The ARRIS TM822G is a phone and internet E-MTA. It has no internal router so I do not have to worry about the bridged mode "doodie" that others have to worry about...personally I don't like the gateways that they have come out with...and the DOCSIS 3.0 in my area is still 8x4 config...I am aware that there are modems/gateways out there that will bond as many as 32 channels down, and 8 channels up (Broadcom modem chips, not Intel here, has the 32x8 config.)
So, until the ISP gets more downstream channels and upstream channels, I do not see any need to upgrade at least yet...but I can test the router on other people's ISP connections that I know and have access to, people which have connections of at least 50 Mbps or faster...even a 10 Mbps DSL line would not show this problem that I am having, as it is below the 25 Mbit threshold that this problem occurs...would be hard if not impossible to duplicate the problem with anything less than 25-30 Mbps...and I do *not* have traffic shaping enabled in my router either.
Never had a use for traffic shaping, as I like to get all of the bandwidth I can get from my setup...even Windows reserves 20% of an Internet connection for the OS alone, but in gpedit.msc (Windows 8.1 Group Policy editor) you can change it to allocate all bandwidth to your applications and the like...so Windows will no longer hog the last 20% of your internet connection.
Everything is configured as it should be...and as far as the line techs in my area from Comcast, they are a joke!
I often end up pointing the problems out to them that they cannot seem to find, but I find easily, like my VOIP was dropping calls all of the time...turned out to be a problem at the cable company headend, and I told them so too, they did not believe me at first, but they did after I pointed out the problem to them and they called back finally acknowledging that they had an internal problem, not a cable plant or cable drop/splitter/wiring issue.
I do not even have cable television in my house, I have RG-6 quad shield cable rated for 3GHz, not the standard 1GHz that many cable plants commonly use, which has a solid copper center conductor, not the copper covered steel that is very common in coaxial cable...all the way from the curb (tap) into my home, grounded at the pole at the back of my house with a coaxial grounding block, there are NO splitters whatsoever, and it has been gone over many, many times by me and Comcast...so I would say the cable/signal part of my Internet connection going to the E-MTA here is up to par...I have also swapped out any cat5e cables from my E-MTA to my router, and tested the structured wiring for any faults with a network wiring analyzer...found nothing...
The modulation error ratio of the E-MTA is a perfect 1.0e-9 modulation error ratio (MER) In other words, no errors/corrected/uncorrected codewords occurring with the Reed-Solomon error correction algorithm on my E-MTA...I have only phone and internet, I refuse to pay the highway robbery of Comcast's CATV...it costs $100 or more alone, and that is not even in HD, nor the highest tier they offer either !!!
SNR on all 8 downstream channels is between 39 to 41 dB, upstream SNR is good as well...Comcast won't tell me exactly what the SNR (signal to noise ratio) is on the upstream channels, but they say it is good whenever I call them to do a health check on my connection...there is no signal ingress/egress anywhere in the cable plant...the cable plant has to be pretty clean to run a modem with a 39 to 41 dB SNR on each downstream channel...many people are blessed to get 35-37 dB SNR...
The signal LEVELS are around a perfect 0 to +3 dBmV (decibels per millivolt) on the downstream channels, and I have a good upstream level that many would kill for...a perfect +35 dBmV across all upstream channels...anything 50 or higher on the upstream is asking for connection loss between the headend's CMTS and the E-MTA/modem/gateway/what have you...
I hope someone can duplicate this problem with a setup similar to mine...that way this can be confirmed as a bug...if there is a bug in any firmware, BIOS code on computers, etc, any software, leave it to me, I will expose it eventually. I am an excellent bug finder.
I programmed all of the BIOS'es in all of my computers with a custom firmware that actually has an audio processor embedded in the BIOS...that I use with a very small FM transmitter to listen to out in the yard/field around where I live...the BIOS version of this program runs much better and faster than anything that runs under Windows and even Linux...yes, executable code can be embedded in computer BIOSes as well...after all it is flashable...the program once loaded from ROM into RAM...runs purely in RAM.
The program is its *own* OS too...no other OS needed...it can be invoked before Windows starts by pressing the F6 key on any one of my computers.
No hard drive or SSD needed at all for this audio processor that I embedded in my computers' BIOSes. Just hope this sheds some light on what I am capable of doing with computers/internet and the knowledge of the technology as well...
Author
Topic: DIR-655 Rev B Only v2.11NA FW Official Release NA region only! (Read 104970 times)