Timed out, waiting for frame ######. Frame dropped.

Alright, so pretty big problem here. I'm using Resolve 14, though the same thing happens in 12.5 too. I'm trying to render a video for YouTube on the 1080p setting. But if I'm trying to render anything of any useful length, basically more than a few minutes, I get an error message that reads "Timeout when waiting for frame ###### for recording, frame dropped." (where ###### is whatever the frame number happens to be). Once I close the dialog box, Resolve all but freezes, and the only thing I can do is stop the rendering (not that it had continued beyond frame ###### anyway) and close the program. Anyone know how to fix this?

Rig:
MSI GTX1080 Gaming 8G
Intel Core i7 5820K
Windows 10-64 bit
MSI X99A Gaming 7 mobo

Last time I had this issue it turned out to be a failing drive (the one I was trying to render to).

Can you try writing to a different drive and see if that works?

[Peter Chamberlain]

Are you rendering to a fixed internal drive or raid file system?

I'm having similar behavior here.

I was just working on a job with about 17hrs of real-time footage. I typically render dallies overnight directly to one of those small bus-powered USB3.0 drives that the client picks up.

The render speeds on my system have jumped quite a bit with v14 and I'm seeing render speeds around 240fps for DNxHD QT's. I did notice that Resolve will blaze through the clip and then have to wait for the clip to write to the USB drive before starting the next clip.

The render crashed when I was rendering overnight to the USB drive so I restarted the render to an internal RAID-0 on the system but it crashed/timed out once as well. Of the 34hrs of dallies/screening copies I rendered, it timed out twice.

My system
Dual Xeon SuperMicro
Dual TitanX Black GPU's
8-drive RAID (external) around 600MB/s write 750MB/s read
4-drive RAID (internal) around 600MB/s read/write
Atto 664 RAID controller
USB3.0 passport drive 60~100MB/s write

Just giving this thread a bump. I've upgraded to 14.0.1 but still having renders stopped with this error. Resolve then completely hangs up and I have to use Win10 taskmanager to force-quit.

[Peter Chamberlain]

Can you provide logs?

viewtopic.php?f=21&t=54509

I ran into a similar problem, and it turned out to be due to the drive that had my source footage on it. I wasn't able to write to that drive at all, so I was rendering to a different drive, and it worked quite nicely for several days; now I'm experiencing media going inexplicably offline while color grading. And it was just simple grading, just to export some stills from my Helium footage... while I was working on a clip, it was fine, but when I moved to the next one, it went offline and I had to re-add it to the timeline after restarting Resolve.

I don't know yet whether or not the two are related, but there was a Windows update release in between the two :-/

I'm still running into this issue on 14.1.1.005. There seems to be no rhyme or reason as to why it happens--it's independent of any format and it's happening an SMB3 share to a ZFS server. Plenty of other footage with identical tech specs renders just fine, but this bug just occasionally pops up. I've sent logs to the support team.

look at taskmangler, see if one of the disks is running at 100%

when i get this error it's always about disk access and Resolve flooding either the source disk, or target disk...

for exports i only use SSD's for transfer disks, due to the relibalaity due to faster i/o speeds

for imports i try to transfer the media from the cleint drive (usualy a usb connection / 5400rpm disk) to a NAS before working

both of those workflows have the occurance of this error down to near zero now

I believe it might be related to how I export the `.recycle` folder on the SMB share, and whether any incomplete or corrupt duplicates of the file that I'm trying to render exist in that `.recycle` folder.

On the ZFS side, I went ahead and `rm -rf`ed the corresponding destination directories inside of the `.recycle` folder. It seems to have done the trick. Maybe that was somehow clogging the 800 MB/s tube...

Are you rendering to a fixed internal drive or raid file system?

Hi Peter,

Do you find rendering to a fixed internal drive helps or hurts this situation?

Thanks,
Clark

Just to follow up on this thread—my network switch handling the connections between the PostgreSQL server and clients had gone bad, and replacing it fixed these issues.

It’s one more thing to check for folks who are struggling with this issue.


Sent from my iPhone using Tapatalk

[Peter Chamberlain]

Are you rendering to a fixed internal drive or raid file system?

Hi Peter,

Do you find rendering to a fixed internal drive helps or hurts this situation?

Thanks,
Clark

Actually these are the two prefered drive connection types, much better than a single non raided drive on a USB.

I am currently seeing this on 15.1.1 Free - on OSX High Sierra.

Source is 50fps Canon Raw, source media is on Raid & Target is also Raid.

I am currently seeing this on 15.1.1 Free - on OSX High Sierra.

Source is 50fps Canon Raw, source media is on Raid & Target is also Raid.

Unfortunately the GT750M is pretty underpowered. Should probably upgrade your hardware to run 15.

https://documents.blackmagicdesign.com/ ... _Guide.pdf


Sent from my iPhone using Tapatalk

I am currently seeing this on 15.1.1 Free - on OSX High Sierra.

Source is 50fps Canon Raw, source media is on Raid & Target is also Raid.

Unfortunately the GT750M is pretty underpowered. Should probably upgrade your hardware to run 15.

That's home machine... work (where this is taking place) is iMac Retina 5K 27" with the AMD Radeon R9 M395 2048 MB GPU.

work (where this is taking place) is iMac Retina 5K 27" with the AMD Radeon R9 M395 2048 MB GPU.

Also quite underpowered, unfortunately. Check official hardware guide for current Mac recommendations for 15.


Sent from my iPhone using Tapatalk

work (where this is taking place) is iMac Retina 5K 27" with the AMD Radeon R9 M395 2048 MB GPU.

Also quite underpowered, unfortunately. Check official hardware guide for current Mac recommendations for 15.

Still chunks through things pretty nicely, I use it for Resolve Color and Deliver work. Edit is done in Avid or PP. VFX in AE.

The Canon C200 4K Raw chokes Avid & PPro - so creating proxies in Resolve is very useful. And it works *fairly* well. (I like DNx MXF OP1a as both PPro and Avid can use them as proxies pretty flawlessly)

The 'Timed out, waiting for frame' occurs occasionally, but not repeatedly.
v15.2 exports corrupt DNxHR MXF OP1a files, if they are above 100MB
and v15.x has an ackowledged '99%' crash when exporting DNx MXF OP1a files.

I've rolled back to 15.1.2 Studio for now. Gives me access to Canon AVCI source clips & non-corrupt DNx OP1a exports.

I don't think any of those are hardware related, so hopefully reporting where these errors occur, and submitting logs when asked for, hopefully will keep Resolve developing.