Video card suggestions

M

marklyn

Pulling my weight
Jun 13, 2015
497
130
Would like to know if buying a video card would improve my overall BI viewing and recording experience when it comes to days where all 15 cameras are picking up a lot of activity (windy days; lots of tree limb and shadow movement).
Most of the cameras I have are ethernet connected and those particular cameras are the ones that see most tree limbs and shadows on windy days.
Right now I have 15 cams on a dedicated BI Windows Intel i5-12400, 32Gb DDR4 memory, running on a Crucial 1Tb SSD drive. My average CPU usage is 2-4% and average memory used is around 40%.
I've followed the BI wiki on making modifications where possible for maximum efficiency.
All in all, I'm fairly happy but would like to hear from anyone who has added a video card and what, if any, improvements you have seen.
I'm considering a GTX 1660 and have one PCIe x16 slot left but don't want to spend $200 or more to not see much of a benefit.
My knowledge and understanding is about average when it comes to this type of hardware so any additional info or thoughts would be appreciated.
 
The first question is do your cameras have built-in AI and if so, using that is probably better than BI motion with CodeProject.

The next question is is the computer pegging out at 100% and getting unresponsive?

The next question is have you spent the time to dial in the motion settings to help knock some of those down? For most people, increasing the make time doesn't miss what they want, but knocks out what they don't want.

If you are not using CodeProject, adding a video card won't gain you anything except a higher electric bill and a warmer computer.

Quick sync/hardware acceleration/offloading video to a video card isn't really needed anymore and it can be an energy hog.

Around the time AI was introduced in BI, many here had their system become unstable with hardware acceleration (hardware decode) (Quick Sync) on (even if not using DeepStack or CodeProject). Some have also been fine. I started to see errors when I was using hardware acceleration several updates into when AI was added.

This hits everyone at a different point. Some had their system go wonky immediately, some it was after a specific update, and some still don't have a problem, but the trend is showing running hardware acceleration will result in a problem at some point.

However, with substreams being introduced, the CPU% needed to offload video to a GPU (internal or external) is more than the CPU% savings seen by offloading to a GPU. Especially after about 12 cameras, the CPU goes up by using hardware acceleration. The wiki points this out as well.

Plus substreams opens up the possibility for older machines to be just fine, along with non-intel computers.

My CPU % went down by not using hardware acceleration.

Here is a recent thread where someone turned off hardware acceleration based on my post and their CPU dropped 10-15% and BI became stable.

But if you use HA, use plain intel and not the variants.

Some still don't have a problem, but eventually it may result in a problem.

Here is a sampling of recent threads that turning off HA fixed the issues they were having....

No hardware acceleration with subs?


Hardware decoding just increases GPU usage?


Can't enable HA on one camera + high Bitrate

And as always, YMMV.
 
As I mentioned, unless you have special use cases or cameras without built-in AI and need to use CodeProject or Deepstack, you may find the camera AI to be better.

Many here use strictly the camera AI and get no false triggers and none of the CPU maxing out during rain events and don't wake up to hundreds of "no object find" triggers either.

You may find that the camera AI has got so good that doing CodeProject is kinda overkill and adds more complexity, time delay, and potential for issues.

Whether to use camera AI or BI AI is obviously up to you, but of course, the AI in the camera may be more than sufficient for your needs without needing BI AI. Do you need the orange box around every object? Do you want to identify animals or logos? Or is just human or vehicle sufficient.

The camera AI is useful to many people, but BI has way more motion setting granularity than the cameras, and some people need that additional detail, especially if wanting AI for more than a car or person. For folks that want AI and alerts on animals or specifically a UPS truck then they need the additional AI.

There isn't really a best practice because every field of view is different and use case and needs are different.

To many here, BI motion without AI is more than adequate for what they do.

To many here, camera AI is more than adequate for what they do.

To many here, using the BI AI adds additional functionality that the above alone can not do.

It comes down to testing with each field of view and which one gives you the most consistent results.


While some of that third party stuff is cool like tagging was it a dog or a bear, I don't need all that fancy stuff. If my camera triggers BI to tag an alert for human or vehicle and BI can accomplish what I need by way of a text or email or push or whatever, that is sufficient for my needs. I just want to be alerted if a person or vehicle is on my property and the camera AI does a fine job with that.

However, I do run BI AI on a few cameras so that it knocks out headlight shine so that the alert image includes the vehicle. The camera AI will trigger for a car, but the alert image was always just the headlights. I also run the ANPR AI module. I am running it with the CPU and not a GPU.

The true test....I have found the AI of the cameras to work even in a freakin blizzard....imagine how much the CPU/GPU would be maxing out sending all the snow pictures for analysis to CodeProject LOL. My non-AI cams in BI were triggering all night. This picture was ran through AI (without the IVS or red lines on it) and it failed to recognize a person in the picture, but the camera AI did. This pic says it all and the video had the red box over it even in complete white out on the screen:

1679354257954.png






See this thread on how using just Dahua AI may be sufficient for your needs (and other cameras with AI would perform similar):

Who uses Dahua AI capable cameras? Reliable AI for triggering events? Pro's/con's?
 
Wittaj, thanks for the detailed response. my cpu never maxes out, ever and I've never had BI console unresponsive. I think the highest I've ever seen it was maybe 24% but it rarely gets to 12% when I'm noticing.
I do have HA (Intel) turned on all of my cameras but turned it off on all cameras a few days ago so see it my CPU or resolutions took a big hit (they didn't). CPU maybe went up 3-5% maybe.
I don't use any of the BI based AI stuff. Have been considering it for certain cameras that are prone to false triggers by shadows or birds or bugs but so far I'm not using it.
I do however use AI on two of my Dahua cameras and let BI trigger from their ONVIF events. That seems to work fairly well but I find myself still tweaking and tweaking until I probably end back up where I started from :)
I did follow as much of the BI wiki tweaks (using substreams, etc.) and feel that helped but that was 3-4 years ago and I still utilize those tweaks on existing and new cameras.
My CPU benchmark is 19492 so that should handle what I throw at it.
I refreshed my memory by revisiting the links you provided so, based on what I've read (again) I think my CPU, onboard graphics and tweaked BI settings are good.
My main issue is on windy days it seems like the camera feeds start some mild pixelation and aren't their usual crisp/clear selves.
when I look at my BI log, I see my total MP/s is around 90 - 100
I may have been looking down the wrong rabbit hole and thinking a high powered video card would resolve this for me but I'm thinking I need to look somewhere else.
Again. Thanks!
 
You must log in or register to reply here.
Top Bottom