HV30 newbie. lots of help needed.
 

Hey guys. Long time reader, first time poster. I was actually in the market for an XH-A1, but I realized just how much of a stretch that is for my price range and inexperience. So I decided on the HV30, solely for the price.

I will be shooting a few promotional short films for a conference in December. So I started reading up on the various things I might need and I found that I was lost in alot of esoteric terminology and confusing equipment names that may or may not be necessary. I understand that there is much more to a great film than "the look'' of it or the equipment behind it, but I would much rather it look as professional as possible, especially for the audience at the conference. I've decided on the camcorder itself, a shotgun mic, tripod. Do you guys have any recommendations on either of those that are both inexpensive and reliable? Can you guys recommend anything else that I might need? I was in the market for some 35mm adapters, but the prices sort of dampened my spirits.

What I found most tricky is the post-production aspect of this. Now, I understand that to import 24p video you need to de-interlace it? What exactly does that mean and what is a JES deinterlacer? I've read about the Pro-res 422 codec on FCP. Does that mean that I could fit HD quality footage on SD DVD also? How could I export HDV footage onto DVD without treading the waters of Blu-ray technology?

As you can see I'm pretty new to all of this, but your help is very VERY much appreciated. Thank you so much.

Daniel, regarding mics, you could do worse than poke around the audio forums, really the answer depends on your budget. I'm a big fan of the Audio-Technica shotguns for price vs quality.Forget the 35mm adaptor unless a 35mm look is essential, which it rarely is outside of fiction. It's a tremendous hassle to use one, you'll need a good focus puller as part of your crew. If by your own admission you are a little inexperienced, save it for later. if you're showing from a video projector, de-interlacing to 24p might not really be necessary. JES de-interlacer is a program that takes interlaced footage, processes it and spits out de-interlced footage (or a bunch of other things). But frankly it may not be worth your time unless you want to work with effects or make a PAL version or unless you absolutely HAVE to have a progressive scan DVD.

What it means is all NTSC video as displayed is 29.97 frames per second. 24p is shown in whats called a 2:3 pulldown. Film gets converted to this pulldown when scanned to video tape. DVDs can retain the true 24p frame rate, but the DVD player adds the pull down when it's sent to the TV (unless you have a true progressive scane DVD player plugged into a true progressive scan TV)

the HV30 shoots at 24p (actually 23.976p) but adds the pulldown when it records it to tape.
If yo (really) want to know more just google 2:3 pulldown. Oh look, I just did it for you.

ProRes 422 is an editing codec, you really don't want to bother sticking it on a DVD. It won't play in any DVD player and will probably be pretty unwatchable (skipping a freezing) in most DVD drives.

An SD DVD holds 4.35GB of footage (single layer) or twice that dual layer. That means a single layer DVD will hold about 4 minutes of 1080i HD pro res. However that DVD WILL NOT be able to play back the data at that rate.

Various people have experimented with burning Blu Ray onto SD DVDs with various results. On Mac you're S.O.O.L as it doesn't support Blu Ray yet (but hey, maybe tomorrow I'll be wrong on that!). DVD Studio Pro does burn HD DVD to SD DVD and HD DVD players will play them back so you could pick up an HD DVD player for next to nothing and use that for your conference.

Alternatively if you've got a fairly Powerful laptop you could play out of the projector via XVGA or DVI to get higher resolution (this time using ProRes, playing from the Hard Drive), in which case I WOULD recommend you de-interlace your footage first.

Hey Daniel:

Here's the skinny: Everything is confusing at first, but once you get the basic concepts, you should be fine:

Here's what you really need to know, and wish I knew when I started:

Rule #1) Good sound is more important than good picture. What that means is that you should budget for a shotgun microphone, a shock mount for it, a lavilier mic (comes in handy even if you don't think you're going to use it) and a pair of headphones. If you can afford it, get a second audio field recorder too - the $200 Samson Zoom H2 gives you good bang-for-buck. You can shoot the graniest, nastiest, most blown-out footage and people will think you're an artistic genius - but a little hiss in the audio track, and you're an amateur.

Rule #2) If you buy the cheap stuff, it will cost you money as you will eventually have to buy the good stuff. There are a couple of products that people swear by for the good "bang for the buck" ratio. Still, if you can afford the good stuff, get the good stuff because you will eventually get it.

Rule #3) Your work looks better and more professional when you do a simple thing well, rather than when you do a complex thing adaquately. You learn more when you do complex things adaquately than doing simple things well. The purpose and use of your shooting should determine what you go for. There's a reason why Vimeo is full of 2 minute "test shots."

Rule #4) Unless you're in a situation where the camera is moving a lot and you can't plan out your shots - say, you're in a riot or warzone - autofocus is not your friend. Always use manual focus when you can.

Rule #5) Learn the Rule of Thirds.

Rule #6) Don't switch tape brands. Seriously, it could gunk up the heads, leading to dropped frames.

Now, here's some specific advice for you:

The Canon HV20/30 family is just about as good as you can get in the consumer camera market. The XH-A1 is better, of course, but again, price-for-performance, if your choices are getting the HV30 now and waiting a year for the XH-A1, get the HV30 now because at the very least it'll help you practice.

Progressive, Interlaced, PAL, NTSC, 24p, 25p, 30p, 50i, 60i - these are terms that you will come to know and loathe.

First, PAL vs. NTSC: When the television first came into being, for technical reasons, different countries decided to make the refresh rate of the TV - that is, the number of "frames" on the screen at one time - different. In the U.S., we use NTSC, and unless you're shooting overseas with an overseas camera, you probably don't have to worry about using anything else. NTSC (the U.S. & Japan) uses 60 interlaced frames per second, while PAL (most of the rest of the world) uses 50 interlaced frames per second.

What does this "interlaced" mean? Well, when TV was invented, they wanted to present a smooth picture, but they could only draw the picture line by line, creating a nauseating "wavy" motion. So what they did was draw every other line of the picture, then went back to the top and filled in the rest.

The opposite of this is "progressive" where the whole picture gets filled in at once, like taking a snapshot for each frame.

Here's where it gets tricky: Interlaced footage sometimes creates interlacing artifacts. If the footage is interlaced at 60i, any movement on camera which takes less than 1/60th of a second will not be in the same place on both of the interlaced frames, so you'll end up with horizontal "fringing." Progressive shooting doesn't have that problem.

The HV30 offers "progressive" recording, but it does so in a sneaky way. The first is 30p, which basically means that the camera records one frame every 1/30th of a second and writes it on both half-frames of an interlaced frame. This is actually simple to recover and you can probably do this in your NLE without any problem. It's simple, it looks crisp and professional and I HIGHLY recommend that you use 30p.

The second is 24p, which is 24 progressive frames per second. That's the same frame-rate as film. That's not to say that it's as good as film, or that it looks just like film. But 24p introduces a little bit of a motion blur that looks like it was shot on film. If you're planning to shoot for the cinema, that might be a good reason for it.

Because the HV30 is a cheap camera, the 24p is recorded on the tape in the 60i format, like a bubble gum wrapper wraps bubble gum. To remove the bubble gum from the wrapper, you must import the bubble gum and then use a tool like JES Deinterlacer to perform an "inverse telecine" and get the gooey goodness inside.

It's a pain in the ass and if I had it to do all over again, I would have shot my movie in 60i.

The XH-A1 does not have these problems - you can import 24p directly from the camera. Additionally, paying $200 for a third-party codec allows you to skip this step. Then again, you're paying $200 for a third-party codec.

Okay, that's the difficult bit:

Tripod: A photo tripod won't do. You'll need a video tripod, one that'll pan and tilt smoothly, instead of the jerky movement of a photo tripod.

ProRes 422 is simply a high quality codec for storing video information that is also fast to edit. That does not mean you can fit HD quality footage on SD DVD.

What you're probably thinking of is H.264, which is a highly compressed video format. With that, you'd be able to cram HD footage as a data file onto an SD DVD, but you wouldn't be able to play it back on your DVD player, you'd only get about a half an hour, and the H.264 format is lossy - it's the same concept as ripping a CD to MP3 - you lose some information. Normally it's information you don't notice - but sometimes it's information you do notice.

You can export HDV footage onto DVD this way for storage but again - you can't play it back on a DVD player.

wow. thank you both so much for the very informative posts.

one more question. so how would i be able to fit my movie onto dvd for distribution? is it possible to at least upconvert SD footage and put it on dvd?

If you're talking about playing DVDs in a DVD player, you are limited to 720x480 in NTSC regions, and 720x576 in PAL regions. If you're an American, you really shouldn't worry about PAL regions at all. What this means is that you are practically limited to 480 lines of vertical resolution - which is standard definition - on a DVD player.

As for upsampling SD footage - this is something that you as a filmmaker have no control over. Upsampling is either done by the DVD player or by the HDTV, and it does not magically turn an SD DVD into HD - it just means that you get the best possible quality out of the DVD by transferring the source material digitally instead of putting it through an analog process. Upsampled DVDs look great - no doubt - but they are not HD.

If you're talking about burning an H.264 file onto a DVD for storage or for mailing to friends, which you cannot play on a DVD player, here's where it gets a little tricky. You can ask "how many minutes can I play" and get 10 different answers - all correct.

A single layer DVD holds about 4.25 gigs of data, a dual layer DVD holds about 8.5. It doesn't matter how much the resolution is, it only matters how much data can fit on the disc.

Now, here's the tricky bit.

The reason H.264 files can show very high res video at relatively small file sizes is because it uses compression. To use a metaphor:

Uncompressed:

Compressed:

The meaning is there, but the subtle nuances are lost.

Now, computers are better than humans at plucking out the nuances that are so subtle that people don't notice them, and so people listen to MP3 files and think it sounds almost the same as a CD - though some prefer the rich sound of the CD - and others still prefer the richer sound of vinyl. Same with H.264 files. But if compression takes too much out of the video, you start to notice it. These little noticable inconsistancies are known as "artifacts" - called that because the compression process has "left behind evidence."

So what this means is that you want to make the movie small enough to store, but do not want to compress it down far enough that it the artifacts it creates detract from the movie watching experience.

Every H.264 file has a "bitrate" - that is, the number of bits used to store a second of video. This, to extend our Gettysburg address metaphor, is kinda like telling the computer: "Okay, you have X number of letters to store the Gettysburg Address" - obviously, the more words you get, (or the more bits-per-second) the more of the original source shows through. So you can make the video any filesize you want and make it fit anywhere - even, theoretically on a floppy disk - it's just that the fewer bits you allow the computer to use, the worse the movie looks.

Where HD and SD enter into this is that when you store SD footage, you can get away with lower bitrates because there's less information to be stored. If you're painting a picture 30 times per second, at 720x480 pixels, you've got 16,848,000 pixels each second to worry about, and each pixel is a different combination of Red, Green, and Blue values. That's a LOT of information. (In fact, it's so much information that even the bog-standard DVD-in-a-DVD-player uses a compression format known as MPEG-2 to store all of it! You don't notice it, because the bitrate is high enough that it leaves very few artifacts.)

But HD, at either 1280x720 or 1920x1080 - uses even MORE pixels! This means that the same bitrate for HD and SD footage will mean that the HD footage will look crappier than the SD footage - because there's just so much of the original information that gets left out.

The long and the short of it is that you can compress any amount of video onto a DVD that you want but the more you cram stuff in there, the worse it's going to finally look.

This is true even for the "miracle" H.264. H.264, as a computer algorithm, distinguishes itself from other compression "algorithms" is that H.264 is "smarter" at removing information that you don't need and retaining information that you do. What this means is that you can lower the bitrate further than you can with other algorithms (like MPEG-2) and still be left with a watchable picture. The tradeoff is that both the encoding machine and the decoding machine have to be "smart" enough to deal with H.264's smart compression... so H.264 requires extra hardware for playback. An H.264 file will have a very hard time playing on an older machine, for example.

Now, this may have discouraged you from taking the HD plunge - allow me to say that as far as recording goes - yes, go HD, even if you only plan to distribute in SD on DVD.

This is for a number of reasons:

1) High Definition Content shrunk (or "down-rezed") to SD sizes looks very sharp and very high quality compared to even high-quality SD cameras.

2) Recording and editing in high definition "future proofs" the material - while BluRay may not take off, or take a while to take off, someday there will be a high definition format. Having a high defintion copy of your stuff even in your own archives will be a godsend when that day happens.

3) On a standard definition camera, it's noticable when you try to "reframe" content by cropping. Sometimes there's something in your shot that you don't want to be in the shot - a boom, for example, or, in some cases, a flasher. If you know you're only going to output to SD, you can cut that out of the shot and still have plenty of visual data left to still make the shot look good. If you recorded the same material in SD, cropping means that you're really left with an even lower resolution than the SD footage. I use this a lot with my current documentary project - I recorded it at 1080p (or 1080 vertical lines of resolution.) Since I'm exporting at 720p, I can basically "zoom in in post" up to 150% of the original image without losing visual quality. If you record in 1080p and output to SD, you can "zoom in" up to 225%.

Bryan did a great job of providing the gory details but the simple answer is:


Render your HD footage to SD DVD with your editing/authoring applications ..

You will find the results to be excellent and somewhat better than SD footage rendered to SD ..

Render your HD footage to SD DVD with your editing/authoring applications ..

You will find the results to be excellent and somewhat better than SD footage rendered to SD


And if you want a really nice SD DVD for playback on an HDTV system using a upconverting DVD player, don't hesitate to crank up the rendering bit rate if your software allows for it.
I have been using 8000Kbs/sec as my preferred render rate. it seems to be acceptable to most Std DVD players at 10,000 Kbs/sec some dvd players start to Choke and you get a jerky playback situation but I have found that almost all will handle 8000 just fine, and the resulting playback on an HDTV is Very nice indeed although you loose disk playback time because the resulting file sizes are considerably larger, but you can easily get 45-50 min on a single layer disk. Short of burning a BD HD disk this method gives the best HDTV playback Ive Seen Yet of SD material, nice and crisp with good background definition and depth.

I realized something while reading through the posts. Would it be much less complicated to shoot everything in SD and then use the Magic Bullet plug in to pretty it up?

SD is always easier than HD.

As to how pretty you can make it, it depends how adept you are with Magic bullet. It's a pro tool and not an easy one to master (render times are horrendous too) and you won't be able to get SD footage to look like HD footage.

You can pretty it up but remember colour correction in video is basically about selectively throwing away image information (be chrominance or Luminance) to achieve the look you want, degrading the image. In many cases it's necessary but remember the "quality" is never as high as it is when it comes straight out of the camera.

You want to make it as pretty as you can in camera before processing it.

If you want to bypass all of the HD hoopla in post editing and you are used to
working with SD then you can use your HV30 footage with " Cineform "....

This will allow you to shoot in HD, let Cineform convert the HD footage to a intermediate
format " AVI " that you edit just like you edit with SD, but the footage is still HD...

Then after the edit you can

1) downconvert to DVD (and the footage does look great)
2) put the HD footage on a standard DVD disk that will play perfect HD footage on a PS3 blu ray player.
3) put the HD footage on a blu ray disk to play on any blu ray player...

Or you can do all three options.. :-)

And Cineform also supports the HV30 24p pulldown perfectly....

I would say no. Magic Bullet Instant HD is nice, but SD material uprezed to HD will not look as nice as HD material downrezed to SD.

Additionally, Magic Bullet takes for-freakin' ever. Brilliant stuff, but I wouldn't use it until the absolute last step of the movie.

There are programs like Cineform that allow you to bypass all the tough stuff - I've found that on a mac, it was better just to jump through the hoops. On a PC? Yeah, I'd get Cineform.