Neither Stable 3.23.4.0 nor Nightly 20230709 attempts 802.11n nor 802.11ac speeds. Using Nightly 20230709 My -nD to nD devices are capable of 130/144 TxMbps link speed. My -ac to -ac and -nD to -ac only attempt 802.11g speeds!
I was hoping to see a MCS8 156/173.3 link between my -ac devices. Especially, I was not expecting to be reduced to 802.11g, 54 TxMbps.
See http://gelm.net/AREDN/whence-ac.html edited.
73, Chuck
54 TxMbps is not a possible 802.11n-ac speed.
https://d2cpnw0u24fjm4.cloudfront.net/wp-content/uploads/2020/09/11102924/802.11n-and-802.11ac-MCS-Chart-1.pdf
73, Chuck
Chuck, TxMbps is not a link speed - it's (mostly) just modulation rate. It looks like the link is running MCS 13 which would be 115.6 Mbps, but since we're running 10 MHz bw, it'd be half that or 57.8 Mbps. I believe the displayed AREDN TxMbps is adjusted for successful packet completion, which in this case is 93% (54/57.8)
Orv W6BI
Hi, Orv:
As advertised in "See http://gelm.net/AREDN/whence-ac.html edited":
nD to ac link: root@NC8Q-LHG5-175:~# cat /sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26/rc_stats best ____________rate__________ ____statistics___ _____last____ ______sum-of________ mode guard # rate [name idx airtime max_tp] [avg(tp) avg(prob)] [retry|suc|att] [#success | #attempts] OFDM 1 6.0M 272 1640 5.2 5.2 100.0 3 0 0 0 0 OFDM 1 9.0M 273 1104 7.7 7.7 100.0 4 0 0 1 1 OFDM 1 12.0M 274 836 10.0 10.0 100.0 4 0 0 0 0 OFDM 1 18.0M 275 568 14.3 14.3 100.0 5 0 0 0 0 OFDM 1 D 24.0M 276 436 18.1 18.1 100.0 5 0 0 1 1 OFDM 1 C P 36.0M 277 300 24.9 24.9 100.0 5 0 0 1 1 OFDM 1 B 48.0M 278 236 30.4 30.4 100.0 5 0 0 28 28 OFDM 1 A 54.0M 279 212 33.0 33.0 100.0 5 1 1 354 359 Total packet count:: ideal 375 lookaround 10 Average # of aggregated frames per A-MPDU: 1.0802.11n MCSs are not attempted, only 802.11g speeds are attempted.
It seems to me that my 802.11n device (RB-LHG-HP5nd) thinks that my RB-SXTsq-5ac device is an 802.11g device.
Why else would only be testing and linking at 802.11g MCSs?
Also note that the LQ and NLQ are all 100%, so there is no multiplication with values < 1.0 happening.
?
73, Chuck
Interesting! Have you tried reloading the software?
73
Orv
"Have you tried reloading the software?"
Yes.
TxMbps 54: Multiple firmware versions => 3.23.4.0, 202308??
I tried to load iperfspeed_0.8, but is seems to be failing to load on the Nightly.
Now, even my GL iNet AR300M16 is doing 54 Mbps with the Nightly.
:-|
73, Chuck
root@NC8Q-5ac:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/64:d1:54:9c:85:af# ls
addba aid delba he_capa peer_debug_trigger tx_filtered
addba_resp airtime driver_buffered_tids ht_capa peer_ps_state vht_capa
agg_status aql dump_tid_stats last_seq_ctrl rx_duplicates
aggr_mode aqm flags num_ps_buf_frames rx_fragments
root@NC8Q-5ac:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/64:d1:54:9c:85:af# cat vht_capa
VHT not supported
root@NC8Q-5ac:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/64:d1:54:9c:85:af# cat ht_capa
ht not supported
root@NC8Q-5ac:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/64:d1:54:9c:85:af# cat he_capa
HE not supported
root@NC8Q-5ac:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/64:d1:54:9c:85:af#
I think
'ht' is 802.11n MCSs,
'vht' is 802.11ac MCSs,
'he' is 802.11ax MCSs.
My RB-LHG-5nD indicates its 'ht' is supported:
root@NC8Q-5nd-0:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26# ls
agg_status aqm he_capa node_aggr rc_stats rx_fragments
aid driver_buffered_tids ht_capa node_recv rc_stats_csv tx_filtered
airtime flags last_seq_ctrl num_ps_buf_frames rx_duplicates vht_capa
root@NC8Q-5nd-0:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26# cat vht_capa
VHT not supported#
root@NC8Q-5nd-0:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26 cat ht_capa
ht supported
cap: 0x19ef
RX LDPC
HT20/HT40
SM Power Save disabled
RX HT20 SGI
RX HT40 SGI
TX STBC
RX STBC 1-stream
Max AMSDU length: 7935 bytes
DSSS/CCK HT40
ampdu factor/density: 3/6
MCS mask: ff ff 00 00 00 00 00 00 00 00
MCS tx params: 1
root@NC8Q-5nd-0:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26# cat he_capa
HE not supported
root@NC8Q-5nd-0:/sys/kernel/debug/ieee80211/phy0/netdev:wlan0/stations/74:4d:28:92:be:26#
Is this the reason my 802.11n devices think my 802.11ac device is only capable of 802.11g MCSs?
73, Chuck
Hi, Orv:
Yes, TxMbps is is a data speed, not data throughput speed.
TxMbps indicates the link speed (MCS) less retries (if any).
iperfspeed would indicate throughput (TxMpbs)*(LQ/100)*(NLQ/100)-overhead(FEC,...)
These are half-duplex radio links so throughput can only approach MCS speed selected.
I am looking at rc_stats and wondering if my new 'ac' device broadcasts that it can do 802.11 encoding.
?
73, Chuck