I have a clean 0.6km link I am using for testing a 4625L with a RFE Horn and and 4625 and am noticing my download speeds are very similar on 80mhz and 160mhz.
I am running TDD mode 75/25, very clean and connection, no interference, MCS is DS11 in both directions at all times downlink is (-46 RSSI, SNR of 50) and I am the only client.
I get 500-600Mbps download in 160Mhz channel size and have actually seen just over 600Mbps with a 80Mhz channel size.
Not interested in running WLR mode but that hits 800Mbps+ which is maxing out that particular towers upload limits.
There have been some comments about 160Mhz not being ready for use with TDD yet… If that is the case, How long before I can see closer to 940Mbps…??
I do hope Cambium is still working on optimizing 80mhz and 160mhz channel sizes for TDD PtMP. On 40mhz I can easily get almost 400Mbps down, but at 80mhz I usually only see around 600Mbps to a single SM in PtMP TDD 75/25 mode. So double the channel size for only 50% more bandwidth. I know a wider channel will get more noise and will not translate to a 100% increase in performance, but I guess I would expect at least an 70-80% increase over 40mhz.
It could be an issue with the 4625 SM’s processing power as the AP seems to get closer to Cambium’s estimates when doing multi-SM link tests even without MU-MIMO.
This is well below Cambium’s own estimates via the ePMP Capacity Planning Tool which shows 80mhz should be near 800Mbps or more.
Well I suppose the 40/80mhz performance is Good/acceptable, but the 160Mhz (Which is why I’m buying) should be advertised as available in a future release.
When we run F4625’s in 80MHz ePTP mode (for point to point), we can usually get around 800mbps out of them in low latency mode… so they’d probably do closer to 1gbps (maximum of their ethernet interface) in high capacity mode.
This is PtMP configuration so lowering power is not an option in the real world, however just for fun a dropped the link to 15db power from 20, now have mcs9 instead of 11 and see the same roughly 500-600mbps downstream (so no difference.)
Not so sure this downstream issue with solid mcs11 is an antenna issue…Don’t forget WLR mode is hitting that 800mbps+ (tower limitation)
I have seen and mentioned this before. When running TDD as you increase channel size the individual SMs do not see a linear bandwidth increase. Meaning at 40mhz you can get close to 88% of the estimated bandwidth (ePMP Capacity Planning Tool) on DL and 91% on UL. Moving to 80mhz and I usually see at best 73% DL but still greater than 92% UL.
Now from the AP side if I run link tests to multiple 4625SMs I can get the AP DL to put out numbers much closer to the ones Cambium lists on the Planning tool. My guess is that the TDD overhead on the SMs limit individual performance at higher channel sizes, but the AP has the processing power to push more as seen when doing Link Tests to multiple SMs at the same time.
Is this a limitation of the processor on the 4625/4525 SM being not able to really take full advantage of the larger channels above 40mhz? Meaning that the larger channel sizes will mainly show their benefit in more capacity for the AP and not necessarily in higher speeds for the SM?
It seems that might be the case as I did a Dual SM 20s link test using two CPEs with similar stats (MCS11 DL/MCS9UL) and I got a combined total of 777DL and 139UL. Using the planning tool with 2 SM it showed I should get a total AP capacity of 765DL and 151UL. That means when doing dual SMs I was getting 101% of max DL and 92% of max UL.
This is also interesting. This performance ‘increase’ only seems to affect single SM link tests. If I do a dual SM link the AP gets more performance at MCS11 than it does at MCS9 but single SM link tests are better at MCS9 vs MCS11.
You’re oversimplifying things in a way that leads to bad conclusions.
The TwistPort is a feed, not “the antenna.” The horn or dish is what defines the radiation pattern, gain, and polarization behavior. Treating the adapter as the antenna and everything else as just a beam shaper isn’t how antenna systems work.
Using a 5 GHz feed on a wideband horn limiting you to 5 GHz isn’t misrepresentation — that’s how every reflector antenna works. The system bandwidth is always set by the narrowest component, and that’s basic RF, not a vendor trick.
Saying RF Elements has the “worst port isolation in the industry” without data doesn’t make it true. Plenty of vendors don’t publish isolation numbers, and lack of a datasheet value doesn’t imply catastrophic performance.
And poor isolation doesn’t fool radios into picking invalid MCS rates. Modern radios adapt based on actual decode performance — retries, EVM, PER. If isolation were that bad, throughput would fall on its face. It wouldn’t magically look better than it is.
In the real world, bad alignment and polarization skew cause far more MIMO pain than small feed isolation differences ever do.
So I ran some tests and noticed the CPU usage on the 4625 was very high during these bandwidth tests so I ordered a 4600C for testing…
While both at DS11 in an 80mhz channel the 4600c connected to the same 4600L as the 4625 saw around 830mbps max, where as the 4625 maxed out at about 650-700Mbps with UDP traffic.
The 4600C hit just under 900Mbps on a 160mhz channel but I don’t have CPU usage data yet for it so not sure if its CPU is maxed or if its some sort of efficiency issue. Either way 160Mhz channels don’t seem to useful at this time in PTMP…
Hopefully the 4616 is available soon to test with as the above results are not particularly exciting based on the radio price points…
Thank you for all your input over the last couple weeks. I am going to be deploying 5 new towers in the next 3 months and upgrading a number of existing. Do you have any interest in sending me a demo unit or two? If your products work as well as you describe, I would be happy to give you my business and provide my findings here to give everyone some good insight.
