but it ain't me missing something funamental.

The ONLY compression being performed is to enable the data stream from the sensors to be written to the tape with the bandwidth limitation that applies.

Neither the analogue Component NOR the Digital Firewire O/P's data streams are compressed, unless their source is the already compressed data on the tape (ie: in playback mode).

Ergo, both sources are uncompressed, as there is no reason for them to be so. The only compression being applied is to get the data onto the tape.

So, why go from the analogue Component O/P's when you can go Digital Firewire with embedded audio?

CS

I haven't worked with ProRes myself yet, but if I remember correctly from the Apple presentation, you can't capture in prores realtime, so if that is correct, FCP has to convert it to prores after capture.
The Aja io hd converts to prores with dedicated hardware.

but it ain't me missing something funamental.

The ONLY compression being performed is to enable the data stream from the sensors to be written to the tape with the bandwidth limitation that applies.

Neither the analogue Component NOR the Digital Firewire O/P's data streams are compressed, unless their source is the already compressed data on the tape (ie: in playback mode).

Ergo, both sources are uncompressed, as there is no reason for them to be so. The only compression being applied is to get the data onto the tape.

So, why go from the analogue Component O/P's when you can go Digital Firewire with embedded audio?

CS

Chris, whatever comes out of the Firewire plug, IS compressed. In an HDV camcorder, what comes out of the firewire plug, whether it's in playback or record mode, is an HDV stream, hence a compressed data stream.

Neither the analogue Component NOR the Digital Firewire O/P's data streams are compressed, unless their source is the already compressed data on the tape (ie: in playback mode).


Chris,

Sergio already corrected you, but I wanted to add that I *wish* that's how it worked. There is no technical reason that manufacturers have to compress live footage to HDV when sending it out via firewire, but they all do it, and none provide any other option. HDV only uses 4.5 MB/s, while firewire (400) has the capacity to carry much, much more data: 48 MB/s. That's not enough for uncompressed 4:2:2 1080, but there are a lot of options. In any case, it would compete directly with ultra-high-margin HD-SDI, and they would need to find a codec that's light on the camera's tiny CPU, so it will probably never happen.

Why no use the BlackMagic Intensity Pro hardware for approx $350?

http://www.blackmagic-design.com/products/intensity/

GAry

Why no use the BlackMagic Intensity Pro hardware for approx $350?

http://www.blackmagic-design.com/products/intensity/

GAry

And then you won't have to use the RCA:BNC adapter cable and RCA Coupler. YES, it will degrade signal slightly, but more importantly, all of those connections are not shielded and can let interference come into play. One connection is better than 3 in your design!

The BlackMagic Intensity Pro sounds great! I am definitely looking for a cheaper solution.

However, I'm looking for a cheaper solution to capture the highest possible quality from my A1.

What connection could I use from the A1 to connect to the Intensity Pro?

Is there such a thing as a Terminal-D Component to HDMI cable?

Or, are there adapters that would allow me to connect the 3 component cables to the HDMI input?

The BlackMagic Intensity Pro sounds great!

What connection could I use from the A1 to connect to the Intensity Pro?

Or, are there adapters that would allow me to connect the 3 component cables to the HDMI input?

Read up and look at the product page... it comes with a breakout cable that allows hookup of the component cables.
http://www.blackmagic-design.com/products/intensity/

I am aware of the HDV 4:2:0 colorspace limitations from XH-A1 captured via fire-wire. I am interested in learning any options for capturing 4:2:2 footage from XH-A1. Do any of the engineer people who've looked at the A1 think the design of the makes it possible to have an add-on jack pack be a component offered by Canon in the future, to upgrade to the functions currently available in the XH-G1? I do know that the camera is ultimately producing a full uncompressed signal, but I'm unsure if it is possible to output it.

I do have an HV-20, and will be setting up a mac pro and black magic intensity pro for capturing 4:2:2 from it in studio via HDMI, primarily for green-screen work. I would ultimately expect to see portable solutions for capturing 4:2:2 HD footage from these cameras directly to disk, bypassing DV compression, hopefully soon.

I'd still like to hear what options people have developed for themselves, and what new equipment they expect will help in portable uncompressed HD capture options or working with some of the newer codecs and capturing techniques.

thanks.

Read up and look at the product page... it comes with a breakout cable that allows hookup of the component cables.
http://www.blackmagic-design.com/products/intensity/
The Intensity has Analog output only. If you want a cheaper alternative to the Kona 3, then the Decklink HD at around $900 is a cheaper alternative.
It works really well for analog capture from the A1 as well as HDSDI capture from the H1.

The Intensity has Analog output only. If you want a cheaper alternative to the Kona 3, then the Decklink HD at around $900 is a cheaper alternative.


You are mistaken. The Intensity has no analog connections at all, it's HDMI only. The Intensity Pro has Analog Component output *and* input. The big difference between the $350 Itensity Pro and it's thrice-expensive brother, Decklink HD, is 8-bit 4:2:2 vs. 12-bit 4:4:4.

As to Chris Carroll's question about portability, there are two options:

1. A portable desktop:

http://www.dvinfo.net/conf/showthread.php?t=97526

2. Laptop with the Magma ExpressBox ($730):

http://dvinfo.net/conf/showthread.php?t=99758

You are mistaken.
Actually..we are both wrong. :)

The Decklink HD is 10bit 4:2:2

Mondays... :)

-will

Excuse me for getting involved in the discussion. The whole discussion is based on the assumtion that the video quality through the component out is of higher quality that the digital HDV stream. How do you know that Canon designed the A1 to output uncompressed video through the component out?
Have someone verified this by checking artifacts by sampling the component out and comparing with the HDV stream?
In an electronics design point of view it would be just as likely (or more likely)that the component out, CVBS out, audio out, and display out were connected after the HDV decompressor.

The Firewire data stream is an HDV (MPEG-2, 25 MBit) or DV (MJPEG, 25 MBit) stream regardless if you are storing to tape or not. The MPEG2 transport stream does not have enough bandwidth to be uncompressed. In fact Firewire does not have the bandwith for uncompressed HD video. Thats why Canon sells the G1.

When using 24F/25F the output from the camera is delayed half a second compared to reality (both audio, video out and display). Why would it be delayed if the camera video output were uncompressed? If the component video output were to be delayed half a second and still be in sync with the audio the camera would need to buffer half a second of uncompressed video. This is not likely since it would just add unneccessary hardware and increase the cost of the product.

Thanks

How do you know that Canon designed the A1 to output uncompressed video through the component out?


I've tested all of the modes (24F, 30F, and 60i) empirically, and they are indeed live. Therefore, they can't be compressed because HDV uses a GOP of at least 15 frames. The fact has been confirmed by many forum members.


In an electronics design point of view it would be just as likely (or more likely)that the component out, CVBS out, audio out, and display out were connected after the HDV decompressor.


The audio is live, as it should be, so that you can monitor and react to what's happening in real time. The viewfinder is very close to live in 30F and 60i modes, but it slows down a bit in 24F mode.

I don't know enough about electronics design to qualify what's easier, but I think the fact that *all* cameras with component output are uncompressed lends credence to the position that it's cheaper or nearly equivalent to make them that way.

[Now that I think about it, compression delay on component output would invalidate its ostensibly principle purpose: monitoring, so that's probably the most important reason.]


When using 24F/25F the output from the camera is delayed half a second compared to reality (both audio, video out and display).


The component output and audio out are *not* delayed in 24F or any other mode. The viewfinder is delayed in 24F. The firewire stream is delayed in all modes.

The component output totally ignores the 24F/30F/60i mode setting: it always pulls 60i off the sensor: there's not even any pull down.

That is exactly what we have seen as well when doing live captures
from analog and HD-SDI from a H1. 60i all the time. No compression.

Hi Everybody,

First post to this great forum - this is a topic of particular interest:

Trying to clarify the information here - Is it feasible to use the component out and feed it into, for instance the AJA io by means of adapters? Like others, my goal is to get uncompressed out without the expense of G1 or H1, ideally with portability too (MacBook Pro).

In the UK at least the io + A1 would be equivalent to the price of H1, but with more flexibility.

Thanks,

Nick.

s it feasible to use the component out and feed it into, for instance the AJA io by means of adapters?
You don't need adapters with the IO-HD. It has component input.

Many thanks Will

Okay, so I'm confused.

With the BlackMagic Intensity, are you able to capture HDV video, then come home and upload it to your computer uncompressed, or does the A1 need to be connected to the computer while you're filming?

It must be connected to the computer.

I have read some discussion on this and seen some comparison stills. But, I was wondering if anyone who has captured 4.2.2 out of their A1 via black magic or Kona cards has an opinion. The few stills that I saw did look slightly more colorful and brighter, but my initial opinion was it was very subtle and maybe not worth the hassle. I am into quality and I know that we are talking about 25% more color with 4.2.2, so I am very tempted but being tethered to a computer and needing an assitant to run the captures seems like a pain.
So I ask, does the bump in color make it all worth it or is it in fact not worth the production pain.

4:2:2 is generally better for chroma keying, as it keeps edges from getting sawtoothed. Better color resolution helps the divisions between colors, basically, reducing artifacts. Personally, I have shot in 24f on my XHA1 and, when lit correctly, it isn't too difficult to key the footage so that it looks really nice. I use Shake and After Effects for keying. I'd think it would matter mostly with really complex scenes, where you are dealing with more than a basic green wall. My experience has been that HDV keys far more nicely than DV ever did. Component would be even better, but I'm not sure it's worth the trade. Depends on what you are going to do.
Additionally, the component out is pre-HDV compression, so there will be some other differences regarding your imagery - mostly related to compression. There are varying opinions on how people feel about HDV compression when it comes to motion, but the common denominator dictates that it does, in fact exist and being able to avoid those artifacts is a real boon for some. Personally, it has never been an issue for me - especially in 24f mode. I shot a ceiling fan once (close-up, high speed) that broke up in 60i, but had no problems in 24f. It's one of those things where if you are going to film so much motion or so many scene complexitites that the compression matters, it would probably be impossible to tether yourself - or you shouldn't be shooting that scene because the basic production value/composition is all wrong.
Again, this is from my personal experience shooting interviews and some light key work, as well as some indy films and events.

Cant you record to Tape simultaneously to get the sound?
perhaps I am thinking in terms of the H1 > iohd

I know you can take in anything with the RCA and than take it in through the firewire, but can i do the same thing with a component?

No. Just like all other HD camcorders, the analog component video connection is an output only.

Actually..we are both wrong. :)

The Decklink HD is 10bit 4:2:2

Mondays... :)

-willSo when using component out from an XH-A1 is it worth it to go with 10-bit 4:2:2 over 8-bit 4:2:2? I mean if I'm going to go through all the trouble of external capture, why skimp on the last two bits?

I realize this is a very subjective question, but would love to hear people's opinions.

Thanks!

Chris,

Sergio already corrected you, but I wanted to add that I *wish* that's how it worked. There is no technical reason that manufacturers have to compress live footage to HDV when sending it out via firewire, but they all do it, and none provide any other option. HDV only uses 4.5 MB/s, while firewire (400) has the capacity to carry much, much more data: 48 MB/s. That's not enough for uncompressed 4:2:2 1080, but there are a lot of options. In any case, it would compete directly with ultra-high-margin HD-SDI, and they would need to find a codec that's light on the camera's tiny CPU, so it will probably never happen.

Pardon me for jumping in on a discussion between much more knowledgeable guys, but isn't this all why some folks buy an XH-G1 rather than an A1 - to be able to access and record/save uncompressed footage? I thought that ALL the outputs from an A1 were either compressed or D-A converted or both.

* Didn't notice or read the next 2 pages of posts on this thread - sorry.

Colin,

I agree with you--that's precisely why we got the G1 instead of the A1--we use the HD SDI to the AJA IoHD, then into a MacBook Pro laptop and save to an external eSATA RAID1 enclosure (via Express34 card). Allows "uncompressed" capture (we actually use ProRes422 HQ, the quality of which is apparently on par with uncompressed.)

I have access to an A1 and an IOHD, and am curious if anyone has found a real working benefit to capture off the component out of the A1 versus pulling the key off the HDV footage. I understand the color space ramifications of each, but wonder if it really makes much difference real world, or vs. something like the G1 or H1 with HD-SDI out.

Or if anyone in the Austin area has an G1 or H1 they would like to loan out for a day...

Thanks

Austin

Does anyone know if the component output on xh a1 is not sure how to say but has not gone thru hdv compression and is uncompressed like the hdsdi on xh h1 or g1. I was thinking of putting a component to sdi converter (either blackmagic or aja) then feeding the sdi signal into macpro with kona lh. I know lh has analog inputs but have read that its better to go into it sdi, plus the advantage of longer cable runs. If it is not any better quality by going to sdi thru converter is it at least better quality compared to hdv. Has anyone done any tests to compare live capture off of hd sdi off of g1 vs component on xh a1. I now should just get g1 or xh h1 but for the price can get a1 and macpro or even iohd. Plus already have 2 a1s.

Check out this thread. Clearly, even though the A1 does a good job with Greenscreen in HDV, component output makes for really clean comps.

http://www.dvinfo.net/conf/showthread.php?t=117788

I've been looking for info on this exact subject., re composite vs firewire capture.

I have the XH-A1 (which I love dearly) and am trying to find a way to get the absolute best captured footage out of it for a project I've currently got going. For this project the camera is locked down in an overhead rig shooting straight down. We don't need sound for it (I usually record sound separately anyway and sync in post), just the clearest richest image we can get under controlled studio lighting.

I have tried capturing directly from On Location via firewire straight to disc, but the footage seems to be exactly the same as that captured from the tape via firewire, which makes sense to me.

However, we're using a 24" Samsung HD touch screen monitor as the preview monitor for the shoot, (the same monitor that this particular interactive HD application will be ultimately delivered on), to get the focus as sharp as humanly possible. We're going into the monitor via component for the preview.

One thing I noticed is a very big increase in the overall visual quality of the signal on that monitor when viewing via component. Colors are richer and more realistically saturated, and strangely, the focus is much more sharp.

When I capture the footage via firewire from the tape, it's noticably duller, on the grey side, and slightly fuzzy, like it's a bit out of focus. Still don't understand the drop in focus though. Why would that be, other than I really may not have had it in focus to begin with; but I had two other sets of sharp eyes to confirm the focus on the monitor.

But hooking the component outs from the camera to the component in on the same monitor at the same time I'm looking at the captured footage in Premiere Pro CS3 shows that same taped footage looking pretty much exactly like it looked when we were monitoring it live. I'd like to be able to deliver that better looking footage as our end result (h.264 full 1080p at a data rate of 10-12Mbps or more), if at all possible.

I'm about to buy a Xena LHe and Cineform Prospect HD in the hopes that I can get the very best out of the camera, since it looks like I might be doing a lot more of these interactive HD projects in the near future.

The increase in complexity and time for the component workflow is not a deterrent to me. Ultimate visual quality is the goal. Even if I had the XH-G1 I'd still have to have a similar capture capability and workflow for the digital SDI footage, correct?

Eventually we'll most likely go to the RED in the future, but that's a whole other story.

Actually, I wonder if there is anyone in the DFW area that has a Xena setup like the one I mention above. I'd be more than willing to pay for the time it would take to capture a test of our latest shoot (approx. 3 minutes) so I can do some more R&D. If anyone has or knows someone locally with such a rig, I would greatly appreciate contact info.

DVinfo is a fantastic educational resource and it has taught me so much. But I've got a long way to go before I understand all this as deeply as I need to.

Thanks for any input from any of you knowledgable types out there.

Hi Doyle, I searched the XH A1 forum for all of the threads I could find which discuss component video capture and have now merged them into one large discussion. You had posted to our XL H1 forum and I have moved your post to this thread since you have an XH A1. Please review the entire discussion thread thread here. Thanks in advance,

Thanks Chris. You're certainly on top of things. I saw my post show up someplace I didn't put it when I went back to the XH-A1 forum and wondered how that happened (still a newbie on these fine boards).

I've been reading that very interesting collection of threads you put together. Very useful. Much appreciated!

Don't know if this really answers any questions, but I found this ( http://www.blackmagic-design.com/products/videorecorder/ ) to be a great option.

this is interesting -
would it allow for HD 4:2:2 recording?
it shows a full res capture preset...
Really?

if you record through the SDI would it record HD
it states 1024 video

ideas?

To me it looks like it does only SD.

Alas,
So close...
and yet,
so far