WNDR 3500L DD-WRT Review

[DaveMcLain]

Great article Tim and I've read in other places how DD-WRT tends to slow down routing performance when compared to stock(slimmer) firmwares. I use DD-WRT a lot on all sorts of generally very cheap hardware not for performance but because it allows me to do new things with old hardware. Being able to do client bridging is a very cool feature. For instance right now at home I have a Netgear WGR614 V8 running as a wireless bridge for my Blu-ray player. I wonder how the wireless performance compares when used in that mode(where the box is not doing any form of routing).

Also I looked on "Myopenrouter.com" and they say that a version of Tomato firmware is also available for the WNDR 3500L. I've heard that firmware is lighter than DD-WRT could you give that a try and see how it seems to run on the review box?

 

[dragonrage]

Well, I assume Tim isn't going to want to get into testing everyone's favorite firmwares since that would end up being a lot of work for him. I do like Tomato and I used to run it over DD-WRT on my WRT54g v4. I also like OpenWRT (with X-WRT for a web UI). I didn't realize that DD-WRT had such poor performance before Tim's review, though I was already upset that they sell DD-WRT for some routers even though the code they use is GPL and they are legally required to release the source for everything they modify. Though I have used it in the past.

 

[YeOldeStonecat]

I also prefer Tomato over DD, seat of the pants performance seems snappier, the web interface is definitely faster, and when you make/apply changes...Tomato doesn't require a full reboot...where as DD requires a reboot every time you even look in its direction.

I've preferred Tomato after comparing the two on over a dozen various routers 'n APs I've used at home.

 

[thiggins]

I might try Tomato if I get a chance. Probably just for routing, though, since the wireless testing takes much more time.

Interesting that there has been no comment from NETGEAR on my findings.

 

[dragonrage]

Interesting that there has been no comment from NETGEAR on my findings.

I don't think I'd agree with them "wanting" you to run alternate firmware with this box. Rather, I think they just figured they'd make a version that should get sales from the people who want to. "Let them run whatever they want as long as we can find a way to get them to buy our product," or something like that.

 

[thiggins]

I don't think I'd agree with them "wanting" you to run alternate firmware with this box.

If they didn't want to encourage people to use alternative firmware, why would they pay to run a whole website devoted to it?

 

[dragonrage]

If they didn't want to encourage people to use alternative firmware, why would they pay to run a whole website devoted to it?

Marketing for the "L" routers to draw people who would have gone to a different brand. I don't know. I just can't imagine them being too concerned about this as long as it doesn't hurt sales. But maybe it will start hurting sales after your review and maybe they will see that and get concerned eventually.

 

[DaveMcLain]

A few tests of some of DD-WRT's other functions would be very interesting too. I'd like to see how well it functions as a wireless client device when compared to say a USB wireless adapter. Does it cost any speed at all when running through the box with DD-WRT to a built in Ethernet adapter on a PC when connecting wirelessly to an AP?

Being able to tap into these other "modes" is what makes DD-WRT very useful for me.

 

[thiggins]

A few tests of some of DD-WRT's other functions would be very interesting too. I'd like to see how well it functions as a wireless client device when compared to say a USB wireless adapter. Does it cost any speed at all when running through the box with DD-WRT to a built in Ethernet adapter on a PC when connecting wirelessly to an AP?

Being able to tap into these other "modes" is what makes DD-WRT very useful for me.

That could be worth a look.

 

[db9]

I know this is an old thread, but ...

Not sure if this thread is about DD-WRT or the 3500L but I'll ask..

I'd love to have one or two 'extra' features available in WRT (open or DD) (I would like to be able to schedule WiFi on/off on my 3700v2 - but the through put performance of the 'standard'

What is it about the DD-WRT code that makes the performance slower than the 'standard' firmware? Is it the 'bulk' or the number of features within DD-WRT causing the slow down or is it the drivers...
Just wondering ..

 

[thiggins]

What is it about the DD-WRT code that makes the performance slower than the 'standard' firmware? Is it the 'bulk' or the number of features within DD-WRT causing the slow down or is it the drivers...

I don't really know.
But many DD-WRT releases have come and gone since the review, so perhaps they have fixed the throughput issue.

 

[deVries]

Before doing the review did you confirm in the Peacock Thread which version of dd-wrt to use? If you did not, then it is very likely you used a beta version that is very buggy, etc. It seems many beta test versions are constantly being used unwittingly, because these beta versions are listed in the database to use for x-model.

You needed to have used a PRE-SP2 version. See ******* below:

RECOMMENDED BUILDS update 12/24/11
The date has been updated to inform you that this is still the current recommended build. The links have been updated to reflect the new directory structure.

NOTE: If you are running K26 builds and have problems with Repeater mode (bridged is fine) please see/update this ticket.

BrainSlayer builds (14929):
ftp://dd-wrt.com/others/eko/BrainSlayer-V24-preSP2/2010/08-12-10-r14929/broadcom/


*******For models that REQUIRE the K26 build (like the WRT610nV2, WNR3500L, RT-N16...etc...see note above)
ftp://dd-wrt.com/others/eko/BrainSlayer-V24-preSP2/2010/08-12-10-r14929/broadcom_K26/*******

or if you need Eko VINT or other special Eko builds:
ftp://dd-wrt.com/others/eko/V24_TNG/svn15230/


If you already have dd-wrt installed on your unit you can update to the Pre-SP2 build through the web gui or using the tftp method (found in the peacock sticky linked below)...If you have a virgin unit (OEM Firmware)...consult the wiki for proper flashing instructions for your model.

For proper methods for flashing and for file information about NEWD and VINT, see the Peacock Sticky at the top of this forum...or follow this link
Actually, that is a GREAT place to start when you are looking for most answers.


If you still have questions...use the Search button and consult the FAQ in the Wiki....if you still haven't found the answer...feel free to ask on the forum.

 

[deVries]

I don't really know.
But many DD-WRT releases have come and gone since the review, so perhaps they have fixed the throughput issue.

The test needed to have been run on a PRE-SP2 version of the software. Your image in the article shows SP2 ... IF that was not PRE-SP2, then you used a buggy BETA test version of the software.

The correct build to have used is 14929 K26

It is admitted K26 is not a stable as build 24 for older slower CPU routers. The above build number is what should only be used for actual use ...not beta. A higher number is beta.

This is the correct version:

ftp://dd-wrt.com/others/eko/BrainSlayer-V24-preSP2/2010/08-12-10-r14929/broadcom_K26/