Skip to contentSkip to conversation

Rate vs Range graphs for ath10k through attentuator.

DISCUSSION Ben Greear 2 participants 4 messages May 15, 2014 – May 16, 2014 Qualcomm Atheros ATH10K wireless driver

lore
Share & tools
lore

Ben Greear

author
#0 lore raw

Just FYI...here's a graph of a UDP upload test through an attenuator. The attenuator goes from 0 up to 95.5, though isolation is not perfect so even at highest attenuation a small bit of signal gets through.

One interesting thing to me is that ath10k does so well at very hot signal strength. ath9k runs like crap with hot signal...

AP is another of our systems, both are using my CT firmware, my 3.14.4+ kernel, and WLE900VX NICs. 32 stations were created and associated on the station machine, but only one was actively sending any significant data. AP system had single active AP vdev, one station interface that was not connected. Traffic flow is station send to AP, AP sends back to station machine on 1Gbps wired Ethernet connection. Both systems are mobile core-i7 CPU, 8GB RAM, etc.

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/

Lots of graphs in the link above..the most useful ones are:

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-0.png

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-30.png

The throughput is a bit spikey, possibly because I am only doing 5 second throughput tests. Will run a 30-second one some day and hope that smooths out.

Thanks,
Ben

Signature · 8 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com


_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
#1 lore raw
Quoted 17 lines · Ben Greear

Just FYI...here's a graph of a UDP upload test through an attenuator. The attenuator goes from 0 up to 95.5, though isolation is not perfect so even at highest attenuation a small bit of signal gets through.

One interesting thing to me is that ath10k does so well at very hot signal strength. ath9k runs like crap with hot signal...

AP is another of our systems, both are using my CT firmware, my 3.14.4+ kernel, and WLE900VX NICs. 32 stations were created and associated on the station machine, but only one was actively sending any significant data. AP system had single active AP vdev, one station interface that was not connected. Traffic flow is station send to AP, AP sends back to station machine on 1Gbps wired Ethernet connection. Both systems are mobile core-i7 CPU, 8GB RAM, etc.

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/

Quoted 8 lines

Lots of graphs in the link above..the most useful ones are:

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-0.png

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-30.png

The throughput is a bit spikey, possibly because I am only doing 5 second throughput tests. Will run a 30-second one some day and hope that smooths out.

Cool charts. Was this over the air or with cables?

Signature · 6 lines
Kalle Valo

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
author
#2 lore raw
Quoted 21 lines · Kalle Valo
Quoted 19 lines · Ben Greear

Just FYI...here's a graph of a UDP upload test through an attenuator. The attenuator goes from 0 up to 95.5, though isolation is not perfect so even at highest attenuation a small bit of signal gets through.

One interesting thing to me is that ath10k does so well at very hot signal strength. ath9k runs like crap with hot signal...

AP is another of our systems, both are using my CT firmware, my 3.14.4+ kernel, and WLE900VX NICs. 32 stations were created and associated on the station machine, but only one was actively sending any significant data. AP system had single active AP vdev, one station interface that was not connected. Traffic flow is station send to AP, AP sends back to station machine on 1Gbps wired Ethernet connection. Both systems are mobile core-i7 CPU, 8GB RAM, etc.

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/

Lots of graphs in the link above..the most useful ones are:

Quoted 8 lines
Quoted 6 lines

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-0.png

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-30.png

The throughput is a bit spikey, possibly because I am only doing 5 second throughput tests. Will run a 30-second one some day and hope that smooths out.

Cool charts. Was this over the air or with cables?

Cables, have to run it through the attenuator...

Ben

Quoted 1 line
Signature · 7 lines
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k
#3 lore raw
Quoted 10 lines · Ben Greear
Quoted 8 lines
Quoted 6 lines

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-0.png

http://www.candelatech.com/examples/ath10k-rate-vs-range-udp-upload/chart-30.png

The throughput is a bit spikey, possibly because I am only doing 5 second throughput tests. Will run a 30-second one some day and hope that smooths out.

Cool charts. Was this over the air or with cables?

Cables, have to run it through the attenuator...

Doh, sorry. I was just looking at the maximum throughput and didn't think at all.

/me crawls back to his hole

Signature · 6 lines
Kalle Valo

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k