nano3D is now shipping

[Dan Keaton]

Dear Daniel,

Thank you for your input.

Your experience is greatly appreciated. I will discusss this with our engineers.



We have programmed "Line by Line", but it is not tested yet.

[Daniel Symmes]

It might be that O/U is more complex (temporal issues) and therefore not possible for the (nearly impossible) NANO 3D.

If this is for monitoring only, no big deal. But I'd not want to record anything that way for professional use.

[Dan Keaton]

Dear Daniel,

We record two isolated, full resolution channels.

We do have an option to record "Combined", but we expect this to see limited usage.

[Bruce Schultz]

Originally Posted by Daniel Symmes

Stereoscopic information is strictly in the horizontal plane (unless the 3D rig is messed up). So preserving the horizontal resolution is a significant consideration. Thus, an over/under (O/U) approach is better than side-by-side (SxS). I would rather lose resolution vertically than horizontally, for 3D.

Great, just the info I was looking for. My Samsung and all of the other 3D HDTV sets that I've used all have both SxS and Over/Under options but from now on I'll default to the Over/Under whenever possible based on this explanation. I guess that would be a great addition to the Nano3D 1.1 firmware upgrade - if possible.

[Dan Keaton]

Dear Bruce and Daniel,

Of course, we can see what we can do (implementing Over and Under) in the nano3D.

But, please remember, that combining is typically only for viewing on set, unless the nano3D is used as a playback device off-set.

As such, this reduces the importance of "Side by Side" versus "Over and Under", unless the extra horizontal resolution will help significantly while on-set.

In post, using the full resolution, isolated channels ("Left Eye" and "Right Eye") one can generate any option one wants.

We have to weigh these decisions carefully, as we want to leave room for more features, for the future.

Of course, if we see that the nano3D is used for off-set viewing, we would want the best options to be available.

[Bruce Schultz]

Dan, maybe a little thinking outside the box is required here.

As I think I mentioned before, I'm doing a shoot the coming week where I'll be sending a muxed 3D signal (1080/60i over/under) with embedded audio via satellite from Los Angeles to a Sony projector in a theater in Madrid, Spain. To accomplish that muxing I've looked into DoReMi and 3ality 3play encoder/decoder/muxers as well as the two brand new BlackMagic HDLink 3D boxes I now have.

It looks at this point as if I'll be using the BM HDLink to send out via satellite because the DoReMi doesn't flip/flop the mirrored camera, and the 3ality box only has HDMI output. Because the satellite feed requires an HDSDI baseband video signal the only option at this point is the BM HDLink 3D. I mention all of this because I'm using a $450 device to encode the 3D signal for transmission and not the $5000 devices listed above because it's the one that does the job.

So I could easily see using a muxed encoded 3D signal from the Nano3D for something other than just monitor viewing on set. Granted, recording and editing two discrete full bandwidth data streams is the aim, but you never know when serendipity and urgent need will weigh in to force the unexpected on you.

Oh, and another muxed feed will be sent to an HDCAM recorder for archival backup and playback viewing on a couple of on-set 3D HDTV's after transmission is over.

[Dan Keaton]

Dear Bruce,

Thanks, those are good points for us to add Over and Under.

[Daniel Symmes]

Bruce -

I'm fully aware of the box. Referring strictly to RECORDING of combined signals.

For on-set/editing monitoring, no problem.

Implementing O/U might be a tough one from a temporal view so I'm not expecting it.

[Bruce Schultz]

Originally Posted by Daniel Symmes

Bruce -

I'm fully aware of the box. Referring strictly to RECORDING of combined signals.

Daniel, just curious as to what your objection or reluctance to recording a muxed signal on set is.

[Daniel Symmes]

Bruce -

As I said, for monitoring no problem. But not for production (especially going out on big screen). And we're referring only to "shared space" combination where resolution is reduced. Anything else is simply a matter of preference.

Not a big deal.

[Dan Keaton]

Dear Bruce,

If one records a combined 3D image, and one image is offset from the other, then the pixels on one side of one image will be black, as the images that are shifted in one direction have an equal number of black pixels added in at the other side.

In post, one would zoom in slightly to great a full image for both the left side and right side.

Of course, for most situations we recommend recording two isolated channels, thus all of the adjustments are done in post.

I hope this helps. And I hope I understood your question correctly.

[Mark Job]

Hi Dan:
A simular question on the recording of 3D digital video and its manipulation for post production came up this Summer at the new Avid Media Composer 5.x product seminar given by Avid personnel out of Tewkesbury, Massachusetts here in Montreal, Canada. Avid concluded it is preferable not to record a *combined 3D image, nor manipulate a combined 3D image in post. It was thought a pre-combined 3D image might suffer from both quality and resolution loss issues, and therefore, 3D images should be acquired and manipulated in post as separate elements in order to have the maximum flexibility for adjustments and to guarantee there would be no possibility of resolution or image quality loss created by post manipulation.

[Dan Keaton]

Dear Mark,

Thank you for the information.

This makes sense to me.

[Daniel Symmes]

Mark -

It's only logical. People are trying to cut corners now, and they buy into the "ease" of combined.

[Luben Izov]

Dear Dan,
I was wondering if a new firmware is available, no matter for 3D or single NF, how is the firmware upgrade performed? Do you need to have both as a 3D unit or you do it separately? Please advise the different options and possibilities. Thank you
Cheers