dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
66
tragik74
join:2012-10-12
Flushing, NY

tragik74 to Dampier

Member

to Dampier

Re: Time Warner Boosting Standard Speed to 15Mbps

this testmy.net is very inaccurate. it will be best use speedtest.net, try out different server in your area

Smith6612
MVM
join:2008-02-01
North Tonawanda, NY
·Charter
Ubee EU2251
Ubiquiti UAP-IW-HD
Ubiquiti UniFi AP-AC-HD

Smith6612

MVM

I seem to have more trouble with that site on their upload test rather than the download test. Chrome seems to mis-report the upload. Firefox seems to give the best results. IE throws up over it. No idea about Safari but if you go too big on it via a mobile device the browser will run out of memory and crash.

I do use that site over speedtest.net though. Lately I've been using Verizon's FiOS300 speed test since it handles higher speed connections as reliably as NDT tests do. Speedtest.net needs to update their applets to make it more efficient.
Mele20
Premium Member
join:2001-06-05
Hilo, HI

Mele20 to tragik74

Premium Member

to tragik74
Speedtest.net is the WORST speed test out there. Don't advise folks to use any Flash test. Always test either by doing a BIG download from a known fast server or a Java test like here but the BEST tests are Visualware's. You need to do a QUALITY test not a CAPACITY test. Almost all speed tests (including Speedtest.net) do capacity tests. That tells you the theoretical capacity of your line. It does NOT tell you the QUALITY of your connection and that is the measurement that matters.

For a big download from a fast server (don't know about right now since hurricane Sandy) one of the best is Optimum on Line's FTP speed test. For someone on 30/5 the download is not really big enough but it is an excellent test for those below 30/5. I've been doing their test since I got Road Runner in 2001. It is listed as OOL Speed Test under links on the left side of the following page:

»OptimumOnline

We have a member here who runs outstanding Quality tests from Visualware. This is the Quality test server in Los Angeles. He also has a Quality test server in Texas and Georgia. You should do both the Capacity test and the Quality test but the really important one is the Quality test.

»www.ispgeeks.com/wild/mo ··· lityTest

If you want more locations for Visualware's superior tests:
Click on a area of the map and choose a yellow pin for Quality testing. You can also do route testing to that server if you get low results and are wondering if the routing is possibly the problem.

»myspeed.visualware.com/index.php

Make sure you click to get the various analyses of the test. You can get a full analysis with complete explanation of any problem areas. There is also an excellent white paper from Visualware explaining why capacity tests are not good tests and do not show you problem areas.

If you want to do a Capacity test then the NDT Web100 tests are still the best out there. They are superior Java tests but test Capacity just like the Flash tests at speedtest.net do. I use this server for 12 years now because it is in California:

»nitro.ucsc.edu/

You can access other public NDT servers around the nation by scrolling down on the page to the list near the bottom. If you have a university near you perhaps you can access their NDT server also.

Thinkdiff
MVM,
join:2001-08-07
Bronx, NY

Thinkdiff

MVM,

said by Mele20:

For a big download from a fast server (don't know about right now since hurricane Sandy) one of the best is Optimum on Line's FTP speed test. For someone on 30/5 the download is not really big enough but it is an excellent test for those below 30/5. I've been doing their test since I got Road Runner in 2001. It is listed as OOL Speed Test under links on the left side of the following page:

Somewhat disagree with you that flash tests are always useless - you just have to know how to interpret the results (ignore powerboost, understand that it's only a small snapshot of your connection, don't take peak performance, look at average performance, etc).

But I do agree the OOL FTP test has been consistently good. If you're looking for one to use with fast connections, they have a 1GB test as well:
»ftp://ftp1.optonline.net/test1024

sivran
Vive Vivaldi
Premium Member
join:2003-09-15
Irving, TX

sivran to tragik74

Premium Member

to tragik74
Testmy.net has never been inaccurate, even when I use a browser the creator of it explicitly states is unsupported.

banditws6
Shrinking Time and Distance
Premium Member
join:2001-08-18
Frisco, TX

banditws6 to Mele20

Premium Member

to Mele20
Wow, that is some pretty good info re: all those tests. Unfortunately they all seem to point to the same thing for me: that while my line tests out at about a 17 Mbps capacity, I rarely achieve anything higher than 4. This bears out what I've been experiencing over the last few works, i.e. slow downloads everywhere I go.

Pretty crappy.

motorola870
join:2008-12-07
Arlington, TX

motorola870

Member

said by banditws6:

Wow, that is some pretty good info re: all those tests. Unfortunately they all seem to point to the same thing for me: that while my line tests out at about a 17 Mbps capacity, I rarely achieve anything higher than 4. This bears out what I've been experiencing over the last few works, i.e. slow downloads everywhere I go.

Pretty crappy.

banditws6 I would recommend you getting a SB6141 or SBG6580 if you want to get your full speed because I am betting you have a DOCSIS 2.0 modem and with all of the data and not to mention the multiroom DVR guide data and box requests being sent over the DOCSIS channels that a single channel in north texas is congested right now we have 6 DOCSIS channels running in all of the nodes in North Texas it should help you speed issues the only thing is you will lose powerboost. I consistently get 30/5 on extreme with a SBG6580.

Thinkdiff
MVM,
join:2001-08-07
Bronx, NY

Thinkdiff

MVM,

Does TWC use the internet DOCSIS channel for their boxes as well? That would be pretty wasteful.

Also, if TWC has allowed it, DOCSIS3 allows D2 modems to switch between different channels (DCC) based on congestion - so even though you can't balance traffic over all 6 channels in real time, your D2 modem should still be able to find an open channel.
Mele20
Premium Member
join:2001-06-05
Hilo, HI

Mele20 to Thinkdiff

Premium Member

to Thinkdiff
Thanks for the link to the larger OOL test. I only have 15/1 but I am doing that test now. I like that it takes much longer so I can better see if there are problems.

This afternoon the regular OOL test started out at 1.2MB/sec. It gradually rose to 1.7MB/sec which is about what it should be the whole time. It got to about 40% and stopped. Several minutes went by and then it started up again but quickly dropped to 84KB/sec. UGH. Stayed there until about 70% done, then it gradually got up to 410KB/sec and stayed there until the end of the download.

Well, this is interesting. The larger test was steady from the very beginning at 1.7MB/sec until it got to about 40% where it stopped. Started again after a minute and quickly dropped download speed to about 85KB/sec at which time the connection was lost and I got an error message. It is 11:50PM here so my problems have nothing to do with prime time usage.

The last time I had a problem like this, this summer, it was like pulling teeth to get Oceanic TWC to fix it but when I finally got a very experienced tech out here (about the third tech the sent) whom I have known for years (before Road Runner existed and he just did TV). He took one look and "I would be furious also if this was happening to me". He notified the person who works on the node and it got fixed in two days. Now I have the problem again for over two months now and I can't get Oceanic to fix it.

The problem I have is very accurately reflected in the analysis that Visualware does. I showed several analyses from different Visualware servers to the tech this summer. He grasped the significance immediately...but the techs sent before him did not and this time no tech has displayed any understanding of the problem. In the fact the last one (sub contractor) called me and said he wanted to come out and show me how to do a speed test...geez...it took all my control to not bite his head off. He wouldn't come out for anything other than to show me how to do a speed test because said his understanding was that I didn't know how to do one.
Mele20

Mele20 to motorola870

Premium Member

to motorola870
said by motorola870:

said by banditws6:

Wow, that is some pretty good info re: all those tests. Unfortunately they all seem to point to the same thing for me: that while my line tests out at about a 17 Mbps capacity, I rarely achieve anything higher than 4. This bears out what I've been experiencing over the last few works, i.e. slow downloads everywhere I go.

Pretty crappy.

banditws6 I would recommend you getting a SB6141 or SBG6580 if you want to get your full speed because I am betting you have a DOCSIS 2.0 modem and with all of the data and not to mention the multiroom DVR guide data and box requests being sent over the DOCSIS channels that a single channel in north texas is congested right now we have 6 DOCSIS channels running in all of the nodes in North Texas it should help you speed issues the only thing is you will lose powerboost. I consistently get 30/5 on extreme with a SBG6580.

He said he went back to Standard at 10/1. He doesn't need a DOCSIS 3modem for Standard tier.

I would not recommend it on Standard tier and especially with the speeds he is getting (4 mbps) he needs Power Boost. His testing that is showing his capacity to be 17MBS probably is due to Power Boost distortion but Power Boost while it screws with most speed tests (Sam Knows tests can filter out Power Boost) is useful on Standard tier. I MISS it on 15/1. I was going back to 10/1 to get it again until TWC decided to up everyone to 15/1 and, I guess, no Power Boost as I have grandfathered Turbo which is 15/1 and there is no Powerboost on it.

Even with Standard tier now boosted to 15/1 there is no need for a DOCSIS 3 modem unless you plan, in the next year or two, to move to a higher tier that requires a DOCSIS 3 modem. Right now, 6141 is way too expensive to justify buying one for 15/1 tier. That SBG6580 has very bad reviews. I would recommend avoiding it and, currently, TWC has only ONE OTHER DOCSIS 3 modem on the approved list ...that is the Surfboard 6141 where price gouging is occurring.

motorola870
join:2008-12-07
Arlington, TX

motorola870

Member

said by Mele20:

said by motorola870:

said by banditws6:

Wow, that is some pretty good info re: all those tests. Unfortunately they all seem to point to the same thing for me: that while my line tests out at about a 17 Mbps capacity, I rarely achieve anything higher than 4. This bears out what I've been experiencing over the last few works, i.e. slow downloads everywhere I go.

Pretty crappy.

banditws6 I would recommend you getting a SB6141 or SBG6580 if you want to get your full speed because I am betting you have a DOCSIS 2.0 modem and with all of the data and not to mention the multiroom DVR guide data and box requests being sent over the DOCSIS channels that a single channel in north texas is congested right now we have 6 DOCSIS channels running in all of the nodes in North Texas it should help you speed issues the only thing is you will lose powerboost. I consistently get 30/5 on extreme with a SBG6580.

He said he went back to Standard at 10/1. He doesn't need a DOCSIS 3modem for Standard tier.

I would not recommend it on Standard tier and especially with the speeds he is getting (4 mbps) he needs Power Boost. His testing that is showing his capacity to be 17MBS probably is due to Power Boost distortion but Power Boost while it screws with most speed tests (Sam Knows tests can filter out Power Boost) is useful on Standard tier. I MISS it on 15/1. I was going back to 10/1 to get it again until TWC decided to up everyone to 15/1 and, I guess, no Power Boost as I have grandfathered Turbo which is 15/1 and there is no Powerboost on it.

Even with Standard tier now boosted to 15/1 there is no need for a DOCSIS 3 modem unless you plan, in the next year or two, to move to a higher tier that requires a DOCSIS 3 modem. Right now, 6141 is way too expensive to justify buying one for 15/1 tier. That SBG6580 has very bad reviews. I would recommend avoiding it and, currently, TWC has only ONE OTHER DOCSIS 3 modem on the approved list ...that is the Surfboard 6141 where price gouging is occurring.

you know what I have a leased SBG6580 and I get my 30/5 constantly the modem is as bad as it seems a lot of the problems are people don't know how to properly bridge the modem.
Mele20
Premium Member
join:2001-06-05
Hilo, HI

Mele20

Premium Member

Yeah...I bet they don't know how to bridge it. I had problems bridging the Netgear modem that Sam Knows sent me (and other poblems with it). They finally sent me a TP-Link already in bridge mode.

If you are leasing that SBG6580 then no problem. I thought we were talking about modems that folks are buying. The recommendation for folks buying a modem for RR to avoid the new rental fee is to buy only the 6141 if you need a DOCSIS 3 modem now or in the next year or so.

motorola870
join:2008-12-07
Arlington, TX

motorola870

Member

said by Mele20:

Yeah...I bet they don't know how to bridge it. I had problems bridging the Netgear modem that Sam Knows sent me (and other poblems with it). They finally sent me a TP-Link already in bridge mode.

If you are leasing that SBG6580 then no problem. I thought we were talking about modems that folks are buying. The recommendation for folks buying a modem for RR to avoid the new rental fee is to buy only the 6141 if you need a DOCSIS 3 modem now or in the next year or so.

well I was. When I got my SB6120 before extreme and ultimate came to my are it was $99 dollars so I think the SB6141 might be a little pricy but still it is a few dollars less than a SB6580 so you might as well buy it. DOCSIS 2.0 is pretty much on its way out if you are buying one of these cheap $50-60 modems you are in a world of hurt because eventually we will probably be seeing 16-24 DOCSIS channels per node and speeds of 300Mbps or more for what ultimate is now at 50Mbps but this is probably at least 2-3 years away in most markets.

banditws6
Shrinking Time and Distance
Premium Member
join:2001-08-18
Frisco, TX

banditws6 to Mele20

Premium Member

to Mele20
Yeah, I'm on Standard 10/1 and that's not gonna change anytime soon, so I don't think a DOCSIS 3 modem is going to do anything for me. I'm currently on a rented Ambit U10C018 which is a DOCSIS 2 modem. It also is a poor quality modem in my opinion due to the number of times I've had to hard-reset it, but I guess it could be worse (at my prior residence I had a COM21 Doxport from Comcast that was truly a piece of junk).

I also tried that OOL large file download test and never got any higher than about 328 KB/sec on it. YouTube is constantly terrible as well (I actually use a video downloader to pull down videos ahead of time so I can watch them without buffering) but I had assumed that was a YouTube congestion issue as many people report it.

Yet, there are a scant few sites where I can pull down about 1 MB/sec., suggesting that perhaps it's not a local issue with my own hardware but a congestion issue, routing issue, or something oddly intermittent.

It's pretty disappointing, to be honest. I was looking at AT&T U-Verse a while back but as long as TWC was operating reliably I thought I had better not mess with a good thing. I'm not sure it's so good anymore. Of course U-Verse has plenty of its own problems, but my coworkers at the office seem to like it...

motorola870
join:2008-12-07
Arlington, TX

motorola870

Member

said by banditws6:

Yeah, I'm on Standard 10/1 and that's not gonna change anytime soon, so I don't think a DOCSIS 3 modem is going to do anything for me. I'm currently on a rented Ambit U10C018 which is a DOCSIS 2 modem. It also is a poor quality modem in my opinion due to the number of times I've had to hard-reset it, but I guess it could be worse (at my prior residence I had a COM21 Doxport from Comcast that was truly a piece of junk).

I also tried that OOL large file download test and never got any higher than about 328 KB/sec on it. YouTube is constantly terrible as well (I actually use a video downloader to pull down videos ahead of time so I can watch them without buffering) but I had assumed that was a YouTube congestion issue as many people report it.

Yet, there are a scant few sites where I can pull down about 1 MB/sec., suggesting that perhaps it's not a local issue with my own hardware but a congestion issue, routing issue, or something oddly intermittent.

It's pretty disappointing, to be honest. I was looking at AT&T U-Verse a while back but as long as TWC was operating reliably I thought I had better not mess with a good thing. I'm not sure it's so good anymore. Of course U-Verse has plenty of its own problems, but my coworkers at the office seem to like it...

6 channels versus 1 will help congestion issues. which you are obviously seeing.

banditws6
Shrinking Time and Distance
Premium Member
join:2001-08-18
Frisco, TX

banditws6

Premium Member

said by motorola870:

6 channels versus 1 will help congestion issues. which you are obviously seeing.

Really, so even though I am not on a DOCSIS 3 tier, the fact that the modem can utilize a wider variety of channels will help because it can select the least congested one? Or would it actually bond channels (even though I'm on the 10/1 tier) until it could deliver the service I'm rated for?

Sorry, I am new to the whole DOCSIS 3 thing and haven't spent much time studying how it works. I had assumed that a DOCSIS 3 modem would not operate any differently than my DOCSIS 2 modem on the Standard 10/1 tier.

Edit: In doing a little reading on this just now, it seems you are right. Huh, frankly I would give up Powerboost to get a consistent 10 megabits. I'm not seeing the benefits of PB right now as it is!

Thinkdiff
MVM,
join:2001-08-07
Bronx, NY

Thinkdiff

MVM,

D2 Modems on D3 systems can use dynamic channel changing to pick a less congested channel. They cannot use multiple channels at the same time (that's why D3 modems exist).

Even on a 10/1 tier, a D3 modem will bond all available channels.

hobgoblin
Sortof Agoblin
Premium Member
join:2001-11-25
Orchard Park, NY

hobgoblin to Mele20

Premium Member

to Mele20
♠"I had problems bridging the Netgear modem that Sam Knows sent me (and other poblems with it). They finally sent me a TP-Link already in bridge mode."

I think you meant Router. Sam Knows were not sending out Cable modems.

Hob
Mele20
Premium Member
join:2001-06-05
Hilo, HI

Mele20

Premium Member

Yes, Sam Knows sends out routers but the discussion was about a weird hybrid router, wireless modem that is junk. So, I was responding to the router aspect of that crappy Motorola hybrid.

DocDrew
How can I help?
Premium Member
join:2009-01-28
SoCal
Ubee E31U2V1
Technicolor TC4400
Linksys EA6900

DocDrew to Thinkdiff

Premium Member

to Thinkdiff
said by Thinkdiff:

D2 Modems on D3 systems can use dynamic channel changing to pick a less congested channel. They cannot use multiple channels at the same time (that's why D3 modems exist).

Small, but important correction: The modem doesn't pick the channel. The CMTS commands the modem to use a specific channel at a specific time or condition.

If the channel is congested the modem won't switch unless the CMTS orders it.

tonyram57
join:2001-11-08
Brooklyn, NY

tonyram57

Member

What about my question? Would a 3.0 modem help with the few second drops.

DocDrew
How can I help?
Premium Member
join:2009-01-28
SoCal
Ubee E31U2V1
Technicolor TC4400
Linksys EA6900

DocDrew

Premium Member

said by tonyram57:

What about my question? Would a 3.0 modem help with the few second drops.

It might. Depends on the cause of the drops.

The drops might be caused by the CMTS telling your modem to switch channels due to congestion or upstream channel signal interference. If it's caused by such switching a D3 modem would help.

It the drops aren't caused by the CMTS switching channels on your modem, a D3 modem may not help.

Thinkdiff
MVM,
join:2001-08-07
Bronx, NY

Thinkdiff to DocDrew

MVM,

to DocDrew
I knew somebody would correct that when I wrote it.. I didn't want to confuse everybody by throwing the CMTS into the mix. But you're correct.

motorola870
join:2008-12-07
Arlington, TX

motorola870 to banditws6

Member

to banditws6
said by banditws6:

said by motorola870:

6 channels versus 1 will help congestion issues. which you are obviously seeing.

Really, so even though I am not on a DOCSIS 3 tier, the fact that the modem can utilize a wider variety of channels will help because it can select the least congested one? Or would it actually bond channels (even though I'm on the 10/1 tier) until it could deliver the service I'm rated for?

Sorry, I am new to the whole DOCSIS 3 thing and haven't spent much time studying how it works. I had assumed that a DOCSIS 3 modem would not operate any differently than my DOCSIS 2 modem on the Standard 10/1 tier.

Edit: In doing a little reading on this just now, it seems you are right. Huh, frankly I would give up Powerboost to get a consistent 10 megabits. I'm not seeing the benefits of PB right now as it is!

I wouldn't of recommend a DOCSIS 3.0 modem if I knew that our market only had one channel. I know there were some rumblings that they were looking to increase the speeds soon due to adding a 5th and 6th downstream. Also the DOCSIS 3.0 modems bond on the upstream here so that would almost be a 100% given you will always get the 1Mbps upload.

here is what North Texas has for DOCSIS channels!

Downstream Bonded Channels
Channel Lock Status Modulation Channel ID Frequency Power SNR Correctables Uncorrectables
1 Locked QAM256 1 783000000 Hz -4.9 dBmV 37.5 dB 391 2
2 Locked QAM256 2 789000000 Hz -5.1 dBmV 37.5 dB 260 0
3 Locked QAM256 3 795000000 Hz -5.7 dBmV 37.1 dB 338 0
4 Locked QAM256 4 801000000 Hz -5.8 dBmV 36.7 dB 638 0
5 Locked QAM256 5 807000000 Hz -6.5 dBmV 35.9 dB 1748 0
6 Locked QAM256 6 813000000 Hz -6.5 dBmV 36.7 dB 1309 0
7 Not Locked Unknown 0 0 Hz 0.0 dBmV 0.0 dB 0 0
8 Not Locked Unknown 0 0 Hz 0.0 dBmV 0.0 dB 0 0

Upstream Bonded Channels
Channel Lock Status US Channel Type Channel ID Symbol Rate Frequency Power
1 Locked TDMA and ATDMA 10 2560 Ksym/sec 33000000 Hz 42.7 dBmV
2 Locked TDMA and ATDMA 9 2560 Ksym/sec 36500000 Hz 42.7 dBmV
3 Locked TDMA and ATDMA 11 2560 Ksym/sec 29500000 Hz 42.7 dBmV
4 Locked TDMA and ATDMA 12 2560 Ksym/sec 24500000 Hz 42.7 dBmV

banditws6
Shrinking Time and Distance
Premium Member
join:2001-08-18
Frisco, TX

banditws6

Premium Member

Sweet, thanks for the details! That's exactly what I was looking for RE: the available channels in this area.

I'm probably going to look at getting a SB6141; despite the expense I do rely on stable Internet connectivity for work-related stuff and it's probably worth it in my case.

motorola870
join:2008-12-07
Arlington, TX

motorola870

Member

said by banditws6:

Sweet, thanks for the details! That's exactly what I was looking for RE: the available channels in this area.

I'm probably going to look at getting a SB6141; despite the expense I do rely on stable Internet connectivity for work-related stuff and it's probably worth it in my case.

the channel id's may vary but yeah that is all of the frequencies in North Texas. Really depends if you are on a Cisco CMTS or a ARRIS CMTS and the ARRIS will have 1-6 on the downstream and the Cisco will have something like 60-67 or 120-127 on the downstream for example. I was on a Cisco CMTS for a long time and then TWC fired up a brand new ARRIS C4 to feed parts of Arlington.