

> I don't have the slightest clue who is this "Hawk".

Well, I (and many others here) have. :)

Hawk, was one of the first persons who managed to surpass the so called semi-pro capture cards with cheap GPUs. A few years ago he wrote the first HQ capture guide, and he (and his friends at Ultimateboard) tested quite some capture cards etc.

If you understand german, here's the sneak preview of Hawk's capture guide 3.0 : http://ookami.iwebland.com/Hawk/kommenar-vdub-filter.html

> Why exactly do you think that DC10+ is not enough for VHS signal? Resolution? Bitrate? Y/C improper separation? Bad dynamic range? MJPEG artifacts at 6 MB/s? Elaborate.

"Not enough" You statement, "overkill", reminded me of quotes like "for VHS capture 320x240 is more than enough, you can't get it better anyway". So, from the enough standpoint, you (or I) could even say, capture at 1/4 resolution is enough.

It is maybe overkill, if you just want to make a quick and dirty capture and convert it to a VCD. But even then it is always better to use a "good as it can get" source then a "what's that?" source :) .

As for the DC10+, the card is simply obsolete in any means. Why to use a card that uses lossy MJPEG (you can't use lossless), that has a average capture quality (comparing it to solutions like Asus 3800+Huffyuv)? I have somewhere the original explanation from Hawk, if I find it, I'll translate it and post it. If you want a quick and simple solution I think there are quite better solutions than such an old capture card.

>>b) VHS captures CAN be improved, if it wouldn't be so then no one would use image improving filters

> True, vhs captures can be perceptually improved but how exactly is this related to capture cards?

How? Simply, if you use a card that can only use a lossy codec for capture, filtering cannot be done the same way as with lossless captures. Simple, is it? If the filters have more information they can work better. More information can be gained by higher bitrate or lossless compression, full resolution, better source etc.

So, if you have card a) (e.g. DC10+) wich can only do MJPEG and has some other flaws and card b) (e.g. Asus 3800) wich doesn't use the BT chips and can be used with lossless codec, there is no doubt that the captures from b) will have a better quality (even without any filtering, later).

But, there is always the question, is it worth it? For instance, I've stopped to do HQ capture because I simply don't have the time and motivation. The few VHS sources that will not have a good DVD tranfer are another story...

Sorry for the chaotic style.

Cheers,

Mijo.

[dvd_master]

Well, I think the one you mentioned is far too much for me....

But i want all my VHS and LD transfered to my computer so i can later burn it to DVD after adding subs, etc. but I want the quality to be a perfect transfer since I know many people have done it before.

[Frisco]

>Hej Ookami

I can se you don’t like the DC10+ capture card.

I have a Asus 7700 deluxe cards and a DC10+ card.

The Asus 7700 +SVHS in+Huffyuv give OK quality, but it is NOT close to the great quality you get whit a DC10+ capture card.

When I chapter, input from HI8 video camcorder, whit the Asus 7700 card and then playing the AVI on the TV, and then compare it whit the camcorder connect to the TV, there is a great difference in the output. The capture quality is not great. When I do the same whit the DC10+ capture card, I can see NO difference in the quality. The quality is great.

A Danish PC-magazine (Magazine name in Danish :”Alt om data”) have make a test of a DC10+ card and the result, white TV-broadcast input (DVD quality) and capturing whit 6MB/s, was that the capture output is in DVD quality.


I hope you understand my bad English.

Hej Frisco

[thorazine]

Interesting points Frisco. I have the same feelings as Ookami against the DC10 (i.e. hardware MJPEG is obsolete, Huffy better than lossy) but I have never seen any real comparison between capture cards that tested them like they do with graphic cards or processors: can anyone point me to any capture card comparison test?

But one question, your comparison may be a bit biased: did you use the same output for both files or used the Asus TV Out for the huffy and the DC10 video out for the MJPEG?
You may be mixing capture (analog to digital) quality with output (digital to analog) quality if you used different outputs.

[Ookami]

>Hej Ookami

Hello Frisco.

> I can se you don’t like the DC10+ capture card.

You see correctly. Like I said, even if I would like the DC10+, there are better solutions out there that work on a similar manner. Why stick with a DC10+ if you can buy a DC30, also cheap?

> I have a Asus 7700 deluxe cards and a DC10+ card.

You have more than I have. :)

> The Asus 7700 +SVHS in+Huffyuv give OK quality, but it is NOT close to the great quality you get whit a DC10+ capture card. When I chapter, input from HI8 video camcorder, whit the Asus 7700 card and then playing the AVI on the TV, and then compare it whit the camcorder connect to the TV, there is a great difference in the output. The capture quality is not great. When I do the same whit the DC10+ capture card, I can see NO difference in the quality. The quality is great.

? Sorry, I think you are making a few things wrong. First, why do you not use your PC monitor for comparison of your captures? If I nderstand you well, your basicly, making a capture from the camcorder to the PC, then playing the AVI through the crappy TV out of the TNT cards The DC10 card should have a much better output than the TNT card. Also, I do not know if the 3800 series is the same as the 7700. When I was buying my 3800, the 7700 already existed, but I didn't want to take the chances. And MJPEG at 6 MB should be "great" for viewing, but I'm trying to make a point here. If you want to filter then you should always go for lossless as possible :) . It is logical, like with audio filters, you can, for instance use MPC for capturing and then say it's great, but capturing lossless is surely better. Now, I am aware that capture cards are much more complicated than this example... With capture cards you have dozens of possible image degrading sources (from cheap cables, different chips on the cards, drivers, sowftare, capture codecs and their settings up to interferences from other electrical components...).

> A Danish PC-magazine (Magazine name in Danish :”Alt om data”) have make a test of a DC10+ card and the result, white

TV-broadcast input (DVD quality) and capturing whit 6MB/s, was that the capture output is in DVD quality.

I don't trust magazine when it comes to such things, anymore. I've read too much rubbish. The german magazine c't is the only one wich has an benefit of doubt in my book :) . Second, most of those "reporters" are even too lazy (dumb?) to learn how to use Flask MPEG correctly and you think they would correctly set up a capture system from a normal GPU? It is surely more easy to install a DC10+ then to fiddle with a normal graphic card with TV in.

> I hope you understand my bad English.

My english is not better, but I think we understand each other.

@thorazine

> have never seen any real comparison between capture cards that tested them like they do with graphic cards or processors.

Me neither. Besides most of the testings are flawed in many ways (even noticable by technically challenged as me). Because of
such things, before buying something, I only trust forums like this! I do Usenet, web and forum searches and the results are always more reliable than those PC magazine tests. Altough, you must know whom to trust (somehow Tom's hardwares "MPEG 4" "tests" come to mind).

On Ultimateboard I've read some comments about the DC10+ and none of them were overwhelming. It is surely no coincidence that almost all Ultimateboard mods that captured switched to the Asus 3800 + SVHS (mostly, Philips VR 1xxx) solution. From ATI all in wonder, WinTV, DC10+ etc. All of them said that the Asus solution is much better. It has its flaws (can be read on those old threads linked in the FAQ) like not possible to capture at e.g. 480x576, no DScaler support etc., so you always have to wage what you want.

Disclaimer:

This is my last comment to this whole "capture card vs. graphic cards with TV in" thing. I'm tired of repeating the same things over and over, and I also think that everybody should use what fits best to him. When I was researching on capture cards etc. I've read dozens (hundreds?) of threads, articles, reviews etc. I first wanted to buy a dedicated capture card like the DC10+ but then, luckily, saw all those limitations of these cards (e.g. DC10+ had the infamous 74 min. bug). I do not gain anything from promoting old graphic cards and I don't care wich logo is on my components... Mind you, I'm a "Asus
Motherboard with VIA chipset" owner :scared: . Also, I will buy a dedicated capture card, but: a) it will surely not be a DC10+ b) it will probably not be the solution for HQ capturing.


Cheers,

Mijo.

[thorazine]

Originally posted by Ookami
@thorazine

> have never seen any real comparison between capture cards that tested them like they do with graphic cards or processors.

Me neither. Besides most of the testings are flawed in many ways (even noticable by technically challenged as me). Because of
such things, before buying something, I only trust forums like this! I do Usenet, web and forum searches and the results are always more reliable than those PC magazine tests. Altough, you must know whom to trust (somehow Tom's hardwares "MPEG 4" "tests" come to mind).

On Ultimateboard I've read some comments about the DC10+ and none of them were overwhelming. It is surely no coincidence that almost all Ultimateboard mods that captured switched to the Asus 3800 + SVHS (mostly, Philips VR 1xxx) solution. From ATI all in wonder, WinTV, DC10+ etc. All of them said that the Asus solution is much better. It has its flaws (can be read on those old threads linked in the FAQ) like not possible to capture at e.g. 480x576, no DScaler support etc., so you always have to wage what you want.

Disclaimer:

This is my last comment to this whole "capture card vs. graphic cards with TV in" thing. I'm tired of repeating the same things over and over, and I also think that everybody should use what fits best to him. When I was researching on capture cards etc. I've read dozens (hundreds?) of threads, articles, reviews etc. I first wanted to buy a dedicated capture card like the DC10+ but then, luckily, saw all those limitations of these cards (e.g. DC10+ had the infamous 74 min. bug). I do not gain anything from promoting old graphic cards and I don't care wich logo is on my components... Mind you, I'm a "Asus
Motherboard with VIA chipset" owner :scared: . Also, I will buy a dedicated capture card, but: a) it will surely not be a DC10+ b) it will probably not be the solution for HQ capturing.


Cheers,

Mijo.

Of course you cannot blindly trust web reviewers but the point is they doesn't seem to exist when it comes to capture cards.
I do also prefer to research usenet and forums but I think you cannot blindly trust bold statements from them because most people only have one particular piece of hardware on a given setup at a time: I don't know others but I only have one capture card, a conexant TV card, I played a couple of days with a all-in-wonder before returning it back, I've seen a friend's DC10 in action and I've seen captures from a dv bridge. I've got the impression that the the DV bridge gives the best quality by far and the dc10 would be a bit better than my conexant when you have very clean sources like digital satellite; but I don't feel legitimated to say "DV bridge is the best" like djtrixx did with the Pro One.
That's why I found so interesting Frisco's comments: because he had tested two cards with the same setup, the same source and the same capture methods (I assume): there's a big difference between that and "My new expensive card is way better than my old cheap card".

[Frisco]

Hello Ookami and thorazine

thorazine
>But one question, your comparison may be a bit biased: did you use the same output for both files or used the Asus TV Out for the huffy and the DC10 video out for the MJPEG?
Ookami
>Sorry, I think you are making a few things wrong. First, why do you not use your PC monitor for comparison of your captures? If I nderstand you well, your basicly, making a capture from the camcorder to the PC, then playing the AVI through the crappy TV out of the TNT cards The DC10 card should have a much better output than the TNT card …… …. Sorry, I think you are making a few things wrong. First, why ….

My tests have one purpose: to see the loss in the video signal.
The video signal form the camcorder to the TV is the best possible, (no loss in video signal in a capture cart and no loss in the video Codecs)
In the test I only use the Asus TV-output to the TV. And my test shows, that there is no difference whet the DC10+.
When I compare avi, from Asus 7700 and DC10+, the result on the PC monitor is the same. The DC10+ is much better.

Ookami
>If you want to filter then you should always go for lossless as possible …..

Yes, but the quality of the caputre Video-signal is so great, whit a DC10+, the loss in quality is so small, close to nothing, that it dossen matter. But I get a loss in the video-signal whit the Asus 7700.

thorazine
>DC10+ had the infamous 74 min. Bug

There is NO 74 min bug, whit the driver for Windows 2000 and XP, you can capture “forever“.

thorazine
>…… because he had tested two cards with the same setup, the same source and the same capture methods (I assume):

Yes, same hardware, same source, HQ video capture : 768 x 576. Whit the DC10+ I use Pinnacle software (it came whit the DC10+), and whit the Asus 7700 I use iuVCR.



Hej Frisco

[Ookami]

Altough I do not want to discuss the x vs. y, I am interested how you did the tests...

Your Asus can capture at 768x476? You use the WDM drivers?

The only "real" comparison would be:

a) capture from the same source (eg SVHS tape)
b) same resolution
c) same output (if you compare on the TV you should use the same output for both the captures)

A few notes: The Asus 3800, has a badder capture quality on Win 2000 and XP, than on 98SE.

The Asus 3800 uses the 704 x 576 resolution (not 768 x 576), so I begin to suspect that the cards do not use the same components.

> Yes, but the quality of the caputre Video-signal is so great, whit a DC10+, the loss in quality is so small, close to nothing, that it dossen matter. But I get a loss in the video-signal whit the Asus 7700.

No, that's simply wrong. MJPEG is lossy. Period. Of course, you can have a better MJPEG capture with a good card and source than a lossless with a bad card and source, but that's logical.

For filters it matters, I have done it and saw the differences with my own eyes. As I said, the question is only: "Is it worth it?".

> >DC10+ had the infamous 74 min. Bug
>There is NO 74 min bug, whit the driver for Windows 2000 and XP, you can capture “forever“.

There was a 74 min. bug (amongst many others), I did that research many months ago (beginning of 2001...).

Whatever, use what you want. I think we should open a TEST thread or something were users with a few capture cards can post their experiences. What do you all think?

@avih

In your sig it says Asus 3800, you also had some Win TV etc, if I remember well... What are your experiences?

Cheers,

Mijo.

[dvd_maniac]

I have purchased the Radeon AIW 8500 with 128mb and was wondering if you all think that it is a bad capture card? I personally love it because of it's dvd quality capture. It even captures audio at 48khz so it can be converted to dolby digital. I would be shocked but interested in a better solution for around the same price.

[Frisco]

Hej Ookami
>> Yes, but the quality of the caputre Video-signal is so great, whit a DC10+, the loss in quality is so small, close to nothing, that it dossen matter. But I get a loss in the video-signal whit the Asus 7700.

>No, that's simply wrong. MJPEG is lossy. Period. Of course, you can have a better MJPEG capture with a good card and source than a lossless with a bad card and source, but that's logical.

I don’t know , but you are wrong. I think you a forgetting the loos in the video-signal in the hardware and only consider the loss in the video signal in the software codes (MJPEG).
Maybe we must misunderstand eachother, so I will give an example :

The hole think is about loss in the Video-signal. We have a clean source like DVD the Video-signal is 100%.

Asus 7700
Input video-signal is 100%, we have a hardware loss (Asus 7700) about 15%, then we have a software loos in the video codes 0% (Huffyuv). The total loss in the video signal is 15%.

DC10+
Input video-signal is 100%, we have a hardware loss (DC10+) about 1%, then we have a software loos in the video codes 2% (MJPEG 6 MB/s). The total loos in video signal is 3%.

The numbers I use in only for the example.


>Your Asus can capture at 768x576? You use the WDM drivers?
Yes

>The only "real" comparison would be:
a) capture from the same source (eg SVHS tape)
b) same resolution
c) same output (if you compare on the TV you should use the same output for both the captures)

Yes, it is the way to do it.

Hej Frisco

[Ookami]

Originally posted by Frisco
Yes, it is the way to do it.

Hej Frisco

Hello Frisco,

then you have done it correctly (as far as I'm concerned)! So you can say that the DC10+ is better than your Asus card.

Thanks for notifiying and all the best,

BTW,

> I think you a forgetting the loos in the video-signal in the hardware

Nope, if you re-read my post(ings) you should notice that I've mentioned that. Altough, my postings are always chaotic, so you have to be very patient to read to the end.

Mijo.