Welcome to our community

Be a part of something great, join today!

  • 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 :)

Epic Audio Problems

Is there a special WAV file that could be recorded, where separate tones for each frame of 23.976 were repeated each second so that it would be easy to look at the first frame and the last and see if there was any drift?

Typically, I recommend that my clients shoot a camera test prior to the first production day where they try out the work flow: run the camera normally, clap a regular timecode slate at the head, roll for at least 2-3 minutes, then clap at the tail. Load the R3D files and the separately-recorded sound files into a laptop, convert the R3D to whatever the client plans to use for editing. Sync up the material and verify timecode matches and sound does not drift. If it's even 1 frame off at the tail, you'll know it.
 
I too have been experiencing sync problems. Scarlet #503 - updated to v3.0.7 recently.

We did a sync test with v2.0.5 - everything was aok.

We are currently tryng to resolve our issue & have done a couple of long records where the sync appears fine. But the production I have pulled this camera off from has recorded several scenes with random sync problems over the last couple of days. The sync is not holding in the guide track or the production track. We are most probably going to have to reshoot ! - I have been wondering if the genlock feed going into the camera has been causing the problem (until I saw these sync posts)

We are shooting with 2 x Scarlets, the other Scarlet #870 sync appears ok.

THIS IS AN EXTREMELY SERIOUS AND URGENT ISSUE, THAT RED NEEDS TO ADDRESS NOW !!!!

Red please respond ASAP.
 
Is this a firmware issue? I am running 3.2.8 but I was not shooting HDR. On Friday, I was asked to send my camera in to have this problem looked in to. But if it is as easy as updating firmware I would rather not send it back.
 
Typically, I recommend that my clients shoot a camera test prior to the first production day where they try out the work flow: run the camera normally, clap a regular timecode slate at the head, roll for at least 2-3 minutes, then clap at the tail. Load the R3D files and the separately-recorded sound files into a laptop, covert the R3D to whatever the client plans to use for editing. Sync up the material and verify timecode matches and sound does not drift. If it's even 1 frame off at the tail, you'll know it.
Excellent!
 
I upgrade to firmware 3.2.13 and ran a test at 3kHD and 5K and sync was fine.
 
I've also had audio synch problems - it seems to be intermittent, rarely happend and then on only on certain clips. The first time I encountered this we were shooting long take interviews, the SSDs only had 9% remaining and the camera also got quite hot (as we had the fans turned down). I was unsure which of these factors might have been effecting the audio synch. Red suggested that there could be a problem with camera's clocks, but I haven't found a week of downtime to send it in for checks.

I have only encountered the problem again once - also shooting long takes, but no heat problem, and will SSD's going beyond 10% remaining. I'm thinking it might be this combination that causes the problem.
 
I've also had audio synch problems - it seems to be intermittent, rarely happend and then on only on certain clips. The first time I encountered this we were shooting long take interviews, the SSDs only had 9% remaining and the camera also got quite hot (as we had the fans turned down). I was unsure which of these factors might have been effecting the audio synch. Red suggested that there could be a problem with camera's clocks, but I haven't found a week of downtime to send it in for checks.

I have only encountered the problem again once - also shooting long takes, but no heat problem, and will SSD's going beyond 10% remaining. I'm thinking it might be this combination that causes the problem.

@ Mark

My issues happened when the card was completely empty and the takes were short takes (1 minute). I have since upgraded to 3.2.13 and sync issue seems to be fixed. What firmware version were you using when you encounter the issue? BTW - I have sent a new log file since upgrading firmware and Red is looking at it. Big shoot coming up hopefully the problem is solved!!!
 
Hey Greg,

Unfortunately... Don't pin 100% confidence on Firmware Version 3.2.13. I've now seen this problem on both this version & the prior firmware version on my EPIC.

Prior firmware version I was shooting testimonials in 5K HD. For me the clips would play in sync for awhile then a chunk of audio would be missing, then audio from 3-4 seconds down the way is playing to a picture that never skipped forward like the audio. Looked at it in REDCine etc. The missing audio chunk is not there. It certainly seems to be a damaged file created somehow by the camera. If it was out of sync I could fix it. Missing chunks of audio I can not.

I put in Firmware 3.2.13 did some testing the night prior everything played back fine. Then on shoot day with the first 64 Clips everything is way out of sync. Missing chunks, it repeats for now reason, etc. On the Same SSD Card the remaining 70 clips play fine. Two testimonial interviews will have to be re-shot again. In neither case was I shooting HDRx.

One interesting thing that I've noticed is it would seem that the bad sections are isolated by where the camera is powered on & off. So in both cases with me between the time where there's problem clips and non-problem clips is where the camera was shut off and powered back up for a different shooting setup. Also in both cases it's at the beginning of a formatted SSD card & not the end as you somewhat noted.

I've never seen anything like this in my 4.5+ years of owning RED Cameras. I currently have no confidence in shooting any Audio with this camera. Trying to explain to clients that we have to pull these people out again, relight & reshoot is certainly embarrassing to us as professionals. Not to mention time consuming & expensive. Also note for us this is the first time ever having to do so in the 15 years of doing this professionally.

My uneducated guess is that it's a Firmware issue (that's perhaps irregularly triggered by camera start up somehow?)

I will say I have never seen these issues shooting regular "5K". So for me I'm going to pin my hopes on that this particular camera setting (5K) doesn't have these major issues.

Has Anyone seen this in regular 5K mode? (Non 2:1 & HD)

I WOULD HIGHLY RECOMMEND THAT YOU TAKE A LAPTOP ON SET & CHECK YOUR TAKES IN REDCINE TO SEE IF YOU ARE HAVING AUDIO ISSUES. THAT WAY AT LEAST YOU CAN RESHOOT then & there BEFORE YOU FIND OUT DAYS LATER IN POST..

One other side note when I shot "5K HD" beside the major audio issues it played horribly choking in both REDCINE with my ROCKET & in Premier CS6. Regular 5K footage plays beautifully. Yet another reason I'm now only shooting only 5K.

I'm a few days late seeing this 3.2.16 Firmware... Guess I'll install this & pray that this takes care of the issues...

It'll be a few months of shooting problem free to restore my full confidence that the bug is gone.

If anyone has Audio Sync Issues in 3.2.16 PLEASE POST SO THAT WE KNOW IT STILL EXISTS!

Russell
 
Are there timecode breaks in your R3D as well?
Just asking in case I recorded a standalone audio track to sync up with picture on an upcoming job, and wanted to know if a lockbox will be needed.. or would work anymore...
I had gotten used to the Epic staying jammed pretty well with a twice a day jam...
 
Same problem... Epic X, FW 3.2.16. All audio trails video by roughly 5 seconds, and slips further every 14 seconds with a hiccup-repeat of a few words. Meters and headphone return from camera were fine throughout the shoot.
 
I'm experiencing the same problems again, when filming interviews - a loss of synch on some takes. Using Epic and 3.2.19. It would be great to hear from someone at Red on whether this is a recognised problem and if there is a solution in the works - it's very difficult to work with.
 
Back
Top