Capture HD other than FireWire?

[Len Capristo]

I've searched this site, but couldn't find the information I want. If someone knows the link, I'd appreciate it.

I have a XH-A1 and find the HDV format not as robust as I had hoped. I have heard that I can buy a Canon HV-30 and convert my recorded tapes made on the A1 into a true, full HD output by using the HDMI output on the HV-30 to bypass the HDV encoding inherent in the A1.

When I spoke to my local video source he said he never heard of such a thing, but anything is possible.

So, is there a way to upgrade my HDV output to a true HD format through the use of a HDMI export, or some other method?

I use a MacBook Pro (first dual core Intel chip, running with 2GB memory and 256MB video RAM, clock speed 2.0Mhz) to render using Final Cut Pro (Studio).

Any help is appreciated. Thank you in advance.

[Adam Gold]

A number of fallacies here.

There is no such thing as "True HD." It's a meaningless marketing term used to lure gullible consumers. What is it you are meaning by that term? What is it you mean by "not as robust as I had hoped"?

You can't add back something that's not there to begin with. Once HDV has been recorded to tape, it's been compressed and dubbing to, or outputting from, another cam won't help you at all. The HV30 and the A1 record in exactly the same format. They compress to tape the same way and output identically, and even if they didn't, what's on the tape is what you get, no matter how you play it back.

HDMI can be better, but only when going out live to a PC equipped with proper capturing software or to an HDMI recorder. Once your video is recorded to tape, it's compressed and there is no advantage to HDMI out -- it's the same signal that goes out via FW, minus the timecode.

There are utilities that "uncompress" HDV for editing, and while they ensure that your footage edits easier and doesn't lose more during the process, they cannot add what isn't there. Cineform is one.

But even these plug-in codecs just allow you to add better new elements in the process -- they cannot improve what is already there. Think of HDV as five gallons of water in a five gallon bucket. Cineform is a ten-gallon bucket. Even if you pour all of your water into the ten-gallon bucket, you've still only got five gallons of water. You can add more, new water to the bigger bucket, but your original amount remains the same. You can't magically change five gallons into ten.

[Noel Lising]

I agree Adam there is no such thing as true HD. The term has been used by Christie Digital for their HD Projectors, the only True HD Projector, blah, blah and to some extent Best Buy sales reps.

[Andy Tejral]

Well, there is in fact such a thing as true HD, it is the same as HD!

If your image is 720 or 1080 lines tall, the image is HD. Period. End of discussion.

Another consideration is resolution which is independent of image size. A cell phone may be HD (true!) but its resolution is going to be a little different than a $50K camera with the same image size.

[Chris Hurd]

For some reason this thread wasn't posted to our Canon XH forum. I've now moved it to where it belongs. I've also edited the title. It was "Can I get True HD from my XH-A1?" which doesn't make any sense, because the XH A1 already records HD.

I've renamed it to "Capture HD other than FireWire?" which is what it's really about. There are a variety of hoops one can jump through to avoid capturing HD via FireWire. They include capturing via the analog component HD video output, which doesn't carry audio, or playing back in an HV series HDV camcorder and capturing via HDMI. Either one will work, but it's the law of diminishing returns... why bother with all that extra expense and workflow just to gain some increase in "image quality" that you're not likely to be able to detect. Regardless, where there's a will, there's a way. There are also plenty of threads that have previously discussed some of these options in great detail.

[Michael Hutson]

Len,

in my opinion, the xha1 takes fantastic footage. If you are shooting in automatic, then you will run into some problems. What mode do you normally shoot in? What do you normally keep your gain set at? What custome preset do you use? What is your lighting conditions? These are all factors on how the images come out. Just look at some of the sample clips that ppl have shot...some are truely amazing.

[Dave Stern]

Len, I think through everyone's passion for the subject, we may have jumped a bit too much on your case, when clearly you are trying to sort though all of this stuff ....

I think Michael said it well, what is it about the footage you don't like or what isn't the camera doing that you would like it to do better? What kind of shots, how are you shooting them, etc. Maybe some of the experts here can help you confirgure the camera and use it to get the most out of what it can do ... for the price, it's pretty amazing, maybe just some tweeking is needed

but it is true that once the image is compressed to HDV and on tape, it can't be 'uncompressed' to put back what't not there ... that said, again, the camera itself can take some pretty incredible shots

maybe you can say more about what you are looking to do...

[Len Capristo]

My thanks to all, and as you obviously tell, I'm not sure myself what I'm talking about.

I received pre-edited video content in DVD format for broadcast on local access TV. I take the DVDs to the TV station, run them through their "ingest" process and the DVD is magically converted into a format that I can edit on FCP, also provided by the TV station. To be clear, I have written authorization to use this format in anyway I would like for the purposes of public TV access.

After doing this for a few years I got the bug to become the next great producer, and took several classes at the TV station - all using their equipment. I learned a bit about lighting (now I know why professional stuff looks so good - it's the lighting for one thing), audio (now I know why XLR connections are used on professional equipment) and camcorders (both analog and digital).

So, I bought a Panasonic GV-180, a fine camcorder that was under $400 with some manual controls, headphone connection and some other things. Clearly knowing that Hollywood was right around the corner, I moved to a used XL2 and XH-A1. I also bought an original 15" Core Duo MacBook Pro, complete with 256MB video card, upgrading the hard drive to 320GB 7200 RPM capability, and bumped up the memory to 2GB, the system limit.

Then I enrolled at Duke University's Center for Documentary studies where I continue to learn something about video, audio and still photography with the intent of telling a story.

Now I find myself in the "dangerous" category - I know something about what I don't know, but haven't any real experience in shooting, editing, or composition - in other words, a beginner with toys.

So my goal is to learn how to cross-convert video - allowing my video to be used on an iPod, or potentially streamed to a website, etc. I even considered a NewTek Tricaster, but I haven't the skill to understand how to best use it.

So the obvious problem is I'm not well educated about video, or what it takes to assemble a basic kit for documentary work, interviews, and to allow me to help non-profit organization (churches, Board of Eduction, museums, etc.) to capture the video they would like and to deliver a final product - DVD, iPod Video, etc.

Where/how do I learn what I don't know? Any suggestions are welcome.

Chris - Thank you for moving me to the correct area, and I appreciate the attempt to classify this post - obviously it's a bigger job than you thought!

[Steve Lewis]

I would beg to differ about the "true HD" designation. I have my own personal idea of what it means: 1920x1080, (interlaced for broadcast, progressive for cinema) and 50Mb/sec or greater. I think that anything that falls below that (like AVCHD, HDV, XDCAM, etc...) is a crippled form of HD. Again, that's just my own designation, nothing universal.

-steve

[Peter Moretti]

Originally Posted by Adam Gold

A number of fallacies here.

There is no such thing as "True HD." It's a meaningless marketing term used to lure gullible consumers. What is it you are meaning by that term? What is it you mean by "not as robust as I had hoped"?

You can't add back something that's not there to begin with. Once HDV has been recorded to tape, it's been compressed and dubbing to, or outputting from, another cam won't help you at all. The HV30 and the A1 record in exactly the same format. They compress to tape the same way and output identically, and even if they didn't, what's on the tape is what you get, no matter how you play it back.

HDMI can be better, but only when going out live to a PC equipped with proper capturing software or to an HDMI recorder. Once your video is recorded to tape, it's compressed and there is no advantage to HDMI out -- it's the same signal that goes out via FW, minus the timecode.

There are utilities that "uncompress" HDV for editing, and while they ensure that your footage edits easier and doesn't lose more during the process, they cannot add what isn't there. Cineform is one.

But even these plug-in codecs just allow you to add better new elements in the process -- they cannot improve what is already there. Think of HDV as five gallons of water in a five gallon bucket. Cineform is a ten-gallon bucket. Even if you pour all of your water into the ten-gallon bucket, you've still only got five gallons of water. You can add more, new water to the bigger bucket, but your original amount remains the same. You can't magically change five gallons into ten.

1. I think what he meant is 1920 horizontal resolution.

2. Upconversion is used throughout various stages of video technology. Examples are pixel shifting and in camera sharpening used by the XH-A1.

3. People have reported improved results capturing out an HDMI port versus firewire. This will of course also depend on the how the HDMI signal is captured. And it's obviously not as good as bypassing HDV entirely, but there are anecdotal accounts of there being a benefit.

4. HDMI is not the same signal minus timecode. HDMI (as used by Canon's HV's) is uncompressed 4:2:2 YUV. It's entirely possible that the HDV to upconversion HDMI adds something that is visually appealing to the picture. But would imagine that even if the there were an advantage, most people would not find it worth the extra effort.

[Tom Fales]

I've spent all afternoon looking through threads on this issue and am still not clear. I think what I was looking for was a discussion on the PROs and CONs of using the HDV/DV firewire connection directly into the PC and alternately using the Component out cables through something like Blackmagic Intensity Pro (what I have).

I am looking for the best quality video to use in my editing system (Avid and Pinnacle 16) to produce the best result. I have a mixture of SD and HD tapes that I need to transfer.

[Chris Medico]

If the video is already recorded into the cameras internal media then taking the firewire signal is as good as it gets. Using any of the other outputs will not improve the signal at all.

If you are going to set up the camera to record new material you can use an external recorder via HDMI to bypass the HDV limitations.

[Eric Olson]

Originally Posted by Chris Medico

If you are going to set up the camera to record new material you can use an external recorder via HDMI to bypass the HDV limitations.

The XH-A1 does not have HDMI out. To bypass HDV on the XH-A1, you need a high definition recording solution with analog component feeds. In my opinion such a setup is useful only for studio green screen work. When shooting on location it is too much trouble to bring the cables, power supplies and computers necessary to bypass HDV recording; when shooting in a studio (except for green screen) the visual environment can be controlled so that 25 mbps 4:2:0 mpeg2 HDV looks great.

Capturing an already recorded HDV tape through HDMI on an HV30 doesn't make sense because this introduces an additional compression cycle into your workflow. Unless something is broken with your hardware or software, the most reasonable approach is capturing the digital video files stored on the HDV tapes directly to your computer through firewire.

[Jeff Pulera]

Hi Tom,

HDV is compressed using Long-GOP MPEG-2 at 25Mbps, with 4:2:0 color, at 1440x1080 (Anamorphic HD). That is what is on your tape, and is transferred bit-for-bit via 1394 into your edit system (unless talking Apple FCP which might transcode). For most applications like weddings and stage events, many editors will be happy with that native format. If doing green screen keying or other compositing work, then HDV does not lend itself very well to that.

Here's my experience. I have shot green screen at 1080i with a Sony HDV camera, and when I captured the footage into Premiere Pro via 1394, the keys I achieved were OK at first look, but magnifying the result would show very blocky, jaggy edges on the subject! I then captured the same footage again, using the camera's analog component output into a Matrox MXO2 Mini capture device. MUCH cleaner edges were the result, producing very nice keys!!

So how can this be? The footage on tape is 1440x1080, 4:2:0, Long-GOP compressed...that color info is GONE, right? Well, yes it is but - the output from HDV cameras is actually 1920x1080, and also with 4:2:2 color (I believe). So consider that the HDV footage is going through an D>A conversion, so the color must need to be interpolated during that process. The Matrox captures full 1920x1080, with 4:2:2 color and I-Frame quality at a high bitrate, 100Mbps versus 25Mbps for HDV via 1394.

While you can't "replace" picture information that was lost during the compression to HDV, you CAN smooth things out with the process I just described. I had not tested to see if HDMI output will yield the same results, but should be similar.

Think of it this way. Due to the digital nature of HDV, and 4:2:0 colorspace that throws away half the color vs 4:2:2, the result is BLOCKS if you magnify the video. Video compression is done in blocks, and color is represented in blocks. So the result when keying is---blocks on the edges. There is not enough information there to produce a smooth edge - the closest it can get is the resolution of those blocks of pixels, rather than that of the individual pixels.

When converting that blocky digital signal to analog, and then capturing to a more robust codec that does better represent the pixel colors with 4:2:2 color resolution, that interpolated result just keys better! Also could be better for other composite work or advanced color correction to reduce banding and such vs. native HDV.

If shooting green screen and you have the opportunity, one should actually take the LIVE output of the camera, HDMI or analog, direct into the capture device to record into edit system on the spot, so that the video bypasses the compression stage in the camera when recording to tape (or cards on newer cameras). There are also portable recorders such as Ninja 2 that will record live HDMI signals at high quality.

The Black Magic capture devices should provide a very similar result to what I got from Matrox, as BM also records to a high-quality 4:2:2 codec.

I wish I could be more technical and show diagrams or samples to reinforce what I'm sharing, but the proof is in the result, it just works!

All of that said, 99% of my editing work is captured from HDV using 1394. I like the Scene Detect feature in Premiere, and the image quality looks fine to me. It is only when keying that I look for that extra edge and capture via analog. So it's up to you - if you want to capture analog or HDMI for everyday editing, and use about 4x as much hard drive space, that is your choice.

I hope this makes sense and helps someone out in making their decision

[Don Palomaki]

I think the message here is that there are several possible methods to take material from HDV on tape to 1920x1080.

The end result will depend on the relative qualities of the various products involved; e.g., is the D->A in the camcorder better that the codec in your NLE? and at what stage in the workflow do you make the conversions.

Other things equal, from an image quality perspective it is generally better to make any upconversions before any editing or effects are applied, and make any down conversions as the last step. The costs may include more storage required and longer processing time.

Run some tests with your setup and material to determine which is best for you.