Tested Capture Cards [was:" DVD quality capture with what capture card ? "]

[^^-+I4004+-^^]

Moderator>I think we should open a TEST thread or something were users with a few capture cards can post their experiences.

so here we go_

[just for start,as this thread continues the thoughts of " DVD quality capture with what capture card ? " i would like to say that there's no way that vhs(or any other analog rip) will look exactly like DVD's_why?because dvd production process is digital from the start(for instance look for the following keywords on the net "philips spirit data cine" to get idea how one transfers film to another media),& there's no noise introduced..on the other hand every analog format introduces noise and digital formats don't like the noise!-so,you need filtering...but filtering doesn't work so well in hi-motion scenes(smoothing & dnr methods all work on same principles:they copy non-moving pixels_good for static images,but also destroys the motion&yes, i tried flaxen,i tried dnr&snr,still temp.smoother&smartsmotherhiQ give the best results_for real noisy video_vhs tends to be that way!)]

my experiences include following cards:they are sorted by quality_topmost place-top quality.......

1_ELSA ERAZOR IIIPRO VIDEO[tnt2pro+itt-micronas vpx3225 capture chip]
{this card gave me the best quality picture i've ever seen on a capturing device->eqc(elsaquickcodec) is nicely optimized for the hardware& outputs by far the sharpest image.......supports a lot of resolutions(no custom resolutions for vdub though_i think that bt8x8 is amongst the few that can do this),has SHARPNESS control(i think no asus card has this.....bt8x8 has it in form of dscaler dll's & vdub's "bt8x8 tweaker" function),2 composit,1 svhs input....}

*** UPDATE->note;i've tested (&used) it mainly on lower res(352x288),at that res. it didn't drop ***

2_ELSA ERAZOR III VIDEO[tnt2+same capture chip]
{this one has the same image quality as my winner,but tends to drop frames a bit.....at worst it drops 0,7-0,8%}_i'm using this one now,as the first one was destroyed by JNC(noname..hehe) ATX PS

*** UPDATE->i've tested it on hi-res lil' more(cause it turned out[in reply to this very post] that some (if not all?) Erazors have trouble keepin stable framerate (ie. they drop frames)) & it seems this is true-Erazor3video IS NOT good solution for hi-res capturing_vdub(or avi_io) don't show drop out numbers correctly ( they don't detect drops with elsacodec),but when you watch recorded video,it skips,it breaks _sometimes more,sometimes less,but anyhow NOT USEABLE material is what you'll get on hi-res_....for lo-res mentioned drop rates (0,7-0,8% across whole video) still apply..... i would also like to APOLOGIZE to anyone i've misleaded ( unintentionaly ) into believing that these are the hi-res capturing devices of some class!(-although i've mentioned that i use Bt card for hi-res capturing &that i use elsa for vhs-rips.....i still use elsa for lo-res capture because it gives much better picture than Bt....) ***

3_PIXELVIEW BT878 Capture Card(no tuner)
{on lesser resolutions it's not that good_image is not sharp( i think if you look at bt produced images long enough you'll soon need glasses to read..hehe)_even if you use 352x576&vertical reduction trick for vdub....
on higher resolutions though i'm experiencing pretty good results:it doesn't drop & picture can be sharpened a bit with bt8x8 tweaker..(if needed)
this is my second capture device:i use it for tv-rips,since ELSA cannot be beaten in lower resolutions(ie. i use elsa for vhs-rips:pal 352x288)_,but elsa NON tnt2pro version (i currently have) drops a bit more on hi-res}


[4]_ASUS COMBO BOARDS[3800 series,6800 series,philips capture chips]
{these aren't my direct experiences,but lookin at other people's work:it seems that asus card owners do not have anything to be proud of:picture(i've seen 704x576 & vcd's made from these cards:my friend used to own both 3800&higher models) is not so sharp&detailed (for example tv-station logo's is a good example:if thats blurred you can be sure that whole image is too_
asus' logo's are awfull!also picture is without details)
seemed to me that better results are acieved on lower resolutions(VCD's were OK.......)_
just like philips' chip is some fawlty version of bt8x8 chip....
we've exchanged quite a few vidcaps&videos and he agreed that elsa looks much better by any merit.....}


5_ATI AIW128 [32MB,RageTheater capture chip]
{this combo board gave me the worse quality picture i've seen(ok,i've seen ati tv cards with bt829 that were even worse!)_picture is pretty blurred,soft,unclear,unsharp!turning on the "noise reduction" increases the problem.........this is good for someone who wants to do encoding onthefly(as it drops very little even when recording directly to divx),& doesn't mind lesser quality that much....}



[?]_CANOPUS ADVC100 [hardware DV codec]
{again,not my tests,but i'll try to show some of the problems that people on NG[hr.comp.video] (i participate in_its Croatian group,so others don't bother_already enough Croats here..huh) experience:
problem1:too bright&contrasted picture_as we all now this flaw is hard to correct in post-processing_decreasing br/contr levels doesn't help much
problem2:doesn't copy vhs so good->if there's little more noise in the signal
(older vhs') there's no picture:screen goes black(!)
(so now guy who is testing it looks for TBCorrector.......TBC for vhs ripping?????)
other than this(but i wonder whats left...,what good is it for?) guy says quality is ok_i think he can guess that picture(720x576 is only res.,round 5MB/s for DV stream)would be ok if it were less contrasted & if it were existant at all(for older vhs rips).....hehehehehe_so i guess i could say that canopus' DV codec is a pile of trash,as we vhs capturers are concerned(ie. what about those old&precious vhs tapes that beg to be converted to more time-resistant format)
.......price here in croatia is round 350-400$}

& for the end i'll just mention my recent experience in coding the 704x576(tv-rip)captured by Bt878:filtered&deinterlaced,at 1400kbit/s(divx311),resized to 576x480(not necessary in that order),video looks stunning!still scenes are completely without a noise(just like on dvd) but some noise occurs on hi-motion(i've talked about this at the beginning)_material was a documentary with quite a few hi-noise/bw images...
without smoothing material would be awfull & i would,t be able to acieve output bitrate of less than 1200kbit/s(i set divx to 1400,but thanks to smoothing bitrate decreased...)_i was also surprised to see how good were scenes with moving camera:there were no blocks at all!
i'll quote "undercut"(involved in nandub development & creator of nandub_onepass) here:
"and generally higher resolution video is more compressable. (less detail per pixel) "

this proved correct indeed!
so,it wasn't dvd,but sure was close....

[Ookami]

Grin. Thanks.

As for BT chip based cards giving a better result than Asus 3800 cards :D . Muahahaha...

> just like philips' chip is some fawlty version of bt8x8 chip....

Now your sounding like one of those video "pros" from certain croatian computer mags :rolleyes: .

I will not go into detail (VCD as comparison?), but I can send you some captures made by myself and then you could send me some stunning BT captures ;) .

As for the Erasor III cards, Genesis promised 200 DM, to anyone who can show a method wich can be used to capture with these cards without problems. No one I know managed to use these cards without problems. All agree that the picture quality is (was) stunning. Drop frames a bit?! This must be an understatement or your the first who got this card working! Please, open a new thread and earn some money :) .

-filters

The problem is that there are so many filters and combinations possible (not to mention the harware, software and source combinations), that it is simply impossible to test it all...

Even people who are doing this for years cannot possibly say "this is it, now I cannot be made better"... A year ago Hawk's guide was unbeaten, now you have dozens of new filters, you have Avisynth, you have averaging and so on.

I agree with your statement about DVD quality from VHS, but I would say near-DVD qualiyt is possible (I did it for myself).

BTW, fISH's denoiser filter is very good (especially as the first in the filter chain). Sadly, only a few people know of it (use it).

> already enough Croats here

Ma da?!

Hehehe...

All the best and thanks for your great posting.

Cheers,

Mijo.

[Chyron]

Sorry for offtopic question, but where can i get fiSH denoiser? Link on Graft's page is dead. Search gave nothing.
If possible, can you mail it me to gabrichyron@yahoo.com?

[BaronVlad]

Think you wont get it there, but the new link for Donald Graft is this one:
http://shelob.mordor.net/dgraft/
For the denoise filter you should have a look at Mijos Posting, cause tHE fISHs site seems to be down at least at the moment:
http://forum.doom9.org/showthread.php?s=&threadid=32521

[Koepi]

Well, i changed my filter chain ... away from vdub!
Find convolution3d in the avisynth forum-it's to be beaten :)

Now we don't need to change color space anymore and can do all the stuff in avisynth (->even a speed improvement).

Just wanted to share...

Best regards,
Koepi

[^^-+I4004+-^^]

O>As for BT chip based cards giving a better result than Asus 3800 cards . Muahahaha...


_i've said something funny?i've received divx's(made from asus),unfiltered,with fairly high bitrate_they looked lousy
also vidcaps of uncompressed video...not much difference



> just like philips' chip is some fawlty version of bt8x8 chip....

O>Now your sounding like one of those video "pros" from certain croatian computer mags

_when we're at chips can i guess that you haven't got the slightest idea what "luma-peaking" is?&can you explain why asus also doesn't know that term?(hehehe)

O>I will not go into detail (VCD as comparison?), but I can send you some captures made by myself and then you could send me some stunning BT captures


_wouldn't it be better if we post those examples HERE?so people can judge for themselves....
(i'll capture some,or find old ones:accepted are the analog vidcaps(TV,VHS,SATV) NO digital sources(DVD,digital cameras...etc)all picture controls to mid-positions)

.

O>As for the Erasor III cards, Genesis promised 200 DM, to anyone who can show a method wich can be used to capture with these cards without problems. No one I know managed to use these cards without problems. All agree that the picture quality is (was) stunning. Drop frames a bit?! This must be an understatement or your the first who got this card working!


_on IntelBX chipset(&now I815) installed drivers from cd-rom(they're better than web dloaded versions)& everything works_if "Genesis" doesn't know how to install & configure hardware then it's his problem.(or if he has via chipset.hehe)..as i've said:PRO version didn't drop(although i didn't test it so much on hi-res with cel400&ata33 disk_i did post one video to hr.misc.binaries_704x576(real-time-deinterlaced) without drops),non pro version drops max 0.8%....(for instance i get 350drops on 65000 frames(0.5%)_65000frames is 43min....i say that's not much....)

i even have "victory erazor"(riva128&the same capture chip as in newer versions)&although it's not "fps monster" i managed to get 23fps with it
(worked better on bx chipset than with i815)_bare in mind that this is an OLD card)

but then again i like better picture& drop here&there than no drops & ugly picture all the way.....(seems like both(no drops&quality) is not possible!)





O>Even people who are doing this for years cannot possibly say "this is it, now I cannot be made better"... A year ago Hawk's guide was unbeaten, now you have dozens of new filters, you have Avisynth, you have averaging and so on.

_i've found good combination:how 'bout you?i did few tests,but i think vdub filters are better than asynth(and i can see why many asynth filter dll's include code from vdub filters....)

O>I agree with your statement about DVD quality from VHS, but I would say near-DVD qualiyt is possible (I did it for myself).

_i've said that it was close to dvd(though i didn't mention that part of the documentary i considered was a VHS-rip.(when i record something important i parallely record onto PC & in VHS so if PC capturing goes wrong(here it did) i have backup copy on vhs)...also turned out very nice!)

O>BTW, fISH's denoiser filter is very good (especially as the first in the filter chain). Sadly, only a few people know of it (use it).

_i doubt that(i doubt any bigger advancements in DNR field_dnr is too weak for my needs_vhs is often crappy to start with)
but i could try it:links to fish's web are broken& dgraft's page also doesn't work(lately often the case)_post it here(vdub filters are usually very small files)




O>All the best and thanks for your great posting.

Cheers,

_dobro Mijo.....


btw.beat dis!(btcaps.zip)

[bb]

I posted the denoise filter by tHE fISH here (in June!):

http://forum.doom9.org/showthread.php?s=&threadid=25317&highlight=denoise+1.0

It's a good filter, but I second Koepi who pointed out that the new AviSynth filter "Convolution3D" by vlad59 is the best choice for denoising nowadays. I did a few TV captures and encoded using Convolution3D. The results were outstanding: There's no more noise, the compressibility increases much (depending on your settings). I used the recommended settings for the full-1-matrix.

If you know of a better filter, then tell me. Although I don't believe you can find one...

bb

[Ookami]

First, I've already stated that I wont participate in those "x vs. y" discussions anymore. Use what you like... Second, this thread is, somehow, getting a "my cable is longer than your cable" taste, wich I dislike. And third, I'm not on a mission to convert non-Asus 3800 believers to Asus 3800 believers (Somehow, I get the feeling that people get this feeling). For the n-th time, I do not care what you use, hell, there are people using Divx fast motion, why not? But I will reply to this thread, so I can defend my honor, I mean, GPU, my captures that I mostly make for myself,I mean :D . See, how funny this is?!

Many months ago (I the year 2000), on a discussion on the german Ultimateboard, I backed someone up who stated that it is stupid to capture at anything higher than 352x288 from VHS, as it cannot be better and it wouldn't matter anyhow (well, or something like that I cannot remember exactly what he wrote)... So, Genesis ( :) ), simply wrote: "test it for yourself". And he posted some facts along, but the line "Test it for yourself" is something that is always repeated on such forums and something I like very much. Now, after I read Kika's explanations I always laugh when I read things like "VHS has a
resolution of 320x240".

When I started (all readers: AAAAAARGH! Not that story, again!), in the DVD backup scene, the test and try idea was standard. Why didn't I want to make a "normal" FAQ (when I did the first one, now obsolete since BaronVlad and I made a better one)? Because, a normal FAQ is always subjective, it says only what the authors believe (think), therefore I liked the idea to link to discussions where people with different opinions discussed and respectfully disagreed (or converted to the other, dark, side "Young Jedi, feel the rage"). What I want to say with this boring and long winded thought (not sure, that I'm sure what I want to say, anyway) is that everyone has it's own opinion and if you disagree then fine. In almost every thread where I said something positive about the Asus 3800 (oh, back on topic!) I posted the cons of it (only a few supported resolutions, incompatible to many software as using a non-BT chip etc.), so I try not to sound like a employee from Asus (and after I bought a Asus MB with a VIA chipset I doubt that I can be someone who will talk positive about them :) )... Like (I think it was Macdaddy), wrote, anything that is incompatible with DScaler is not interesting. So there, different opinions. So, now we can all hold our hands and sing "We are the world" (or a song from OK Computer).


> i've said something funny?i've received divx's(made from asus),unfiltered,with fairly high bitrate_they looked lousy
also vidcaps of uncompressed video...not much difference

Yeah? Well, then I must say, Asus sucks! Don't buy it and don't use it! All those Asus 3800 users suck and they can capture directly with Divx as there is no much difference to lossless captures anyway! Not to mention that people who are wasting their time with square and non-square pixels, field swaps, dropped frames, aspect ratios etc. should get a life and start using their VCRs...

>_when we're at chips can i guess that you haven't got the slightest idea what "luma-peaking" is?&can you explain why asus also doesn't know that term?(hehehe)

As a matter of fact, I had a idea, but didn't knew it until I looked that term up. Strange, my captures are not getting better after knowing the term, can you explain that ;) ?

>O>I will not go into detail (VCD as comparison?), but I can send you some captures made by myself and then you could send me some stunning BT captures


> _wouldn't it be better if we post those examples HERE?so people can judge for themselves....(i'll capture some,or find old ones:accepted are the analog vidcaps(TV,VHS,SATV) NO digital sources(DVD,digital cameras...etc)all picture controls to mid-positions)

No, that wouldn't. Because:

a) Pictures are not good for video quality comparisons (ever saw a audio comparison made with separate (and maybe even lossy compressed) tones?), for instance Doom9 (Yay!) does all his comparisons looking at encoded videos

b) The only objective testing would be if you would test cards under the SAME circumstances, like same PC, same OS, same source etc. So if your pictures would be better that could mean a many things like: You use a lower JPEG compression :D , you have a better source, your cables are better, you have no interferences from other devices and so on. There is no guarantee that it would mean that the quality of your card is better, and another problem with these cards is that the quality can be much worse wich different drivers, software, OS etc. So, if I upload a crappy capture made with my Asus 3800 and my equipment, does that mean that all Asus 3800 cards produce, always, such a pictures or that, for example, derKarl couldn't do it better? When you buy a dedicated capture card, you are mostly spared from many of these problems (or possible problems).

c) All below Huffyuv is unacceptable (ever tried to upload a 30s Huffyuv clip with dial up ;-) ?), BTW, can anyone download anything from http://www.ookami.iwebland.com/ ? $#/%&/)# If you use compressed videos and the results suck that can also mean that you don't know how to use the used encoder properly
.
.
.
x) See my disclaimer in one of my earlier postings

LOL, I could've said "no" and it would be the same thing.

However, as you live in Croatia, I am willing to send you some Huffyuv; Divx ;-) and SVCD samples. Then, after you see the crappy quality (BTW, do you know the croatian word for special insulated antenna cable ?), you can come here and write "Asus 3800 sucks", "Ookami doesn't knows what he's talking about". BTW, I hereby state that I don't have the slightest idea what I'm talking about, and that I've, extensively, only used my Asus 3800 for captures... And anyone who is on this forum long enough knows already, that I have no clue, but as I'm a mod no one dares to write that (aaand strike out!)... Muhahahaha...

Regardless the quality of my samples, the a,b,c - x facts don't change a bit.

>_on IntelBX chipset(&now I815) installed drivers from cd-rom(they're better than web dloaded versions)& everything works_if "Genesis" doesn't know how to install & configure hardware then it's his problem.(or if he has via chipset.hehe)..as i've said:PRO version didn't drop(although i didn't test it so much on hi-res with cel400&ata33 disk_i did post one video to hr.misc.binaries_704x576(real-time-deinterlaced) without drops),non pro version drops max 0.8%....(for instance i get 350drops on 65000 frames(0.5%)_65000frames is 43min....i say that's not much....)

Don't remember what card exactly they used, but I know that at least 5-6 people used a specific Elsa Erazor card and they could get it work reliable (!). It would work sometimes, but mostly not. And mind you most of these people were professional PC admins, so I think they know how to set up a PC :) . And, that was about 2 years ago, so it was probably a BX chipset or similar. If I can get Genesis's email I will ask him if he's still interested in your explanation (altough you've seem to work with the card without any change and tweaks, so I can't see what you could explain to him in this thing).

-Just wanted to stop re-reading this post and save the .txt file as I saw this "real-time-deinterlaced"! RT deinterlacing

DURING capturing with a Celeron 400? Er, you are aware of what you are writing? Please, post your capture method in a new thread if this is correct (and please use a new thread as we, already, gone way off topic here).

>but then again i like better picture& drop here&there than no drops & ugly picture all the way.....(seems like both(no drops&quality) is not possible!)

No drops? A quote from the new FAQ:

"First lets clear some misconceptions that many users have. Dropped frames are not a disease that needs to be cured. It is simply a solution to maintain the A/V sync by dropping or doubling a frame. The catch is, when you (oops, here should be a "have") dropped frames that are caused by other things than the A/V sync algorithm."

Since when have dropped frames something to do with the picture quality itself? Or do you mean you get more dropped frames when using full PAL (as opposed to low res caps)? If so, you should optimize your system. But, you will probably find nothing new in our FAQ here *nudge, nudge*.

>_i've found good combination:how 'bout you?i did few tests,but i think vdub filters are better than asynth(and i can see why many asynth filter dll's include code from vdub filters....)

Oh. Now you're coming with the "I know it all" attitude?! (...deine Springerstiefel...) "I did a few tests", and in the same line you put a pseudo-provocative question like "How 'bout you?". Tsk,tsk,tsk...

I have done a bit more than a few testings (I'm here quite some time and did my share of captures *puts away his note where he wrote how long his cable is*), but I am not even near to persons like Luke, fISH, derKarl, Kika, Genesis etc. So, if you want someone you can proove your knowledge on capturing,I'm too dumb and unexperienced for that. And I mean that, seriously.

And to imply you have a perfect combination for all things can simply not be true and correct. You have, probably, found something that fits your needs, but person Z maybe doesn't have the time for filters, or isn't satisfied with the quality by your method.

And not to mention (oops, did it anyway), methods that use different approaches, like averaging etc. Wotef did a short explanation in one of the recent threads, and Poptones did the long winded version :) ...

O>I agree with your statement about DVD quality from VHS, but I would say near-DVD qualiyt is possible (I did it for myself).

>_i've said that it was close to dvd(though i didn't mention that part of the documentary i considered was a VHS-rip.(when i record something important i parallely record onto PC & in VHS so if PC capturing goes wrong(here it did) i have backup copy on vhs)...also turned out very nice!)

Like I said, I agree with you. BTW, I did the same thing, only that I use SVHS when I have a SVHS tape ready :) (that's showing off and yeah, my dad's stronger than yours!).

O>BTW, fISH's denoiser filter is very good (especially as the first in the filter chain). Sadly, only a few people know of it (use it).

>_i doubt that(i doubt any bigger advancements in DNR field_dnr is too weak for my needs_vhs is often crappy to start with)
but i could try it:links to fish's web are broken& dgraft's page also doesn't work(lately often the case)_post it here(vdub filters are usually very small files)

Why, before you even tried? This filter is mostly used only as the first in a filter chain (for Divx (as you seem to use that) e.g. Denoiser, Deinterlacer of your choise,Null transform,rmPAL, HQ smoother, precise bicubic; no this is no perfect combo, either) Too crappy to start with but a few lines above you wrote "turned out very nice"? Long story short, many people (including myself) misused the Deflicker filter for normal VHS captures. So, fISH took this approach to write the Denoiser filter (hope that's the correct name).

@Koepi

See, that's what I mean... Were discussing here and Koepi uses Avisynth, probably all YUVy, and dreams about

"convolution3d"... Koepi, koenntest Du mir bitte ein Beispiel Skript und ein paar Zeilen an meine E-mail Adresse senden?

"Teste es selber?" Ja, mein Gott, denkst Du das ich an das glaube was ich von mir gebe? Du lernst schnell Grashuepfer! Im Ernst, wuerde meine Tests erleichtern bzw. schneller machen wenn Du mir unter die Arme greifst. To all the people who don't speak german: I just proposed to try the "strike" button on all the users that disagree with me in the next postings... We will not press it more than three times in a row!

@bb

I think I also uploaded it somewhere, here. But because of the pruning, I'm mostly too lazy to do a forum search for such things...
I will try the mentioned filter, ASAP.

Summary:

I've enjoyed writing this and putting my weird humor into this posting (doubt that anyone will laugh, if someone did, we could write to Fox together and demand a "Married with children" DVD release). I even laughed quite some times when writing a new line or putting new dots (for those that didn't noticed the dots, they look like this -> ... <-) and weird sentences together. Maybe I should translate my sensational "Capture Reportage" and... Whatever, I wanted to reply as short as possible and now I've broke my record, I think.

Nice chatting with you, but no more replies from me, here (unless it's a mod thing). And welcome to this board!Hope you'll stay here and be a long-time member! Now I can (er, must!) go back to my learning...

Cheers,

Mijo "Spelling is for wimps!" Rebic.


>Cheers,

>_dobro Mijo.....

:)

>btw.beat dis!(btcaps.zip)

Let me guess, you've uploaded a few bitmaps to show how good the BT chips do their capturing?!

"Prestar sam da podem s tobom, ali poslusaj ovaj moj savjet: "...""

[Swede]

Ok guys, time to cool down a bit. AND stay on topic. AND keep the length of your posts to something more readable.

[Darksoul71]

Hi !

LOL ! I never had so much fun reading a thread like this since month.

@^^-+I4004+-^^:
1) Learn to make your postings more readable. Make more line feeds and paragraphs. Thiswholewritteninablock-thing hurts my eyes while reading

2)I fully agree to Ookami:
It´s impossible to do serious testing unless you test ALL cards in the SAME system with the SAME surrouning (same source, cables, etc). Everything else will make it impossible to judge quality differences. Using different surroundings will let you only judge BIG differences (like a strong smoother, chroma artefacts, etc) but not the small and nifty differences (like lack of details). So let´s forget about this "let´s compare our cable length" stuff. Comparing encoded samples is even worse as there are billions of filter combinations.

3) I´m not going to this "my cable is longer than yours discussion" but I have a few things to say:

On Asus: I own an Asus 6600 Deluxe which uses the same ADC (Philips 7113) that my much more expensive (once it costed 500€) hardware MJPEG card has. The capture quality kicks ass and can compete with much more expensive cards like DC10, DC30, etc.
Don´t judge on VCD samples, etc.

On BT 8x8 based cards: LOL, a BT8x8 based card beats an Asus quality wise ? Get real....

On Canopus ADVC-100: I own one myself. Just a few words: Only gets "black" if you have a macrovision protected tape. No problems with contrast or brightness (at least not at the last 4 VHS tapes I captured). VERY good quality even with VHS. Even small res DivX look much sharper compared to my previous MJPEG captures.

Believe me: With a good source an ADVC-100 is able to get "near DVD quality". A friend of mine has also an ADVC-100 and does captures via S-VHS from D-Box (a german pay-TV DVB receiver). He did a few MiniDVD & SVCD encodes with CCE in 3 Pass mode. They all kicked a.. and looked stunning near DVD. Of course every analog signal has noise in it but from a perfect source it´s very low.

So, this is enough on this "discussion". Now you can overthrow every of my opinions.

-D$

[Owen]

Hummmm.
Cards I have used.

Miro/Pinacle DC10 and DC30

Both outstanding quality due to high quality ADCs and good board design.

Several BT8xx based TV cards Pixelview, Dynalink etc.

These tend to give fuzzy pictures and suffer from noise pickup from the computer.
The fuzzy picture can be fixed by using the drivers from here.

http://btwincap.sourceforge.net/

These allow you to adjust sharpness.

The noise. Well every PC I've had from Cel450, PIII. 700,900,1g
all with BX chipsets and Athlon 1.33g with VIA chipset all had noise problems.
I now use a P4 2.1gig with SIS chipset and noise is gone completely.

I think this is due to the low power usage, lowering the Radio Frequency Interference and the high clock speed moving what is left out of band. (just my theory)

There is also a problem with Luma clipping with this chip.
If contrast is set to high/bright. Detail is lost in the near white or bright areas of the picture.

I find that reducing the contrast control to about 85% (or about 75% of default 100% value) is required to fix this.
The resulting picture is a little darker in the peak white areas that the original but retains the detail.

The best resolution for PAL capture is 720-576.
768-576 only increases noise pickup.

After all that, the BT8x8 can work very well.

What next.

The new Conexant chip TV cards from MSI, Pixelview and others looks really good with 10bit ADCs in place of the 8bit on the old BT8x8.
Cant wait to get my hands on one.

Has anyone tested one yet ????

Regards,

Owen

[thorazine]

Originally posted by Darksoul71

On BT 8x8 based cards: LOL, a BT8x8 based card beats an Asus quality wise ? Get real....


Isn't that the same "my asus is bigger than your conexant" attitude you talked about a few lines above?

Seriously, what are the technical advantages of one chip versus the other? I compared data sheets a while ago and found them to be quite similar (at least from the specs I understood). Are there any hidden quality gain from transfering video over the AGP bus instead of using the PCI bus?

Originally posted by Darksoul71
On Canopus ADVC-100: I own one myself. Just a few words: Only gets "black" if you have a macrovision protected tape. No problems with contrast or brightness (at least not at the last 4 VHS tapes I captured). VERY good quality even with VHS. Even small res DivX look much sharper compared to my previous MJPEG captures.


Regarding macrovision, I remember reading in rec.video.desktop that you can disable it on all canopus bridges if you hold the ON/OFF button for 30 seconds

(edit: now, I'm not sure if that's the right button, you should be able to find the right procedure archived in google)

[Darksoul71]

@thorazine:

Isn't that the same "my asus is bigger than your conexant" attitude you talked about a few lines above?


Yes, you´re right ! And I must excuse myself for "comparing cable length". I must be listening to too many people claiming that BT/Connexant chipsets provide good quality. I owned two BT 8x8 based TV cards. One Askey and one Haupauge WinTV. Both had a crappy quality compared to EVERY I owned and I owned/tested a lot of VIO graphic adapters: Elsa Victory Erazor, Voodoo³ 3500 TV, Erazor 3 Pro, Asus V3800 VTR, Asus V6600 Deluxe, MSI GeForce3 Ti200, etc.

I know what I see and unfortunatly I don´t have a clue why the BT 8x8 has such a low capture quality compare to other VIO solutions out there.


Are there any hidden quality gain from transfering video over the AGP bus instead of using the PCI bus?

This is no "hidden" gain but something to free more bandwith on PCI bus. Although I´ve been able to capture up to 10 MB/s using my WinTV and a SB Live on a KT266A based motherboard capturing via graphic adapter/AGP frees bandwith on PCI bus. More bandwith means less chances for dropped frames. At least this are my experiences concerning PCI capture cards.

Although this is OT:
Regarding macrovision, I remember reading in rec.video.desktop that you can disable it on all canopus bridges if you hold the ON/OFF button for 30 seconds

Hm, some guide mention a input select button but my ADVC only has a power button and you method didn´t work out. At least as I tried to capture from a Macrovision protected buy cassette the picture turned "black" (read: lowered contrast & brightness). Do you know what to do ? I didn´t find anything on google.

(edit: /me stupid. I found the button and it worked. I´ve successfully tested it with a MV protected tape. Thanks for the hint)
TIA,
D$

[^^-+I4004+-^^]

*Good to see thread moving again(after moderator's essays & my "quick encyclopaedias"...heheh)*

Let us talk:



>@^^-+I4004+-^^:
1) Learn to make your postings more readable. Make more line feeds and paragraphs. Thiswholewritteninablock-thing hurts my eyes while reading

_it's not me,it's notepad!(hehe)

>2)I fully agree to Ookami:
It´s impossible to do serious testing unless you test ALL cards in the SAME system with the SAME surrouning (same source, cables, etc).


_how many cables did you tried & what differences did you saw?
(i mean,really?tests include VHS VCR's tuner cabled to chinch(V-in) to cards....you used golden connectors,shielded cables,high% of copper?->can i bet that you don't even know how does shielding work?_every so often i'll get the urge to pinch you(or him) just a bit-when you try to sound too pro....don't take this too seriously,just bare in mind where you are& what are you into->this is NOT pro's stuff...this is to get better results encoding video at HOME...)

by looking at the picture (in some of the capturing formats huff,mjpeg..pitty though that huff wont play so it's nearly the same thing if i post pix..)you can say A LOT ...(off course knowing encoding technique)

this is my view on the subject:i KNOW my surrounding the best(same equipment is used ALWAYS,just the cards change.same vhs,same cables.....results are always different.....if i had the lab then i would posts test done with pro testing equipment & more detiled in technical manner_but then i wouldn't be on this forum,now would i?)


>3) I´m not going to this "my cable is longer than yours discussion" but I have a few things to say:

_someone misread stuff i wrote:i had few cards,i said my opinion-compared one to another.....i also saw results from some other people (unfiltered) & read some (advc100) on NG(& pretty reliable sources as it turned out_more later...)


>On Asus: I own an Asus 6600 Deluxe which uses the same ADC (Philips 7113) that my much more expensive (once it costed 500€) hardware MJPEG card has. The capture quality kicks ass and can compete with much more expensive cards like DC10, DC30, etc.
Don´t judge on VCD samples, etc.

_vcd was just one of the things(one of the videos.....)_also on much expensive models there are Bt8x8 as input chips(i think hauppauge & even some pinnacle cards had it.....) this alone doesn't say much on either philips or Bt...

DC10+ for example uses philips' chip:not same model,but question is how much different "a/d engines"(so to call them) they have (note:chip on dc10+ is about '95. "baujahr"...don't know if they changed it recently..)

btw. you're telling me your cable is longer???heheheh

[also:notice the asus' number is in parenthesis?this means: not my tests,but still i think of them worth mentioning]

perhaps i'll correct myself when Ookami sends me some samples...off course nothing will be deleted in my post:just updated! & maybe then no parenthesis]


>On BT 8x8 based cards: LOL, a BT8x8 based card beats an Asus quality wise ? Get real....

_post vidcaps with better picture quality (and in the same way captured) than i did & i'll agree....(picture on "arte" on my system has lil bit more noise..),no problem there at all!My source was analog SATV....

i'm judging chip's quality_not the tuner one's_i even don't have tuner on that card..it's pretty decent capturing_

(i'll say it again:ASUS DOESN'T HAVE SHARPNESS CONTROL_Bt can be configured so it DOES....have i lied?_i like a bit sharper image than softer:encoding will soften it enough)

[see,i'm paragraphing....

oups...hehe]

>Believe me: With a good source an ADVC-100 is able to get "near DVD quality". A friend of mine has also an ADVC-100 and does captures via S-VHS from D-Box (a german pay-TV DVB receiver). He did a few MiniDVD & SVCD encodes with CCE in 3 Pass mode. They all kicked a.. and looked stunning near DVD. Of course every analog signal has noise in it but from a perfect source it´s very low.


_i believe that!i believe also i mentioned older & noisier vhs:those that really need to be archived!they make problems!(& NOT in conjuction with particular VCR's brands!problems on more than one VCR!)

lately (in "hr.comp.video") Damir Hamberger (of "Hamberger" & Canopus Hr(Croatia)) publicly (on NG) declared that_(quote,translated from Croatian)
"If the VHS signal is not stable,DV compression brakes,as does MPEG"(end of quote)

so all agredd that in these cases TBC is needed.....
( take this in account when buying equipment:you'l need TBC too..no more,no less)

if the canopus dealer is truthful enough to have said that,i think we can all
agree.....

[i DON'T have ANY reason to question the honour of mr.Hamberger or the guy who was testing the ADVC at his home....i believe them completely..]

(the guy that started the tests,now backed off the canopus & uses Digital8(sony) camera to convert as it does have TBC built in_interesting solution i may add:if you need converter& the camera..)


>Now you can overthrow every of my opinions.

_you've said something?(heheh)

{ as i can see u used erazorIIIpro...what are your experiences? }




1>These tend to give fuzzy pictures and suffer from noise pickup from the computer.
The fuzzy picture can be fixed by using the drivers from here.

_i wasn't considering the use of built-in tuner!

1>http://btwincap.sourceforge.net/

These allow you to adjust sharpness.

_so do "dTV" dll's for vfw drivers!
many will still argue that vfw is a way to go with these cards(as opposed to wdm model)_myself,i haven't tried but am satisfied with vfw drivers...


1>I now use a P4 2.1gig with SIS chipset and noise is gone completely.

_interesting!however i would recommend some tuner completely away from the machine.....(if it's possible for you...)

1>I think this is due to the low power usage, lowering the Radio Frequency Interference and the high clock speed moving what is left out of band. (just my theory)

_not bad,not bad.....into account should be taken not just tuner shielding but also signal plains from tuner to chipset.....guess you're right.....

1>There is also a problem with Luma clipping with this chip.
If contrast is set to high/bright. Detail is lost in the near white or bright areas of the picture.


_as far as i know that's the case with all of capture devices:too much light/contrast=>less detail (same is with TV.....& everywhere else..)
no news there

1>I find that reducing the contrast control to about 85% (or about 75% of default 100% value) is required to fix this.
The resulting picture is a little darker in the peak white areas that the original but retains the detail.


_if you're talkin bout Bt,i use all controls to mid-positions & it's ok_
darker picture can be brightened on encoding(filters) or at playback(plb filters)...._no need to let the capture card do all the picture adjustments_just keep it lil less brighter (as with all cards)



1>After all that, the BT8x8 can work very well.

_you've said something bad?(heheheh)





1>The new Conexant chip TV cards from MSI, Pixelview and others looks really good with 10bit ADCs in place of the 8bit on the old BT8x8.
Cant wait to get my hands on one.

_i also saw some news on that_yeap it would be interesting to compare those two (do you know the price for MSI card?)





2>Seriously, what are the technical advantages of one chip versus the other? I compared data sheets a while ago and found them to be quite similar (at least from the specs I understood). Are there any hidden quality gain from transfering video over the AGP bus instead of using the PCI bus?


_thats not the problem.(agp is faster,but pci is enough with 133(as i remember)MB/s)..some cards have& some do not "luma peaking"(find that in datasheets) controls- this relates to sharpness of the image.....
off course,they don't use the same ADC's so quality cannot be the same...
i guess there could be found more stuff if we got into the sheets deeply enough




2>Regarding macrovision, I remember reading in rec.video.desktop that you can disable it on all canopus bridges if you hold the ON/OFF button for 30 seconds

_problem wasn't related to this:home made VHS tapes in question.....




wow,again monstruous reply!hope Swede will not take it as bad as we were more focused on the subject
[btw.Swede what happened to my examples in asynth forum?_someone feeled touched by it so kept it for themselves?hehehehe]

cheers

[Herske]

>Seriously, what are the technical advantages of one chip versus the other?

The biggest problems of bt8x8 chips (up to 878 -> that's what I've used) are those:

1. Bad y/c separation
2. Most cards with bt8x8 are low quality and rather noisy.
3. The dynamic range is somewhat limited.

It's a decent consumer chip, yet nothing to write home about.


> Are there any hidden quality gain from transfering video over the AGP bus instead of using the PCI bus?

No.

[Swede]

[btw.Swede what happened to my examples in asynth forum?_someone feeled touched by it so kept it for themselves?hehehehe] Well, this forum is *not* meant to be a place for trading *anything* so pls, feel responsible for having the admins lowering the limit of posting down to <200kb.
wow,again monstruous reply!hope Swede will not take it as bad as we were more focused on the subject I would really like to have seen that but I'm sorry, your post are still toooooooo long and tooooo unreadable for me so I'll just stop reading your posts.

[Owen]

Im back again.

First, Im not here to get into an argument.
I'm too old for that.

Second, I think for this thread to be of any use, it is important for people to restrict there comments to hardware that they have personally used.

It is also important NOT to use the Tuner on the capture card you are evaluating.
You want to test the capture hardware not the Tuner.
If you want to make comment on Tuner quality, state it separately.

In my tests on capture hardware I use Svideo inputs on all cards for all tests and suggest that others do the same for best quality and consistency.

For BT8x8 cards, there is no reason to use VFW any longer as WDM drivers on Win 2000 or XP work fine now, with no capture file limits on NTFS drives.

I would also like to restate the importance of CONTRAST settings on BT8x8 based capture cards.
The default settings with the sourceforge.net drivers (and must others) are much to high.
Don’t complain about quality with BT8x8 cards unless you have carefully adjusted Contrast and other settings.
Failure to do this will result in a very unnatural looking capture.

I should also mention that that YUV2 capture seems works much better than RGB24.
I can’t get a good looking capture with RGB24, no matter how much I try.

Hope someone can post a tests on a card based on the new Conexant CX23881 chip.
This should be the chip we have all been waiting for.

Regards,

Owen.

[Darksoul71]

Hi !

Again @^^-+I4004+-^^:

>_it's not me,it's notepad!(hehe)
Huh ? Nothing funny here ! Your postings are a pain in the a$$ to read.
Even the crappiest texteditor (also Notepad) give´s you the option
to make paragraphs by hitting the <enter> key

As I said before: Thiswholewritteninablock-thing hurts my eyes while reading

>_how many cables did you tried & what differences did you saw?
At least two extremely different cables convinced me that the connection is
a very critical point in doing capturing. I onced used a pretty cheap
Audio cinch cable with a thin, shielded wire. The image was somewhat fuzzy,
unsharp and I got serious distortions (hope this is the right word).

I soldered my own cable (for Video only) using thin coax cable and solid,
shielded metal cinch connectors (not gold needed here) and tell you what:
Sharp image with no distortions. Although the cable is only 2 m long.

So, YES, I´ve tested at least two cables that provided a big difference in
quality.

>can i bet that you don't even know how does shielding work?
Well, you CAN bet but you´ll loose. I have a solid egineering education with
some very nice knowledge about electronics and HF stuff. Do you
know about HF, cables, capacity, complex resistance of cables, etc ?

Oops, just took out my VGA cable to compare length ! Sorry...

>this is NOT pro's stuff...this is to get better results encoding video at HOME...
Wahoo, what´s different to "pro´s stuff" ?
As Ookami said be4: You can easily beat "pro´s equipment" when using the right
techniques and equipment.

And at least we can learn more here compared to many "pro" forums where people
don´t even know how to deinterlace correctly without Adobe Premiere. :D

by looking at the picture (in some of the capturing formats huff,mjpeg..pitty though that huff wont play so it's nearly the same thing if i post pix..)you can say A LOT ...(off course knowing encoding technique)

IMO you can´t say anything on a card as long as you don´t test it in the same enviroment.
This would mean:
-Same source (DVD, satellite, etc)
-Same cable connection
-Same PC config (Another OS, driver release can easily f... up your capture quality)

So IMHO without this the comparision would be useless.

this is my view on the subject:
i KNOW my surrounding the best(same equipment is used ALWAYS,just the cards change.same vhs,same cables

Yes, but please judge only cards that you´ve tested yourself under the same circumstances.
This "I´ve read it in a newsgroup" stuff is ridiculous.

"If the VHS signal is not stable,DV compression brakes,as does MPEG"(end of quote)

After all your comments on the ADVC100 (without builtin TBC) I did a few tests with some
VERY old VHS tapes (15 years old and cheap tapes). Audio was pretty messed up and picture
is no where near watchable. I used my ADVC with composite In and NO frames were dropped during
100 min of capture using ScenalyzerLive.

My conclusion: If a VHS tape is so screwed up (in terms of signal stability) that the DV codec
of the ADVC-100 fails, then you´ll also get serious problems in doing captures using a BT 8x8 based
card and the special sync version of Virtual Dub (by Andreas Dietrich).

I also tested a newer 2nd generation copy VHS tape with the ADVC100 where my MSI GeForce4 VTR with
Philips Chipset failed (looses sync after 10 min of capturing). The ADVC100 kept sync audio
till the end of the movie.

So, a TBC isn´t really needed.

>as i can see u used erazorIIIpro...what are your experiences?
Crappy piece of hardware. Good signal quality but dropped frames even at
Quater CIF (352x288) where the Erazor Victory captured without problems.
Didn´t run at 75 MHz AGP clock. Crashed in OpenGL with Q3A (even at normal
AGP clock).
I returned this piece of sh.. to my local dealer and went home with a Voodoo³ 3500 TV
which I used for quite some time and which did a much better job in the video & game
department.

_thats not the problem.(agp is faster,but pci is enough with 133(as i remember)MB/s)..
In theory, at least in theory PCI is very fast but try the following combination:
-VIA based chipset (KT133/KT266/etc)
-SB Live/Audigy soundcard
-TV card capturing 720x576
-RAID PCI controller with two HDs as capture drive connected

Have phun..... ;)


2>Regarding macrovision, I remember reading in rec.video.desktop that you can disable it on all canopus bridges if you hold the ON/OFF button for 30 seconds

_problem wasn't related to this:home made VHS tapes in question.....


This was just for me and not related to the TBC stuff you´ve mentioned.

Now I´m off awaiting another monsterous posting from you :)

-D$

[^^-+I4004+-^^]

>Even the crappiest texteditor (also Notepad) give´s you the option
to make paragraphs by hitting the <enter> key

_that's why your posts look so funny?every other line broken at the start,sometimes even 1 word = 1 line...if you look moderators posts
you'll see that they more resemble to mine than your way of writing...
[although i didn't copied that from moderators:my first & most recent posts (on usenet & anywhere) looks the same:perfectly readable...long,but readable....]
you just make too many paragraphs_paragraphs are used to distinct different thoughts in message..... you use it all the time (i'm not sure why?i'm hell sure you arent saving any space or making it more readable:i'm used to read&write the old way:just as in books & newspapers....i've never seen that someone hits <enter> in the middle of the line as you.....quit
it
already
!
!
!(see....not pretty at all!)...hehehe i say again....)

[this altogether doesn't matter as long as i can read your & you mine posts_ and i guess we can(?)]

>Do you
know about HF, cables, capacity, complex resistance of cables, etc ?

_a thing or two:electrotechnic is in the name of my occupation...
i use UNSHIELDED pieces of wire for cables (i've just soldered conectors on some wire from toy-telephone...)->i've never had any interference problems & you can say that my room is full of electrosmog....(so in my case cables are of NO importance...)
i make all my cables(that way) for vcr,SATV,audio,etc._never ONE problem with them....(perhaps i have equipment thats shielded enough for itself....)

>As Ookami said be4: You can easily beat "pro´s equipment" when using the right
techniques and equipment.
^^^^^^^^^^(hehehe you've said it..)
_so perhaps we can expect some vhs-dvd releases from Ookami???PIH!
what "pro" (person or equipment) is used (uses) for analog capturing???NONE!(NOTHING)
in pro world vhs(or analog capturings for that matter) cannot be used for quality tv or media production (THIS CAPTURING WE TALK ABOUT? PURELY AMATEURS STUFF!)
he just said that cheap devices can beat expensive ones if used correctly
there isn't a category called:professional video capturing...(although some will claim that there is...)
dvd's,vhs',tv-programmes are made in different manner to this!


>And at least we can learn more here compared to many "pro" forums where people
don´t even know how to deinterlace correctly without Adobe Premiere.

_yes,you can learn here...
(btw. where is deint. used in pro-video production???i doubt you can cnnect those two words in one sentence...real pro goes to work every morning-doesn't sit at home makin videos_to clarify my vision of "pro"..)

>So IMHO without this the comparision would be useless.

_you're just splitting hairs&quoting someone else....(although all stuff i mentioned was on same machine,OS.. etc.)

if drivers for one platform(OS) are fawlty that is pretty soon KNOWN FACT...(so you can do dual boot & check if it's true...although for capturing win98 still should be the way to go:no dilemmas there...)

cables i've mentioned already (but if you're the one that can "hear" different cables when listening audio equipment,then i can't explain anything to you!)

& one who captures SHOULD know the quality of the source!
(we're not blind,are we?picture on tv can be reference&good parameter on what to expect....(good tv-set...i have panasonic& sony:if you have "SEG",disregard this...))


>This "I´ve read it in a newsgroup" stuff is ridiculous.

_why would this be?i think i explained good enough,if you know the croatian you could see for yourself what has been written/said...

i'm an active participant of mentioned group and i can guarantee you that stuff's TRUE!no mumbo-jumbo there....by now i know what to expect from what person...
(you CANNOT say that("ridiculous") without reading thread on canopus first!!!you simply cannot!you're jumping into words of canopus dealer in croatia?if you saw his DV FAQ,perhaps you would think twice before questioning his competence.....or are you telling me that you're questioning my truthfulness?in that case we've got nothin to talk about anymore!)


advc doesn't drop frames,it looses the picture completely (black screen)_on the same parts of the tape Bt based card drops more than usual,but at least it has picture:one more thing:(hope you can get this) problems with canopus were always with the same scenes on vhs recording!
(ie. no soft,or hardware problems in capturing system:particular signal (particular place/scene on vhs tape) always kills advc.....guess you know what that means...)

guy who tested canopus was potential customer:gave it up after this to buy something that costs twice as much!so it seems he didn't gave it up because of price!


i don't doubt your test methods,but perhaps you don't have really unstable vhs signal (noisy & unstable vhs isn't the same thing!look at the link in capture faq(here) on tbc to see what unstable means...i think i saw it there...-this guys vhs tape looked ok on tv...noisy vhs' DON'T look fine on tv!)

it MAY as well be that this is 1 in 100 cases,but would i risc getting expensive equipment just to find out that i cannot capture vhs of particular "vintage"?


>>as i can see u used erazorIIIpro...what are your experiences?
Crappy piece of hardware. Good signal quality but dropped frames even at
Quater CIF (352x288) where the Erazor Victory captured without problems.


myself couldn't get ("Elsa Victory Erazor(riva128)4MB '99.or even '98.")
over 23fps anyhow!(are we talking about the same card????)quarter cif worked without dropping (Era3Pro) on the same win98 & BX mobo...


>Now I´m off awaiting another monsterous posting from you

some silly uk commercial said"it's good to talk"(bt prehaps?)
i say "it's good to write!A LOT"hehe

[Swede]

Since you can't keep on-focus and stay away from the bashing... I'll now close this thread.