Just came out with this version of my 3D beamsplitter. I spent a long time on the design to make something that's easy to use and delivers consistent results. Please respond with any questions about the rig. It's just out so I will either give you answers or find them out. I'm also looking for beta testers interested in a discount in exchange for feedback on the rig.

The Robert Rig - 3D Made Easier - YouTube

Looks good, Jesse! Please put me on the list, as someone who's interested in beta testing!

Glad you like it Matt! I'll PM you with details on Beta testing.

Ah, Looks compact & mobile, but what of veiner adjustments? However, with the RECTIFY ability in the new EDUIS 6.5, This could be completed in post to some degree.

Let me know of the beta testing possibilities also.
Best,
Carlton

I'm not familiar with Veiner adjustments. Can you elaborate? Seems like you'd be a great candidate for beta testing. I'll send you the scoop.

Great looking rig, Jesse. It's definitely cheaper than most rigs.

Can one vary convergence? What if someone was used to setting interocular distance on the standard scale? Thanks for sharing!

I shoot everything parallel and that's what the rig is built around. However, you can manually converge the cameras if you want; there's just no dial-able convergence knob...yet. The rig is built for upgrades and, if there is demand, I have plans to build a convergence thumbscrew into the rig. This upgrade will be back compatible with previous models.

Yes! You can use whatever IO scale you want. The IO scale on the rig went through a ton of revisions before I finally found a design I liked. What's great is it's actually a custom IO scale holder. So, you can use the ones that come with the rig, download and print extras scales that either I or someone else creates, or make your own. It's totally customizable. In my dreams users are creating scales and posting them here. "Here's an IO scale that works great for stuff going out to 3DS." and "Here's a short I did, I used this scale to set the depth on all the shots." etc...

Thanks Jessie,
I will be be Beta testing this with two of the new SONY RX100's.
BTW- the RX100 has an onboard electronic leveler indicator. Although the camera was not designed for 3D use, the leveler is a unexpected benefit for specific 3D work.

I will upload some samples once I get EDIOUS 6.5 in full swing.

If possible, I would also like to participate in the beta testing. I shoot with two Sony TD10...

Thanks Wolfgang! Send me an email at jesseATgoatandyeti.com and I'll fill you in on the details.

You can send me the information to videotreffpunktATaon.at.

I will be be Beta testing this with two of the new SONY RX100's.
Off-topic: I am very curious how you sync and align these.

Good questions Matt. I'm working on an alignment how-to video right now and will post as soon as it's done. For sync, I've used sound sync, an IR splitter to trigger both cameras, and I'm waiting to test a sync controller for my Canon's. Of course, if you're shooting with Genlocked cameras it's a non-issue.

Matt,
I at this point,I still align manually, and sync to the closest frame in Post. (see the dance piece in the 3D reel thread).

EDIUS 6.5 offers a RECTIFY feature, ( which aligns most of the aspect of 3D).

I am still working on a way to sync the clips before stereopticly setting the paired clips.
EDIUS may come out with a method of their own soon, (I am sure they wiil)

There is no good way to pair clips in Edius, if they do not have a real timecode - what is true for all consumer AVCHD cameras. Beside the synchronisation - what means that the reading from the CMOS starts within a certain time frame - there is no synchronisation of the start point of recording. So you can end up with the situation, that two clips have difference start points - one clip starts maybe 2 or 3 frames earlier, compared with the second clip. For my two TD10, it can be zero, but can also go up to 2 or 3 or 4 frames also.

If that happens you have to adjust that.

Not the focus of that thread, but what seems to work is that - as a first workflow:

What I did was to put the two clips in the timeline - in 1VA and 2VA. Then adjust the clips according to the waveform - what is not really new in s3d without a timecode. Then I trim the clips at the end and at the beginning in the timeline - so both clips have the same length. Then I move both clips into the bin (there seems to be no way to pair the clips in the timeline) - with the mouse by drag and drop. And pair them in the bin. The paired bin can be used in a s3d project - the original clips in the timeline can be deleted.

Not very comfortable - but the best what can be done if you do not have timecodes and genlock.

This process seems to be not foreseen in Edius, since there is no way to pair the two trimmed clips in the timeline. This workflow is similar to what I do in Vegas - but more complicated since here it is foreseen to pair the clips in the timeline. So if you work with two consumer AVCHD camcorders, the support by Edius could be better.

If there is a better way how that can be done I would be happy to learn that I am wrong. Please tell me.

Looking forward to learn how you adjust your heights and tilt.

You really need to totally block light from entering the back of the mirror box. It will seriously degrade the mirror performance and the top camera will be able to see the bottom of the mirror box.

Thank you Alister for pointing that out!

I forgot to mention that the rig is frequently featured in the video without the all of the back and top light blocking pieces in place. (I found that the black on black was difficult to see at times.) Calibration video coming soon.

Wolfgang, I do it similarly. I first name the pairs whatever_L/(or R). Place them ontimeline and align by audio. I then shorten the one starting sooner, and return into bin. Kill the one with the same name, originally in the bin. I don't bother with the end at this tme, and stereo pair the two, now starting at the same time. That's it, it is actually quite quick, because I use cricket clicker to mark the sound on both cameras. Not automatic, but just as fast as with any other stereo editor I used (4). I generally cut-off just about all the clip ends once I have them paired and on the timeline. Stereo cameras of course don't need that step. Some editors, like Stereo3D Toolbox do auto-align (with FCP-X).

I use cricket clicker to mark the sound on both cameras.

Ah, so that explains the mysterious clicks in your videos. I have been wondering what they were for quite some time, but now, mystery solved. :)

Yes, I should do a cleaner job and edit the click out, but I am just having fun. Apparently "Pluraleyes" for Edius will also align the tracks, but I didn't spend much time with it and I never figured out how to use it. It is normally used for multicam sync. (Sorry Jesse for hijacking your thread. This is the last, I promise and apologize.)

So, I have paid and ordered the rig yesterday. I am keen to see how it works..... :)

Here's a quick general assembly video which goes over mounting the camera plates and other general features. Robert Rig Assembly Video - YouTube

One of the big breakthroughs we made on the Robert Rig was in isolating the camera axis for alignment. Check it out here. Robert Rig Calibration Video - YouTube

Looks great - thank you for the video, that seems to be a great concept to adjust the rig. Maybe that has to be done with the cameras installed again (my experience is that the cameras are also not 100% equal).

Looks great - thank you for the video, that seems to be a great concept to adjust the rig. Maybe that has to be done with the cameras installed again (my experience is that the cameras are also not 100% equal).

Yes. Depending on your set-up you can run the cameras through StereoCat or a monitor for finer alignment. This will get you close enough to fine tune the alignment in post. What's nice is you can actually tune the rig to accomodate a known discrepancy in your cameras (i.e. I've shot with different brand 28mm lenses). So, as long as you mount them the same way each time you'll be set. Of course, if you're using zooms that can change over the course of the zoom. So, it's not a solution for every situation.

I am aware of that geometric issues for cameras, Jesse. That is why I ended up to adjust my two consumter TD10 cameras at full zoom, that brings you a limited error at full wide angel.

That is also something that will be necessary I think - to allocate the cameras to L and R in the rig, and to fine-tune the adjustment using the cameras.

My rig has arrived today. I am keen to start to test it - but first I will have to assemble it (what is not the standard situation since we have decided to split it up in the parts to minimise the transportation costs).

Any news /reviews of the Robert Rig?

What do you wish to know? I have done some tests here, for my two consumer-cameras TD10 Robert had to send me some specific plates. The geometric adjustment works fine so far but needs some time, even if I still did not have the time to perform the final tests due to workload and absence driven by my job.

Thanks for checking in Maheel,

Lots of great feedback from my beta testers on the rig. I've made the following improvements:
- Upgraded Matte Flaps to aluminum
- Upgraded top and rear light shields to aluminum
- Improved internal and front light blocking
- Durable vinyl 3D alignment chart
- Vinyl IO scales
- Eyebrow shim
- Glass set-pins changed to black and matching allen wrench adjustment
- Double nylon washers for shoulder bolts, easier to adjust

I'm working through some other tweaks to the rig and will keep the forum updated as to my progress.

Thanks Jesse, Wolfgand,

It is good to know that the rig received positive feed back. Like to see some HQ footage from the rig.

Maheel

Here is a piece I just shot with the Robert Rig in anaglyph. It's a teaser for a feature length movie I'm working on. If you have something in particular you'd like to see, I can do a test. Just let me know.
Chompers 3D Bathtub on Vimeo

Here is a piece I just shot with the Robert Rig in anaglyph.
LOL! Fun stuff, Jesse!