Jump to content

WARNING: Audio jumps in Zoom F6 recording.


Sound

Recommended Posts

9 hours ago, Joe A said:

No, nothing directly hardware or operation related, or with storage choices.   Did you receive a new unit yet from Amazon?

 

 

No, still have the second unit with the same issues. The ifs version is not available anymore.

Link to comment
Share on other sites

Report of another late yesterday, dual speed recording, F6, a good ExPro card, ....  It's the most recent posting on the Facebook group F8n-6-3-2 forum.  I suspect there are a good number more with machines that are not being pushed.  I posted the usual and a reference to this thread.  You'll go nuts there trying to calm things down between the various posts and comments.  Zoom has to get this under control and issue a SW fix or a recall.

Link to comment
Share on other sites

Has to be the HW/components, yet I don't see any in dealers used equipment inventory, and on auction and big sales sites the F6 listings are New at or above list price.  By the way, the "GE" might be Germany models.  I see "US" here in the old model codes.  "IFS" is unknown, or maybe a Italy France Spain code; all leading back to, same batch/same problems.  You cannot be so lucky to receive so many duds out of the number that are surely being sold.

Link to comment
Share on other sites

Up very early...   Operation issues:

 

Do these recorders you've received fail out on the box or during use?  Use of 5min, ...?  How long? 

 

Back in the beginning with my F6 copy, I kept losing date/time and emailed support.  Turns out, despite my having AAs and an NP-F batteries loaded, they needed to be on USB power, they said, overnight or a day or so. I forget.  Yesterday, after sitting these last several months unused, but with usual charged batteries, date/time was gone again.  All good this AM after it's USB bath.  No documentation for this.

 

Next, do they fail during use?  Duration?

Do they fail under specific loads or conditions or settings?  6 mics?  3 MS pairs?  Ambisonic?  You mentioned dual mode.  With/without timecode?

 

Fill in anything that has been done and repeated.

 

This is a new recent problem. Maybe a printed circuit or surface component after it heats up?  Longer in the cold, shorter inside?  Powering, since there is a cascade order of 3 possible sources?  I read earlier today of a problem with an F4 powering-up.  Still use mine at times, and it lost date/time last summer, but it's always on external DC.  Undocumented/unknown there too.  

 

??

 

Link to comment
Share on other sites

5 hours ago, Joe A said:

Up very early...   Operation issues:

 

Do these recorders you've received fail out on the box or during use?  Use of 5min, ...?  How long? 

 

Back in the beginning with my F6 copy, I kept losing date/time and emailed support.  Turns out, despite my having AAs and an NP-F batteries loaded, they needed to be on USB power, they said, overnight or a day or so. I forget.  Yesterday, after sitting these last several months unused, but with usual charged batteries, date/time was gone again.  All good this AM after it's USB bath.  No documentation for this.

 

Next, do they fail during use?  Duration?

Do they fail under specific loads or conditions or settings?  6 mics?  3 MS pairs?  Ambisonic?  You mentioned dual mode.  With/without timecode?

 

Fill in anything that has been done and repeated.

 

This is a new recent problem. Maybe a printed circuit or surface component after it heats up?  Longer in the cold, shorter inside?  Powering, since there is a cascade order of 3 possible sources?  I read earlier today of a problem with an F4 powering-up.  Still use mine at times, and it lost date/time last summer, but it's always on external DC.  Undocumented/unknown there too.  

 

??

 

I think the problem can show with a lot of settings. But the more data you record, the more likely it gets. So with six channels dual recordings dropouts seem more likely, but I had the same with 24 bit only recording. I think it was six channels again which I always turn on for a better workflow. (Its easier to batch assign track labels and so on if all clips have the same number of tracks and you dont have to restart recording and set all levels manuell again every time you need another track during the shoot..) Its always six mono channels. Maybe - but only maybe - its less likely in 32 bit as in 24 bit only but I am not sure yet.

The problem described was beause of the failing full card test. I tested almost all my cards and as always - ALL SDXC-Cards fail. (And again: Its a LOT of high class and tested cards - they are not the problem.)

So I guess there will be the same issues.

Link to comment
Share on other sites

Thanks.  At this point the test is loading the recorder with mics in 24, and 32, and in dual modes (three tests), and running the recording until/if it breaks.  I could do the same here with my unit and maybe my F4 or F8n or all three as a concurrent control.  What does a break exhibit while the recorder is running?   (I am not going to listen to hour after hour of daily living sounds in hope of hearing the event, as pleasant as that would be.)

Link to comment
Share on other sites

If you run all of your recorders at the same time with the same timecode recording the same audio source, you can just sync them via timecode, scroll to the end of each audio file of the f6 and see if they are still in sync. If not you can just go back and see if its in sync at the beginning of the file and see where it started to jump..

Link to comment
Share on other sites

The delay here is taxes.  Got the last two income statements yesterday and Wed, and needed to break in hard today; two weeks to go, Fed and state.

Some ground setup questions: I am going to use one card, even in the 8n, transferring and reformatting in-machine.  You have a choice of the ExPro 95MB/s or the 170 label, both SDXC UHS-I, both 128GB.    I will set TC to RTC internal, and I typically use 29.97d frames, but this is a test so your choice over all.  I have 8 Senn 80xx and will choose either 6 or all 8 in the 8n, and use them in the same first-6 order and port number in the F6, all plugged directly into the recorder - no cables.  24bit/48, or different (of course, both in the F6)?  Individual tracks, no LR.  Same Zoom Hirose DC cable power source.  Add anything else.

Link to comment
Share on other sites

Results.   In the most demanding tests in the F6 - 6 tracks, dual 24-36/48 + LR-float (forgot to disable LR),  all 13 files end at the same timecode value, of timecode generated in the F6, RTC.

I used a new ExPro 128GB SDXC UHS-I formatted in the F6.  I am using DaVinci Resolve Studio to display the beginning and ending TC values.

 

Link to comment
Share on other sites

This result does not rule out a HW or FW problem with the newer copies, and perhaps many of these copies have not been pushed hard enough to reveal the condition.  Yes, the frustration is understood.  I will rerun the above more times, pushing up towards dual/192, or whatever the limits are.  I'll post back here if a failure is triggered, but check in otherwise for positive outcomes.

Link to comment
Share on other sites

23 hours ago, Joe A said:

Results.   In the most demanding tests in the F6 - 6 tracks, dual 24-36/48 + LR-float (forgot to disable LR),  all 13 files end at the same timecode value, of timecode generated in the F6, RTC.

I used a new ExPro 128GB SDXC UHS-I formatted in the F6.  I am using DaVinci Resolve Studio to display the beginning and ending TC values.

 

 

Results 2:    Conditions:  Same card, reformatted in the F6; timecode from F6 as RTC; DaVinci file timecode reader; recordings: dual 24-linear / 36-float + LR-float, giving 13 files/14 tracks; all tracks at 96k sample rate.

 

                >>>>  All files display the same beginning and same ending timecode values.

 

  Although 192k sample rate is available, there are several limitation conditions that prevent choosing the output of all 13 files during a single recording pass, specifically, the recording options of using "dual" or a setting of "float".

 

  This 96k test puts the greatest stress on F6 internals, and on the quantity of simultaneous file output written to the SD card.

Link to comment
Share on other sites

That was my original question: how would you detect the small gap during live operation, so it would not be necessary to listen to hours of recordings to hear a gap?  

 

I use continuous timecode to sync audio recording segments to video segments, and feed that unique timecode to both the recorder and the camera.   In this case, my recorder generates the timecode continuously, and I would feed that timecode stream to my camera as it records streams of video.

 

If it is possible that the fault you see with your F6 is also in my F6, and the evidence was in the recording files by reading the beginning and ending timecode values in each 2G break file, then, no, my F6 does not have that defect.  Ending minus Beginning values are the same for each, and each file is exactly 2G, except for the last file with <2G of ending odd storage space.

 

 

 

Link to comment
Share on other sites

I am afraid that wont work.

If you feed timecode to you recorder or camera, most of the time only the starting timecode is recorded frome the generator - or converted later in software in case its a LTC-Timecode.

When recording is running, cameras just count their frames and audio recorders count their samples to continue.

It doesnt matter whats being recorded.

 

So as I said: you need a reference audio.

Both recorders need to record the same audio signal at the same time.

When they are timecode synced and you hear the source in sync at the end of every clip, there are no audio jumps.

Link to comment
Share on other sites

I use Tentacles normally to feed the camera and jam the recorder, and Blackmagic for one brand, makes this distinct with a feed track for the Tentacle and a separate feed from the recorder or an external or internal microphone.  I'll even use a splitter on stereo TRS feeds with the Tentacle if that is the only option.  Never had a problem like your example with Tentacle TC or recorder TC.  Same solid sync result at post.  But the F6 got a good workout.  

Link to comment
Share on other sites

I don’t exactly get it.

If you feed audio ltc timecode to your recorder on an audio track and the recorder tc has been synced previously to your f6 then you could compare audio tc with file tc.

but you have to split the file first towards the end while keeping file tc and after that read the audio tc of this part only. As the audio tc is read only at the start of the clip.

did you do that?

Link to comment
Share on other sites

I have the same problem! I bought my F6 unit in 2020-05-20, so I don't think that it's related to the new components. Anyone found how to find these shifts easily by waveform or TC? My videos are not always easy to sync by image, so it's so big hassle to sync audio afterall, as I found - that if you record 2 hours of video / audio it may jump 3-4 times during this time.

 

I record in 96Khz / 32 bit / 2 Mics

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...