//Tales from the Towers: Chapter 65 – Picky, Picky

Tales from the Towers: Chapter 65 – Picky, Picky

Dumping time or as I like to call it, “what the heck were the engineers drinking when they came up with this”. Then again, it should also read, “business people really need a second opinion when engineers bring something for their approval. Sometimes the chasm between what engineers think is a good idea and the ability of business people who agree with them because they are clueless and don’t want to look stupid is HUGE!!!! I’m not going to lie, most of the products we get from manufacturers today are amazing. Yea, we all get ticked off at one product or another that doesn’t seem to fit our idea of what it should be, but in general, except for a few sniglets, most of the products are pretty good and there are even more amazing ones coming.

Certain products jump the shark and can take many of us to the next tier. I meant products that change the way we deploy our models or cause a fundamental change in our operation, profits, and the ability to compete. In order of products that jumped the shark for me I would say:
1) Ubiquiti Nanostation 2/5
2) Ubiquiti M series
3) Mimosa A5-360
4) Ubiquiti AF24
5) Mikrotik Routers
6) Mimosa B5c
7) Mimosa A5c/C5c series
8) Ubiquiti Unifi Mesh
9) RF Element Horns
10) IgniteNet 60GHz

And to a smaller extent:
1) Netonix switches

There are other products that we deploy to solve one-off kinds of issues but the top ones are the key ones that made us rethink our models and deployment strategies and helped us achieve amazing results over the last 3 years. Some of these may have been on the market for years such as MIkrotik, but they were brought into our use as we tested them and figured out how to integrate them. Some newer deployments that we did, using things like Cambium 450 900MHz and Siklu, have started to make some headway into our network but they are isolated deployments for now.

The next major product I see jumping the shark is IgniteNet 60GHz 2.5Gbps PTMP systems. We have started deploying them with speeds of up to 300Mbps. As we test them further, we may sell 500Mbps residential packages. And did I mention they are 2.5Gbps. Yea, they are a little more expensive but when you can ARPU $160 per month for 300Mbps, it’s cheap, cheap cheap!

Last time I checked though, it’s 2018 and even though many of these products are still being sold and as much as we like them, we limit our use of some of them because using them is liking going back in time and not efficient. It’s like, hey we have a great product so we don’t need to develop it further. When you are the only game in town, that’s great. But ignore that product long enough and when we find something better, we are moving on.

BTW, this isn’t about a product having a technical or firmware problem. Those can be resolved and manufacturers fix them. We notify the manufacturer of the issue and in most cases, they have been extremely responsive. New products, firmware repaired, critical features needed before we can deploy, are all things we deal with but most manufacturers have pretty good support for that. This is more about things where manufacturers just made a decision that I can’t figure out and gets my panties all in a bunch.

So onto my sniglet list. I’m not complaining, well, yes I am a little, but it’s constructive complaining. A constructive criticism is like when your wife tells you that your cologne went out of style in 1968 and you smell like a wet dog so you should probably change it. I think 1968 is also when some of these dashboard interfaces were designed like Siklu and Cambium 450 and guys, yes, you should probably change it. Do that and we will consider using more of the product line. Just a suggestion, dashboards are for quick, visual assessment of the most important real-time information. They should not look like an owners manual from a 1974 Altair 8800 and require the user to have an engineering degree and MathLab to simply find out how much traffic is moving through the AP. The AF5x had a good dashboard, but its feature set and manual entries for frequency (2003 just called again), turned us off compared to devices with auto-frequency. Haven’t checked out the AF5HD yet though.

Then there is Dragonwave. I think I just threw up in my mouth a little. Although their GUI interface is slightly better, and I mean less than the thickness of Subways see-through meat slices, their operation leaves me completely baffled. Take the scenario of a simple power supply failing on one side. The other side starts randomly rebooting multiple times and until you log into it, you can’t get it to reconnect. That would be the side you can’t get to reconnect because it’s rebooting and you have to drive out there to fix it. I don’t know what Rocket Scientist Engineer thought that was a good idea, but termination wouldn’t even come close to the punishment I was envisioning when this happened to us not once, not twice, but 3 times. BTW, all of those Dragonwaves are now disconnected for other reasons, but there is no way on the planet we would ever use something like that again. And to make it worse for them, Bridgewave is now going to heavily invade their space and we are on that bandwagon so far. You snooze, you lose.

Reboots. Making even something as simple as a frequency change takes at least 60 seconds for reboot, if you use auto-channel, 20-30 seconds more, and then the dreaded DFS wait. Then you have to wait for the client radios to find the new channels and start negotiating up. Ubiquiti has done a good job of upgrading their firmware so things like password changes don’t require reboots, but if you use the term “carrier class” and you are offline 2-3 minutes for certain things, it’s more like “pigeon carrier class”. And God help you if auto-frequency finds the wrong channel! You have to lock it out, and then do this again. Ubiquiti isn’t alone in this as every product from Mimosa and Cambium’s ePMP and 450 line have this same issue. If you can put a secondary 2.4GHz Wi-Fi chip in a 5GHz outdoor radio, I would think you could figure out how to put a second 5GHz chip in the radio or possibly another method, so important changes can be made without knocking users offline. MImosa has minimized the DFS channel change thing as long as the AP does it automatically though. But if you do it manually, it’s another 60 seconds offline and yes I know the rules, but Mimosa figured out partly how to get around that.

As I write this, I realize my list could go on and on. Most of us have sniglets we deal with daily. And there are many products out there that we might use more if the original design had taken into account some of our requirements, but that’s more product design than sniglets. We need to accept that every manufacturer has something we don’t like or some bad experience we had with them in the past so we haven’t looked at them since. I never thought I’d use Cambium, but the 900MHz 450i has us partly re-evaluating that (when they upgrade the dashboard) and we have 820’s in the air we are finishing up this week. Ignitenet had some stumbles early on with 60GHz but through their efforts, we are now deploying a bunch of them and it will be very profitable. Just revisit each manufacturer periodically and maybe their sniglets are gone and they can now fit into your toolbox.

By |2018-07-23T20:54:55+00:00July 23rd, 2018|Tales From The Towers|