CodeProject.AI Version 2.0

Hello!

I've tried 2.1.9. and now 2.0.8. I always only get "Client request 'detect' in the queue (...65c264)" but never a response. Tried reinstalling both versions, restarted the computer (Win 10). I just installed the CP.AI Version and went straight to the /vision.html and tested the AI. But theres never a response.
What am i missing?
 
FYI: Chis has now added a Coral library to CpdeProject-AI, it also is available in windows.
For $30you can get a Coral that slots into your WiFi slot (Chris uses a throttled library so should be fine), I went with the external USB, as I didn't want to risk my motherboard, power requirements can be high,

1689201743888.png
 
  • Like
Reactions: Pentagano and David L
spammenotinoz said:
FYI: Chis has now added a Coral library to CpdeProject-AI, it also is available in windows.
For $30you can get a Coral that slots into your WiFi slot (Chris uses a throttled library so should be fine), I went with the external USB, as I didn't want to risk my motherboard, power requirements can be high,

View attachment 167583
Click to expand...
Have you really purchased one for $30? That has been the "theoretical" list price for a while, but actual retail has been over $100, if you can even find one at all.
 
jrbeddow said:
Have you really purchased one for $30? That has been the "theoretical" list price for a while, but actual retail has been over $100, if you can even find one at all.
Click to expand...
The M.2 module is $24 and in-stock (key B+M), the WiFi Module (same price) is out of stock.
The Dual M.2 module is still $39 (in-stock), CodeProject AI will only use 1 of the modules. Ideally you need to purchase your own heatsink, but as CodeProject AI throttles, it's probably okay.

I also purchased the USB version (yes that's about double $59.99), but that includes the enclosure, cables, heatsync.
Coral USB Accelerator - Seeed Studio
 
  • Like
Reactions: teejnut, fenderman, David L and 1 other person
spammenotinoz said:
The M.2 module is $24 and in-stock (key B+M), the WiFi Module (same price) is out of stock.
The Dual M.2 module is still $39 (in-stock), CodeProject AI will only use 1 of the modules. Ideally you need to purchase your own heatsink, but as CodeProject AI throttles, it's probably okay.

I also purchased the USB version (yes that's about double $59.99), but that includes the enclosure, cables, heatsync.
Coral USB Accelerator - Seeed Studio
Click to expand...
I stand corrected, thanks. I guess I haven't looked in the last couple of months.
 
Last edited:
But is this one coming from the Chinese warehouse so would take some time to arrive? My son is currently in Florida but only for a short time.
What is the delivery time on this?

1689255769932.png

oh dear 15-30 days so no chance
Normal Orders (shipping from CN): US/GB/DE/AU POST, Singapore Post.

*US/GB/DE/AU Post includes the available country as United States, United Kingdom, Germany and Australia.EU Post includes 30 countries in Europe.

*Limiting for battery, includes pure battery and battery used products.

The above shipping methods normally take 15 to 30 working days to reach majority of the world.
Click to expand...
 
Just ordered a coral usb from amazon - but was 83 USD! Got fed up with waiting for over 2 years for one.
I use a local service here (tiendamia) that handles all the shipping and customs to Uruguay from Miami. Got another sff thinkcenter while I was at it to add to my proxmox cluster taking advantage of the shipping as they put them together.

If the coral usb is plugged into my windows BI pc I wonder if I can connect to it on the lan for frigate on another pc also. ??
Have to do some research on that, basically sharing it
 
Last edited:
What kind of performance are you guys getting with the coral? It's pretty abysmal (~300ms) on my end unless I set the models to "small." It ships with them defaulting to "medium," so I'm confused about the amazing performance Chris was getting.
 
tofu said:
What kind of performance are you guys getting with the coral? It's pretty abysmal (~300ms) on my end unless I set the models to "small." It ships with them defaulting to "medium," so I'm confused about the amazing performance Chris was getting.
Click to expand...
I think Chris was using small or tiny
 
MikeLud1 said:
I think Chris was using small or tiny
Click to expand...

Well that makes sense. He should have said so lol. A few guys are bewildered why the performance is terrible.

I assume frigate uses the 300x300 models as well, unless they had a coding breakthrough of some sort
 
tofu said:
What kind of performance are you guys getting with the coral? It's pretty abysmal (~300ms) on my end unless I set the models to "small." It ships with them defaulting to "medium," so I'm confused about the amazing performance Chris was getting.
Click to expand...
yep 300ms is awful. Lightning fast on Frigate! Mine will arrive in 10 days or so. Will find out then.
If it's worse than my gpu I'll use it for my Frigate I have on another pc.
Maybe the new model for coral needs refining/developing??
 
Pentagano said:
yep 300ms is awful. Lightning fast on Frigate! Mine will arrive in 10 days or so. Will find out then.
If it's worse than my gpu I'll use it for my Frigate I have on another pc.
Maybe the new model for coral needs refining/developing??
Click to expand...

Well, on small I seem to be having accuracy issues. It's very easy for me to test because I'm on a high traffic street. I get a lot of false "bus" and "cat." Not looking too good here ;(
 
  • Sad
Reactions: Pentagano
tofu said:
Well, on small I seem to be having accuracy issues. It's very easy for me to test because I'm on a high traffic street. I get a lot of false "bus" and "cat." Not looking too good here ;(
Click to expand...
Lets hope its just to start of their development and will catch up with the correct models that will work well(better than GPU's) with CPAI.
 
You must log in or register to reply here.
Top Bottom