1. How Can Cambium Offer Competitive Options To Mimosa Users? It seems the ePMP 3000 product is heading in an excellent direction, but there are no ways to move to it if you are using Mimosa hardware currently. There are only 2 ways to transition:
Rip and replace (Too expensive to justify)
Duplicate installation (Often too many radios for microPoP or other deployments)
The easy answer to this is:
SUPPORT WIFI MODE!!
In the WISP industry, it is common knowledge that WiFi mode is of reduced performance relative to [conceptually] all other TDMA modes. However, Mimosa managed to implement WiFi mode, and on a speed-basis, they have often outperformed ePMP for many years. There are fundamental adjustments they made to reduce hidden-node issues, and there are many systems running this mode successfully. [I would prefer an elevate option, but I understand that is not possible for the Mimosa hardware.]
2. How Can Cambium Further Ease the Transition? Just supporting WiFi mode will help tremendously, but the process could be further improved. Mimosa users running the connectorized radios have existing 4x4 or multiple 2x2 antennas in-place and already aimed. There are Mimosa specific antennas, and there are Mimosa approved antennas. Currently, it seems Cambium will not approve the use of any of these existing antenna configurations.
CAMBIUM SHOULD TEST AND APPROPRIATELY APPROVE THESE ANTENNAS!
The transition would be further helped by actually testing these antennas and making the necessary adjustments for the ePMP 3000 to behave when using them!
Cambium, please, we have several Mimosa deployments we are considering changing to ePMP. We cannot do a rip-and-replace, we cannot run dual 5 GHz systems from almost all of the locations, and we have many custom 4x4 antenna installations that cannot be readily changed. The only path forward requires some improved method to transition... or I have to continue to run Mimosa areas and ePMP areas separately.
So unfortunatly, I've talked to the dev's and there's no planned WiFi mode for e3k.... and there's never going to be an elevate build for Mimosa because they use chips from Quantenna... so your only option is rip and replace.
Your best bet is to reach out to your Cambium regional sales manager (RSM) and discribe the situation and see if you can get a pricing exception (PE) for all the gear you need to do a rip and replace. Usually a deal can be made.
So unfortunatly, I've talked to the dev's and there's no planned WiFi mode for e3k.... and there's never going to be an elevate build for Mimosa because they use chips from Quantenna... so your only option is rip and replace.
I've talked to the devs too, and recognize there is not a current plan for WiFi mode for e3k... but I'm suggesting there should be. I understand their reasoning, but even at an imaginary 50% discount, I couldn't justify the cost and downtime associated with a rip and replace strategy.
[I have also received the explanation for why they will not be able to elevate Mimosa. I think elevating would be the preferred process, but I understand the limitations they've presented.]
This thread is to fundamentally request that there will be some pursuit of a better process for swapping Mimosa. Without a better process, I likely won't be able to justify switching, and I expect others are in a similar situation.
I agree on the WiFi mode to enable forklifting over from Mimosa. Maybe they just don't think that many people would be willing to leave Mimosa for ePMP 3000 ?
That said, just curious why you are replacing Mimosa. I didn't like their stuff when it first came out and I haven't messed with it since then but I was really hoping it had matured by now and would possibly become something I might want to replace ePMP with.
Maybe the old Motorola team at Ubiquiti will roll out some nice PMP gear.
I think the biggest problem would be finding a replacment for the A5.
Did anyone test an ePMP3000 with an N5 with regards to mu-mimo?
A few comments -
1 - Elevate for Mimosa is technically a very complex undertaking due to fundamental differences in the hardware. Elevate is possible due to commonality of hardware. When that goes away it is simply not justified
2 - WiFi mode for inter-operability is certainly doable but when we did wifi mode originally for ePMP 1K/2K we noticed that interim operation in wifi mode was just a nightmare as it was WiFi. Thus we went ahead with Elevate
3 - We are working with KP to have a true Mumimo omni antenna. We are very excited about this offering from KP to work with the e3K as it will be one of its kind and actually do MUMIMO in an omni.
I agree on the WiFi mode to enable forklifting over from Mimosa. Maybe they just don't think that many people would be willing to leave Mimosa for ePMP 3000 ?
It seems to me that nobody would switch from Mimosa without some reasonable path. They would continue to run and expand Mimosa while possibly adding ePMP3k in new areas. That feels like a lost opportunity for Cambium.
...just curious why you are replacing Mimosa. I didn't like their stuff when it first came out and I haven't messed with it since then but I was really hoping it had matured by now and would possibly become something I might want to replace ePMP with. ...
We have a mix of ePMP and Mimosa. The majority of our outage related service calls are from Mimosa areas, and our primary challenges with installations are from Mimosa. Now, we do have some Mimosa setups that give us almost zero problems, but Mimosa recently decided not to do multicast at all... and we are about to run IPTV as multicast on our system.
I think the biggest problem would be finding a replacment for the A5.
Did anyone test an ePMP3000 with an N5 with regards to mu-mimo?
A few comments -
1 - Elevate for Mimosa is technically a very complex undertaking due to fundamental differences in the hardware. Elevate is possible due to commonality of hardware. When that goes away it is simply not justified
2 - WiFi mode for inter-operability is certainly doable but when we did wifi mode originally for ePMP 1K/2K we noticed that interim operation in wifi mode was just a nightmare as it was WiFi. Thus we went ahead with Elevate
3 - We are working with KP to have a true Mumimo omni antenna. We are very excited about this offering from KP to work with the e3K as it will be one of its kind and actually do MUMIMO in an omni.
1 - The justification for no elevate feature for Mimosa does make sense, and I accept that.
2 - As much as I don't like Mimosa right now, their WiFi mode works very well (I think much better than ePMP did). I don't think WiFi mode should be ignored. I would think reviewing the methods Mimosa used would reveal the necessary improvements / changes to get it to work. At a bare minimum, SMs could support WiFi mode and connect to a Mimosa AP until all SMs were changed out... then the AP could be replaced.
3 - Mimosa already has an omni antenna that does MU-MIMO called the N5-360. That is what Ryan is mentioning above. I'm sure the KP unit will be very nice, but please, please, test ePMP3000 with the Mimosa N5-360 omni and all other Mimosa approved antennas so we can just swap radios instead of having to change antennas as well.
2 - WiFi mode for inter-operability is certainly doable but when we did wifi mode originally for ePMP 1K/2K we noticed that interim operation in wifi mode was just a nightmare as it was WiFi. Thus we went ahead with Elevate
...
...
2 - As much as I don't like Mimosa right now, their WiFi mode works very well (I think much better than ePMP did). I don't think WiFi mode should be ignored. I would think reviewing the methods Mimosa used would reveal the necessary improvements / changes to get it to work. At a bare minimum, SMs could support WiFi mode and connect to a Mimosa AP until all SMs were changed out... then the AP could be replaced.
...
It occurs to me that WiFi mode for just the new ePMP 300 series SMs would be enough to facilitate transition (I was picturing a e3k AP with WiFi mode originally). As long as some manual adjustments are possible to match up with the Mimosa AP, the ePMP 300-x SMs could register with a Mimosa AP. Then, ePMP 300-x SMs could be installed in place of the Mimosa SMs, until they were all replaced and the AP could be the last piece to be changed.
.. WiFi mode for inter-operability is certainly doable but when we did wifi mode originally for ePMP 1K/2K we noticed that interim operation in wifi mode was just a nightmare as it was WiFi. ...
Sakid,
I had to purchase more Mimosa hardware this morning for an area that is mostly ePMP. If ePMP Force 300 radios had WiFi mode, I could be ordering Force 300 radios instead and finish transitioning to ePMP in this location.
I know you mention that the WiFi mode didn't work well in 1k/2k... but Mimosa managed to solve getting WiFi mode to work effectively within the standards in most situations. I suspect Cambium gear would work just as well as Mimosa in WiFi mode as long as the necessary adjustments are available.
This is what Mimosa is expecting for 3rd party devices to connect to their AP:
Don't hate me for this but I just can't see spending the effort to optimize WiFi mode to inter-operate with Mimosa. You see the threads around here with issues we are focusing on the e3K, peoples request for flexible mode and many other features. It just boils down to how much resource we have and what we can spend it on.
I will say that if there is something we can do commercially for you or anyone ready to do a complete flip of your 3rd party network to ePMP we can talk. I totally get the truck roll costs etc but we can at least have a conversation around it and see if we can come up with something. Thx Sakid
...I just can't see spending the effort to optimize WiFi mode to inter-operate with Mimosa...
To be clear, I'm talking about Mimosa because I have Mimosa. I didn't mean to suggest optimizing WiFi mode for Mimosa, just to implement WiFi mode in general for the SMs. If standards-based WiFi mode was implemented, then ePMP SMs could be attached to Mimosa (facilitating migration), MikroTik (facilitating migration), Radwin (facilitating migration) etc.
...You see the threads around here with issues we are focusing on the e3K, peoples request for flexible mode and many other features. It just boils down to how much resource we have and what we can spend it on...
I certainly understand the resource limitations. While ePMP 3000 still isn't done, I agree that it definitely makes sense to focus on that product. Perhaps when it is further along this could be entertained.
...I will say that if there is something we can do commercially for you or anyone ready to do a complete flip of your 3rd party network to ePMP we can talk. I totally get the truck roll costs etc but we can at least have a conversation around it and see if we can come up with something...
I'd be willing to have this conversation, but an all-at-once approach is probably not financially viable, especially since Mimosa PoE differences would require in-home work and coordination. Who would I contact to entertain such a thing?
Mimosa and Cambium have both been quite competitive in this space. I have one project going now with ePMP (and 450i on 900), and another just about to install Mimosa's newer generation, the A5c/N5-360 and C5x. The older A5 was a bit weak, with four SISO 1-quadrant sectors and MIMO only if you overlapped them. The newer method has four 2-quadrant sectors, so it's always MIMO, and beam steers to the right pair. And Cambium seems to be doing that too with a new antenna so they are watching each other! ;-) The old C5 was flimsy plastic junk; the new C5x is the opposite, a really nice solid metal body, and a clever design with screw-on passive antenna elements. Cambium let down some users with the Force 300 mount, which was cheaper rather than better than the Force 200. I'm sticking with Force 200s and ePMP 2000 for now -- it's in a tough area where we'd never get the higher speed anyway. But oh I miss the full-time spectrum analyzer (which I guess the ePMP 3k has too).
