I'm interested in a crowd-funded router built with OpenWrt recommended chips, branded with an approved OpenWrt logo and preloaded with the latest OpenWrt release. The cost would include a 20% OpenWrt project donation, shipping and 4 routers to the developers who agree to support/maintain the build.
Even if I wanted an OpenWrt branded device (no, I really don't want to put that burden[0] on project and developers), mt7621a+mt7603e+mt7612e (in the best case, worse if the stated mt7620a SOC would be true) would not be on my wish list[1], even less looking 6-24 months into the future.
--
[0] and it is a burden, even if the financial side of it would be covered, the project's reputation and legal status is still affected by such an endeavour; mt76's support for mt7603e doesn't magically get better, just because you put an OpenWrt sticker on the device.
[1] that would more look like Xiaomi AX3600, Xiaomi AX1800 or Redmi AX1800 (so ipq807x based), respectively something based on mt7622+mt7915e+mt7915e.
I would suggest that is a lot of work and risk for little benefit.
I suggest if you find a router that you wish was supported then do some research and ask questions here if someone can/would like to port openwrt for that device.
Buy one for yourself and if your feeling generous for the developers that offered to help with the port
Once it is confirmed to be working add it to TOH etc.
Recommended it on the forum .
With regards to donations.
Design some openwrt stickers or good quality labels and sell the stickers donate the proceeds of the stickers to openwrt.
Now anyone can place the sticker on their favourite router , AP fridge whatever.
The developers take on that burden with each device that it committed. What would be different about this is that the developers are also generating financial support (servers, meeting support, etc). The developers reputation is on the line with each release.
Likely less work as they start with known specifications (ones they picked) and more benefit as it generates financial support for the project.
Can you expand on "different level of support"? The project already responds to security issues and bugs. And to reiterate: the hardware is supported by the manufacturer.
'purpose built' you say... well... i'd kinda like to have some superman undies... i'm sure a few of my mates would like them too... let's do a bulk order then... choose the 'best' feasable qualities out of what's contemporary... and we'll have something capable, practical and proven.
order time... few are keen... why...;
some wanted boxers
some wanted lycra
some changed their minds and prefer gummibears
half couldn't get in their size / color / time frame
returning them for warranty / alteration is impractical
too specific nobody is interested... not specific enough it strays from the original intent.
seriously though... if you are serious...;
adopt the disney model
project gives a [tick] sticker for 2% earnings from vendors
members get a discount
vendors vendify and developers develop
and the only thing that really makes something like this appeal to me is being able to get say ipq807x with open / well supported nss blobs... or franken sbc rpi4+8port-managed-switch+2x10GbE like results. point is... without an 'edge'... you don't really have the product differentiation to succeed.
Well, if we are going to be semi-realistic, then "purpose built" means "adapted from some SoC reference design".
Which leads us back to the numerous "which is the currently recommended device for OpenWrt" threads. AFICS, none of these have come to an absolute answer yet. There have never been a SoC with 100% pefect open source score. There have been, and still are a few nice, candidates. But my claim is that there will never be a perfect SoC as long as there are no SoC vendors actively participating in OpenWrt development. My proposal is that no hardware design is blessed with the "OpenWrt" name unless the SoC vendors BSP/SDK is identical to the current OpenWrt main branch.
Taking some OpenWrt snapshot and doing in-house development on it for years just doesn't cut it, as Qualcomm so nicely have demonstrated. There is still an infinite amount of work left before those SoCs can be considered supported in OpenWrt.
And depeloping gigantic drivers completly outside the Linux and OpenWrt communities leaves even more work to be done, as Mediatek still shows us.
And taking just one design and opening up some of the drivers, without any documentation of the vast number of firmware interfaces at all, will not magically create any community. Offering free hardware to developers makes absolutely no difference under such conditions, as Marvell have shown us.
Having support for old designs without following up with new designs doesn't do much good either, as Intel has shown.
There really isn't much to say about Broadcom, except it's a wonder they till exist. I have no idea why, but there isn't a chance in hell it's going to last so there is no point in considering them for any future OpenWrt work
When your looking at HW projects like this you would also be hard pressed to have enough information of the costs involved before setting an initial target, for example if you said (for example) you waned to mirror the R7800 in specifications how much would that cost ? do you get a fully customised PCB manufactured (expensive) or "buy" an off the shelf product from an ODM (would almost 100% lead to compromises)
And thats before you have to contend with vendor BSP's and how they integrate with OpenWRT mainline (if at all), if you choose to not use things like NSS on qcom (just so you keep totally clean OpenWRT mainline) the device will perform worse than the next manufacturer over.
EDIT : I would actually say that instead of making a "OpenWrt purpose built router" it would be better marketing such a project as some "Open Router Dev Kit" that would ideally cost in line with other routers, but is designed to allow you to easily allow building / flashing different images / OS's (like OpenWRT / QSDK in this example / Something totally custom). So support becomes less of an issue (as its on a consumer device off the bat) but would work for anyone wanting to get into router development.
If there is interest, I will. I also know I can't do it alone. It will be key to get approval from the project, input from the developers and negotiate (someone fluent in Chinese) a solid contract with the manufacturer. The GoFundMe type site needs to have a history of transparent accounting and honesty. Also need to work out distribution of the device.
This is kinda what I had in mind although I wanted to avoid bleeding edge hardware that increases the risk of a poorly performing device. The development hardware should be flexible enough to be flashed with stable OpenWrt firmware for day-to-day use or a master git pull for development. There should be enough flash/ram that the device will not be obsolete in the near future. I'm also open, based on demand/developer recommendations and the availability of open source support without non-disclosure agreements, to consider Qualcomm.
The first step in all this is gauging interest. If the interest is not there, no need to bother the project/developers, search for a reliable site to host the project and work with a manufacturer.
if you want a sdk that is well tied with openwrt qualcomm is the way to go... the only problem is that they are stuck at an old kernel. Aside from that if someone wants to really work on it he can just port the needed code and package to the new kernel and you are done (like we are doing for the nss project)
Hmm ... I don't see any benefit. Let's say you pick ~3 routers at a price range between ~80-150$/€. Also let assume that this 3 candidates are currently well supported by OpenWrt. You have exactly the same discussions like in the according threads about which devices are the best for a given budget/requirements. At the end it will be hard to find the "golden device" for everybody. So an extra picked device by the "community/OpenWrt-developers choice" would not make any difference on specific requirements for "casual" users (in fact they are not casual because they are installing OpenWrt). At the end you would make it "only" for the enthusiast/nerds here.
The only way I would go or I can imagine is if OpenWrt is choosing some devices which they sell pre-flashed with OpenWrt and maybe with a OpenWrt sticker like "driven by OpenWrt". A certain amount of money will be donated to the OpenWrt project. So ppl. have a bit of a choice.
If I look around there are a lot of ppl. offering such devices and they are making money with that. And if I remember correct DD-Wrt did something like that in the past. I don't know if they are still selling pre-flashed devices. But in the past they had only a few hand picked devices. Maybe in future it is possible to find a device manufacturer which is interested on a partnership and we get the "golden device".
But that is not my decission and I would never force nor expect ppl. driving this project in their spare time without getting money to make those decissions which would in fact mean investing more time for things which (maybe) make no fun in spare time. Like creating a online-shop, setting up a billing system, doing marketing, responsibility, pressure, etc.
I was thinking to set up the infrastructure for a single production run and was leaning toward cheaper devices initially. If it works, then another run could be set up incorporating advances in hardware or those interested in higher performance devices. The first run will determine if it is viable and a cheaper device lessens the risk.
OpenWrt would provide the firmware for the manufacturer's initial flash. The boot loader would not need any special hacks (see the Xiaomi 4a thread) for the purchaser to update to a subsequent release or ^master.
I don't want to be the opposition guy nor the pessimist here. I appriciate new ideas/thinking.
Just an example: If I look around here in Germany most people just using limited ISP modem and sometimes an additional router for extending their WLAN/Wifi. Advanced users (without deeper knowledge) using FritzBox mostly (AVM is the market leader here). Those users would (IMO) be the target for such a project. If you compare FritzBox to OpenWrt I would miss at least a Guest Wifi with just a on/off button/click and a "Kidsafe". And if I compare OpenWrt to FritzOS I see some features which would advanced user want to have nowdays which are not supported by OpenWrt out of the Box yet (e. g. Mesh with a few clicks (I know this is not true; it is more complicated but it is working within FritzOS ecosystem; it would be an option for a common hardware base for OpenWrt), monitoring/administrating SmartHome devices).
What I want to say is you have to do at least a bit market research (not only in terms of hardware). This forum is (IMO) not representative. You should know what functions people want/expect and what they would pay for that first. Overall I have the impression that most people here are asking questions about older devices which they have already. Which mean they are not buying a new device. It's more like extend the usage of the old device. Those people usually don't have deeper knowledge about OpenWrt. But maybe there is already a statistic for which devices/chipsets firmware is downloaded mostly? Define the devices/chipsets what are you aim for and cummulate the amount of downloaded firmwares to see how large is your potential market. For this I would choose the downloads after a stable release. Maybe within the first month?
I cannot elaborate more/deeper ... It's simply too much to say/think about for such a project.
IMHO those stats are not representative.
They contain downloads by bots and double downloads by users.
Do not trust any statistics that you havn't forged yourself.
I don't really need another router currently, nor would I be using as primary router.
However I would be happy to have such an option and I would chip in to support the project.