Apple TV won't remember WPA key

[qoncept]

I ran out of cat5 in my living room and had to start using my Apple TV with a wireless connection. It works fine, but every time it reboots it forgets my WPA key. The power has been going out quite a bit and I'm going to pull my hair out if I have to use the terrible Apple TV keyboard interface again. Please save me!

 

[Chipskip]

First- if your router is "n" than wireless is faster than cable with the TV

Second- I would suggest doing a factory restore and seeing if that helps.

Third- is this a newer TV still under warranty? If so try contacting apple, but I bet they will have you try a factory restore first.

Fourth- If the power is going out a lot you want to shutdown your computers equipment (including TV) when you are not using it. Maybe even get a battery backup system if you think the power issue will continue.

Fifth- Make your passcode easier! make it so that it is easy to type in via the remote. Just don't use 12345

 

[joudbren]

This problem usually seems to be router specific. I had a Dlink DIR-655 which worked flawlessly with the ATV and now I have a Netgear WNDR3700 as I wanted dual radio support for both 2.4ghz and 5.0ghz. I was initially having problems with the Netgear which were solved with a firmware flash.

It was doing pretty much the same thing, forgetting the password, forgetting my iTunes password as well and dropping the connection all the time when the ATV wasn't being used. That has settled out quite a bit with the updated firmware but I still seem to have to wake it up all the time when I go to use the ATV and that takes about 30 seconds. I'm connecting on the 5ghz radio in N mode only. With the Dlink I never had any of these issues.

It's ok now as at least it's remembering the passwords but safe bet you have a router compatibility issue or possibly some interference from something else in the house or a neighbour's wireless setup. You don't mention what brand of router you have and if it has the latest firmware. You should also be using WPA2 if possible and AES encryption only, again, if possible. If you have multiple devices connecting to the router that may be a little tricky for older stuff.

That's all I have! Cheers!

James

 

[qoncept]

The router is an Asus WL500W running the latest version of DD-WRT firmware. And yeah, I'm actually using WPA2 encryption, not WPA. I was using some sort of Netgear with DD-WRT before the radio in it died and wasn't having this issue, so you're right, it probably is router specific.

I'm planning on running cat6 throughout the house before I drywall my basement so I'll probably just add an extra run to the living room as I prefer to use a wired connection for anything I'm not moving around frequently. In the meantime, it sure is annoying.

 

[fpnc]

First- if your router is "n" than wireless is faster than cable with the TV...

That's often not true. Under many situations even 100Mb, wired ethernet will be faster than 802.11n. Wireless has a theoretical advantage but in the real world it gets only a small fraction of its rated speeds.

As for the OP, I use to have this problem until I updated my Apple TV to the latest software (the "fix" happened, I think, somewhere in the 2.X release series). What version of the Apple TV software are you using? If you're running the latest software, then I'd try a full factory restore as an earlier reader suggested (but make certain you have backups of all of your content/media before you do the factory restore on the Apple TV).

 

[Chipskip]

That's often not true. Under many situations even 100Mb, wired ethernet will be faster than 802.11n. Wireless has a theoretical advantage but in the real world it gets only a small fraction of its rated speeds.

Although I can not speak for other setups, I have tested my system and three other networks in which I have setup for family/ friends. If everything in the network is wireless then you can bog down to a lower speed. One fix is a dual antenna or even dual band router. Another is to wire what you can (that makes since), however the TV is restricted to 100MB NIC but can see speeds up to 300MB wireless (via N). If you have a wireless N network and are seeing speeds less than 100MB then you have issue with your network.

I test my network regularly and usually see 160MB and sometimes up to 300MB true file transfer speed from laptop (wireless) to TV (wireless)

Now all that said I will be removing the wireless card from my TV to add a CrystalHD card. So its either time to do some hard wiring or an airport express, either way I am restricted to the lower speed.

 

[fpnc]

...I test my network regularly and usually see 160MB and sometimes up to 300MB true file transfer speed from laptop (wireless) to TV (wireless)...

Okay, then maybe you can explain the wireless throughput results from testing organization like the following (where maximum throughputs range from 50Mbps to 70Mbps -- best case):

(2.4GHz band, four different 802.11n routers)
http://www.smallnetbuilder.com/wire...n-router-with-gigabit-switch-reviewed?start=1

(dual-band, Belkin Play Max Wireless Router)
http://www.smallnetbuilder.com/wire...to-the-charts-belkin-play-max-wireless-router

(or this page where you can compare about 30 different routers)
http://www.smallnetbuilder.com/component/option,com_wireless/Itemid,200

(a different source -- CNET -- 5GHz band, latest Apple AirPort Extreme-N )
http://reviews.cnet.com/wireless-ac...base/4505-3265_7-33549749-2.html?tag=txt;page

Or, to quote CNET:

The AirPort Extreme performed well in the 5Ghz band, scoring a 66.48Mbps on our throughput test, where we put the test client just 15 feet away from the router. This means it would take about a minute to transmit 500MB of data. This wasn't the highest score, as the D-Link DIR 825 scored a 80.96Mbps. On our range test, however, where the client was 100 feet away, the D-Link's speed declined significantly to 36.48Mbps, while the AirPort Extreme remained high at 59.04Mbps.

As for my setup, when running from room-to-room (about 35 feet) over an Airport Extreme-N I get maximums on file transfers of about 50Mbps (that's megaBITS, in case you are wondering). However, over 100Mb wired ethernet I get consistent results at about 90Mbps+ (amazingly close to full spec. -- wired ethernet is a mature technology and works very well, something that can't often be said about 802.11 wireless)

 

[fpnc]

Here is a "bump" on this thread because I'd really like to know how Chipskip is seeing:

160MB and sometimes up to 300MB true file transfer speed from laptop (wireless) to TV (wireless)

Frankly, without further confirmations or examples or data I've got to assume that those number are wrong and perhaps incorrect by nearly an order of magnitude (and even here I'm assuming that when "MB" is written that Chipskip means megabits and not megabytes). I guess Chipskip may be talking about "link rate" which really isn't the same as throughput as link rate does not report true data throughput.

Frankly, I've never seen real-world results like Chipskip is reporting for wireless file transfers and even when a basestation is sitting only a few feet from a wireless client the throughputs will generally still be less than 100Mbps. Well, I have read about one exception to that type of throughput, when using a high-priced, commercial wireless router but that's not the type of equipment that most people have in their homes (and certainly not what Apple's AirPort Extreme or most typical wireless basestations can do).

In any case, here is an FAQ that talks about real-world throughput when using 802.11N (it also discusses link rate):

http://smallnetbuilder.com/wireless...-smallnetbuilders-wireless-faq-the-essentials

Interestingly, this FAQ also makes the following statement:

I want to stream HD video wirelessly. What N router should I buy?

Simply put, none. There is no product on the market today that can reliably deliver trouble-free HD content over a wireless connection, even if the distance is just a few rooms. You really need to use an Ethernet connection. 10/100 is fine. You don't need Gigabit Ethernet.

And this statement pretty much backs up my claim that under almost all conditions 100Mb wired ethernet will perform better than 802.11N wireless.

 

[Phantom Gremlin]

Frankly, I've never seen real-world results like Chipskip is reporting for wireless file transfers and even when a basestation is sitting only a few feet from a wireless client the throughputs will generally still be less than 100Mbps. Well, I have read about one exception to that type of throughput, when using a high-priced, commercial wireless router but that's not the type of equipment that most people have in their homes (and certainly not what Apple's AirPort Extreme or most typical wireless basestations can do).

When I bought my Airport Extreme Base Station (late 2009 version) I did some performance tests between it and a vintage 2008 Macbook. Using AJA System Test (a disk stress benchmark) between the wireless Macbook running 802.11n on channel 44 (5 GHz) and an iMac hardwired to the Extreme, I achieved rates of 15.8 MB/s reading and 16.8 MB/s writing. That's with a few feet of wireless. So that matches Chipskip's claimed "160 MB" which I concur can only mean 160 Mb.

As you might expect, those numbers drop dramatically when the Macbook is moved about 50 feet away, with a few walls or floors in between. The lower rates were about 4 MB/s. Also I've run some tests with Apple's Airport Express (latest n version) and the throughput is much lower than the Macbook can achieve. Which makes sense once you read that Apple didn't upgrade the CPU hardware in the Express when they went from 802.11g to 802.11n, they just upgraded the wireless circuitry.

Anyway, the point is the Apple hardware has no problem achieving very impressive throughput "when a basestation is sitting only a few feet from a wireless client". But that's not why I like their wireless hardware; the reason is that most other consumer wireless hardware is the cheapest sort of cr*p, frequently stalling, frequently dropping out, frequently needing to be power cycled.

 

[fpnc]

When I bought my Airport Extreme Base Station (late 2009 version) I did some performance tests between it and a vintage 2008 Macbook. Using AJA System Test (a disk stress benchmark) between the wireless Macbook running 802.11n on channel 44 (5 GHz) and an iMac hardwired to the Extreme, I achieved rates of 15.8 MB/s reading and 16.8 MB/s writing. That's with a few feet of wireless. So that matches Chipskip's claimed "160 MB" which I concur can only mean 160 Mb...

Thanks for those results. However, having your wireless client sitting a few feet away from the basestation hardly makes much sense particularly when you are talking about the Apple TV. Beside that, Chipskip claimed up to "300MB" which hardly seems likely (and your measured 16.8MB/s maximum equals just 134Mbps which is smack dab between my admitted 100Mpbs over a few feet and Chipskip's lower claim of "160MB"). I think Chipskip was using the link rate, which isn't the same as real throughput. In fact, I see link rates up to 300 on my Airport Extreme but that doesn't mean I'm going to get 300Mbps file transfers (not even remotely close to that figure).

Furthermore, I'm not certain that results gathered through AJA System Test would sit well against Chipskip's claim of measuring "true file transfer speed from laptop (wireless) to TV (wireless)." Lastly, what began this conversation was Chipskip's claim that a 802.11N wireless connection to an Apple TV would be faster than 100Mb wired ethernet which I still maintain will hardly ever be the case (i.e. stick with wired ethernet if you can).

 

[Phantom Gremlin]

(i.e. stick with wired ethernet if you can).

That's good, practical advice. But wireless has at least the potential to work just fine in streaming/viewing HD. Here's another personal example.

TiVo HD <-- Airport Express 5 GHz --> AEBS simultaneous dual band <-- TiVo 802.11g USB adapter 2.4 GHz --> TiVo HD

I just yesterday transferred a file from one TiVo to the other. These TiVo boxes are quite underpowered and normally will max out at about 27 Mb/s over wired Ethernet when otherwise "idle". Mine were moderately loaded at the time (e.g. viewing and recording some programs).

I achieved 19.33 Mb/s transfer rate, a 1 hr HD program of 6.38 GB transferred in 47 min, 14 sec. I know this because the TiVo lets you view the statistics for the last transfer you did.

That's faster than real time for "true" HD. (The MPEG-4 stuff normally used by e.g. the Apple TV is probably less than half that bit rate.) And that's with 802.11n from one TiVo to the AEBS and 802.11g from the AEBS to the other TiVo. I.e. "simultaneous dual band". And that's through a few walls and one floor.

The potential for quite usable wireless is here now. The real question everyone should be asking themselves about wireless is why the transfer rates encountered by the "average person" in the "real world" are so bad.

The 50 Mb/s rates you report in your setup should be perfectly adequate for most HD. I would love to run Cat 6 around the house "just because", but the little woman doesn't approve of the aesthetics.

 

[fpnc]

That's good, practical advice. But wireless has at least the potential to work just fine in streaming/viewing HD. Here's another personal example.

TiVo HD <-- Airport Express 5 GHz --> AEBS simultaneous dual band <-- TiVo 802.11g USB adapter 2.4 GHz --> TiVo HD...

I never suggested that wireless couldn't be used for streaming to the Apple TV (although the quote from Small Net Builder wasn't too positive on using wireless for HD). I use 802.11N wireless for my Apple TV and it works pretty well, but I maintain that Chipskip was incorrect to state that wireless-N was faster than 100Mb ethernet and that the Apple TV could "see speeds up to 300MB wireless (via N)."

In fact, Chipskip said the following:

If you have a wireless N network and are seeing speeds less than 100MB then you have [an] issue with your network.

Which, in my opinion is misleading if not outright wrong.

My only point is this -- under almost all conditions 100Mb ethernet will outperform 802.11N wireless (unless you are using high-priced, commercial WiFi equipment or are within a very few feet of the wireless base station).