.comment-link {margin-left:.6em;}

High Definition Video for Independent Filmmakers
A How To Guide for Digital Filmmakers
Welcome all! This is my blog to share my latest research,
thoughts, etc. on utilizing HD for independent filmmaking.

YES, I am available for consulting
Contact me at mike@hdforindies.com

All content copyright 2004-2007 Mike Curtis.

Monday, January 30, 2006

DV's Four Camera Shootout - All the sub-$10K HD cameras you care about 

DV - Features

OK, this is MUST READING for anybody thinking about shooting a feature with a sub-$10K HD camera - as in the HVX200, the XL H1, the GY-HD100U, and/or the Z1U.

Long, long, long article, registration required. This link takes you to the features page, there is no direct link, but the article in question is titled "Four Affordable HD Camcorders Compared" and is presently the top link.

There is SO much good info in here, this is mandatory reading.

It talks about the true measured resolution off of all of these cameras, and all kinds of killer details about how they compare and how it all looked in usage.

The interesting bit of summary was how close they all came out in the end in terms of quality, and therefore workflow and shooting ease start to get much more important.

Go read it all, it'll take you a while.

I'm planning on incorporating much that they learned into my own testing to be done in a few weeks.

-mike
Comments:
I still have to read it but have been reading reactions over at DVXuser.com.

It seems that HDV compression was not factored into the test and makes the Panasonic HVX200 look poorer in the end. I dont know about you guys, but I do documentary. Portability is the only way to shoot.

If it is the case, DV magazine should make it very very clear that they are only talking about uncompressed output and NOT what a user would actually experience following a subject down a staircase at fast speeds.

-Christopher
 
Christopher,

Compression quality is a much smaller factor here than most assume. If the tests where only done compressed, some would agrue that the low resolution of the Panasonic was due to DVCPRO-HD lower luma sampling, and that uncompressed would give better results. This was a good way to measure camera performance. Compression performance is more subjective, as both compressors throw data away, up to the viewer to determine which looks more natural (and it is not obvious.)

David Newman
CTO, CineForm
 
If you read the article again, you will see that Adam Wilt, the author, states that there are at least 100 more tests that need to be run on all the cameras. This was just a starting point.

The first thing that does need to be tested is motion and how each camera handles it.

Good start though and I am glad that Adam wrote the article and not one of the websites that promotes one camera over another.
 
The article does clearly state that all of their findings are based on uncompressed output, but it was still surprising how poorly the HVX200 looks considering it is supposed to have a 1080p native sensor. While these tests are certainly quite superficial in nature, it's nice to see the side by side footage from the cameras. Things that I took away from this: The F900 is clearly on a different level from the rest of the cameras, and 24F is definitely not 24p in terms of resolution. Now if they could only get their hands on one of those RED cameras...
 
Compression is a huge factor. It can effect color, resolution, and can create very stronge artifacts in the image. If you dont think so, you have no idea what your talking about.

I'm sure they will go into in future articles but, for some things HDV compression can be unusable. The P2 camera would clearly come out on top there. Compression problems are very content dependant. Try taking a headshot of a static actress then compare it to a shot with a horizonal pan across a parked race car. Watch all those logo's and stickers tear in the image. Its hard to miss. HDV completely falls down on horizontal movement in detail heavy scenes. Its not at all subtle.
 
I would like to deposit my experiences with HDV material.
On IBC 05 on the AVID booth there was a comparison of footages between HDCAM DVCPROHD and HDV. It was shown on 50' inch HD plasma display and I was watching about 5 feet from the display. When pans with normal action were projected, if you looked really close you could see a bit more noise on the HDV clip compared to the other two. Only when heavy camera movement (the operator was shaking the camera literaly) was introduced along with within the frame heavy action (stunts with bicycles on the air) did the picture become obviously softer. According to the Avid guy SKY CHANNEL was accepting HDV originated material but demanded uncompressed-or at least higher than HDV compression - post production workflow. And of course went forward to introduce to us their codec HDNXHD.
By the way, check http://www.ctt.ru/
A camera with a recorder that is up to the specs of VIPER or GENESIS with the addition of high speed (250fps at 1080) and with a price tag of USD18000 plus USD4800 for the field recorder. And they will support MJPEG2000 in the near future. Personally I'm impressed

esp esp@easy.com
 
Usual "HDV can't to horizontal pans with detail rubbish"

Total BS- HDV works fine- if you turn down detail level on the camera, most HDV cameras come with it set way too high.
 
Rubbish? Go try it yourself. Your either blind or never have. Detail doesn't effect the detail in the image. Its sharpening. Sure, it may make it more or less noticible but, the artifacts are created none the less. I own a Z1U and a Varicam and use both shooting motorsports for a living. Unless you can claim more experience dealing with HDV and horz pans I think I'll call BS on you.
 
I deal a lot with vintage aircraft like Mustangs flying at speeds of 300mph and taking off on grass strips(lots of detail in grass!) at 100mph, if there was a major issue I think I would have noticed it. We use the Z1 as a third unit with 2 HDCam cameras and sure there is some softening in a pan, even the HDCam cameras soften when you pan with the detail up, but I haven't seen ANY artifacting in our footage.
 
Try this. Put your z1 on a tripod and point it at about 10' from a parked mustang with lots of detail behind it. Tress, hangers whatever. Now slowly record a pan from one end of the plane to the other. Look at your playback on a good monitor and closely observer any markings on the plane with strong contrast and detail. Numbers and decals work great for this. Enjoy HDV the artifacts
 
Thanks for the tip, strangely enough we do that all the time and still no issues. I also get to watch back the footage via HD-SDI to a quality CRT monitor and still no problems apart from slight softening on the pan.

Guess I must be doing something wrong
 
I'm not talkng about HD-SDI captured material but, HDV playback from tape in case there was some confusion. BTW, how are you watching HD-SDI playback from a Z1U ?

Next time when actually watching playback from tape check any strong vert/diag/horz lines (such as black numbers on a white background) Watch the edges as you pan. If you can't see not much I'm going to say that will change your mind.
 
I mean Decklink HD-SDI monitoring of captured HDV material using the HDV codec in FCP.

I honestly haven't seen any artifacts and there are a lot of vertical black/white lines on WW2 aircraft.

Your experience is clearly different. I'm not saying you haven't seen it. I'm just saying I haven't.
 
uh, those demo frames of the russian HD camera sure *dont* look nice.. what's all this color noise?! (really poor debayering or what?)
still, it makes me a bit more at ease with a possible price of the RED.

as for HDV compression:
i havent got much experience with the cameras, but setting detail lower should definitely help a bit to avoid artifacts, since the codec can compress it easier (this even works with jpeg and the like).
if you sharpen again in post (preferably in an uncompressed codec) you'd also make the artifacts more visible again, but it should still look better than if you did it in camera pre-compression.

worth a try
++ chris
 
HDV doesn't suck and it works fine. Especially with the Z1 in good hands. Pans and tilts and even running with the camera shots look fine if you aren't a terrible operator.

I found it a little interesting how badly the HVX200 performed in these tests, though. Certainly, I'll be renting one before we commit to using this thing on a real project. And I wonder why Panasonic refuses to tell us the CCD specs.
 
I love how people just refuse that there could possible be artifacts in HDV footage. Its not really an issue up for debate. Trade offs were made in the compression scheme to get the datarate low enough to record to DV tape.

Nobody that developed and now sells this format will ever claim that there are no artifacts in HDV recorded footage caused by the compression. There are. Its just a matter of if the viewer will find them objectionable or not.

I think everyone here saying 'I've never seen artifacts in HDV footage on my camera" and "Must be the operator." Should really take a couple of workshops on video engineering (BAVC offers a good one) or maybe just stop trying so hard not to see them.

The question is not,"Are there artifacts in HDV footage?" It's, "Is HDV good enough to use on my project?" For a lot of people the answer is yes.
 
There are "artifacts" in every compressed tape format even HDCam it's just whether they are obvious or not. Some artifacts may be more obvious to an engineer, but fortunately I work as a creative person and not a number cruncher and it's the pictures I see and what my audience see that interest me, not the numbers.

Everyone I 've worked for using HDV has been delighted with the quality, we've even had footage broadcast on the Discovery Channel without ANY problems.

Theoretically HDV is crap compression - in the real world it looks great and it will only get better as they refine the compression as in DV.

HDV cameras are out there working for thousands of people as we speak if a few elitist tech heads turn their noses up at it, that's fine by me, doesn't affect the money I'm earning with it one bit

Long live the HDV revolution!
 
HDV is not a revolution. Its a off ramp on the road to where we are going. It is not really what you would call great. Its just sorta OK. That's not where we want to be.

After you've had your hands on a direct to disk/P2/BlueRay/HD-DVD/whatever camera that records in a less compressed format you'll look back on that HDV foootage and wonder what you were thinking at the time.
 
"HDV is not a revolution" - IMHO you're wrong on that

Sure HDV is just a transition - EVERY format is a transition. but It's great because it's CHEAP and it's available NOW, it democratises HD production for filmakers on a budget in the same way DV did - it's very simple HDV is the HD equivalent of DV, it has limitations but it's affordable and that is ALWAYS going to be a comprimise

Sure I'd rather have HDCam or DVCPro 100 but it's very expensive, sure the HVX is cheaper and I know it's the start of the next revolution, but P2 is too expensive and impractical for a lot of people at the moment. Do you think for one moment Panasonic would have released this camera if HDV hadn't arrived?

OK there's better stuff coming, but there always is, if you wait around for it moaning how bad HDV is then you might miss the boat.

I'm a filmmaker that has worked with most HD formats including uncompressed 10 bit HD, so I think I'm in a reasonable position to comment on this.
 
It sounds like you agree with everything I said but, still want to call it a revolution. Fine go ahead. I don't see HDV holding much market share outside of the people that can't afford to upgrade a year or so from now. Its not a long term format. Just look at the numbers of VTR's out to support HDV. If I was a betting man I'd say NAB 2007 will have no new HDV cameras.
 
Post a Comment


Links to this post:

Create a Link

This page is powered by Blogger. Isn't yours?

Listed on BlogShares