Phone lines have a small amount of voltage which runs along them, the amount varies depending on line. Some modems in their stats will report the voltage on the line (as an example when i was ADSL with a Netgear DG834GT with DGTeam firmware that did). The higher the voltage generally the more stable a line is. Not always the case, you can have too much but thats easy for an engineer to fix. A neighbour in my street (3 doors away) on ADSL could get almost 3Mb more than i ever could even trying various routers and it was all down to their line having something stupid like half a volt more power.
You did not upset me, i assure you its a software issue with the device the hardware has been tested independently to make sure it works. The software is based on opensource code, which do you think is more likely to need fixing, hardware which has undergone independent tests before it can be produced and sold, or opensource software which has been modified even further by Asus for the device?
Its not just the modem chip which would had been tested either, every component in that device would have to go for testing before it can be sold. The power supply, the wifi, everything right down to the screws have to be certified. The plastic case is even likely to have undergone test involving setting it on fire to make sure the plastic is not too oily and encourages it to burst into flames/flammibility. This is not some knockoff fleabay chinese shed made stuff its a mass commercial product.
It would have even had to be tested in its final design, Mediatek would highly unlikely allow them a license to use their chips in a chipset which did not meet their approval either.
Hardware nowadays for a worldwide market is subject to more testing than you can imagine.
The very fact the device becomes more stable for many just by disabling the spectrum monitor/stats pages in the SOFTWARE, even shows that there are software issues with the device. To blame the hardware with no evidence is a bit silly. The fact with some software releases its made the device either more or less stable for some people also points back to SOFTWARE.
Which does not show anything, there have been users with problems with skys router (see their independent forum) which have gone away with another device. Also although you can argue it may or may not be more stable than the Asus still does not explain why for some that switch to say a Openreach Huawei modem instead of the Sky 102 find the Huawei more stable. The chipset is basically the same in both devices, one over the other should not perform significantly better. Yet they do.
Its no different (apart from how reliable you deem it to be) with the Asus situation. No doubt out there their is some other Mediatek device (probably bundled gear in the far east for some fibre service) and that gives no issues.
Im not defending Asus, the software is a bug ridden mess, FEC stats are obviously reported wrong, having to disable part of its software (IE spectrum) to make it more stable is daft and thats just issues we know about. The problem if any is Asus have gone for a chipset and then basically ported the software from another device of a differing chipset (IE the non DSL AC68u) and just encoded the modem driver in with that software..... It should had been tested more before they released it.
It could be so bad that it would be easier to start from scratch, without all the code at present i could not say, either way within a week or so i will hopefully know when i get one of them and dump all the software via TTL. From there ill probably laugh, rip a load of redundant cack out and fix it. Though no doubt then people will want certain things which make it buggy put back in.... Sometimes as a software dev you cant win or ever please people, more features in software ALWAYS EQUALS more problems.
.