Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Major Sync issues

S

Simon N

Well-known member
Joined
Apr 15, 2009
Messages
251
Reaction score
0
Points
16
Location
Liverpool UK
I start a 3 month shoot next week and the final day of testing has thrown up some significant problems.

The post workflow is nailed so there are no options to transcode in redcine or other options.

The workflow is epic shooting r3d's and out of the sdi into a pix 240 recorder to record dnxhd36 for dailies. Sound is being recorded seperately onto a 788t. Both pix&788 have ambient time code boards so that should not be a problem.

We have tried many configurations of timecode varying from camera generating it to 788 generating & pix also. The problem is that if there was a frame discrepancy with every slate then we could set an offset and account for that in the edit. The problem is though that the sync changes from slate to slate.

As an example our latest test is this:

We set the camera as master timecode using time of day. We then attach the pix to the camer via sdi, timecode & sync. We then shoot say 25 5 second clips of a clapperboard. The pix records simultaneously. We then bring the 25 r3d files into premiere ( to test, the production is using Avid). I then sync both the r3d's and the pix mov files. This is what I don't understand. Out of the 25 clips, 20 sync up fine but 5 of them are one frame out. The pix is 1 frame behind for these 5 slates. Of course one solution is to manually resync the rogue slates, however post are reluctant in case it creates any issues down the line.

What I need is for the timecode to sync no matter what which is what should happen regardless of which camer is used? My thoughts are is it something to do with the timing out of the camera, using the old interlaced fields analogy, is the camera outputting a timecode on the wrong"field" dependant upon what moment the trigger occurs? This is only a theory but hopefully someone has experienced this and can
Point me in the right direction.

One further point, my Internet has gone down so I am not easily able to research this via a 2g iPhone!! So I can't upload at the moment and files or logs though if someone thinks it is crucial I could go somewhere else to try and upload.

Thanks
 
Afaik the camera does not output timecode. You need to feed it timecode from the pix, using the pix as the master recorder, not the camera. You will need a sync cable from the red store. The one with the 3 bnc's on it.
This is what has worked for me. I have an epic and a pix 240.
 
Hi Shane

Thanks. I can set the camera to take an external sync (lockit box/788/Pix etc..) which then sends timecode via either the SDI or timecode out (the sync cable you refer to). The pix then picks up the timecode and I can jam the pix to the camera. I am also able to set a timecode start delay in the pix which at the moment is +1 frame. This has enabld me to bring the camera r3d's and pix into sync. The main thing that is baffling me is that when I sync 25 clips, 20 are in sync and the rest are 1 frame out. I am testing this by shooting these 25 clapperboards one after the other with only a few seconds break in between each shot. The consistency is that the out of sync shots are always 1 frame out (Pix 1 frame behind the r3d's)

My issue is mainly the inconsistency - if I have to make a global change then that is correctable, what is not is random clips being out of sync
 
As long as you are correctly using the gear, Epic has no problem syncing with Pix240 or any external audio recorder. Are you trying to visually match with the slate? if so the issue is latency.
Have you tried exporting a clip out of RedcineX and auto matching timecode. If you are correctly synced to the camera everything should be perfectly in sync.

EDIT:
Are you syncing to the SYNC IN on the camera?? Its a small lemo connector on the back near the SDI out.
You will need this cable or similar:
http://www.red.com/store/products/3bnc-to-00-lemo-sync-cable

if you are not using this you are not correctly jamsyncing the camera.
You must also go into the menu and select jam sync.

NOTE: If you dont see SYNC displayed in green on the cameras LCD you are NOT in sync. This should be the first place you look.
 
Don't jam the pix from the camera. Jam the camera FROM the pix. The pix has a much better timecode generator.
Use freerun timecode mode. Once you jam the pix ( from your audio mixer) using the timecode in, set the timecode menu in the pix to timecode bnc: out. I know the pix will pick up the timecode from the camera through the sdi but it's not consistent as you are discovering.
Another advantage of jamming from the pix is that it will keep timecode sync over a battery change, for a while at least.

Also, make sure that you have the latest software on the pix. That's not relevant to the issues that you are having but there are some really cool updates in there.
 
My knowledge and experience with synch is not extensive, but i will offer what I think I know FWIW.

You didn't describe your method for synching the slate. Did you synch slate? What method did you use?

It seems asinine to state it, but if you want camera sound and slate to be in synch, then all three of them must be able to set and maintain accurate synch, and they must all be set to the same synch source. You can not allow the Pix and the 788 to run off of "ambient timecode boards" (whatever that means) and expect them both to be in perfect synch together. You have to use one as a Master to set synch for the others and you have to do it properly.

It is common for sound dept to manage synch, so pro sound recorders have accurate synch built in and other devices are typically designed to accept synch signals. The 788 is a good master on set.
 
thanks for the replies to date.

Shane.... yes I have the latest firmware V2.0 and I also have the sync cable direct from Red.

We have tried syncing with varying degrees of success:-

788T to Lockit boxes on the cameras (yes there are 2 cameras, which is going to be a completely seperate issue). So the Lockits are jammed to the 788, sending sync to the camera, which in turn outputs to the pix. Various sync issues here, though Red did say the model of lockit (202) can cause problem.

788T Direct into camera. Was quite successful, however we did not do multiple tests which later showed sometimes we'd have sync and others we wouldn't

Pix into camera with both timecode & Sync. Red t/c, sync & gen icons all lit up green. Still had issues where the picture on the pix was varying degrees of frames out from the red pictures

camera into Pix - this has been the most stable, but still random slates are out of sync, but always by 1 frame.

I have not had any internet access so have not been able to get the manual for latest firmware, am doing so as I write.

Again the main problem is not necessarily that I cannot get sync - I can - but some slates with exactly the same settings are in sync, the others out. So for example, the 25 clips I shot all lasted about 5 seconds and were of a clapperboard clapping. The sync was done automatically by timecode in Premiere, out of these 25 clips 20 were rock solid, and 5 were 1 frame out. This is what my issue is, why the inconsistency?
 
Simon N said:
thanks for the replies to date.

Shane.... yes I have the latest firmware V2.0 and I also have the sync cable direct from Red.

We have tried syncing with varying degrees of success:-

788T to Lockit boxes on the cameras (yes there are 2 cameras, which is going to be a completely seperate issue). So the Lockits are jammed to the 788, sending sync to the camera, which in turn outputs to the pix. Various sync issues here, though Red did say the model of lockit (202) can cause problem.

788T Direct into camera. Was quite successful, however we did not do multiple tests which later showed sometimes we'd have sync and others we wouldn't

Pix into camera with both timecode & Sync. Red t/c, sync & gen icons all lit up green. Still had issues where the picture on the pix was varying degrees of frames out from the red pictures

camera into Pix - this has been the most stable, but still random slates are out of sync, but always by 1 frame.

I have not had any internet access so have not been able to get the manual for latest firmware, am doing so as I write.

Again the main problem is not necessarily that I cannot get sync - I can - but some slates with exactly the same settings are in sync, the others out. So for example, the 25 clips I shot all lasted about 5 seconds and were of a clapperboard clapping. The sync was done automatically by timecode in Premiere, out of these 25 clips 20 were rock solid, and 5 were 1 frame out. This is what my issue is, why the inconsistency?
Click to expand...

Looks like you've tried everything I can think of. The inconsistency is strange, especially if you are feeding the camera from the pix. Try to email Matt Meyer at sounddevices. He's helped me in the past. I'll do some tests myself today.
 
Scott Crawley said:
My knowledge and experience with synch is not extensive, but i will offer what I think I know FWIW.

You didn't describe your method for synching the slate. Did you synch slate? What method did you use?

It seems asinine to state it, but if you want camera sound and slate to be in synch, then all three of them must be able to set and maintain accurate synch, and they must all be set to the same synch source. You can not allow the Pix and the 788 to run off of "ambient timecode boards" (whatever that means) and expect them both to be in perfect synch together. You have to use one as a Master to set synch for the others and you have to do it properly.

It is common for sound dept to manage synch, so pro sound recorders have accurate synch built in and other devices are typically designed to accept synch signals. The 788 is a good master on set.
Click to expand...

That's what I was saying, sync the pix from the sound recordist and then use the pix to feed the camera. I just meant that the pix has a more accurate timecode generator than the epic. (made by ambient)
 
Curious - possibly check to see if sound is recording a different timebase timecode than your project settings in the Epic? Heard of a similar issue when sound was generating a 29.97 clock reference, and it was used to jam cameras shooting in a different timebase.
 
Possibly a bad cable?
I assume you have one on each camera?

Just so we are clear, is the problem a visual match with the slate or have you tried actually syncing in post?

Have you selected jam sync in the menu?
Do you have the green sync indicator in camera?
 
Greg M said:
Possibly a bad cable?
I assume you have one on each camera?

Just so we are clear, is the problem a visual match with the slate or have you tried actually syncing in post?

Have you selected jam sync in the menu?
Do you have the green sync indicator in camera?
Click to expand...

Hi. The issue is a post problem. We are synchronising timecodes in the avid and without changing any setting some takes are in sync and others are one frame out. We have all green lights on the epic. Tc sync & gen
 
please be very specific in describing what you are doing, its hard to help without knowing.
Is it all one camera, or both cameras. If you are testing two cameras, could one be off and one on (are the inconsistencies specific to one camera)? How are you syncing in avid?
are both cameras recording on pix240's(2 pix 240's) if so, is the problem associated with one or both recorders? Have you tried syncing the r3d's?

I would start there. Try syncing to the r3d and this will tell you if the camera sync is off or the Pix240 recording. You will first need to sort out where the source of the problem is coming from.

BTW- are you aware the PIX240 has an offset adjustment, has it been adjusted?
 
Hi

My Internet is completely screwed. I will try to elaborate tomorrow with screen grabs but we have had 4 of the top techs from the post house scratching their heads. What we are trying to achieve is no different to what multiple other productions do on a daily basis but the inconsistencies still prevail.
 
Well I can assure you syncing cameras is easy and rock solid on the Epic. So you are either doing something wrong or you have a bad piece of cable or gear in the mix. Isolating the source of the problem is your only hope.
Until you determine if the problem is the PIX240, the master clock, the cable or the Epic its hard to offer much more advice.

You should also read pages 24-31
http://www.sounddevices.com/download/guides/pix_en.pdf
 
Thanks. Appreciated. We will test more tomorrow and Internet permitting report back, but like I said earlier the issue is more to do with intermittent takes being OOS rather than nothing syncing at all. If we could understand why 5 out of 25 takes don't sync we'd be on our way to finding a solution. Please keep suggesting solutions.
 
Its important that you have a single master clock, and that no other device is doing the same. Intermittent OOS issues are not un-common when something is not connected correctly or you have the wrong settings/dip switches on lockit box, camera, SD, etc etc.. Ive even seen guys hook up the lockit backwards and swear the camera was faulty. I hate to say it, but from my experience 90% of the tc sync issues are operator error. So dont get hung up on the fact that its not consistent, double check each device in the chain, something is not set up correctly.
 
Greg said it perfectly. Just like in the Highlander movie, with Timecode THERE CAN BE ONLY ONE!!! Master clock reference that is. And in your scenario, THE reference should be the Sound Devices 788t. Dont daisy-chain timecode the way you have described (SD 788 to Epic to Pix). Make it a spoke and hub model, with the 788 at the center. Jam sync all devices from the 788. That will give you the best result.
 
This is what is great about Reduser... if you have an issue, someone has experienced it before, if not it is an amalgamtion of experience to solve a problem. It is late on a Saturday nght in the Uk now, so I will not try and solve the problem without a clear head. I will test tomorrow and report back on where I am at and maybe ask a few more questions. Thanks for the help to date.
 
You must log in or register to reply here.
Back
Top