The General Technology Thread

[KLB]

It's been a while...



Indeed.



I have ethernet running to most places in the house.



Right now, I've got an old WRT access point (DD-WRT) in the living room. It's just good enough, but only 2.4GHz and only one channel - set to mirror the guest network on the router. The problem is, if you connect to that, it will happily let you connect to my server (assume the server likes your credentials).



I'm not an IT pro, but know my way around linux and such. I just have no idea what's available today, product-wise.

Devices connected to different SSIDs on the same network are going to be able to talk to each other.

Now we are back to how much do you want to spend?

To do what you want with the guest network you will need something that can support VLANs and has some firewall abilities.

Another option could be to create two networks with devices that both NAT traffic going to your router.

Either option is going to be more complicated than you are used to and more expensive.

 

[Mr Evilwrench]

But adding gain on only one side is only partially beneficial. To get full effect you have to raise it on BOTH sides. Yes you will get SOME increase in range. But if the other end lacks the same, its lack of radiating power still wont talk loud enough to reach the high gain receive antenna when it gets to a certain point.

To be more specific, (in generalities) if you start with a standard antenna and have range of 200 yards, and add a high gain antenna to both sides you can double the range to 400 yards because you are increasing receive gain as well as radiating efficiency. But take away one side and the range drops to 300 because although you are increasing the antenna sensitivity, its still not adequate to achieve the full distance. Its still better than no gain, but just doing one side cant get you the full benefits.

Maybe my decades of RF engineering have confused me, but if you put a passive antenna with gain over the default antenna on one end, it will be symmetrical. You'll get a transmitter that yells a little louder, but it listens a little louder, too. Yes, you'll see the most benefit if you put the gain antennas at both ends, but you have to understand, an antenna works both ways. Most important is to understand the radiation pattern of the gain antenna, and point the peak at the signal's destination. If your peak is 10dBi, it's going to be pretty pointy. From there, it works both ways. The more gain your antenna has, the sharper and narrower the peak. Some of them are a little funky looking. If you're wondering, I'm enough of an RF geek that I have a Smith chart t-shirt.

 

[Cameramonkey]

Maybe my decades of RF engineering have confused me, but if you put a passive antenna with gain over the default antenna on one end, it will be symmetrical. You'll get a transmitter that yells a little louder, but it listens a little louder, too. Yes, you'll see the most benefit if you put the gain antennas at both ends, but you have to understand, an antenna works both ways. Most important is to understand the radiation pattern of the gain antenna, and point the peak at the signal's destination. If your peak is 10dBi, it's going to be pretty pointy. From there, it works both ways. The more gain your antenna has, the sharper and narrower the peak. Some of them are a little funky looking. If you're wondering, I'm enough of an RF geek that I have a Smith chart t-shirt.

Guess my decades of real world deployment and use of these wifi devices is lying to me then and the behaviors I have seen in the reals (as the other CM would say) are just an illusion.

 

[JettaKnight]

Maybe my decades of RF engineering have confused me, but if you put a passive antenna with gain over the default antenna on one end, it will be symmetrical. You'll get a transmitter that yells a little louder, but it listens a little louder, too. Yes, you'll see the most benefit if you put the gain antennas at both ends, but you have to understand, an antenna works both ways. Most important is to understand the radiation pattern of the gain antenna, and point the peak at the signal's destination. If your peak is 10dBi, it's going to be pretty pointy. From there, it works both ways. The more gain your antenna has, the sharper and narrower the peak. Some of them are a little funky looking. If you're wondering, I'm enough of an RF geek that I have a Smith chart t-shirt.

That's my thought.


But, we both know that RF is black magic, and I was not good in those necromancy classes.

 

[maxwelhse]

Devices connected to different SSIDs on the same network are going to be able to talk to each other.

Now we are back to how much do you want to spend?

To do what you want with the guest network you will need something that can support VLANs and has some firewall abilities.

Another option could be to create two networks with devices that both NAT traffic going to your router.

Either option is going to be more complicated than you are used to and more expensive.

Isn't that the entire point of the guest network mode of the router? That the devices can't see each other and they're actually on divorced networks through a common router?

 

[Mr Evilwrench]

Guess my decades of real world deployment and use of these wifi devices is lying to me then and the behaviors I have seen in the reals (as the other CM would say) are just an illusion.

Well, as I said, you stack on a bundle of gain, the pattern gets pointy. I didn't mention things like reflection, refraction, and diffraction. That stuff gets more significant as you go up in frequency, and most of what I've done has admittedly been 900MHz and below, but all that crap goes into the final actual gain out of the antenna, as well as the radiation pattern and aim point. I would be willing to bet that if you put it out on an antenna range it'll do what I said it'll do. The signal at the 50Ω output of the amplifier just can't know how much gain it's going to see on its way out the antenna. Now, I could see where the interfering materials in one direction could be different than the materials in the other direction, but if you hang an analyzer off both antennas in each direction, it won't be too far different than equal

Just so you don't think I'm just being hardheaded, I have seen some interesting propagation. One was an uninterrupted QSO between CB radios (I won't comment on their eligibility) from California to Queensland. That's in Australia. They were playing on the grey line, but they weren't messing with their antennas. Another one I was actually involved in was at about 460MHz through an atmospheric duct. I was standing in my driveway with a handheld radio, and the traffic helicopter for WSPD in Toledo was giving me full quieting. I waited until they were done and tried to hit him, but he just complained about some noise. 20dB quieting one way, **** in the wind the other way.

 

[Cameramonkey]

Well, as I said, you stack on a bundle of gain, the pattern gets pointy. I didn't mention things like reflection, refraction, and diffraction. That stuff gets more significant as you go up in frequency, and most of what I've done has admittedly been 900MHz and below, but all that crap goes into the final actual gain out of the antenna, as well as the radiation pattern and aim point. I would be willing to bet that if you put it out on an antenna range it'll do what I said it'll do. The signal at the 50Ω output of the amplifier just can't know how much gain it's going to see on its way out the antenna. Now, I could see where the interfering materials in one direction could be different than the materials in the other direction, but if you hang an analyzer off both antennas in each direction, it won't be too far different than equal

Just so you don't think I'm just being hardheaded, I have seen some interesting propagation. One was an uninterrupted QSO between CB radios (I won't comment on their eligibility) from California to Queensland. That's in Australia. They were playing on the grey line, but they weren't messing with their antennas. Another one I was actually involved in was at about 460MHz through an atmospheric duct. I was standing in my driveway with a handheld radio, and the traffic helicopter for WSPD in Toledo was giving me full quieting. I waited until they were done and tried to hit him, but he just complained about some noise. 20dB quieting one way, **** in the wind the other way.

Yep. You are a little out of touch (no offense intended) because I'm talking about dealing with 2.4Ghz-60Ghz. A TOTALLY different animal. I dable a bit in 900mhz too. That stuff almost bends around things and is pretty cool Hard to interrupt that signal. But you get above 5Ghz and its a totally different animal due to signal absorption, etc.

Hell, you get into the 60s and you have to start worrying about short range rain fade.

 

[JettaKnight]

Yep. You are a little out of touch (no offense intended) because I'm talking about dealing with 2.4Ghz-60Ghz. A TOTALLY different animal. I dable a bit in 900mhz too. That stuff almost bends around things and is pretty cool Hard to interrupt that signal. But you get above 5Ghz and its a totally different animal due to signal absorption, etc.

Hell, you get into the 60s and you have to start worrying about short range rain fade.

Don't forget about adding in MIMO... I remember studying that in grad school... real voodoo.


Anywho, my current router is a Netgear N600.

 

[Cameramonkey]

Don't forget about adding in MIMO... I remember studying that in grad school... real voodoo.


Anywho, my current router is a Netgear N600.

Hey, if it works for you, great! Most people just need 20mb throughput or less in reality. (surfing the web, watchign youtube, etc.

Another really cool thing for PTP links is GPS sync'd full duplex. Ubiquiti has a REALLY cool 24Ghz product called AirFiber. 1GB/sec actual throughput over wireless. (1.25 actual including overhead) They have a pro unit that is faster but I dont recall that throughput off the top of my head.

Anyway, Airfiber uses GPS to sync the clocks of the two radios. Once sync'd it is able to transmit packets simultaneously so that the packets pass each other in mid air. (normally one side transmits,waits for the other side to ack, then the other side sends, etc) that sync is a game changer.

 

[KLB]

Isn't that the entire point of the guest network mode of the router? That the devices can't see each other and they're actually on divorced networks through a common router?

That would depend upon the device you have, and if you want to extend it beyond your one device. If you want to add another AP in your house that has this isolated guest network that makes it more complicated.

 

[Mr Evilwrench]

No, no, I'm not denying that things change in different bands. I tried to hint at that with paths differentiating. I've had a little 20 or so Ghz as well, and chased signals down waveguides. I've also, I'll have you know, done superheterodyne within the visible light spectrum, much fun. The thing is, the same effects exist from DC-daylight. I smiled when I realized that. The difference is the materials you use for a given effect, or conversely the different effects you get with given materials.

 

[JettaKnight]

I think I'm going with a Ubiquiti, or similar, solution a sensible wired router, with one or more access points.


Winter's coming, so it's a good time to run a new cable to the attic for a ceiling WAP.


EDIT: My coworker recommended Netgate's SG-1100, which apparently is more open - just tack on the software you want to make it want you want.

 

[JettaKnight]

OK, I got a Ubiquiti AC-LR access point, an Edgerouter-X-SFP and a 16 port TP-Link switch.

So much better. The one access point is mounted at the top of a cathedral ceiling in the middle of the house and reaches everywhere with good signal strength.
It's not easy to setup, but very powerful, and relatively inexpressive.


Plus, it doesn't look like a Klingon Bird of Prey.

 

[IndyBeerman]

OK, I got a Ubiquiti AC-LR access point, an Edgerouter-X-SFP and a 16 port TP-Link switch.

So much better. The one access point is mounted at the top of a cathedral ceiling in the middle of the house and reaches everywhere with good signal strength.
It's not easy to setup, but very powerful, and relatively inexpressive.




Plus, it doesn't look like a Klingon Bird of Prey.

On your TP LInk Switch, did you get a Un-managed one?

I have a 32 port TP-Link Un-managed switch, love it. I upgraded to the 32 port when I started getting close to running out of ports on my 16, It was a few $$$'s more and my thought was not to run out because, I'm always adding stuff.

 

[JettaKnight]

On your TP LInk Switch, did you get a Un-managed one?

I have a 32 port TP-Link Un-managed switch, love it. I upgraded to the 32 port when I started getting close to running out of ports on my 16, It was a few $$$'s more and my thought was not to run out because, I'm always adding stuff.

Yeah, the $50 unmanaged one.


The AP physically plugs into the router, so think I can do VPN stuff there, should the need arise. And there's three spare ports on that router.

I might set up one such that when I plug in my laptop, it connects to the work network automatically; no more VPN client on the laptop.

 

[Cameramonkey]

Unifi products are awesome, and a reasonable price. And the visibility rocks.

For those less interested in the complexity, Amplifi is a great product too. Same guts, but with an integrated controller.

https://amplifi.com/

 

[Phase2]

There is a widespread outage on YouTube. No estimate on when it will be fixed at this point. You can follow any status updates on their Twitter account here: https://twitter.com/TeamYouTube

 

[NKBJ]

Yep, youtube's got some MIA's don't they.

 

[jamil]

It seems to be working now.

 

[jedi]

There is a widespread outage on YouTube. No estimate on when it will be fixed at this point. You can follow any status updates on their Twitter account here: https://twitter.com/TeamYouTube

Good! Down with YouTube and their censorship!!!