Linksys EA8500 - First MU-MIMO Enabled 802.11ac Router Set to Shipby Ganesh T S on May 7, 2015 7:00 AM EST
- Posted in
- Qualcomm Atheros
Readers following our wireless networking coverage must be quite familiar with the developments in 802.11ac land. As a recap:
- The first 3-stream 802.11ac routers (based on the Broadcom BCM4706 chipset) started shipping in May 2012, with Buffalo Technologies and Netgear leading the way. There were marketed as AC1750 units (3x3 5 GHz providing up to 1300 Mbps and 3x3 2.4 GHz providing up to 450 Mbps). The street pricing of the AC1750 routers started to go down once Qualcomm started shipping their first-generation 802.11ac router platforms in products such as the TP-LINK Archer C7 in April 2013.
- In Q4 2013, a wave of AC1900 units started to hit the market. Netgear's Nighthawk R7000, a popular member in this series, still continues to sell well in the market. These units still used a 3x3 5 GHz radio for up to 1300 Mbps, but the 2.4 GHz had some proprietary Broadcom extensions (TurboQAM) to support up to 600 Mbps in an end-to-end Broadcom environment.
- In 2014, Quantenna began to bring their technology lead in the Wi-Fi space to the market with their Wave 2 4x4 802.11ac chipset. Despite being announced at CES 2014, the Asus RT-AC87U began to ship in July with the QSR1000 radio coupled with a Broadcom SoC. Netgear started to ship the same radio coupled with a Qualcomm IPQ8064 SoC in the Nighthawk X4 in September 2014. The QSR1000's Wave 2 feature set includes support for MU-MIMO, but neither of these routers shipped with MU-MIMO enabled in the firmware (citing lack of client support at that point of time). Given the 4x4 configuration, these routers were tagged as AC2350 / AC2400 units (600 Mbps in the 2.4 GHz band and 1733 Mbps in the 5 GHz band)
- In the midst of these Wave 2 product launches, Qualcomm announced plans for their Wave 2 802.11ac platforms (both routers and clients) in April 2014. Broadcom, being late to the Wave 2 party, decided to bring out the XStream platform with dual 5 GHz radios (six-stream configuration) also in April. They were able to get it quickly into the market too, with the Netgear Nighthawk X6 R8000 becoming available in June 2014.
- At CES 2015, we saw the first set of routers based on Qualcomm's QCA9880 4x4 Wave 2 802.11ac router platform getting announced. Vendors such as TP-LINK and TRENDnet were expecting their AC2600 class routers to ship towards the middle of the year.
Coming back to today's announcement regarding the EA8500, Linksys is becoming the first vendor to start shipping a QCA9980-based router. The important aspect here is that the router will ship with MU-MIMO enabled. Client devices with MU-MIMO enabled are also coming into the market just now (more on that in our detailed review). Hopefully, this announcement will spur both Netgear and Asus to enable MU-MIMO in the firmware for their Quantenna-based routers.
The salient features of the EA8500 router are listed below:
- 4x4 802.11ac with MU-MIMO and beamforming support
- QCA9980 radio + 1.4 GHz dual-core IPQ8064 SoC
- Simultaneous dual-band operation (5 GHz 802.11ac - 1733 Mbps, 5 GHz 802.11n - 600 Mbps, 2.4 GHz 802.11n - 800 Mbps)
- 1x USB 3.0, 1x USB 2.0 / eSATA
- 4x 1Gbps LAN, 1x 1Gbps WAN
- Adjustable antennas for optimal performance
One important point to note is that MU-MIMO capabilities will translate into real benefit only when there are multiple MU-MIMO-capable client devices talking to the router simultaneously. This will probably be a common scenario in enterprise Wi-Fi deployments soon. However, in a typical household, we would imagine that a situation involving multiple MU-MIMO clients is probably at least a good year away. Qualcomm is targeting an end-to-end play here, as the announcement made last year included a number of products in the client space with MU-MIMO capabilities.
Linksys is accepting pre-orders at $280 for the EA8500 today with shipments slated for May 10, 2015. Wide retail and e-tail availability is also expected starting next week. There is no doubt that MU-MIMO will be the most efficient way to boost Wi-Fi performance in the near future, and it is good that Linksys and Qualcomm are teaming together to accelerate its market adoption.
Post Your CommentPlease log in or sign up to comment.
View All Comments
dreamcat4 - Friday, May 8, 2015 - linkstun. I can tell you some of it:
* Linksys EA8500
This one is most like the existing Linksys WRT1900ac. And can be considered it's replacement. Although we don't know for sure (because it's not released yet). It seems to have same cores and other internals hardware etc. Except for the radios. Which they have changed out.
* The WRT1900 currently supports manufacturer firmware, OpenWRT and DD-WRT. But not Tomato.
Problem is: all Wave2 radios are Quallcomm new chipset. And therefore closed source driver = probably no wireless in the open firmwares support.
This applies even more so to the X4 (R7500) because not only has that 'brand new' Qualcomm Wave2 radios. But actually the CPU SoC is Qalcomm too. So it's only got official Netgear firmware on it. Which is basically terrible. And that is why to avoid the X4 for a very long time (probably years). Despite it being the fastest / technically most superior hardware than any of these routers.
X6 - No point don't bother. All the extra antennas are a waste of money.
Apple Airport - is very good AC wifi. And a good product. But no open firmware support whatsoever. So you are at the mercy of Apple's firmware updates etc.
Netgear R7000 Nighthawk. Is best supported for open firmwares. All of them right now straight away. Including Tomato too. Therefore, the R7000 is the one to buy ATM. It also gives excellent Wifi ac performance. Cheaper / reduced in price a bit now. So is best value for money. And mature well tested product - lots of knowledge / quirks are well known.
WRT1200ac - not released yet. But will have same open firmware support as the WRT1900ac. It's just fewer radios. These are also technically a VERY good product but sorry not 1st choice when you can't run tomato on it.
In router space - the hardware moves very quickly. Software moves very slowly. So all the newest and last years router are not well supported in open software. Until several years away. And it is worse with companies like qualcomm who really want to protect their IP / new innovations from being copied their competitors. = no open source drivers. But they will often help release stuff later on, once they obselete it with even newer products. So if anything, these new MIMO products are good. Because it means that the previous ones it replaces (like Linksys WRT1900) can become more well supported now in the open source lands.
DanNeely - Friday, May 8, 2015 - linkJust curious about your strong preference for Tomato because I've never used it. Does Tomato do some things significantly better than DD-WRT; or is it just a case of sticking with what you tried first because it does everything you need it to do. (Why I've never tried anything but DD-WRT.)
dreamcat4 - Friday, May 8, 2015 - linkTomato is most similar to DD-WRT features. So most user may use either DD-WRT or Tomato. But the reason it's good to have both is a freedom to switch. Since unfortunately these days DD-WRT has gotten a bit complex for it's own boots. And new commits breaking features without knowing. A lot of that is because DD-WRT support so many different hardware.
Wheras Tomato only work on relatively fewer routers. This means that the tomato developers are in a situation where they can test and re-test their firmware more thouroughly on those fewer products. Hence it is more stable / less broken features / confusion hassle trying to make settings work that cannot because there is some unknown bug.
But DD-WRT is not a bad firmware at all. Especially if it happens to be all working, for the precise commit that you decided to install.
Another feature of tomato is Jacky's 'advanced Tomato v2'. Which is just a fancy ajax Web 2.0 interface bolted on top of Shibby regular enhanced tomato firmware. Which graphically, for the WebGUI puts all other router firmwares to shame. Of course, that's just a bonus.
Tomato also seems best choice for QoS. If that is important to you.
This is why I say 'just get the Netgear R7000 Nighthawk'... because than you don't have to worry or decide which firmware you like best. You can try all of them, and then finally settle on which one you prefer. Including the manufacturer. So if 'no tomato' = fewer options. Not the ideal situation.
I would mention ASUS. But their product has a slower CPU than the R7000 Nighthawk. Which is a disadvantage for certain CPU - heavy features. Such as VPN server, etc. Also their USB 3 port is not shielded correctly and it interfers with the Wifi signal. Also their Asus Merlin CFW Firmware - I'm not so easily impressed as others as since it's mostly a manufacturer firmware (meaning basic and lacking certain advanced features of DD-WRT or tomato).
zodiacfml - Saturday, May 9, 2015 - linkI disagree with the X6 comment. It is actually more useful than all the MU-MIMO routers you can buy because MU-MIMO needs client support which is almost none except for some new Acer laptops I've heard.
The X6 is basically two 5Ghz WiFi routers which explains the number of visible antennas.
If you have multiple devices connected to the router of around one dozen, then there's a benefit. The reason to have two routers is to segregate the devices sharing a single WiFi signal.
In a highway, we have multiple lanes for multiple speeds. In the X6 or other Tri-Band routers, they put the fast WIFI devices on one signal, the other on the medium speed, while the slowest on the 2.4 Ghz.
dreamcat4 - Saturday, May 9, 2015 - linkThat is a good point.
However there was good reason for my comment. It is twofold. 1) R8000 is not so well supported for open firmwares as the R7000. 2) For the extra cost of the R8000. You can either buy about 2x R7000. Or 1x R7000 + 1x Wireless repeater. Which combined effectively gives you the only extra benefit of more X6 radios. Yet without X6 you have the better open firmwares support plus also the freedom to position the 2 hotspot in different locations to one another. Rather than being stuck in the same single physical device.
geekfool - Friday, May 8, 2015 - linkomg just tell us what the heck mu mimo is.
UtilityMax - Tuesday, May 12, 2015 - linkWell, the key improvement of the wireless-n and wireless-ac standards over the old g standard is that they allow several simultaneous streams of data (each stream much faster than the old wireless-g), even over the "narrow" 20MHz channel. (The other improvement is that you can use wide channels). However, the issue is that 2x2 MIMO (2 steams) is only now becoming standard, and 3-stream adapters are installed on very high end laptops (like MacBook Pro), while at the same time there are many routers on the market with 2-3-4 stream MIMO. Well, it turns out that with the old (now current MIMO) the router takes turns talking to each client device instead of talking to them at once. So a three-stream router talking to one-stream device, can't talk to another one or two-stream device at the same time. The MU-MIMO (multiuser MIMO) fixes that. Arguably, that's quite an improvement. But it's not worth buying a MU-MIMO router right now, because MU-MIMO client hardware practically does not exist. I'd way a couple of years until I have MU-MIMO capable phones, tablets, and PCs before I buy a router with MU-MIMO.
Lucky Stripes 99 - Tuesday, May 12, 2015 - link$280 is rather steep for a device that doesn't have either a fixed 10GbE port or a SFP/SFP+ port. You can pick up MicroTik switches with at least one 10GbE port for just under $200 these days.
UtilityMax - Tuesday, May 12, 2015 - linkYou are really confused if you think that not having 10Gbit switch is a limitation, much less the biggest limitation of this router. Considering that the wireless-ac does not sustain or only barely approaches real world data transfer rates near 1Gbps (that would be with a MU-MIMO clients, which don't even exist), 10Gbps WAN port on this router would be a complete waste of resources.
Lucky Stripes 99 - Tuesday, May 12, 2015 - linkI would want the 10GbE port on the LAN side for a file server or as an uplink to a workgroup switch. Saturating a 1Gbps port is easy these days with several wired clients.
Having a SFP port on the WAN side could allow this router to directly connect to a FttH/FttP network using either active or passive optical Ethernet, which is increasingly common these days.
Asking that the 10GbE port be a SFP+ port might be a bit much for a sub-$300 device, but it isn't without reason. I am seeing people increasingly skip CAT6a and CAT7 copper for OM4 multimode fiber for long runs. 10GbE over legacy CAT6 is limited to around 170ft/50m, over CAT5e is limited to around 130ft/30m, more or less depending on cable quality and the presence of adjacent cables.
Please don't ever call me confused again.