Version 6.0.1

[Tinman]

Yep definitely holding off until things settle out

It does allow you to keep using your existing CPAI running (see arrow) BUT the link does not work for me, tried it on 3 other computers and no luck.

I dropped this on to a shared drive if you want to download it. Let me know when you get it so I can stop sharing.

Awesome...got it, Thanks !

 

[hapstabu]

Weird, I still get this when I click on the link to download the models?

View attachment 234471

FYI, with an unrelated issue, I found that Comcast DNS servers do not have DNS entries for some domains and servers.
I was able to resolve my issue by changing the default DNS settings on my router to Cloudflare.
YMMV

 

[Tinman]

FYI, with an unrelated issue, I found that Comcast DNS servers do not have DNS entries for some domains and servers.
I was able to resolve my issue by changing the default DNS settings on my router to Cloudflare.
YMMV

That was exactly what my issue was, my Mediacom DNS server could not find it yet. Popped in 1.1.1.1 and away it went. Thanks for the tip! Now as far as the new AI, it brought my 12th gen to its knees on my demo machine first time around....still playing with it.

 

[raptor_demon]

i know its early days for this but does it support the Coral TPU's?

 

[hapstabu]

5.9.7.3 working well for me.
I'll watch and wait for now.
"If it ain't broke, don't fix it."

 

[jeremypsl]

Hello, I tried to put yolo11x.onnx model in the folder and choose it in AI interface. It seems to work perfectly, and fast with a RTX3060 12Gb

Same for me, YOLO11 is working on BI. What is your consumption of graphic ram ? I'm at 6,8/8Go with L model. Not sure X will pass on my card.

 

[Il_Pres]

How many possibilities we have to fine tune ALPR ? With the CPAI module you could set the size of plate, size of chars etc. to get the italian plates to be correctly recognised I had to change the default settings a but (ie 3.4:1 ratio)
Thanks

 

[itzaudro]

I just installed 6.0.1 and with my RTX 5090 speed is terrible

View attachment 234458

My times are much lower on an older card... I have an Nvidia Quadro M2000

What device is set in the AI tab?

 

[MikeLud1]

My times are much lower on an older card... I have an Nvidia Quadro M2000

What device is set in the AI tab?

View attachment 234498

My times are now where I expect them to be.

 

[Vettester]

My times are now where I expect them to be.

So what did you do to fix it?

 

[MikeLud1]

So what did you do to fix it?

I missed one of the AI server IP settings that I was using to test an AI server.

 

[Tinman]

So does the new 6.0.1.1 change anything for the good? I can't test it too much now that its night, will try it tomorrow.

 

[CrazyAsYou]

I'm running the latest 6.0.1.3 version and have disabled my CP.AI service in favor of testing the new built-in AI - I'm using the medium model (yolov8m) and so far things seem to be working well and at decent speeds. One thing I don't like is there are no options to fully disable faces, plates etc (models I don't want/use) What I've done is delete all the files from the AI folder and have only left yolov8m.onnx and coco.names - I have also edited coco.names file and have commented out with a - (hyphen) all the objects I don't want to detect (almost all of them).

FYI - deleting the unwanted models makes BI ask if I want to download them again if I go into the AI setting tab (I guess is scans for a list of known files) but other than that everything works as it should.