Showing Results: Most Recent
This review is from: LINKSYS WUSB6100M Next-Gen AC AC600 Mu-Mimo USB Adapter
Pros: Setup was extremely easy -- plug it in and windows 7 found it right away and installed the drivers. There is a disc included in the box, and I upgraded the drivers from linksys' web page "just because", but I did speed testing beforehand using a 5.8GHz MIMO network and found absolutely no speed difference vs. the stock drivers.
Ran some throughput tests on this device and was happy with its results. First test was a local LAN filesystem throughput test, and wired gig could write at 706Mbps and read at 550Mbps, and via the AC adapter was able to write at 142Mbps and read at 160.7Mbps. Those are sustained throughput numbers over five minutes and include all the losses associated with using a filesystem with SMB3, so while not coming anywhere close to wired gigE they were still quite good.
The test I expected to faceplant was the VOIP jitter test. Here I simulated 50 simultaneous VOIP calls for one minute. Usually on a wireless network this is enough data that the jitter will be all over the place. Happily, the jitter over my MIMO network was an average of 0.6ms, while the wired ethernet had jitter of 0.5ms. That's far far below the threshold where it'll matter for VOIP or for gaming, and I was very happy to see the network card performing so well so consistently!
Size of unit is a bit of a wide wedge; if you have USB ports close to each other horizontally or stacked vertically on top of each other (such as on a laptop), you're going to have a bad time and you'll lose access to one of those ports. You can fix that with a 3" USB extender cable, but who wants to do that?
You must plug this into a USB3 port to get anywhere near its maximum speed. That should be obvious, but it's worth mentioning. It still functions in a USB2 port, but at reduced maximum throughput.
The largest complaint I had with the unit (and this happened both with windows 7's built in drivers and linksys' drivers), if my desktop PC went into a deep sleep one out of every four times or so the adapter would NOT come back up when resuming from sleep. It was hard to pin down exactly what was failing, because it didn't fail every time the computer went into S3 sleep, only maybe 25% of the time. If I would pull the adapter out and plug it back in then it'd fire right back up, but that shouldn't be necessary; it should come back from deep sleep EVERY time. (I didn't think to try disabling and re-enabling the adapter when that happens, I'll have to try it and see what happens.)
Other Thoughts: Very good throughput and excellent jitter make this a great option for those with a MIMO network and can't do wired. While the throughput still can't touch wired gig, the reliability is top notch.READ FULL REVIEW
Pros: MAN I want to like this switch. I really really *really want* to like this switch. It's perfect for people like me, who have central PoE switches (especially folks like me with 802.3at switches) but might need to power more than one PoE device on a single cable run. Plus the switch itself is powered via PoE! This should be a slam dunk!
No power supply at all; the switch gets its power solely from your upstream PoE switch via port 5. But you do get four ports for you to use, two of them that could possibly have PoE flow through, in certain situations (see below).
No fans, so completely silent operation.
Nice metal case, gives it a bit of heft that feels like quality when it's in your hands.
Has five ports instead of the usual four, so that you can plug in four devices (max 2 PoE).
Can be easily buried or hidden behind a desk due to its small size. (About the size of a small paperback book.)
Runs very cool. Despite sitting directly on my carpet, the GS105PE is room temperature after running about a week continuously. No worries about overheating here!
Cons: Unfortunately this switch is not nearly the slam dunk that it should be, mainly because the software programming the switch is neutered to the point of being nearly useless. Netgear in their infinite wisdom has decided to omit nearly everything from the switch's control capabilities, including PoE. So you're truly flying blind with regards to PoE, as far as this switch is concerned.
Let me explain. PoE devices come in many different power classes (think of a power class like an energy budget). The problem happens when you get something that declares that it's a class 0 device, which is essentially "classless" and can by definition draw up to 12W. What my main 802.3at switch does when it encounters a class 0 device is it powers it up and then *measures* how much power is used. I can tell at a glance from my main switch that my power injectors use about 5W each. (Not a lot, considering 802.3at can deliver 25W!) So theoretically I *should* be able to hook two power injectors to this switch, have it power itself up and both of them, and be a happy camper.
Unfortunately it's not to be. First off, the gs105PE has no ability whatsoever to show any information, PoE or otherwise, about its connected devices! This blows my mind. The entire reason for being for this switch is PoE power passthrough. There is simply no way to get the amount of power used by a device, what PoE classification the device is presenting to the switch, *or even its MAC address*!! Worse, there's no way to unpower a PoE device short of physically pulling the plug! THIS IS NOT ACCEPTABLE FOR A POE SWITCH! I literally had to plug my devices into my regular switch to find out their PoE classifications and draws! (and MAC addresses, for that matter!)
More to the point, the GS105PE isn't intelligently programmed, either. It knows it's hooked up to my main 802.3at switch and can draw up to 25W of power. (That much was negotiated with my main switch, and I could see on the other end.) My main switch is telling me that the *combined draw* for the GS105PE and one of my power injectors hooked up to it is 7.1 watts. So I have 25-7=18W of power left, far more than the maximum 12.95W a single 802.3af could ever draw. But the GS105PE steadfastly refuses to power up both of them at the same time. One or the other it's fine, both and it's no soup for you!
Other Thoughts: Switch software has almost no features. It's actually faster to list the features that it does have rather than the features that are missing. Software supports port mirroring, per-port and 802.1Q vlaning, multicast, 802.1P/DSCP QoS rate limiting, IGMP snooping, EEE, and *some* version of loop detection though it's not stated anywhere what kind. (Guessing it's regular old spanning tree, but haven't bothered to break out wireshark).
Software does not support logging of anything to anywhere (not even locally!), NTP, LACP, LAG, LLDP, SNMP, RADIUS, TACACS, IPv6, voice VLAN, ARP table (it obviously *has* an arp table, but no way for the user to see what's connected to what), DiffServ QoS, 802.1X, ACLs, and MAC filtering. And those are only the features I found missing when comparing to my GS108T! So you can imagine, the control interface is really really sparse.
Pretty much you plug it in and in an emergency use it to split PoE for two native PoE devices (i.e. no injectors!) and could mirror a port (or three) with it. Anything more complex you're pretty much out of luck. The GS105PE could have been a great PoE passthrough switch with a software upgrade, or even a solid good by simply giving some insight into how its PoE is functioning. But as it is with firmware 188.8.131.52, it's merely OK. It's certainly fast enough, but having no visibility whatsoever is simply not OK for a PoE passthrough switch!
Pros: Fully modular power supply; you can even disconnect the ATX motherboard power connector (though I don't know why you'd want to as you wouldn't be able to start the power supply). Lots of plugs for everything; includes plenty of SATA and old school 4-pin drive plugs, along with two floppy connectors! The cables are flat rather than round which might be a ding for some, but look stylish in black.
No fan noise unless you're doing gaming. For regular PC work, the fan doesn't even spin! (There's a sticker pointing out that the fan not spinning is normal!) Once you fire up a game and your graphics card's draw really fires up the fan needs to spin to cool it, but by that time your graphics card's fans completely drown out the power supply's fan. So you effectively have a completely silent power supply, even though it's literally not true.
Packaging is a bit over the top... who needs a velvety bag for a power supply? Are people taking their power supplies out for nights on the town? Maybe I'm old school but I take a power supply, I attach it with screws, and until it dies I never look at it again.
80 Plus Gold certified, so if you buy a larger power supply than you need (which is what most people do) you're at least not thowing money away hand over fist with an inefficient power supply. So no matter how much load you demand (...or more importantly, DON'T demand) you're going to be at least 87% efficient. Nice! Granted it's not platinum or titanium, but those levels cost substantially more while only getting to 89% or 90% efficiency, respectively.
I don't know what it is about this design, but it seems as though it's more sleek than a regular power supply. I actually held it up to the power supply I removed that I was previously using, and it's the same size, it just somehow feels more sleek.
2x dual pci-e connectors for graphics cards. So you can run some monster rigs with this one.
Cons: Not 80 plus titanium. Not a real con, as that's kind of a ridiculous standard for most people.
Other Thoughts: If this power supply was stolen from me, and my previous power supply was broken, I'd likely buy this one as the replacement. Silent operation is very VERY nice, and it's only when you don't have it that you realize how much noise ruins experiences. I have absolutely no reason to swap this power supply out now that I've reviewed it -- it's just that good!READ FULL REVIEW