Kitz Forum
Broadband Related => Broadband Technology => Topic started by: kezzaman on May 16, 2012, 12:07:40 AM
-
Hi guys,
Wondered, if there's alot of bitswapping on a line does it indicate that there might be an issue?
and does all that swapping affect latency, like are packets having to wait while bits are being swapped, or am i just not getting what bitswapping is about? :D
-
. . . or am i just not getting what bitswapping is about? :D
Possibly. ;)
Kitz has a very nice write-up (http://www.kitz.co.uk/adsl/adsl_technology.htm#bit_swapping) on the subject.
-
just too add. no it wont affect latency...
as you talk of latency I suspect your a gamer? Gaming latency relies on many small packets (per second) being sent & received, this uses a fraction of your available bandwidth.
# think of your broadband connection like a multi-lane highway (many hundreds of lanes), with broadband speed (throughput) its how many cars you can get down that highway in X amount of time, dodging traffic and roadworks on the way, you can send many hundreds of cars all behind each other until all your available lanes are full, and however many get down that highway in X amount of time is how fast it is... now with Latency its how quick one single car get down the highway and back on its own - so when your gaming its like a small convoy of cars all running in a line one after the other down just a couple lanes on this wide open highway, if traffic becomes too much on this highway then it can affect each cars finish time (response time) , where as how fast you can download is like sending every car you have until all lanes are full - i think thats a good metaphor... :) - full download speed is congesting your highway to get as many cars through as poss and latency is just a small trickle of cars going through.
Anyway, like I say Bit-Swapping is normal and will not affect your latency, well, unless you had a very long line with a really slow & noisey connection and the bits constantly being swapped consisted of a large percentage of all available bits/tones, and the router cpu was slow at swapping them?? ..perhaps anyway, maybe someone can clarify that? :) - but as far as Im aware it will not
-
Good analogy there snadge with the car lanes :)
Latency is how long it takes a single packet of data to reach from point A to point B. Latency should remain the same regardless of your adsl speed. eg someone on a 512kbps connection could very well have exactly the same latency as someone on a 24Mbps connection.
adsl works by opening up lots of different channels (lanes) and allowing more data packets to traverse at the same time which is what increases the bandwidth speed. Think of a hosepipe.. it takes the same time for a drop of water (data packet) to reach the tap to the end of the hose, but if you have a wider pipe then more water (datapackets) can pass through the hose at the same time.
>> all that swapping affect latency,
No, bit swapping makes sure that the maximum databits can pass through on each channel (lane) and if one channel cant handle x no of bits it swaps them to another channel if that has any spare. kinda like lane hopping.. except that your router is constantly monitoring the bit allocation and is in advance thinking which traffic (bits) to send down which lanes (channels). It would be more like a policeman directing traffic to the best route.
Interleaving and error correction can affect latency though. This is because of 1) the slight extra time whilst data is interleaved by the router and reassembled at the other end 2) Error correction carries redundant data - therefore less useful data per transmit (speed).. and of course the time taken by the routers at both ends to code/decode the data.
>> unless you had a very long line with a really slow & noisey connection and the bits constantly being swapped consisted of a large percentage of all available bits/tones, and the router cpu was slow at swapping them?? .... maybe someone can clarify that
Correct, bit swapping only works to a certain extent.. if the line becomes too noisy, then theres not enough 'spare' for bits to be swapped to... and data may have to be retransmitted... and yes lots of bitswapping can cause a router cpu to work hard.
-
Thx guys :)
-
Could you say that these highway lanes are tones?
-
Yes. :thumbs:
The tones (as in DMT) or discrete channels. ;D
-
Cool :)
Lets say that on an adsl2+ line (with tones going from 33-511) i was downloading a file which was only downloading at half the possible speed, and lets assume that each tone is equal in size (same amount of bits). would this download use half the tones, for example 33-272, or would it use half of each tone, for example all the tones 33-511 but only using half of each 1?
-
ummmm good question. :-[
Ive no idea which tones it would use. DMT just makes the tones available for use if needed. It opens up the lanes on the highways with a maximum speed limit but doesnt dictate which lane the traffic will use.
I suppose theres many instances where traffic doesnt traverse at its full potential say for example you have network cable and a network switch that is gigabit ethernet.. but you are restricted to 10/100 because of the NIC.
-
Ooh. A good question. But one that I do not know the answer. :no:
I best keep watch . . . ready to learn.
-
>> I best keep watch . . . ready to learn
lol me too..
I really dont know.. but to me, logic would seem to point to it filling the bins in order... starting at the lower end of frequencies until each bin is full.
I would imagine the router having to work out and constantly monitor the existing throughput speed, and divide it up between the total no of bits available... and then start calculating say 2 bits per bin then a fraction of a second later 3 bits per bin would add unnecessary additional work for the router.
-
Still none the wiser despite googling... it would appear there are several different methods.. some of the tech stuff is over my head... but it would seem to imply each bin is analysed in order.
However whats the waterfill approach? anyone know what that means?
-----------
edited to add
>> whats the waterfill approach?
I think it means filling the lower (better) channels first... or at least it does when talking about power allocation.
-
If i had to guess i would think that filling each bin in order might create issues if theres alot of bitswapping going on. For example, if the 1st couple of bins are full but bits need to be swapped around in them it could create abit of lag perhaps.
>>> whats the waterfill approach?
But this "waterfill" suggests to me that yea each bin is filled up and then when its starts overflowing it goes into the next bin, :-\
... Unless each bin has tiny little holes in that the water is allowed to seep through, filling each bin equally. ???
L0L
-
>> whats the waterfill approach?
I think it means filling the lower (better) channels first... or at least it does when talking about power allocation.
I think i get that, imagine turning the DMT graph upside down so the more stable bins become deeper channels.
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg848.imageshack.us%2Fimg848%2F8824%2Fdmtwaterfill.png&hash=d8800abb4766097e1712aa5dccbf1cf06d0d3f7d)
-
>>> but bits need to be swapped around in them
bits can be swapped from any subchannel in the whole frequency range, not just adajcent. When the bit table is allocated theres normally some spare SNR reserved for the bit-swapping process.
By filling the early (low frequency) bins first...
1) Usually these are the more stable channels.
2) if the line is only running at half speed... there will be plenty of empty bins doing nothing in the lower end where bits can be swapped from.
3) if the line is running at full pelt all bins are full so it will utilise bins with any spare SNR. At this point power output can be increased to gain a bit more SNR. (real SNR not SNRM)
Bit swapping happens really fast.. and 1 bit of data will pass through so quick, that it will be 'gone' and dealt with in less than the blink of an eye anyhow.
Thinking about it, if you need to take bits from bins.. then IMHO it would be better if possible to 'take' from the lower end (higher frequencies) where things are likely to be less stable anyhow, rather than borrow from the higher frequencies.
Waiting to see if asbokid (or anyone else) can add anything. He may be able to make sense of some pretty complicated algorithms.. cause thats where I tend to get lost. :/
>>> imagine turning the DMT graph upside down
ah... yes I see that :D
-
Can't claim to know anything about bit allocation algorithms but attached are two excerpts from books by those who do know!
Bingham, John (2000) ADSL, VDSL, and Multicarrier Modulation, ch.5: Fundamentals of Multicarrier Modulation
Jacobsen, Krista (2006) Fundamental of DSL Technology (ed Goulden, P, et. al) ch.6: Fundamentals of Multicarrier Modulation
Bingham references John Cioffi, Emeritus Professor of Engineering at Stanford. Cioffi's (old) course website carries several bit allocation algorithms coded in MATLAB:
http://www.stanford.edu/class/ee379c/
cheers, a
-
Thanks asbokid. It was late last night when I was in bed and it dawned on me that the algorithms actually were for bit allocation (ie working out how many bits the subchannels can use) rather than actually loading bits for transmission into the bins.
One of those links you provided though does give more info on the bit swapping process.. and it talks purely about the bit swapping process using power gain/decrease to adjust the SNR. So the case (3) I mentioned above method is used regardless of what bins are loaded with however many bits.
However... Yet another link (http://www.stanford.edu/class/ee379c/) would seem to show a function for the router on how to actually fills the subchannels with bits... which I believe is the key to the answer for kezzamans question :)
Having had a quick look at this code, it would appear the router does indeed use the 'waterfilling' process to decide which bits to load into the subchannels.
This waterfilling process would seem to analyse all the channels, and determine the subchannels with the best SNR as those being the ones to fill first.
-
Do you guys ever check how many bitswaps occur in 1 hour on ur line?
-
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fpgg999.co.uk%2Fstats%2Fimages%2Fst546.bitswap.3d.jpg&hash=8f7b43d4b636d048d3e75a58e817907bf1926247)
-
Does fibreoptic broadband use bitswap?
-
Does fibreoptic broadband use bitswap?
GOOD QUESTION!!
someone much more skilled than I will answer in full and 100% correct, but for now I will attempt some explanation... NO.... lol at least I dont think it does, only FTTC does because it is still a form of xDSL, its fibre to the cab, but from there its VDSL2 to the premises, so i just like bringing the exchange closer to you - Virgin Media's network is a HFC Network (Hybrid Fibre/Coaxial) in that it also has fibre nodes (cabs) near to the home and from these nodes is coaxial copper cable, but these typically have amplifiers en route beyond the nodes towards the premises and inbetween premises which 'boost' the signal power to keep it strong en route to all customers, also the coaxial cable is a much better quality copper cable and has some shielding too, Virgin Media's network uses DOCSIS (as opposed to xDSL) and I THINK this protocol is designed to be used with such a Network (HFC) ad operates at 'Fixed' speeds only..and therefore no bit-swapping is required as its not designed to 'adapt' to the lines conditions (like in xDSL) because the line has to always be in good working order able to receive the max speed (e.g. 100Mbps)...or not... there is no "in-between" where speed can be 'shaped' around the condition of the line ...something like that anyway
there is some info
here -> http://en.wikipedia.org/wiki/Hybrid_fiber-coaxial
and here -> http://en.wikipedia.org/wiki/DOCSIS
And same applys for pure fibre networks where they exist...
A lot of people THINK (due to Virgin Media advertising lies) that they get superfast fibre right into their homes....when they dont. Virgin used to criticise BT's copper network calling it "infrerior" and listing problems with it in full knowledge they too used copper as last mile...just a better type of copper that was less affected due to the way it was being used.
when are OFCOM gunna force Virgin to wholesale its network? cant wait for that day!!! - even though HFC/DOCSIS outperforms xDSL slightly.. I still prefer DSL because its massive acheivement in technology...theyve been able to match Virgin speeds (almost) over 100 year old copper network...without the use of higher grade last mile wiring or amplifiers en route too boost power.
...oh and when Vectored VDSL2 (http://www2.alcatel-lucent.com/blogs/techzine/2012/vdsl2-vectoring-in-a-multi-operator-environment-separating-fact-from-fiction/) comes into play you will see speeds faster than Virgin's current 100Mbps max (over 100Mbps for those 400m and less from the Cab) - Vectored VDSL2 makes use of Vectoring all the Pairs together to cancel out cross-talk at the Dslam by calculating the noise over the pairs and injecting "Anti-Phased" noise into the lines which cancels out the noise over all the lines...enabling even faster speeds ...something like that anyway , for it too work at its best all the lines in the cable need to be vectored, a few strays that arent can affect performance, all you need to do this is upgrade software on Dslam and pack all the VDSL2 pairs in their own shielded cable away from other ADSL1/2/2+ pairs...so it would seem , see video below
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fwww2.alcatel-lucent.com%2Fblogs%2Ftechzine%2Ffiles%2F2012%2F03%2FTZ-Fig-1-v4-Vectoring-restores-VDSL2-lines-to-near-optimal-noise-free-performance.jpg&hash=1dd5f78f0e8c3624da71cd2f85633b48335a07b2) [youtube]http://www.youtube.com/watch?v=zkVmej4urx4[/youtube]
-
>> Does fibre optic cable use bitswap.
No.
>> Does vdsl (FTTC) use bitswap
Yes.
------------------------------------
Bitswap is an essential part of xDSL which uses DMT technology.
The reasons why are all explained in more detail in these sections.
What is DMT (http://www.kitz.co.uk/adsl/adsl_technology.htm#dmt).
What is DMT modulation (http://www.kitz.co.uk/adsl/adsl_technology.htm#dmt_modulation).
What is bit Loading (http://www.kitz.co.uk/adsl/adsl_technology.htm#bit_loading)
What is bit swapping (http://www.kitz.co.uk/adsl/adsl_technology.htm#bit_swapping)
Further info on those pages helps slot understanding into place.
But to simplify things to the most basic of levels :-
~ DMT technology is used between the dslam and CPE (your router)
~ Without DMT, telephone cable to the home is limited to 56kbps,(dial up)
~ DMT modulation is splitting of frequencies so each subchannel can carry 'up to' 56kbps (if the bin is fully bit loaded)... and merging to provide higher bandwidth.
The downside being the higher frequencies are more subject to signal deterioration.
~ DMT modulation is why DSL needs a modem (adsl modem/router) rather than a 'standard' router used for cable broadband.
~ Without bitswapping the line would lose sync if the SNR in any of the subchannels dropped too low and became unstable.
~ Bitswapping can be performed between any of the available subchannels
Cable broadband doesnt use DMT at all.
Optical Fibre uses Frequency Division Multiplexing* or Wave Division Multiplexing.
(See Wave Division Multiplexing (WDM) (http://www.kitz.co.uk/adsl/21cn_network.htm#WDM)).
Although fibre can be subject to attenuation, it is minimal when compared with copper.
Co-ax is used for the entry into the home, which as snadge mentioned is better quality copper than the twisted pair... and use of amplifiers if needed to ensure against signal deterioration. - Think how ethernet doesnt need/use bitswap either. This length of co-ax only has a relatively short run.
*DSL also uses FDM to separate the upstream and downstream channels. DMT splits it into subchannels.
IIRC Bitswapping is a standard in g.DMT not just the rate adaptive products - its interleaving/FEC that doesnt occur on the traditional fixed rate products.
I think adsl2/2+ may use a different/better bitswap algorithm as it handles SNR slightly differently.. and bit loading of min 1 bit per bin (3dB SNRM) when compared to min 2 bits (6dB SNRM) of adsl1.
-----
edited to add
Excellent and informative post by snadge - thank you :)
-
Hi Kitz,
The other day i disabled bitswap and it solved my gaming lag issue, i just can not believe it after all this time :no: :lol:
Now im interested to know does bitswap just have a bad affect on my line or is it everybodys line... (when it comes to gaming).
If you know any gamers on adsl or vdsl get them to disable bitswap and see if it improves for them.
When i did it my connection was alittle jumpy for about an hour and speedtest.net results were all over the place but then it settled down and now... im just so happy ;D
-
Hmmm, four years ago Talktalk rolled out an update to their MSAN's which turned off bitswapping on 999 exchanges. This led to customers experiencing intermittent sync, dropping connection, and a degradation of throughput speeds. Further updates were rolled out over two evenings to put this right.
If bitswapping is not enabled then, as line conditions change, SNR will drop, BER will increase and eventually the line will drop.
-
:-\
Whats BER?
If disabling bitswap causes issue wouldnt raising snr profile fix it?
-
:-\
Whats BER?
BER is the Bit Error Rate (http://en.wikipedia.org/wiki/Bit_error_rate).
-
hmmmm also from me too. Disabling bitswap isnt something I'd ever recommend. :(
Its there for a purpose and without it the line in theory should become more unstable - although saying that it wont if the SNR (not SNRM) across ALL subchannels remains fairly constant, but I would imagine over the course of the day just about every line will experience some fluctuations, bearing in mind we are talking a heck of a lot of subchannels on adsl2+.
Granted at sync time the Bit Allocation Table (BAT) will include a small safety margin for BER, but eventually over time that will deteriorate too.
Even if you do compensate by increased SNRM, the BER is at some point going to cause lag in itself, and eventual deterioration of the line :(
I really dont understand how disabling bitswap can have improved your line, the router will constantly be loading the datastream into the different subchannels as defined by the BAT anyhow - regardless if Bitswapping is on or not.
Bitswapping isnt the shuffling around of data, its actually telling the modem where to load the bits and its a calculation that happens so fast it shouldnt have any impact on the data throughput.
-----------
>>get them to disable bitswap and see if it improves for them
No ty ;D :D
My own SNRM as reported by my router is rock steady - its practically flatlined for the past 4 yrs and never moves by more than .5 dB
To be perfectly frank, I dont think you will find many lines any more stable than mine* (see graph below).
My Error rate over the past 13 days 16hrs is
208 CRC
0 ErrSec
0 FEC (obviously as im not interleaved)
Yet if I switch DMT tool on bitswapping is happening ALL the time... and I can see that the real SNR fluctuates on a couple of the subchannels very briefly for more than what my SNRM is.
Without bitswap my BER would be silly stupid figures... and without a doubt its bitswapping that keeps my line stable at such a low SNRM. I reckon I'd probably last about an hour (if that) before loosing sync. :(
*Admittedly I have an problem with physical line degradation of about 1Mbps of sync per annum that BTw wont do anything about yet because the line is still classed as way above average, but its unrelated to this topic of conversation
-
Hi Kitz, just a quick question (sorry for the hijack), what router do you use and how do you graph your SNRM over time?
-
Hi Kitz, just a quick question (sorry for the hijack), what router do you use and how do you graph your SNRM over time?
Currently using a ST585v6 - which unfortunately is going to have to be retired pretty soon as the network switch doesnt work properly and access is via wireless :(
Although I have a new router, ive not yet installed it, cause I will be really sad to see my faithful ST go.. one of the things I will lose is the MRTG graphing and monitoring which you see above :(
-
But Kitz how many bitswaps to u get in a day or an hour or minute?
-
But Kitz how many bitswaps to u get in a day or an hour or minute?
A fair few.... I dont have any way of measuring, other than monitoring via DMT tool and the bit swap notification that flashes up.
-
The other day i had 720 downstream and..... (drum roll).... 2880 upstream bitswaps in 1 hour...
Thats seems like a ridiculous amount to me.
Other days about 500 downstream per hour and 0 upstream.
and on occassions with bitswap disabled im still getting upstream bitswaps, at a similar rate of 2880 per hour. Its usually either that or zero.
-
upstream is on lower frequencies which attenuate less, they are strognger and less susceptible to noise unlike the higher frequencies - so it would seem odd that your having so many bit-swaps in upstream... infact (i may be wrong) but as upstream is such a small amoutn of total bandwidth sitting on the strong part of the spectrum there should ZERO (or very little bit swapping) - this to me would indicate a problem you have on your line thats causing massive amounts of noise on the low end... the upstream doesnt carry many bits too start with and your having almost 3000 swaps per hour??? sounds like a fault/problem to me
can you post up you connection stats and SNR/BitSwap/QLN/Hlog graphs using our graphing scripts (http://forum.kitz.co.uk/index.php/topic,11216.0.html) (assuming you have a compatible netgear) - if not, but you can use DMT Tool, I could graph them out for you if you hand me the logs.. let me know if you can run DMT Tool and will send you the destructions on what to do... you could also post up a screenshot of DMT in the meantime
...just too add, if your getting bit-swaps with it turned off...then it aint really turned off , also, if it were turned off then its likely your connection would just degrade and keep doing so until its unable to sync - bits/tones will become un-usable and will not be recovered because BS is turned off - i think anyway
-
>> he other day i had 720 downstream and..... (drum roll).... 2880 upstream bitswaps in 1 hour...
TBH I really dont know for sure.. but Im unsure if it is excessive. Bitswapping happens so very fast. Bit swap is one of the first lines of defence in keeping a line stable and therefore I wouldnt be too surprised if a bitswap count far exceeds any error count on some lines on some occasions... I guess it depends on the individual line and the type of noise fluctuation?
Added:- just found this (http://flylib.com/books/en/3.206.1.25/1/)
Eventually, depending on the speed of swapping (usually 20-100 swaps per second is a good number), the optimum 19.3 dB will be restored, although standards only mandate an ability to implement 1.25 swaps/second.
However, a vendor's modem can request swaps more frequently and if they are implemented by the other modem, profit from the increased speed of swapping. For 20 - 100 swaps/second, the new bit distribution in our example would be implemented in a couple of seconds without any bit errors or loss of service or additional errors (beyond those that occur for other reasons). This bit swapping process is seamless without errors, and without service interruption or need for retraining.
Even your highest figure is within the 1.25 swaps per second.
>> infact (i may be wrong) but as upstream is such a small amoutn of total bandwidth sitting on the strong part of the spectrum there should ZERO (or very little bit swapping)
Funny enough I suspect (but cant prove) that most of my bitswap occurs on my upstream - as my upstream SNRM is more variable than my down.
It would only take tiny fluctuations of SNR for bitswap to kick in, and if you have a few subchannels that constantly fluctuate (rather than a specific noise burst) it could materialise as lots of bitswaps but show no CRC or more serious type errors.
>> if your getting bit-swaps with it turned off...then it aint really turned off , also, if it were turned off then its likely your connection would just degrade and keep doing so until its unable to sync - bits/tones will become un-usable and will not be recovered because BS is turned off - i think anyway
Totally agree.
-------------
Edited to add.
Dont know enough about this.. and Im out of time now and have to go out.
but a thought just occurred.. no doubt someone will correct me if Im wrong, but from scanning the link, its possible that low level cross talk could be causing the bitswapping?
This all be moot anyhow.
-
This is my SNRM with the bitswapping stoppping at 05:44 after i disable it with command - adsl configure --bitswap off
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg39.imageshack.us%2Fimg39%2F5144%2Fupbitswapbeforeandafter.jpg&hash=161e4e4a61288805752dad3b1fb033d750693c21)
When i disable bitswapping it will only sometimes stop bitswapping on the upstream like i said before, but it always stops on the downstream.
Snadge i struggled to get those graphing scripts to work, the reason why i got them was to record a bitswap graph but i worked it out manually, i get approx 48 bitswaps per minute or none at all on upstream.
Heres my stats, 1st image is 2 mins after a reboot the 2nd is 4 hours (bitswap disabled):
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg69.imageshack.us%2Fimg69%2F9320%2Fafterreboot.png&hash=6b621614031a9407e3cc144f2486a770f92ad3cf)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg31.imageshack.us%2Fimg31%2F6889%2Fupstreambitswaps.png&hash=4b78c94d73911cdb2675e548f98f9adae5c23bc7)
-
Just a sudden thought.
The US frequencies are lower than those of DS. In other words, they are somewhat "closer" to the very low frequencies used by the human voice. Could a substandard (or failing) micro-filter / centralised filter / SSFP be responsible? Would a "swap-out" prove to be a useful experiment? :-\
-
Just a sudden thought.
The US frequencies are lower than those of DS. In other words, they are somewhat "closer" to the very low frequencies used by the human voice. Could a substandard (or failing) micro-filter / centralised filter / SSFP be responsible? Would a "swap-out" prove to be a useful experiment? :-\
Wouldn't that only apply though if the bitswapping was occuring when the 'phone was in use ? You can use an ADSL(2/+) modem/router on a line without a filter as long as you don't want to use a phone on it too.
-
Wouldn't that only apply though if the bitswapping was occuring when the 'phone was in use ? You can use an ADSL(2/+) modem/router on a line without a filter as long as you don't want to use a phone on it too.
Well, yes, that is utterly true. Including VDSL2. After all, the "filter" (whatever sort it may be) is only in series with the telephony connection.
Hmm . . . But what if the passive components of the filter are "breaking down"? ???
-
Added:- just found this (http://flylib.com/books/en/3.206.1.25/1/)
Eventually, depending on the speed of swapping (usually 20-100 swaps per second is a good number), the optimum 19.3 dB will be restored, although standards only mandate an ability to implement 1.25 swaps/second.
However, a vendor's modem can request swaps more frequently and if they are implemented by the other modem, profit from the increased speed of swapping. For 20 - 100 swaps/second, the new bit distribution in our example would be implemented in a couple of seconds without any bit errors or loss of service or additional errors (beyond those that occur for other reasons). This bit swapping process is seamless without errors, and without service interruption or need for retraining.
Even your highest figure is within the 1.25 swaps per second.
I was thinking about this 1.25 per second figure, do you think they might have just been talking about Downstream with its 478 Bins because it did mention 19.3dB which im guessing is output power?
So what if it depends how many bins are in use for calculating whats a reasonable amount of bitswaps per second...
I am just guessing but if what im thinking is true, then by dividing 1.25 by 478 we can work ot how many bitswaps per second per bin is acceptable and then multiply that by the number of bins in the upstream (25) we can compare it to my bitswap per second.
1.25 / 478 x 25 = 0.065 Bitswaps per second
and im getting approximately 0.8 bitswaps per second
Obviously all this could be irrelevant if my theory about number of bins mattering, but its just something i thought about.
-
its possible that low level cross talk could be causing the bitswapping?
Say if it is crosstalk what show i do next?
-
Did you get disconnected as the DS connection rate (19695 Kbps) is higher in the first picture than the second (19563 Kbps).
-
Did you get disconnected as the DS connection rate (19695 Kbps) is higher in the first picture than the second (19563 Kbps).
those differences dont mean anything really...its never exactly the same each re-sync, had it been several hundred KB's or MB's then there may be cause for concern,
does the bit-swapping affect the actual upload throughput? have you tried upload testing? you should get about 0.8Mbps with a 1050k sync (I get 0.9Mbps with 1170k sync)
I dunno what problems you have with the graphs.. if you followed the instructions and installed the folder too the root of the C: Drive (not D or any other letter) then the scripts should work , well, if your using a netgear and login is admin/password
-
>>> Did you get disconnected as the DS connection rate (19695 Kbps) is higher in the first picture than the second (19563 Kbps).
No.
>>> does the bit-swapping affect the actual upload throughput? have you tried upload testing? you should get about 0.8Mbps with a 1050k sync (I get 0.9Mbps with 1170k sync)
No i get about 0.84 which is about right.
I could proberly get ur graphs to work if i went back and followed the instructions properly, but what information are you hoping to get from it?
-
the QLN graph may show something (Quiet Line Noise - Router Stats and DMT dont graph these important stats)
it may show something happening on low end which may tie in with the over actiove bitswapping on your low end
-
I followed the instructions for using those graphs.
The graphs appeared in the current stats folder but they dont seem to be plotting anything, i ran the start log the cmd script ran but after that the graphs stayed blank. I dont know what im doing wrong...
My router is netgear dg834gt.
(the folder is in my C: drive folder.
-
your folder should be like:
C:\ADSL_Stats\Scripts
you havent named the root folder yourself have you? it must be structured like so
C:\ADSL_Stats\
in which is
C:\ADSL_Stats\Scripts
C:\ADSL_Stats\Apps
C:\ADSL_Stats\Current_Stats
C:\ADSL_Stats\Ongoing_Stats
if its not structured like that (e.g. C:\ADSL_Stats\ADSL_Stats\Scripts etc) then it will not work because the scripts look for files specifically targetted with fixed urls.
is your router login
name: admin
password: password
it should be for the scripts too work
did you download & install v1.2 of the scripts? (latest version)
sounds like a log in problem too me
-
When i run the "get current stats" i did notice it say in the cmd bit something like, couldnt find folder C:\ADSL_stats, creating new 1 and then later on it would say somthing about adslctl info --stats not working so then it choses default with 256 tones.... something like this anyway...
Ill have another go.
UPDATE:
"C:\ADSL_stats" DOES exist
... but subfolder does not exsist, i guess thats normal? so then it makes the subfolder but then again when it comes to the adslclt info --stats bit it doesnt seem to work, is this maybe the problem?
the username and pass match the 1s in "apps"
-
When i run the "get current stats" i did notice it say in the cmd bit something like, couldnt find folder C:\ADSL_stats, creating new 1 and then later on it would say somthing about adslctl info --stats not working so then it choses default with 256 tones.... something like this anyway...
Ill have another go.
UPDATE:
"C:\ADSL_stats" DOES exist
... but subfolder does not exsist, i guess thats normal? so then it makes the subfolder but then again when it comes to the adslclt info --stats bit it doesnt seem to work, is this maybe the problem?
the username and pass match the 1s in "apps"
I take it you are getting to this point then:-
************************************************************************
* The ADSL Mode could not be determined from "adslctl info --stats" *
* The data is either not present, or is unreadable *
************************************************************************
Defaulting to produce graphs scaled for G.DMT Mode (256 Tones)
In this folder - C:\ADSL_Stats\Current_Stats
You should see at least one subfolder named something like - Current_Stats_20120607-1705
In that subfolder you should see, at the very least, a text file named something like - Plink_20120607-1705.log
If that is there, double-click it to see what it contains.
It should start off something like:-
=~=~=~=~=~=~=~=~=~=~=~= Plink log 2012.06.07 17:05:31 =~=~=~=~=~=~=~=~=~=~=~=
BusyBox v1.00 (2008.10.21-10:23+0000) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
# adslctl info --Bits
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 0
Channel: FAST, Upstream rate = 1153 Kbps, Downstream rate = 13054 Kbps
Tone number Bit Allocation
0 0
1 0
2 0
3 0
4 0
5 0
Are you getting at least that far?
-
It was a login issue, even though i use admin as the username to login onto my router there seemed to be another username, which was my ISP login something like ....@talktalk.net. But anyway i got "get current stats" to work in the end and i got all graphs except, bit-loading..
This is what i got from plink regarding adslctl info --bits:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg52.imageshack.us%2Fimg52%2F3909%2Fplinknotfound.png&hash=a371454f39dbf66e4975eb36c0bec7343ba17514)
Also would anybody be able to tell me what LATN(dB) and SATN(dB) means?
-
Just another thought..............
In this folder "C:\ADSL_Stats\Ongoing_Stats" there should be a plain text file named "Error.Log"
There may be a clue or two in there as to what is not working.
-
It was a login issue, even though i use admin as the username to login onto my router there seemed to be another username, which was my ISP login something like ....@talktalk.net. But anyway i got "get current stats" to work in the end and i got all graphs except, bit-loading..
This is what i got from plink regarding adslctl info --bits:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fimg52.imageshack.us%2Fimg52%2F3909%2Fplinknotfound.png&hash=a371454f39dbf66e4975eb36c0bec7343ba17514)
Also would anybody be able to tell me what LATN(dB) and SATN(dB) means?
Ah. It looks like you need a newer firmware version for the router, either official Netgear or DGTeam should do it.
snadge may be able to assist with that.
From what I can gather, older firmware versions do not report the Bits data, even via Telnet access.
Although that logging in still looks a little dodgy.
LATN(dB) = Line Attenuation
SATN(dB) = Signal Attenuation.
Can we see the graphs you have managed to create so far?
-
Im pretty sure i have the latest official netgear firmware and i just tried the command adslctl info --Bits in telnet and it worked fine. Looking at the plink log isnt it strange to have the password joined up with the command like that? Ive had both the dgteam firmware and the official firmware at some point and did u know with the official firmware telnet doesnt require a login? Maybe thats whats causing issues?
Here are the graphs i got:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg513%2Fscaled.php%3Fserver%3D513%26amp%3Bfilename%3Dqln201206080831.png%26amp%3Bres%3Dlanding&hash=97d13f691b5cd910dc905018708e496d7c0466cd)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg560%2Fscaled.php%3Fserver%3D560%26amp%3Bfilename%3Dsnr201206080831.png%26amp%3Bres%3Dlanding&hash=7a5316de2db84f7d9d86050ecbe9468ccc56c263)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg442%2Fscaled.php%3Fserver%3D442%26amp%3Bfilename%3Dhlog201206080831.png%26amp%3Bres%3Dlanding&hash=35c4b090889dcaa2c3ac909422f340e39218af03)
-
Im pretty sure i have the latest official netgear firmware and i just tried the command adslctl info --Bits in telnet and it worked fine.
That's good then.
Looking at the plink log isnt it strange to have the password joined up with the command like that?
That's the part that is bothering/confusing me.
I don't have an ADSL connection to test any of this on, but I THINK that could just be due to the scripts initially being written with half a mind on making it easy to convert for Linux use, an attempt was made to strip Windows line endings from the collected data.
Could you copy & paste the first few rows from a Plink log for me to examine more closely (as actual raw text rather than a graphic)?
Ive had both the dgteam firmware and the official firmware at some point and did u know with the official firmware telnet doesnt require a login? Maybe thats whats causing issues?
Yes, I'm aware of that, but I thought the attempted login would be quietly ignored when using "official" firmware.
I could soon tweak the script for official firmware, but I would somehow like to be able to have just the one script that works with official AND DGTeam firmware.
I thought that snadge had tested with both firmwares as O.K.
The montages should be small enough to post as downloadable attachments when we have ironed out these last little snags.
-
Could you copy & paste the first few rows from a Plink log for me to examine more closely (as actual raw text rather than a graphic)?
I think this is what you meant?
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg220%2Fscaled.php%3Fserver%3D220%26amp%3Bfilename%3Dplink1.png%26amp%3Bres%3Dlanding&hash=c61f7ddb5a6cfbd60bf8380df60ad7cb2de886bb)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg341%2Fscaled.php%3Fserver%3D341%26amp%3Bfilename%3Dplink2.png%26amp%3Bres%3Dlanding&hash=d37a0f99b12fca86d5d0ff4311f5f0ee725d336a)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg593%2Fscaled.php%3Fserver%3D593%26amp%3Bfilename%3Dplink3.png%26amp%3Bres%3Dlanding&hash=e1ec3965f7d6932a1f2928727b413aa49d229d7a)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fdesmond.imageshack.us%2FHimg27%2Fscaled.php%3Fserver%3D27%26amp%3Bfilename%3Dplink4.png%26amp%3Bres%3Dlanding&hash=eaa1e29f75d6728a0164d30c502ba40173233f43)
-
I actually meant like this:-
=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2012.05.14 20:28:57 =~=~=~=~=~=~=~=~=~=~=~=
Login: admin
Password:
BusyBox v1.01 (2008.11.28-10:31+0000) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
# adslctl info --Bits
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 0
Channel: INTR, Upstream rate = 448 Kbps, Downstream rate = 6944 Kbps
Tone number Bit Allocation
0 0
1 0
Or like this (i.e. text that I can actually copy & paste into Notepad etc.):-
=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2012.05.14 20:28:57 =~=~=~=~=~=~=~=~=~=~=~=
Login: admin
Password:
BusyBox v1.01 (2008.11.28-10:31+0000) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
# adslctl info --Bits
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 0
Channel: INTR, Upstream rate = 448 Kbps, Downstream rate = 6944 Kbps
Tone number Bit Allocation
0 0
1 0
-
=~=~=~=~=~=~=~=~=~=~=~= Plink log 2012.06.08 08:31:48 =~=~=~=~=~=~=~=~=~=~=~=
BusyBox v1.00 (2008.10.21-10:23+0000) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
# admin
/bin/sh: admin: not found
# passwordadslctl info --Bits
/bin/sh: passwordadslctl: not found
#
# adslctl info --linediag
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 8000
Channel: FAST, Upstream rate = 1019 Kbps, Downstream rate = 19864 Kbps
Down Up
SNRM(dB): 8.9 11.5
LATN(dB): 14.5 6.6
SATN(dB): 14.5 7.5
TxPwr(dBm): 0.0 11.7
ATTNDR(Kbps): 23008 1108
Tone number SNR
0 0.0000
1 0.0000
2 0.0000
Like this?
-
Hmmmm.
It wasn't that then.
It looks like "admin" is correctly ignored (not found), but "password" & "adslctl info --Bits" have been joined together.
Have you edited LOGIN2.TXT at all?
As you don't need to login using your official Netgear firmware, you could try the attached version of the script to see if that works for you.
-
I just ran that official get stats, it created a current stats folder but that was it, nothing inside.
I did change the login originaly because i was using a different password, but i changed it back to "password" on both my router and "login2.txt" when i was trying to get it to work.
Perhaps i should go back to login2 and leave a space after password... :-\
-
Nope leaving a space after the password didnt work but i managed to get u this if it helps.
The ADSL MODE was determined to be ADSL2+
Creating Bits-20120608-1329.png
gnuplot> plot 'Bits.txt' w boxes fs solid
^
line 0: warning: Skipping data file with no valid points
-
The attached logins seem to work with Netgear Firmware & the script as downloaded via snadge's link (not the one I posted earlier today).
Spaces are no use. It is more to do with Linux/Windows line endings in the Login.TXT files.
This is how the Plink log should look (everything on its own line & showing admin & password as ignored):-
=~=~=~=~=~=~=~=~=~=~=~= Plink log 2012.05.20 12:01:25 =~=~=~=~=~=~=~=~=~=~=~=
BusyBox v1.00 (2005.07.29-07:22+0000) Built-in shell (ash)
Enter 'help' for a list of built-in commands.
# admin
/bin/sh: admin: not found
# password
/bin/sh: password: not found
# adslctl info --Bits
-
I cant open or unzip that file
! C:\Users\kezza\Downloads\Logins.zip: The archive is either in unknown format or damaged
-
It opens from the attachment for me.
Possibly something at your end?
I have attached it again though, just in case.
-
Yep its all working now ;)
I wanted to ask you, should i be concerned that the LATN and SATN on the upstream arent equal because they r on the downstream?
-
Yep its all working now ;)
Phew!
So, to recap, you have a different username & password for accessing the modem's normal GUI where you can view/change the usual settings?
"admin" & "password" are still used for accessing stats directly via telnet or the scripts, except username & password are just ignored (not found) because you are using official Netgear firmware?
Let's see the evidence then, attached as a portrait montage (as an attachment, not a link) :)
I wanted to ask you, should i be concerned that the LATN and SATN on the upstream arent equal because they r on the downstream?
I'm not sure regarding ADSL. Mine aren't equal on my VDSL2 connection, but that happened very suddenly, so it might be a physical cable problem - I don't know yet.
Now you have done the easy bit, have you got the ongoing stats (modem_stats.log) being updated every minute?
-
i was just about too say... if your router is from TALK TALK you will NOT have official Netgear firmware, however if your usernam and password was
admin
password
theres no reason it shouldnt work. also, if you edit the LOGIN files yourself SOMETIMES they dont work because they are formatted in special format (cant remember name of ASCII was it?) so they work, when you save them in windows notepad SOMETIMES they no longer work, sometimes they do... so if you edited LOGIN1 , LOGIN2 then theres a chance they wont work as the formatting is removed, I had this happen to me but second time I tried it they worked.. this is why the versions BA sent you work...cos they will have the proper formatting and reason your not getting bit loading is probably cos its Talk Talk firmware
you should install Netgear FW, I can supply links to the files - as its TT you may have to flash it from Windows XP with proper flah util, unsure though
anyway, your QLN graph looks alright too me, your line looks good...
-
@ snadge,
You're with Sky aren't you?
Don't they have "special" username & passwords?
Do you use a different GUI username & password to the ones you use for the scripts (admin & password), or have you changed the ones used by the scripts to match the GUI username & password?
-
if you edit the LOGIN files yourself SOMETIMES they dont work because they are formatted in special format (cant remember name of ASCII was it?) so they work, when you save them in windows notepad SOMETIMES they no longer work, sometimes they do... so if you edited LOGIN1 , LOGIN2 then theres a chance they wont work as the formatting is removed
Linux lines are terminated with a line feed (lf) or in linux terminology "\n"
Windows/DOS lines are terminated with carriage return & linefeed (CRLF) or in linux "\r\n"
Ditch notepad and use textpad instead, it can save in the required format thus preserving (or changing) the line endings.
-
Iam with sky and sky's default login for their routers is
admin
sky
I dont use that, also sky dont let you telnet the routers data either.
the ones on the script are default (which is what I use on my non-sky netgears)
admin
password
his troubles where because of what i explained, he had different login, but then editing the login1 & login2 files would have stopped them from working due to the formatting issue, all he had to do was change the login info in his router to:
name:admin
password:password
and would work - but as his router is from Talk Talk it will have a Talk Talk fw on it - I think anyway, ISP's normally have their own version of fw installed... ive not gotten a router yet that hasnt - oh except my first 834G from AOL
Tiscali
o2
Sky
all had their own firmwares
@ BlackEagle = cheers, is that what it was? following advice from someone else I had to use Notepad++ and convert them into something, cant remember, the posts on forum somewhere - I later re-saved them in notepad and they didnt work, but later I done it again and they did work? wierd...
-
Ditch notepad and use textpad instead, it can save in the required format thus preserving (or changing) the line endings.
I'll second that sentiment.
I now use Textpad for most plain text file processing & found it really useful for developing the batch files, converting Linux stuff that looked like it was all on one line (when viewed with Windows Notepad) to a readable/saveable format, reading binary files, using line numbering, the cursor remembering which column it is in, search & replace etc. etc. etc.
Apart from the odd reminder to purchase, it is completely free & uncrippled.
-
-------------
Edited to add.
Dont know enough about this.. and Im out of time now and have to go out.
but a thought just occurred.. no doubt someone will correct me if Im wrong, but from scanning the link, its possible that low level cross talk could be causing the bitswapping?
This all be moot anyhow.
Doesnt crosstalk only affect the higher frequencies?
-
FEXT (http://en.wikipedia.org/wiki/Crosstalk_%28electronics%29#Crosstalk_in_cabling) can happen at any frequency, its just the higher the frequency, the more susceptible it is to XT because the higher frequencies attenuate more and are therefore 'weaker'
-
FEXT (http://en.wikipedia.org/wiki/Crosstalk_%28electronics%29#Crosstalk_in_cabling) can happen at any frequency, its just the higher the frequency, the more susceptible it is to XT because the higher frequencies attenuate more and are therefore 'weaker'
If it was FEXT crosstalk would that mean it was coming from inside or around my house?
-
crosstalk happens in the cable between neighbouring pairs...so if it was XT then it would be in the run of cable somewhere on the line, FEXT is nearer the customers home NEXT is nearer the DSLAM/MSAN
-
Would the cabinet be classed as FEXT?
Just to clarify when FEXT does affect the lower frequencies would it have to affect the higher 1s to, or could it just leave the higher 1s untouched?
-
the cabinet is just a patch point when its non-VDSL2 - however, if its VDSL2 then NEXT would be the cab and FEXT still the customers home, VDSL2 brings the DSLAM up into the cab
XT could happen at any frequency as far as Im aware, maybe someone else like BS knows more about it - but Interference could happen at any point in the spectrum.
-
As snadge says.. it can happen anywhere across the frequency range.