Add Comment
|
Related Links
|
TrackBack
Related Content
need a new router
i'm tired of my befrs41 and lately it's been acting up. had it for years.
any recommendations? 4 ports minimum and wireless N.
charlie
Wed, 06 Jan 2010 18:53:33 +0000
I am contemplating the airport.
Sarcast
Wed, 06 Jan 2010 20:29:34 +0000
No specific recommendations... bestsellers at amazon...
Bestsellers in Routers
1) d-link dir-655 is very popular with good ratings based on the number of reviews
2) people seems to like the linksys wrt160n, provided that it's v2 and upgrade to dd-wrt
3) netgear wndr3700 seems ok but pricey
gnowk
Wed, 06 Jan 2010 22:39:35 +0000
I bought a refurbished Netgear WNR834B for about $20 a year ago:
Recertified Netgear WNR834B Wireless N Router $19.99 AR
It has ran flawless for the past year. Never had trouble with it. I checked earlier this week, this router is going for $80 refurbished. But Geeks has
it for $29.99 (through Amazon) or $22 direct. I've updated the link
above.
I've heard that this firmware on this router is hackable. But I have
never tried it myself, nor have I really looked into it. Let me know
what you ended up with.
Chieh Cheng
Thu, 07 Jan 2010 00:21:37 +0000
I did a quick search of "Netgear WNR834B firmware hack" and the first
page on Google has Netgear WNR834B loaded with dd-wrt. I've looked no
further than that. So you'll have to do more legwork if that's what
you are interested in, before buying the router.
Chieh Cheng
Thu, 07 Jan 2010 00:51:38 +0000
I bought the Netgear WNR834B v2 from Newegg in July for $25 after tax/shipping.
I haven't powered it on for awhile. The range wasn't that great for me compared to my Linksys WRT54g. Plus I was only getting good throughput under 802.11n when I don't use any encryption. I upgraded to dd-wrt, but it didn't seem to help.
How's the throughput for you?
gnowk
Thu, 07 Jan 2010 02:41:40 +0000
I don't know. If you show me how to test throughput, I'll test it for
you. I don't use encryption.
Maybe you should hand it over to Charlie. ;-) Or off-load it to him for $30.
Chieh Cheng
Thu, 07 Jan 2010 04:20:44 +0000
If you want to see power consumption, I've listed it here:
Kill a Watt Measurements: Network Peripherals
If you guys have a Kill a Watt, feel free the list your network
equipment (or other devices) on my Kill a Watt pages.
Chieh Cheng
Thu, 07 Jan 2010 06:30:47 +0000
To get a sense of the throughput, you can start a large xfer over wifi from one box to another. While the xfer is going, monitor the xfer speed....
Under osx, you can just use Activity Monitor and switch to the Network tab.
Under windows, I use NetPerSec, but I can't find a free download link right now. Netmeter should work (http://www.metal-machine.de/readerror/index.php?action=tpmod . . .).
gnowk
Thu, 07 Jan 2010 09:08:49 +0000
gnowk, maybe you can do what Chieh does.
Sarcast
Thu, 07 Jan 2010 11:09:52 +0000
ok.... i think i'll get this cause it's $25 and has N.
Netgear WNR2000 Wireless N Router - 300Mbps, 802.11n, 4-Port, Recertified
charlie
Thu, 07 Jan 2010 13:54:54 +0000
The shell looks exactly the same as the WNR834B, except it is black.
Do you know what is the difference between this Netgear WNR2000 Wireless N Router and the WNR834B?
Chieh Cheng
Thu, 07 Jan 2010 16:51:55 +0000
although it's a refurb, this router is $25, when it sells for $79.73 on sale. that's all i know. i never heard of this one or the 834B.
charlie
Thu, 07 Jan 2010 21:48:00 +0000
I tried a test run on a wired desktop to a wireless-G notebook on
Windows Vista. The Vista transfer is weird. It starts really slow and
the slowly ramp up over time. So I don't see how I can test the
throughput with any real sense of speed (my wireless-N laptop is
Windows 7, which is the same as Vista).
Chieh Cheng
Thu, 07 Jan 2010 23:57:01 +0000
The router Charlie is looking at is the WNR2000. It's newer than the one that Chieh and I have. WNR2000 uses the Atheros chipset, while the WNR834B uses Broadcom chipset. The reviews show better throughput for the WNR834B, but they were written more than a year ago, so I am not sure if this still holds true with the newest firmwares.
Review for the WNR2000 (dated 10/08):
http://www.smallnetbuilder.com/wireless/wireless-reviews/30615-netgear-wnr2000-wireless-n-router-reviewed
Review for the WNR834B (dated 9/07):
http://www.smallnetbuilder.com/wireless/wireless-reviews/30152-netgear-wnr834b-rangemax-next-wireless-n-router-review
gnowk
Fri, 08 Jan 2010 00:36:04 +0000
I don't know how you read the charts, but the first one I encountered
in the two reviews is that WNR2000 has better throughput.
By the way, on a related, but tangent note, I don't really see how we
can test throughput at home or why it's even worthwhile. The equipment
we use at home are all completely different and doesn't necessary
complement one another.
One primary example is when I got fiber with 25 Mbps down and 5 Mbps
up. First test the service tech and I ran was using SpeakEasy's DSL
tester. We were able to sustain 25 Mbps down and 5 Mbps down on my
Pentium 4 Wired desktop.
I had not been able to sustain that throughput on any other computers
at home, even though they are using the 54 Mbps/108 Mbps wireless-G
connection, which should in fact, out run the 25/5 Mbps limit.
I've found that on different laptops the sustainable throughput are
different. Based on my earlier test, I would say you would have to
first max out your CPU/system bus/IO throughput before attempt to max
your Internet and network throughput.
To date, out of approximately 10 computers we have at home. Only one
computer is able to achieve the highest Internet performance. My take
is that talking about throughput is moot. Just pick a router based on
the reviews and be happy. For me, I just pick a router based on price,
theoretical performance, and (mostly) reliability.
Chieh Cheng
Fri, 08 Jan 2010 01:30:24 +0000
I was referring to the wireless performance (broken down by encryption used). The first one you saw is the routing performance.
I don't have FIOS at home, so I don't have anything to add there. But I do care about wireless performance within my home network. You can only see that by measuring throughput. Theoretical performance doesn't tell you anything. With some of the 802.11n routers out there, you're only getting 802.11g speed. So why bother upgrading?
gnowk
Fri, 08 Jan 2010 03:11:29 +0000
> I was referring to the wireless performance (broken down by encryption
> used). The first one you saw is the routing performance.
Ok. I'll try running it. But I can't guarantee you will really get any
results, as the only wireless-n laptop is only a netbook. And I have
no doubt it is not optimized to be good at anything. Any difference
your'll see is probably bottleneck on the netbook.
Personally, I think this test will only be meaningful if you have a
computer that is fully mixed out on throughput and network connect is
its only bottleneck.
> I don't have FIOS at home, so I don't have anything to add there. But I do
> care about wireless performance within my home network. You can only see
> that by measuring throughput. Theoretical performance doesn't tell you
> anything. With some of the 802.11n routers out there, you're only getting
> 802.11g speed. So why bother upgrading?
I didn't realize some routers are only giving you 802.11g speed.
I think we are looking for different things. The only reason I
upgraded was because two of my verizon routers are completely flakey.
I have to reboot one of them every two days. The wireless on the other
could crap out at undeterministic periods. I was running three routers
(acting as AP) just to keep availability up (as I'm running web sites
at home).
So when the WNR834B was staying up all year round, I was pretty much
content and couldn't care less about performance. The wireless-N
upgrade was a first strike at the future, as I don't have any
equipment (other than my netbook) to take advantage of it. And even
with my netbook, I can't take advantage of it, because all my servers
are wireless-g or slower. Netbook was only used on the road.
Chieh Cheng
Fri, 08 Jan 2010 06:01:31 +0000
ordered. =)
charlie
Fri, 08 Jan 2010 07:32:33 +0000
Did your throughput test match the reviews?
Chieh Cheng
Fri, 08 Jan 2010 08:28:35 +0000
I have to test again to compare. All I can recall is that the performance was just a tad better than my WRT54g. I was expecting a bigger improvement.
As for performance among the 802.11n routers, sometimes the best router can be twice as fast as the worst performer in certain scenarios:
NETGEAR WNR2000 Wireless-N Router Reviewed - Wireless Security Performance, Wireless Competitive
gnowk
Fri, 08 Jan 2010 11:23:37 +0000
I ran the SpeakEasy speed test with my wireless N netbook:
Speakeasy Speed Test
The result is
Download Speed: 19149 kbps (2393.6 KB/sec transfer rate)
Upload Speed: 4225 kbps (528.1 KB/sec transfer rate)
Close, but could not sustain the full fiber speed. This is without any
encryption at all. I'll try it with encryption later tonight and see
how that works.
Chieh Cheng
Fri, 08 Jan 2010 14:18:39 +0000
A note on why you can't get 25Mbps on your 802.11g equipment.
Your FIOS is rated by the effective data transfer rate (overhead already excluded), while the advertised 54Mbps for 802.11g doesn't account for overhead.
Typically, your effective data rate is about ~22Mbps according to http://en.wikipedia.org/wiki/IEEE_802.11#802.11g
gnowk
Fri, 08 Jan 2010 15:01:41 +0000
The previous test result (quoted below) was done with 130 Mbps, which
I believe is compatible with wireless g devices.
The result for 270 Mbps mode (which I believe is not compatible with
wireless g devices) and no encryption is:
Download Speed: 17592 kbps (2199 KB/sec transfer rate)
Upload Speed: 4400 kbps (550 KB/sec transfer rate)
WEP 64-bit Encryption, 270 Mbps:
Download Speed: 15712 kbps (1964 KB/sec transfer rate)
Upload Speed: 4299 kbps (537.4 KB/sec transfer rate)
WEP 128-bit Encryption, 270 Mbps:
Download Speed: 16452 kbps (2056.5 KB/sec transfer rate)
Upload Speed: 4350 kbps (543.8 KB/sec transfer rate)
WPA TKIP, 270 Mbps:
Download Speed: 17925 kbps (2240.6 KB/sec transfer rate)
Upload Speed: 4239 kbps (529.9 KB/sec transfer rate)
WPA2 AES, 270 Mbps:
Download Speed: 19126 kbps (2390.8 KB/sec transfer rate)
Upload Speed: 4425 kbps (553.1 KB/sec transfer rate)
Ok. I think that's all the encryption modes. Let us know what you
gleamed from all this data and from your own router.
Chieh Cheng
Fri, 08 Jan 2010 16:58:42 +0000
I put it into a table for easy reference here:
Netgear WNR834B Wireless-N Router Network ThroughputNetgear WNR834B Wireless-N Router Network Throughput
Based on the table, it seems that WPA2 is the way to go if you want
throughput. It's even slightly faster than no encryption.
What encryption mode were you using?
Chieh Cheng
Fri, 08 Jan 2010 18:00:43 +0000
Ok, I bought the idea of effective transfer rate at first. But then I started think about the 5
Mbps up connection. What's the reasoning that I can't achieve 5 Mbps
up with an effective transfer rate of ~22 Mbps? Don't tell me it's my
Internet connection, because like I have said, I sustain both 25 Mbps
down and 5 Mbps up using the wired desktop computer with the same
speed test.
Chieh Cheng
Fri, 08 Jan 2010 18:47:45 +0000
how was your wired desktop set up? was it connected directly to the fios modem? or was it connected to one of the ethernet ports on your WNR834R?
gnowk
Fri, 08 Jan 2010 21:39:46 +0000
That's a good point. I don't remember how it was connected before, but
my network has changed several times since then. Now that you reminded
me, the Pentium 4 desktop I have now was not the same desktop I used
to performed the test.
Currently the desktop is connected directly to the fiber optic modem.
So I just ran a test on the desktop and here is the result:
Download Speed: 19253 kbps (2406.6 KB/sec transfer rate)
Upload Speed: 4585 kbps (573.1 KB/sec transfer rate)
Then I reconnected the cable so that the desktop goes through the
Netgear WNR834R router and then into the fiber optic modem. The
result:
Download Speed: 19270 kbps (2408.8 KB/sec transfer rate)
Upload Speed: 4524 kbps (565.5 KB/sec transfer rate)
Based on this test, I found that the extra link on the wired
connection is a non-factor.
However, I am no longer achieving the 25/5 Mbps connection that I've
achieved before. The Pentium 4 computer is faster than my previous
desktops. So I don't know what changed.
Chieh Cheng
Fri, 08 Jan 2010 22:33:48 +0000
I updated the wiki with the file transfer throughput between my desktop computer and the netbook.
Chieh Cheng
Sat, 09 Jan 2010 00:21:49 +0000
After performing the throughput tests and updated the wiki page to the current state, I feel there is no practical value in these home throughput tests. Everything plays a factor: CPU, operating system, router rate mode, encryption mode, surfing the Internet, which computer initiates file transfer, which direction the file is being transferred, whether I'm moving from seat to seat (effecting antenna), whether a client adapter works nicely with the router, where is the wireless computer placed. The difference in throughput is so insignificant that it's unlikely you are only going to transfer files using one single performance method over convenience.
Based on this experience, unless you are using wired connection, wireless throughput test outside an artificially controlled lab is pointless. And when are you going to install a controlled lab at home/office? Or operate in your home/office as if it was a controlled lab? In fact, I would take throughput tests posted on various web sites with a grain of salt, or even ignore them all together.
Personally, I think I'm just going to continue to use manufacture published data, convenience, and personal reliability experience to decide which router I use.
Chieh Cheng
Sat, 09 Jan 2010 01:17:57 +0000
There will always be flutuations between your data points. But you
have only collected samples for one router.
To find meaning with the data, you have to collect samples for another
router and compare the results.
My original point was that this Netgear 802.11n router doesn't perform
significantly better than my existing WRT54g when encryption is
enabled (WEP in my case), while the general assumption presume 802.11n should be noticeably faster. Your data points will show whether my
WNR34b is problematic or the product itself is problematic.
gnowk
Sat, 09 Jan 2010 02:24:01 +0000
I added Wireless-G devices to the wiki page. It shows that how you pick your test client and environment makes a big difference in throughput. Perhaps MSi Wind U100 is not the best test candidate.
> There will always be flutuations between your data points. But you
> have only collected samples for one router.
> To find meaning with the data, you have to collect samples for another
> router and compare the results.
Yes. And my point is that with the amount of permutation and with the number of client increasing, the point of doing that (unless performance is absolutely critical) is pointless.
Just look at the amount of permutation I have on the wiki now. I spent 4 hours gathering maybe 10% to 25% of all different permutation (just for one client). And it's showing me completely different results due to external factors.
So let say if you were able to test two different routers, using the same client. The two are likely going to be slower/faster in different areas. You'll have to compare 100% of the data to make a choice, which likely result in insignificant performance difference. Now, you look at the comparison online and they are also no where near capturing 100% test point either. So it's a waste of time.
The manufacturer might come close in a full-on controlled test lab.
> My original point was that this Netgear 802.11n router doesn't perform
> significantly better than my existing WRT54g when encryption is
> enabled (WEP in my case), while the general assumption presume 802.11n
> should be noticeably faster. Your data points will show whether my
> WNR34b is problematic or the product itself is problematic.
Or your environment is problematic.
It's true you can get a ball-park estimate of whether something is wrong with your set-up. But based on my test results, it's unlikely for you to determine whether it is your router, because of the external factors. It could be the test client you are using. Or the amount of metal in your room. Or maybe it's the electricity pole outside of your house. The best you can do is get another router and made a determination whether the Netgear router performs well in your environment or not compared to other routers..
Chieh Cheng
Sat, 09 Jan 2010 03:38:04 +0000
Did you test your router? What's the result?
Chieh Cheng
Mon, 18 Jan 2010 20:13:39 +0000
For my Macbook Pro, I am getting 5-6 Mbps (40-48 mbps) without encryption. I am seeing similar throughput for WPA2 AES.
For my MSI Wind running OSX, I am getting 3.5-4 Mbps (29-32 mbps) without encryption, 2.7 Mbps (22 mbps) with WPA2 AES.
Both of these were downlink throughput from my NAS in the same room.
The reviews show bigger degradation w/ WEP and I suspect that's what I used last time, but I can't get my clients to connect with the AP when it's set to WEP.
gnowk
Wed, 17 Feb 2010 22:25:01 +0000
Hmmmm. I could test MSi Wind OS X without encryption.
I didn't realize MSi Wind OS X supports WPA. I thought WPA was unavailable with the Ralink wireless card. Did you have to install something special? Or did it just work with Mac OS X? I think Mac OS X sees the Ralink driver as a Ethernet card, rather than a Airport card.
Chieh Cheng
Thu, 18 Feb 2010 01:16:07 +0000
I bought a replacement Dell wireless card when I first got the MSI Wind. It works with OSX's native drivers.
gnowk
Thu, 18 Feb 2010 01:49:12 +0000
I tried the Mac OS X test today. But the Finder transfer the file using estimated time without the KB/s. Where is the Activity Monitor that you were talking about?
Chieh Cheng
Thu, 18 Feb 2010 03:03:15 +0000
Look under Application -> Utilties
Spotlight should be able to find it too.
Alternatively, there's something call iStat Menus (menubar resource monitors):
http://www.islayer.com/apps/istatmenus/
gnowk
Thu, 18 Feb 2010 04:13:19 +0000
Thanks. I'll try it when I'm home.
My spotlight has been disabled with Spotlight Manager because of MagicMenu:
http://www.cynosurex.com/Software/MagicMenu/
http://www.cynosurex.com/Software/Spotlight%20Manager/
Chieh Cheng
Thu, 18 Feb 2010 06:12:22 +0000
amazed and scared.
i have my wifi router in my the bottom floor room, with the door closed. i get only one bar on the 2nd floor, understandable, with my linksys. i got my new netgear and i now get 4 to full 5 bars which is great, but all that wireless power it takes to get up here scares me. should i lower the power and live with less signal? almost makes me think i should go wired only. won't help with my neighbors APs i can see though.
charlie
Thu, 18 Feb 2010 08:50:24 +0000
I say you forget about it. Don't bother turning down the power, but
just be prepared for your head to grow three folds like the "grey"
aliens.
The point is even if you turned your power down, everyone else is
still using their technology at full blast. You are going to be
affected by the strong router signal, cell phone signal, HDTV signal,
etc. no matter what. It doesn't matter what you do. Unless you can get
everyone to stop using these wireless devices, you might as well jump
on the bandwagon.
One more thing you can do to get out of this situation . . . move to
rural area of Colorado, Wisconsin, Montana, or somewhere like that.
Chieh Cheng
Thu, 18 Feb 2010 10:02:33 +0000
On different wavelengths over EMFs
gnowk
Thu, 18 Feb 2010 12:21:35 +0000
Add Comment
|
Related Links
|
TrackBack
Related Content
Did your message disappear? Read the Forums FAQ.
Add Comment
Spam Control | * indicates required field
TrackBack
TrackBack only accepted from WebSite-X Suite web sites. Do not submit TrackBacks from other sites.
Send Ping |
No TrackBacks yet. TrackBack can be used to link this thread to your weblog, or link your weblog to this thread. In addition, TrackBack can be used as a form of remote commenting. Rather than posting the comment directly on this thread, you can posts it on your own weblog. Then have your weblog sends a TrackBack ping to the TrackBack URL, so that your post would show up here.
Messages, files, and images copyright by respective owners.