Help me decide best router for home/office use (very urgent and important)

Status
Not open for further replies.
[OT] with due apologies to Futureized :P

blr_p said:
Beamforming is MIMO. You already have it in existing n-stream devices. MIMO improves reception by making better use of reflected signals. It does not actually increase range but rather decreases signal loss. The unintuitive bit with MIMO is you must have 'some' obstacles in the way so as to increase reflections to get better speed than a clear line of sight (!) MIMO is useless in an open field or less useful in a very large room.
Click to expand...
MIMO is not necessarily beamforming, just as 4 tyres on a car doesn't necessarily make it 4WD.

MIMO has a number of transmission modes:-
1. Transmit diversity is as you describe - bouncing multiple copies of the same signal around. You might be mixing transmit diversity for beamforming.

2. Beamforming adds phase to the signals making it narrow and amplified, and proper beamforming (like Ruckus does) uses antenna arrays.

3. Then there's spatial stream multiplexing. All n devices get their rated speeds from the number of streams being handled simultaneously - 1 antenna:1 stream = 150Mbps, 2A:2S = 300Mbps, 3A:3S = 450Mbps, etc.
I have no idea when exactly (or if at all) these devices switch over from stream multiplexing to sending out a single stream with diversity or beamforming. AC devices in contrast can match N bandwidth with a single stream giving manufacturers the flexibility to use other antennas for beamforming instead.

For example:- the N66U uses the Broadcom BCM4331. It clearly doesn't specify beamforming, and lists "space-time block coding" which is used in transmit diversity. While the AC66U using the Broadcom BCM4360, does list beamforming. I'm not saying n devices/chipsets don't support beamforming, there are a couple of Ralink chipsets (RT3593) which do. Beamforming was an optional feature on .11n and was mostly not implemented, which the .11ac standard seems to rectify.


I've often wondered why this isn't the case presently with existing N 5 Ghz routers. Why can't they compensate somehow so that range on both bands is similar.
Click to expand...
Forget 5GHz, even 2.4GHz devices seem to be dropping power compared to devices from before. Probably stricter FCC regulations on transmit power?

[/OT]
 
One way to settle this.

Check out the difference between the ac & non-ac performance of the dark knight in Tims's house.

Amazing how AC manages to keep performance up over distances even up to Location F (!) so yes i do stand corrected :)





For reference
Location A: AP and wireless client in same room, approximately 6 feet apart.

Location C: Client in upper level, approximately 25 feet away (direct path) from AP.
One wood floor, sheetrock ceiling, no walls between AP and Client.

Location D: Client in upper level, approximately 35 feet away (direct path) from AP.
One wood floor, one lower level sheetrock wall, sheetrock ceiling between AP and Client.

Location F: Client on upper level, approximately 65 feet away (direct path) from AP.
Four to five interior walls, one wood floor, one sheetrock ceiling between AP and Client.
 
just to add my comment on your technical stuff above.
People on other sites told me not to go for AC as this is not required and it is slow compared to RT n66u
 
Crazy_Eddy said:
Edit: bottle says 802.11ac devices seem flaky. Get the N66U then.
Click to expand...
Wonder why he said that ?

.11ac devices will appear flaky if mixed with n devices as the n forces the router to drop speed when communicating.

Basically was he referring to a mixed device environment ?

Mixed device environment is bad for throughput, better to partition into VLANs and allow the slower devices to associate with a router of their kind and the same with the faster devices.

Must not replace old tech router with new tech router, must hold onto old router if there are older devices and convert it into a wired AP to service them. Use the new router to handle the faster clients.
 
^ He was looking at both the N66U and AC66U and picked up the N66U after seeing quite a few user reviews citing .11ac devices as being flaky. By flaky, not just speed drops but connection drops as well. As you said 11ac is still in the draft stage, so will take a while for devices to get polished in terms of hardware and firmware.

In any case if you're spending a lot on a router and it can't handle a mixed device environment, its pointless isn't it.
 
Crazy_Eddy said:
^ He was looking at both the N66U and AC66U and picked up the N66U after seeing quite a few user reviews citing .11ac devices as being flaky. By flaky, not just speed drops but connection drops as well. As you said 11ac is still in the draft stage, so will take a while for devices to get polished in terms of hardware and firmware.
Click to expand...
This is a firmware issue most likely, they have to work out the bugs. You buy something that's this bleeding edge then you have to wait until they work out the kinks.

Though i think the solution in a mixed environment is to use the 5ghz with ac clients only and leave the other band for n devices and another router entirely for g and lower.

Crazy_Eddy said:
In any case if you're spending a lot on a router and it can't handle a mixed device environment, its pointless isn't it.
Click to expand...
We buy the latest protocol router to get 'blazing fast' speed so if performance matters I don't think there is any way around it but there are some qualifications from what i can understand.

This is why you do not replace an older router but re-purpose it as a wired AP to the network so the older devices can work with it and leave the newer router for the faster devices. A dual band router helps somewhat here by relegating the older b&g to the 2.4ghz and leaving the 5ghz for newer but its range is shorter. More radios is the solution. Either dual band or more than one AP.

A slower device will not affect throughput if it only associates with the router, but will if it initiates a data request and the router radio will communicate at the clients speed for the duration of its conversation. Everything else that is on the network gets reduced to the speed of the slowest link. But once the communication is over then things revert back to normal. This has to be as the frame format is different for b,g,n & ac. Newer devices are backwards compatible and can talk older protocol but at a reduced speed.

The routers out of draft seem to have more flexibility here. They can handle a mixed environment better but they will still slow down if a slower device starts talking.

Now if the slowest device interacts only for short durations and there are no other high data uninterruptible transfers going on then its not so bad. It could for all intents and purposes go unnoticed if the requirement is basic intermittent web browsing.

But if there were data transfers or movie streaming ie constant data flow going on then they will be slow downs, lagging etc. Mixed environments are a no-no for those types of activities. I would not be surprised that lots of complaints of constant disconnections are due to a mixed environment. All routers will have trouble here and pretty much attract the same kind of complaint.

Its a mess. Vendors provide mixed ability and backwards compatibility as selling points but the user had better be aware of the trade-offs. A mixed environment is more complex and the bugs harder to fix as well.

Put a b device in a g network and the speed drops from 54/108 to 11mbs for the duration b is active.
Put a g device in a n network and the speed drops from 65/150/225/300/450 to 54mbs etc
Put a 20Mhz only channel width client and there goes your 40Mhz 2.4Ghz network. Do not use apple devices in a such a network !
Put a a device in a n network and there goes your 5 ghz network from 150/300/450 to 'a' speed.

And for the kicker...
Put a b/g device in a n network and even their speed gets affected unless the router talks b or g specifically as opposed to mixed. have read comments to the effect how the new 'n' router is slower or range is shorter than the older g router. Or how the g devices do not work as fast with the new n router as they did with the older g router.

This new n router sucks. Nah, it does not, just don't mix different protocols and expect one radio to handle it without tripping up.

I have a doubt with MIMO. Does sticking non-MIMO n devices in a n network with other MIMO devices also reduce the speed to non-MIMO n ?
 
Status
Not open for further replies.
Top Bottom