First look at Sony XDCAM HD

[Wayne Morellini]

Originally Posted by Simon Wyndham

It wouldn't work quite as easily as that. There will always need to be innovation. But there are also so many other aspects to consider when making a camera.

It is how they stay in business and keep making a living. If they gave everything all at once they wouldn't be able to keep innovating.

That is what I was talking about, the situation you quoted where they give everything, which means every innovation, nothing left over to innovate.

Such a camera would be relatively weightless, warp space to capture and focus the image (whoops maybe I shouldn't mention that one, let the cat out of the bag) record anything, anywhere, forwards or backwards in time, and be able to run the universe, or something like that ;). And I forget, editing by re-arranging the universe (very realistic 3D effects) etc. Of course, I know you didn't mean anything like that, but you get the idea, once you get a camera too good, repair and replacement.

But seriously, the way old Eng cameras are design is rapidly becoming history. Now days there are many are programmable circuits with immense power over these Eng cameras, designed by tens or hundreds of people, that are produced by the millions for many different applications, to reduce the price. So most of the design has been done and is spread across the volume pricing. When one person is putting together and customising the components into a camera, he can work for free.

Red, on the other hand, are, at least, customisations of components, if not some complete new components. So, their costs, startup costs, and volume manufacturing and marketing cost, should be significant. But in an established volume manufacturer, there should be systems to get the parts cheaper (or manufacture them cheaper), volume manufacturing capability, established marketing channels (a real killer to try to establish) servicing, etc. So costs of established volume manufacturers should be much lower than for Red (but that doesn't mean they aren't goign to be wasteful anyway).

You can see why so much more is now becoming possible with so much less over the old Eng cameras, but while people are used to paying the old prices the companies might continue to charge them. There is significant investment against starting out in the market.

This process is so advanced, that an Eng camera circuit could be fit inside the size of the Sanyo, leaving room for drives and batteries. toughen cases are cheap enough. Lens systems, my own design idea will potentially greatly increase light gathering and latitude, with less weight.

We already had one engineer design a good 720p digital cinema camera, used to film the German feature film "Drachen Feder" in prototype. They hope to have more camera of commercial release.

Making a steadicam out of existing parts (and maybe bending some piping) is cheap enough and possible.

Seriously, listening to Ken is a good idea, engineers, that are familiar with volume engineering/manufacturing/marketing processes, tend to have some insight into the process.

[Wayne Morellini]

Ken, maybe you should look at doing camera systems. We have had a number of projects around here, but it is usually the engineers that have the ability and know how.

I personally want to see some low cost solutions to high quality recording, we have enough mid priced ones.

I can confidentially name a few.


Thanks

Wayne.

[Mark Donnell]

Reading through this thread has been interesting. On a separate thread I mentioned a "Maximum PC" magazine article from this month which compared the Blu-ray and the HD-DVD formats. The maximum data transfer rate for both was reported at 36 Mbs. If this is true, the best match to a camcorder system would seem to be the XDCAM at 35 Mbs. The P2 system undoubtedly gives a great picture, but at present it will have to be compressed 3:1 to make it onto the proposed HD DVD system. Maybe Sony should consider a light portable prosumer fixed lens (16x would be great) XDCAM-type camcorder with a built-in hard-drive recorder.

[Simon Wyndham]

XDCAM is faster than 35mbps. On the single laser pickup devices such as the XDCAM camcorders it currently goes up to 80mbps after the last firmware upgrade. Dual pickup devices such as the decsk go twice as fast. Thats one of the main differences between XDCAM and consumer Blu-Ray.

One difference between Sony and Panasonic is that Sony currently have no intention to replace all their tape based cameras with XDCAM. So I wouldn't hold out hope for a small version. It isn't impossible though as they did once consider bringing out an 8cm XDCAM disc.

[Simon Wyndham]

But seriously, the way old Eng cameras are design is rapidly becoming history. Now days there are many are programmable circuits with immense power over these Eng cameras

Wayne, as I stated the likes of Sony and Panasonic don't try and give camerapeople bad backs on purpose. If they could make them lighter and smaller they would. And indeed they are. The XDCAM is much smaller than some previous ENG cameras. And the new F900 is reduced to be more like the 750 in size. What you say in your post might make sense at first. But the problem is that you forget that Sony and Panasonic are businesses. A business exists to make money, as much of it as possible. They are not evil becaue of this. It is just the nature of the beast. If the methods you talk about made more business and financial sense then Sony and Panasonic, JVC etc would all be using those practises. If there is a way of making more money then they will do it.

volume manufacturer, there should be systems to get the parts cheaper (or manufacture them cheaper), volume manufacturing capability,

In the camera that I make my living from I don't want cheap components. I want reliable ones.

[Wayne Morellini]

I'm sorry, Simon, but my knowledge is not insufficient to judge these things. Even though you verify what I say by saying the replacement cameras are smaller, it is likely they are still a generation or two behind the sort of technology used in the Ambarella chip, and hard-drive technology have practically/economically moved into the uncompressed RAW recording realm in the last few years. The reliability is very high, if you want it to be. But we are talking about what is best for us, not what is best for the companies, so despite what they might wish, or what they might in reality choose to do, the discussion is about what is possible.


Coming attractions:
About $600 of drives will give approx 5 1/2 hours uncompressed RAW 1080/25p footage. Even without backing, up offline, to a cheaper tape/disk process, but buying drive after drive, it is economical for a feature, At 6:1 advanced visually lossless, 33 hours, that would be economical for budget, and even Eng. But for Eng, even 12:1 might be practical on an advanced codec, or just use a more standard codec. Within a year or two you might find the capacity has doubled (1 Terabyte drive planned) and half the cost. If the In-phase (I think that was the one) Holographic optical disks do go 600GB this year (1.6TB eventually) then the possibilities might become cheaper sooner (but I imagine bulkier at first).


But back to the past:
You should also find that the tape mechanism and batteries are themselves a big limiting factor, once they go to hard disk and latest battery technology, and give up the old manufacturing lines for these items, you will find what I said. The problem is companies have hundreds of millions invested in these lines, and selling media, and it is more desirable for them to keep them profitable than to recommission them to another production process. The delusion is, sold by marketing rather than practical realities, is that the desires of us videographers are anything more than secondary to profit.

Even 8Mp has been done a number of times, Olympus/Olympic, had an 8Mp Eng camera a few years ago too. I have already pointed out one approx $7.5K Altasens 1080p (I understand there is a special binning interpolation process for 720p as well) based RAW uncompressed (Most accurate) product. I have not seen the product, but understand it is a dual box product, but the reality is the PC part can be shrunk into a very small PC box, say smaller, in total, than a XDCAM HD, and possibly similar to a Z1 without lens (But I doubt SI is that far ahead in thinking) . The reality is, that if only SI would take the finale step to make a smaller ENG model, with an option for a open source FPGA codec, the XDCAM HD would be history, as far as status (realising Sony would still have the momentum and convenient ease, to sell truck loads to Eng only shooters). But with technology that is used in Ambarella, and CELL microprocessors, little more is needed than a cigarette packet sized board and one or two drives. In reality, with the existing production lines to support, the companies will give you what they can make a lot of profit out, instead of a lot for a little profit (realising that bloat, high wages, in companies/marketing/sales teams can eat up a lot of profit as well).

So, there is room for small, but well funded, unburdened companies to maneuver in the shadow of colossal giants of companies, which Ambarella seems to have done in image processing.

Now as far as businesses and evil goes, it is immoral to profiteer off the poor, and in effect, that is the case for the majority of video camera owners, they are not well paid camera people. There are many other legal issues, that I do not want to go into here. Needless to say, have a look at the DVD set, the corporation, and look through the cut interviews on the extras disks (several hours all up) to find out a little of what is really happening out there. It is moral to care for the poor and take a (reasonably sized) portion for doing it and for them to care for you, I believe it is not a money takes all proposition.

I also do not wish to be drawn to more argument on this, as it is useless trying to argue me into the ground about these things, I've seen too much, and it is really getting off topic.

I however, would really welcome a hard disk based 100-200Mb/s (Mpeg1) or V35Mb/s XDCAM HD, small Z1 like camera for no more than (and preferably a lot cheaper than the Z1, and at least 10 stops). In reality, I think that the XD CAM has sufficient bandwidth for Blu-ray disk, and stacks compared to existing HDTV, but half of what you would really desire for a feature. So, Sony still has room to maneuver there, and drop a XDCAM HD Hard disk down into the realm where the pro-sumer (where they can still make Blu-ray drive sales) and us, can author on Blu-ray, and have the more expensive pro cameras at double the quality. In this way, our material also will have sufficient quality for better HDTV codecs. With the pro cameras having sufficient quality for next generation Blu-Ray and HDTV codecs. That would be one way to ward off potential new competitors from taking business.

Let's just wait to see what direction the market twists and turns, and if the main players, can change their strategy to ward off new competitors from taking business.

[Ken Johnes]

I'm a little late, but I've been busy.

Originally Posted by Simon Wyndham

Have you ever looked inside the average 2/3" camera? It isn’t a huge empty space in there. The electronics is literally shoe horned into them.
.

That doesn’t prove anything. There are some parameters missing in your argument except weight and space. I bet that if their life would depend on this, they would make it half the size and weight in 2 months! And profitable too!

Originally Posted by Simon Wyndham

On top of this there are the lenses. The lenses, if designed to perform optimally would be much BIGGER than they are at the moment, not smaller. All lenses are a compromise in some way or another. Its the laws of physics and the way things work.
.

And if we seek for quality, are we doomed to deal forever with large and heavy lenses? Is it really the nature laws that restrict us or is it our imagination?

Originally Posted by Simon Wyndham

There are hundreds of people involved from the design, through to R&D, construction, and a whole lot of other stuff.
.

Unfortunately innovation does not increase proportionally to the number of people working in a company -far from this. A company will be very lucky to have a couple of true innovative people in good positions. And I am not talking about the "inventions" that get massively registered from big companies like IBM with 3500+ "inventions" per year. This is a paranoia of our time that must be resolved soon before we'll be forced to resolve, from the crisis that will arise.

Originally Posted by Simon Wyndham

Lets take Steadicam as another example. These products are low yield. Have you ever tried machining even a simple component for use in such a device?
.

Yes, since you asked I have designed and made quite a few original constructions in the past, some totally by me or by ordering custom components, even for professional
use, including a thermal binding machine for photo-albums production and a gold-printing modification for the same company that worked flawlessly. I love mechanical
design, it's a challenge and process which I could describe as ...beautiful!


In general I agree with you about the hidden costs.
My point is that what currently *exists* on the market today does not necessarily reflect what is *possible* today, for many reasons. One of them is the very nature of business today: Everything is profit driven (I am not against profit of course, but I think we all have lost control). So a company will not try to make what is possible but *just* what is profitable. I emphasize the word *just*. And most companies will invest more money in research only under the fear of competition and only if they have to. Not to please You and Me with a great product or to achieve the ultimate in technology or to ...save the world! Instead they are destroying our world *right now* for profit, but that's another story...

...which brings us to the subject of policy or...stupidity!
Do you think big companies are clever in general? If you do, I won't try to change your mind, just read the very interesting and invaluable book "in search of stupidity -over 20 years of high-tech marketing disasters" and you'll be surprised!


To finish my point, there is also a universal conservative tendency to keep things about the same and repeat the creative work of others for decades. Usually they do not dare to innovate in a revolutionary new way! If you want to innovate, you have to forget "how things work", clear your mind and start from zero. They don't do it ...they are afraid!

[Ken Johnes]

Originally Posted by Wayne Morellini

Ken, maybe you should look at doing camera systems. We have had a number of projects around here, but it is usually the engineers that have the ability and know how.

I personally want to see some low cost solutions to high quality recording, we have enough mid priced ones.

I can confidentially name a few.


Thanks

Wayne.

Wayne, I would LOVED to, really, BUT I'm trying to enter the software business now and I don't have time at all! Maybe in the future though.. I have some other ideas too that have to see the light...

[Simon Wyndham]

And if we seek for quality, are we doomed to deal forever with large and heavy lenses? Is it really the nature laws that restrict us or is it our imagination?

Because the only way to let more light in through a lens is to have a bigger front element. Of course the lens can be engineered better to have less light loss by the time it gets thrown onto the CCD's. But there are limits.

Going back to size for a moment, do you think that the Dalsa camera is the large size it is because the makers want to hold back? I doubt it very much. The size has been a problem for production, and the makers will know this. They didn't make it big on purpose.

[Ken Johnes]

Originally Posted by Simon Wyndham

Because the only way to let more light in through a lens is to have a bigger front element. Of course the lens can be engineered better to have less light loss by the time it gets thrown onto the CCD's. But there are limits..

Sure, except... there is another way! Think about it!

Originally Posted by Simon Wyndham

Going back to size for a moment, do you think that the Dalsa camera is the large size it is because the makers want to hold back? I doubt it very much. The size has been a problem for production, and the makers will know this. They didn't make it big on purpose.

Such a world-class company, they have provided ccd sensors for Mamiya and even for Mars robots but they also rent(!) other manufacturer's cameras???

Their "Origin" camera looks interesting (in specs) but about the size, if you look at the image it seems they didn't care to take advantage of the space, but rather they prefered to make it ...stylish!

http://www.dalsa.com/shared/content/...panel_500w.jpg

It's an impressive camera though, and they target the filmmakers, so it seems they do their best! ;>)

[Wayne Morellini]

Interesting, in another thread, where there where some, demanding, people, I talked about my idea for a technique to get 3D part surround footage from 2D footage. I also mentioned that a 8 foot+ mirror lens could be used to get the surround. Basically, even a 60 foot mirror lens could be used at a fraction the lens length and weight. The lens is my own unique design idea to overcome all the mirror lens faults (and add everything a conventional lens will do, including aperture). Since I started researching for the Digital Cinema Camera projects I came up with a number of unique new design solutions for most things involved in a camera. For moving, or other awkward shots, in reality, you would use a number of smaller cameras. Theoretically, with such great light gathering power, and the high latitude image sensor design, you could get things like 32 bit video with 200db latitude (over 30 stops) that will work on less than starlight. People that know me here, know that I don't quote fanciful science stuff here, but stick to what is practically possible base/worked off of present technology, materials and science.

Wow that Origin camera is small, for 4K 16-bit, it probably is also from a previous era, at least as far as hard drives are concerned. Do they store the hard drives in that back end, or a separate unit?

On what Ken said. The reason that they choose a bigger chip is to get the 35mm frame for lens. But the reality is that the larger sensors have different advantages too, like larger sensor pads. The larger pixel sensor pads allow for a larger well capacity (the number of electron volts it can store converted from the photons that hit it) which gives larger latitude (12 stops, but it seems that you will get 15 stops if you include the extremes) but also the less the noise floor is compared to the signal, so you can shoot in lower light, which is what Ken was meaning I think. I'm looking at a sensor that has over 300K ev- at the moment, big sensor pads, a good chip will have over 60K ev-. Probably over 13 stops of useful latitude once the noise is included in (I doubt I can source a camera cheap enough though).

Forgot to mention, they would build the Origin big, even if they had the chance to make it small, just because it looks more impressive to that market. I also have to consider how to make my camera designs bigger, if that is what the market is looking for. Not much use making a mini camera strapped to the back of your wrist if they want to shoulder mount it for stability, so putting batteries etc, in the extra body/rail struts is desirable.

[Simon Wyndham]

Forgot to mention, they would build the Origin big, even if they had the chance to make it small, just because it looks more impressive to that market.

I doubt it. It has come in for a lot of criticism because of its bulky size. It makes shooting very awkward at times. No company in their right mind would build a camera that was bulky and awkward with a view that it 'looks impressive to the intended market'. It doesn't work that way, and feature filmmakers don't decide on equipment usage in such materialistic ways.

[Ken Johnes]

Originally Posted by Wayne Morellini

Interesting, in another thread, where there where some, demanding, people, I talked about my idea for a technique to get 3D part surround footage from 2D footage. I also mentioned that a 8 foot+ mirror lens could be used to get the surround. Basically, even a 60 foot mirror lens could be used at a fraction the lens length and weight. The lens is my own unique design idea to overcome all the mirror lens faults (and add everything a conventional lens will do, including aperture). Since I started researching for the Digital Cinema Camera projects I came up with a number of unique new design solutions for most things involved in a camera. For moving, or other awkward shots, in reality, you would use a number of smaller cameras. Theoretically, with such great light gathering power, and the high latitude image sensor design, you could get things like 32 bit video with 200db latitude (over 30 stops) that will work on less than starlight. People that know me here, know that I don't quote fanciful science stuff here, but stick to what is practically possible base/worked off of present technology, materials and science.

Wow that Origin camera is small, for 4K 16-bit, it probably is also from a previous era, at least as far as hard drives are concerned. Do they store the hard drives in that back end, or a separate unit?

On what Ken said. The reason that they choose a bigger chip is to get the 35mm frame for lens. But the reality is that the larger sensors have different advantages too, like larger sensor pads. The larger pixel sensor pads allow for a larger well capacity (the number of electron volts it can store converted from the photons that hit it) which gives larger latitude (12 stops, but it seems that you will get 15 stops if you include the extremes) but also the less the noise floor is compared to the signal, so you can shoot in lower light, which is what Ken was meaning I think. I'm looking at a sensor that has over 300K ev- at the moment, big sensor pads, a good chip will have over 60K ev-. Probably over 13 stops of useful latitude once the noise is included in (I doubt I can source a camera cheap enough though).

Forgot to mention, they would build the Origin big, even if they had the chance to make it small, just because it looks more impressive to that market. I also have to consider how to make my camera designs bigger, if that is what the market is looking for. Not much use making a mini camera strapped to the back of your wrist if they want to shoulder mount it for stability, so putting batteries etc, in the extra body/rail struts is desirable.

Wayne, I cannot understand what you mean about the 3d technique you mentioned. You speak about a 8 foot or 60 foot(!) mirror lense? You mean foot as a measure of length (or diameter) or something else? Can you explain?

Anyway, the magic keyword is "mirror" as the answer to Simon's argument about heavy lenses! That's what I meant!

[Simon Wyndham]

Tell you what. You guys go and make your 3" long camera using mirrors and 3D holographic image capture with a RRSP of $1.50 and then come back when you've done it.

In the mean time I'll just enjoy the cameras that I can actually go out and buy right now.

:)

[Ken Johnes]

Originally Posted by Simon Wyndham

Tell you what. You guys go and make your 3" long camera using mirrors and 3D holographic image capture with a RRSP of $1.50 and then come back when you've done it.

In the mean time I'll just enjoy the cameras that I can actually go out and buy right now.

:)

I was editing my post to say more, but after I saw your reply I changed my mind. I must admit this is the healthier way Simon, besides, research requires a bit of masochism! :>)