What to look for in a DOCSIS modem?

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
This ... Also, XYZ in disguise, I'm pretty sure it was pointed but be aware your ISP will update and bounce the bridge at their leisure whether it is yours or theirs. They own the network and manage anything attached directly to it. Purchasing your own device will not change that at all.
If it's a gateway item, yes. For straight modems, it's basically just the firmware, etc that they mess with it and if you can see the admin screen. Otherwise, it's a straight modem so no real issues. That's why we moved to a plain modem in the first place--just trying to get rid of the same thing the OP is.
 

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
It's a modem. You input power and coax, and it spits out ethernet. There's not really much to "do wrong" on consumer modems. If it's so flakey that it needs special handling and can't handle small drift in signal or attenuation, then, it's not something I'd suggest folks go out and purchase, since it's a sensitive snowflake that could very well have problems.
I agree with you, and there's a lot of people that have had exactly that experience--it just works. I think the special snowflakes are the ones that think they've got some sort of defective item that's going to win them some pennies at the end of a class action lawsuit, and want the whole world to know about it.
 

zandor

2[H]4U
Joined
Dec 14, 2002
Messages
3,825
Generally I'd say DOCSIS 3.1 is what to look for unless you're planning on being on the slow boat for a good long time. Depending on where you live you may not have a choice about that. Other than that, I dunno. I had one of those Intel Puma devices for a few years and never had any trouble that I'm aware of. I like my Netgear CM2050V better though. It's a Comcast exclusive, but basically just a CM2000 + Comcast voice support. The main thing I like about it is that it reboots really fast and the web interface is much more responsive. I guess I can't really give any advice on what's best since I just haven't been in a situation where I needed to bother thinking about that. I know sometimes people have issues, but I just haven't had any that weren't Comcasts' fault.. and, thankfully, fixed by Comcast. Guess I'm lucky for now?
 

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
I know sometimes people have issues, but I just haven't had any that weren't Comcasts' fault.. and, thankfully, fixed by Comcast. Guess I'm lucky for now?
I'd hedge a bet to say that you're not lucky, but just the typical scenario where stuff just works.
 

Valnar

2[H]4U
Joined
Apr 3, 2001
Messages
3,831
It's a modem. You input power and coax, and it spits out ethernet. There's not really much to "do wrong" on consumer modems. If it's so flakey that it needs special handling and can't handle small drift in signal or attenuation, then, it's not something I'd suggest folks go out and purchase, since it's a sensitive snowflake that could very well have problems.
They aren't that simple. There are quality differences in every product ever made.

Once again, Intel admitted it's an issue, and there are reports all over of people experiencing the issue, besides this thread.

Eveyrone else is basically drinking the coolaid just parroting what they're read online. I've actually done enough testing to debunk what most people are falsely calling an issue.
Seriously? I have first-hand experience with this, as do many others. Why would anyone NOT buy a Broadcom based modem and simply not worry about it?
 

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
They aren't that simple. There are quality differences in every product ever made.

Once again, Intel admitted it's an issue, and there are reports all over of people experiencing the issue, besides this thread.


Seriously? I have first-hand experience with this, as do many others. Why would anyone NOT buy a Broadcom based modem and simply not worry about it?
True, but less so in the world of modems since basically arris makes everything now. If there was such a big brew haha about it, Intel would have had to send out some sort of fix like on the cpu issues, but there's been no official notification that I've gotten.

And I doubt the scenarios in which you and others had this issue. When I first ran the dslreport puma test, both modems were showing issues and I thought, 'oh no'. Then I ran it on the stock comcast router and it also showed there were issues, which wasn't correct. I had to disconnect everything on the network and have just the modem and a laptop to run the puma test and then everything was fine on all 3. Testing scenarios makes a huge difference.

Why go with one vs something else? For me, the main driver is cost. Because of being bashed online ad infinium, the sb6190 can be found used at the same price as the sb6183 or even sb6141, which presents a great value, whereas generally almost all other modems with the same bandwidth capability cost 2x as much. Sure, it's not much in absolute dollars ($30 or so), but if there is no real issue, there's no reason to throw that money away.
 

Eulogy

2[H]4U
Joined
Nov 9, 2005
Messages
2,637
They aren't that simple. There are quality differences in every product ever made.
Look at who and what I was replying to for full context. SamirD was suggesting that consumers were "doing something wrong", which, it's pretty difficult to, again, given that consumers plug in coax, and get ethernet in return. For them, there's not anything to "do wrong".
 

Eulogy

2[H]4U
Joined
Nov 9, 2005
Messages
2,637
True, but less so in the world of modems since basically arris makes everything now. If there was such a big brew haha about it, Intel would have had to send out some sort of fix like on the cpu issues, but there's been no official notification that I've gotten.
From the intel lawsuit (which I see you still haven't bothered to read on):
Intel and Arris tests indicated that the SB6190 Modem experienced occasional UDP latency spikes over [Redacted] milliseconds, from a baseline average of [Redacted] milliseconds. ECF No. 180-2 at 139 (Arris tests); ECF No. 172-18 at 7 (Intel tests)
Nonetheless, Newman testified that the SB6190 Modem's over [Redacted] millisecond UDP latency spikes during the Intel and Arris tests "can cause significant problems when you're doing - when you're using UDP," such as in certain video games. Newman Depo. II 134:17-135:17; 209:5-15. For an even higher and thus worse UDP latency spike, Newman describes the result for end users as annoying and frustrating. Specifically, Newman opines that a latency of "over about 200 milliseconds becomes annoying to the point of materially degrading the experience" of video and audio conferencing. Newman Supp. Decl. ¶ 40. Such an experience would be "annoying" and "frustrating.

And, basically, Intel's argument is that latency isn't tantamount to a defect requiring response or recall, or that the consumer good is lacking. Which is why you haven't heard anything. Not even sure why you'd think Intel would contact you directly on such a thing, even if they lose the lawsuit.
 

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
Look at who and what I was replying to for full context. SamirD was suggesting that consumers were "doing something wrong", which, it's pretty difficult to, again, given that consumers plug in coax, and get ethernet in return. For them, there's not anything to "do wrong".
In terms of testing to determine that the modem is at fault, yes, consumers can most definitely do something wrong.
 

SamirD

Supreme [H]ardness
Joined
Mar 22, 2015
Messages
5,200
From the intel lawsuit (which I see you still haven't bothered to read on):



And, basically, Intel's argument is that latency isn't tantamount to a defect requiring response or recall, or that the consumer good is lacking. Which is why you haven't heard anything. Not even sure why you'd think Intel would contact you directly on such a thing, even if they lose the lawsuit.
Hell no I'm not going to bother reading a lawsuit that doesn't affect me or many others. Definitely have better things to do.

So the problem is on UDP and most internet traffic is TCP/IP. Got it. I guess that's why most people won't see the issue, including myself since most traffic is TCP vs UDP. So if one is using anything that uses UDP, I guess you need to avoid it. I'm sure no one has done comparative test on UDP for other brands as a comparison, but why bother since bashing is fun. :rolleyes: And this still doesn't explain why my two modems were 100% clear on the dslreports puma report...
 

Eulogy

2[H]4U
Joined
Nov 9, 2005
Messages
2,637
Hell no I'm not going to bother reading a lawsuit that doesn't affect me or many others. Definitely have better things to do.

So the problem is on UDP and most internet traffic is TCP/IP. Got it. I guess that's why most people won't see the issue, including myself since most traffic is TCP vs UDP. So if one is using anything that uses UDP, I guess you need to avoid it. I'm sure no one has done comparative test on UDP for other brands as a comparison, but why bother since bashing is fun. :rolleyes: And this still doesn't explain why my two modems were 100% clear on the dslreports puma report...
Most gaming, streaming, and VoIP is UDP. Not a trivial thing. I doubt anyone else has done as much testing, because they didn't have faulty units that required testing for a court case.
 
Top