moiré effect on pdx/trv950 at DVinfo.net

Love Mov · April 23rd, 2004, 07:03 PM

The worst thing I don't like my older camcorder is the moiré effect, (jaggle line or grid lines in some pattern when paning)..well, not sure what's exactly the name should be, I hope you know what I am saying...
Home video looks bad because of this, never see such effect on a DVD, although the resolution are the same, leave me wonder if the problem is DV format itself or camera.
Any thoughts?

Ignacio Rodriguez · April 23rd, 2004, 09:58 PM

Yes. Minimize edge enhancment and your PDX10 should be much better in this sense (softer jaggies) than your average DV camera because of the higher resolution of it's CCD array.

Love Mov · April 23rd, 2004, 11:06 PM

So, you mean lower the sharpness? Or other parameters?
Thanks.

Shawn Mielke · April 23rd, 2004, 11:39 PM

Precisely. Having lowered Sharpness to it's minimum several months ago and for good, my attention is rarely drawn to jaggies anymore.

Mike Sanchez · April 24th, 2004, 03:17 AM

Shawn and Ignacio are correct. Lowering the sharpness down to its lowest setting will eliminate the Moire that I can see in the TRV-950 when S-video to the TV.

However, you don't see this in DVD outcomes because, as far as I can tell, all AVI to MPEG2 converters use a low pass filter on the video during rendering. I am not sure about the high end like Canopus (too much money for me) where you may be able to turn off or customize the low pass nature of the conversion.

Anyway, for Pinnacle I noticed that, in lowish light, frame grabs from the DVD compared to the AVI were significantly lower noise and signficantly softer (but look better). This is indirect indication of the lowpass filter effect (and why Moire is absent in the Pinnacle rendered outcome). I ran a series of experiments with my cam sharpening, then rendered to find the best settings POST RENDERING.

At the moment, I set the TRV-950 down one notch during capture for sharpening. Pinnacle low pass then softens it a bit more and boom, sharp enough outcomes, low noise and no Moire.

So, when setting the sharpness setting you may want to do it as a system experiment in conjunction with your rendering platform.

It is a good thing you see Moire though. This is the equivalent of aliasing of high frequency components. Meaning.....the TRV-950 and PDX-10 appear to have a higher resolution than NTSC video....not a bad thing.

Ignacio Rodriguez · April 26th, 2004, 06:15 PM

> It is a good thing you see Moire though. This is the equivalent
> of aliasing of high frequency components. Meaning.....the TRV-
> 950 and PDX-10 appear to have a higher resolution than NTSC
> video....not a bad thing.

Hmm... but I thought the downsampling process in itself should act as a good LPF... thus most moire must come from the sharpening algorithm. It has been commented on several threads that the PDX10's lowest 'sharpness' setting seems to be the most natural setting. So like Shawn I have kept it at minimum.

Still, it makes a lot of sense to check your final output when looking for the ideal settings, for this and other parameters. I guess it's not easy to get a TV station to output a siginal from your camera while you tweak the setting though, which is what I would have to do since most of my current work is for a TV program :S

Anyway I will try now setting it to a little higher than minimum and see what it comes out like when broadcast.

Way to go, Mike!

Mike Sanchez · April 28th, 2004, 01:52 AM

Ignacio,

Thank you and you must be correct that the downsampling procedure would remove frequencies above NTSC. Pretty much has to I guess, by definition. So, the sharpening must be the cuprit and, since the Moire goes completely away at minimum sharpening when my cam goes direct to TV this is another indicator that it is just the sharpening.

Don't see why a big TV station cannot broadcast your experiment to all of us.....I'm in!

Ignacio Rodriguez · April 28th, 2004, 09:46 AM

> Don't see why a big TV station cannot broadcast
> your experiment to all of us.....I'm in!

Jajja, I am pretty far away in a hidden corner of the world so it would have to be a really big big TV network for that to happen. Now if by chance know of a vacancy at the local branch of CNN... <grin>

April 23rd, 2004, 07:03 PM	#1
Love Mov Regular Crew Join Date: Apr 2004 Posts: 97	moiré effect on pdx/trv950 The worst thing I don't like my older camcorder is the moiré effect, (jaggle line or grid lines in some pattern when paning)..well, not sure what's exactly the name should be, I hope you know what I am saying... Home video looks bad because of this, never see such effect on a DVD, although the resolution are the same, leave me wonder if the problem is DV format itself or camera. Any thoughts?

April 23rd, 2004, 09:58 PM	#2
Ignacio Rodriguez Contributor Join Date: Sep 2003 Location: Santiago, Chile Posts: 932	Yes. Minimize edge enhancment and your PDX10 should be much better in this sense (softer jaggies) than your average DV camera because of the higher resolution of it's CCD array. __________________ Ignacio Rodríguez in the third world. @micronauta on Twitter. Main hardware: brain, eyes, hands.

April 23rd, 2004, 11:39 PM	#4
Shawn Mielke Major Player Join Date: Jul 2003 Location: Santa Rosa, California Posts: 745	Precisely. Having lowered Sharpness to it's minimum several months ago and for good, my attention is rarely drawn to jaggies anymore. __________________ Breakthrough In Grey Room

April 24th, 2004, 03:17 AM	#5
Mike Sanchez Regular Crew Join Date: Jan 2004 Location: Rochester, NY (USA) Posts: 54	Shawn and Ignacio are correct. Lowering the sharpness down to its lowest setting will eliminate the Moire that I can see in the TRV-950 when S-video to the TV. However, you don't see this in DVD outcomes because, as far as I can tell, all AVI to MPEG2 converters use a low pass filter on the video during rendering. I am not sure about the high end like Canopus (too much money for me) where you may be able to turn off or customize the low pass nature of the conversion. Anyway, for Pinnacle I noticed that, in lowish light, frame grabs from the DVD compared to the AVI were significantly lower noise and signficantly softer (but look better). This is indirect indication of the lowpass filter effect (and why Moire is absent in the Pinnacle rendered outcome). I ran a series of experiments with my cam sharpening, then rendered to find the best settings POST RENDERING. At the moment, I set the TRV-950 down one notch during capture for sharpening. Pinnacle low pass then softens it a bit more and boom, sharp enough outcomes, low noise and no Moire. So, when setting the sharpness setting you may want to do it as a system experiment in conjunction with your rendering platform. It is a good thing you see Moire though. This is the equivalent of aliasing of high frequency components. Meaning.....the TRV-950 and PDX-10 appear to have a higher resolution than NTSC video....not a bad thing. __________________ New to Video but Learning Fast. Sony TRV-950 Canon ES-970 Canon Elan Canon AE-1 Canon Z-135 Yashica Mat124G

April 26th, 2004, 06:15 PM	#6
Ignacio Rodriguez Contributor Join Date: Sep 2003 Location: Santiago, Chile Posts: 932	> It is a good thing you see Moire though. This is the equivalent > of aliasing of high frequency components. Meaning.....the TRV- > 950 and PDX-10 appear to have a higher resolution than NTSC > video....not a bad thing. Hmm... but I thought the downsampling process in itself should act as a good LPF... thus most moire must come from the sharpening algorithm. It has been commented on several threads that the PDX10's lowest 'sharpness' setting seems to be the most natural setting. So like Shawn I have kept it at minimum. Still, it makes a lot of sense to check your final output when looking for the ideal settings, for this and other parameters. I guess it's not easy to get a TV station to output a siginal from your camera while you tweak the setting though, which is what I would have to do since most of my current work is for a TV program :S Anyway I will try now setting it to a little higher than minimum and see what it comes out like when broadcast. Way to go, Mike! __________________ Ignacio Rodríguez in the third world. @micronauta on Twitter. Main hardware: brain, eyes, hands.

April 23rd, 2004, 11:06 PM	#3
Love Mov Regular Crew Join Date: Apr 2004 Posts: 97	So, you mean lower the sharpness? Or other parameters? Thanks.

April 28th, 2004, 01:52 AM	#7
Mike Sanchez Regular Crew Join Date: Jan 2004 Location: Rochester, NY (USA) Posts: 54	Ignacio, Thank you and you must be correct that the downsampling procedure would remove frequencies above NTSC. Pretty much has to I guess, by definition. So, the sharpening must be the cuprit and, since the Moire goes completely away at minimum sharpening when my cam goes direct to TV this is another indicator that it is just the sharpening. Don't see why a big TV station cannot broadcast your experiment to all of us.....I'm in! __________________ New to Video but Learning Fast. Sony TRV-950 Canon ES-970 Canon Elan Canon AE-1 Canon Z-135 Yashica Mat124G

April 28th, 2004, 09:46 AM	#8
Ignacio Rodriguez Contributor Join Date: Sep 2003 Location: Santiago, Chile Posts: 932	> Don't see why a big TV station cannot broadcast > your experiment to all of us.....I'm in! Jajja, I am pretty far away in a hidden corner of the world so it would have to be a really big big TV network for that to happen. Now if by chance know of a vacancy at the local branch of CNN... <grin> __________________ Ignacio Rodríguez in the third world. @micronauta on Twitter. Main hardware: brain, eyes, hands.