AC Routers a veiew.

[ulaganath]

Hi Guys lets start a fresh look at ac routers reviewed in this forum when launched and does the con's are really fixed over the period and keep answering the below question.
1.what makes the best wireless router in ac series .
2.Does MU MIMO is far better than SU MIMO or just a gimick .
3.How far the Gen1 AC routers can handle and be able to support.
4.Does the Wave 2 is really a next generation.
5.Do we really need Quad stream technology. Can there be usb/PCI e client released for that in coming years.

Do post your current model and your thoughts over the period of usage till now. Are you satisfied with current or will to upgrade so as the current one cant handle today's needs.

 

[L&LD]

The best AC router is the AC1900 class with the RT-AC68U with RMerlin software being the best in my experience, all things considered.

MU-MIMO at this point is just a marketing ploy - there are no clients to play with. But it will be important in the future.

Gen1 AC routers like the RT-AC66U are way past their prime vs. what is possible with the latest AC1900 and AC3200 class routers available today. And no, they will never support anything more current because of their low performance chipsets.

Wave2 AC was supposed to be the new generation, but it didn't happen. What did they miss? 10GBe ports, greater than 4 antennae / 4 streams, MU-MIMO enabled hardware and software along with the required clients to take advantage of all this networking power.

Wave2 AC has come and gone. Wave3 AC routers and client devices should attempt to provide what we were promised with Wave2, but I'm not holding my breath for that until I'm holding and using a Wave3 router myself.

I don't need clients to support 3 or 4 streams, though that would be nice. The router needs to have as many antennae / streams as possible (8 is what I'm holding out for) to fully exploit the benefits that MU-MIMO is designed to offer.

I am personally using an RT-N66U still and the performance while adequate is not what I want to be stuck at for another year. But that is not up to me; that is up to the manufacturers offering a solid device at any price.

Today, I can upgrade to an RT-AC56U (and have done this for many customers with routers less capable than my RT-N66U) for about $80 when the sales hit. I don't do it though because that is still money better spent (if the product is worth it, of course) on the routers we'll be seeing in the next few months. Even if the new products are 5 times the cost, depending on the performance and features they offer, it will be worth it for me to buy at that time. (Because I will buy again sometime in 2020 again).

Keep in mind that most routers give good enough speeds for most people because they have no way to compare to something better or worse.

What I'm looking for is enough (wired and wireless) performance for my personal needs so that I don't need to be testing every new model that comes out. When I can increase the performance of my current solution past a certain level (let's say 20 to 30%), while keeping the stability I've come to expect from my current and past networking equipment; that is when I'll be sold on a new model.

Until then, marketing fluff, classifications and implementations will be all taken with a grain of salt. Just like with everything else that is offered for sale.

 

[htismaqe]

What I'm looking for is enough (wired and wireless) performance for my personal needs so that I don't need to be testing every new model that comes out. When I can increase the performance of my current solution past a certain level (let's say 20 to 30%), while keeping the stability I've come to expect from my current and past networking equipment; that is when I'll be sold on a new model.

I had to pull this quote out of the entire post because I love it.

Outside of transition periods where I was looking to upgrade and maybe tried a few models of router before settling on one, I have really only owned 4 routers in over 10 years.

Linksys WRT54G gave me about 4 years. Got another 4 or so out of a Zyxel P-330W. Replaced that with a Netgear WNDR3700 and it too last about 4 years. I'm into year 2 of the Linksys WRT1900AC and don't anticipate upgrading anytime soon.

 

[L&LD]

We have roughly the same experience (ownership) in the same amount of time too.

What the box says in bold letters with bright stars and upwards graphs doesn't sway me in the least. Even professional reviews are something to be verified by me in my environment and with my equipment.

Buy the product, bring it home and compare (objectively) to what I had running previously. In five minutes I know if I'll keep testing. By the end of the return period, I'll know if I'll be happy vs. what I had, performance wise, of course.

Each time I upgraded, the performance has been much more than the 20 or 30% I indicated in my post above (for at least one or more aspects of my network usage). And today it is a far cry from the WRT54G standard that I replaced a few times over the first few years because anything else was effectively a downgrade in one way or another.

When I upgraded my WNDR3700 to the RT-N66U I could not believe how much the Netgear router was holding my network performance back (both wired and wireless). I am expecting to see the same jump (almost there with the RT-AC68U or the RT-AC56U already) when I buy my next router too.

 

[ulaganath]

Cool man that's Right only we decide what's best and right for us. I had owned 3 adsl modem and it's no longer in use. So far 3 netgear, 2 linksys and 1 tplink router all n series and latest 1 is netgear ac router R7000.

Most router were absolete except my centria. As the only thing that pushed me to upgrade was frequent wan disconnect. It was bad that some time it happened every hour and sometimes every day. I tried downgrade some settings but it's doesn't fix the problem. Though this was designed for Nas like performance it also holds 1ghz dual-core cpu as it performance is best in router attached storage with 90MB READ and 33 MB write. With USB 3.0 60MB read and 45MB WRITE. RELIABLE OVER LAN wireless is out of hands though.

My recent switch to AC ROUTER I was wondering what if even that wan disconnect happens so that I might have mistaken my centria but to my surprise it's not. It's been awhile say 4+ days up and running with no reboot at all. So definitely centria could not handle it. Though I was surprised say my isp speed is merely 4Mb/s both up and down. So not sure even very much less speed it can't handle for 1 day.

Now I am happy with new ac router.

Keep sharing yours the reason to switch and does that really solves the purpose.

Sent from my ASUS_Z00AD using Tapatalk

 

[L&LD]

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance

http://www.snbforums.com/threads/does-an-ac-router-improve-n-device-performance.19533/


The first link to the article above and the discussion of that article is mostly what you're asking for (I think).

 

[ulaganath]

Let discuss here. A fresh start to ac

Sent from my ASUS_Z00AD using Tapatalk

 

[ulaganath]

I was just wondering what is actual throughput for LAN - LAN and LAN to WAN, LAN to Wireless.

1.As per the theoretical speed of actual in realy time MB/s rather Mb/s. with a 1020Mb connections we should be getting a close of 128MB/s for Lan to Lan transfer if the devices are Category A high quality switch/router connected with cat 6 cable.

2. When translating the same of WAN - LAN the max speed is 100 Mb/s reached on an average and only in certain countries gigabyte speeds are offer with a maximum throughput of 600Mb/s reached.

3. What is the ideal speed for WAN to LAN/WIFI. Say if a 1gb file is transferred from internet with 100Mb/s which approximately 8-12MB/s which both Lan & Wifi can handle with even wireless n 300Mb/s connection.

4. When the same 1gb file is transferred between Lan connected PC vs Wirelessly connected PDA the rates are obviously vary to high margin. So the transfer is not expected more than 20Mb/s for wireless N and 50Mb/s for Wireless AC.
But with lan the highest possible upto 800Mb/s and above is possible.

5.Same is applicable to trasnfer a data between devices connected to home say i video of 500Mb/s is transferred to connected storage of router. It needs to be downloaded by the client to upload to router rather not directly transferring as this reduces half of the available throughput.

 

[zakoh]

Only relatively few high end systems, such as discrete Wi-Fi adapters, premium laptops, such as MacBook Pro, or wireless bridges have three stream implementations of 802.11ac This means that for a small home or office, an AC1200 router could be just fine. However, AC1750 class routers are often relatively affordable, and have beefier internals, so the consensus that they may be worth the price. As for tri-band, MIMO-MU, or AC1900 class routers, you're basically throwing lots of money at little or no improvements at all in real life performance. For example, the AC1900 differs from AC1750 in that AC1900 router uses TurboQAM modulation in 2.4GHz spectrum, increasing the peak throughput only of that band from 450Mbps (for three streams) up to 600 (hence, the bump 250 "AC units" from AC1750 to AC1900). But there are almost no clients implementing TurboQAM now, and it is a proprietary extension, so it's possible that some clients, like Intel, will never implement it. Perhaps some of those quad-stream routers could be useful as a means of wirelessly bridging two LAN's together, but with ordinary clients there is no benefit from them.

MU-MIMO is also gimmickish because it does not optimize for wireless device to wireless device data transfers. It optimizes simultaneous router to wireless device data transfers (downlink only).

Personally, I have been very happy with a TP-Link Archer C8 (AC1750). I had it for a month and it has been pretty stable running the latest TP-Link firmware. Looks like a winner to me, considering you can often find it on sale only $10 more than an Archer C7.

As for wireless data transfers, I am seeing 10MB/second data transfer rate between two wireless AC1200 clients using http and CIFS with no line of sight to the router, so I am guessing it would have been somewhere in the neighborhood 20MB/second if the hard drive or the file server was attached directly to the router. This is certainly nowhere near "gigabit wireless" claim, but at least we have matched and beaten the twenty year old 100 megabit Ethernet (woohoo! cry from joy). Now I know that if I get a 100Mbps broadband, it won't be a waste (got 20ish now)

 

[ulaganath]

It could be taken into point but are ac router just bandwidth oriented. It's keep adding features over the period of time and I don't believe even ac quad based router bridge or ac1900 wds bridge is as effective as wired unless you want Internet of 1gbs to reach wired only devices. Mostly internal Nas or media server is quite common. For example to connect such devices one don't have to spend double the amount instead a PowerLine adapter is sufficient. I do agree ac1750 itself an overkill but keeping in mind it's first generation ac means as of now the support would be obsolete. So ac1900 now pretty common and quad and even. 3.2gb is getting evolved into the market. One need to decide how far the model will support considering the fact.

Sent from my ASUS_Z00AD using Tapatalk

 

[sfx2000]

MU-MIMO is also gimmickish because it does not optimize for wireless device to wireless device data transfers. It optimizes simultaneous router to wireless device data transfers (downlink only).

No, MU-MIMO is not a gimmick... It's serious stuff, and with the right clients, it does make more efficient use of Air Time on the downlink, which is most important for most folks.

The devil is in the details - it's not about increasing speed for a specific MU-MIMO client, but improving capacity and speed for all Clients on the WLAN.

 

[ulaganath]

Say as per the current wave 1 gen the speculation about the connection between a single ac client and router is full ac1300 even though the client support only 1x1 antenna. If 3 similar clients connect to the above router the router can't deliver say 20MB/s download be achieved the combined through put of 60MB/s can't be achieved instead it would be reduced to half or below that as it can do simultaneously serve the client needs all at once.

Sent from my ASUS_Z00AD using Tapatalk

 

[ulaganath]

Was that your MUMO DOES THAT

Sent from my ASUS_Z00AD using Tapatalk

 

[zakoh]

No, MU-MIMO is not a gimmick... It's serious stuff, and with the right clients, it does make more efficient use of Air Time on the downlink, which is most important for most folks.

The devil is in the details - it's not about increasing speed for a specific MU-MIMO client, but improving capacity and speed for all Clients on the WLAN.

The gains exist but they are are marginal right now IMHO. To take advantage of the MU-MIMO downlink performance, you need to have the WAN connection that's actually faster than say a single consumer level wireless-ac device. One issue now is that, at least in North America, even 100Mbps broadband is still an exotic thing and relatively expensive. According to some study I recently heard mentioned, the average broadband speed in the USA is still below 10Mbps. Although it seems like some major ISPs are planning to roll out consumer gigabit broadband at least in the major cities some time soon. So the usefulness of MU-MIMO is primarily through the downloads (but not uploads) from (but not to) devices that are directly attached to the router, like file or media server.