Which is better? DV or Huffy?

[mojo]

I've been capping laser discs for conversion to DVD and have been using the Dazzle DV Hollywood Bridge. The laser discs have a very clean image, but the encoding has been surprisingly difficult. The bitrate needs to be VERY high to get a decent image - it is making me wonder if the compressed DV format is introducing enough subtle artifacts to make the compression more difficult.

I know DV is a high-quality format, but it is lossy and definitely introduces artifacts to the video (it may be subtle, but it's still there). I'm wondering if capping with an S-video capture card using Huffy would be a better choice to insure the cleanest possible video.

Anyone have any ideas?

[vhelp]

mojo,
Yes, go with Huffy. Analog capture card route, that is.

I see you also suffer the color washout look too, and am finally fed-up
with it. Good for you. Nothing will be a good analog capture. So, go
for it and do your best.

Remember, it's not only the capturing (or however way you bring in the
video source ie, DV or Analog capture) but HOW you process it to a final
keep.

Which is the better route, DV vs. Analog ??
Analog, of course!! DV in itself, is a compression. Period. So, there
is bound to be some quality loss.
I don't know who told you that DV is a high quality format though. It's
a good comprimise but not something I would keep things in, with the
exception of miniDV tapes, because I know from experience that it will
come out perfect if I use the Analog Capture card route on it later. It's
only when you finalize it to DV format, THAT YOU LOOSE quality. Just
because it's digital doesn't mean its high quality You are finally
wising up to it now, hence your concirn here.

DV is an alternative like Huffy is to Analog Capturing. It's a medium.
So, which is the better medium ?? You guessed it.. Huffy!! Whos' compressed ??
that's right. Ok then.. Huffy it is. So, go Huffy on along now

My vote goes with Huffy or uncompressed.

-vhelp
PS: maybe you could share a sample or too of your work, to help us
understand your DV quality issue.

[Innershield]

I have done extensive testing with huffy and my AIW 7500 and my Sony 740 with pass-through. My conclusion is that DV is better. The colors were more natural and showed no signs of compression (not saying that they weren't there). Huffy made the colors look more like a cartoon and I could definitely see signs of compression. That's my 2 cents worth

[mojo]

Of course, 2 replies and 2 different answers

I have done some reading on Huffy, and, if I am to believe what the main site says, if you use the lowest compression it is supposedly a LOSSLESS format. If this is so, Innershield - are you using the right settings or is Huffy not lossless?

[Innershield]

No compression is absolutely 100% lossless. You can only achieve close to lossless. When they say that Huffy is lossless, they mean that it is as close as you can get to lossless. But if something is let's say 99.9% lossless or 99.8% lossless, can you really tell the difference. Especially not on the television.

[319]

I'd have to put my vote in for capturing via Dv instead of through the video card using HuffyUv. While I've found that capturing a laserdisc the dv did introduce some obvious artifacts, my captures using my Ati Radeon video card through rca cables had more of a general video noise to them. The real difference is noticable once encoded into mpeg, where the Dv captures to me looked much better.
While I've run into some other unwanted issues with my dv capture, namely that the mpeg file when encoded will skip and stutter occasionally (perhaps this is due to the one or two dropped frames that sometimes occur) the overall quality is quite good.
I think with a laserdisc you could capture either way and the quality would be decent- but when capturing from vhs or tv, the dv captures to mpeg encodes are much better.

319

[yg1968]

Most of your loss will occur when you convert to mpeg-2 or mpeg-1 as it is much more compressed than DV. Any compression will introduce small subtle artifacts that you can see on a computer screen (but these are very hard to notice on a TV). Incidentally, installing the Canopus DV codec will give you less artifacts upon playback than using the Microsoft DV codec.

vhelp,

Mojo never mentioned anything about colour washout.

[cosmichippo]

Hello all,

I too have been very curious about this as well.

I had considered the Hollywood DV Bridge and then found out about the Canopus 100 that offers better quality and less headaches.

During this time I had heard about the HuffYUV encoder. In reading about it it seems to offer the best consumer level conversion. I don't believe that at it's non-compressed setting it is compresses the siginal at all. It is conveting to digital so you do loose some minuet detail. Analog is continuous and digital is a series of sample points. I know that over simplifies but it is basically the deal.

Anyway, what analog capture card is the choice of most quality seeking folks? Obviously the convertors built into AIW's and Nvidia Personal Cinemas aren't really up to what Huffy can handle. I've heard some good stuff about the Pinnacle cards but haven't seen anything produced by one that impressed. I think that letting the analog signal get into the noisy PC case itself may offset the benefit of not compressing the capture.

Anyone have specific experience with any good to great analog capture cards that are suitable for use with HuffYUV?

Glad this thread got started.

Mike

[yg1968]

It remains a subjective debate. Huffy is compressed 3:1. So capturing to Huffy requires a lot of diskspace. My experience with analog cards hasn't been a good one. I had an ATI AIW 128 but I got sick of having to deal with dropped frames and out of sync issues. So I stopped using and decided to get a DV camcorder with passthrough instead. I haven't regret it. It has been a hassle free experience to use DV. DV requires 13 gig per hour of disk space but it is only a temporary file. I don't know if DV is the best option but it is certainly one of the better ones. No dropped frames and no out of sync issues! I haven't noticed any colour washout issue either.

[thxkid]

well even having a very good MPEG2 capture card I found out that my laser discs look a heck of a lot better if I record them directly to my Panasonic DMR-E20, then its a one step process and at .87 cents per DVD-R, Star Wars never looked so good, Lion King came out great also, wish my blasted pioneer laser Disc player would not have stopped working and using these sub $500 laser disc players have nothing on my top of the line pioneer 990 digital series player.

[Bondiablo]

I found out that my laser discs look a heck of a lot better if I record them directly to my Panasonic DMR-E20

Yeah but are you recording on DVD RAM? Going from laser disc to DVD RAM is kinda like going from beta max to laser disc.

I suppose, strictly speaking, huffy is slightly better than DV. However, your source, capture card, even the cables you use to connect them, etc are also going to be factors. Considering how much larger the file size will have to be for it to be truly lossless and the problems involved in analog captures, even if you get it working properly you may still have trouble with dropped frames and audio video sync problems, you're probably better of with DV and an imperceptible loss of image quality, 12GB per hour file sizes and no sync problems.

[mojo]

Thanks for all the great responses, everyone!

So ultimately what it comes down to is am I better off sticking with the DV Bridge or plugging in this ATI Radeon 7500 AIW card so I can use Huffy?

I'm reading that a lot of people have had hassles with AIW cards. To get an analog capture that looks better than the DV route, what am I going to have to spend, and is it worth it?

[vhelp]

mojo,

I'm reading that a lot of people have had hassles with AIW cards. To get an analog capture that looks better than the DV route, what am I going to have to spend, and is it worth it?

About $59 bucks

I have the WTVGO card, that seems pretty good for most of my needs.
I find that for the and framedrop'less capturing, you can't use codecs.
Rather, just capture uncompressed. But, in order to do that, for say, an
1 hours capture, you'd need to have 60+ gigs of hd space. But, this is
hard-core stuff. If you are cerious about total quality, and must have
that last bit of ounce of quality, then bump DV asside for now, and go for
the WTVGO card, get any bugs worked out (like I did) and finally, begin
your ultimate LD capture projects - a well worth endeavor IMO. But,
that's my opinion.

On the other hand, it could all be your LD player. I have a suggestion
though. But, let me ask you a question first. Did you D/L the two divX
samples that flaninacupboard - seg28d.avi and seg22D.avi
had provider at this thread on 2nd page:
Star Wars Conversions-Hit me with your best shot...

If his ftp is worked out, try: --> ftp://192.168.0.2/ for the 2 files.

If not, and samples are still there, just D/L them and have a look at them
to see (gauge) your process quality. Of course, to best gauge yours vs.
his, set a 640x298, and crop. Actually, he went into details.. and do the same. Again, it's a divX sample, and probabably not
the best quality being that it was for demo purposes, but the quality was
pretty good when I played them in BSplayer. This app will widescreen
some samples pretty well, ie, those two 640x298 divX samples.

Remember, you just wanna gauge your level of quality.

-vhelp

[mojo]

Hey Vhelp

Thanks for that link! None of the addresses you gave me seemed to work for the files but the discussion is a big help.

I did an test earlier tonight and made a big discovery - the Dazzle DV Hollywood bridge SUCKS!! At least the 2 that I have tried...

Just for fun, I took a Sony DV camcorder and recorded the laser disc to a DV tape inside that. I then fed the camera via firewire into my computer and capped the same footage that I had capped through the Dazzle. I compared a frame from each capture and, much to my surprise, the image from the camcorder looked MUCH BETTER. The Dazzle image had it's contrast pumped up way too high, and the artifacting was highly apparent. Edges were ragged and the entire thing looked like a poorly compressed JPG. On a scale of 1 to 10, I'd say the camcorder footage was an 8 and the Dazzle was a 5.

I then did an MPG compression test on the camcorder cap and WOW it's so much better.

I'm going to return the Dazzle for another one in the hopes that it's a faulty unit, but I have a feeling I'll get the same results.

Perhaps other firewire capture cards offer better image quality?

[ronnylov]

Perhaps it is possible to use your camcorder directly as a DV bridge. Instead of recording to tape, record it in real time with your . But it may not be possible with direct pass through on your camcorder but it might be worth a try.

[yg1968]

Most Sony Digital camcorders have a passthrough feature. All you have to do is enable DV out on the camcorder's menu. In other words, if you use the passthrough feature, you skip the intermediary step of taping on a dv tape. You are essentially passing through the analog signal directly to your computer in a DV format.

[Bondiablo]

- the Dazzle DV Hollywood bridge SUCKS!!

That's pretty much the consensus. Either get a DV cam that you can pass thru or a Canopus ADVC-100. Here are some samples I just did with the Canopus, Scenalyzer Live and TMPGEnc Plus 2.56.

These are from the rooftop shoot out in The Matrix. I did some of the Morpheus Neo fight too but don't have room to put both up and thought since this scene has a lot of sky as background if there were any problems it would really show up there.

1 second DV capture 4MB:
http://home.att.net/~bonfiles/human.avi

12 second MPEG2 at Project Wizard default DVD NTSC VBR 3000 settings 5MB:
http://home.att.net/~bonfiles/human.mpg

[yg1968]

vhelp,

Here is an article on DV codecs that might explain your colour washout problem (I don't claim to understand it...):

http://www.sonyusadvcam.com/content/article_39.shtml

[Peka]

Just for fun, I took a Sony DV camcorder and recorded the laser disc to a DV tape inside that. I then fed the camera via firewire into my computer and capped the same footage that I had capped through the Dazzle. I compared a frame from each capture and, much to my surprise, the image from the camcorder looked MUCH BETTER...
...Perhaps other firewire capture cards offer better image quality?

It's not the firewire card that has given you the better image quality. Firewire is just a transport mechanism - it just allows the transfer of the DV from the camcorder. ie, like a file copy. In the test you did above, the DV camera is your capture device. Since all a DV camera does is basically convert analogue video to DV and write it to tape, you should expect it to do a much better job than a $100 capture card, which is going to rely on the rest of your PC hardware/software to get decent results.

________________
Good drugs -

[mojo]

Peka:

True, firewire is just a means of import/outport. However, I was referring to firewire CAPTURE cards, which I assume use a hardware DV codec to translate the incoming video.

Clearly the schemes used by different DV devices to translate video vary wildly in quality. The difference between the Dazzle box and the Sony camcorder was astounding.

Does anyone know if ALL Sony products would give the same quality of encoding? Are the Canopus products just as good, or better? I'd bet my wasted $300 on the Dazzle DV Bridge that the $300 Canopus does a much better job.

I'd love to see the same test frame captured by different DV devices!

Is there someplace here I can post the frame that shows the difference between the Dazzle and Sony?

[vhelp]

mojo,

I've ben recommending the ADVC-100 to others here, even though I didn't
have it, because I was basing it on the user reviews/comments, but I
finally decided to take the plung and pick one up.

First, I'd like to say, I was wrong about the DV color issue in 1 sense.
But, mojo.. ..
as you said, different manf. vs. dif. makes and modles, and devices too..
all make a difference. My Canon ZR-10, when I firewire it through it's
DV, the quality is auwful. But, if I hook up A/V to it's S-Video, and play
the miniDV tapes, and catpure it w/ an analog capture card, the quality
is outstanding.
Now, after messing around with the ADVC-100, and comparing it to my
Canon ZR-10 using Firewire, I can say w/out a doubt, that the ADVC-100
wins hands-down, AND, matches my analog capture cards quality
as well**. So, I'm surprised, AND excited too. Oh, forgot ta mention, NO
FRAMEDROPS !!!

I'm still messing around with the ADVC-100, but so far, I am am now
please with it. I like it very much, he, he... It makes me VERY HAPPY, he, he...
I have so many light-bulbs going ON in my head, I can't stop.. ..

Well, you get my enjoyment. But, I still stand by my statement, that
nothing (not even DV) will beet an analog capture cards' quality.
But, I must say, that the ADVC-100, and in some of the tests that I've
done so far, have proven worthy. I mean, I can't tell the difference between
my analog captures (**well, almost) It's that good. Boy, am I glad I picked
one up. So far, I haven't regreted it, though I did have some problems
for a little while, when I first got it.

ADVC-100 ?? ..VHELP DEFFINATELY recommends it. And, for your LD
projects, scrap the Bridge and Haul in for the ADVC-100 !!!!!

-vhelp

[vitualis]

Just some clarification:

Huffy can definitely be lossless in the sense of the word. That is, if you encode digital video with it, you can decode it and get bit for bit back.

If you use Huffy to encode the video to YUY2 then there is a slight loss in the RGB to YUY conversion.

In terms of capturing video from an analogue source (i.e., if you connecting a composite video cable or S-Video cable into the video-in on your PC), the concept of "lossless" doesn't really exist. There will be a step of analogue to digital conversion (done by your capture card) and how good this looks will depend on your card. A-D conversion will always be "lossless" in a sense as the only way to get "all" the data is to use an infinite sampling rate. Obviously, this is not the case and we settle for something where can can't notice the difference from the original.

However, the "loss" due to the A-D conversion has nothing to do with HuffyUV itself. For the raw digital data coming out of the capture card, Huffy can encode this losslessly.

Regards.

[vitualis]

But, I still stand by my statement, that
nothing (not even DV) will beet an analog capture cards' quality.

To carry on from my post just previously, this will depend on the quality of the video analogue to digital conversion unit. Obviously, the A-D conversion on a miniDV camcorder occurs INSIDE the camcorder (which then compresses the digital video into the DV format and writes it on the digital tape).

With an analogue capture card, you are inputing analogue video and use the A-D converter on that card to do the job.

I wouldn't be surprised if the A-D unit on a high end analogue capture card was superior to that in a miniDV camcorder. It's a matter of price. However, I think that it is fair to say that your average miniDV camcorder may have a superior A-D unit to say a TV capture card...

Regards.

[sch_m5]

ADVC-100 ?? ..VHELP DEFFINATELY recommends it. And, for your LD
projects, scrap the Bridge and Haul in for the ADVC-100 !!!!!

... and use Canopus dv codec (playback is free). Most likely, you will not notice any differences when you just play DV avi in ,let's say, media player (maybe small differences in colors). But if your final format is vcd/cvd/svcd or divx (~1.5-2kbps), the difference is there.

Note: to playback DV avi using canopus dv, avi is to be converted to canopus avi format (also free D/L from canopus web site). The time spent on coversion to canopus avi format (~ 35 minutes per hour footage on my 845g 2.1G comp) pays for itself.

[mojo]

Vitualis:

I would LOVE to cap these laser discs with Huffy and avoid DV compression - but the problem is I don't have an analog capture card, and from what I've read, a good one that provides perfect A-D is very expensive. They've been saying at least a grand!

As far as I know, Virtual Dub won't recognize a DV device to capture from, and even if it did (as I think you mentioned) the incoming signal would most likely be compressed DV anyway.

To be honest, the quality I'm getting using the Sony camcorder as a DV passthrough is very, very good. And from what Vhelp (and others) say, the Canopus ADVC-100 is at least as good, probably better. And at $300, it's a fair price and a lot better than the Dazzle.

The thing is, for $450 you can buy a cheap DV camera now! For the extra $150, isn't it worth getting a whole camera? I'm sure every manufacturer uses the same DV codec in all their products, regardless of price (the higher prices ones have better lenses, 3 chips, etc).

Doesn anyone know what sort of DV codec Canon, JVC and other non-Sony camcorders use?

[yg1968]

vhelp,

I haven't compared the two but I would not be surprised that a Sony miniDV camcorder does a better job of converting analog to digital than your Canon. Like I said I have never noticed colour washout with my Sony TRV-25. I have decided to use the Sony DV codec in order to avoid any codec issues but I can't say that I have noticed any problems with other DV codecs. Perhaps that I am not as difficult as others. Who knows?

[energy80s]

I have just recently bought a Canon DV camcorder for use (among other things) as an analogue to digital converter. The quality is excellent. I have an ATI AIW card and have been using that for over a year and I can tell you that the difference between ATI caps and DV caps is very noticable. Even the raw picture display on the ATI is very soft / artifacty.

The best way to cap digital TV is to use a D-Sat or DTT card and grab the raw MPEG2 data stream out of the air! Failing this, a DV cap is the next best thing. Analogue caps trail a poor third in my book (and I have the evidence to prove it!!)

Also, I have never experienced the "colour washout" problem with ANY DV camcorder firewire capture/transfer. I wonder is this an NTSC problem?

[yg1968]

I don't think it's an NTSC problem. My Sony TRV-25 is NTSC and I haven't experienced this problem. It is probably just a problem that is specific to vhelp (defective hardware or software issue, etc.).

[digiflix]

For what it's worth, I have been capturing my old Laser Disks with a Pioneer CLD-1090 Video Disc Player. Video connection is through S-Video into a Matrox RT2500 Capture Card. Capture Software is Adobe Premiere (V6.02). Captures are flawless and the uncompressed end-product is perfect (Pink Floyd Delicate Sound of Thunder rocks even without THX!). This way I can really customize things. I also have a separate satelite feed for all those Sopranos episodes. The real pain is exporting to MPEG format for burning to DVD. A 2 hour video can take +8 hours to encode (depending on what the source is etc...). I'd like to be able to capture right to MPEG and burn to DVD. Is there a way? (I digress).

I have done some pass through testing with a DV Cam, and it worked great - somewhat cludgy though - in terms of using the videocam as a player... I saw no real diference in quality. So, for me, it became an issue of practicality.

I'd appreciate any comments on capture > encode > DVD burning...

Oh yes, I have tried Virtual Dub, but cannot get the capture driver to work "No Capture Driver Available". Any comments on that?

[energy80s]

Oh yes, I have tried Virtual Dub, but cannot get the capture driver to work "No Capture Driver Available". Any comments on that?

I have always got the same response! Even with the so-called "wrapper". Virtual Dub is not a great tool in my opinion, as it seems to cause more problems that it fixes.