IVS event skipping frames on the local microSD recording

[rado stefano]

I have a few IPC-HDW5231R-Z with 256gb Sandisk microSDs.
I have been reading through the forum and trying to find more information about the frames skipping when there is an IVS event. Most of the posts are talking about skipping when using a NVR. So I decided to just use the cameras with internal microSDs. Come to find the issue exist when recording on board as well. It is very unfortunate considering IVS event is the time when everything should work fine...
The cameras around my house are skipping exactly when a face is supposed and should be seen.
This is my question:
Is anybody else experiencing the IVS skipping when recording directly to onboard microSD?
Any way to fix that issue?
Is buying a certain NVR with certain firmware going to help?
Thanks

 

[Crazykiller]

As you read the other posts you might have seen, that's a general problem how Dahua records events. They don't buffer the full frame and don't start recording just in time instead with a full frame. You might decrease the gap a little bit by reducing the iFrame rate to the same value as the fps is. So you only one second instead of two seconds while you are "blind".

 

[rado stefano]

have this always been the case with the dahua cameras?

As you read the other posts you might have seen, that's a general problem how Dahua records events. They don't buffer the full frame and don't start recording just in time instead with a full frame. You might decrease the gap a little bit by reducing the iFrame rate to the same value as the fps is. So you only one second instead of two seconds while you are "blind".

 

[rado stefano]

However, members here have come up with a number of workarounds that are hopefully temporary before Dahua engineers develop some kind of solution.

What solutions?

 

[SquareEyes]

What solutions?

Mate, Crazykiller is one member has been trying to help you and he is a member that has found a workaround.

If you want solutions, maybe try working with the people that are trying to work with you. Nobody is getting paid to help you, they are just good people.

 

[aristobrat]

Is buying a certain NVR with certain firmware going to help?

Not that I've seen...

I switched to Blue Iris. Blue Iris can't see IVS events on the cameras, so I turned them all off and use Blue Iris's built-in Motion Detection. For my setup, Blue Iris Motion Detection is on pretty much on par with Dahua's IVS. I've never had Blue Iris miss an event, but Blue Iris will sometimes trigger when the scene goes from sunny to cloudy real-quick (or vice-versa). IVS on my Dahua cameras would not trigger on that.

They are a huge PIA that could likely be easily remedied if the cameras could be configured (providing they have ample buffer) so that there was a couple of seconds pre-record before the IVS event triggered.

IIRC, if not recording continuously, there is an option where you can specify how many seconds of video to record before the IVS event.

When recording continuously, I wish they'd do what Blue Iris does and just have one recording, and separately keep "bookmarks" to when IVS/MD events occurred in that file.

 

[TechBill]

It a bug. Even the color flag for motion detect and IVS is wrong too. I am taking some screenshot and video to send to Andy hopefully they will resolve it soon.

 

[SquareEyes]

IIRC, if not recording continuously, there is an option where you can specify how many seconds of video to record before the IVS event.

Had to recheck myself as I record continuously. Turned of continuous recording and still missing the start of the IVS event.

The only place that I know where you can Pre-event Record is:

Even at max Pre-event it still misses the start.

 

[rado stefano]

So I disabled IVS recording trigger and there are no more skipped frames.
I get a good push timings on the gDMSS mobile app. Starts a little earlier and shows the entire event.
I wish there was a way for the SmartPSS to emulate what the gDMSS is doing with push notification videos. If I could export the times from the mobile app in to the SmartPSS.

 

[TechBill]

I finally gotten around to hooking up my NVR. I have no skipped or dropped frame issue at all when I am playing back with SmartPSS on my laptop. Also on the SmartPSS, the IVS show up as orange flag in the playback timeline which is the correct color. In the SD card playback it was showing up as yellow flag.

 

[rado stefano]

I finally gotten around to hooking up my NVR. I have no skipped or dropped frame issue at all when I am playing back with SmartPSS on my laptop. Also on the SmartPSS, the IVS show up as orange flag in the playback timeline which is the correct color. In the SD card playback it was showing up as yellow flag.

Great news. I will have to try that...
What NVR model?

 

[TechBill]

Great news. I will have to try that...
What NVR model?

This one

Dahua NVR5216-4KS2 / NVR5216-16P-4KS2

 

[EMPIRETECANDY]

This one

Dahua NVR5216-4KS2 / NVR5216-16P-4KS2

Thanks Bill!!
I also get some solutions from dahua guys, hope can help something out, tomorrow i may get some pics for further checking.

 

[Mike Mo]

I finally gotten around to hooking up my NVR. I have no skipped or dropped frame issue at all when I am playing back with SmartPSS on my laptop. Also on the SmartPSS, the IVS show up as orange flag in the playback timeline which is the correct color. In the SD card playback it was showing up as yellow flag.

You must be an aberration. Everyone else that I've spoken to or emailed has a problem with IVS recorded events. There is a fundamental problem with how dahua records these events. The code they use has a serious bug.

 

[rado stefano]

What is the status on this???

 

[Crazykiller]

What is the status on this???

I have no issues with the actual firmware for the NVR, except that I couldn't control multiple PTZ now, which really bugs me. Hopefully they'll release an official, functioning update soon...

 

[carteriii]

Rather than create a new thread, I'm replying to this old one to ask if anyone is aware of any fixes to IVS missing the start of a recording? Has Dahua even acknowledged this problem?

I do record 24x7 onto Samsung Endurance Pro cards but lose 1-2 seconds with every IVS event, which is ironic since that's arguably the single-most important 1-2 seconds to capture. Other than the advice above of using the lowest possible iframe rate (matching the fps), is there any newer advice to workaround or minimize the problem?

 

[excel]

Any updates on this issue? I have the same problem with 2 recently purchased IPC-T5442TM-AS units with 128GB MicroSD cards installed. (thanks @EMPIRETECANDY)

I have both continuous recording and IVS Tripwire/Intrusion recording rules set. What I am finding is there is sufficient overlap between the regular and motion recordings to stitch together a continuous video, but not without a lot of manual effort. SmartPSS and the mobile apps should be able to do this but they struggle at the moment.

Looking deeper into the Playback tab in the WebUI, this is what I see:

Event 118: 12:32:12 >> 12:32:29 (motion)
Event 119: 12:32:24 >> 12:34:48 (general) -- note this file should be over 2 min length, but when exported from SmartPSS the file only runs to 12:32:29, which is precisely when the prior event ends! Exporting from WebUI gives the full file.
Event 120: 12:34:41 >> 12:35:00 (motion) -- again SmartPSS export runs to 12:34:48 (end of previous event) only
Event 121: 12:34:54 >> 12:35:27 (general) -- again SmartPSS export runs to 12:35:00 (end of previous event) only
Event 122: 12:35:22 >> 12:35:39 (motion) -- again SmartPSS export runs to 12:35:27 (end of previous event) only
.. etc

If I clip the entire period and export as a single file, the playback skips around a lot - it skips back 5 seconds, repeats, then jumps forward, resulting in the behaviour many have reported during regular playback in SmartPSS. The mobile apps behave similarly, though sometimes they quit playing entirely at one of these overlap points.

The only way to extract the full video is to download each clip individually from the camera UI, convert to AVI/etc and stitch manually.

I suppose this then points back to an issue with the stitching logic in SmartPSS and the mobile apps. It appears from other threads that this was solved in recent NVR firmware - does this suggest we are awaiting the same fix applied to SmartPSS and mobile directly connected to cameras? @EMPIRETECANDY can you help here?

 

[excel]

Any updates on this issue? I have the same problem with 2 recently purchased IPC-T5442TM-AS units with 128GB MicroSD cards installed. (thanks @EMPIRETECANDY)

I have both continuous recording and IVS Tripwire/Intrusion recording rules set. What I am finding is there is sufficient overlap between the regular and motion recordings to stitch together a continuous video, but not without a lot of manual effort. SmartPSS and the mobile apps should be able to do this but they struggle at the moment.

Looking deeper into the Playback tab in the WebUI, this is what I see:

Event 118: 12:32:12 >> 12:32:29 (motion)
Event 119: 12:32:24 >> 12:34:48 (general) -- note this file should be over 2 min length, but when exported from SmartPSS the file only runs to 12:32:29, which is precisely when the prior event ends! Exporting from WebUI gives the full file.
Event 120: 12:34:41 >> 12:35:00 (motion) -- again SmartPSS export runs to 12:34:48 (end of previous event) only
Event 121: 12:34:54 >> 12:35:27 (general) -- again SmartPSS export runs to 12:35:00 (end of previous event) only
Event 122: 12:35:22 >> 12:35:39 (motion) -- again SmartPSS export runs to 12:35:27 (end of previous event) only
.. etc

If I clip the entire period and export as a single file, the playback skips around a lot - it skips back 5 seconds, repeats, then jumps forward, resulting in the behaviour many have reported during regular playback in SmartPSS. The mobile apps behave similarly, though sometimes they quit playing entirely at one of these overlap points.

The only way to extract the full video is to download each clip individually from the camera UI, convert to AVI/etc and stitch manually.

I suppose this then points back to an issue with the stitching logic in SmartPSS and the mobile apps. It appears from other threads that this was solved in recent NVR firmware - does this suggest we are awaiting the same fix applied to SmartPSS and mobile directly connected to cameras? @EMPIRETECANDY can you help here?

After more experimentation it appears the serious issue with video skipping back and forth is limited to Intrusion type IVS rules. After changing to a Tripwire style rule the problem is much less pronounced -- only about 0.5 seconds like everyone else reported. Still an issue but far less problematic.