GH1 Stereoscopic Rig Collaboration

I mean that it does not matter if you are recording on the cameras - or - with external HDMI recorders. The issue remains that the cameras MUST be sync'ed. It is likely that GH2's can be sync'ed at power-on using the simple method of turning external power on to both cameras simultaneously (as is described earlier in this thread). Even with power-on sync, however, the cameras will drift out of sync over time. It remains to be seen how bad the drift will be with GH2s.

I plan to test GH2s as soon as possible.
 
Sorry, been away a while

Sorry, been away a while

Alan Hunter's Syncamajig (http://samj.snappages.com/) project is perfect for an off-the-shelf sync / record controller. Alan, you still listening in on this thread?

I did buy a second GH1, but the pair I have drift so rapidly as to be useless. Like 450 milliseconds per minute. Not just too fast for shooting 3D footage; too fast to even do much useful experimentation. So I'm going to buy one more, this time with a serial number very close to one of the GH1's I have, and see if I get lucky. I plan to do enough programming this weekend and next to be ready to ship the first few Syncamajigs. To patient beta-testers. I really should make some videos showing how they work and how to use them. Alan
 
I did buy a second GH1, but the pair I have drift so rapidly as to be useless. Like 450 milliseconds per minute. Not just too fast for shooting 3D footage; too fast to even do much useful experimentation. So I'm going to buy one more, this time with a serial number very close to one of the GH1's I have, and see if I get lucky.
So don't distant serial numbers work?
 
I did buy a second GH1, but the pair I have drift so rapidly as to be useless. Like 450 milliseconds per minute. Not just too fast for shooting 3D footage; too fast to even do much useful experimentation. So I'm going to buy one more, this time with a serial number very close to one of the GH1's I have, and see if I get lucky. I plan to do enough programming this weekend and next to be ready to ship the first few Syncamajigs. To patient beta-testers. I really should make some videos showing how they work and how to use them. Alan
And about the GH2? Should the GH2 work as same as the GH1?
 
I plan to do enough programming this weekend and next to be ready to ship the first few Syncamajigs. To patient beta-testers. I really should make some videos showing how they work and how to use them. Alan
When can we buy the first Syncamajigs production units? Or can anyone be beta-tester?
 
It would be interesting to see if we could introduce an external sync with the hack. There is an external remote for the GH1. With the remote connected to both cameras, is there a difference ? For example, if we use the remote to trigger the shutter on both cameras at the same time, does it force them to sync up ?
 
serial numbers and good drift rates

serial numbers and good drift rates

So don't distant serial numbers work?

I don't know. It could be that I have an outlier. But it is logical, as David Cole has said, to expect greater similarity in component values among units that were built in the same factory at about the same time. The capacitors and crystals and so on are more likely to be from identical batches. I (or someone) would need to measure the drift rates of many GH1's (as with any other camera) against some common timebase to get a good idea of the typical variance. I've used a blackburst generator myself to compare drift rates among my own cameras, and I suppose I could mail it out along with a Syncamajig for people to measure the drift of their GH1 with respect to it. Then they could post their values online somewhere, and people with well-matched GH1's could trade or buy and sell among themselves. But I'd want a hefty deposit. I haven't thought about this much yet.
 
And about the GH2? Should the GH2 work as same as the GH1?

I haven't read the manual yet. If it has a toggle-type power switch, and if it outputs composite video or composite sync from the AV port while shooting, and if this doesn't make the viewfinder go dark, probably. Even the manual may not tell the tale, since the GH1 has this undocumented feature of putting out composite sync while you shoot, but only after holding down the "trash / delete" button for about 10 seconds. Each camera needs to be tested. Some cameras start up putting out sync, and then reset sync after a few seconds, so a delay needs to be programmed into the Syncamajig to wait until the sync is steady before measuring it.
 
I haven't read the manual yet. If it has a toggle-type power switch, and if it outputs composite video or composite sync from the AV port while shooting, and if this doesn't make the viewfinder go dark, probably. Even the manual may not tell the tale, since the GH1 has this undocumented feature of putting out composite sync while you shoot, but only after holding down the "trash / delete" button for about 10 seconds. Each camera needs to be tested. Some cameras start up putting out sync, and then reset sync after a few seconds, so a delay needs to be programmed into the Syncamajig to wait until the sync is steady before measuring it.
And me?

You didn't respond to my post:

When can we buy the first Syncamajigs production units? Or can anyone be beta-tester?

And I was the only one to wonder and ask about your product!... : )
 
external sync using hack - unlikely

external sync using hack - unlikely

It would be interesting to see if we could introduce an external sync with the hack. There is an external remote for the GH1. With the remote connected to both cameras, is there a difference ? For example, if we use the remote to trigger the shutter on both cameras at the same time, does it force them to sync up ?

Just working the shutter release does not affect sync. I thought that Vitaliy had already said that he thought this approach was highly unlikely, but I just scanned all the earlier posts looking for that post and didn't see it.
 
serial numbers and drift

serial numbers and drift

I did buy a second GH1, but the pair I have drift so rapidly as to be useless. Like 450 milliseconds per minute. Not just too fast for shooting 3D footage; too fast to even do much useful experimentation. So I'm going to buy one more, this time with a serial number very close to one of the GH1's I have, and see if I get lucky.

OK, I was lucky and bought a 3rd GH1 on eBay, with the same first 5 characters in the serial number as one that I already had, and only about 120 units apart in the digits. I first measured the drift for an hour, and it was a remarkably steady 33 microseconds per minute or so; the next day I loaded the same firmware into both cameras, and measured a steady 45 microseconds per minute or so, over the course of an hour. So it appears that 1) getting a pair with close serial numbers is probably a good bet, and 2) the firmware can have some effect on the sync frequency. I'll put my odd GH1 up for sale on eBay this weekend, and do some more testing. Maybe using MJPEG will drift differently from using AVCHD, or something.
 
I've decided on two GH2's due to the improved image quality / light sensitivity and the 1:1 crop mode (I also need to shoot good stereo stills with them). I probably won't buy for another month or two so I'm very interested if the sync trick will work with the bodies.

I'd be even happier if it worked with auto-focus lenses, or if they can at least be made to work by say taping up some pins. Auto-focus is not essential but handy, and the Panny prime I'm interested in will have lens distortion correction (applied by the body).
 
Is it really worth worrying about m4/3 lenses if they cause so much trouble ?
Actually yes, if you want an ultra-small and lightweight rig, and/or care about low-light performance.

I'm planning to use the Panny 14mm F2.5 prime pancake lense - it barely exists it's so small!




Their other pancake (20mm) is almost as small, but F1.4. With the GH2 1:1 crop mode you also get a 2nd focal length (2.6x zoom) from the primes with no reduced quality, and no need to swap lenses.

I expect to mount the GH2 bodies facing each other, lenses aimed at small mirrors angled at 45 deg - this should reduce the min lens separation significantly, and doesn't require a large beamsplitter rig.


@David, if you get your system on the bench again, these tests with AF lenses would be v. interesting:

1) Boot up without lenses attached, then attach them (theory: focus calibration at boot randomly delays video timer initialisation, after boot it may not).

2) Match the focus of both lenses, then reboot (theory: focus calibration time may be consistent on both bodies).

3) Place lenses in MF mode, then reboot (does the body remember focus modes across reboot)?
 
Last edited:
As soon as i have 2 GH2's in house, I'll set the test rig back up. I'm happy to give these AF tests a spin.

Awesome. How's your GH1 rig working out now that it's used in anger? Are you finding the sync reliable over all shots?
 
Back
Top