Conversion to Support Progressive Scan DVD

[wwaag]

I'm not sure if this is the right forum, but here goes anyway. Using a HD monitor and a progressive scan DVD player, commercially produced DVDs look really great--at least on the showroom floor--as compared to a normal DVD player and standard monitor. It's my understanding (please correct me if wrong) that such images are first, non-interlaced, and second, they are usually displayed at a frame rate of 24fps.

My question. Can you--and if so, how--take your home videos, captures from laserdisc, satellite, etc. and achieve similar improvements in quality? I understand that since the source in these instances will never be as good, the results will never be as good (GIGO). However, can you get some improvements in the quality of the displayed image using a progressive scan DVD player and HD monitor?? If so, how do you convert from captured DV source material?

Thanks in advance.

wwaag

[wwaag]

Hey guys!!

Surely someone must have some info on my question regarding conversion for use on a progressive scan DVD player. Thanks.

wwaag

[Quigonsean]

Well, you can first DeInterlace you captures, either during encoding, or some capping utilities such as WinDVR which I use DeInterlaces on the fly as it caps, this makes your video a progressive video, however if your TV does not support Progressive frames even a progressive DVD play (Which you need to play it in progressive mode) if you tv is not progressive then it won't display it progressive though I don't know exactly what it does I think it displays each fram twice.

If the video you capped was originally filmed at 23.97 fps then you can go back to that by running IVTC filter such as the one in Tmpegenc (IVTC = Inverse Telecine) Then use Pulldown.exe to add the 2/3 flag so it playes it at 29.97fps. You cannot do this with a home movie because it was not done in 23.97 when it was shot. To do IVTC without jumpiness and yucky stuff the flick had to be filmed org in 23.97 if I'm remembering correct. Though I could be wrong on that. Now if the org filming was in 23.97 which most movies are most are of the opinion that it will improve the quality.

Sean

[Ste]

Keep the video interlaced. The job of a progressive scan DVD player is to do the de-interlacing for you. That's what it does. All commercial DVD's are 480i (i = interlaced). The progressive scan DVD player takes this data and converts it to a 480p signal (p = progressive).

Also, if your progressive scan DVD player has 3:2 pulldown, then you don't have to worry about the FPS. It'll take 29.97 and bring it down to 24 for you.

[malducci]

[Keep the video interlaced. The job of a progressive scan DVD player is to do the de-interlacing for you. That's what it does. All commercial DVD's are 480i (i = interlaced). The progressive scan DVD player takes this data and converts it to a 480p signal (p = progressive).

Also, if your progressive scan DVD player has 3:2 pulldown, then you don't have to worry about the FPS. It'll take 29.97 and bring it down to 24 for you.]

I agree that if your source is interlaced, then you might want to keep it interlaced to keep things simple ( less work ). *BUT* if your viewing this material on a HDTV or monitor, then you do have the option of using a more complex deinterlacer program or filter ( I don't mean blend fields or discard 1 field).

I do not agree that a progressive scan DVD player has a higher quality deinterlacer, then some of the OFFLINE intelligent deinterlace programs or filters. Also, not all DVD's are interlaced. Actually, most DVD movies are progressive scan with 2:3 pulldown . As far as taking 29,97fps to 24fps - no DVD player that I know of can output 24p video. I have read about 72hz HDTV's that can take 3:2 pulldown video and show the video in its native 24p ( each frame lasting 3 frames 72/24=3 ) giving smooth playback, but these are not on the consumer market yet.

Some newer HDTV sets and HD monitors claim to have 3:2 pulldown, but this advertisement is misleading. What they actually have is an IVTC chip - my Hitachi has it ( its only 7point IVTC were the newer sets are 20+ point ). 3:2 pulldown is not the same as IVTC. IVTC recreates 24p video from 29.97 Telecined video. After the HDTVs IVTC chip decodes the video to 24p(23.976fps progressive), it applies a 2:3 pulldown( not 3:2 pulldown ) to the video and displays it. My TV activates this by putting the set in "movie" mode and you can see when it sometimes misses a frame ( interlaced artifacts ). Most newer HDTV's and HD monitors have adaptive deinterlacing instead if line doublers, but it still doesn't look that good - most commonly leaving temporal artifacts or smearing.

Check out www.100fps.com - eventhough this site is for PALl video - the same techniques apply to NTSC. The 60fps video is great if you plan to keep it on a computer. [/quote]

[FulciLives]

Hello

As far as I am concerned you should leave it interlaced UNLESS you can do an IVTC on it (inverse telecine) in which case you can try doing that but this can't be done with stuff shot on video camcorders including DV cams. You might be able to do IVTC on sources such as VHS, TV programs, LaserDisc, etc.

With video camcorders you would have to deinterlace which is different from doing IVTC. Generally speaking deinterlacing will give you less quality then just leaving it interlaced. The only time I would deinterlace anything is if I have too (for instance it will be viewed on a PROGRESSIVE only display).

I believe that all if not most HDTV televisions are capable of both interlaced and progressive playback. So it rarely is worth it to deinterlace and although some disagree I don't really see the point in doing IVTC either. In my opinion when you IVTC something it is never as "sharp" looking as a studio mastered progressive DVD so if you aks me it just ain't worth it.

- John "FulciLives" Coleman

[malducci]

Exactly,

IVTC is not for DVcam video or true interlaced video. But in cases which involve capturing from VHS, Laserdisc, and satellite movies - there is a good chance the video might be Telecined. You would then definitely want to use IVTC if the movie is telecined.

although some disagree I don't really see the point in doing IVTC either.

If you understand what IVTC is, then you should understand its benefits. If a movie is original telecined from 23.976fps progressive to 29.97fps interlaced and you encoded the movie in 29.97fps interlaced, then you just wasted 20% of your bitrate bandwidth. That extra 20% could be used encode a smaller video file with the same bitrate or keep the file the same size and up the bitrate.

In my opinion when you IVTC something it is never as "sharp" looking as a studio mastered progressive DVD so if you aks me it just ain't worth it.

True: any analog capture or even satellite capture is not going to look as sharp as a studio mastered DVD. But other advantages to IVTC your telecined video: is if you have a progressive TV, your filters for cleaning up the video have a 20% less work load, your encode time is 20% faster( if your system isn't on the slow side). Also, IVTC is so easy to use too. Two simple lines in avisynth and your on your way. BTW, progressive video is convert to interlaced video by the dvd player if you using RF, composite, S-video, and 480i component. So even interlaced TV setups can still benefit from the bitrate savings.

I believe that all if not most HDTV televisions are capable of both interlaced and progressive playback.

Obiviously you don't have a HDTV or HD monitor. If you could manage to disable the onboard deinterlacer, the video would look like cr@p. HDTVs have either 1080i(960i on some) or 720p as there high resolution and 540p or 480p as there low resolution. But, 480i which is NTSC resolution
is up-converted to one of these resolution mention for HDTV. Have you ever watched interlaced captured video on your computer? It looks like cr@p because you can see the interlaced lines - same thing happens on a HDTV since the resolution higher like a computer monitor ( I use that analogy loosely ).

There are several reasons why a true interlaced NTSC TV hides its true nature - mostly notably is that TV manufactures overlay PART of even field scanline with the odd field scanline. Also, the phospherus(SP) of the previous field is dimmer than the new field draw - strange that no one has create a filter to emulate this before field blending. Anyway, this effect does not scale well past 25" TV and worst for 32"+ up. But a computer monitor and a HDTV don't present the video this way, a scanline from each field is perfectly aligned with one-another.

So to my first point, if you want to view interlaced video on a progressive TV (EDTV or HDTV) - you'll either loose half your vertical resolution as cheap realtime deinterlacer will through out 1 field and interpolate the other field to fit the vertical resolution, or you'll get temporal artifacts, smears, or trails from an adaptive deinterlacer. And this is all assuming you progressive scan DVD player does try to help you out be using its own style de-interlacer. So why not use a smarter de-interlacer with custom filters before you encode.

All this writting and I haven't even mentioned about video noise reduction. I guess this was just a waste of time since most people have standard TV sets and this pretty won't help them out if there doing realtime capturing and encoding.

I'm tired...goodnight :)

[FulciLives]

A progressive DVD has an internal frame rate of 24fps (23.976fps to be exact) but an interlaced signal has a frame rate of 30fps (29.97fps to be exact).

Now I understand that IVTC takes a 30fps source and converts it back to 24fps

But whenever I do this it never looks as good on my interlaced TV (no, I do not have a HDTV) then if I left it interlaced. Why is this?

I'm not sure how the IVTC process works exactly but consider this. An interlaced video signal (that can be IVTC'ed) has 30 frames per second in a 3:2 pattern meaning there are 3 progressive frames followed by 2 interlaced frames followed by 3 progressive frames followed by 2 interlaced frames etc.

So in one second you have only 18 progressive frames and 12 interlaced frames. So when you do IVTC is can't just remove the 12 interlaced frames or you would end up with only 18 progressive frames. Where do the other 6 progressive frames come from to get a frame rate of 24fps? My guess is the IVTC process somehow "creates" these progressive frames using the interlaced frames. Again, my guess is your first 3 progressive frames are frames 1, 2 and 3 and then the 4th and 5th interlaced frames get combined into one progressive frame and this becomes frame 4 in the new progressive video. Then the next 3 progressive frames are used and then the next 2 interlaced frames become one again etc.

Am I wrong about this?

My guess is the original 2 interlaced frames becomming one progressive frame is a process that is less than perfect ... hence the reason why a 29.97fps video IVTC'ed to 23.976fps does not look as sharp as a studio mastered progressive source which is a straight conversion ... each frame of film is a progressive frame in the video.

It is for this reason that I don't like to do IVTC because the output, while progressive, looks soft. When you then watch such a signal on a normal interlaced TV it has to be converted back to 29.97fps and the result is worse than if you just left it that why to begin with.

Perhaps doing an IVTC can help improve the quality for HDTV playback but it sure doesn't have that effect when watching it on an interlaced TV.

I also understand (I could be wrong) that most modern HDTV sets have no problem doing IVTC on a NTSC source when that is possible so again why should you do IVTC? And yes I understand the less info to encode stuff but still that doesn't make up for the softness of the IVTC process.

- John "FulciLives" Coleman

P.S.
Can a HDTV display interlaced video as interlaced or MUST it do some sort of deinterlacing/IVTC/line doubling since it is a progressive only (is it?) display device???

[DJRumpy]

Telecined material on a DVD is not interlaced and it is not 29.97 fps. The pulldown flags make it appear as such. When you run pulldown (or select '3:2 pulldown when playback' for you TMPGenc users), you simply add a flag into the MPEG stream, that says 'duplicate this frame during playback'. The stream is still 23.976 fps. It is telecined on the fly to 29.97. A progressive scan DVD player simply ignores these pulldown flags, outputing a 23.976 video (progressive).

Performing IVTC will not dull your output. It does not make it less 'sharp'. It simply removes these duplicated fields. They are exact copies of fields already present in your video. The only affect IVTC could have on an output video, is to make it jerky. This happens when the filter finds the wrong 'duplicated' field, causing it to remove the wrong field, and a jerk in playback is the result. The new crop of filters are very good at finding these duplicated fields though. They can have difficulty with mixed sources (telecined, with true interlaced frames in the source), but as a rule, most sources don't do this (dvd, vhs, etc will usually be telecined, or true interlaced, but not normally both). In situations where they are hybrid, I would probably just leave the source as is.

So in one second you have only 18 progressive frames and 12 interlaced frames. So when you do IVTC is can't just remove the 12 interlaced frames or you would end up with only 18 progressive frames. Where do the other 6 progressive frames come from to get a frame rate of 24fps?

This isn't accurate. The Telecine process adds duplicated fields from the existing content. It adds one additonial frame for every 5, and blends two frames (interlaces) as a result. It's easier just to show you (I hope my ansi art is up to snuff ). It does not destroy any of the original frames or fields in the process. It simply reorders them for playback.

A progessive video stream is still made up of 2 fields. Top, and Bottom (denoted by a T, and B. The frame number in parens ( ) ) like the following 4 progressive frames:

(1) (2) (3) (4)
1T 2T 3T 4T
1B 2B 3B 4B

When the telecine process is applied, some of these fields are copied so an additional frame is created. No frames are deleted, but three frames (1,2,3) are spread out to create a 3rd frame. It looks like this for a telecined source
:
(1) (2) (3) (4) (5)
1T 1T 2T 3T 4T
1B 2B 3B 3B 4B

Only frames 2 and 3 are modified. 1, 4, and 5 are left untouched. 4 and 5 are bumped to the right by the additional frame, becoming frame #4, and #5 respectively. If you look at this example, our 2nd frame's top field (2-1T) is a copy of frame #1's top field. Our frame #2 still has the same bottom field (2-2B). Examing the 3rd frame will show that frame #3 has a copy of the second frames top field (3-2T), but it keeps the original 3rd frame bottom field. The 4th frame (used to be frame #3) is untouched. So is the 5th frame (used to be frame #4).

To undo this process (IVTC), the top frame of frame #2 is removed (1T). The bottom field of frame #3 is also removed (3-3B). By removing these, Frame #3 shifts to the left, becoming Frame #2 again, with 2T for the top field, and 2B for the bottom field. Frame 4 also shifts to the left, becoming field 3 again (3T, and 3B). Frame 5 becomes Frame 4 again. Think of the process like that old game 'connect 4'. If you remove a block from a wall, the blocks above that would drop down to take it's place. A rough analogy, but it should get the point across.

Regardless, the IVTC process will simply remove the duplicate fields. No video data is lost under normal circumstances, since we are simply putting things back to normal (normal being 23.976 fps).

My guess is the original 2 interlaced frames becomming one progressive frame is a process that is less than perfect ... hence the reason why a 29.97fps video IVTC'ed to 23.976fps does not look as sharp as a studio mastered progressive source which is a straight conversion ... each frame of film is a progressive frame in the video.

It is for this reason that I don't like to do IVTC because the output, while progressive, looks soft.

It is perfect. It's a simple matter to split out the top and bottom fields, and combine them in any manner you like. They are distinct and seperate in your video. If your experiencing 'soft' video, then something else is at fault. IVTC will not soften your video. It will discard dupicates. The remaining fields are not modified in any way to make them 'fit' with their old counterparts. You actually increase quality, due to the fact that you have 20% less data to encode. That bitrate can be applied to the remaining frames, to increase their quality, or to simply reduce the size of your output.

I also understand (I could be wrong) that most modern HDTV sets have no problem doing IVTC on a NTSC source when that is possible so again why should you do IVTC?

If you own an HDTV, you would notice occasional artifacts. The IVTC chipsets are not perfect, and they have very little time to do what they do. They do not have the leisure to look at the video source in detail to find and remove these duplicate fields. A filter process peformed on the source does have the time to do this. If it misses a frame, it can go back and re-examine/fix the issue. Television broadcasts don't allow this.

I would look at your conversion process to find what has created your 'softness' problem. It's not IVTC.

[malducci]

EDIT: Hehe.. didn't realize DJRumpy replied before I did.

Can a HDTV display interlaced video as interlaced or MUST it do some sort of deinterlacing/IVTC/line doubling since it is a progressive only (is it?) display device???

Yes, all sets that I have encountered have the deinterlacer enabled by default or hard wired. Some manufactures have service menus in which a service technician can disabled the deinterlacer, but only for trouble shooting. Although HDTV high resolution is 1080i( some sets only have 720p as high resolution ) and this is an interlaced mode is not evenly divided by 480i and thus not applicable for upconverting AND keeping it interlaced. NOTE: Some Sony units that I was looking at when I was purchasing a HDTV, had a 960i mode which is evenly divided by 480i, but the unit still deinterlaced before upconverting. DVD players do not output 1080i component, they use 480i or 480p component. All HDTV have some sort of progressive scan mode, weather its 480p or 720p or 540p. If the resolution does not match, then its upconverted to the nearest progressive mode ( some sets allow you to set the TV to 1080i for all incomming signals for upconversion).

I'm not sure how the IVTC process works exactly but consider this. An interlaced video signal (that can be IVTC'ed) has 30 frames per second in a 3:2 pattern meaning there are 3 progressive frames followed by 2 interlaced frames followed by 3 progressive frames followed by 2 interlaced frames etc.

So in one second you have only 18 progressive frames and 12 interlaced frames. So when you do IVTC is can't just remove the 12 interlaced frames or you would end up with only 18 progressive frames. Where do the other 6 progressive frames come from to get a frame rate of 24fps? My guess is the IVTC process somehow "creates" these progressive frames using the interlaced frames. Again, my guess is your first 3 progressive frames are frames 1, 2 and 3 and then the 4th and 5th interlaced frames get combined into one progressive frame and this becomes frame 4 in the new progressive video. Then the next 3 progressive frames are used and then the next 2 interlaced frames become one again etc.

First, lets get the 29.97fps out of the way. There really aren't 29.97fps - it's 59.94 fields per second @ 720x240 (262 but 240 is visable). There are no frames in NTSC video, only fields. Computer video editing software works with frames. Since there needs to be some sort of conversion, two fields are captured at the same time giving a 720x480 frame. You could capture each field as a frame and interpolate the 720x240 to 720x480 and get 59.94 frames per second. I don't want to go too indepth here, but it makes for extremely smooth video from true interlaced video if you plan to keep it on the computer.

Each field is either on a even scanline or an odd scanline. Thus it's possible to have 29.97fps progessive frames in a 59.94 field based video, but keep in mind that when a new field is being drawn, the old field is fading away and because most normal TV's overlap the scanlines of each field to further help hide the interlacing artifacts - your 480 vertical resolution has not dropped down to ~ 330+ vertical resolution. This is even worst on VHS where is has a low vertical Luminance resolution, and a lower vertical chrominance resolution. Most people refur to the luminance resolution when describing XXX x XXX (i.e. 720x480 ).

When a DVD player decodes a interlaced 29.97fps video, it seperates the frames back into fields @ 59.94 fields per second. If you had a 29.97fps progressive video and you played it back on a field based TV, then the video would look like this when converted:

ABCDE 'frame A through E in 480 res @ 29.97hz

A-B-C-D-E 'Top field of a frame with 240 res @ 59.94hz
-A-B-C-D-E 'Bottum field of a frame with 240 res @ 59.94hz

The "-" represents time that the field is not being drawn and is actually fading.

OK, all the info out of the way. Now, This is how a 23.976fps progressive movie is telecined to 59.94 fields per second.

First, the original 23.976 movie:

ABCD 'Full frames in 480 res @ 23.976hz

Second, this is how it looks in Telecined fields per second:

A-A-B-C-D- 'Top field of frame in 240 res @ 59.94hz
-A-B-C-C-D 'bottum field of frame in 240 res @ 59.94hz

Third, if we were to capture its as interlaced 29.97 frame per second @ 480 res:

AABCD - field 1
ABCCD - field 2

Since two fields are captured in a single frame in 480 res, field 1 is on the even lines and field 2 is on the odd lines. You can see that Field 1 of frame A is duplicated and field 2 of frame C is duplicated.

A[A]BCD
AB[C]CD

What inverse telecine does is remove the duplicate fields and reorder them.

ABCD
ABCD

Thus when in order, you have full 480 vertical res progressive frames:

ABCD

And since you discarded two fields and reorderd them, you now have 1 less frame. There is a reason why I showed 6 frames or 12 fields in the examples.

The telecine process creates a creates 2 new frames in place of the 1 old frame, in a group of 4 frames. This now gives you a group of 5 frames - multiply this by 6 and you get 30fps. The reason why we say 23.976 or 29.97 is because the 30fps is slowed down by 1%.

Eventhough 2 new frames were created in the telecined process, the original frame was not destroyed - just a bit of shifting and field duplicating.

I can't really think of why you are experiencing problems or softening of the video. If your capture card captures video with the field order reversed, then you would HAVE to reversed the field order before running IVTC.

My guess is the original 2 interlaced frames becomming one progressive frame is a process that is less than perfect ... hence the reason why a 29.97fps video IVTC'ed to 23.976fps does not look as sharp as a studio mastered progressive source which is a straight conversion ... each frame of film is a progressive frame in the video.

Actually, nothing is lossed when deleting the dup fields and reodering them because all you are doing is reordering them - nothing else - no blending, interpolating, or resizing. You almost got it right: a studio mastered 23.976fps DVD movie with 2:3 pulldown flags is superior to any analog captured video for the fact that the film is scanned at its full color resolution and pixel resolution. There are MANY other factors, but IVTC is not one of them.

To go in depth about understanding the inner working of NTSC analog video signal, luminance resolution versus chrominace resolution, noise and interferrence effecting the signal and color subcarrier signal - would be beyond the scope of the thread.

There are so many conversions that happen before the signal is received at your destination - even if you have digital cable or satellite. IVTC is just a single step in repairing the artifacts of these conversions. And don't forget the 20% savings in bandwidth going from 29.97 to 23.976. If your running filters to clean up noise, color shifting, etc - then you just saved some workload and some filters work better in progressive scan.

Oh yeah, not to mention that when you capture video - it should be resized before encoding, to compensate(sp) for the overscan area of the TV. Every brand DVD player is different and some put more of the picture in the over scan area. If your video is partially interlaced because of the telecined process, you can not resize the video ( well, its HIGHLY recommend not to resize interlaced video ). If you have done VHS to VCD, SVCD/CVD, or DVD then you'll know what I'm talking about ( overscan correction ).

I also understand (I could be wrong) that most modern HDTV sets have no problem doing IVTC on a NTSC source when that is possible so again why should you do IVTC? And yes I understand the less info to encode stuff but still that doesn't make up for the softness of the IVTC process.

Yeah, most newer HDTVs have built in IVTC when put in the right mode, but why waste the extra 20% bitrate when it can be used to fit more video at the same quality or same duration of video but with higher bitrate for better quality

Before I bought my HDTV - I had a 32' normal TV and the video didn't look softer to me when IVTC was applied.( and I'm anal about video ).

But hey, if your capturing video with less than 480 lines of vertical resolution or your capturing with a mpeg codec with ATI progy, Cyberlinks progy - then you can throw this out the window because this does not apply to IVTC.

Whoa, that was a lot of type...

See ya.

[puddy]

Hope I'm not to late to jump in here and get a little advice from you obviously expert folks.

I'm planning on buying the Mitsubishi 65" (the one they always have in the Best Buy fliers) or the 73" version which Best Buy doesn't carry.

My question is, in anticipation of owning this TV combined with a progressive scan DVD, how should I encode my captures?

In some cases I'm making SVCD's for things like 30 min episodes of kids programs taken as TV captures . Currently I'm capturing at 480x480, no 3:2 pulldown, no deinterlacing, MPEG2. Then I do some quick trimming in TMPGEnc using the MPEG tools and then burn to CDR using the SVCD template in the newest Nero.

In other cases I'll be doing VHS and Hi8 captures to be authored for DVD.

All are NTSC.

I guess I'm not sure if you guys are saying don't deinterlace or pulldown or anything during capture or afterwards as long as you are not resizing because the hardware (TV and/or DVD player) will take care of making it "progressive" for nice, smooth display on the widescreen.

Is that the correct assumption?

Man, I've confused myself with my own question.

I just want to know what's the best to do for TV caps and regular video caps (not from a DV source) in anticipation of having the Mitsubishi widescreen and a progressive scan DVD.

Thanks!
Puddy

[DJRumpy]

I never deinterlace. It just looks bad. Interlaced encodes will look fine on your tv (HDTV, or regular TV).

If your captures are telecined (you didn't say), then you should perform IVTC, and then simply add the pulldown flags to make it appear as 29.97. This type of encode will playback as progressive when combined with a progressive scan player,component outs, and a progressive capable television. The component inputs may not be a requirement...that depends on your television, and dvd player.

If your source is interlaced, leave it that way. If your not sure how to identify the different types, look here: www.lukesvideo.com

[malducci]

Yeah, if you keep the source interlaced when you encode - the progressive DVD player ( with component out ) will more than likely deinterlace it for you and if it doesn't, then the HDTV will. Quick and simple and the conversion is done for you. All assumeing that the source is primarily interlaced.

But, and I do mean BUT...

What kind of deinterlacer does you DVD player have? Some DVD players with just interpolate one field. Are you happy with the method a DVD player deinterlaces the video? More than likely the DVD player will do the de-interlacing and not the TV since it will convert to pregressive scan and is first on the chain.

I recommend reading the following links.

http://www.hometheaterhifi.com/volume_10_1/dvd-benchmark-guide-to-progressive-scan-sho...ut-1-2003.html

and

http://www.hometheaterhifi.com/volume_7_4/dvd-benchmark-part-5-progressive-10-2000.html

65" HDTV is pretty big, but 73" is huge and will show more imperfections - point being make sure your setup has the best deinterlacer available. I have a Hitachi 53" HDTV and personally rather use a combonation of smart filters to do the deinterlacing to acheive the best quality. Then again the video I'm converting is shot using a DVCPRO and shot as if it was film - converted to progressive scan and converted to 24fps ( very heated debut on some forums ) - so my prioritys are different than most.

[puddy]

So, I think what you are saying is that unless I'm a total freak about quality (and I'm not quite to that point yet) then just let the DVD or the TV deinterlace and/or do the 3:2 pulldown stuff and I'll be fine.

I guess I got confused because a lot of the software wants to deinterlace for you automatically. I think it assumes you plan on watching it on a computer monitor? And since a progressive scan TV has characteristics like a computer monitor, I just wasn't sure.

Hey, thanks again. This place has been such a cool experience. There is so much knowledge and nice people here. Thanks.

Puddy

[DJRumpy]

So, I think what you are saying is that unless I'm a total freak about quality (and I'm not quite to that point yet) then just let the DVD or the TV deinterlace and/or do the 3:2 pulldown stuff and I'll be fine.

Exactly. I only suggest doing IVTC on telecined material, as it can improve the output, especially on low bitrate formats like VCD/CVD/SVCD, and even that is optional. If your not too concerned, leave it the way it is, and just enjoy

[hpv]

Keep the video interlaced. The job of a progressive scan DVD player is to do the de-interlacing for you. That's what it does. All commercial DVD's are 480i (i = interlaced). The progressive scan DVD player takes this data and converts it to a 480p signal (p = progressive).

Also, if your progressive scan DVD player has 3:2 pulldown, then you don't have to worry about the FPS. It'll take 29.97 and bring it down to 24 for you.

A progressive scan DVD player plays an interlaced DVD at 30 fps by pushing out the two fields of the frame at the same time. It does NOT convert the video to 24fps if that is what you mean by "de-interlace". If you notice, the ATSC (that's HDTV standard, kids) is 30fps progressive. It doesn't change the frames or undo the interlacing. It just shoves it out all together so that you don't get the flicker you normally would from interlacing as the fields aren't being drawn in alternating odd and even lines, but in consecutive lines. That's ALL progressive scan DVD players are going to do for you.

[malducci]

If you notice, the ATSC (that's HDTV standard, kids) is 30fps progressive. It doesn't change the frames or undo the interlacing. It just shoves it out all together so that you don't get the flicker you normally would from interlacing as the fields aren't being drawn in alternating odd and even lines, but in consecutive lines. That's ALL progressive scan DVD players are going to do for you.

Actually the ATSC standard has nothing to do with the deinterlacing process. Either the DVD player or TV will convert the 480i into a 480p signal if either are progressive capable ( requires progressive component interface ). If the DVD player does not deinterlace the video, then the TV will. And yes, there is more than 1 way to deinterlace video. Not every HDTV or progressive scan DVD player use the same deinterlacing algoritham(s). So, yes there are progressive scan DVD players that will deinterlace a interlaced video internally.

Not sure what you mean "shoving it all out together" - I think your confusing 29.97p video inside a 59.94i signal with field based video. We are talking about 29.97i fps which is really 59.94 fields per second. If you put two fields together, you get interlaced video frames - so just "shoving it all out together" would not deinterlace a 29.97i video.

BTW, The video standards and what HDTV's are capable of are two different things - my Hitachi can do => 480p (upconverted 540p or 1080i) @ 60fps.