[ 386.4 alpha Build(s) ] Testing available build(s)

[jsbeddow]

Pros and cons to dropping the AC in favour of AX - the AC stuff would be forced to become nodes in AiMesh configs as they'd be feature-frozen (until...), AX stuff could be focused on as you've said...hmmm.
with @L&LD 's glowing report and whole-hearted endorsement of the AX68, this could be the lateral move while we wait...
the AX86 is sold out at my local retailers, and the $300+ Canuckistan kopeks to obtain one is a bit dear for me at the moment IF one were available (what's the difference between the 86u and 86s?)...the ax68 is only 60-70 kopeks less, still a reach but stretchable...
I will wait. there are only 3 AX-enabled devices (2 iPhones and an asus laptop) on my network at this time

Yes, the choices will be hard. I have been tempted to pick up an RT-AX86U, and they have been available in my nearby BestBuy, but...they also just went up about 12% in price about two weeks ago ($250 to $280 USD now). Not a lot in the grand scheme of things, but annoying for an item I am on the fence about to begin with, as it isn't a true "need" at the moment. My RT-AC86U is still working fine, but is approaching three years old, and is a model known to suffer from sudden/premature death.

Sorry for the off topic rant...

 

[hancox]

Again, everyone drawing conclusions.

Why must it be broadcom? There is another tied-in vendor, who has already caused issues with this forum/build/etc, that nobody is mentioning.

Why not TrendMicro? Honestly, it doesn't really matter, but let's not all assume it's Broadcom at fault, when we really don't know.

 

[sorachan]

One thing that confuses me - isn't ASUS firmware based on an open source project, and therefore requires ASUS to release a GPL that tracks with the actual firmware release (although potentially time shifted)?

One encouraging thing I see is that many of the more current routers are now on the same firmware release.

A comment I'll make is that I'm in favor of reducing the number of supported routers if that would make it possible to continue the project. I'm sure a lot of people wouldn't like this, but this is a freeware project, and RMerlin's time is limited. Could drop everything but AX routers.

Because ASUS GPL included closed source SDK from Broadcom.

Broadcom complained to ASUS, so ASUS deleted all the source code archives. Merlin actually also deleted some his source codes probably also to satisfy Broadcom.

Unless Broadcom is fully satisfied you won't see any GPL to release any time soon.

As you can see, there are many open source firmwares such as OpenWrt, and so on, none of them support any new router because all wifi chips need closed source SDK. There is nothing to support if the chip manufacturers don't want to share them.

 

[sorachan]

Again, everyone drawing conclusions.

Why must it be broadcom? There is another tied-in vendor, who has already caused issues with this forum/build/etc, that nobody is mentioning.

Why not TrendMicro? Honestly, it doesn't really matter, but let's not all assume it's Broadcom at fault, when we really don't know.

This time it is Broadcom, not TrendMicro.

------

Before you ask why, I saw Broadcom SDK rumor back in August. That has now long been proved, no longer a rumor.

 

[elorimer]

(what's the difference between the 86u and 86s?)..

2 core vs 4 core; 512mb vs 1gb memory; no 2.5mbs LAN/WAN port; USB3+USB2 rather than 2USB3.

 

[capncybo]

Because ASUS GPL included closed source SDK from Broadcom.

Broadcom complained to ASUS, so ASUS deleted all the source code archives. Merlin actually also deleted some his source codes probably also to satisfy Broadcom.

Unless Broadcom is fully satisfied you won't see any GPL to release any time soon.

As you can see, there are many open source firmwares such as OpenWrt, and so on, none of them support any new router because all wifi chips need closed source SDK. There is nothing to support if the chip manufacturers don't want to share them.

Indeed this does seem to be a slippery slope.
In the name of security & (to my limited understanding) recently mandated by certain countries governments...
The latest & greatest WiFi technology does seem to be closed source.
From a technology standpoint Intellectual Property usually starts of as protected IP.
And IMO from a hacking standpoint, open source code is most likely easier to defeat or circumvent.
So I suppose from a router standpoint...
What's most important to you, FEATURES, SPEED, SECURITY ???
Open Source Code probably helps improve features.
Speed is likely where the BIG $$$$ IS.
And serious arguments can be tied to: Maintaining Security via Closed Source.
Hmmmmmm

 

[sorachan]

Indeed this does seem to be a slippery slope.
In the name of security & (to my limited understanding) recently mandated by certain countries governments...
The latest & greatest WiFi technology does seem to be closed source.
From a technology standpoint Intellectual Property usually starts of as protected IP.
And IMO from a hacking standpoint, open source code is most likely easier to defeat or circumvent.
So I suppose from a router standpoint...
What's most important to you, FEATURES, SPEED, SECURITY ???
Open Source Code probably helps improve features.
Speed is likely where the BIG $$$$ IS.
And serious arguments can be tied to: Maintaining Security via Closed Source.
Hmmmmmm

Well, whatever the opinions you and me have, they have absolutely zero impact on chip manufacturers business model and decisions.

There is no wifi chip supplier providing new chip SDK to public, even the eventually released open source SDK for older chips are not as stable as closed source firmwares, probably intentionally.

I doubt this will ever change, otherwise US and China would not need to start a trade war partly based on telecom chip supremacy.

 

[SwampKracker]

Also, drawing the line at AX routers was just a starting point for discussion. Perhaps there is zero benefit to RMerlin by dropping any of the currently supported routers, although I think the AC88U/AC3100 and RT-AC5300 are effectively EOL devices. There are many better alternatives out there, and I would drop those if there was any hassle with keeping those supported.

Which is unfortunate. I may not get much use from my RT-AC3100 lately, but it is a perfectly good, 4x4 access point.

 

[princi]

Quite frankly, I'm about to give up there. This has been going on for over 4 months now, with the last official GPL being from last spring. I asked once again yesterday for an update following the last communication I had received three weeks ago that was another "almost there, but still more work to do", and got another "still more work to do", without any form of ETA. So, I told Asus that this had become seriously problematic, as I can't expect all users to wait 2-3 months every time to get access to publicly revealed security fixes. I haven't heard back from them yet.

So yeah, there is a distinct possibility that this project might have to get shelved indefinitely at this point unless something changes. I'm back at the same situation that I was last year when the 384 releases were all over the place, with some models not getting updated code for a long time. This project simply cannot survive without updated code from Asus, as too much of the code relies on their closed source portions.

I agree. Don’t let them treat you that way.

You have done amazing work in the past, but nothing lasts forever.

It goes without saying that your work, and what you created, won’t be forgotten.

 

[Treadler]

How is fttp I'm still waiting on a planned upgrade for my area, I'm on fttn my ping for the most part is ok, the only upside is that I've got an old plan with Optus so the price is relatively affordable.

FTTP has been bulletproof reliable. The only outages have been network wide ones. Very infrequent.
An ISP that is generous with the dreaded CVC provisioning is vital to avoid peak time slowdowns though.
A bit of experimentation was required in the early days. 4th ISP was the charm…..

 

[Vexira]

FTTP has been bulletproof reliable. The only outages have been network wide ones. Very infrequent.
An ISP that is generous with the dreaded CVC provisioning is vital to avoid peak time slowdowns though.
A bit of experimentation was required in the early days. 4th ISP was the charm…..

Impressive, I'm looking forward to it I'm still sitting on my hands till it happens to be upgraded day and I'm excited to trim my ax88u through its paces.

 

[Wisiwyg]

Quite frankly, I'm about to give up there. This has been going on for over 4 months now,

Don't sweat it, @RMerlin . I know that's easy for me to say b/c I don't have all the impatient users pinging me all the time.

You've been very clear - this issue lies with Asus and you have no control over it.

That should be enough and people should have faith, after everything you've done, to sit back and trust that when you have something ready you'll put it out there. So don't let them put the weight of this on your shoulders. They're broad enough to carry it as you have, but lighten the load. Many of us recognize your contribution, especially those like me that couldn't even begin to do it for ourselves. Whenever you release a new version you make my day, litterally. And I enjoy the extra features every time I think about logging onto the console.

So ignore the noise, you are appreciated for what you do for everyone. Anytime I talk with someone about wifi routers I recommend an Asus and in the same breath tell them to come here and load the latest Merlin.

 

[jj22038]

I worked for a mid-size company that got a very large customer. That customer demanded the GPL bits (as is their right under the GPL). There were two important aspects: conforming to the GPL (proprietary code had to be strictly separated from GPL code), and satisfying the very large customer's very large set of lawyers. They had a list of "GPL" packages that weren't really free, and we had remove them from our product. We couldn't ship until that was done.

Wrt the chip vendors' SDKs: chips have become very cheap. A company such as Broadcom can no longer pay development costs strictly from chip sales. The high cost of an SDK is a way to make up for this. Economics has driven proprietary code.

 

[thecheapseats]

so... how's that asus stock factory firmware lookin' these days?... /sarc

 

[dave14305]

The problem is that GPL releases are all-or-nothing. They were asked to pull off existing archives due to licensing reasons, and they can't publish any further GPL archive until all licensing issues are resolved to the satisfaction of the complainee. So far, every time Asus thinks that they have sorted out all issues, they are told that there are still unresolved issues, so they have to go back to it to address them. Until all issues are resolved to the satisfaction of that other party, they can't release anything.

What are the chances that future GPLs won’t even be “buildable” by you after the cleanup is done to the complainant’s satisfaction?

 

[XIII]

so... how's that asus stock factory firmware lookin' these days?... /sarc

Quite afraid of that question…

What if there’s a CVE fix that is so serious that we have to revert to stock?

 

[shabbs]

What if there’s a CVE fix that is so serious that we have to revert to stock?

Yeah, that's the one scenario that does worry me. A major exploit that needs to be fixed and our only option is to revert to stock.

 

[laracroftonline]

I don't think the decision is hard. I'm thinking about this in terms of: 1) how RMerlin could cut down on his time, and 2) what routers are most likely to be part of a GPL release. Assuming that ASUS begins to release GPLs again (I think they will), these are two key barriers to keeping the project alive. I get that the AC68U and AC86U are popular (these are what I have), but I think the reality is that if dropping AC-support ensured survival of the project, there would be a big increase in AX downloads.

Also, drawing the line at AX routers was just a starting point for discussion. Perhaps there is zero benefit to RMerlin by dropping any of the currently supported routers, although I think the AC88U/AC3100 and RT-AC5300 are effectively EOL devices. There are many better alternatives out there, and I would drop those if there was any hassle with keeping those supported.

So you want to eol devices in this time of huge chip shortages,

 

[sorachan]

What are the chances that future GPLs won’t even be “buildable” by you after the cleanup is done to the complainant’s satisfaction?

I don't know whether anyone knows the answer, but my understanding is that the upstream is not happy to see their SDK released to public, thus people can even port ASUS AX router firmwares to Netgear models such as RAX50 and RAX80. If you build a router yourself using the same hardware, I guess you can port ASUS firmware to your self-made router as well.

Obviously, router manufactuers benefit from this due to increasing sales, so it is in their interest to keep things as is.

 

[Wadadli]

I don't know whether anyone knows the answer, but my understanding is that the upstream is not happy to see their SDK released to public, thus people can even port ASUS AX router firmwares to Netgear models such as RAX50 and RAX80. If you build a router yourself using the same hardware, I guess you can port ASUS firmware to your self-made router as well.

Obviously, router manufactuers benefit from this due to increasing sales, so it is in their interest to keep things as is.

Stand by for the wrath of @thiggins !