Digital Betacam vs MiniDV
 

I've observed footage from various MiniDV camera from US$400-$3500 on a LCD NTSC monitor. The quality is poor compared to what I see on TV. It appears that high quality TV shows like Young and Restless (see below) are recorded on Digital Beta. Cameras start at around US$25,000. Compression is much lower than MiniDV. How much better quaility is Digital Beta than MiniDV?

Article below: Young and Restless is recorded in Digital Beta. The light levels used are actualy quite low, around 100-150 LUX. OK, I admit to watching this show. It's my lunch time enjoyment. The quality of the picture is amazing even on my US$120 TV.

http://www.highdef.org/magazine/arch...-Dec_2001H.pdf

http://www.adamwilt.com/DV-FAQ-tech.html#FormatTable
The format itself is better than mini-DV if you are viewing the footage on high end monitors with a very clean signal chain. In real world scenarios where you are sitting at a distance from your TV, there is very little difference between the formats. Blue screen work is an exception, but there isn't a lot of that in soaps.

Thw two major differences between digibeta and mini-DV are production values and difference in the cameras. Digibeta productions usually have a lot of money put into them and are usually in nice, controlled conditions. Digitbeta cameras usually have ~$15k Fujinon lenses and 2/3" CCDs. They have more depth of field, no "optical" defects (vertical smear for example), superior low light, superior color accuracy, and better manual controls.

High End MiniDV
 

Thanks Glenn for pointing me to this article. I'd seen it before but forgot about it.

So is it true then that the camera makes a far greater difference on final video quality than the video format.

Using Adam Wilts numbers from 1-10, he rates VHS at 3.5. If I compare VHS movies I rent or broadcast TV saved to VHS this would be a maximum of 3.5 according to him. If I compare this VHS image to video produced on an under US$3,500 video camera, I would have to give prosumer MiniDV about 1.0 and cheaper camcorders about 0.5. But Adam Wilt gives MiniDV a number of 9.0. Does MiniDV quality vary that much, from 0.5 all the way to 9.0? What type of MiniDV camera even come close to 9.0?

Yeah, I bought a MiniDV camcorder because I saw numbers like this, but when I got the camcorder, I realized that common MiniDV cameras aren't anywhere near this quality.

Adam's ratings are actually very fair, but he's comparing video formats NOT cameras.

Say for instance, you have a most excellent video camera, but it's just a camera - it produces a video signal you're going to record onto tape. You have a choice of tape formats you can plug your camera into. Adam is rating the tape format for their picture quality in recording a source.

In most situations, there is very little quality difference between miniDV and DigiBeta - but only if they are recording the same source video or are connected to the same brilliant camera.

All miniDV cameras record with equal quality in terms of tape format, but it's the lenses, CCDs and other circuitry, and the skill of the cameraman which determines the finished quality which you are comparing, and as you can see, there are way too many variables to compare easily.

Graeme

The numbers only apply to the format. If you feed the format a perfect picture, then the results should stack up close to his ratings. However, the format the pictures were shot is just one factor in the overall quality of the image. Things like color "correction" can make dramatic improvements to image quality. Lighting is also very important.

Decent mini-DV cameras would be the VX2000/PD150 and DVX100. That gives you a high quality image for a few thousand. Cameras like the Sony DSR500 ($10k or $20kish, not sure) have interchangeable lenses so you can put a nice $15k lens on, 2/3" chips, and other features that make it a high end camera.

mini-DV is the best consumer format right now if buying new. If you're ok with converting hi8 to DV then you can pick up a 3CCD hi8 camera, but the hi8 format is not as good as mini-DV for getting footage. You might be able to pick up a nice hi8 camera though with things like real manual controls, lenses that can rack focus, 3CCDs, etc.

If this helps any:

As a cinematographer, I would request copies of projects I have worked on in DigiBeta (they usually originated on 35mm or 16mm). The process of building demo reels from this footage was previously cumbersome and costly, since I would have to beg, borrow or steal my way into a Digitbeta online suite. I was preparing to sink 20K+ into a Media 100 edit system a few years back when (thankfully) Final Cut 1.0 came out and I got a chance to see what it could do with Digibeta footage transferred to MiniDv. I bought a system with a DSR-11 DVCam deck, and starting cutting my reels at home.

With pristine film footage dubbed to DV, the image is lovely. Yes, if placed side by side I can see the difference between that and Digibeta, but viewed by itself it's impressive, and I considered the price effectiveness to be worth it. Nobody has ever noticed the compression to be a problem when viewing my reels. Ironically, DVD is just starting to become a viable format to distribute demo reels, but believe it or not 3/4" is still the predominant format out there!

So to echo what has been said before: it's all about the source material (camera or original format of footage). DV is surprisingly good for the price point.

http://www.matternvideo.com/formats.htm

I am surprised 3/4 inch is listed as the same S/N ratio as Hi8. I always thought Hi8 was noisy, but I guess it was just the cameras matched to it? In any case, Hi8 is dropout city.

Nice link, but I can't believe they have all their facts straight - they list a different resolution for DV as for DVCAM - they're in fact identical.

Also, they have D-1 as a lower resolution than DV, which again, I can hardly believe - they are both going to record to the same 601 format digital video, just with compression on the DV (and those 6 extra lines missing, but they don't effect horizontal resolution)

If you're going to compare formats, you have to do it with equivalent cameras. Comparing DVCAM shot with a $4,000 camera to DigiBeta shot with a $50,000 package is not very meaningful. If you take a camera like the DSR500/570 (about $18,000 plus lens...you can get a fairly nice setup for around $25K or so) and recored something to DVCAM and at the same time record to a Digibeta deck via the 26 pin output, then play the tapes back, the Digibeta will look somewhat better, but not overwhelmingly so under many conditions. Still, the less compressed the format, the better the look. A DV50 format recorded with the same camera will look better than DV25. But, if you took a low end DV camera, say a VX2000, and recorded to Digibeta with it and recorded the same object to DVCAM with a DSR500, I would guess that most people would like the look of the DVCAM over the Digibeta. It's the camera more than it is the format these days. Unfortunately a lot of people who knock DV as poor quality usually have only seen stuff shot with the lower end cameras.

Oh certainly yes, mini DV with a high end camera would easily without question crush DigiBeta from a VX2000.

DigiBeta not only has less compression but also has more color bandwidth.

Think of miniDV as being like audio MP3 at 128kbps and think of DigiBeta as a real CD. You can perceive a difference if you look for it, but the difference is not obvious to most people nor annoying or detrimental to the experience.

But a VX2000 to DigitBeta would be like making a CD from a recording from a SoundBlaster microphone, while a high end camera to miniDV would be like listening to an MP3 file of a recording with all high end microphones and preamps.

MiniDV compression
 

Thanks for all the feedback. This is a big revelation that Digital Beta is "really" just a hair above MiniDV in quality if shot with the same camera. So it's the camera quality that counts most. I wish there was a list that showed how different cameras compared to each other in quality all the way from low end comsumer to highest end commercial. I thought a US$3,500 camera such as the Sony VX2000 was already nearing top quality MiniDV. I guess not.

I read Adam Wilts comments about DV artifacts such as mosquito noise, quilting, and motion blocking. I sounds like he thinks that these artifacts are mostly caused by the compresssion of the MiniDV format. What does compression have to do with camcorder quality? So here is my big question. Do high end MiniDV cameras compress the DV image better than lower end cameras?

http://www.adamwilt.com/DV-FAQ-tech.html#FormatTable
http://www.adamwilt.com/pix-artifacts.html

Well, no and yes. All MiniDV camera's use the exact same DV
format and algorithm. BUT, if a camera has better optics, better
electronics and better software algorithms this will yield a far
better picture (usually less noise etc.) that actually might change
how it is compressed as well etc. Basically there is a DV codec
inside each camera and we know for a fact (when looking at the
difference in DV codecs on computers!) that how well a codec is
written will have a huge impact on how well it compresses
footage (not in the amount of compression, but in quality).

DV Codec
 

Thanks Rob for this information on DV codec. Could it be that many prosumer camcorders have reasonable lenses and CCDs, but that the manufacturers are just dumping cheap software into these cameras? To push people on an endless upgrade path maybe? I tell you my Panasonic PV-DV852 camcorder has about the same glass quality and same area of each photosite as my Nikon CP5400 digital still camera, but the Panasonic has vastly worse video and still pictures than the Nikon. It looks to me like the Panasonic has done a horrible job of compression, exactly the same defects as I've seen on images taken with early digital still cameras.

I don't think there is a difference in DV compression between the high end or low end. If there is, it seems like no one has noticed and that Sony isn't advertising it.

Images from high end cameras should be slightly easier to compress. They don't have image sharpening turned up a lot, which would reduce mosquito noise slightly. Less noise overall makes the images easier to compress. I don't think the difference in compression makes too much of a difference though. I never notice any of these artifacts in real life footage.

Ben, your Nikon still camera is a 5.1MP camera and the Nikon optics are superior to what is on your Panasonic. You really can't compare them. Digital files are resolution limited, meaning poor resolution will show first. Noise is important, but most digital files, given proper exposure, will show little noise. The low resolution CCD and optics have the greatest effect on your image quality.

Actually the VX2000 or VX2100 has a very decent picture quality... it's closer to the upper end than to the lower end of the DV spectrum. The same image quality as the PD150/PD170. In the same lot you have the DVX100 and XL1... with the difference that the XL1 has a real lens mount. Conditions have to be very special for more expensive cameras to produce much better video than these models. The important step upwards is having a 'native' 16:9 imaging block, which with one or two exceptions costs much more money.

More confused than ever!
 

I compared my Panasonic PV-DV852 to a Panasonic DVX100 camcorder and viewed on a LCD NTSC monitor with S-video connection (not the highest possible quality connection) in a camera store. Under reasonable indoor light, the general picture quality of my PV-DV852 was close to the the DVX100, both cameras in automatic. Even the DVX100 picuture I saw, I wouldn't consider to be the same quality as a VHS tape I rent. The artifacts on both these cameras are horrible compared to commercial VHS tape productions I view. According to Adam Wilt's chart, if VHS tape tops at 3.5 I'd give the DVX100 about 1.5, and my PV-DV852 about 1.0 and lower cost DV camcorders 0.5.

Now Ignacio claims that the DVX100 puts out a quality of almost 9.0 under average conditions, but I'm seeing only about 1.5. What is going on?

Ben:

What are these "commercial VHS tape productions" that you are viewing? Are they movies that originated on film? If so, as discussed earlier, they will always look different (most would agree better) than the output of a DV camera.

Again, there are two entirely separate technologies being discussed here: DV as a recording format vs VHS recording, and DV cameras such as the DVX100 vs camera X. If camera X happens to be a 35mm motion picture camera, it will win regardless of the recording format (VHS, DV, DVD etc).

Yeah, it sounds like there is confusion here between the source and the media.

Commercial VHS recordings
 

Charles, yes the commercial VHS recordings are usually shot initially with 35 mm film. But I find that high end TV productions (like Young and Restless) shot with studio video cameras are almost equal in quality to those shot with 35 mm film (viewed on cheaper TVs). So I am assuming both would have about the same quality recorded to VHS tape (they are when I tape them).

The quality of 35 mm film and high end video converted to VHS appear to me as way above the quality coming from a $US1,500-3,500 video camera.

Am I missing something. Can video from a $US3,500 video camera be highly processed to eliminate 95% of the mosquito noise, quilting, and motion blocking that this price range of camcorder produces, to almost equal commercial high end video? That would require a 10x increase in video quality.

You probably watch such video from a camcorder on TV every day and not notice it. When a decent camcorder is used with skill and after being broadcast on TV, it's nigh on impossible to tell. The TV broadcast process (digital or analogue) really ruins picture quality to such a point that all you're seeing is the grossest differences in picture (ie video v film).

The quality of even the cheapest DV camera far outweighs VHS in every aspect . The only reall difference is that on TV productions they have people behind the cameras who know what they're doing. You can try and quantify the difference between DV and DigiBeta all you want, or wether it's the camera not the tape format that makes the difference, but really, when it comes down to it, it's the person operating the camera that makes the most difference.

Graeme

Probably the easiest way to compare the images from a studio camera and a DV camera is to go to a convention like NAB, Showbiz Expo or the DVExpo. The major manufacturers have their cameras lined up side by side shooting the same set and you can just walk alongside the monitors comparing image quality.

My personal feeling is that while there are certainly quantifiable differences between the two levels of camera, it's probably not as much as the price difference. DV cameras are amazingly good and getting better each generation.

Graeme's points about craftsmanship are well taken also.

Ben, you can see mosquito noise, quilting, and motion blocking on DV video? With most real life footage those artifacts don't appear.

IMO, most of the difference in quality is due to color correction, superior lighting, and steady camerawork. You can gain dramatic improvements from all three.

Here's one example of decent footage shot on a PD150 and the DVX100. http://www.kenstone.net/fcp_homepage...dvx_pd150.html

Do I have Alien Eyes?
 

Glenn says that mosquito noise, quilting, and motion blocking don't appear with most real life video. Everything shot with my Panasonic PV-DV852 has massive amounts of this. Indoors, outdoor, perfect lighting, tripod shooting, not moving the camera, slowly steadily panning shots. And every US$3500 and under camcorder footage I have seen has massive amouts of this compared to stuff I see on TV. Am I an Alien from another planet who has eyes that see things no one else does?

PD150 and DVX100 footage
 

Glenn, I watched the footage from your cameras. I'll agree that I don't see many artifacts in your clips. Did you process the MiniDV footage at all?

Those aren't my cameras. (I wish they were :P)

I think they processed the footage, but I'd have to read carefully to check. But anyways, you'd definitely process footage before final output.

Processing video from MiniDV
 

Is it possible that mosquito noise, quilting, and motion blocking can be largely removed from MiniDV and that is why my footage looks terrible because I don't?

I don't think anyone does that, and I'm not aware of any good tools of dealing with those artifacts. Most of the time those artifacts are not a problem. Maybe your camera is messed up or you're trying to shoot in low light and picking up a lot of grain (which you must confuse as mosquito noise???). DV also suffers from aliasing/stair-stepping, but in most real life footage you won't have a problem with that. Maybe you should be a little more specific as to the artifacts you are seeing.

Artifacts
 

Glenn what I see with my Panasonic PV-DV852 is the following:

1. Picture basically nowhere near as clear as I see on TV, even when picture is taken of non moving objects in normal sunlight. My pictures look out of focus, but they aren't. I think this is what is called mosquito noise, it reduces the clearness of the picture.
2. A halo around all objects. I think this is called quilting. For example shooting birds flying in the sky, one can see a white ring around the bird against the blue sky. Moving cars have same. Don't see this at all on TV.
3. Massive motion blocking, mainly inter-line twitter with pictures shot inside and outside. For example shooting a picture inside a camera store one sees the edges around the lights moving rapidly back and forth. Some pictures have so much of this it is hard to even view the picture, the whole picture seems to be vibrating. See little of this on TV.
4. All pictures I'm comparing are at less than +9 Db gain, and same thing happens outdoors in normal sunlight as well.
5. I've had the camera in to Panasonic twice and they say it's as good as it gets. All these artifacts are normal on even slighlty expensive cameras.

1- No that's not mosquito noise. Mosquito noise is explained pretty well on Adam's website. It's the crud around the text. Blurring your picture reduces mosquito noise.

I think your camera is just blurry. You can test the resolution of your camera if you like. http://www.bealecorner.com/trv900/re...x.html#EIA1956. You can kinda compare with the images at this Japanese site: http://babelfish.altavista.com/babel....html&lp=ja_en. Don't compare the numbers you get unless you take into account the Kell factor. Just judge the subjective blurinnes/sharpness.

2- Ok that has nothing to do with the DV format. Panasonic turned the sharpness setting up on the camera to fake sharper pictures. The side effect is halos on areas with a high difference in contrast. Your TV may also have image sharpening turned up, so this makes the halos even worse. You should be able to turn sharpening down if your camera has that manual control.

3- Nothing to do with the DV format. It has to do with the electrical system in your country (25hz AC or 30hz), what kind of lights you are shooting, and the shutter speed of your camera. Some cameras have special shutter speeds that compensate for this.

The interlacing of the NTSC video format can also cause some problems. Thin high-contrast 1-pixel lines will flicker on a TV set. You can fix this in post or shoot progressive.

Motion may appear strobey if you shoot with a fast shutter. Films on TV are always shot with a slow shutter, and you can too. Nothing to do with DV versus digibeta.

4- Any gain above 0dB will add video noise (not really grain as I mentioned earlier). Nothing to do with DV versus digibeta, just how your camera works.

I'm thinking you are watching some interlacing artifacts and
whatnot. Do you have the ability to put some samples up on
a website for us to see? Raw DV would be preferred or
uncompressed BMP stills. If you put up compressed movies it
will introduce new problems which aren't in the footage.

Basically I've almost never seen any of what you describe with
my XL1S and I think the image looks WAY better than what I'm
getting on my analogue cable TV.

Basically talking about such issues is very difficult without the
pictures behind it. If you can't host pictures yourself we can
probably work out some other place to host them so that
everybody can see!

Sorry if this is a stupid question but are we talking about DV or could this be video coming out of the camera in MPEG4 or some other low bandwidth codec? What you describe sounds surprisingly similar to my camera's MPEG1 memory stick video mode.

I'm talking about DVout and S-video out
 

Ignacio, the output isn't MPEG4. Basically it's the image stored on my MiniDV tape. My output from the camera is generally via the composite AVout port on the camcorder and to the AVin on the TV, VCR, or NTSC monitor, but I've also viewed the output via S-video out on the camcorder to S-video in on a NTSC monitor which is a bit better.

Here is a 3 second DV clip showing inter-line twitter
 

Here is a 3 second clip (9 Mb) of some footage from my Panasonic PV-DV852. Generally right mouse click over the hyperlink will allow you to save to your hard drive.

The footage is stock AVI (same as MiniDV I'm told) and was not compressed in any way. The clip was actually extracted from my camcorder via the Firewire, or Ilink port into my computer. Quality was the same with two different trial capture software, one expensive and one innexpensive, and is typical of what I see on the TV, and NTSC monitor as well.

There is a lot of inter line twitter of the picnic table. This is typical of pictures taken indoors and outdoors at all different settings.

Note that many computer monitors will show even more inter-line twitter than a TV. Better TVs and NTSC monitors show a little better quaility but the effects are still very much there.

http://www.benwiens.com/temporary.html

XL1S
 

Rob you mentioned: "Basically I've almost never seen any of what you describe with my XL1S and I think the image looks WAY better than what I'm getting on my analogue cable TV."

That is what's so frustrating. People are mentioning that their MiniDV footage is superior to what they see on TV, yet my footage is much worse that what I see on TV. Yet my camcorder is a high end 1CCD, suggested retail price was originally around US$1,600.

Maybe it's broken?

If I play something I've shot off my XL1 or my ZR40 right to my TV, the picture will be crisper and sharper than any VHS tape I own.
However, despite its crispness, it may not be as pretty if that VHS tape is showing a movie that has been shot on 33mm film, with a professional lighting crew and DP.

1CCD cameras need to have a low pass filter, which is designed to blur high frequency information (to prevent false colors). Unfortunately they also blur some low frequency information, which makes the image blurrier.

3CCD cameras like the VX2000/PD150 and DVX100 tend to give better pictures than 1CCD cameras for various reasons.

The weirdness on the brown picnic you see is probably a combination of a number of things.

As to why your footage doesn't look as good as TV, it's not because it wasn't shot on digibeta. Lighting, post processing, steadiness, and the camera are major factors in determining image quality.

Ohkay, I checked the footage and it basically it looks very good
on my laptop. Yes I see some stairstepping on the table which
is probably due to some added sharpness your camera is doing?
Have checked the manual to see if you can change this?

Is my XL1S footage looking better? Yes, but not as much as the
difference between VHS and your footage. Basically, I don't have
any VHS tape or TV transmission that is better than your video
in overal quality. Your footage is a bit overexposed (hot) though
and has some minor problems like stairstepping (aliasing) on
the table and a bit of flickering in the fence.

But those are only some minor problem that might disappear
if you can turn the sharpness down or perhaps expose a bit
better?

It sounds like somehow you are seeing a lot of extreme things
that I'm not? Perhaps something else is wrong?

I checked the footage again in my NLE and it has indeed a bit of
extreme stairstepping / pixelation. But only on those non-straight
lines. But on a TV you should not even see those (normally).

I'm not too familair with your camera, so I don't know if this is
normal for that cam or not. Again, this might be due to heavy
sharpening the camera is performing. Check for that first!

you shoot against the sun.
no camera will produce good image that way friend