Phantomski
Regular Contributor
Apologies for bit of a clickbaity headline. I was looking into new models for myself, also friend asked me for a recommendation so I was trying to figure out if the one I have (AX88U Pro) is still a valid recommendation and it got me through all sorts of rabbit holes and spreadsheeting 
But mainly it got me thinking. I know it's difficult to say and there are many factors at play, but overall, in more of a broad brush strokes - what goes into choosing which model makes it so to speak onto the Merlin's list of supported models? Or even in more general sense, what makes it onto Asus internal list (well apart from 'it sells well')?
While it might sound like a stupid question (I was always told there's no such thing as a stupid question
), just please hear me out. I'm genuinely interested in the process of selecting which particular router is worth investing time in for Asus, for Merlin and ultimately, what makes it the model that is worth recommending and stands the test of time. I know, consumer routers, 2-3 years, 5 if you're lucky. Who cares, buy one, next year will be different. I get it. But still, there are reasons why some specific routers (AC68U, AC86U, AX86U, AX88U Pro, ...) are better choice and more preferred by people here than others. Why they got on Merlin's desk, he had a play, said 'hmmm, that's interesting, worth my time' and the rest is history?
So, if I can kindly ask you - what exactly goes into the selection process - what are the challenges that make a particular router difficult to support or unsupported model and conversely - what makes it a long term success?
1. Is it the CPU?
Since the actual relevant lineup of BCM4906, 08, 12 and 16 are basically some variation of ARMv8 Cortex-B53 cores with various networking "bolt ons" I would assume, this part could be relatively easy - but is it? Is it easy to port the code from 4906 all the way to 4916 and it just "runs faster" or is better hw accelerated, or does it still rely on SDK from Broadcom, proprietary closed source code, Asus being helpful, information that's hard to get, etc?
2. Is it the SoC comms processor?
Now this I always thought is the secret sauce that is hard to dig deep into, because that's the competitive advantage, so companies get secretive and obstructive. Some SoCs are a paper success, but then don't quite stack up in real life. Some are average, but very stable and popular. Which are those that so far stood the test of time and why? Is the SoC the main stumbling block in terms of making it into supported models list? Is it as simple as Broadcom provides a blob to ASUS, they implement it into FW, Merlin if he's lucky gets it to incorporate into the custom FW and that's about as much as can be influenced along the line?
3. Is it as "simple" as having a clear line of communication with relevant parties so the final Merlin's FW can be developed, tested and produced and because that's hard or even impossible for the above combinations of CPU/SoC, the model gets dropped or even is a non-starter in the first place?
4. Or even simpler reason - only model Merlin gets his hands on is evaluated and if all of the above works, makes it onto the list?
5. All of the above?
6. How does GT-BE98 Pro with its all new chipset fit into all this?
I don't want you to spend too much of your precious time trying to elaborate on all of those. Maybe it's a simple explanation, maybe not. But I think it could be also a helpful insight for people here searching for recommendations into what makes a model a good relatively future proof choice. Basically some models stood the test of time. Just because they sell well is an easy answer, but why they sell well?
Thanks.
But mainly it got me thinking. I know it's difficult to say and there are many factors at play, but overall, in more of a broad brush strokes - what goes into choosing which model makes it so to speak onto the Merlin's list of supported models? Or even in more general sense, what makes it onto Asus internal list (well apart from 'it sells well')?
While it might sound like a stupid question (I was always told there's no such thing as a stupid question
), just please hear me out. I'm genuinely interested in the process of selecting which particular router is worth investing time in for Asus, for Merlin and ultimately, what makes it the model that is worth recommending and stands the test of time. I know, consumer routers, 2-3 years, 5 if you're lucky. Who cares, buy one, next year will be different. I get it. But still, there are reasons why some specific routers (AC68U, AC86U, AX86U, AX88U Pro, ...) are better choice and more preferred by people here than others. Why they got on Merlin's desk, he had a play, said 'hmmm, that's interesting, worth my time' and the rest is history?So, if I can kindly ask you - what exactly goes into the selection process - what are the challenges that make a particular router difficult to support or unsupported model and conversely - what makes it a long term success?
1. Is it the CPU?
Since the actual relevant lineup of BCM4906, 08, 12 and 16 are basically some variation of ARMv8 Cortex-B53 cores with various networking "bolt ons" I would assume, this part could be relatively easy - but is it? Is it easy to port the code from 4906 all the way to 4916 and it just "runs faster" or is better hw accelerated, or does it still rely on SDK from Broadcom, proprietary closed source code, Asus being helpful, information that's hard to get, etc?
2. Is it the SoC comms processor?
Now this I always thought is the secret sauce that is hard to dig deep into, because that's the competitive advantage, so companies get secretive and obstructive. Some SoCs are a paper success, but then don't quite stack up in real life. Some are average, but very stable and popular. Which are those that so far stood the test of time and why? Is the SoC the main stumbling block in terms of making it into supported models list? Is it as simple as Broadcom provides a blob to ASUS, they implement it into FW, Merlin if he's lucky gets it to incorporate into the custom FW and that's about as much as can be influenced along the line?
3. Is it as "simple" as having a clear line of communication with relevant parties so the final Merlin's FW can be developed, tested and produced and because that's hard or even impossible for the above combinations of CPU/SoC, the model gets dropped or even is a non-starter in the first place?
4. Or even simpler reason - only model Merlin gets his hands on is evaluated and if all of the above works, makes it onto the list?
5. All of the above?
6. How does GT-BE98 Pro with its all new chipset fit into all this?
I don't want you to spend too much of your precious time trying to elaborate on all of those. Maybe it's a simple explanation, maybe not. But I think it could be also a helpful insight for people here searching for recommendations into what makes a model a good relatively future proof choice. Basically some models stood the test of time. Just because they sell well is an easy answer, but why they sell well?
Thanks.
Last edited:
