List of compatible modems for Fusion X2 pair bonded service

[Duncan]

But WiFi is also a factor, and it's moving much more quickly than DSL technology.

Ok, maybe that accounts for all the different models. I never consider it because I like to keep my Wifi equipment separate from the DSL modem component. Obviously most other people just want a single device and therein lies the need for variety (sort of).

[wa2ibm]

The innovation cycle for the wireless is faster than for the DSL.

This is why I use separate modems, routers, and WAPs. I have switched out each one individually as technology and my needs change. I realize this isn't the "norm", but it works for me. All of my network gear is in an out-of-the-way (hidden) location so the clutter isn't a factor.

The only drawback is when my ISP demands that I use, and pay for, an all-in-one CPE. I end up disabling or bypassing most of its function. So far I haven't run into this, but it's beginning to look like I'll be going there soon.

[Guest]

Or what about the same basic box but with, say, a changeable WAN interface module?

There's nothing "basic" about a firewall/router with interchangable WAN interfaces. You're talking about enterprise equipment, easily $500 and up for the SOHO/SMB space. When companies can get the same performance (but not necessarily the same quality) from Chinese manufacturers, what do you think they'll choose?

[Duncan]

Or what about the same basic box but with, say, a changeable WAN interface module?

There's nothing "basic" about a firewall/router with interchangable WAN interfaces. You're talking about enterprise equipment, easily $500 and up for the SOHO/SMB space.

I'm not talking about a Cisco router with plug-in modules, but a design for consumer equipment where the overall product remains the same but the section with the WAN interface could be swapped at the factory before it's closed up. Granted, some companies do just that (and sometimes even leave a blank port left over in the housing as a result) but it's certainly not the norm, in my experience. Too many companies seem to clutter up their product matrix with 'distinctions without a difference' marketing tactics for the sake of boasting how many models they offer.

I look at this and wonder how many of those models simply overlap functionality with others by the same vendor, with perhaps a specific feature withheld entirely for marketing purposes:

https://wiki.sonic.net/wiki/Category:DSL_Modem

Anyway, we were talking about AT&T specifying modems for their own end-to-end service, and I can't imagine why they couldn't consolidate just that collection down to two or three models.

[Guest]

I'm not talking about a Cisco router with plug-in modules, but a design for consumer equipment where the overall product remains the same but the section with the WAN interface could be swapped at the factory before it's closed up.

I'm not convinced that's a compelling way of doing it anymore. It's much more cost-effective to have tightly-coupled devices. For consumers, it's even more important to get the cost down. Just look at people balking at line items at $5-$10 each month. I wonder what the cost of designing/testing what you'd want would be, accounting for every module. Is there any licensing required for the interface? Maybe spend some time rolling your own? What happens if a new standard requires a totally new design? It gets tricky to guess how many you'd need to sell so the amortization of the original design pays off. There are some schools of thought where even in software it's better to code what's required now rather than spend the time and energy to code for future expansion. The rationale is due to added complexity and time required your test cases may not even do what they're supposed to because there is no real-world use case. Code it wrong and you could potentially have a security vulnerability.

This decision would be a non-starter if you have bean counters running companies. More so since they have golden parachutes and if they're not the original founders.

[digitalbitstream]

If there are no compatible modems, then Sonic should just fess up, and say that x2 requires modem rental and inside wiring coverage period. Don't pretend it's otherwise.

There must be other compatible equipment, we are not doing anything non-standard.

So far nobody on this list has identified compatible equipment. If rental is required for X2, just say that. But saying it's possible is a little like selling property on the courthouse steps at 3am and claiming you had a public auction.

Sonic_Dane_X2.png (63.84 KiB) Viewed 38646 times

The moral high ground is to either produce a list of modems known to be compatible, or officially say X2 requires inside wire insurance. If approximately zero Sonic customers are supplying their own modem for X2, it's not really legitimate to advertise BYO as an option.

Or save a hassle and offer two levels of add-on fee: deluxe (end to end including inside wire insurance), and self-support (support only to a hobbled Sonic supplied modem operating in gateway mode, and only then when plugged into the MPOE).

[polpo]

Perhaps Sonic doesn't have a list of modems known to be compatible that one can buy because they only ones they know to be compatible are the ones they rent?

Anyway, looking at the manual for the Kasda KW5225 (http://www.kasda.cn/Downloads.asp?FileN ... _Specs.pdf), it supports being set up with PTM, despite it not being mentioned on the main product page (http://www.kasda.cn/product_info.asp?id=248). They're on Amazon: http://www.amazon.com/Kasda-KW5225-Giga ... B00HIXDL8U

The Zhone 6729-W1 also supports PTM: http://www.zhone.com/products/6729-W1/

[Duncan]

I'm not convinced that's a compelling way of doing it anymore. It's much more cost-effective to have tightly-coupled devices. For consumers, it's even more important to get the cost down. Just look at people balking at line items at $5-$10 each month. I wonder what the cost of designing/testing what you'd want would be, accounting for every module.

Look at the radio chipset used in today's smartphones. Qualcom makes an all-in-one solution that works with CDMA, HSPA, EV-DO, LTE, etc., etc. yet most of those bands will never be used by any single customer.

And again, I'm talking about AT&T only here, who owns the entire network from CO to premise, and they can't narrow down the equipment requirements for their own pool of millions of customers? That's just poor engineering, or lazy management.

As a customer, I'm tired of constantly juggling countless incompatible devices or purchasing/maintenance decisions for every damned piece of technology I come across.

This decision would be a non-starter if you have bean counters running companies. More so since they have golden parachutes and if they're not the original founders.

Yeah, isn't it all wonderful?

[Guest]

Look at the radio chipset used in today's smartphones. Qualcom makes an all-in-one solution that works with CDMA, HSPA, EV-DO, LTE, etc., etc. yet most of those bands will never be used by any single customer.

Point taken. But compared to DSL there is massive profit driving R&D resources to that platform.

And again, I'm talking about AT&T only here, who owns the entire network from CO to premise, and they can't narrow down the equipment requirements for their own pool of millions of customers? That's just poor engineering, or lazy management.

I would like to think it's due to lazy management but the answer is probably a bit more complicated than just standardizing all CPE. SBC has acquired other companies and standardizing those COs/RTs is probably quite costly and time consuming. Not as much as upgrading to fiber but it's probably a pretty good chunk of change.

[virtualmike]

SBC has acquired other companies and standardizing those COs/RTs is probably quite costly and time consuming. Not as much as upgrading to fiber but it's probably a pretty good chunk of change.

Not only that, but when any telco goes out to buy a stock of switches, RTs, VRADS, etc., it requests that suppliers submit bids for equipment that meets the current specifications. Last year, perhaps it was Lucent. This year, maybe Northern Telecom. Tomorrow, who knows?