RF will always require decent SNR, bandwidth and a path. No matter how advanced the radios get, they will be up against Nyquest and Shannon and Maxwell.
And like many other industries, the bandwidth requirements and new products available to the customers will not stagnate either. I see that RF is continuing toward the asymptote that terminates in various versions of LTE. Look at the bits per hz evolution and LTE is kinda the end of the practical road. In the meantime, cellular carriers are dropping microwave backhaul like it has ebola and connecting towers with fiber. I don’t doubt that RF will eventually allow cost effective CPE to get a reliable 100 Mbps to every house if the APs are close, interference is down and the APs are lightly loaded. And people will pay good money for that for years to come. In the meantime, I am rolling out GigE over fiber to the customer and a nearby competitor, Veracity, is advertising 10 Gpbs. And we have 4K TVs on the market with higher resolutions in the pipe. Look at the number of people on this list that are doing fiber. Be fun to graph that over time. Anyone remember the mile long threads where it was me VS ---- ----- on this subject? 10 years ago people thought I was telling tall tales. And ---- was sure that we would all die if we didn’t do usage based billing. From: Josh Reynolds via Af Sent: Sunday, October 19, 2014 3:28 PM To: [email protected] Subject: Re: [AFMUG] Pmp450 vs epmp pros vs cons Need to get something out... Our industry, like many others, doesn't stagnate for long. Right now we're on the edge of a new generation of radio tech, with bitlomat (cross-vendor software with performance gains, gps sync, 802.11ac but down to the PHY layer) ubiquiti (airprism, tdma offloading, *possible* gps sync), and mimosa (too much to list) leading that charge. At least two of those vendors have had working beamforming for several years now (both TX and antenna based). These three are currently working on things that are outpacing MikroTik, the ePMP product, and others by anywhere from 1/2 gen to 1 full hardware gen beyond everybody else. I'd guess we're about 1 year off to being able to provide reliable 100Mbps to the home, with 250Mbps (or more) roughly 2-3 years behind that in PtMP, depending on various factors. That said, this is JUST the 802.11ac based stuff. There is at least one vendor up there on that list who will be manufacturing their own radios soon as well. Vendors will continue to come out with new tech every ~ 3 years. Their tech may be light years ahead of other vendors, and it's very likely they won't play well together in terms of general base tech or GPS sync. Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com On 10/19/2014 02:11 PM, Rory Conaway via Af wrote: Yea, I covered that in one of my articles. I just didn’t see everyone sitting around the campfire singing Kumbya. Another reason I don’t worry about GPS. My next article covers my main reason. Rory From: Af [mailto:[email protected]] On Behalf Of Josh Reynolds via Af Sent: Sunday, October 19, 2014 1:56 PM To: [email protected] Subject: Re: [AFMUG] Pmp450 vs epmp pros vs cons LOL :) Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com On 10/19/2014 08:13 AM, Rory Conaway via Af wrote: I’m assuming all 12 WISPs cooperate with each other? Rory From: Af [mailto:[email protected]] On Behalf Of Mike Hammett via Af Sent: Sunday, October 19, 2014 5:33 AM To: [email protected] Subject: Re: [AFMUG] Pmp450 vs epmp pros vs cons Entirely not true spoken by a WISP that has up until this point used Mikrotik and Ubiquiti in rural and suburban markets with 12 WISP competitors. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com ---------------------------------------------------------------------------- From: "Mark Radabaugh via Af" <[email protected]> To: [email protected] Sent: Saturday, October 18, 2014 3:52:03 PM Subject: Re: [AFMUG] Pmp450 vs epmp pros vs cons And now your completely out of spectrum and can't deploy anything new. I suppose the good part for you is nobody else can do anything given the amount of noise your making. Mark On 10/18/14, 1:27 PM, Josh Reynolds via Af wrote: You just hit the nail on the head why we have never considered deploying 450 (and similar) in the past: By the time "you" (relative term) have the cashflow to pay for those sectors, "we" (another relative term, for people deploying UBNT or similar) have already thrown up 4-6 shielded sectors and at least 10 clients per. If we don't think we can hit a decent sub density or at least make the site a valuable repeater, then we don't go there. Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com On 10/18/2014 09:01 AM, Kurt Fankhauser via Af wrote: I prefer sectors too but math doesnt always work out. I'll put the omni in to get the site up and once the customers are there change it to sectors. The 450 platform is very easy to drop sectors in and have the existing clients link right up. I have a couple sites with existing customers i am dropping a two sector 450 system in with 120 segree KP antennas. cant afford any more sectors than that per site right now... Sent from my iPhone Kurt Fankhauser Wavelinc Communications P.O. Box 126 Bucyrus, OH 44820 http://www.wavelinc.com tel. 419-562-6405 fax. 419-617-0110 On Oct 18, 2014, at 11:21 AM, Mike Hammett via Af <[email protected]> wrote: I've noticed a lot of PMP operators are deploying omnis (presumably because they can't afford 4 APs. Give me TDMA Atheros with sectors over omnis on anything any day. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com ---------------------------------------------------------------------- From: "Kurt Fankhauser via Af" <[email protected]> To: [email protected] Sent: Saturday, October 18, 2014 8:38:14 AM Subject: Re: [AFMUG] Pmp450 vs epmp pros vs cons TJ, No difference between the 3 different frequencies bands (other than NLOS range) as far as the product itself they are all the same animal. 2.4ghz NLOS is slightly better than 3.65ghhz. They all function the same and have the same expected throughputs per channel width. They all use the same firmware and i love the interface being the same across all 3. The only major difference is the 5ghz is V/H versus slant on the other two. That just translates to the 5ghz omni being ALOT smaller and lighter. There are some places that i wish the 2.4ghz woulda been V/H because of the omni size but overall I am still very happy with the 2.4ghz 450. Kurt Fankhauser Wavelinc Communications P.O. Box 126 Bucyrus, OH 44820 http://www.wavelinc.com tel. 419-562-6405 fax. 419-617-0110 On Sat, Oct 18, 2014 at 4:57 AM, TJ Trout via Af <[email protected]> wrote: Kurt, Any pros and cons on 450 between 2ghz, 3.65 and 5? Any differences at all? Range vs throughput? Obviously 2ghz penetrates better, 3 is licensed and 5 has more spectrum but anything else? All bands are open for me Thanks On Fri, Oct 17, 2014 at 11:20 PM, Kurt Fankhauser via Af <[email protected]> wrote: I started the spring deploying 450 in 2.4ghz, 5ghz, and 3.65ghz and then middle of the summer deciding i had to"try" some ePMP because the cost was so low I couldn't resist.... I can say now that I am fairly certain I will probably stick with the 450. There are many small reasons that when I considered them all i came to this conclusion. Here are my reasons: 1. ePMP latency starts to go up quickly once you have more than 10 clients on an AP. Once you get over 20 clients the latency is pretty much 25-30 ms. Cambium was honest about this at the road tour and they noted if you want the best latency to stick with the 450. 2. Sync between the two platforms is not there yet. If you have adjacent towers on the different platforms that can see each other you won't have sync. 3. No remote spectrum analyzer for clients. This is HUGE for when the clients fire up their wireless camera and baby monitors and trash the whole spectrum. 4.No burst bucket on CPE's 5.EPMP Interface is SLOWWW. Cambium explained at the tour they were offloading alot of processing power to the PC you are viewing the interface with and i can't be taking a quad core machine up a tower to work on these radios and do site surveys. I am working with a Panasonic Toughbook and takes FOREVER to log into the EPMP radios. 6. Fore some reason site surveys are a PITA with ePMP. Think its a combination of many factors here... slow interface one of them... 7. EPMP in 5ghz DFS band has really low power output. Something like 13-14db. When using an omni antenna you can't get maximum legal EIRP out of the ePMP. 8. 450 link tests and SM modulation is pretty stable and predictable. EPMP seems like its all over the place. I don't think I have yet seen EPMP linktest get full up or down outside of a lab environment. There might be other reasons but I'm pretty tired and was heading for bed. Kurt Fankhauser Wavelinc Communications P.O. Box 126 Bucyrus, OH 44820 http://www.wavelinc.com tel. 419-562-6405 fax. 419-617-0110 On Fri, Oct 17, 2014 at 5:05 PM, TJ Trout via Af <[email protected]> wrote: I haven't been keeping real up to date on current generation ptmp offerings but we have a new site going up and I need to decide pretty quickly on some equipment. For the guys who have been using both 450 and epmp do you have any pros and cons ? Any reason to spend the extra money when epmp seems to have the same if not better performance , sync, etc? My gut says 450 is going to be my best long term solution but with all of the positive epmp feedback it's hard to justify the extra money? -- Mark Radabaugh Amplex [email protected] 419.837.5015 x 1021
