HDX 1080 not a native resolution?

[philnerd]

Philip Williams
I said they gave the cammera to a typical customer. He shot with it for a weekend and posted on the web all his test footage... with Canon's approval I assume.

GG

Well, firstly I think you're referring to Kaku's footage (did I spell his name right??). He's an incredibly helpful guy and also got a lot of Sony FX1 footage out very early (I think it was released in his native Japan before the US). By the way, he's not a typical consumer, I believe he actually works for a video magazine.

As I stated in my original post, sometimes footage is released from someone that gets their hands on a model early. Kaku's footage is a good example and is actually what I had in mind when I made that statement.

Anyway, I'm gonna bow out of this discussion now, as its obviously dead end. Some readers just aren't going to be happy unless Panasonic hands them footage and CCD specs a month+ before the camera is released. Its certainly within one's rights to think Panasonic is "secretive" or strange for not releasing footage yet. No problem.

Good luck.

Philip Williams

 

[Emanuel]

Kaku's footage is a good example...

...for us - the filmmakers not the videoish fans - in order to say: No thanks!

But I think there will be surely other talented digital cinematographers that can sell the cam...

 

[mezelf27]

I think it would have been so much easier if the guys who make CCDs hadn't called the elements on 'em "pixels", because now people expect that there's a 1:1 relationship between CCD pixels, frame pixels, and final displayed format pixels. And there simply isn't.

Up-rezzing is a totally bogus term and doesn't apply to what's happening with a CCD. A CCD is an analog device. It is not digital, and there is no digital access to the CCD pixels. I'm trying to think of a way to describe this, and about all I can think of is like in audio -- the CCD is basically like a video version of a microphone. The microphone senses the air pressure of a sound wave and turns it into voltage. Well, a CCD senses the light values in an image and turns it into voltage. Not pixels, because a CCD doesn't output pixels. It outputs an analog signal, a voltage signal, sort of like how a microphone doesn't output an audio sample, instead it outputs voltage.

People seem to understand the concept that a music system "samples" the audio, at maybe 48,000 times per second or maybe 32,000 times per second, at maybe 24 bits or maybe 16 bits or maybe 12 bits. But for some reason they don't make the same connection with CCDs. It works in basically the same way: an A-to-D circuit samples the CCD output. It samples it at maybe 1440x1080 at 8 bits, or maybe at 1280x720 at 10 bits, or maybe 720x480 at 10 bits, whatever. But the CCD doesn't output pixels, it outputs voltage. That voltage gets sampled into pixels. They are not the same. There is not a 1:1 relationship between them.

Nobody ever asks if a microphone is 16-bit, do they? Of course not. But that microphone will get sampled into 48khz/16-bit audio, right? Yet a microphone doesn't have "bits", right? So is this a problem? Of course not. So why such a hangup on CCD pixels? I can assume that it's only due to the misconception that CCD pixels are the same (or somehow related to) the frame pixels, and they're not.

Euhm, here's something wrong, Barry... Obviously, the light and it's spacial intensity passing through the lens are analogue. The CCD can be seen as an analogue device because it's voltage-output isn't normed (yet - that happens in A/D), but problem is: it is a MATRIX of pixels, and each pixel works as you explained the whole CCD works. All the info is read out in a string, but already quantisized to the individual intensities of pixels/line.

So: since the quantisation is there uprezzing is the only true term in fact. The difference with a microphone is that it is a menbrance that ossilates, and that ossilation can modulate the current passing true. A microphone is therefore a true analogue device in every aspect, a CCD isn't (just the pixels of CCDs). It is more like recording stereo with 1 microphone for instance or dolby 5.1 with a stereomicrophone.

I can defenately agree to people being unhappy about this 'secrecy', no one here can honestly say they would like it if thge car salesman would say "you want to know what kind of engine is in there? Ah, but I'm not telling you! You can drive all you want and that should tell you wether it is a good car or not.". No one likes that. And on top of that, if they would do it to someone who needs a car but isn't really interested in cars.... But this is a professional product. Users have a certain amount of expertise on these things but aren't entitled to the info to judge for themselves?

 

[Barry_Green]

" Manufacturers actually almost NEVER release footage from cameras."

Canon did to a typical customer. And it looked pretty good and made a lot of people pretty happy about the cam.

I just want to say once more my thoughts and I will leave the topic as is. It's not about the actual specs, but the fact they are being evasive where I would expect them to be proud and to the point.

Not an answer you want, but something I'll just point out -- you use Canon as a good example, and then condemn Panasonic for not releasing pixel specs. You do realize that Canon went over six years without ever releasing (and to this day, never have released) the pixel dimensions on the XL1/XL1s CCD, right?

If the camera is so great, then shoot something for 10 minutes and show it to the world. Or maybe no one there knows how to shoot with a camera?

If I was egotistical, I'd take the blame for that. I've been lobbying Panasonic for months to get ahold of the first HVX and be the first guy to release footage, because frankly I'm a little tired of seeing trade-show convention floor footage being the first footage released! And yes, I posted the first trade show convention footage released for the DVX100 back in August 2002, and yes I posted the first trade show convention footage released for the HD100, and I very nearly posted the first trade show convention footage released for the XL-H1 (and would have succeeded, if the darn Sony could have read the 24F tape!)

Convention trade-show floor footage is common, but it's really no way to judge performance at all, except for convention shooters I guess. So yes, I do expect them to give the camera to those who know how to shoot. But there are a lot of people in line for that, and there are a lot of people who come before that too (like I said earlier, there are manufacturers who need to get ahold of the units as soon as possible so they can measure and test and make sure their accessories will work, and those are considerations that are going to come before shooting some test footage for web release!)

It's not the answer we want to hear, but I think it's the answer that's closest to the truth. They'll put out some killer footage when they can, but I don't think it's at the top of their priority list; others need to get ahold of the product first.

 

[Barry_Green]

Are you referring to the XLH1? If so, the initial footage wasn't from Canon. I can't recall Canon ever releasing any footage whatsoever. If I'm wrong, I'm more than welcome to correction and/or a link to official Canon footage.

Philip Williams

Canon arranged for Kaku Ito to get one of the pre-release XL H1's in Japan. So no the footage wasn't from Canon themselves, but it was facilitated by Canon to make sure it happened.

 

[taubkin]

It's pretty obvious that the numbers will be unimpressive. If it was something that could make us really excited (like DVCPro HD, P2, variable frame rates, etc) it would be out by now. So I am expecting a smaller resolution CCD that would interpolate information through pixel shift (Don't forget that even the CCDs being lower res, there are 3 of them!) and bring us a very nice HD picture that will fulfill our inicial expectations (since there are already comments about the image of the camera, and all of them are very good). Let's not forget that Reel Stream is getting a very nice 720p image out of the DVX sensors!

In my opinion Panasonic already said (by not saying) that this will not be a number that will impress us. On the other hand, the image might still be good, and there is a reason for that. Do they need a true HD resolution CCD? I mean, more pixels equals smaller pixels, and in a 1/3" CCD size project that could bring several drawbacks. So less and larger pixels equals more sensitivity, more latitude, better S/N, and if they can coupe that with a nice progressive HD image, who cares what the numbers are? Maybe it is more intelligent and more efficient to have a CCD with less pixels (actually 3). So instead of trying to talk people into believing it, they just ommit the information until they have more ways of arguing (like showing the actual footage, for instance).

Of course we will have to wait and see. And that is really all there is to it.

 

[Stevet]

oh man.... here we go again.
I find it no big deal that Panasonic has not released all THEIR specs.

When the camera is AVAILABLE on the market, we will all know. Than,
one can decide if it's right for them. Assuming you believe in specs alone,
which I don't!

 

[obin]

Barry your flat out wrong, a CCD or CMOS HAS pixels, or "photosites" as some call them, some have more 6million pixels, and some have less like 1 million pixels..pleadon't go around saying CCD chips and CMOS chips don't have photosites on them!!!

if you still think your right just make a simple google.com of how a CMOS or CCD chip works.

 

[obin]

Also guys don't forget what reel-stream.com is getting out of the SD dvx100!! if they can do that I would think that Panasonic can do the same thing with a lower res CCD block in realtime...this could be the reason for no ccd pixel count....wait till you see the camera images!

 

[mezelf27]

Not an answer you want, but something I'll just point out -- you use Canon as a good example, and then condemn Panasonic for not releasing pixel specs. You do realize that Canon went over six years without ever releasing (and to this day, never have released) the pixel dimensions on the XL1/XL1s CCD, right?

Huh? It's always been available. XL1 was 270k and XL1s was 440k and XL2 is 680k (on the top of my head). That's always been public...

 

[mezelf27]

There are indeed a lot of comments on the quality of the footage that has been shown. But it seems the footage was showed on a 1366x768 display, so actually showing 720(p - I guess). How could anyone judge the quality of the 1080p mode on that? We'll have to wait and see wether the 1080p mode will be anything like impressive. With 960x720 CCD, I guess they'd better stick to a 720p camcorder, but automatical uprezzing might have its perks, I'm guessing...

 

[Shiloh Arts]

Ok ...I'm lost :grin:

to much techyness :shocked: wow... "head spinning"

 

[mezelf27]

sorry :undecided

 

[GGman]

Not an answer you want, but something I'll just point out -- you use Canon as a good example, and then condemn Panasonic for not releasing pixel specs. You do realize that Canon went over six years without ever releasing (and to this day, never have released) the pixel dimensions on the XL1/XL1s CCD, right?

"They'll put out some killer footage when they can, but I don't think it's at the top of their priority list; others need to get ahold of the product first.

Barry,

With all due respect, I am not condeming Pana, but pointing out my opinion of how I feel they are not behaving in a professional manner over information on their "finished product", which is packed on a pallete ready for shipping. As I have said, it's not what the actual specs are, but the behaviour and attitude of holding back on information. After all, it's not NAB on Dec 5th.

Also, I am not concerned with what Canon has done in the past with other cameras. What really matters is now. I am only pointing out that I feel Canon did a better thing by letting a person shoot with their new camera and post the results before they released the camera to the public. They were confident with what they made and they even did not seek out a DP to shoot with it. And, I give credit to everyone here for having enough intelligence to make decisions/oipinions and know the difference between the results of who shot what and how they went about doing it. The Canon footage was presented to everyone with a full disclaimer on who shot, how he shot it and that he was a not a typical DP at it's best. I bet everyone would be happy to see Jan shot out a window for a few minutes like before. Maybe even the guy in shipping who packed the pallette.

It should be at the top of their priority list, IMO. And I can wait for "killer fooatge" when they get around to it. I would like to have respect for a company who has integrity. I would rather have techs talking instead of reps/salesman.

Best regards,
GG

 

[Luis Caffesse]

With all due respect, I am not condeming Pana, but pointing out my opinion of how I feel they are not behaving in a professional manner over information on their "finished product", which is packed on a pallete ready for shipping.

Just for the record, the cam is not yet "packed on a pallete ready for shipping."
According to Jan she just got a finished cam in her hands yesterday (or at least she was hoping to, not sure if she actually got it). She has also mentioned that the first batch of cameras to be distributed in December were going into production this week.

So there is no 'finished product' just yet.

 

[wabbit]

Wanna know when they gonna release the footage? Who doesn't believe it will be on November 17 (or maybe a couple of days after)?

Now that the camera is "finished" the releasing of specs is a marketing/hype decision. If the footage had been released a month ago, we would be bored talking about it already. You BUILD the hype. If you peak too soon, you lose the enthusiasm.

So if you were in charge at Panasonic sales, would you:

A. throw every possible bit of infomation at the public as it becomes available, giving you nothing left to reveal in the final months?

or

B. strategically put out crumbs of info to keep people interested and talking about the camera until it's release (and also hold on to the biggest crumb, the footage, to be premiered at the same time as your competition releases it's camera).

I could be wrong, but I bet that is what they have been planning all along.

Best

 

[GGman]

Just for the record, the cam is not yet "packed on a pallete ready for shipping."
According to Jan she just got a finished cam in her hands yesterday (or at least she was hoping to, not sure if she actually got it). She has also mentioned that the first batch of cameras to be distributed in December were going into production this week.

So there is no 'finished product' just yet.

You will have to take that issue up with the Pana rep who said that last night to a crowd of 25 people. He said he personally saw the pallete full of final HVX200 cameras being made for shipment to fall in line with their Dec 5 LA show.

If this is not true, than it adds more weight to what I have been saying about getting info from a company.

GG

 

[Barry_Green]

Barry your flat out wrong, a CCD or CMOS HAS pixels, or "photosites" as some call them, some have more 6million pixels, and some have less like 1 million pixels..pleadon't go around saying CCD chips and CMOS chips don't have photosites on them!!!

Obin, I said nothing of the sort. First of all, I never said anything about CMOS. Secondly, I said that yes CCDs have pixels, but that the camera cannot directly access them. They get read out as an analog voltage signal, not as pixels. So please don't go putting words in my mouth, especially when followed by multiple exclamation marks!

 

[Barry_Green]

Huh? It's always been available. XL1 was 270k and XL1s was 440k and XL2 is 680k (on the top of my head). That's always been public...

Yes, a total quantity, but not an array. How are the 270k divided up? What's the array? Why are they keeping it secret? Etc... they never bothered to say. Canon says 1440x1080 on the XL H1, but they never said "anything x anything" for the XL1.

 

[Barry_Green]

But it seems the footage was showed on a 1366x768 display, so actually showing 720(p - I guess). How could anyone judge the quality of the 1080p mode on that?

Well, the footage that was shown was 1080i, not p. And it was shown on two different displays, one was a 720p-native LCD monitor but then it was also broadcast on a projector. I don't know what type of projector...