Not so sure about that. I would rather say if both cameras are already focused, then camera can start shooting immediately, so there is no change in timing between both cameras. On the other hand, if you press shooting before previous auto-focus the cameras might have different actual timings, since the lenses have not the same speed, they are not previously at the same position, etc.
GH1 Stereoscopic Rig Collaboration
- Thread starter David Cole
- Start date
Not so sure about that. I would rather say if both cameras are already focused, then camera can start shooting immediately, so there is no change in timing between both cameras. On the other hand, if you press shooting before previous auto-focus the cameras might have different actual timings, since the lenses have not the same speed, they are not previously at the same position, etc.
No, focusing does not change anything in Vsynch signal. However, I had a feeling that I should not pre-focus in the first few seconds after the camera powers-on (when it reads a data from SD card).
David Cole
Well-known member
My DC couplers from ebay seller in Japan just arrived today. In case you are wondering, the source seems to be trustworthy and they are the genuine Panasonic article.
http://cgi.ebay.co.uk/GENUINE-Panas...226?pt=LH_DefaultDomain_0&hash=item20b84d2dda
http://cgi.ebay.co.uk/GENUINE-Panas...226?pt=LH_DefaultDomain_0&hash=item20b84d2dda
I can not see any mis-sync at 1:16 and 1:27 at all. There are abrupt horizontal shifts which mean just depth changes in sync. More detailed description of your check result will be appreciated.Now it is much better. However, have you checked clip around 1:16 and 1:27? Observe the clip e.g. in anaglyph form and check edges of the flower in wind. Otherwise it seems perfectly (or at least very-well) synched.
Is this realized by the function of the 3DSLR Master system? Could you summarize the function of 3DSLR Master system in relation to the 3D sync.
I had some difficulty when I try to sync 24P clip recorded on GH13 (NTSC, not PAL), where 1/60 Sec shift may cause different frame L/R pair at the same time code in 60P sequence, which results in horizontal shift i.e., parallax oscillation and is not acceptable for S3D. Can 3DSLR Master avoid this situation in GH13(NTSC)? If yes, it will be nice! This is one of reasons I shifted to GH2 with 60fps sensor output from 24P GH13.
With my method first rec on just after power on requires more frame shift in the post processing to achieve perfect sync than the every succeeding rec on. I am sure the consequent sync after post adjustment is 100% perfect, no exception I have encountered as far as judging from the audio tracks .
Hi, David
I made a brief experiment for you how half push does affect the sync with my twin GH2 system and found an interesting result. As I have been reporting, half-push and then full-push realizes 100% sync with post processing. however full push without half push results in 1/120 Sec mis-sync at a probability of less than 50% but near 50% and the rest, more than 50% are just in perfect sync! This kind of mis-sync is worst and cannot be fixed in post at all. I guess 120fps sensor output for speeding up of AF in GH2, of which gl have pointed out, may be related to this mis-sync. Therefore Half push to complete the AF before start recording makes the situation simple and may be a must condition for synced recording when AF lenses are being used, I think.
inoue_k3D
You are right. It seems that extreme depth of the scene (by using relatively large stereo base even when using extreme zooms) is producing so large horizontal shifts even though the actual stereo distance of the flower is not changed so much. My mistake and relief as well, since it proves that reliable synch can be obtained even with Panasonic lenses.
It is still under "construction". It offers shifted (in time) power-on (and power-off) of both cameras, shifted remote signals for both cameras, reading composite video from both cameras, displaying FPS and synch in ms (resolution 0.01ms). I'll have to add time lapse and user-defined parameters with some menu system similar to "3D LANC Master".
I don't know. For PAL it's easier, since I can see if synch changes when starting shooting. However, for NTSC signal it might happen that the signals are in synch while the actual footage is not. :-(
Anyway, once I'll have to check 24FPS (NTSC) settings as well.
Exactly as I thought! Now it's clear what could cause missynchs in David's shots with Panasonic lenses.
Hi, All
I made a sync test using an external flash.
The result as shown in the following SBS photo, is approximately 2.6% of the frame duration (=0.4mS), a little more than expected.
If the acceptable limit is assumed to be 1mS, twelve minutes clip can be shot on my particular pair of GH2's.
Any comment will be much appreciated.
Full HD photo have been added to the end of 3D Rig Photos;
http://v-galleries.com/images/Twin_GH2_Rig/index.htm
inoue_k3D
I made a sync test using an external flash.
The result as shown in the following SBS photo, is approximately 2.6% of the frame duration (=0.4mS), a little more than expected.
If the acceptable limit is assumed to be 1mS, twelve minutes clip can be shot on my particular pair of GH2's.
Any comment will be much appreciated.
Full HD photo have been added to the end of 3D Rig Photos;
http://v-galleries.com/images/Twin_GH2_Rig/index.htm
inoue_k3D
Have you posted the original "flash" picture or already aligned one? Namely, missynch should be measured from the original pictures.
Vitaliy Kiselev
Carbonite Member
I am really envy reading all this
Btw I made special category on my site - http://www.personal-view.com/talks/categories/3d , so if you want to discuss deeply technical things or just have separate thread about your specific rig, you are welcome.
Btw I made special category on my site - http://www.personal-view.com/talks/categories/3d , so if you want to discuss deeply technical things or just have separate thread about your specific rig, you are welcome.
This is a very good result, especially for video. My cameras have about 0.7ms missynch after power-up. With additional (artificial) delay the missynch after power-up becomes 1.7ms and then in about 2 minutes it reaches 0ms and in additional 2 minutes about 1.7ms in another direction. I'm planning to modify the frequency of one camera soon in order to have lower drift. I also think that in most situations (not all!) 1ms missynch is very good for 1/50 or 1/60 shutter speed.
Vitaliy, this is a very good idea about special category for discussing more technical things. Thanks!
g.l
Well-known member
Actually I would prefer to switch on +ve, however I don't understand how to do it. My electronics knowledge is patchy (I'm a software guy). I can do simple things with microprocessors, but I have no idea how to get the negative voltages etc. that the +ve switching seems to require (with MOSFETS at least).
Switching on ground is simpler. I think I know how to do it now, but if you have any advice I'd appreciate it
.I think it will be fine, as I will use separate MOSFETS for each remote line (so there is no cross-conduction).
g.l
Well-known member
Interesting, how will you do it?
g.l
Well-known member
Crunchy, I just read your post on Vitaliy's pages about the GH1 not being able to use microphone & remote at the same time. What happens if you try to connect both?
Actually on the GH2 it wouldn't be a disaster, the built-in mic is surprisingly good (but unfortunately picks up wheel clicks etc). But I was hoping to use my Sony stereo mic with it.
That mic has a 3.5mm stereo jack, and works great on its own with a generic 4pin 2.5mm adapter I have. But it fails with another one I bought (wrong pin assignments, seems to be used with mobile phones):
Actually on the GH2 it wouldn't be a disaster, the built-in mic is surprisingly good (but unfortunately picks up wheel clicks etc). But I was hoping to use my Sony stereo mic with it.
That mic has a 3.5mm stereo jack, and works great on its own with a generic 4pin 2.5mm adapter I have. But it fails with another one I bought (wrong pin assignments, seems to be used with mobile phones):
Hi, Damir and All
Thank you for the information on initial sync of your GH13 based rig. Are your sync values after offset adjustment using 3DSLR Master?
My experience on LANC based 3D Handycam(HC3) with LANC Shepherd, initial sync is below 0.00mS as shown below (I found it was almost 4 years ago!)
http://v-galleries.com/index.html#LANC
Therefore I felt the value 0.4mS is somewhat large and understand due to no offset compensation in my GH2 system. Is it right?
With 3DSLR Master can I expect the initial sync will be set to zero and then drift like my result with twin HC3?
I look forward to such function of 3DSLR Master with which no frame shift in post will be required.
Anyway, as 60i footage on my twin GH2 show acceptable and reproducible sync after editing, I would like to start shooting 3D video hereafter for my after-retirement pleasure.
Thanks and good luck to this wonderful forum !
inoue_k3D
Thank you for the information on initial sync of your GH13 based rig. Are your sync values after offset adjustment using 3DSLR Master?
My experience on LANC based 3D Handycam(HC3) with LANC Shepherd, initial sync is below 0.00mS as shown below (I found it was almost 4 years ago!)
http://v-galleries.com/index.html#LANC
Therefore I felt the value 0.4mS is somewhat large and understand due to no offset compensation in my GH2 system. Is it right?
With 3DSLR Master can I expect the initial sync will be set to zero and then drift like my result with twin HC3?
I look forward to such function of 3DSLR Master with which no frame shift in post will be required.
Anyway, as 60i footage on my twin GH2 show acceptable and reproducible sync after editing, I would like to start shooting 3D video hereafter for my after-retirement pleasure.
Thanks and good luck to this wonderful forum !
inoue_k3D
Hi Katsuhiko,
Yes. Originally it is about 0.7ms and it reaches 0ms in 1 minute and 0.7ms in the opposite direction. In order to have more time for preparation of the shot (after power-on) I additionally added 1ms offset. If I wanted, I can reduce initial offset close to 0ms, but then I have to be very quick in taking shots.
By using initial time compensation with 3DSL Master, both cameras can start with 0ms synch. However, it does not mean that both cameras will have 0ms synch after starting actual shooting. In most cases they are in synch, but sometimes it happens that they differ. If I see that they differ (on a 3DSLR Master display), I stop taking video and restart the cameras again.
This one-frame-drift is really strange thing. Can you tell me how did you realise remote control for the cameras? If it is too technical we can discuss on Vitaliy's forum.
Go for it and show us some more clips!
Yes. Originally it is about 0.7ms and it reaches 0ms in 1 minute and 0.7ms in the opposite direction. In order to have more time for preparation of the shot (after power-on) I additionally added 1ms offset. If I wanted, I can reduce initial offset close to 0ms, but then I have to be very quick in taking shots.
By using initial time compensation with 3DSL Master, both cameras can start with 0ms synch. However, it does not mean that both cameras will have 0ms synch after starting actual shooting. In most cases they are in synch, but sometimes it happens that they differ. If I see that they differ (on a 3DSLR Master display), I stop taking video and restart the cameras again.
This one-frame-drift is really strange thing. Can you tell me how did you realise remote control for the cameras? If it is too technical we can discuss on Vitaliy's forum.
Go for it and show us some more clips!