Stutter lag, how to fix? HDW5231R-ZE
- Thread starter m3tpe
- Start date
I don't know if this would cause it, but for VBR I've always set the Max Bit Rate to the high-end of the Reference Bit Rate (which is 6912 in your screenshot above). Maybe try bumping the Max Bit Rate up.
Both of your values seem reasonable. I don't think they absolutely have to be the same (Dahua ships them from the factory where the iFrame rate is set 2x the FPS rate) but they being the same shouldn't cause any problems.
When you say they're stuttering/lagging in Blue Iris, how are you watching them? In the BI console? Smartphone app? On a web browser on another computer connected to the UI3 web server?
CCTVCam
Known around here
- Sep 25, 2017
- 2,875
- 3,758
Stuttering might be the frame rate depending on what it's affecting. At 10fps, there will be gaps between each recorded frame. The length will depend on light conditions (and thus shutter speed). Don't know if you can set shutter speed independently in BI.
Whilst that 10fps is fine for pedestrian prowlers, if you're looking at a faster moving object eg a speeding car, it can have travelled some distance in the time between frame captures, which results in it appearing to jump or stutter down the road. To put it another way, at 10fps each frame is potentially 1/10 of a second in duration or 0.1 seconds in duration. Apart from the fact that the shutter is probably faster than that. At a 1/50th second, each frame will be (if my maths is correct - it's late at night here and I'm tired), 1/5th of 0.1 seconds = 0.02 seconds in duration. Between each frame therefore, 0.08 seconds goes un-captured. Any significant movement of an object during the period when the camera isn't recording will result in a visible jump of the object or stutter, as it moves down the road. That is why motor racing videographers etc. don't set overly fast shutter speeds. There's also the rule of setting the shutter at 180 degrees. A slower shutter helps reduce the interval between captures at the cost of a little blur.
Not saying it is this, but just be aware. Example here (bear in mind this is probably shot at 30fps):
You can see how a faster shutter = more gaps between recorded frames, starts to introduce stutter and this is most probably at 30fps, not 10 fps.
If this isn't you problem, have you checked your pc specs and tested it's performance? There might be a processing or read/ write issue.
Whilst that 10fps is fine for pedestrian prowlers, if you're looking at a faster moving object eg a speeding car, it can have travelled some distance in the time between frame captures, which results in it appearing to jump or stutter down the road. To put it another way, at 10fps each frame is potentially 1/10 of a second in duration or 0.1 seconds in duration. Apart from the fact that the shutter is probably faster than that. At a 1/50th second, each frame will be (if my maths is correct - it's late at night here and I'm tired), 1/5th of 0.1 seconds = 0.02 seconds in duration. Between each frame therefore, 0.08 seconds goes un-captured. Any significant movement of an object during the period when the camera isn't recording will result in a visible jump of the object or stutter, as it moves down the road. That is why motor racing videographers etc. don't set overly fast shutter speeds. There's also the rule of setting the shutter at 180 degrees. A slower shutter helps reduce the interval between captures at the cost of a little blur.
Not saying it is this, but just be aware. Example here (bear in mind this is probably shot at 30fps):
You can see how a faster shutter = more gaps between recorded frames, starts to introduce stutter and this is most probably at 30fps, not 10 fps.
If this isn't you problem, have you checked your pc specs and tested it's performance? There might be a processing or read/ write issue.
SouthernYankee
IPCT Contributor
But a video on youtube with an example of the stuttering from your cameras. Post it here.
No camera data should be passing through the router.
Are you using a POE injector or are you using a POE switch. I recommend a quality POE switch.
No camera data should be passing through the router.
Are you using a POE injector or are you using a POE switch. I recommend a quality POE switch.
Heres' how the stuttering looks like
https://photos.app.goo.gl/AGB6ZBT4zGTMj5mc7
I'm using https://www.amazon.com/gp/product/B0796ZJPJ7/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1
should I get AV2000 PoE and separate injector?
Last edited:
As an Amazon Associate IPCamTalk earns from qualifying purchases.
bigredfish
Known around here
For faster moving things like cars, like @CCTVCam mentioned, a faster frame FPS can help. There is a 25 MPH road in front of my house and that cam is set to 20 FPS and cars can still look a little jerky/stuttery when playing back a video, but the still images are fine so I haven’t adjusted it higher. My other cameras only record slower things (cars pulling into driveway or people walking by house). Those I have set at 10 or 15 FPS.
So if you look at the timestamp for 1:01, the black car that goes downhill. Isn't that lag? The time ticks on the second, but no movement of the car.
There's 2 cams that are running a bit weird and have this same issue. Other cams are running fine and the time stamp is ticking by the second. But these 2 cameras would get stuck and skip 4-6 seconds before skipping a frame. I wonder if it's the Powerline bottlenecking. My main powerline adapter is AV2000, but have 4x AV200 because PoE and powerline is built in. The two cameras that are lagging and stuttering are connected to the AV200
Last edited:
CCTVCam
Known around here
- Sep 25, 2017
- 2,875
- 3,758
Reviewing your video, it looks like some kind of bottleneck or lag. The even stuttering could be as I described but then in your video there's several places where the cars, especially the black one tag a huge jump instead of a regular one like you'd expect from a shutter issue. This to me looks more like dropped frames which implies a bottleneck in processing or writing the data.
You should probably post your pc specs and runs some performance tests on your pc. It could be anything on the pc from processing to memory to write performance to even a failing hard drive.
One way of seeing if it's a bottleneck or the cameras might be to disconnect all your other cameras other than the affected two and see if it still happens. You could even test each camera individually and then add others back one at a time. If it doesn't happen with just the two, it implies a performance issue. If it still happens, it implies it may be something unique to the 2 cameras and more testing is needed.
I also second the opinion of others in that the bit rate looks awful in any event.
BTW, one last suggestion, you mention you suspect a faulty AV2000 of which you have 4. Checking this is easy, just switch it with another one and see if the cameras affected change with the switch. If they do the AV2000 is faulty. If they don't then you need to look elsewhere.
You should probably post your pc specs and runs some performance tests on your pc. It could be anything on the pc from processing to memory to write performance to even a failing hard drive.
One way of seeing if it's a bottleneck or the cameras might be to disconnect all your other cameras other than the affected two and see if it still happens. You could even test each camera individually and then add others back one at a time. If it doesn't happen with just the two, it implies a performance issue. If it still happens, it implies it may be something unique to the 2 cameras and more testing is needed.
I also second the opinion of others in that the bit rate looks awful in any event.
BTW, one last suggestion, you mention you suspect a faulty AV2000 of which you have 4. Checking this is easy, just switch it with another one and see if the cameras affected change with the switch. If they do the AV2000 is faulty. If they don't then you need to look elsewhere.
Last edited:
I have a TP-Link AV2000 Powerline Adapter kit. Then I bought TRENDnet Powerline 200 AV PoE+ Adapter to use on 4 other cameras because they are situated in individual locations where i couldn't use a switch. From what I read, even if your using AV2000 powerline, adapting them to AV200 would lower the speed. So I'm guessing the AV200 is bottlenecking the entire system?
SouthernYankee
IPCT Contributor
I have tested wire line power units. What you are see is what you will get. Hardwire with a real ethernet cable.
I got jerky video in a test in the same room less than 10 feet apart.
I got jerky video in a test in the same room less than 10 feet apart.