What's new

Asus moving some devices to EOL while issues still remain

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

I guess it's either sell it, or ewaste the inventory.

Okay, sell it for $50 inventory clearance price as EoL product. Not at a premium $190 with thank you and good luck.

Announcing officially EoL in a link most users don't even know exists and continuing salles of the same product on Asus official store doesn't really give grounds for much appreciation. RT-AX68U was officially recommended upgrade model just a few days ago, now it's EoL. In my opinion something to appreciate will be EoL minimum 2 years after EoS. This is what you get with good brands on commercial gear market. You spend the money, but you know when is EoS coming and when EoL so you can plan replacement accordingly and don't end up with a brand new unsupported model.
 
In an ideal world, Broadcom and Qualcomm would actually start trying to cooperate with the OSS community, and provide open source drivers to the community, so that projects like OpenWRT can cover for manufacturers who eventually drop the ball on a perfectly functional piece of hardware.

Qualcomm and MediaTek are engaged with the FOSS community - MediaTek more so that QC... mt76 along with ath9k/10k are out there - QCOM could do better with the 11ax ath11k, and there's still issues with Qualcomm's NSS subsystem (ubi) for flow acceleration - at least there, MediaTek has been opening the kimono there, exposing enough interfaces that the Linux fastpath can work with it.

Broadcom is in an odd place - we have brcm_fmac and brcm_smac drivers out there, but they are more focused on the client station side, much like Intel...

The Broadcom WL drivers likely will never be "open" in the sense of the mac80211 layer inside the kernel - it's not that they can't open up the drivers, it's the interfaces back into the kernel that are problematic...

And the fastpath inside the Broadcom SoC's is still very closely held.

Just my $0.02 worth of opinion here.
 
Obviously combined modem/routers are terrible and I would ask any ISP that provides one of those to put them in bridge mode and invest in a third party router. However, I know many people are happy with whatever their ISP provides and it also shows how little awareness there is among non techy people about the risks involved with these types of devices.

Interesting enough - I've seen a significant change here in the carrier provider Residential Gateways in the last couple of years here in the US.

They changed significantly over the last couple of years - I think this is much due to upgraded requirements and test specifications from CableLabs and the Broadband forums where the operators are active.
 
Broadcom is in an odd place - we have brcm_fmac and brcm_smac drivers out there, but they are more focused on the client station side, much like Intel...

And I would add that in the client space - both desktop and mobile - things are a bit complicated due to Broadcom (Avago) spinoffs - some of the IP went over to Cypress, and that got traded off to to NXP - so it became a bit of a orphan that nobody developed, but had to maintain... and there the focus was more on Android...

Armbian code based on upstream - yes, it's a bit of a *hit*how - the driver there is a mess...

Code:
$ dmesg | grep dhd
[    0.218689] Wifi: bcmdhd_init_wlan_mem: [dhd] STATIC-MSG) bcmdhd_init_wlan_mem : 101.10.361.10 (wlan=r892223-20210623-1)
[    0.221537] Wifi: bcmdhd_init_wlan_mem: [dhd] STATIC-MSG) bcmdhd_init_wlan_mem : prealloc ok: 8162304(7971K)
[    5.318485] [dhd] _dhd_module_init: in Dongle Host Driver, version 101.10.361.24 (wlan=r892223-20220913-1)
               drivers/net/wireless/bcmdhd compiled on Jun  6 2023 at 08:16:57
[    5.320317] [dhd] ANDROID_VERSION = 13
[    5.320818] [dhd] dhd_wlan_init_gpio: WL_HOST_WAKE=-1, oob_irq=71, oob_irq_flags=0x8
[    5.321687] [dhd] dhd_wlan_init_gpio: WL_REG_ON=-1
[    5.322312] [dhd] dhd_wifi_platform_load: Enter
[    5.322910] [dhd] Power-up adapter 'DHD generic adapter'

the WL DHD driver is Android - I see it over in the hobby boards - it's the Android driver, which is a bit of a mess, but it's impossible to migrate the RPi driver over due to licensing issues...

At least RPi get's it right, but you know that Broadcom and the RPi folks are best/favorite friends..

Code:
$ dmesg | grep brcm
[    9.979487] brcmfmac: F1 signature read @0x18000000=0x15264345
[    9.999378] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43455-sdio for chip BCM4345/6
[   10.000385] usbcore: registered new interface driver brcmfmac
[   10.290855] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM4345/6 wl0: Nov  1 2021 00:37:25 version 7.45.241 (1a2f2fa CY) FWID 01-703fd60
 
In these recent cases, I don't know why a number of fairly high end models received very infrequent updates over the last year. It does not match the planned release schedule Asus tsaid they were aiming for a few years ago, so something has changed. I can only speculate. Maybe the plan got derailed between the large number of SKUs they released these past 3 years, and the amount of work involved in maintaining 384, 386, 388 and 102 code branches at the same time over the past 12-18 months. While having a shared common firmware code between all models should make it easier (tho the current heavy branching isn't helping there), maybe the QA is where they can't keep up with so many devices that need testing before a public release is issued?

I would also have expected more frequent updates for all these very recent models. Maybe end users should start asking them for it (which is far more likely to produce a positive result than coming on SNBForums to complain about it, where nobody from Asus will ever see it).


At least that gets you more up-to-date code, as I get the same GPL version for all models. I generally ask for a GPL update every alternate release.


It was easier for you guys tho, since you had like three SKUs to put through QA before a firmware release.

I agree that this is definitely something many customers appreciate. If I were in Asus' shoes, I would have made that a marketing point with their ExpertWifi devices - promise firmware updates on a two months cycle. The same could have been done for the Pro SKUs as well, tho these are really just upgraded variants of already existing models, and not a separate product line.


I also remember when Belkin made a splash by announcing "the new WRT54G is here - the WRT1900AC will get full open source support", only to discover after the announcement that Realtek had yet to provide OSS drivers to the OpenWRT devs. That took quite a few months if I recall.
Well, I can only agree with you here, it's a missed opportunity for Asus and I have zero insight into what's going on. Might take your advice and send their support an email though.
 
I guess it's either sell it, or ewaste the inventory.

Odd that they didn't tie the remaining inventory to their EOL date. Could be that Intel themselves decided to EOL the software stack. Would be another example of where I feel that the SoC supplier is probably often to be blamed with early demise of certain products.
I think this has something to do with Intel selling the entire BU to MaxLinear.
 
Interesting enough - I've seen a significant change here in the carrier provider Residential Gateways in the last couple of years here in the US.

They changed significantly over the last couple of years - I think this is much due to upgraded requirements and test specifications from CableLabs and the Broadband forums where the operators are active.
That's at least good news, as whatever my cable ISP provided in Taiwan was... outdated to say the least.
I also got a crappy old 802.11ac router included with the service I had in my previous place here in Sweden and it was so locked down, you couldn't even change the WiFi channel... It also had about half the range of the GT-AX6000.
 
I think this has something to do with Intel selling the entire BU to MaxLinear.

If I recall, that was only the Lantiq gateway solutions group - client silicon is still in-house with the teams in Oregon and Haifa
 
I think this has something to do with Intel selling the entire BU to MaxLinear.

It had something to do with a weird looking device not too many people liked.
 
Nah, they have a license to use some Atom cores, weirdly enough.

AnyWan was the ex-Lantiq stuff - remember the PUMA mess?

Few years back, I interviewed with them and walked away as there was no clear roadmap as to what Intel was planning to do other that just shipping a product "now"...

For history buffs - Lantiq was spun out of Siemens/Infineon - ran as an independent business for a while, Intel bought them - made a good go of it, not for a lack of effort - but with Intel focusing efforts away on telecom in general, they sold the BU off to MaxLinear, which is likely in my opinion, a better fit for that BU and the sector they're in - MXL is a great company that few folks know about.

TBH - it was never big business for Intel in the larger scope... and talking with a few that were there - lots of politics around having to glue an x86 Atom core on to their existing architecture -

Similar to the 3G/4G Baseband/Modem efforts - Intel bought the modem group from Infineon, and eventually sold most of the group to Apple
 
AnyWan was the ex-Lantiq stuff - remember the PUMA mess?
Yeah, they just changed the name though.
Few years back, I interviewed with them and walked away as there was no clear roadmap as to what Intel was planning to do other that just shipping a product "now"...
Yeah, I was talking to an Intel disty about using one of the upcoming chips in a project and it just got delayed and delayed and...
TBH - it was never big business for Intel in the larger scope... and talking with a few that were there - lots of politics around having to glue an x86 Atom core on to their existing architecture -
At least they sold it, rather than just killing it, like so many other companies do. They made a huge loss though.
$345 million purchase price in 2015 https://en.wikipedia.org/wiki/Lantiq
Sold for $150 million 5 years later... https://www.techpowerup.com/265525/...alifornia-based-maxlinear-for-usd-150-million
 
Anyways - Broadcom is questionable - mainly due to the client silicon swaps across vendors, and the parent company, Avago, is more focused on the cloud...

Qualcomm-Atheros and Mediatek seem more focused, and fierce competitors - from Mobile to WiFi clients and AP's...
 
What's a bit worrying about that list is the AX86U is not listed as a replacement for anything.

RT-AX86U/S as well as RT-AX88U are not manufactured anymore. Replaced by RT-AX86U Pro and RT-AX88U Pro.

Makes me wonder how much time it has left?

Will perhaps last for few more years along with other models locked on 3.0.0.4 Asuswrt 4.0 firmware. Expect fewer firmware releases though.
 
RT-AX86U/S as well as RT-AX88U are not manufactured anymore. Replaced by RT-AX86U Pro and RT-AX88U Pro.



Will perhaps last for few more years along with other models locked on 3.0.0.4 Asuswrt 4.0 firmware. Expect fewer firmware releases though.
Well that was a pretty short manufacturing life vs the AC68U......................
 
Yes, compared to that model, everything is.
 
Well that was a pretty short manufacturing life vs the AC68U...

Not really. Asus just kept going with RT-AC68U as model number, but there were many hardware changes with different radios (C1), different CPUs from 800/1000MHz to 14000MHz (AC68P h/w B2) up to completely different NHD platform (V4). If you noticed later firmware for RT-AC68U was double size - it had images for different hardware.
 
Last edited:

Similar threads

Latest threads

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top