Personal View site logo
GH1/2 3D Sync Controller
  • Applied Logic Engineering, Inc. has built a sync controller around the basic power-sync process that all the GH 3D'ers are using. This is a preliminary run-thru of the bare board:

    I should have a first article to test very soon. I'm fairly confident in the design as I've been in touch with the good folks at Applied Logic from the very beginning. I basically begged this product into existence.

  • 34 Replies sorted by
  • @David_Cole

    great news, i was interested in something like : LANC for video cameras, but just for GH2, remote control type, as this press for focus, and record ...

    but this one is much more advanced :-)

  • It is very, very good thing.

    Thanks for info, David.

    It could be really cool to have some genlock like think in GH3 for 3D.

  • +1 GH3 Genlock (or sync)

  • +infinity for genlock !!!!!

  • Very, very interesting! Thanks for sharing! Genlock on the gh3 would be immense...

  • Hi. Sometimes i use 2 Gh2 for live action3d with a genus mirror rig with similar lenses. I have one pixar remote with dual cable to start and end the record. The big problem is the Sync between both, Recording in 50p helps a lot and as a shutter upper 60. This card show the disparity of the signal of the cameras. My question is: Any possibility to conform the both signals with sync?

  • You will achieve MUCH better results if you use this controller or make your own to power the cameras up at the same time. This is a time-tested technique to achieve very tight initial sync (typically less than .05 ms). Your cameras will drift apart after startup, and there is currently no way to fix that. I find that I have very acceptable sync with both my GH1s and GH2s for 30 min or more after power-up sync. I purposely buy cameras with close serial numbers to increase the odds of success. Your results may vary.

  • Thanks David.

  • is there an expected release date and price? Perhaps an intro price for personal=view persons!

  • Quick status update.... My test copy is in bound now. As soon as I can get it tested, Applied Logic will open up the order page. Should be just a few days. The user guide is available now... http://www.appliedlogiceng.com/index_files/Page1834.htm

  • I've now tested the controller and it works extremely well! I've asked Applied Logic for a FAQ section in the user's guide and a VERY minor change to the firmware to increase the shutter close time during time-lapse operation. Neither should hold up delivery at all.

    I'm guessing they'll start shipping next week.

  • @David_Cole

    How about deals for our members?

  • I'll certainly ask!! Standby...

  • .. strums fingers .. I am keen to try this

  • @AppliedLogic

    I hope you do not sell it without any case?

  • It's a way too expensive if sold without any case. If they persist with such a price, I'll have to publish FW and circuit boards of my solution (like with 3D LANC Master) sooner than I planned. Here I repeat what have I mentioned on 3DTV Yahoo group:

    I've made my own synchronizer (3DSLR Master) for video and time-lapses which is a little more sophysticated than the one already shown in this thread. First I've made it on ARM uC, then I developed it on AVR controller due to better "freeware" SW support.

    The 3D video of some smaller subjects (flowers) was made on GH1 (the morphing effect from 2D to 3D is made by slightly moving left camera during taking video):

    http://dsc.ijs.si/damir.vrancic/down_3d/Movies/

    The file is relatively large (848MB), since XViD quality settings were relatively high and resolution is 3840x1080. Therefore I added another smaller file for preview (with 50% resolution).

    There you can also find pictures of older 3DSLR Master prototype and beam-splitter rig. The newest 3DSLR Master is using only one switch for navigating through menus and powering on/off the cameras (upper-right one).

  • Hmm... I think it's priced ok. I've built plenty of my own controllers - but - would certainly buy this one instead (in fact, I DID buy one). The LANC Shepherd is $430, as a comparable product.

    It does need an enclosure, however. I'm pretty sure Applied Logic got the message on that one.

    I REALLY don't want to bust their balls too much... it was difficult to get anyone interested in building these things and I'm very thankful they made the investment. I really DONT have a dog in this fight... as I said, I've built my own controllers. Not everyone has the time or inclination to do that, however (I posted the schematics for a simple, functional design almost 2 years ago!). There are only a handful of folks that I can tell have ever made any real headway with stereo on the GH platform and that's really a shame. It's a GREAT camera for a lot of stereo use. A ready made controller might help get this out there.

  • I think price is ok .. with caveat .. they provide no details about how critical having 2 x GH2 with super close serial numbers is .. is it still critical .. as the development threads indicated? It would be sad to buy it, then say I can only get sync 1 in 100 times and be told that I need to get 2 consecutive serialed cameras. @David_Cole .. How critical are the serial numbers now?

    In Australia I have made enquiries with several shops/suppliers, and nobody can even guarantee me two cameras from the same batch

  • Unfortunately, no one really knows how critical "serial proximity" is. All of my GH1s and 2s were purchased as sets and are very close. It's probably not a huge issue for most users. Even though my cameras will maintain acceptable sync for upwards of 30 min, I NEVER have takes that long and I reset after every take. If you're buying in the same general lot (made in the same 6 months), your probably FINE.

    This, by the way, is the exact same scenario that LANC-sunk camera users have faced forever.

    A bigger issue is that actual M4/3 lenses (that have an electrical connection to the camera) can introduce non-deterministic delay. In my tests (somewhat extensive) this results in a 30% mis-sync. For me, it's just easier to use Nikons and an adapter.

  • I too will only use manual lenses.

    I am keen to get into 3D .. is there any chance that you could do a test of teh new controller with non consecutive GH2s .. if you have access. A sample size of maybe 50 -100 starts .. more if you could spare time.

    @AppliedLogic .. do you have any statistics for non-consecutive cameras?

    I realise that my results might be different, and understand that the cameras will vary .. any statistics would help me make a decision.

    And I ask because, I only have one GH2 and my cashflow does not really allow for purchase of two new GH2s even if I could find them

  • @AppliedLogic .. if you do a case, I would request Metal or Diecast Aluminium style with a 1/4 tapped hole in all 4 side and the rear .. supply a bung for all the holes .. else I'd rather the bare board as you are offering in preference to a plastic housing.

  • There is only some slightly higher chance to get larger synch drift with quite different camera batches. On contrary, you can get quite large drift with "consecutive" cameras. Everything depends on difference of crystal frequencies. So the best solution is to take syncroniser and video cables with you and connect the cameras. Select the cameras with lowest synch drift.

    Concerning synch when shooting: I have very good results with a pair of GH1s and 14-42mm lenses (in PAL mode). I don't have good resuts with 20mm lenses (the synchonizer shows missynch in majority of cases when starting shooting). On contrary, I just tried synch with a pair of GH2s with 20mm lenses (in 24p mode; PAL settings). First five tests are encouraging. I'll try to measure synch on a CRT screen in the following days.

  • I wish could get a comment from @AppliedLogic on results with non-almost-consequtive camera

  • I think that it depends. You can have almost the same frequency if you are lucky. The best way is to have synchronizer with you when buying the cameras or simply buy more cameras and find a pair with lowest drift.