What do you mean faking the extra resolution? if it looks good to the eye then the resolution is there! Those two words are almost an oxymoron. Since the JVC is already in 720P, there is no need to uprez it using S-Spline. I do'nt see the "cooking" of the test as you refered to. As far as I can tell, it was a simple re-size. You can't judge the colors of the DVX purely by this single example shot. There has been PLENTY of impressive full-rez samples of DVX footage online and not so much good full-rez ones of the JVC. If you want to see examples of how DVX handles color look no other than this guy's website:

http://www.phocinema.com/vietnam/index.htm

Those are perhaps some of the best examples of what Digital camcorders can do. But eithercase... you'd be foolish to say a single CCD video camera can produce colors and latitude better than a 3CCD. Also you pointed out that the walls and the chair gives away the differences between the two cams. Yeah, the chair on the JVC is a total blur! However, if you look at the computer monitor in both shots, you'll see almost the same amount of details, with the JVC perhaps a little better (But that's to be expected). Now I'm not proclaiming the S-Spline is a perfect solution, since the rescaling takes so long.. but on the other hand, so does Magic bullet, and alot of people can't live without it. So I guess if your goal is for a HD projection, then its definitely worth it to spend the extra time at the end to do the s-spline up rez. I can't wait to see what other clever algorithms lies ahead for us.

I mean the extra resolution is faked. It's clearly not in the original camera footage because it cannot be. The extra pixels are interpolated from existing data and then sharpening is applied. The resulting footage is much more than what came from the camera.

Yang, you appear to feel like it's proper to compare two cameras by applying unlimited processing to one and none to the other. More power to you, but I certainly won't rely on your judgement for my opinions. If processing were allowed on both, I'm sure the JVC would have showed better, especially in color. I never said processing was bad, I simply pointed out it was unequal.

I watched all the provided footage on a very high res computer monitor at double size. The chair back in the JVC footage is vastly superior to to the DVX. The back wall is noticably different, some much so that some speculated there was a DOF issue. I don't believe so since the computer monitor was similarly fuzzy in both shots. The digital display and the chair are both in the foreground and there's a noticable difference in quality.

My point about the color is that I didn't like it in the demo shot. That opinion is in contrast to others who've said it was better. If I were processing the footage, I'd adjust the saturation in both shots but in opposite directions. I'd also fix the white balance on the DVX. There were color artifacts in the DVX that I didn't like (the right shoulder, for example).

I didn't say "a single CCD video camera can produce colors and latitude better than a 3CCD". I said that believing 3 CCD's is inherently better than 1 is misguided, especially for subsampled chroma formats. I let the engineers decide how to build a camera then judge its performance. This is in contrast to 3CCD fans who summarily dismiss a camera based on a specsheet. I personally believe that the advantage of 3CCD is in sharpness. That's certainly the case in the still world where there are definite analogs.

The dvx footage WAS SHARPENED. The unsharp mask feature was USED in the blowup to 1280 size. How much more clear can that be? I tried it without the UNSHARP MASK sharpen, the results are slim to none as far as getting detail like the JVC cam.


As Craig stated, it's unfair to provide 6 seconds of CPU image processing to each frame of the DVX and none to the JVC. To be fair, the same process should be done to both.
Some might say ' the JVC is at 720p already!'. Well, I say , why limit it to 1280, baby! Bring on 1910.

I did some JVC stills at 1910, and the wonderful s-spline pro can make that look pretty stunning as well.

All that aside, it takes a bunch of work to actually do the S-spline pro uprez-sharpen to the footage. It's not a drop in, turn on plugin. You have to use still frames. It does not work on movie files.
I think S-spline pro, and some of the other advances upres sharpening tools out there are very good tools. Perhaps in the future that will make a NLE plugin as well.

Yes, the sky is still falling on the DV res ( web cam res ) res world. Even if it has 8 chips ;). That's technology.

-Les

Les Dit

why dont you try down-sample your hd1 to sd and sharpen it and compare to dvx to see if the sharpening function does so much different as you claim?

Why would you want to sharpen the downsampled HD footage to compare it to the DVX footage? The HD footage already looks considerably sharper than the DVX footage, at least in this example:

http://66.78.26.9/~fiercely/DVXvsJVC/DVXvsJVC.avi

It does. Sharpening, by its very nature, creates detail that isn't already there. Downsampling, then sharpening, proves nothing.

What we have here is upsampling through interpolation followed by sharpening. The upsampling creates higher potential bandwidth in the format. Sharpening fills it in. The added detail is false detail but is pleasing due to the nature in which it's added. The resulting image has higher frequency content that the original. I guarantee it.

I don't understand why this is controversial. No one said the technique is bad, but you can't compare the performance of two cameras that way. People here want to accept this as proof that the DVX has nearly the resolution of the JVC when it does not. It is proof that you can get good image quality at HD res from the DVX through careful upsampling. It is equally true that you can do it to the JVC, like Les said.

why cant compare this way?
dvx can do HD by just some resizing, and its image quality blows the hd1 in HD away, this is done by some software tweaking.
as for the hd1, you cant get a 1ccd look like a 3ccd by color correcting(software tweaking).

I can't see how you arrived at either conclusion. The DVX didn't do HD at all and the JVC wasn't color corrected, so there's no evidence to support either of your claims. As for image quality, there's no doubt in my mind that the JVC looked better in both resolution and color than the DVX.

To answer your question, you can't compare two cameras "straight out of the camera" by heavily post-processing one but not the other. I think that's plain enough.

For all those that claim that the DVX resolution is as good as the JVC, consider the following:

All the detail in the DVX footage is contained with the DV format. If it is the equal of the JVC, that says that all the JVC's detail could be contained in the DV format even though it's in HD. The logical conclusion is that you could downconvert the JVC to DV resolution, then upconvert that DV footage back to HD without losing any detail. Let's see it.

Anyone with the proper equipment could confirm that this not the case without wasting time on it. The HD format has much higher bandwidth potential than DV, so the JVC must not be using any of that if the DVX fans here are correct.

In my opinion the uprez'ed DVX footage looks good because the talent is very close to the camera. As you can see, no matter what you do you can't pull detail out of a blurry background, no matter what. If you are going to shoot a project where you always have tight shots, this may work, but throw in a wide or long shot and you will see that s-spline isn't going to save your SD resolution.
So far all of the A/B comparisons (Jon's included) have not been done in a way that allows people to see the resolution differance. The focus of the shots have always been a few feet away. Hell my 1chip consummer cam looks great when I shoot close shot.

This is what I would like to see as a definative A/B.
Shoot a person 25-30 feet away, against a wall. This will do away with DOF debates. Center the shot so that the head and foot room of the subject are the same from both cams. Do not worry about width matching unless you are using an anamorphic adaptor. Keep the shots in there origional size. Do not resize or resample. Let us compare a 1280x720 with a 720x480.
And judge!

I think it will be plain to see the resolution advantages of the JVC. No amount of s-spline is going to save the SD footage.
Ken

It's important for the test subject to have detail that challenges both cameras and I think that shooting at a distance will help provide that.

<<<-- Originally posted by Craig Jones : I can't see how you arrived at either conclusion. The DVX didn't do HD at all and the JVC wasn't color corrected, so there's no evidence to support either of your claims. As for image quality, there's no doubt in my mind that the JVC looked better in both resolution and color than the DVX.

To answer your question, you can't compare two cameras "straight out of the camera" by heavily post-processing one but not the other. I think that's plain enough.

For all those that claim that the DVX resolution is as good as the JVC, consider the following:

All the detail in the DVX footage is contained with the DV format. If it is the equal of the JVC, that says that all the JVC's detail could be contained in the DV format even though it's in HD. The logical conclusion is that you could downconvert the JVC to DV resolution, then upconvert that DV footage back to HD without losing any detail. Let's see it.

Anyone with the proper equipment could confirm that this not the case without wasting time on it. The HD format has much higher bandwidth potential than DV, so the JVC must not be using any of that if the DVX fans here are correct. -->>>


as i said, one can get a better video by software tweaking as the other cant no matter what "heavily post-processing" you apply , you cant make a 1ccd to look like a 3ccd. that is the difference.
the color thing maybe subjective, but the advantage of 3ccd over 1ccd is pretty like common sense.

or you can show us your color correcting skill that makes the hd1 look like the dvx, post here, then you can call your 3ccd hd1, and be proud of it. we can then get rid of the expensive 3ccd and go for 1ccd.

Then again, we're debating HDV vs. DV. That's like comparing, say, a 3 CCD DV to a 3 CCD Digibeta...Maybe a bit extreme, yes, but it's essentially two different formats.

heath

Seems like a dangerous thread to post opinions on. I have been staying away until now.

I don't think it matters how much processing you use to upconvert DV to HD, you can still favourably compare to HD source material to the up-res'd image. Using Barry's latest DVX100 to GR-HD1 comparison (the nice S-Spline version), I have high-lighted the obvious issues with both sources. In the end you can only get HD detail (not talking sharpness) if you start with an HD source.

See comparison:-
http://members.cts.com/crash/d/dan/temp/DVX100vsHD1.png

Whoa -- didn't mean to start so much trouble...

My point was, since the DVX was going to be upscaled (and in the first test I used Vegas) I thought it prudent to point out that much, much better upscaling software was available.

Remember, the original post included two clips: DVX and JVC at DV resolution, and DVX and JVC at HD resolution. I posted comparisons of down-rezzing as well as up-rezzing.

Remember, you CANNOT compare the two cameras straight off the tape, because they're very different formats! You've got to change one to match the other in order to compare them.

No other image manipulation was done: no color correction, no contrast, nothing. In one case, you have JVC downrezzed and compared to native DVX, in the other you have DVX uprezzed and compared to native JVC.

Obviously the JVC is going to hold an edge in resolution: you're talking about 864,000 pixels vs. 380,000 pixels! What's interesting here is that, using the proper software for resizing, you can get a much more comparable image from the DVX. The original clip I posted, where I used Vegas for the uprezzing, was simply unacceptable, very very soft. I think just about everyone would agree on this one point: the S-Spline up-rez is much more pleasing than the Vegas up-rez!

Diu,
You are following the sheep. 3ccd 3ccd 3ccd , that's old school buzz word stuff, the stuff people with a marginal grasp of the technology keep repeating like some kind of mantra.
Look, I can show you dozens of examples of a single ccd making a better image than a $100,000 3 chip camera.

Whaa Whaattt ? !!! ( my best Homer Simpson sound bite )

Yea, they are called digital still cameras. Oh my god!

So put aside the 3ccd issue, new things are happening.
3ccd in video cams is going to 'melt away', as readout rates and CCD tech gets cheaper-faster-better.
I'm not saying the JVC has the perfect single chip 720p implementation, but it's going in the right direction.
There are some mpeg2 issues on the JVC, sometimes I can see the flat areas quantize a bit, but on the whole, it's a refreshing image to look at because it *kinda* matches scanned film in detail.
Far less jaggies and other low res artifacts than DV footage.

Diu, did you see what effect the SAME PROCESSING does on some JVC footage? How did you come to the conclusion that " no matter what heavy processing" can't get the JVC looking better?
Look before you leap.
Did you test it?

-Les




<<<-- Originally posted by Diu Hai :
as i said, one can get a better video by software tweaking as the other cant no matter what "heavily post-processing" you apply , you cant make a 1ccd to look like a 3ccd. that is the difference.
the color thing maybe subjective, but the advantage of 3ccd over 1ccd is pretty like common sense.

or you can show us your color correcting skill that makes the hd1 look like the dvx, post here, then you can call your 3ccd hd1, and be proud of it. we can then get rid of the expensive 3ccd and go for 1ccd. -->>>

Barry: Please tell the forum that you used the 'graphic' preset on S-spline, AND THAT DOES INDEED DO MORE THAN A RESIZE.

Sorry for the caps, but it DOES AN UNSHARP MASK SHARPEN as well.
Maybe the caps will help people see that :)

If you leave off the sharpen, the effect is slim to none, over a bicubic upres.
That would be a more fair test.

<<<-- Originally posted by Barry Green : Whoa -- didn't mean to start so much trouble...

No other image manipulation was done: no color correction, no contrast, nothing. In one case, you have JVC downrezzed and compared to native DVX, in the other you have DVX uprezzed and compared to native JVC.
-->>>

Les & Barry,

As I just commented (see http://members.cts.com/crash/d/dan/temp/DVX100vsHD1.png), it really doesn't matter whether you do any post processing, it is still possible to tell whether a clip has been upconverted. S-SPline does allow a DV image to have equivalent sharpness, but not equivalent detail.

Stilll, it is a great technique if you need to mix SD material within an HD production. I have been doing slow motion experiments mixing 480p60 with 720p30, the 480p60 could do with some sharpening. Note: The slow motion still looks awesome.

The problem people are upset, thinking that this operation is cheating as a form of comparison. Maybe a fair test for those who question the technique is to upconvert them both 1920x1080 and compare them at that res. :)

David,
Yes, I saw your test pic snippets, they were good and to the point.

I like your "both to 1910 upres" idea as well, I think I mentioned that as well, in one of my rants!

-Les

<<<-- Originally posted by David Newman : Les & Barry,

As I just commented (see http://members.cts.com/crash/d/dan/temp/DVX100vsHD1.png), it really doesn't matter whether you do any post processing, it is still possible to tell whether a clip has been upconverted. S-SPline does allow a DV image to have equivalent sharpness, but not equivalent detail.

Stilll, it is a great technique if you need to mix SD material within an HD production. I have been doing slow motion experiments mixing 480p60 with 720p30, the 480p60 could do with some sharpening. Note: The slow motion still looks awesome.

The problem people are upset, thinking that this operation is cheating as a form of comparison. Maybe a fair test for those who question the technique is to upconvert them both 1920x1080 and compare them at that res. :) -->>>

The hi-res test would be interesting.

David, good choice of sample area. Points 2 and 3 were what really bothered me.

Diu Hai, old prejudices die hard, especially when you use prejudice itself as proof. I'm not predicting the future of 1CCD vs. 3CCD, but they are just design decisions that engineers make. In the still world, resolution and color are king, yet engineers use 1CCD almost universally. Let the engineers do their job and judge the cameras rather than the specs.

What is it specifically about 3CCD that can't be duplicated with 1CCD? I wouldn't want to make the JVC look like the DVX in this test.

<<<-- Originally posted by Les Dit : Barry: Please tell the forum that you used the 'graphic' preset on S-spline, AND THAT DOES INDEED DO MORE THAN A RESIZE.
-->>>

I used the "photo" preset. Whether it does other processes or not, I don't know. All I know is that it made an attractive, and fairly competitive, up-rez. Obviously not as detailed as the true HD of the JVC (as easily discerned on the diagonal lines) but not too shabby either.

<<<-- Originally posted by Les Dit : Diu,
You are following the sheep. 3ccd 3ccd 3ccd , that's old school buzz word stuff, the stuff people with a marginal grasp of the technology keep repeating like some kind of mantra.
Look, I can show you dozens of examples of a single ccd making a better image than a $100,000 3 chip camera.

Whaa Whaattt ? !!! ( my best Homer Simpson sound bite )

Yea, they are called digital still cameras. Oh my god!

So put aside the 3ccd issue, new things are happening.
3ccd in video cams is going to 'melt away', as readout rates and CCD tech gets cheaper-faster-better.
I'm not saying the JVC has the perfect single chip 720p implementation, but it's going in the right direction.
There are some mpeg2 issues on the JVC, sometimes I can see the flat areas quantize a bit, but on the whole, it's a refreshing image to look at because it *kinda* matches scanned film in detail.
Far less jaggies and other low res artifacts than DV footage.

Diu, did you see what effect the SAME PROCESSING does on some JVC footage? How did you come to the conclusion that " no matter what heavy processing" can't get the JVC looking better?
Look before you leap.
Did you test it?

-Les

-->>>

put the 3ccd's future blah blah aside for a moment.
just make your 1ccd look like 3ccd, or at least make the color of hd1 comparable to dvx.
oh and remmeber to post it here when your done

<<<-- Originally posted by Craig Jones : <<<-- Originally posted by David Newman : Les & Barry,


Diu Hai, old prejudices die hard, especially when you use prejudice itself as proof. I'm not predicting the future of 1CCD vs. 3CCD, but they are just design decisions that engineers make. In the still world, resolution and color are king, yet engineers use 1CCD almost universally. Let the engineers do their job and judge the cameras rather than the specs.

What is it specifically about 3CCD that can't be duplicated with 1CCD? I wouldn't want to make the JVC look like the DVX in this test. -->>>


what we are arguing is that you said it was not fair because no tweak was done to the hd1, and that is why dvx blows hd1 away. and i asked you to do some tweak to the hd1 so that its color can be comparable to the dvx. if you can tweak hd1 to look like the 3ccd dvx, it wins, or it loses. its that simple.

My conclusion so far:

If you have optimum lighting, you can get stunningly detailed images with the JVC, which are unachievable by any DV camera, but it is technically difficult to get consistently good footage. There may be some issues with colour, but there is some disagreement about this.

It is much easier to get good video footage with the DVX, and, when this is up-rezzed to 1280x720 p using s-spline pro, it looks superficially like HD. Some people prefer the up-rezzed DVX to the native JVC 1280x720.

It is possible that, with image processing software, the subjective quality of the JVC HD footage could be improved further. (and probably the DVX too)

Fair?

Patrick

Diu,
Is the color the biggest issue for you?
Specify what elements of the DVX images were causing you to think the DVX 'blows away' the JVC.
Perhaps you don't care about extra detail?
Perhaps you would prefer to ignore that aspect, I don't know.

BTW, the s-spline 'photo' preset does an unsharp mask of radius 2.5, so yes, it *is* more than just upresing.

Also, it's not about 'wining' or 'losing', jeez, it's a tool, a camera. What is more important is what you use it for.
It's like the still photo consumers, many of them are constantly getting better lenses, bodies, accessories, in the vain hope that somehow it will help their photographs like like the ones they was in a book store of exhibit. It doesn't work that way.
But this is a technical forum, and it's fun to talk tech issues too.

I'll see if I can post a still from that JVC footage that looks very close to the undoctored DVX . I did an experiment before, and it turned out if I did a BLUR of 1.8 pixel radius on the JVC, it knocked out enough of the detail to make it look very similar. I saturated the color a bit to get you those candy coated colors the dvx makes.

Cheers,
-Les




<<<-- Originally posted by Diu Hai : <<<-- Originally posted by

what we are arguing is that you said it was not fair because no tweak was done to the hd1, and that is why dvx blows hd1 away. and i asked you to do some tweak to the hd1 so that its color can be comparable to the dvx. if you can tweak hd1 to look like the 3ccd dvx, it wins, or it loses. its that simple. -->>>

<<<-- Originally posted by Diu Hai : <<<-- Originally posted by Craig Jones : <<<-- Originally posted by David Newman : Les & Barry,


Diu Hai, old prejudices die hard, especially when you use prejudice itself as proof. I'm not predicting the future of 1CCD vs. 3CCD, but they are just design decisions that engineers make. In the still world, resolution and color are king, yet engineers use 1CCD almost universally. Let the engineers do their job and judge the cameras rather than the specs.

What is it specifically about 3CCD that can't be duplicated with 1CCD? I wouldn't want to make the JVC look like the DVX in this test. -->>>


what we are arguing is that you said it was not fair because no tweak was done to the hd1, and that is why dvx blows hd1 away. and i asked you to do some tweak to the hd1 so that its color can be comparable to the dvx. if you can tweak hd1 to look like the 3ccd dvx, it wins, or it loses. its that simple. -->>>

I did not say that, because the DVX does not blow the JVC away. In fact, quite the opposite. You assert that the JVC's color cannot possibly be as good as the DVX simply because no one has has modified it to look the same. The JVC's color is better than the DVX in this clip as far as I'm concerned.

As far as your rules about who wins and who loses, it doesn't matter. You define what's good simply as whatever the DVX produces. As I said before, old prejudices dies hard. The JVC wins this contest and doesn't need to be modified to do so.

The JVC wins this contest and doesn't need to be modified to do so.

Amen to that brother! :)

Troy

In terms of VIDEO a 3CCD system setup will be superior to a 1CCD chip in terms of colour, why beacuse a photosite (pixel) is a transducer it converts the photon that hits it to a small charge, the more the photons the larger the charge, anyway like many transducers the response may not be linear it may be some other non-linear response curve...
ok ok cutting to the chase the pixel represents an aprroxmimate luminance level, this is all very fine if you want to record black and white images but what about the colout? well back in the old days (near the begining of TV) one method was to spin a transparent mult-coloured wheel that had different coloured segments (red, green, blue) in front of the camera lens then on the TV sets it too would have the same wheel and buy spinning the wheel at the same phase (ie when the segment say was red past the camera lens at the same moment in time the wheel in form of the TV would also be red) this was a very crude method NOW back to the present day when using a 1CCD chip in order to get colour information you have to mask/filter off some of the pixels or photosites and you guys know the rest... now this method means that the colour information is only an aprroximation, but WAIT engineerings came up with a much better solution way not split the light into its primary colours and record them individually and thus avoid all the filtering and bayer matrix and pixel shifting issue all together.

Now you get some people who say " my digital camera can make images that are like waaay superior to any video camera and ITS a 1CCD" thats because its DESIGNED to take STILL images. It has many many more pixels then any video cameras so even after considering that not all of these pixels are used effectively (becasue you have to filter/mask some of the pixels to derive colour information) the large pixel count compensates for that, thats the reason why you need such high megapixel count to get good images. And since you are not demanding video the DSP has TIME to gather enough information to make a good image.

Anyway if you dont believe me then why cant a digital camera be used to make video and vise versa, i.e try making a video from a digital still and you will see that it is by now way at any level to compete with even an old analouge camcorder.

And what if you had a digital still camera that had 3CCD's it would diffinately blow away all these single chip digital still cameras(now dont give me all this stuff about using a 3CCD camcorder to take a digital still, because a video camera is designed for VIDEO and a still camera is desgined for a STILL image you get it now)

To conlude:

a 3CCD system is better than a 1CCD sytem period.
Comparing a digital still is a null issue because they are designed for a different purpose(for stills) [hint: thats why they suck when it comes to taking videos]

one last comment about the whole dvx vx hd1

universal law: grabage in garbage out

therefor even if the JVC has a higher resolution format you can not say that it has a superior lens than the DVX,
secondly in terms of colour the dvx has 3CCDs, the JVC may be recording (or trying to) a higer resolution but it has to sacriface a lot of pixels to achieve colour samples, as a result you are goinging to be losing resolution (add the fact you've got a poor lens) then to top if all of HD needs a higer bandwidth and mpeg2 just doesn't cut the musturd beacuse for gods sake during acqasition stage you want to reatin as much data as possible and using mpeg is a lossy process you are just throwing buckets of data away.

(hmm interesting note, if 1CCD was soo good why does CiniAlta use three?)

the JVC cam is like on of those cheapo digtal still cameras that you see advertised in newspapers you the one that are made by some cheapo company that claim to be 4megapixel at only £100 or something like that, then after you buy one and take a picture and compare it with like a 4meagpixel sony or say Kodak you realise how much the Sony/Kodak is superior. Becasue the cheapo camera is not actually 4meagpixel (its hardware interpolated) and the lens is poor quality it may not even have CCDs (most likely cmos).

So When JVC said it had a HD cam at a consumer price, lots of silly people got excited and thought that they had just brought a ciniAlta ...

Look i'm all for HD i would love to have HD but even HD is being replaced takea look at whats happing with Dalsa.

i'm tried of typing.

See if I've summarized Anhar's points correctly:

- 3CCD is better than 1CCD because 3CCD's are better.
- Still imager design is irrelevant because still imagers aren't designed for video.
- Video cameras can't take good stills and still cameras can't take good video.
- People who think still and video technologies are related are fools.
- JVC video is garbage.
- JVC's mpeg2 compression cannot possibly be good, therefore it is garbage.
- JVC's lens in not superior to the DVX, therefore it is poor.
- JVC sacrifices a lot of pixels and its lens is poor.
- DVX's color must be better because it's 3CCD.
- 1CCD can't be good because CineAlta uses 3.
- The JVC doesn't do what it advertises, doesn't have the claimed resolution, and probably doesn't have a real CCD in addition to its tiny lens.


I can see that you've done a thorough objective analysis and have a firm technical grasp of the issues. That the JVC is such a "piece..." doesn't reflect well on the DVX since it was totally outclassed in this comparison.

Just a few interesting facts for you:


- 3CCD cameras may eliminate the Bayer pattern, but not the color filtering. The prism has problems of its own.
- Both cameras here subsample chroma. The JVC format offers nearly as much chroma bandwidth as the DVX has luma bandwidth!
- The Nikon D2h processes in excess of 32Mpixels/second. The DVX does at most 10Mp/s per CCD and the JVC does 30Mp/s. Still cameras have no problem keeping up with video speeds.
- Curious that the DVX at its pixel rates can't do HD. So much for the superiority of 3CCD.
- There are digital video systems with pixel counts in the high MP range.
- The Nikon D100 uses an interline transfer CCD, and architecture DESIGNED for video cameras.
- The optics of video and still cameras work the same way.
- There are very few aspects of imager design that restrict their use to still or video.
- No one who bought a JVC thought they got a CineAlta.

Everything you posted, Anhar, is based on prejudice and heresay. Old prejudices die hard. I suggest you read JVC's discussion on it's 1CCD imager, only this time with an open mind.

Everyone,

Let's stay polite here; we strive to keep this intelligent, informative, and a fun debate.

Just a friendly reminder.

heath

- The optics of video and still cameras work the same way.

Craig, I agree with most of your major points, but I think you need to add still digital cameras. Optics designed for film still cameras may show anomalies when used on digital still cameras.

<<<-- Originally posted by Jeff Donald : Craig, I agree with most of your major points, but I think you need to add still digital cameras. Optics designed for film still cameras may show anomalies when used on digital still cameras. -->>>

You are right. I just meant the optical principles don't change. The light doesn't know whether it's being recorded in full motion or still!

Thanks for the claification.

Anhar - OK, we all get your point.
3CCD still beter than 1CCD still.
3CCD video better than 1CCD video.


Unfortunately it isn't that simple.
See, what is being compared here is DV format to HDV format:
DV compression to mpeg2ts compression, 480p to 720p, 16x9 to 4:3. 1chip vs 3 chip.
All of these factors add up to the final image.
Out of all of the deficientcies this camera has, colour is not one of them. As well neither is resolution, which you say, should be gobbled up by the lack of 3CCD's.
Many, and my self included, agree that the colour from the HD10 is more real world then the DVX100's. Add this to the 720p resolution and you have a hyper realistic image.

I would also like to point out that not all CCD's are equal. There are some 1CCD DV format cams that have very comparable or better picture than some 3 CCD cams. And out of all cam makers JVC is known to be one of the best for 1chip designs.

Ken

Just one point on the optics issue:
One of the biggest jokes in the science of optics is selling video camera people lenses than have resolutions 5X what is needed for the format. It is pretty easy to make a lens that does a good job at 360 line pairs of resolution. But the Gucci effect really sells.
Wow, by camera has an XYZ lens, hey that means I'll be able to take better video with it! Rubbish!

The lens is not going to make much difference at 720 across.
Look at it this way, a $10 disposable camera with a plastic lens gets you way more resolution then *any* DV camera can record.

Yes, there are issues with how the focus ring works ( servo vs. manual ) and whatnot, but you don't need Gucci brand names with some kind of voo doo to make that work. Besides, the typical big names that Jap. cam makes use for 'lens endorsements' aren't very good at motor controlled lenses anyway. It's not there bag.
-Les

Look at it this way, a $10 disposable camera with a plastic lens gets you way more resolution then *any* DV camera can record.

There's more to quality optics than resolution. Lens designers also consider contrast, sharpness, color saturation, and control of flare and other optical defects.

Are you saying that the MTF of the plastic lens on the disposable lens is higher than most video lenses? Reading your post one might get that impression.

That's kinda what I'm saying, the MTF on the $1 lens is more than enough to support a web cam image ( DV ).
Notice that I said "can record" above.
Sure, lens design is more than sharpness, and they are interrelated in a strong way. But consider that in video cams, people ( including myself ) use filters ( Tiffen ultra contrast, black fog, whatever ) for reducing the high contrast, since the film plane can't take the contrast anyway.
The Tiffen Ultracontrast is basically a dirty lens element.

-Les

>>>>>>>>The JVC wins this contest and doesn't need to be modified to do so.

- Was just wondering if anyone has addressed how the HD10 renders vertical lines. I have yet to see the camera draw any vertical line as a straight vertical line. Everytime I have seen a live demo of the cam, vertical lines are never straight - they are drawn out very squigly, at times looking almost like a vertical high frequency sine wave or even a slalom ski course. Anyone notice this at all?

Also, has anyone been able to tame the "star-cross filter effect" anamoly often seen with small hot spots in the frame?

Not to start a war or a negative discussion here, but I haven't seen these things happen at all with the DVX100. I pointed this out to a DP friend of mine at the DV Expo and his response was "Thank you, I'm glad that I am not the only one who's seeing these things!"

If these side-effects are to be expected with HDV, then I would much rather shoot with a 'full-on' DVCPRO50 or 'DVCPRO100' camera with a sweet lens. Hopefully, this will improve with future HDV camera models?

>>>>>>>> The lens is not going to make much difference at 720 across.

- This is incorrect. We have seen repeatedly that increasing the resolving power and rectilinear response of the glass does have a noticeable effect - such as raising the degree of sharpness in a very measurable, tangible way and also decreasing chromatic aberrations. Photographers will agree that "it is all in the glass".

- don

I think you're greatly mistaken, Les. When tests have been done on different lenses mounted on an XL1s different lenses have made a profound difference in image quality. The tests (both objective and subjective) done by American Cinematographer several years ago bare this out.

My own personal shooting experiences using different lenses confirm this. Using various cameras, 2/3 inch chip and 1/3 inch chip, show vastly different quality images with different lenses.

Filters are one way of controlling contrast, as well as exposure and lighting techniques. But not all video users are looking to reduce image quality with various so-called enhancing filters. I think you need to take this into account in your sweeping generalizations related to DV and optical performance.

As for lenses, the requirements for a lens for a full HD camera (I'm talking the Panavision ones here, as they're the only ones I've used and know about) are greater than that of a 35mm camera. That's why Panavision came up with a whole new series of lenses for their HD cameras, because their 35mm lenses were not good enough...

I would then have to say that whatever lens you have on a DV camera, you'd need a better one on a 720p camera to do any justice to the resolution. Perhaps that's why the HD1 always has the sharpness set too high....

Graeme

Is their any objective basis for concluding that the JVC lens is inadequate?

Sharpness is one the areas that the two JVC differ. I personally believe that the shortcomings of the JVC are related to its shortage of pixel pounding DSP power. We know that it was a great challenge for JVC to put enough processing power into that size/weight/cost/power package at all. I would attribute the lack of user control and image quality issues to that fact rather than to the type of imager and lens though it's just speculation on my part.

I believe the simplest explanations are usually the right ones. With the JVC we have optics and an imager that aren't really that different, but they are coupled to a massive computer and compressor by DV standards. My wishlist for improvements would be double the processing power and improvements in manual control before a better lens and imager. That's not to say that lens and imager improvements wouldn't make a difference, of course, but I bet there's image quality left on the table simply due to DSP limitations.

I've still not seen any pictures or video from the HD10 to see how it's sharpness compared to the HD1.

More DSP power in any camera will help the picture, as well as having more pixels on the CCD than it needs for it's rated resolution.

What I don't get is why 18mbs for the MPEG2 stream on the HDV? It's on DV tape, which we know can cope with 25mbps, so why not use them all and squash some of those MPEG artifacts we're seeing? Why not record real 24p and save some more of those precious bits per second for the picture?

Also, can someone shoot some resolution charts with the HD10 and also some colour charts, so we can see truely what the colour on the camera looks like?

Graeme

I'm confident that the JVC only uses 80% of available DV tape bandwidth because they didn't have enough DSP power to generate the full 25Mbps steam. I believe JVC said that themselves.

I also believe the shortage of DSP power is the reason other manufacturers haven't pursued consumer HD cameras yet. They clearly didn't think it was possible.

When you realize the processing required to do HD and the MPixel/second read rates, you understand the decision to use 1CCD. 3CCD's at the desired resolution would be generating 100 MPixel/second of raw data. It's quite a leap to process that much data in the current size/heat/power package the JVC targeted. JVC's raw pixel rates are comparable to existing 3CCD DV cameras. It's processing requirement is greater because of the mpeg compression.

Eventually, silicon process technology will allow DSP's to provide the processing necessary and consumer 3CCD HD cameras will be possible. It remains to be seen whether engineers will take the 3CCD route or not (or how long before they do). Right now the problem is computing, not optics and CCD.

i didnt say that the camera(stills) and video camera are un related of course they use the same technology and are very related, BUT they are designed for different PURPOSES, thus excel in the area they where designed
Take this example :

a loud speaker and a microphone BOTH use the same technologies are very related yet you try using a loud speaker as a mirco phone and vise versa get the point

about the Nikon D2h it uses JFETS so whats your point?

"- Curious that the DVX at its pixel rates can't do HD. So much for the superiority of 3CCD."

do you some have an intimate knowledge of the DVX's DSP module, if JVC could use a DSP to pump out HD from a 1/3" 1CCD your saying that Pana couldnt not be able to use an DSP that is incapable to produce HD from 3CCD?


"- JVC video is garbage."
if you say so Craig thats not what i said i said Garbage in Grabage out *univeral law*

"- 3CCD cameras may eliminate the Bayer pattern, but not the color filtering. The prism has problems of its own."

so your saying the 1CCD has solved the filtering system?

"- The Nikon D100 uses an interline transfer CCD, and architecture DESIGNED for video cameras."

so why cant nikon just use an approprite DSP and record 30fps of 4K by 2k frames ?

"- There are very few aspects of imager design that restrict their use to still or video."

Oh apart from a DSP, if this is so , please please tell me why we are not recoding video with frame sizes that digital still cam produces?

"- The Nikon D2h processes in excess of 32Mpixels/second. The DVX does at most 10Mp/s per CCD and the JVC does 30Mp/s. Still cameras have no problem keeping up with video speeds."

and what does the end result? a few still frames not video, hmm...

the DVX's DSP is its bottle neck its format is now way gonna compete with HD granted, but MPEG2 to record HD content c'mon it aint good enough, i've gotta go i'll finish this later

There are 5 megapixel digital still cameras which are better than the current CineAlta, which I think is 2 megapixels, right?

heath

ps-Let's stay cool, everyone!

Anhar,

A fundamental assertion of yours was:
a 3CCD system is better than a 1CCD sytem period.
You dismissed analogies with digital still cameras because, as you say:
they are designed for a different purpose
It's clear that the digital still arguments don't support your claims. Ultimately, the only difference between still and video that you (or anyone else) can come up with is the issue of high-res/low framerate versus low-res/high framerate. None if this matters when it comes to generating the best possible image in a frame. That's what it's all about for video or still. There are technical considerations (such as interline transfer CCD's, pixel rates, DSP power, etc.) which I've attempted to address. Other differences between video and still---lens zoom ratios, ergonomics, storage devices---are driven by user needs, not fundamental technical differences. It is the optimization of the entire product that differentiates video and still, not some imaginary internal technical incompatibility. It is possible for a single device to do exceptional video and still but it's unlikely that the resultant tradeoffs would make the product attractive to buyers.

3CCD and 1CCD are two different approaches to achieving color out of monochrome sensors. Each has it's strengths and weaknesses. In the JVC case, it wasn't realistic to process 100MP/s using available size, weight, and power so 1CCD was the only choice. That's not proof that 3CCD is better but it's certainly proof that 1CCD is better in this specific case. Time will tell how popular 3CCD wil be in consumer HD cameras but it is out of reach today.

I suggest again that you read JVC's discussion of how its 1CCD design works. It is my belief that 3CCD's big quality advantage comes from it's reduced dependency on antialiasing rather than the improved raw color bandwidth that's commonly believed. The greatly reduced antialiasing means that the image can be inherently sharper. I also believe that 3CCD camera manufacturers intentionally crank up saturation since it plays to the buyer's belief that 3CCD designs provide "more color". I would be happy to hear other's opinions on this. After reading the JVC 1CCD paper, everyone should understand that the JVC design produces more color bandwidth than the format can use. In practical experience, it is not color but image sharpness that has been the big issue. That does not surprise me at all.

I don't see how you think 3CCDs need more DSP power to process the info. Surely, to extra colour information from 3CCD is going to be a lot easier than from a 1 chipper where you actaully have to do a fair bit of processing.

Also, surely the greatest use of DSP power in the JVC is to produce an MPEG based data stream rather than a DV one. Inter-frame encoding taking more processing power than Intra-Frame encoding.

Graeme

You have 3 times the raw data rate with 3CCD assuming the same resolution sensors. You don't have to do the demosiac but you do have to have data paths and hardware to process three sets of data streams rather than one. I have no idea how data from 3 independent channels gets processed into a single, compressed stream, but if you discard the extra data up front there wouldn't be a color advantage. If you carry the extra data through the process you clearly have more signal processing.

In any event, I agree that the mpeg encoding is what took a lot of DSP power. It's not just DSP, though. There is size, heat and power consumption to worry about. Each CCD and its associated ADC's and data paths consumes power. I think it was power, heat, and DSP that drove the issue as much as cost.

I have a PDX10, which has 3 CCDs, each of which is >1mega pixel. It uses the extra pixels to provide a better quality image by subsampling, and proper 16x9. Also, through posts on this forum, we've found that it also has a superb digital zoom as it uses all those extra pixels for greater resolution. From this example, I don't see that the CCD's or their DSP is the issue here. It's most likely to be the MPEG 2 processing that is eating the battery and processing. MPEG2 has come a long way, but 18mbps doesn't cut it for high def.

I guess we'll only find out what the HDV format is capable of when it gets used in a professional camera, rather than the JVC.

Graeme

Don - "I would much rather shoot with a 'full-on' DVCPRO50 or 'DVCPRO100' camera with a sweet lens."

So would most anyone. But I think going from $3000 to $12,000 is a price differance keeps that keeps this option unrealistic.
Also, in regards to the wavey vertical lines you mentioned, can you refer me to an example or explain more. You are the first who has mentioned this and it has me puzzled and concerned.
Ken

I don't know how the PDX10 processes the pixels coming from its CCD's, but I doubt it carries a full 1MP from each each sensor all the way through the process. It's possible the subsampling is done "on-chip". i.e. through prgrammable circuitry that connects the CCD to the ADC. I doubt any more data is processed than necessary. These devices are light on power.

Future cameras should always be capable of better than current ones.