XL H1 broadcast quality?

Vince Gaffney · February 11th, 2007, 06:29 PM

Quote:

Originally Posted by Steve Rosen

So the drill is to capture via component rather than FW... Ok-a-y....

HD SDI is even better. Although I'm currently ingesting from component.

vince

Steve Rosen · February 11th, 2007, 07:18 PM

Yeah, that's what I'll have to do (HD SDI) because my DeckLink HD Pro card doesn't have Component in.. which means I'm going to have to use the H1 as a deck, which I'm not too happy about...

I'm starting shooting a feature length documentary next month that is going to be labor intensive. I will be shooting over 100 tapes, so that's a lot of wear and tear on my H1... and I will be shooting 24f - I had hoped to use my A1 or buy an HV10 as the deck, but this changes things..

Do you know if this can be done effectively (converting 24f m2t files to DVPRO HD or whatever) from a Firestore?

Kevin Munn · February 11th, 2007, 08:13 PM

If you can wait for the hv-20 it has hdmi out, so you could capture through that rather than using the h1 w/ hd-sdi.

Marty Hudzik · February 11th, 2007, 09:03 PM

Quote:

Originally Posted by Kevin Munn

If you can wait for the hv-20 it has hdmi out, so you could capture through that rather than using the h1 w/ hd-sdi.

Forgive me for the ignorant question, but what benefit does capturing via HDMI out from an HV20 have if you are using it to play previously recorded HDV material from tape? Other than convenience? I could see a live feed going to the PC/MAC to be encoded directly to a superior codec....but once it goes to tape as HDV is there any real "quality" benefit to capturing vis SDI or HDMI? Isn't the color information gone at this point and the compression already in place? If so why not just convert to Intermediate codec after capture?

The only potential benefit I can see is real-time conversion as it is ingested to the PC/MAC to a better codec. However I find that Aspect HD on the PC converts HDV to Cineform files in almost realtime. In fact, I actually end up with the original .m2t files along with the Cineform .avi files. I can easily archive the .m2t as they are much more "size friendly" and never have to go back to the original tape, unless there is a real emergency.

Si am I missing something here?

Steve Rosen · February 13th, 2007, 08:16 AM

Kevin: Yeah, well, the HV-20 advertises 24p... if that means what I think it means, it is the end of 24f...

Marty: See, this is just what I mean when I say I get confused by this conversation... I would assume that HDV is HDV and that other than avoiding additional compression generations there would be no particular advantage..

But I have read many threads warning against editing native HDV - I have done it successfully, as I said above, for a year, but if there is a better way I want to know what it is...

And if I do chose to batch capture HDV into FCP conventionally, how do I batch convert those files to AIC or DVPRO HD? Do I have to sit there and do it one clip at a time?

John Richard · February 13th, 2007, 10:22 AM

"Yeah, well, the HV-20 advertises 24p... if that means what I think it means, it is the end of 24f..."

Not sure that this concern is warranted. I believe when the H1 was introduced there was a patent issue with the person who developed the method of getting 24p out the video stream - so Canon opted for the legally safe 24f nomenclature?? I remember reading that someowhere. Now maybe that issue has been resolved and they can now safely call 24f - 24p.

Chris ????

Tony Tremble · February 15th, 2007, 03:06 AM

The 24F is dead from the point of view that the HV20 is natively progressive as it uses CMOS chip and thus will record 24P.

The F in 24F/25F/30F only really refers to the way the XL-H1 and XH-A1/G1s produce their progressive footage from interlaced CCDs and has little to do with the legalities of using 24P as a nomenclature it has more to do with being honest with the consumer. Which I applaud.

As we all can see 24F is 24P no matter how it is generated.

TT

Andrew Davies · February 15th, 2007, 05:09 AM

Does the interlaced nature of the 24F cause artifact problems with fast pans?

Tony Tremble · February 15th, 2007, 05:54 AM

Quote:

Originally Posted by Andrew Davies

Does the interlaced nature of the 24F cause artifact problems with fast pans?

Not at all.

There are no interlace artefacts at all just a slight drop in resolution. There is a slight stair stepping of bright diagonal edges but no worse than on the HVR-V1.

The "F" just means the progressive frames were cleverly created from and interlaced CCD and _not_ that it is faux progressive.

The green ccd is clocked a field out of sync to the red and blue ccds and then the dsp gets to work reconstructing a full colour progressive frame. Think of it as temporal pixel shifting.

Canon can do this because they are using very high res CCDs in the first place. Other manufacturers are using lower res CCDs and CMOS chips and even though Canon lose 10% of res in progressive mode the outcome is still as good if not better than the competition.

F=P

TT

Steve Rosen · February 15th, 2007, 08:18 AM

The point being, however, that the HV20 will probably not serve well as a deck for footage shot 24f w/ the H1/A1/G1... Many have been using the Hv10 for that purpose.. hopefully that little camera will not go away immediately..

Also, it seems likely that the much contemplated successor to the H1 will have progressive CMOS chips and offer true 24p and at even higher resolution than 1080i...

Chris Hurd · February 15th, 2007, 08:34 AM

Quote:

Originally Posted by Steve Rosen

the HV20 will probably not serve well as a deck for footage shot 24f w/ the H1/A1/G1... Many have been using the Hv10 for that purpose.. hopefully that little camera will not go away immediately

What makes you say that, Steve? The HV10 and HV20 share the exact same tape transport. There's no absolutely difference between them as far as playback capabilities are concerned, and there's no logical reason why an HV20 wouldn't serve well as an HV10, as a deck for footage shot in Frame mode from the bigger Canon camcorders.

Quote:

it seems likely that the much contemplated successor to the H1 will have... even higher resolution than 1080i...

Then it wouldn't be HDV. It's still a little early for that yet.

Tom Roper · February 15th, 2007, 09:03 AM

Quote:

Originally Posted by Tony Tremble

The 24F is dead from the point of view that the HV20 is natively progressive as it uses CMOS chip and thus will record 24P.

The F in 24F/25F/30F only really refers to the way the XL-H1 and XH-A1/G1s produce their progressive footage from interlaced CCDs and has little to do with the legalities of using 24P as a nomenclature it has more to do with being honest with the consumer. Which I applaud.

As we all can see 24F is 24P no matter how it is generated.

TT

I think honest misinformation takes on a life of its own. Not that Tony is wrong, but theories abound:

1.) CMOS pixels are byte addressable
2.) That this or that sensor is natively progressive

I won't accept these distinctions until proved. As to the first point, who really has any earthly idea how a chip is scanned, i.e. shift register versus byte addressable? It's one thing when you have a product data sheet on it, which in the case of proprietary chips made by Canon themselves for their consumer products, you don't! Until proved otherwise, (and I'd love to be) this information just comes down the same old marketing pipeline.

As to the second point, it flies in the face of logic that so called natively progressive sensors find first applications in interlaced cams, witness FX7 and HV10.

I think at most what could be said is that a chip is inherently neither progressive or interlaced. It's just what is done with it.

Tony Tremble · February 15th, 2007, 09:59 AM

Tom

There are progressive CCDs and interlaced CCDs used in HVX 200 and XL-H1 respectively. How they create non-native streams is down to the intellectual wherewithal of the manufacturer.

You were right to pick me up on saying that CMOS are natively progressive. I should have qualified what I meant to avoid ambiguity. Clearly Canon differentiate between 24F and 24P themselves. I suspect the HV20 will scan its CMOS chips progressively when recording 24P thus Canon use the P. 24F being derived from interlaced CCDs they feel unable to label it 24P progressive even though it is.

In 60i/50i I would expect the HV20 CMOS chips to be scanned interlaced. But there are several ways to skin a cat in that regard. It could scan 60P/50P and leave the interlacing to the encoder. Who knows?

TT

John Benton · February 15th, 2007, 10:20 AM

Sorry to bring this up again, buy last year a Canon Rep told me at a Canon showing off of the H1 that you could record to tape and yet capturing through the HD-SDI port to another codec would allow for more robust info to come in than normal HDV.
Now, some people have claimed this, but this has generally been shot down by everyone who knows anything.
I have a feeling this workflow was what the person was referring to

Chris Hurd · February 15th, 2007, 10:52 AM

Quote:

Originally Posted by John Benton

...record to tape and yet capturing through the HD-SDI port to another codec would allow for more robust info to come in than normal HDV.

Scott Billups told me face-to-face about this method and swears by it.

February 15th, 2007, 10:20 AM	#104
John Benton Major Player Join Date: Oct 2005 Location: NYC Posts: 658	Boy, I hate to open up a can of worms, but... Sorry to bring this up again, buy last year a Canon Rep told me at a Canon showing off of the H1 that you could record to tape and yet capturing through the HD-SDI port to another codec would allow for more robust info to come in than normal HDV. Now, some people have claimed this, but this has generally been shot down by everyone who knows anything. I have a feeling this workflow was what the person was referring to

February 11th, 2007, 07:18 PM	#92
Steve Rosen Major Player Join Date: Dec 2005 Location: Monterey, California Posts: 895	Yeah, that's what I'll have to do (HD SDI) because my DeckLink HD Pro card doesn't have Component in.. which means I'm going to have to use the H1 as a deck, which I'm not too happy about... I'm starting shooting a feature length documentary next month that is going to be labor intensive. I will be shooting over 100 tapes, so that's a lot of wear and tear on my H1... and I will be shooting 24f - I had hoped to use my A1 or buy an HV10 as the deck, but this changes things.. Do you know if this can be done effectively (converting 24f m2t files to DVPRO HD or whatever) from a Firestore?

February 11th, 2007, 08:13 PM	#93
Kevin Munn New Boot Join Date: Jan 2005 Location: Southington, CT Posts: 19	If you can wait for the hv-20 it has hdmi out, so you could capture through that rather than using the h1 w/ hd-sdi.

February 13th, 2007, 08:16 AM	#95
Steve Rosen Major Player Join Date: Dec 2005 Location: Monterey, California Posts: 895	Kevin: Yeah, well, the HV-20 advertises 24p... if that means what I think it means, it is the end of 24f... Marty: See, this is just what I mean when I say I get confused by this conversation... I would assume that HDV is HDV and that other than avoiding additional compression generations there would be no particular advantage.. But I have read many threads warning against editing native HDV - I have done it successfully, as I said above, for a year, but if there is a better way I want to know what it is... And if I do chose to batch capture HDV into FCP conventionally, how do I batch convert those files to AIC or DVPRO HD? Do I have to sit there and do it one clip at a time?

February 13th, 2007, 10:22 AM	#96
John Richard Major Player Join Date: May 2006 Location: Incline Village, Nevada Posts: 604	"Yeah, well, the HV-20 advertises 24p... if that means what I think it means, it is the end of 24f..." Not sure that this concern is warranted. I believe when the H1 was introduced there was a patent issue with the person who developed the method of getting 24p out the video stream - so Canon opted for the legally safe 24f nomenclature?? I remember reading that someowhere. Now maybe that issue has been resolved and they can now safely call 24f - 24p. Chris ????

February 15th, 2007, 03:06 AM	#97
Tony Tremble Regular Crew Join Date: Oct 2006 Location: United Kingdom Posts: 451	The 24F is dead from the point of view that the HV20 is natively progressive as it uses CMOS chip and thus will record 24P. The F in 24F/25F/30F only really refers to the way the XL-H1 and XH-A1/G1s produce their progressive footage from interlaced CCDs and has little to do with the legalities of using 24P as a nomenclature it has more to do with being honest with the consumer. Which I applaud. As we all can see 24F is 24P no matter how it is generated. TT

February 15th, 2007, 05:09 AM	#98
Andrew Davies Regular Crew Join Date: Jan 2007 Location: Wales Posts: 104	Does the interlaced nature of the 24F cause artifact problems with fast pans?

February 15th, 2007, 08:18 AM	#100
Steve Rosen Major Player Join Date: Dec 2005 Location: Monterey, California Posts: 895	The point being, however, that the HV20 will probably not serve well as a deck for footage shot 24f w/ the H1/A1/G1... Many have been using the Hv10 for that purpose.. hopefully that little camera will not go away immediately.. Also, it seems likely that the much contemplated successor to the H1 will have progressive CMOS chips and offer true 24p and at even higher resolution than 1080i...

February 15th, 2007, 09:59 AM	#103
Tony Tremble Regular Crew Join Date: Oct 2006 Location: United Kingdom Posts: 451	Tom There are progressive CCDs and interlaced CCDs used in HVX 200 and XL-H1 respectively. How they create non-native streams is down to the intellectual wherewithal of the manufacturer. You were right to pick me up on saying that CMOS are natively progressive. I should have qualified what I meant to avoid ambiguity. Clearly Canon differentiate between 24F and 24P themselves. I suspect the HV20 will scan its CMOS chips progressively when recording 24P thus Canon use the P. 24F being derived from interlaced CCDs they feel unable to label it 24P progressive even though it is. In 60i/50i I would expect the HV20 CMOS chips to be scanned interlaced. But there are several ways to skin a cat in that regard. It could scan 60P/50P and leave the interlacing to the encoder. Who knows? TT