WinTV-PVR-250/350/500 Which one to get?

[SatStorm]

We don't capture the VHS / SVHS info at 720 x 480/576, but the picture canvas that includes that info.

So, the difference we have, is that you keep the canvas as is, but I don't! I like to extract the VHS / SVHS info from the canvas.
BTW, this is my overall problem with all this hype regarding those "PVR"s cards.

In case you don't know, Mainconcept 1.4.2 has a great feature and architecture, that's why it is so heavy as a program. It captures in a determined by the user framesize, frameserve internal what it captures realtime to the encoder, filter it and encodes to mpeg 2.
So, you can capture at 768 x 576 and encode to 352 x 576 realtime. This produce far better visual results than capturing and encoding realtime at this framesize
This is a benefit those PVR cards don't have.
Of course, Mainconcept is not a popular alternative, because it is not so good with the NTSC sources and most posters in this forum are from countries that use NTSC. But for us, the PAL users, it is a great alternative and IMO, if you have a CPU > 2Ghz, a far better one!

[davideck]

We don't capture the VHS / SVHS info at 720 x 480/576, but the picture canvas that includes that info.

So, the difference we have, is that you keep the canvas as is, but I don't! I like to extract the VHS / SVHS info from the canvas.
BTW, this is my overall problem with all this hype regarding those "PVR"s cards.

What is this "picture canvas" you are referring to?

[FulciLives]

We don't capture the VHS / SVHS info at 720 x 480/576, but the picture canvas that includes that info.

So, the difference we have, is that you keep the canvas as is, but I don't! I like to extract the VHS / SVHS info from the canvas.
BTW, this is my overall problem with all this hype regarding those "PVR"s cards.

What is this "picture canvas" you are referring to?

I find it clear what he is saying.

He is saying tha when you capture, for instance, a PAL VHS/S-VHS video, you really only need a resolution of roughly 352x576 or you could say that this resolutioin (known as Half D1) is more than good enough for a PAL VHS/S-VHS video.

So he captures at 720x576 and then resizes it to 352x576 thus "extracting" the needed resolution rather than leaving it at a more "wastefull" 720x576

The only argument here is that some find that 720x576 looks better than 352x576 even with a low quality source like a PAL VHS video. However one could argue that the higher resolution only looks better due to the way capture cards work and/or is a "trick" ala a televisions "sharpening" control which appears to sharpen the picture but really just adds "video noise". Anyways that's my own analogy.

Food for thought.

- John "FulciLives" Coleman

[davideck]

So he captures at 720x576 and then resizes it to 352x576 thus "extracting" the needed resolution rather than leaving it at a more "wastefull" 720x576

I am not sure what you mean by "wasteful". At a given bitrate, the Hauppauge converts directly from an Analog Video source to an MPEG file. What's "wasteful" about using a higher resolution during that process?

By "extracting" do you mean low pass filtering (smoothing)?

The only argument here is that some find that 720x576 looks better than 352x576 even with a low quality source like a PAL VHS video. However one could argue that the higher resolution only looks better due to the way capture cards work and/or is a "trick" ala a televisions "sharpening" control which appears to sharpen the picture but really just adds "video noise"

One could also argue that a higher resolution preserves more actual picture detail. One would certainly expect so and hope so. Yes, it may also preserve more noise, but one would expect more noise within a wider bandwidth.

[edDV]

What's "wasteful" about using a higher resolution during that process?

I think "wasteful" is a legacy concept from when DVD blanks cost > $2.

I just purchased 50 Chinese blanks for $2.94 after the rebate settles. That is $0.06 a blank. The remaining problem is small 4.35 GB capacity is beginning to look like a floppy disc.

[lordsmurf]

I find it clear what he is saying.

He is saying tha when you capture, for instance, a PAL VHS/S-VHS video, you really only need a resolution of roughly 352x576 or you could say that this resolutioin (known as Half D1) is more than good enough for a PAL VHS/S-VHS video.

So he captures at 720x576 and then resizes it to 352x576 thus "extracting" the needed resolution rather than leaving it at a more "wastefull" 720x576

The only argument here is that some find that 720x576 looks better than 352x576 even with a low quality source like a PAL VHS video. However one could argue that the higher resolution only looks better due to the way capture cards work and/or is a "trick" ala a televisions "sharpening" control which appears to sharpen the picture but really just adds "video noise". Anyways that's my own analogy.

Food for thought.

- John "FulciLives" Coleman

Yes. 8)

Also I'd say go for the PVR 250.

As far as Hauppauge blurring on 352, I've seen it do it a few times. But usually on clear source that was higher res. That's to be expected. One of those things, have to play around and not try to find errors where none exist.

[SatStorm]

There are plenty reasons the Full D1 picture compared the half D1 looks better on some cases.

- Most DVD Standalone's mpeg decoders, don't care to playback the best possible way half D1. This is the most usual reason.
This "problem" is even more noticable when you use interlace 352 x 288 (a PAL trick): All the DVD Standalones have to support this framesize, but since it is very rare at the commercial discs, they playback it really bad.

-Resizing is also a pain in the ass. When you resize the wrong way, you loose something. Lanczos is not the holly grail unfortunatelly.

"Extracting" the VHS / S-VHS infro from the Canvas is a complicated, time consumer and difficult task. And you are not able to do this with an mpeg 2 source (a compress source). You can do this with uncompressed avi, Huffyuv and maybe Mjpeg (when you use the compression to 19). Also, you can do this with sources from DV converters. And the steps to do it includes filtering, resizing, encoding, etc. One hour of video in 6 hours at best!

The bottom line is that the results gonna be visually identical with the realtime mpeg 2 full D1 captures you do with those PVRs, but in much more smaller filesizes (4 hours of S-VHS on 1 DVD5).

This knowledge, technics, alternatives, etc, must be kept on the surface! We must teach / inform the newcomers that they do exist. By keep telling them and advice them to "Use PVRs/ DVD Recorder Standalones, etc" we keep them in the dark of this hobby! We create "stupid" users and I can see in 2 - 3 years, the forum members heres knows nothing for all this! Do we want that?
I'm here long enough to see forum generations come and go. And I have to ring the bell right now: Newcomers learn less than 3 years ago! The overall level is lower right now than before. We must change somehow that. If we don't, in 2 - 3 years this forum gonna turn a niewbie forum only. And I don't wish to see this (already it is called like this on some other places...).

[davideck]

"Extracting" the VHS / S-VHS infro from the Canvas is a complicated, time consumer and difficult task. And you are not able to do this with an mpeg 2 source (a compress source). You can do this with uncompressed avi, Huffyuv and maybe Mjpeg (when you use the compression to 19). Also, you can do this with sources from DV converters. And the steps to do it includes filtering, resizing, encoding, etc. One hour of video in 6 hours at best!

The bottom line is that the results gonna be visually identical with the realtime mpeg 2 full D1 captures you do with those PVRs, but in much more smaller filesizes (4 hours of S-VHS on 1 DVD5).

What is this "Canvas" you are referring to?

[SatStorm]

Open a dictionary, or better ask a Photoshop user to explain you in detail.
Or someone with better english skills than me.

[davideck]

This knowledge, technics, alternatives, etc, must be kept on the surface! We must teach / inform the newcomers that they do exist.

I am willing to learn, and your english is fine...
Please explain the "Canvas" with respect to an Analog Video signal.

[BrainStorm69]

This knowledge, technics, alternatives, etc, must be kept on the surface! We must teach / inform the newcomers that they do exist.

I am willing to learn, and your english is fine...
Please explain the "Canvas" with respect to an Analog Video signal.

I'm not sure, but I think Satstorm is thinking in this way (or at least this is how I think of it when he talks about the "canvas"):

Suppose your vhs resolution is about equivalent to 240 lines. That means that 240x480 is really the maximum resolution you can get on your "canvas." The "canvas" is the area this 240x480 resolution "picture" is being "painted on" - i.e., your capture resolution (e.g. - 352x480, 720x480). The 240x480 resolution of vhs (the actual "canvas" size) can be "stretched" to fit your capture resolution (or "frame" if you want to continue the painting analogy), but that isn't really creating any additional detail, it's just "stretching" the actual, existing resolution across a bigger "frame" and creating a bigger (but not more detailed) picture.

[davideck]

The Vertical resolution of an Analog Video signal is fixed by the number of active lines per field.
The Temporal resolution of an Analog Video signal is fixed by the number of fields per second.
But the Horizontal resolution of an Analog Video signal is determined by its luminance and chrominance frequency bandwidths. These bandwidths vary among the different tape formats.

The VHS luminance bandwidth provides a horizontal resolution of up to 240 lines.
Since the horizontal resolution is specified per picture height, it must be multiplied by the aspect ratio to obtain the equivalent number of pixels or samples required per line.

240 lines X 1.33 = 320 samples per line.

352 samples per line is therefore adequate for preserving the VHS luminance horizontal resolution.


The S-VHS luminance bandwidth provides a horizontal resolution of up to 400 lines.

400 lines X 1.33 = 532 samples per line.

352 samples per line is therefore not adequate for preserving the S-VHS luminance horizontal resolution.
720 samples per line is adequate for preserving the S-VHS luminance horizontal resolution and therefore represents the better choice for S-VHS recordings that contain high frequency detail.

No amount of "filtering, resizing, encoding, etc." can preserve 400 lines of horizontal resolution with only 352 samples per line. It is all just fundamental sampling theory. No "canvas". No "overkill". No "hype".

[lordsmurf]

Your "samples" is only theory. Good on paper, not in real life. Not how things really work. All sorts of minor things can change the way it works, from the quality of the tape, to the quality of the hardware, to the mode of recording. Analog was controlled chaos, it was not an exercise in precision. Digital is precision (close enough, anyway).

When you think of the canvas, you also have to consider the 4:3 aspect. I guess it just takes a certain kind of thought process to understand all this. Some can do it, some just get confused. I'm having no trouble following satstorm. It all makes perfect sense.

S-VHS is about 400-500x480 equivalent recording quality, in a digital term. 352x480 can generally get enough of the data to look transparent to the source. Sometimes you need to do Full D1. It all depends on what you have to work with and what you see.

[BrainStorm69]

352 samples per line is therefore not adequate for preserving the S-VHS luminance horizontal resolution.
720 samples per line is adequate for preserving the S-VHS luminance horizontal resolution and therefore represents the better choice for S-VHS recordings that contain high frequency detail.

No amount of "filtering, resizing, encoding, etc." can preserve 400 lines of horizontal resolution with only 352 samples per line. It is all just fundamental sampling theory. No "canvas". No "overkill". No "hype".

I agree that for an S-VHS source, I would use full D1 to ensure maximum detail. In fact, I use full D1 for VHS unless it's a long capture that I need to use a low bit-rate to allow it to fit on one DVD without excessive macroblocking. I don't necessarily trust that all DVD players will play half D1 well.

I was just trying to explain how I interpret Satstorm's "canvas" analogy.

[SatStorm]

It is impossible to capture all the 400 horizontal lines from a S-VHS tape without hi-end equipment. We need hi-end cables, component or RGB connections and of course hi-end cards (those from Matrox I think are the most cheap ones we can get). A good alternative for the PAL users is DV, but then again, on extreme situations, with DV you get artifacts (compression aftifacts)

It's the psychological need to know that you did the best possible with your tapes, that's why we end up encoding at full D1. The bad mpeg decoding the DVD standalones do helps that direction too. If people see a difference, why to blame his loved equpment? He blame the framesize...
People can't accept this fact easy (neither can accept, that a PAL interlace 352 x 288 with a 1800kb/s picture bitrate is visually indentical a PAL VHS tape).

From a practical point of view, even if I somehow manage to capture those 400 lines, I prefer 48 less horizontal lines, than 320 fake ones. Don't mention the wasted bitrate. But that is just a matter of taste. Being a PAL user may also has something to do with it.

Davideck, it seems that you have a problem with the practical aspect of this hobby. It seems that you are too much of a theory person. Back in 2001, when we use to learn TMPGEnc-ology to convert our tapes, you would have a great problem accepting things....

[davideck]

Your "samples" is only theory. Good on paper, not in real life. Not how things really work.

Real life limitations almost always prove theoretical calculations to be optimistic. The samples_per_line figures calculated above represent theoretical minimums for the resolutions specified.

S-VHS is about 400-500x480 equivalent recording quality, in a digital term. 352x480 can generally get enough of the data to look transparent to the source. Sometimes you need to do Full D1. It all depends on what you have to work with and what you see.

In general, I can agree with these statements. Specifically, with respect to the Hauppauge PVR-250 and my S-VHS source tapes and my JVC HR-S6800U VCR, the 720 captures are noticeably better. No question about it.

Davideck, it seems that you have a problem with the practical aspect of this hobby. It seems that you are too much of a theory person.

SatStorm, we do indeed approach this hobby from opposite ends. I am most interested in preserving as much original detail and picture quality as possible in the shortest amount of time. You have acknowledged that the Hauppauge cards capture better at 720 and I agree. But that doesn't make them "wasteful" or "overkill" or "fake". The amount of time that you spend filtering, resizing, encoding, etc. seems "wasteful" to me, but it doesn't to you. Trying to pack 4 hours of Video onto a single DVD5 seems like "overkill" to me, but it doesn't to you. (Bitrate is a different discussion, because we must then consider the effects of motion and compression. Perhaps we can debate that another time...)

Can we agree that both approaches to this hobby are equally worthwhile? The Hauppauge cards may not be a good choice for you, but that does not make them a "bad" choice for everyone else.

[SatStorm]

Davideck we don't "debate" on something here. We simply testify what we know about a hobbie we both love! I don't have to proove something to you, neither you have to proove something to me. We have our results, thoughts and tests and simply post them so others to read and get informed! They gonna choose, we won't tell them what is right and what is wrong!

And yes, the use of a framesize higher the original info and the use of the bitrate that is needed to support it, for me is a waste of filesize, a framesize overkill and a fake thing.
For you, what I do in 6 hours and you do realtime but on 4X filesize is a waste.
We have a different approach here. Maybe because for you 1 hour per DVD5 is enough, while for me is too little.

And of course both approaches are equally worthwhile. Some wants more per disk, some wants faster results. Both want the best possible and both approaches succeed that.

What bothers me, is the fact that my approach doesn't have any voice in the forum any more. Not because there is no interest about it, but because noone mention it. So the newcomers don't learn about it!
I'm the minority here, but since this approach exist, why not to mention it?

Regarding the Hauppauge PVR cards, I like what they do (the realtime mpeg 2 thiny) but as capture cards I don't like the quality they offer.
For example when I compare the software realtime mpeg 2 results of my old trusty Asus 7700 Deluxe (produced with mainconcept 1.4.2 at 720 x 576) with the results I get from my Hauppauge PVR 250 at the same framesize / bitrate, my Asus card wins hands down.

Finally, regarding what I believe, here it is: I believe that those hauppauge PVR cards are a bad choice for those wishing for perfection in less filesize.
They are good choices for those that don't care about filesize.
The way I see it, PVR cards don't exist to serve achival porpuses, but time shifting needs. In a couple of years they gonna be a distant memory, because software solutions gonna do the same better. Just like built in mjpeg cards don't exist anymore (remember Matrox G400? Sefy's favorite capture card!)

[edDV]

Hey. This is getting fun

Davideck we don't "debate" on something here. We simply testify what we know about a hobbie we both love! I don't have to proove something to you, neither you have to proove something to me. We have our results, thoughts and tests and simply post them so others to read and get informed! They gonna choose, we won't tell them what is right and what is wrong!

And yes, the use of a framesize higher the original info and the use of the bitrate that is needed to support it, for me is a waste of filesize, a framesize overkill and a fake thing.
For you, what I do in 6 hours and you do realtime but on 4X filesize is a waste.
We have a different approach here. Maybe because for you 1 hour per DVD5 is enough, while for me is too little.

And of course both approaches are equally worthwhile. Some wants more per disk, some wants faster results. Both want the best possible and both approaches succeed that.

So we have the archivers wanting to save every last possible drop of quality, vs. the compressors willing to invest hours-days-weeks of rendering to get acceptable video into the smallest flashcard (or IPOD video).

Defining the outer bounds is usefull but most people must choose a mid course. In the old days we didn't have the technology to get to theoretical quality at the consumer level. Compromise was needed at the capture point, in compression and in storage. The capture devices were crude, the computers lacked power for adequate quality during compression and we were limited to tight storage media targets (floppy, then CD then DVD5) The progress of technology extends to both extremes: 1) you can save more quality -OR- 2)you can compress playable video into smaller and smaller media. Both goals are valid and both are pushing the envelope and generating new uses.

What bothers me, is the fact that my approach doesn't have any voice in the forum any more. Not because there is no interest about it, but because noone mention it. So the newcomers don't learn about it!
I'm the minority here, but since this approach exist, why not to mention it?

The technical action is in the extremes. The middle course becomes more accessible with less effort. Capture devices are getting better, CPU power is getting cheaper, hard drives are approaching US$0.015/GB. DVD5 is approaching US$0.009/GB. The problem with DVD is limited capacity more than cost. Double the capacity at double to tripple the cost would be welcome and would save time.

Regarding the Hauppauge PVR cards, I like what they do (the realtime mpeg 2 thiny) but as capture cards I don't like the quality they offer.
For example when I compare the software realtime mpeg 2 results of my old trusty Asus 7700 Deluxe (produced with mainconcept 1.4.2 at 720 x 576) with the results I get from my Hauppauge PVR 250 at the same framesize / bitrate, my Asus card wins hands down.

That is because the PVR series design is 5yrs old and Connexant hasn't focused on higher quality, just cost reduction. ATI has better quality MPeg2 hardware encoding technology but can't seem to master the necessary software. Meanwhile software encoding technology is advancing (Mainconcept) and there is no clear winner. Over time, the winner will be hardware in applications that have the base customer volume.

Finally, regarding what I believe, here it is: I believe that those hauppauge PVR cards are a bad choice for those wishing for perfection in less filesize.
They are good choices for those that don't care about filesize.
The way I see it, PVR cards don't exist to serve achival porpuses, but time shifting needs. In a couple of years they gonna be a distant memory, because software solutions gonna do the same better. Just like built in mjpeg cards don't exist anymore (remember Matrox G400? Sefy's favorite capture card!)

I agree that Hauppauge cards are best for timeshifting and only adequate for archiving, but I differ that software will overtake hardware. I truely believe hardware will win for VHS to DVD archive and for the PVR timeshift space. ATI may miss the opportunity as they fumble with software applications and drivers for their Theater 550/Elite hardware. Meanwhile other hardware boxes will overtake computer solutions for mass market VHS archive and PVR..

[lordsmurf]

So we have the archivers wanting to save every last possible drop of quality, vs. the compressors willing to invest hours-days-weeks of rendering to get acceptable video into the smallest flashcard (or IPOD video).

No, that's not it.

When it comes to DVDs, there are really just 3 people:

(1.) Anal retentives that insist on using the highest resolution and high bitrates, all in the hopes of retaining theoretical data that may or may not exist. Many of them just don't know any better. Some of them have a valid reason for what they do.

(2.) Normal folks that want high quality, but without getting into overkill, and without filling their house with disc after disc after disc. You can put 2-4 hours if you have the right hardware/software, and if the source allows for it. It's about slight compromises that really don't affect quality perceptions. A 3-hour DVD in Half D1 will look good from most sources most of the time. This is an optimal setting for home users. Again, assuming hardware and software allows for it.

(3.) Compressors. These people suck. They destroy video by compressing the crap out of it, and toss out huge amounts of data to make their crappy little VCDs, KVCDs/KDVDs, XVID/DVIXs and other junk. These people should lose their video privileges. Morons.

..........

This thread discusses a little bit of #1 (not to the extreme though, just a touch), and then discussions on #2 (whether the card in question would allow for high quality work).

I don't think anybody has said the card is bad. Simply that some people may sometimes see a software output on some sources, so use your best judge between Half D1 and Full D1 on the Hauppauge cards.

[edDV]

(1.) Anal retentives...

(2.) Normal folks...

(3.) Compressors. These people suck.

HA !

I did say "Defining the outer bounds is usefull but most people must choose a mid course." but in a less judgemental way.

[davideck]

(1.) Anal retentives...

(2.) Normal folks...

(3.) Compressors. These people suck.

I am in category 2. I try to maximize the duration of each DVD without noticeably sacrificing picture quality or detail. One hour per DVD is best when capturing my S-VHS masters with my JVC HR-S6800U. Two hours per DVD is fine when capturing with my HR-S9600U (with TBC/DNR), as this VCR provides less detail with less noise.

[SatStorm]

Regarding the capture devices, they are all the same the last 4 years. What changed, is that they sell them cheaper. So, what you had in 2001 is the same you have today. The chips used for capturing are the same.
After all, capturing analogue has no interest anymore. In USA it's all about HDTV now and in Europe is all about DVB /s or /c. Don't mention that the big ones wish to close the so called "analogue hole" once and for all. (most later capture cards sense macrovision...)
So, IMO, I don't see the technology for analogue capturing get more advance in the future, regarding the hardware.
I expect software solutions for the future. Latest mainconcept PVR proove that (it produces excellent results with any bt8xx based card on any - more than 2Ghz CPU - system).

Regarding the courses:
Using those mpeg 2 PVRs to capture realtime mpeg 2, is not the mid course. It is the easiest course, like using a DVD standalone recorder.

Capture analogue (avi) with later technology capture cards or DV converters is the mid course.

This is my opinion of course.

The compressors are sick people: From a point and beyond, you can't compress without harming your project. I'm not among them. I'm among those few that want 100% of what this technology really offers.
The determination of this 100% for me, means: "Best visual quality with the needed bitrate, the needed framesize, and I don't care about the time to succeed it".
For others, the 100% means "best visual quality, fastest passible way, don't care for filesize".

On both cases, you have best visual quality results. There are both alternatives to succeed the same!

edDV, the prices you mention, exist in USA and they don't apply for top quality media (those last in time - or we hope so). 8X TY media for example in Greece, costs a half euro (bulk). For your information, the typical Greek salary for a 30 year old, is 540 euros per month. And the 30 year olds without a job, are about 23% of the population. It is the same everywhere in Europe right now... (the salaries on some countries are higher, but the taxes are also higher, so you end up on about the same....)

BTW,the latest maincocept PVR costs 50$ and a typical BT8xxx card cost 30$. So, with 80$ you can have something which produce better picture results than a Hauppauge PVR card. Of course, you need a 2GHz CPU for full (or crop) D1 capures. This is cheaper a PVR card, right?

[FulciLives]

These days I'm using a stand alone DVD recorder but I miss the old days of using my BT chipset PCI card and using AviSynth and TMPGEnc or CCE to encode etc.

Sometimes (especially with VHS sources) I wish I still did it "the old way" but the major flaw with the old way was always A/V sync and I gave up on it.

One day (sooner than later perhaps) I'll get a Canopus ADVC-110 or DataVideo DAC-100 and go back to the old way of capturing and filtering and doing a VBR software MPEG encode.

But that takes WAY too much time. OK for pre-recorded stuff like VHS or LaserDisc but just not reasonable for TV recording.

I record 1 show on Monday (60 minutes). I record 2 shows on Tuesday (30 minutes each). I record 2 shows on Wed. (60 minutes each). I was recording a single 60 minute show on Thursday but that was canceled. I record a single 60 minute show on Friday. There is a show on Cartoon Network's Adult Swim that is 30 minutes and I must record that nightly from Monday - Thursday. Then there is the occassional movie I record from FLIX or SHOWTIME etc. not to mention specials I like to record from time to time on channels such as the HISTORY channel or the NATIONAL GEOGRRAPHIC channel etc.

No way I could do that the "old" way and keep up!

The missig link as far as I am concerned is some sort of hardware device that can do video noise reduction that works similiar to Convolution3D or some sort of temporal filtering etc. and although I do like my stand alone DVD recorder (for the most part) I do miss a true multi-pass VBR encode.

As far as capture cards go I never really could understand why someone never came out with a card that allowed uncompressed or slightly compressed video (HuffyUV or PICVideo MJPEG) with audio included and "locked" so we don't have the deal with the damn video clock and sound clock being off thus causing A/V sync issues.

Closet capture card to A/V sync lock is either a hardware MPEG capture device or the DV decives like the Canopus or DataVideo I mentioned before but with one you get a capture with a bit too much compression (MPEG even at 15,000kbps can look bad with a less than perfect source) and of course DV introduces the 4:1:1 "bug". The old PCI cards with HuffyUV or PICVideo MJPEG have A/V sync issues and even if you can get perfect sync you end up with an "odd" frame rate (like 29.972 or 29.968 instead of 29.970 dead on) and then that has to be adjusted afterwards.

No wonder so many people love the Canopus/Datavideo DV devices or MPEG hardware capture cards (and now stand alone DVD recorders).

- John "FulciLives" Coleman

[edDV]

Regarding the capture devices, they are all the same the last 4 years. What changed, is that they sell them cheaper. So, what you had in 2001 is the same you have today. The chips used for capturing are the same.
After all, capturing analogue has no interest anymore. In USA it's all about HDTV now and in Europe is all about DVB /s or /c. Don't mention that the big ones wish to close the so called "analogue hole" once and for all. (most later capture cards sense macrovision...)
So, IMO, I don't see the technology for analogue capturing get more advance in the future, regarding the hardware.

Probable advances, 3D comb filer for Y/C separation, maybe 9 or 10bit A/D, better hardware filtering (proc amp, TBC, DNR, etc.) built into capture device.

HDTV will need hardware support for encoding and decoding.

edDV, the prices you mention, exist in USA and they don't apply for top quality media (those last in time - or we hope so). 8X TY media for example in Greece, costs a half euro (bulk). For your information, the typical Greek salary for a 30 year old, is 540 euros per month. And the 30 year olds without a job, are about 23% of the population. It is the same everywhere in Europe right now... (the salaries on some countries are higher, but the taxes are also higher, so you end up on about the same....)

Half Euro (bulk) is still a good price here as well. These crazy prices of the past few days are only store promotions to start the Christmas buying season. Sort of like a Le Mans running start. Prices go back up tomorrow.

BTW,the latest maincocept PVR costs 50$ and a typical BT8xxx card cost 30$. So, with 80$ you can have something which produce better picture results than a Hauppauge PVR card. Of course, you need a 2GHz CPU for full (or crop) D1 capures. This is cheaper a PVR card, right?

If they ever get the PVR-150 working right, they cost ~$99 list $69 on sale. It comes with some software but to get a better PVR you need shareware or commercial solutions like Beyond TV.

[SatStorm]

FulciLives, you are very unlucky with your capture cards, what to say... I never had lipsynch issues myself, but I'm Pal and that can makes the difference (it is all about 25fps for me. Much easier!).
Regarding capturing the TV shows, I grabb direct the stream "as is" from satellite (what I watch is on satellite only). I do that 7 years now with a push of a button (Nokia 9600 with dvb2000 or with my Sky Star 2). Of course those are European things, I don't know if those are options for you, the NTSC people. The point is that we have other alternatives here to do those things and quality wise, in a lossless way! (DVB /s or DVB / t cards)

edDV, 3D comb filer for Y/C separation existed since the ISA days. The 9 and 10bit A/D chips are from 2000. TBC existed from the 70s, Proc Amp use to be mainstream from the 80s (vidicraft, etc) and DNR exists on VCRs from early 90s!
The technology remained the same. Nothing new added! What changed is that it is cheaper and some cards are capable to combine elements. But the price is about the same bying seperate units (and seperate units offers more!). And, the reason TBC turned neccessary, is macrovision and fake macrovision (latest nvidia capture drivers, ATI legentary problems with this issue, etc). Even latest canopus cards detect macrovision and need TBC for by passing it!

Regarding HDTV, in Europe we can grabb HDTV Satellite channels with a typical Sky Star 2 card and progdvb (or other related programs), even with a 1.5 CPU PC. The problem is that we don't have TV screens to playback them (1080 just appeared and they are very expesive) and there is no way to store those grabbed files to anything else than HDs (on DVDs, we end up to something that reminds me the SVCD days...). For your information, some mpeg 4 DVB tests made by Premiere World (Astra series 1 Satellites, this is a german subscription service) also grabbed easy with the same card and modified software. The cost of a Sky Star 2 card? 60 Euro!

Encoding to HDTV is a very distant need: We don't have home-made sources for that framesize. At the time being, we have very few satellite HDTV channels in Europe, which we can grabb and burn on DVDs as data, really easy and cheap as is!

Finally, we can use GBPVR and a plug-in (both freeware) and turn any analogue capture card to PVR those days (if we have a more than 2Ghz CPU). You don't need one with built-in mpeg 2 hardware abilities anymore! Don't mention that the heart of modern HTPC in Europe is a DVB /s (or /t) card and not a PVR one.

Because those things are non existed in USA, you don't mention it. And because of this, this forum lost the international appealing once had. This is a huge problem IMO. We need more European users but the more you don't talk about those things (and don't even mention it, you don't even know about them!!!) the less non Americans come here.
Maybe it is time to search for a new and international forum to gain again interest for this hobby. It's time to think a bit for my future here.

[lordsmurf]

DVB is largely lost on North America.
That's a European luxury that we don't have.

[Wilbert]

Whoa long thread.

@Satstorm

From a practical point of view, even if I somehow manage to capture those 400 lines, I prefer 48 less horizontal lines, than 320 fake ones. Don't mention the wasted bitrate. But that is just a matter of taste. Being a PAL user may also has something to do with it.

You don't create 320 fake ones. You also don't keep those 400 lines intact, because they are analog and no upsizing is taken place. Even if they would be pixels, and cap cards would upsize, they don't use a point sampler for that.

[SatStorm]

Wilbert, we always capture full frame (the canvas). Then, we can choose: Or we extract the 400 lines and loose 48 horizontal lines so to end up to 352 x 576 (which DVD supports) or we emulate the canvas and we keep everything inside it.
The second option is easier, have the same visual results, and is possible to made it realtime. The only minus is the filesize (the more bitrate we need)
Also, I doubt that we can capture all those 400 horizontal lines inside the 720 canvas. If you know a way, in practice and not in theory, then please suggest the combo: That includes a card capable to capture all those horizontal lines, the needed cables to carry the signal, the needed sources (best SVHS deck), etc. I wish to see a solution for this, costing less than 1500 euro!
If you know something, enlight me.

[davideck]

Wilbert, we always capture full frame (the canvas). Then, we can choose: Or we extract the 400 lines and loose 48 horizontal lines so to end up to 352 x 576 (which DVD supports) or we emulate the canvas and we keep everything inside it.

The number of horizontal lines (240, 400, etc) is simply a unit of measurement that describes the frequency bandwidth of an analog video signal. These lines do not actually exist along the horizontal direction in the same way that lines exist along the vertical direction.

There is no sampling structure in the horizontal direction inherent in an analog video signal.
Once the video is sampled by a capture card, the frequency bandwidth preserved will be proportionate to the number of pixels sampled per line. At that point, a horizontal sampling structure now exists.

[SoCalMK]

The 500 has no remote or IR though.
If you want a remote, and the ability to watch/record two channels at once, get two PVR-150's.
One retail version with remote, and one MCE version with FM.
This gives you lots of options for additional inputs as well as the remote, IR, and FM radio.

The 250 is being phased out, replaced by the 150.
The 350 is just a 250 with additional hardware mpeg-2 DEcoding. It's output is extremely good, but ONLY for mpeg-2, so if you want to watch avi's on the TV, you should just stick with a 150 and a good video card with TV out.

Couldn't you get the 500 and use a PC remote like Firefly to control it?