Why not WNDR3700L instead?

12 posts / 0 new
Last post
PaulC
PaulC's picture
Why not WNDR3700L instead?

Given the audience that an open-source labeled product is aimed at, why choose the WNR3500 as the platform instead of the more capable (& faster) WNDR3700?

Is it because of the Atheros vs. Broadcom chipsets? Given that the Atheros folks seem to be more open source friendly than Broadcom, wouldn't an Atheros-based router make more sense for an open source platform than a Broadcom-based one?

It would also avoid the grief Netgear is getting because of Broadcom's poor open source policies...

-- Paul

PaulC
PaulC's picture
I do 'know my way round a bit

I do 'know my way round a bit', so that looks encouraging (I develop embedded linux devices for a living). Though it is a shame that it's not a more recent version of OpenWRT.

As far as the binary packages go, it bothers me personally much less than some in the community, so long as it's things I'm unlikely to want to modify (e.g. wireless baseband firmware). Plus some of those listed in the thread look like they're either minor (LED control) or not really 'binary' just 'not built from source', e.g. samba-scripts.

In my professional life, I've interacted with both Atheros and
Broadcom (though ended up going with another vendor). I got the clear
impression that Atheros understood the open source community, and saw
the synergies (within their legal and business constraints), Broadcom
not so much. Somehow it seems ill-advised for the community to be
favoring a silicon vendor that doesn't recipricate. And nothing sends a
clearer message than voting with your wallet :)

But the real point of my post was a question for Netgear - why put your resources into producing a '-L' version of the 3500, when I'm sure most card-carrying geeks would rather be playing with the 3700?

-- Paul

PaulC
PaulC's picture
@Som Pal Choudhury:

@Som Pal Choudhury: Appreciate you taking the time to reply. It's great to see your involvement here.

Obviously your understanding of the market you're targeting would be way better than mine :)  though I'm not so sure the OSS enthusiast segment is very price-sensitive for smallish deltas in price ($20-40 on a ~$140 product) when the underlying hardware has clear advantages. Though if the hope is that the WNR3500L will eventually dip below the 'pyschological barrier' of $100, then it makes more sense to me.

And the competition point is well-taken. On the other hand, sometimes it's better to strike at a segment that's underserved, with weak/little competition.

I'd love to contribute to WNDR3700 support, though having just joined an early-stage startup, I'm not sure how much spare time I'll have. Not much, I suspect...

When all's said and done, I don't mean to sound critical - I'm very glad that Netgear is engaging with the community. I still have my KWGR614, and wish that had enjoyed the same success.

@Borromini: Could be that those modules are not modified. As I understand it, the GPL only requires that modifications are made available in source form on request, so if those are unmodified, they're not violating the GPL by not providing the source. While it's good practice, there's no legal obligation. On the other hand, the 'wndr3700' prefix does imply that they're specific to the wndr3700 somehow...

I've not looked at OpenWRT that closely, but it's common practice for the kernel to have the module version check turned on, so it will only load modules compiled for the exact same kernel version.

Frankly I've no idea why anyone would want to use 2.4 for new product development. 2.6.18+ has so many advantages over 2.4 for embedded development.

Traek
Traek's picture
@Som Pal:

@Som Pal:

I completely agree that many more people are looking into using open source firmware on the WNDR3700. When the WNDR3300 came out, CapnBry (on the DD-WRT forums) did some serious work and had to jump through several hoops to get the open source firmware to pass CRC and hack through a ton of code to get it working.

While I certainly applaud Netgear for engaging the community on the WNR3500L, I would think a better approach would be to document (and make available) more code for the 3700 and subsequent Linux-based firmware routers and certainly make the process less painful for upgrading to 3rd-party, open source firmware.

You said:

It does not mean WNDR3700 will never be enabled with DD-WRT and Open-WRT. I believe work is already ongoing by the different firmware communities in this regard. These firmwares will be posted on the MyopenRouter.com community site when available.

It seems if you're willing to post 3rd party firmware here, you'd have already posted (unless I just can't find where you've have it posted) the DD-WRT firmware for the WNDR3300, though I did see references to Tomato, DD-WRT and other open source firmware projects. My guess would be you haven't posted it since you can't simply upgrade... it requires a lot of non-friendly (though arguable with the given community) steps to accomplish.

How about this: release a firmware that AT LEAST opens up the WNDR3700 to 3rd party firmware (skipping the hacks and TFTP requirements so the WebUI can be used to upgrade) as well as provide a dump of that firmware so it can be modified more easily by the community. You'll see sales increase on a higher-end router as -- like I said before -- there is a LOT of interest in this router if it can be opened up by the amazing features of some very mature firmware projects out there.

All that said, I love the WNDR3700's performance and would love help bring this product to open source. I just think Netgear needs to engage the community more instead of passively waiting for someone to hack at it and selfishly post their results in myopenrouter.com when they did nothing to help get it there (other than create the router in the first place). Let's see how Netgear can really engage the community, now!

Here's hoping to the beginning of a great relationship...

Rosenstand
Rosenstand's picture
Another

Another

Rosenstand
Rosenstand's picture
Another "me too." Was just

Another "me too." Was just about to purchase a 3500L when I saw the initial press, but after seeing that it was Broadcom-based and hence would lock me into the ancient and unmaintained 2.4 kernels I quickly turned away.

Had my eyes out for the 3700 for a long time - great looks, powerful hardware, and a much better candidate for an "open platform" due to the Atheros chipset. Unfortunately it's not supported by OpenWrt and it will surely take a long time since none of the developers have access to one.

Instead I went for a Linksys WRT160NL which seems very popular with the OpenWrt developers. Had the 3700 been supported, I would clearly have chosen that due to the (even) better hardware. Price is not really a concern in this (open source enthusiast) market - our software is free so we've got much more money to spend on nice hardware :-)

Traek
Traek's picture
Rosenstand said:

Rosenstand said:

Instead I went for a Linksys WRT160NL which seems very popular with the OpenWrt developers. Had the 3700 been supported, I would clearly have chosen that due to the (even) better hardware. Price is not really a concern in this (open source enthusiast) market - our software is free so we've got much more money to spend on nice hardware :-)

Can't agree with you more. In fact, I almost bought the 160NL for the same reason but thought I might luck out and see some progress on OpenWRT, DD-WRT, Tomato or SOMEthing else. I have a few weeks left on my receipt and if it doesn't look like true open source firmware is on the immediate horizon, I'm going to have to ditch the Netgear as well in favor of that Linksys.

Trust me, I have no love for Linksys and would just as soon gouge my eyes out. Still, it's never been their hardware that I couldn't stand, but their firmware (which isn't a problem with these open source firmwares). Hopefully people at Netgear are looking at this thread and step up to the plate for the community.

Dragon
Dragon's picture
Theres been some slow

Theres been some slow progress on openwrt on the 3700 in openwrt forums, a couple of people are now poking at it and slowly getting things working on it ( kernel, some lan functionality, leds etc ), though nowhere near any usable state yet, only just got wireless at all active it seems, but apparently not working very well yet. Still, looks like there's some hope.

https://forum.openwrt.org/viewtopic.php?id=22311

metamatt
metamatt's picture
Yeah, I'd really like to see

Yeah, I'd really like to see Netgear actively supporting this too.

PerS
PerS's picture
fyi: http://wiki.openwrt.org
Grexe
Grexe's picture
great! now compiling...:=)

great! now compiling...:=)

Jamby
Jamby's picture
I'm the network admin for a

I'm the network admin for a coffee company doing importing, and roasting of coffee. We also have a successful, busy, coffee shop. I am using the Netgear WDNR3700 for both our office, and a guest network for customers. It works great, with changes to many settings. I would like to migrate from the stock firmware to more powerful open source. Netgear recommended the 3500L, which I will not move to. Any updates on the OpenWRT, DD-WRT or Tomato experience with the 3700?