this may be of use - mt7988a-bananapi-bpi-r4-pro.dts
1 Like
Thanks,did not see it yet,but seems they link an AN8831X phy
Updated my branch to add this dts and fix some errors
Am I wrong or the comments on the phy section mention airoha while the phy driver is referencing aeonsemi…
That is what i mean
Bpi write this:
But r4pro dts in mtk-sdk seems to use airoha phy. Till i get my board i ignore comment (lets see which phy is found) and use the AS21010 driver.
Seems the comment is outdated…i see this which seem to match the as21010 phy
1 Like
I just need this thing released! Need to build a router to get rid of this POS ATT rig, and this thing is about as perfect a solution as one can get!
You should wait till software support is done completely. This normally takes at least a year.
to make the confusion about phy used perfect:
[kernel-6.6][mt7988][eth][Add Airoha AN8831X 10G PHY DTSO and driver package]
[Description] Add Airoha AN8831X 10G PHY DTSO and driver package.
but lists autobuild/unified/global/common/files/package/kernel/as21xxx/source/as21xxx.c
defconfig change also only adds this driver and the firmware package
i tried to find a driver for the airoha phy…it looks like both are using same driver
mainline linnux clearly states it as “Aeonsemi AS21XXxX PHY Driver” no note about airoha
so maybe airoha builds same (or compatible) phy
Sometimes SFP+ ports are 10G/1G which means you can insert a SFP+ (10G) or SFP (1G). At least with R4 you could replace the other 10G to install a 2.5G port, I vaguely remember this as a copper port if I am not mistaken…
Are you a member of the Discord group that is dedicated to this. I failed trying to replace it on my 1 Gbps AT&T service…
I tried looking for airoha 10g phys on their docs but didn’t find any referencing an8831x… i started thinking airoha and aeonsemi are the same company but chatgpt says no 
I guess we just ignore the comment line in the DTS as all the rest is about aeonsemi
8311? Yup. Just got my “internet whisperer” last week
I guess that does make sense. I started looking for a board to use, rather than the almost $600 TPlink router, and the Pro popped up as the #1 search result. It looks to be exactly everything I was looking for
Sorry to say that, but this is just a board which is not released yet. And after release, it will still be just a board with an OpenWrt version which is under development for Wi-Fi 7. For me, the best-case scenario is bugless hardware board that will have a good working OpenWrt (open source) version in two years… (my opinion). If you have the time to wait (if you’re not developing it yourself), it may be worth it. If you’re looking for something in the next month… bad day!
Maybe a bit faster as mt7988 itself is mostly upstreamed, phy too and afaik only 10g switch and dts missing. And of course the pcs upstreaming issue we have on R4 too,but openwrt should be faster as 10g switch driver already exists.
1 Like
I totally get what youre saying. Just never really thought that far into it. This venture is a bit deeper into the weeds than Ive ever been before, so I put more energy into finding when I can get my hands on “the perfect product” when I should have been thinking more about what product is actually able to do what I need.
Personally, this is exaclty what I’m looking for! 
It took me several months to get the R4 working as I want to and… I enjoyed the journey.
I learned a lot during the process and I look forward to do it again with the Pro!
I can’t wait to put my hands on it to start working on having it run with vanilla OpenWRT! Ahahah.
@sinovoip… any spare prototype boards? 
I hope so! I’m much more euphemistic about the timeline than I wrote 
.
But what makes me a bit concerned:
I. Wi-Fi range:
At this point in the investigation, it is (not?) clear what the main causes of the limitations are.
- Are they located on the R4?
- Or on the BE14?
For an open-source router, I see a lack of transparency:
- I haven’t seen Simon or anyone else from Banana Pi in this discussion.
- And even if they did write some statements… it’s “just” a forum post, which is not linked to the product and not part of the documentation.
II. Bug handling:
If I wanted to buy the BE14, no one could tell me what condition the board is in, because there’s no unique serial number on it. After the bugs with the R4 and its Wi-Fi board, I would expect an individual number to trace bugs and support recall campaigns. THIS caused reputational and financial damage!
III. Advertising on https://www.banana-pi.org/:
“Banana Pi open source hardware community”
“The development documents, software, and hardware (including schematic diagrams) are all open. The purpose is to allow all developers around the world to participate.”
This advertising attracts people who are looking for fully working open-source hardware and software routers.
Fairly said: On the R4 page, the phrase “development board” does appear.
→ But I think that’s not enough for someone to realize they’re in the wrong place…
The frustration of those who expected a fully working board right from the start had an impact on me as well.
They also provide an OpenWrt image that includes a closed-source driver?
Story of my life: I bought a GL.iNet router because I was interested in OpenWrt. Only later did I learn that there’s a Qualcomm chip inside which will never be fully supported by OpenWrt. Just one more word like “fork” or “not supported” would have been enough to clearly signal a red flag for enthusiasts…
I believe there are ways to present Banana Pi products in a way that everyone understands at first glance what they are intended for!
The End
So, it’s not just about the R4 Pro board itself – it’s about the ecosystem.
There’s a high chance I will buy this router…
but Banana Pi should really provide more information about the board’s power consumption 
.
There are only the first samples ready,and i guess some things will change and not all ist tested in hw yet. I hope i get my board next week and can try out what i have prepared so far.then i can also measure the power consumption (with my simple wattmeter),measure temperatures and such.
Sounds good 
I’m concerned about the 10 GbE chips…
(I was even planning an undercover mission with my hot air gun to desolder those little bastards 
)
But maybe my concerns are unfounded!
Would this issue (Significant issue with its Wi-Fi emission power) will be resolved ? Making sure you're not a bot!
It is written in the text: → “The vendor informed me that the problem is not with the antennas, but with the router’s BE14 chip.”
The BE14 is already in production, so the answer is no.
However, the person who wrote that text did not mention the noise issue, which comes on top of the existing “possibilities”. Additionally, there’s something else in the text that isn’t quite right:
→ “There is no other Wi-Fi 7 module at present.”
In fact, the AW7990 from AsiaRF does exist — but the driver looks for basic support:
→ https://patches.linaro.org/project/linux-wireless/list/?series=265107
1 Like