and it doesn't *always* just use that one for a variety of reasons. 



----- 
Mike Hammett 
Intelligent Computing Solutions 

Midwest Internet Exchange 

The Brothers WISP 




----- Original Message -----

From: "Matt Hoppes" <[email protected]> 
To: "AnimalFarm Microwave Users Group" <[email protected]> 
Sent: Thursday, January 23, 2020 4:50:44 PM 
Subject: Re: [AFMUG] The Future 

Exactly. And then you always know the IP of Netflix traffic on your network. 

> On Jan 23, 2020, at 5:47 PM, <[email protected]> <[email protected]> wrote: 
> 
> Can't you get a netflix box from netflix? 
> 
> -----Original Message----- From: Matt Hoppes 
> Sent: Thursday, January 23, 2020 3:42 PM 
> To: AnimalFarm Microwave Users Group ; Darin Steffl 
> Subject: Re: [AFMUG] The Future 
> 
> We have a Netflix CDN... I'm literally trying to get a MikroTik 
> consultant who can write a simple queue. 
> 
> I'm finally just giving up on a consultant and just need to find the 
> time to do it myself. 
> 
> It's really not that hard... you know the IPs for your Netflix CDN... it 
> doesn't change. You need a queue that matches to the source of that CDN 
> going to each end user. 
> 
>> On 1/23/20 5:04 PM, Darin Steffl wrote: 
>> Matt, 
>> 
>> Caching is something from the 90's unless it's with a direct CDN provider 
>> like Akamai or Netflix. If you're trying to do this yourself without having 
>> an appliance or server farm from a CDN, you're dreaming. This is why you 
>> won't find a Mikrotik consultant that can do the job for you. 
>> 
>> Gone are the days of T1's and HTTP caches so get with the times. Upgrade 
>> your middle mile circuits so this isn't a concern. 
>> 
>> On Thu, Jan 23, 2020 at 3:35 PM Steve Jones <[email protected] 
>> <mailto:[email protected]>> wrote: 
>> 
>> you have to log into the netflix account and set it off auto for 
>> each sub account 
>> 
>> ours was 512k max. maybe resolution on a tv would have been 
>> pixelated we did our verifications from PC 
>> 
>> On Thu, Jan 23, 2020 at 3:25 PM Adam Moffett <[email protected] 
>> <mailto:[email protected]>> wrote: 
>> 
>> Interesting. 
>> 
>> On 1/23/2020 4:23 PM, Matt Hoppes wrote: 
>> > The cell carriers have caching servers installed so they can 
>> > manipulate the traffic. We are working on doing something 
>> similar... 
>> > but so far every MikroTik consultant I've hired has screwed 
>> me on this. 
>> > 
>> > On 1/23/20 4:21 PM, Adam Moffett wrote: 
>> >> If I recall correctly, they have lower quality levels for 
>> mobile 
>> >> users on 4G. I wonder how they know you're mobile and 
>> whether you 
>> >> can trick the system into counting your user as mobile. 
>> >> 
>> >> 
>> >> On 1/23/2020 4:17 PM, Mathew Howard wrote: 
>> >>> Yeah, last I looked that's what they said the lowest 
>> quality needed. 
>> >>> A few years back I did some testing with various speeds, 
>> and I think 
>> >>> I got down to somewhere around 500k before Netflix would 
>> break. But 
>> >>> even then, the picture quality was getting pretty ugly. 
>> >>> 
>> >>> But seriously... if Netflix defaulted to lower quality (not 
>> lowest, 
>> >>> but in the middle), and made you set it higher if you 
>> wanted, most 
>> >>> people would never know or care... and it'd save a lot of 
>> bandwidth. 
>> >>> 
>> >>> On Thu, Jan 23, 2020 at 3:14 PM Adam Moffett 
>> <[email protected] <mailto:[email protected]> 
>> >>> <mailto:[email protected] <mailto:[email protected]>>> 
>> wrote: 
>> >>> 
>> >>> I'm pretty sure the lowest quality level on Netflix 
>> needs 0.7 
>> >>> mbps. If your rule ended up giving them 256k+512k then 
>> it would 
>> >>> have worked. 
>> >>> 
>> >>> 
>> >>> On 1/23/2020 4:10 PM, Steve Jones wrote: 
>> >>>> Way back in the day, when powercode had the old type 
>> queue, we 
>> >>>> built our basic one to buffer at 512 long enough to 
>> maintain a 2 
>> >>>> hour sd stream at 256k with periodic 512k bucket 
>> refills. so 
>> >>>> really it was 512k effectively. It may very vell be that 
>> >>>> expectations of "standard" definition were different 
>> back then. 
>> >>>> but I thought that was an actual resolution standard 
>> >>>> 
>> >>>> On Thu, Jan 23, 2020 at 2:58 PM Ken Hohhof 
>> <[email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> I don’t remember ever being able to stream Netflix 
>> on 256K. 
>> >>>> 1M maybe, and 1.5M still gives you decent SD. You’re going 
>> >>>> to need at least 2.5M though for HD. So that’s 
>> one part of 
>> >>>> the answer is HD. Some streaming services, like 
>> DirecTV On 
>> >>>> Demand, don’t have adaptive video quality and want 
>> a minimum 
>> >>>> of 5M to stream. Another factor is “live” video, 
>> which is 
>> >>>> compressed on-the-fly and probably not as 
>> efficiently as 
>> >>>> pre-recorded content. 
>> >>>> 
>> >>>> Of course, if the customer has more, video streams 
>> will 
>> >>>> happily use it. 
>> >>>> 
>> >>>> *From:* AF <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> *On Behalf Of *Steve Jones 
>> >>>> *Sent:* Thursday, January 23, 2020 2:29 PM 
>> >>>> *To:* AnimalFarm Microwave Users Group 
>> <[email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>>> 
>> >>>> *Subject:* Re: [AFMUG] The Future 
>> >>>> 
>> >>>> we are at the end of the wireless backhaul road. 
>> when I 
>> >>>> started 15 or so years ago, we were just moving off a 
>> >>>> handdful of random T1s to a bonded 6mb circuit 
>> backhauling 
>> >>>> that was nothing. Now we have two gig circuits on 
>> separate 
>> >>>> parts of our network, and we are a tiny WISP in 
>> podunk USA.. 
>> >>>> We dont put less than 1.2gbps backhauls in for 
>> core backhauls 
>> >>>> now. The existing technology for distance in a 
>> single unit us 
>> >>>> roughly 2gbps when trying to cover any distance of 
>> merit. 
>> >>>> Sure you can do more than that, you can cheat 
>> outside link 
>> >>>> budgets and ignore your rain region. But if youre 
>> talking 
>> >>>> about most temperate region backhauls with legitimate 
>> >>>> reliability thats the wall. 
>> >>>> 
>> >>>> we keep poking a little more bits/hz out, but that 
>> not really 
>> >>>> new tech, its all dependent upon smaller and 
>> smaller path 
>> >>>> budgets, that eventually wont be attainable. so 
>> you have to 
>> >>>> start doing shorter shots, with more radios, more 
>> channel 
>> >>>> size, etc. eventually you hit the point where its 
>> no longer 
>> >>>> economically viable to keep throwing radio and 
>> lease costs at 
>> >>>> it and youll have to put glass in the dirt. 
>> >>>> 
>> >>>> Duct is whats future proof, fiber is just the 
>> current best 
>> >>>> long term option for transport. pending some 
>> breakthrough 
>> >>>> tech, its the only real long term cost effective 
>> future 
>> >>>> proofish option. 
>> >>>> 
>> >>>> We will hit a wall on demand at some point in the 
>> near term 
>> >>>> as we run out of things to connect. 
>> >>>> 
>> >>>> Can anybody answer why 256k used to be able to 
>> deliver a 
>> >>>> decent SD netflix stream and now i need multiple 
>> mbps for the 
>> >>>> same thing? asking for a friend 
>> >>>> 
>> >>>> On Thu, Jan 23, 2020 at 1:40 PM Carl Peterson 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> "Elon started it as a project to raise money, 
>> yes. Morgan 
>> >>>> Stanley is up valuing it because they don't 
>> understand 
>> >>>> technology. This project is not even close to 
>> spacex's 
>> >>>> purpose for existing. If it disappeared it 
>> would not have 
>> >>>> any real effect on their overall mission." 
>> >>>> 
>> >>>> This isn't really true. There was one primary 
>> driver. 
>> >>>> 
>> >>>> 1) You need to bring down the cost of launch 
>> considerably 
>> >>>> in order to expand the launch market to a size 
>> where 
>> >>>> developing and maintaining a reusable rocket 
>> fleet makes 
>> >>>> sense but you can't bring down the cost of 
>> launch till 
>> >>>> you have customers to fill the launch 
>> manifest and that 
>> >>>> spool up will take years. SpaceX thinks they 
>> have solved 
>> >>>> this by becoming their own customer for all 
>> their extra 
>> >>>> launch capacity for the foreseeable future. 
>> >>>> 
>> >>>> When they looked at #1 above they 
>> realized that there was 
>> >>>> a huge potential market there and even a a few 
>> % of the 
>> >>>> global internet market could be a cash cow for 
>> years to 
>> >>>> come. 
>> >>>> 
>> >>>> On Tue, Jan 21, 2020 at 9:13 PM Jason McKemie 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> Elon started it as a project to raise 
>> money, yes. 
>> >>>> Morgan Stanley is up valuing it because 
>> they don't 
>> >>>> understand technology. This project is not 
>> even close 
>> >>>> to spacex's purpose for existing. If it 
>> disappeared 
>> >>>> it would not have any real effect on their 
>> overall 
>> >>>> mission. 
>> >>>> 
>> >>>> On Tuesday, January 21, 2020, Robert 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> um, no, Starlink is now becoming the 
>> primary 
>> >>>> reason for the huge run-up in 
>> valuation for 
>> >>>> SpaceX... 
>> >>>> 
>> >>>> 
>> >>>> 
>> https://www.businessinsider.com/spacex-future-multibillion-dollar-valuation-starlink-internet-morgan-stanley-2019-9
>>  
>> >>>> 
>> >>>> On 1/21/20 4:15 PM, Jason McKemie wrote: 
>> >>>> 
>> >>>> The difference being that this is 
>> a side 
>> >>>> project for one of the main 
>> businesses, not 
>> >>>> their primary purpose. At best I 
>> don't think 
>> >>>> this is going to be anything 
>> besides a better 
>> >>>> alternative to other satellite 
>> internet 
>> >>>> options. 
>> >>>> 
>> >>>> On Tuesday, January 21, 2020, 
>> Darin Steffl 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> Guys, lots of misinformation here. 
>> >>>> 
>> >>>> They are NO plans nor hints of 
>> >>>> integrating Starlink antennas 
>> into Tesla 
>> >>>> cars. It may happen but no one 
>> has hinted 
>> >>>> of this happening. All Tesla's 
>> have 3G or 
>> >>>> 4G modems already built-in to 
>> them along 
>> >>>> with WiFi. Updates are sent 
>> via WiFi 
>> >>>> first and after the fleet has 
>> received 
>> >>>> the updates, they eventually 
>> push it to 
>> >>>> cars via cellular data that 
>> haven't 
>> >>>> updated via WiFi. 
>> >>>> 
>> >>>> Regarding B2B backhaul, I 
>> don't believe 
>> >>>> you'll see this as an option 
>> anytime soon 
>> >>>> for WISP's or other ISP's. They're 
>> >>>> targeting residential and small 
>> >>>> businesses as well as 
>> >>>> government contracts. The cost 
>> if they 
>> >>>> did offer B2B backhaul 
>> services would 
>> >>>> likely be higher than fiber to 
>> your 
>> >>>> network. Please stop thinking 
>> this will 
>> >>>> happen as I bet it will not. 
>> >>>> 
>> >>>> They may offer a self install 
>> option but 
>> >>>> they'll also have a contractor 
>> to perform 
>> >>>> most installs for a cost is my 
>> guess. 
>> >>>> Maybe they'll send a self 
>> install kit for 
>> >>>> X price and if you can't get 
>> it working, 
>> >>>> they'll schedule a contract 
>> install for 
>> >>>> XX price. 
>> >>>> 
>> >>>> I'll also say that you should 
>> not doubt 
>> >>>> Elon's passion to achieve 
>> great things. I 
>> >>>> have a Tesla and it's a work 
>> of art and 
>> >>>> by far the best vehicle I've 
>> ever driven. 
>> >>>> 99% of people who have driven 
>> one also 
>> >>>> think this. Tesla is 
>> succeeding, SpaceX 
>> >>>> is on it's way there, The 
>> Boring Company 
>> >>>> is half done with their Vegas 
>> tunnel, and 
>> >>>> Starlink will likely be a viable 
>> >>>> competitor for us. 
>> >>>> 
>> >>>> On Tue, Jan 21, 2020 at 4:48 
>> PM Ryan Ray 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> Can you link that? What 
>> exactly were 
>> >>>> they testing? 
>> >>>> 
>> >>>> On Tue, Jan 21, 2020 at 
>> 2:36 PM 
>> >>>> Robert Andrews 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>> wrote: 
>> >>>> 
>> >>>> Somehow they passed a 
>> first 
>> >>>> review from US DOD... Can't be 
>> >>>> all smoke 
>> >>>> and mirrors in space... 
>> >>>> 
>> >>>> On 01/21/2020 12:18 
>> PM, Ryan Ray 
>> >>>> wrote: 
>> >>>> > I'm still very wary 
>> of this. 
>> >>>> There seems to be a lot of 
>> >>>> over-promising 
>> >>>> > under delivering. In 
>> typical 
>> >>>> Elon fashion, no 
>> details but the 
>> >>>> world runs 
>> >>>> > with it and puts out 
>> all these 
>> >>>> data models that make 
>> it seem 
>> >>>> like the 
>> >>>> > second coming of christ. 
>> >>>> Customer CPE is a 
>> pizza box ufo 
>> >>>> <$200 and they 
>> >>>> > are starting in 
>> 2020, but 
>> >>>> there's no pictures or 
>> details. 
>> >>>> How is that 
>> >>>> > even possible? We're 
>> buying 
>> >>>> 450b at a more 
>> expensive cost and 
>> >>>> there 
>> >>>> > ain't no phased 
>> antenna with 
>> >>>> motors in it. 
>> >>>> > 
>> >>>> > Then all you read 
>> online is the 
>> >>>> cult following of 
>> spaceslax who 
>> >>>> takes a 
>> >>>> > twitter post as 
>> gospel and just 
>> >>>> keeps perpetuating the 
>> same tired 
>> >>>> > information. 
>> >>>> > 
>> >>>> > 
>> >>>> > 
>> >>>> > On Mon, Jan 20, 2020 
>> at 10:02 
>> >>>> AM Bill Prince 
>> >>>> <[email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> > 
>> <mailto:[email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] 
>> <mailto:[email protected]>>>> wrote: 
>> >>>> > 
>> >>>> > If the SpaceX 
>> Starlink 
>> >>>> system works at 50% of 
>> what it's 
>> >>>> hyped, it will 
>> >>>> > become the 
>> future of rural 
>> >>>> internet. Urban is 
>> still going 
>> >>>> to be 
>> >>>> > dominated 
>> (eventually) by 
>> >>>> fiber for the 
>> foreseeable future. 
>> >>>> Higher 
>> >>>> > speed 
>> >>>> > wireless will be 
>> very, very 
>> >>>> local. 
>> >>>> > 
>> >>>> > 
>> >>>> > bp 
>> >>>> > 
>> <part15sbs{at}gmail{dot}com> 
>> >>>> > 
>> >>>> > On 1/19/2020 
>> 6:29 PM, Matt 
>> >>>> Hoppes wrote: 
>> >>>> > > I don’t know 
>> why, but 
>> >>>> this evening got me 
>> thinking about 
>> >>>> > broadband 
>> delivery over the 
>> >>>> past 30 years and the 
>> future of 
>> >>>> broadband. 
>> >>>> > > 
>> >>>> > > First we had 
>> nothing, 
>> >>>> then along came 
>> dial-up and 
>> >>>> that was 
>> >>>> > amazing and many 
>> companies 
>> >>>> sprung up offering the 
>> service. 
>> >>>> Giants 
>> >>>> > like AOL and 
>> Prodigy. 
>> >>>> > > 
>> >>>> > > Then DSL and 
>> Cable came 
>> >>>> along as well as 
>> wireless and 
>> >>>> dial-up has 
>> >>>> > all but died. 
>> >>>> > > 
>> >>>> > > Now DSL is 
>> basically 
>> >>>> dead, cable and 
>> wireless have 
>> >>>> gone through 
>> >>>> > several 
>> iterations and we 
>> >>>> are seeing a push to 
>> fiber. 
>> >>>> > > 
>> >>>> > > What’s the 
>> possibility 
>> >>>> in the next 10 years 
>> cable and 
>> >>>> wireless 
>> >>>> > will be dead 
>> technologies 
>> >>>> with fiber at the fore 
>> front? 
>> >>>> Possibly. 
>> >>>> > > 
>> >>>> > > But then..... 
>> is fiber 
>> >>>> really future proof? 
>> We are 
>> >>>> talking about 
>> >>>> > investing 
>> hundreds of 
>> >>>> millions into fiber 
>> >>>> infrastructure, because 
>> >>>> > it’s “the 
>> future”. But is 
>> >>>> it? 
>> >>>> > > 
>> >>>> > > So far every 
>> technology 
>> >>>> delivery mechanism to 
>> date has 
>> >>>> become 
>> >>>> > obsolete in as 
>> little as 
>> >>>> 6-10 years. 
>> >>>> > 
>> >>>> > -- 
>> >>>> > AF mailing list 
>> >>>> > [email protected] 
>> <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> <mailto:[email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>>> 
>> >>>> > 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> > 
>> >>>> > 
>> >>>> > 
>> >>>> 
>> >>>> -- 
>> >>>> AF mailing list 
>> >>>> [email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> 
>> >>>> -- 
>> >>>> AF mailing list 
>> >>>> [email protected] <mailto:[email protected]> 
>> >>>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> 
>> >>>> 
>> >>>> -- 
>> >>>> Darin Steffl 
>> >>>> 
>> >>>> Minnesota WiFi 
>> >>>> 
>> >>>> www.mnwifi.com <http://www.mnwifi.com> 
>> <http://www.mnwifi.com/> 
>> >>>> 
>> >>>> 507-634-WiFi 
>> >>>> 
>> >>>> <http://www.facebook.com/minnesotawifi> 
>> >>>> Like us on Facebook 
>> >>>> <http://www.facebook.com/minnesotawifi> 
>> >>>> 
>> >>>> -- AF mailing list 
>> >>>> [email protected] <mailto:[email protected]> 
>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> 
>> >>>> 
>> >>>> -- 
>> >>>> Carl Peterson 
>> >>>> 
>> >>>> *PORT NETWORKS* 
>> >>>> 
>> >>>> 401 E Pratt St, Ste 2553 
>> >>>> 
>> >>>> Baltimore, MD 21202 
>> >>>> 
>> >>>> (410) 637-3707 
>> >>>> 
>> >>>> -- AF mailing list 
>> >>>> [email protected] <mailto:[email protected]> 
>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> 
>> >>>> -- AF mailing list 
>> >>>> [email protected] <mailto:[email protected]> 
>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>>> 
>> >>>> 
>> >>> -- AF mailing list 
>> >>> [email protected] <mailto:[email protected]> 
>> <mailto:[email protected] <mailto:[email protected]>> 
>> >>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> >>> 
>> >>> 
>> >> 
>> 
>> -- AF mailing list 
>> [email protected] <mailto:[email protected]> 
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> 
>> -- AF mailing list 
>> [email protected] <mailto:[email protected]> 
>> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
>> 
>> 
>> 
>> -- 
>> Darin Steffl 
>> Minnesota WiFi 
>> www.mnwifi.com <http://www.mnwifi.com/> 
>> 507-634-WiFi 
>> <http://www.facebook.com/minnesotawifi> Like us on Facebook 
>> <http://www.facebook.com/minnesotawifi> 
>> 
> 
> -- 
> AF mailing list 
> [email protected] 
> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 
> 
> -- 
> AF mailing list 
> [email protected] 
> http://af.afmug.com/mailman/listinfo/af_af.afmug.com 

-- 
AF mailing list 
[email protected] 
http://af.afmug.com/mailman/listinfo/af_af.afmug.com 

-- 
AF mailing list
[email protected]
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

Reply via email to