For those who are interested, I did some testing comparing the pi-zero, pi-3A+, and pi-3B+ and wrote it up in a blog post. You can access it here:
http://coloradodigital.net/03-11-2019.html
Feedback welcomed!
pi testing
pi testing
Robert (Bud) M. Talbot III
W0RMT
Colorado Digital system admin
http://coloradodigital.net
DMR: BM 31088
DSTAR: XRF/DCS303D
YSF: 99256
NXDN: 31088
P25: 31088
W0RMT
Colorado Digital system admin
http://coloradodigital.net
DMR: BM 31088
DSTAR: XRF/DCS303D
YSF: 99256
NXDN: 31088
P25: 31088
Re: pi testing
Excellent article, Bud!
Previously, I had been replacing my RPi Zero W-based hotspots with the new RPi 3A+ in order to take advantage of its better performance while still enjoying its relatively compact form factor. Based on what I learned from reading your article, I was inspired to go ahead and try replacing the RPi 3B+ in one of my main hotspots that I use for nets (running Pi-Star 4.0.0-RC3) with the RPi 3A+, too. My results were excellent. Previously, that hotspot was running quite warm when I used it on a longer net (~59 C). Yesterday afternoon, I had it linked to a very busy talkgroup for several hours, then switched to the Colorado Hotspot Discussion net (via BM TG 31088) yesterday evening, and averaged ~45 C with no noticeable change in performance. The only downside of the RPi 3A+ is no Ethernet port, but since I always run over WiFi, that's fine for me.
Thanks again for the great article!
Edit: My observations are all with the Pi-Star dashboard open in a browser. I really like having the dashboard open, especially when I participating in a net.
Previously, I had been replacing my RPi Zero W-based hotspots with the new RPi 3A+ in order to take advantage of its better performance while still enjoying its relatively compact form factor. Based on what I learned from reading your article, I was inspired to go ahead and try replacing the RPi 3B+ in one of my main hotspots that I use for nets (running Pi-Star 4.0.0-RC3) with the RPi 3A+, too. My results were excellent. Previously, that hotspot was running quite warm when I used it on a longer net (~59 C). Yesterday afternoon, I had it linked to a very busy talkgroup for several hours, then switched to the Colorado Hotspot Discussion net (via BM TG 31088) yesterday evening, and averaged ~45 C with no noticeable change in performance. The only downside of the RPi 3A+ is no Ethernet port, but since I always run over WiFi, that's fine for me.
Thanks again for the great article!
Edit: My observations are all with the Pi-Star dashboard open in a browser. I really like having the dashboard open, especially when I participating in a net.
Last edited by KE0FHS on Sat Mar 16, 2019 2:54 pm, edited 1 time in total.
73, Toshen, KE0FHS
Playing with Pi-Star (unofficial notes about setting up and using Pi-Star):
https://amateurradionotes.com/pi-star.htm
Playing with Pi-Star (unofficial notes about setting up and using Pi-Star):
https://amateurradionotes.com/pi-star.htm
Re: pi testing
Thanks, Toshen. And thanks for sharing your findings about the operating temperature. Yes, the lack of Ethernet Port is s bummer for more permanent benchtop uses, but not a dealbreaker, I think.
One other thing of note from some testing this morning on the processes running on a pi-star based hotspot:
- pi-zero W w/N5BOC duplex board, no display attached
WIth the dashboard open and being served by the hotspot, 1' load average are over 1.4. With the dashboard *not* open and no dash being served, 1' load averages are about 0.9. the php processes are really taxing the system in order to serve up the dashboard. Take home message: don't have your dash up if you don't need it!
One other thing of note from some testing this morning on the processes running on a pi-star based hotspot:
- pi-zero W w/N5BOC duplex board, no display attached
WIth the dashboard open and being served by the hotspot, 1' load average are over 1.4. With the dashboard *not* open and no dash being served, 1' load averages are about 0.9. the php processes are really taxing the system in order to serve up the dashboard. Take home message: don't have your dash up if you don't need it!
Robert (Bud) M. Talbot III
W0RMT
Colorado Digital system admin
http://coloradodigital.net
DMR: BM 31088
DSTAR: XRF/DCS303D
YSF: 99256
NXDN: 31088
P25: 31088
W0RMT
Colorado Digital system admin
http://coloradodigital.net
DMR: BM 31088
DSTAR: XRF/DCS303D
YSF: 99256
NXDN: 31088
P25: 31088
Re: pi testing
Hello Toshen and Robert,
thanks for your testing AND presenting to us the really interesting outcome.
Best Regards
Kai
dh3ybe
thanks for your testing AND presenting to us the really interesting outcome.
Best Regards
Kai
dh3ybe