how-to block ads
|
Johkal
Cool Cat
Premium,MVM
join:2002-11-13
Happy Valley
clubs:
1 edit | reply to dscline
Re: [News] PowerBoost
Cool! I think we can all agree and go on to discuss other points.
Thanks! | |
dscline
join:2001-09-01
Atlanta, GA
| reply to Johkal
said by Johkal  :I like posts short and to the point. I lack that ability. 
But your point is valid. Not all speedtests are going to be able to saturate your pipe. You need to use several, and try to get a feel for which ones seem to have high enough capacity to minimize the variables that could skew the results. Even your own throughput could vary if you're on a congested node. But if you get familiar with the results that you typically get from various sites, they can be a quick and convenient tool to confirm that there's been a change in your service, for the better, or for the worse. | |
Johkal
Cool Cat
Premium,MVM
join:2002-11-13
Happy Valley
clubs:
 ·Comcast Digital Vo..
 ·Comcast
·Vonage
| reply to dscline
I like posts short and to the point.
This is the simplest analogy of speed tests:
I use speed tests like you would use an idiot light on a car to tell you if your oil was low.
If you test within 80% of your advertised rate, you have no issues. If you fall below 80% of advertised, you may have an issue. Even this is by opinion.
Example:
Advertised 6000
If I test anywhere between 4800 - 6000+, I'm good
It is that simple. Speed tests should be used as basic "idiot lights".
--
Write me up a 125.......I Can't Drive 55 »redrocker.com/ »cabowabo.com/ | |
dscline
join:2001-09-01
Atlanta, GA
| reply to dadkins
said by dadkins :User comes here, new to Comcast, posts a speed test that is 1-2mbps below what they are supposed to be getting(like the Testmy.net results I posted above), WTF are *WE* supposed to say? "Sorry, you have to call Comcast because that speed test shows a problem."? No! We should yell them that speed tests are not to be trusted(exactly what you are complaining about me doing!). What are you supposed to tell them when they say ftp downloads are 1-2mbps below what they should be? I hit real-world downloads well below my available bandwidth all the time, because the site I'm pulling from simply doesn't have the bandwidth. When you get outside of major corporations that supply a lot of data, MANY companies you may be downloading information from have upload capacity that isn't sufficient to saturate the pipes of all their customers. That's not a problem with our connections, it's simply the nature of the beast. Buying 8mb service doesn't mean you'll get 8mb from everywhere. In fact, depending on your usage, it could be the exception rather than the rule.
The problem here is, you are attributing discrepancies specifically to the fact that the timed download is a "speedtest" site. What difference does it make? The same factors that make real world transfers vary can also make speed tests vary. Speed tests that are a great distance from me, or travel over low bandwidth links, or are from services that don't really have the capacity to saturate high bandwidth connections, don't accurately reflect the true bandwidth potential of your connection. Just like ftp downloads from sites with the same limitations. As far as I can see, it has NOTHING to do with the fact that it's a speedtest, but you seem to insist that it is.
You made a comment way back in the beginning of this debate that you've run speedtests where the results didn't correspond to what was reported by netpersec. THAT was what I was curious about, and was interested in discussing. Because I'd like to understand why. If that's the case, there's got to be a reason behind it, and that would suggest a truly inaccurate result, as opposed to an accurate result that doesn't necessarily represent your true bandwidth potential, or represent average results of downloads with different variables (not the least of which is sample size). But rather than discuss this odd phenomenon, you just continually jump up and down yelling "speedtests are inaccurate", and repeatedly show results to prove your belief (and ironically the results prove nothing, since the speedtest result is all that's posted).
said by Johkal :I can not understand why some folk in this thread do not understand the gross inaccuracies of speed tests. The problem is, some folk don't seem to be able to distinguish the difference between an inaccurate result, and a result that is simply not representative of your bandwidth under different conditions. The many variables between you and the server can affect the result, and now with bursting and boosting, the size of the transfer is another variable to consider. If a speedtest that is based on, say, a 10MB transfer, says you have a 15mb transfer rate, then an ftp download of a 20MB file averages out to an 10mb transfer rate, which one is inaccurate? Not necessarily either one... the speedtest result should accurately represent your average transfer rate for a 10MB file, and the ftp download should accurately represent your average transfer rate for a 20MB file. If your real world use has you downloading 4MB MP3s from fast servers, you could find that your real-world use actually exceeds the speedtest results.
Now, if the speedtest results don't correspond to the average netpersec results during that time slice, then that is interesting info. It's not what I experience, but then again I don't have significant bursting in my area. Perhaps there IS a bug in the speedtest, and it truly IS inaccurate. Or perhaps the speedtest in question was simply designed for a time when bursting wasn't around, and they therefore show you the peak values, assuming that the valleys were due to other variables that were momentarily bogging down the transfer. Regardless, it would be something worth understanding if we could have intelligent discussion on the topic. But sadly, we can't get past the gross generalizations that "speedtests are wrong for almost everyone".  | |
joekuz
join:2003-03-13
Calistoga, CA
| reply to Johkal
I think some just don't understand the difference between
intial/burst rates vs sustained rates. Somewhat resembles
the confusion with audio amplifiers. You may have an
amplifier that is say 400W peak and 250W rms (continuous)so
people believe that it will put out 400W. When in all
actuality it will be outputting 250W and occasionally 400W
for only a split second.
--
AMD 3700+ (san diego)@2739
Asus A8N-sli deluxe
1GB Corsair PC3200
Chaintech 6600GT @540/1.2
WD Raptor 36GB
Antec Trupower 430W
etc..... | |
Johkal
Cool Cat
Premium,MVM
join:2002-11-13
Happy Valley
clubs:
·Comcast Digital Vo..
·Comcast
·Vonage
1 edit | reply to dadkins
In the Tweaks forum, we get people who routinely use test servers that are on the other side of the continent from them & people in other countries who use our servers here in the US. Needless to say, the results are less than accurate. Like you pointed out: "User comes here, new to Comcast, posts a speed test that is 1-2mbps below what they are supposed to be getting". That is the exact scenario. These people use the test sites recommended on the first tool page, from a FAQ or "What to Post" for help question. The sites are not always the best or recommended site for their location. Many times I have gotten these people to test via several sites close to them, gotten much better results & the issue is resolved. I definitely agree that test sites are not to be trusted. When I was testing Adelphia's 16/2, I would get extremely low rates when using their own servers. Here, the servers had issues with reporting accurate speeds with the new tier. This reminds me of what people are seeing with Boost, but the opposite(i.e. too high speeds).
So, I can not understand why some folk in this thread do not understand the gross inaccuracies of speed tests.
--
Write me up a 125.......I Can't Drive 55 »redrocker.com/ »cabowabo.com/ | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
| reply to motoracer
Since people come here all the time posting slow speed test results, what exactly am I (or anyone else) supposed to tell them?
You're the expert... tell me.
User comes here, new to Comcast, posts a speed test that is 1-2mbps below what they are supposed to be getting(like the Testmy.net results I posted above), WTF are *WE* supposed to say?
"Sorry, you have to call Comcast because that speed test shows a problem."?
No!
We should yell them that speed tests are not to be trusted(exactly what you are complaining about me doing!).
We are supposed to tell them to try several tests and/or to start a download from a known fast server.
When we find that them doing a download shows speeds within 80% of advertised speeds... Tell me, were we right to tell them not to trust speed tests? Uh huh.
If it bothers you so much, either don't read anything I post, or try helping these people yourself!
You will find yourself posting the exact thing the rest of us do - Speed Tests are not to be trusted!
God forbid you actually grasp this concept, helping that is.
Tell ya what pal, I'll stop posting in Comcast for a while... see how many others take up the slack.
Will you be one of them?
If so, after a few hundred posts of people using a test to gauge their speeds, and that test is seriously wrong, what will you tell them?
Have a Great Day! 
--
Think outside the Fox... Opera | |
motoracer
join:2003-09-15
Valencia, CA
·VOIPo
 ·AT&T DSL Service
| reply to dadkins
said by dadkins :Uhhh... then don't read or reply to my posts? Just an observation of many people getting flaky results, and many others asking about/claiming connection problems because of poor speed test results. Yet, when they do try an actual download, speeds are fine. Since most of us are here to try and help, informing people not to trust the skewed results is the logical response, right? Thanks for your input, but I will continue to tell everyone that with Comcast(at least), speed tests are not to be trusted! I ask YOU friend, do you think this test is correct: :::.. Download Stats ..::: Download Connection is:: 6614 Kbps about 6.61 Mbps (tested with 5983 kB) Download Speed is:: 807 kB/s Tested From:: » testmy.net/ (Server 1) Test Time:: 2006/07/04 - 10:52am Bottom Line:: 115X faster than 56K 1MB Download in 1.27 sec Tested from a 5983 kB file and took 7.411 seconds to complete Download Diagnosis:: Awesome! 20% + : 27.49 % faster than the average for host (comcast.net) D-Validation Link:: » testmy.net/stats/id-WOQGDK3H6User Agent:: Opera/9.00 (Windows NT 5.1; U; en) [!] Should I call Comcast? Is my connection hosed? Is this test correct: Last Result: Download Speed: 29931 kbps (3741.4 KB/sec transfer rate) Upload Speed: 681 kbps (85.1 KB/sec transfer rate) Don't think so... How about an actual download? See pic. In YOUR opinion, which "test" should people trust? Have a Great 4th!  I'm not saying that the speed tests are accurate - in fact I know that they aren't accurate always. You just always sound like a broken record Oh well, I said what I wanted to. Have a good night. | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
| reply to cracker 52
ROFL! Ok, now try all the rest.
Try single digits, double digits.
Remember, Comcast is in 33 States, each of which has a few ZIPs.
Perhaps it's a bigger "problem" than I suspected!
I grow weary of this! LOL!
Y'all go ahead and be happy with your results, and I will be happy with mine.
...and when someone comes here asking if their connection is hosed because of a trashed speed test, I'll be sure and point them in your direction.
It's been "fun" guys! LMAO!
Have a Great 4TH ALL!
--
Think outside the Fox... Opera | |
cracker 52
@comcast.net
| reply to dadkins
I'm glad you're referring to the archive. So let's see the results of some of Comcast markets, with the first 3 digits of the city's area code:
Jacksonville, Fla - 322
»/archive/comca···?zip=322
just saw one skewed result there
Augusta, Georgia - 309
»/archive/comca···?zip=309
just one here also.
Savannah, Georgia - 314
»/archive/comca···?zip=314
Miami, Florida - 330
»/archive/comca···?zip=330
Fort Lauderdale, Fla - 333
»/archive/comca···t=Search
Knoxville, Tenn. - 379
»/archive/comca···t=Search
Charleston, S.C. - 294
»/archive/comca···t=Search
Los Angeles, Calf. - 900
»/archive/comca···?zip=900
just one skewed result there.
Seattle, Wash - 981
»/archive/comca···?zip=981
Detroit, Mich. - 482
»/archive/comca···?zip=482
Portland, Oregon - 972
»/archive/comca···?zip=972
Dallas, Texas - 752
»/archive/comca···?zip=972
Only one skewed result there
Except for the ones I noted, none of the above cities have skewed test results, using the same archive you referred to. No doubt some areas have significant/prolonged bursting, and other areas are getting powerboost. It appears to be mostly in the northeast/mid-Atlantic states (Verizon's territory), parts of Midwest, parts of Chicago and certainly where dadkins lives. But the areas I listed above don't have it, except for a few selected CMTS or headends. | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
3 edits | reply to cracker 52
said by cracker 52 :
There's nothing wrong with tha Atlanta systems. As I posted before, we have as good hybrid fiber plant (750 Mhz abnd 860 Mhz) as that good as any other Comcast market. Before being acquired by Comcast. ATTBI completed a major upgrade of the headend, plant and network systems.
Sorry, dadkins, but there's no speed or system problems (other than normal local problems that pop up periodically jsut like anywhere else ) here in Atlanta.
Whatever friend!
Go here: »/archive?zip=1···t=Search
Change that number to whatever from 0 - 9.
See where your area is in the grander scheme of Comcast subs that are testing here..
Sure, you may be getting all of the 8mbps that you are paying for... where's the burst that everyone else is getting?
The burst that most have been getting for nearly a year...
EDIT:
Know what! I really don't care whether you are or are not getting bursting.
Botom line is, speed tests are skewed, one way or the other - For most Comcast customers.
If yours is acceptable to you, Cool! Have at it!
Have a Safe 4th!
David
--
Think outside the Fox... Opera | |
cracker 52
@comcast.net
| reply to dadkins
There's nothing wrong with tha Atlanta systems. As I posted before, we have as good hybrid fiber plant (750 Mhz abnd 860 Mhz) as that good as any other Comcast market. Before being acquired by Comcast. ATTBI completed a major upgrade of the headend, plant and network systems.
Sorry, dadkins, but there's no speed or system problems (other than normal local problems that pop up periodically jsut like anywhere else ) here in Atlanta. | |
dscline
join:2001-09-01
Atlanta, GA
| reply to dadkins
said by dadkins :Well then, something is wrong with the ATL market. Seeing as nearly every other Comcast market is getting bursts, and you are not...  So which is it? Are your high speedtest results due to bursting, or are they inaccurate? I'm really quite puzzled by your stance. When YOU get high results, you say the speedtests are inaccurate. When our speedtest results correspond with our subscribed speed tier AND ftp downloads, you say something is wrong with our market. So exactly WHAT result would you be happy with, that would make you feel everything is OK?
And how does a feature not being enabled equate to something being wrong with the market? And what is your source that states nearly every other market besides Atlanta has bursting? | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
| reply to cracker 52
Well then, something is wrong with the ATL market.
Seeing as nearly every other Comcast market is getting bursts, and you are not...
Sorry!
--
Think outside the Fox... Opera | |
cracker 52
@comcast.net
| reply to dadkins
Obviously, no one should contact Comcast if the download speeds are skewed upwards. It should only be done if speeds are well below advertised tiers, using both java and FTP tests. (From my past experience, Comcast CSR asked me to use Speakeasy speed test to verify speed levels when I had upload speed problems earlier this year. Try asking your local Comcast CSR what test they use with their customers.)
My Speakeasy Atlanta test results are still consistent:
Download Speed: 8489 kbps (1061.1 KB/sec transfer rate)
Upload Speed: 720 kbps (90 KB/sec transfer rate)
Results from U. of Michigan network tester aren't skewed either:
CP/Web100 Network Diagnostic Tool v5.2.0f
click START to begin
Checking for Middleboxes . . . . . . . . . . . . . . . . . . Done
running 10s outbound test (client to server) . . . . . 730.56Kb/s
running 10s inbound test (server to client) . . . . . . 8.05Mb/s
Your PC is connected to a Cable/DSL modem
Information: Other network traffic is congesting the link | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
| reply to motoracer
Uhhh... then don't read or reply to my posts?
Just an observation of many people getting flaky results, and many others asking about/claiming connection problems because of poor speed test results.
Yet, when they do try an actual download, speeds are fine.
Since most of us are here to try and help, informing people not to trust the skewed results is the logical response, right?
Thanks for your input, but I will continue to tell everyone that with Comcast(at least), speed tests are not to be trusted!
I ask YOU friend, do you think this test is correct:
:::.. Download Stats ..:::
Download Connection is:: 6614 Kbps about 6.61 Mbps (tested with 5983 kB)
Download Speed is:: 807 kB/s
Tested From:: »testmy.net/ (Server 1)
Test Time:: 2006/07/04 - 10:52am
Bottom Line:: 115X faster than 56K 1MB Download in 1.27 sec
Tested from a 5983 kB file and took 7.411 seconds to complete
Download Diagnosis:: Awesome! 20% + : 27.49 % faster than the average for host (comcast.net)
D-Validation Link:: »testmy.net/stats/id-WOQGDK3H6
User Agent:: Opera/9.00 (Windows NT 5.1; U; en) [!]
Should I call Comcast? Is my connection hosed?
Is this test correct:
Last Result:
Download Speed: 29931 kbps (3741.4 KB/sec transfer rate)
Upload Speed: 681 kbps (85.1 KB/sec transfer rate)
Don't think so...
How about an actual download? See pic.
In YOUR opinion, which "test" should people trust?
Have a Great 4th!
--
Think outside the Fox... Opera | |
motoracer
join:2003-09-15
Valencia, CA
·VOIPo
·AT&T DSL Service
1 edit | reply to dadkins
said by dadkins Uh huh... Oh yeah! They're accurate, huh? LMAO!
For some odd reason, I seriously doubt that Comcast is hitting me with ~5 MegaBytes per second.
No one is forcing you to read or reply to anything I (or anyone else for that matter) posts. Are they? 
I'm not saying that they are accurate or not - that's another topic. I'm just being honest and letting you know it gets tiring to see you always preaching about their inaccuracy. | |
dadkins
Can you do Blu?
Premium,MVM
join:2003-09-26
Hercules, CA
·Comcast
2 edits | reply to motoracer
Accurate? |
said by motoracer :Dadkins, Can you stop preaching about how inaccurate speedtests are? Everyone, I mean everyone, knows your view of them, so just drop it once and for all please. Thanks Uh huh... Oh yeah! They're accurate, huh? LMAO!
For some odd reason, I seriously doubt that Comcast is hitting me with ~5 MegaBytes per second.
No one is forcing you to read or reply to anything I (or anyone else for that matter) posts. Are they?
--
Think outside the Fox... Opera | |
dscline
join:2001-09-01
Atlanta, GA
| reply to cracker 52
said by cracker 52 :
In some of the denser areas, capacity could be close to straining but, as the case with dscline, I've been getting consistent, full Gold tier speeds day and night. My parents live in an older community part of Atlanta, where you won't find many power users, and they don't get any bursting. When I first moved in to this house 13 years ago, this was somewhat the hood (near Little Five Points). I was really surprised when I got cable internet back around '98 before many of the areas that you'd consider were probably much bigger markets (back then, it was MediaOne, before it became AT&T, then Comcast). For the longest time, I was probably one of the few in my area to have broadband. In the past couple of years, this has become quite the trendy (and much more expensive) place to buy, and there are huge new lofts and other developments going up all over. I'm fully expecting the near spontaneous influx of likely broadband customers to catch the local infrastructure by surprise, but thankfully it hasn't appeared to happened yet. | |
cracker 52
@comcast.net
| reply to dscline
Actually here in Atlanta, several techs I spoke to said ATTBI spent about 2 million dollars upgrading the headend, network and plant just before being acquired by Comcast. The hybrid plant is either 750 Mhz or 860 Mhz. If you look at BBR Comcast TV board, most people are envious of the Atlanta market having the one of most extensive channel lineups in the country (just ask the knowledgeable people in New Jersey who have the smallest lineups).
In some of the denser areas, capacity could be close to straining but, as the case with dscline, I've been getting consistent, full Gold tier speeds day and night. My parents live in an older community part of Atlanta, where you won't find many power users, and they don't get any bursting. | |
|
