1. Welcome to Verizon Forums - the unofficial Verizon community! Have a question about Verizon? Click HERE to get started.
    Dismiss Notice
  2. Expecting Cell Phone Forums? We recently moved Verizon specific content to VerizonForums.com. If you previously had an account on CPF, it has been transferred!
    Dismiss Notice

How fast is the Verizon wireless service for my laptop? (CT/NYC)

Discussion in 'alt.cellular.verizon' started by Kab, Mar 11, 2004.

  1. Jesse McGrew

    Jesse McGrew Guest

    N9WOS wrote:

    > From Washington to San Francisco California to Palo alto California to
    > Washington
    > to Tokyo in 160ms.
    > and a total distance traveled of about 20,000 miles.
    > I wish all normal internet traffic was that fast.
    > but it is not.
    >
    > You are using a tracert program that sends packets out that are
    > marked as time sensitive.
    > (ie) VoIP, and the like.
    > Normal WWW traffic flow through the system is not that fast.
    > What you are seeing with those numbers is the speed that
    > voice data is routed.


    I'm using the tracert program that comes with Windows XP. I get the same
    results with the diagnostic traceroute on my Linksys router, and with
    the standard traceroute on a Linux box at work (on a T1, four hops from
    Verio). What traceroute program do you suggest I use?

    Jesse
     



    › See More: How fast is the Verizon wireless service for my laptop? (CT/NYC)
  2. N9WOS

    N9WOS Guest

    > I'm using the tracert program that comes with Windows XP. I get the same
    > results with the diagnostic traceroute on my Linksys router, and with
    > the standard traceroute on a Linux box at work (on a T1, four hops from
    > Verio). What traceroute program do you suggest I use?
    >
    > Jesse


    This one here.
    http://www.pingplotter.com/downloads/pngplt_1.exe
    is one of them.

    The high priority, and normal priority packet decision is
    something the programmer makes when he writes the software.
    They don't say what types of packets they use,
    but the results sure tell the difference.
    One gives the user a nice fuzzy feeling (Ow look at the speed),
    the other is more realistic.

    The time of the high priority ones mainly tell you
    the distance between the jumps.
    But the normal priority ones mainly tell you what
    the actual day to day speed of the system is.
    Ping plotter is mad to test and plot variations in speed from
    hour to hour, and day to day, which depends on network loading.
    And to depict trends in link speed change.
    To do that, he has to use normal priority packets in the ping engine.

    Here is the tracert on my xp machine that uses high priority packets.
    The tracert that comes with the computer.

    C:\>tracert 129.250.4.174

    Tracing route to p16-0-2-2.r21.tokyjp01.jp.bb.verio.net [129.250.4.174]
    over a maximum of 30 hops:

    1 184 ms 187 ms 185 ms 199.69.83.73
    2 187 ms 187 ms 187 ms 199.69.83.65
    3 296 ms 187 ms 187 ms gbr1-p51.sl9mo.ip.att.net [12.122.252.229]
    4 296 ms 187 ms 187 ms tbr2-p013502.sl9mo.ip.att.net
    [12.122.11.113]
    5 187 ms 187 ms 187 ms tbr2-cl7.cgcil.ip.att.net [12.122.10.45]
    6 187 ms 187 ms 187 ms ggr2-p390.cgcil.ip.att.net [12.123.6.37]
    7 187 ms 187 ms 187 ms att-gw.chi.verio.net [192.205.32.194]
    8 234 ms 223 ms 233 ms p16-0-1-1.r21.sttlwa01.us.bb.verio.net
    [129.250.
    2.6]
    9 234 ms 233 ms 234 ms p16-7-0-0.r81.sttlwa01.us.bb.verio.net
    [129.250.
    2.239]
    10 327 ms 331 ms 327 ms p16-0-2-2.r21.tokyjp01.jp.bb.verio.net
    [129.250.
    4.174]

    Beyond the 188ms delay at the modem, I only have about 130ms delay,
    for the rest of the round trip.

    Here is the normal priority tracert
    same time, same computer, same connection.

    Target Name: p16-0-2-2.r21.tokyjp01.jp.bb.verio.net
    IP: 129.250.4.174
    Date/Time: 3/13/2004 11:33:08 AM

    1 182 ms [199.69.83.73]
    2 219 ms [199.69.83.65]
    3 249 ms gbr1-p51.sl9mo.ip.att.net [12.122.252.229]
    4 280 ms tbr2-p013502.sl9mo.ip.att.net [12.122.11.113]
    5 312 ms tbr2-cl7.cgcil.ip.att.net [12.122.10.45]
    6 343 ms ggr2-p390.cgcil.ip.att.net [12.123.6.37]
    7 404 ms att-gw.chi.verio.net [192.205.32.194]
    8 453 ms p16-0-1-1.r21.sttlwa01.us.bb.verio.net [129.250.2.6]
    9 499 ms p16-7-0-0.r81.sttlwa01.us.bb.verio.net [129.250.2.239]
    10 580 ms p16-0-2-2.r21.tokyjp01.jp.bb.verio.net [129.250.4.174]

    After the 180ms modem delay, I have 400ms round trip delay.

    Here is one to goggle.

    C:\>tracert www.google.com

    Tracing route to www.google.akadns.net [216.239.41.104]
    over a maximum of 30 hops:

    1 183 ms 187 ms 187 ms 199.69.83.73
    2 187 ms 187 ms 186 ms 199.69.83.65
    3 187 ms 187 ms 297 ms gbr1-p51.sl9mo.ip.att.net [12.122.252.229]
    4 187 ms 187 ms 187 ms tbr2-p013502.sl9mo.ip.att.net
    [12.122.11.113]
    5 187 ms 187 ms 312 ms tbr2-cl7.cgcil.ip.att.net [12.122.10.45]
    6 187 ms 187 ms 187 ms ggr2-p3120.cgcil.ip.att.net [12.123.6.69]
    7 187 ms 187 ms 187 ms dcr1-so-3-3-0.Chicago.cw.net [208.175.10.93]
    8 202 ms 202 ms 327 ms dcr1-loopback.Washington.cw.net
    [206.24.226.99]

    9 202 ms 203 ms 201 ms bhr1-pos-10-0.Sterling2dc3.cw.net
    [206.24.238.38
    ]
    10 202 ms 202 ms 202 ms csr11-ve240.Sterling2dc3.cw.net
    [216.109.66.82]

    11 203 ms 217 ms 218 ms 218-google-exodusdc.exodus.net
    [216.109.88.218]

    12 218 ms 219 ms 1843 ms 64.233.174.133
    13 217 ms 202 ms 219 ms 216.239.49.37
    14 217 ms 219 ms 217 ms 216.239.41.104

    Trace complete.

    besides the 187ms modem delay, I just got
    a sub 30ms google ping.
    Yea right!!!!!
     
  3. Jesse McGrew

    Jesse McGrew Guest

    N9WOS wrote:
    >>I'm using the tracert program that comes with Windows XP. I get the same
    >>results with the diagnostic traceroute on my Linksys router, and with
    >>the standard traceroute on a Linux box at work (on a T1, four hops from
    >>Verio). What traceroute program do you suggest I use?
    >>
    >>Jesse

    >
    >
    > This one here.
    > http://www.pingplotter.com/downloads/pngplt_1.exe
    > is one of them.


    I hate to break it to you, but I get nearly identical results with that
    program:

    Target Name: p16-0-2-2.r21.tokyjp01.jp.bb.verio.net
    IP: 129.250.4.174
    Date/Time: 3/13/2004 12:46:46 PM

    1 3 ms 0 ms 0 ms 0 ms [192.168.1.1]
    2 * * * * [-]
    3 13 ms 9 ms 10 ms 9 ms [12.244.20.97]
    4 16 ms 17 ms 15 ms 18 ms [12.244.0.17]
    5 23 ms 17 ms 15 ms 19 ms [12.118.106.5]
    6 17 ms 16 ms 17 ms 17 ms tbr1-p012402.st6wa.ip.att.net
    [12.122.5.174]
    7 33 ms 34 ms 33 ms 33 ms tbr2-cl1.sffca.ip.att.net
    [12.122.12.113]
    8 33 ms 32 ms 38 ms 35 ms ggr2-p390.sffca.ip.att.net
    [12.123.13.194]
    9 35 ms 34 ms 35 ms 35 ms
    p16-0-1-1.r20.plalca01.us.bb.verio.net [129.250.9.73]
    10 34 ms 33 ms 34 ms 33 ms
    xe-0-2-0.r21.plalca01.us.bb.verio.net [129.250.4.231]
    11 56 ms 63 ms 56 ms 56 ms
    p16-0-1-1.r20.sttlwa01.us.bb.verio.net [129.250.5.82]
    12 79 ms 77 ms 79 ms 77 ms
    p16-6-0-0.r81.sttlwa01.us.bb.verio.net [129.250.2.219]
    13 156 ms 151 ms 152 ms 151 ms
    p16-0-2-2.r21.tokyjp01.jp.bb.verio.net [129.250.4.174]


    Target Name: yahoo.com
    IP: 66.218.71.198
    Date/Time: 3/13/2004 12:50:30 PM

    1 1 ms 1 ms 0 ms 0 ms 0 ms [192.168.1.1]
    2 * * * * * [-]
    3 14 ms 8 ms 9 ms 8 ms 8 ms [12.244.20.97]
    4 16 ms 24 ms 16 ms 16 ms 17 ms [12.244.0.17]
    5 16 ms 16 ms 18 ms 22 ms 18 ms [12.118.106.5]
    6 27 ms 26 ms 18 ms 18 ms 25 ms
    tbr2-p013802.st6wa.ip.att.net [12.122.5.158]
    7 21 ms 17 ms 15 ms 18 ms 20 ms ggr1-p3100.st6wa.ip.att.net
    [12.123.44.133]
    8 17 ms 27 ms 16 ms 16 ms 17 ms att-gw.sea.level3.net
    [192.205.32.22]
    9 24 ms 17 ms 16 ms 23 ms 19 ms
    ge-6-1-0.mp1.Seattle1.Level3.net [209.247.9.49]
    10 35 ms 36 ms 34 ms 34 ms 34 ms unknown.Level3.net
    [209.247.9.118]
    11 34 ms 33 ms 33 ms 35 ms 33 ms
    ge-9-0.ipcolo3.SanJose1.Level3.net [64.159.2.9]
    12 35 ms 38 ms 37 ms 34 ms 35 ms unknown.Level3.net
    [64.152.69.30]
    13 33 ms -32764 ms -32764 ms -32764 ms 42 ms
    UNKNOWN-66-218-82-226.yahoo.com [66.218.82.226]
    14 34 ms 35 ms 54 ms 34 ms 35 ms w1.rc.vip.scd.yahoo.com
    [66.218.71.198]

    Maybe you should sign up with Comcast if these results are such a
    surprise to you. ;)

    Jesse
     
  4. N9WOS

    N9WOS Guest

    > > This one here.
    > > http://www.pingplotter.com/downloads/pngplt_1.exe
    > > is one of them.

    >
    > I hate to break it to you, but I get nearly identical results with that
    > program:


    Hmmm.......interesting.
    That tells me that the isp has probably fixed
    the TOS(type of service) setting that your computer
    attaches to packets on the internet.
    Probably has something to do with the routing overhead
    of the cable system.

    > Maybe you should sign up with Comcast if these results are such a
    > surprise to you. ;)


    Not a chance in hell. :)
    BTW, no cable company serves are area.
    They think we are not worth their time.
    And they can shove their cable up their A$$ for all I care. :-o
     
  5. I was about to let this go - but the ridiculous assertions and facts in this
    will probably lead to a bunch of other folk having no idea what they're
    talking about...so better to set the record straight (and if you read on,
    you'll see how this might affect cellular folk).

    So here are some facts:

    - Traceroute does not use some sort of high priorty packet. Even if they
    did, the fact that the internet has some level of "QoS" (quality of service)
    implemented in any but the most specific of circumstances is pure folly.
    There's definitely no cross provider QoS occuring as of yet. Traceroute (an
    age old program) uses UDP or ICMP packets - in the former, no service
    guarantees exist at all (the fact they get delivered at all is a miracle) -
    in the latter, may parts of the internet block such packets.

    - Sure, there may be versions of traceroute out there that try to use higher
    priority packets - but they are of limited usefulness (only on those
    networks that support the type of QoS they support) and certainly none of
    the common linux/windows/unix/macintosh standard traceroutes (or tracerts)
    are of that ilk (all of them in fact share a common code base whose
    operation is well known, has been around for many years, and works with no
    priority whatsoever).

    - Traceroute is really a lousy tool for determining anything to a laymen
    except perhaps that there's a potential (I say potential) for high latency
    on a path and perhaps the location of a network outage. These potentials
    are only realized if a) all the hops on the route actually respond to ICMP
    or the UDP actually makes it through all the hops; and more importantly b)
    the traceroute shows you the same route your packets actually take (those
    who know the quagmire that is/was springnet understand this).

    - Traceroute over wired networks is tough - over cell networks, is absolute
    insanity. Think about it (and N9WOS - shame on you for not focusing on
    this) - networks communicate via radio waves - whether they be over a wire
    (traditional networks) or over the air (802.11, cellular and others). We'll
    leave the explanation of other wireless types for more appropriate forums -
    but with cellular your performance on traceroute will depend very highly on
    many thing associated with your celluar link and having little to do with
    Verizon's connection to the internet.

    * For example, your latency will tend to increase as you start moving.
    So, for those sitting at their desks in home with their cell phones attached
    to their desktops, your may having some decent latency. But if you use your
    cellphone for data attached to your laptop travelling 125 mph down the
    eastern corridor, your latency is likely to reach into the 500 ms range
    (where not only will you have propogation and interference issues, but also
    issues with cell handoff).
    * Your latency will depend highly on the number of people using the same
    cell you're using (both voice and data). In QNC mode (yeah, I'm talking
    verizon here - but the prinicples apply across the board), data is usally a
    second class citizen....so if you have a lot of voice connections, you are
    probably going to experience high latency (if you can connect at all).
    * Your latency will depend on the environment - is it raining? Is the
    sun out? Are you on 800 Mhz or 1900 Mhz? Are there people walking around
    you? Do you live in a house made of metal? Are you inside or outside?
    These are just a few of the examples of things that can make your latency
    vary second to second (not to mention the whole host of technical issues
    like multipath, etc).

    I've experienced both your latency profiles on Verizon's network...in most
    cases, the latency is fine for things like web browsing (although sometimes
    it get's a little harrowing)....interactively, it takes a lot of patience.
    Course, I haven't tried the new broadband stuff yet - maybe that'll be
    better.

    So next time n9wos, please be sure you know what you're talking about - you
    throw about a lot of terms, but for the most time, they're meaningless....

    I don't know what the original question was - but latency on a cell network
    (esp if you're mobile) will vary widely...

    Good luck.


    "N9WOS" <n9wos@nobug.worldnet.att.net> wrote in message
    news:4AL4c.16103$H44.295870@bgtnsc04-news.ops.worldnet.att.net...
    > > > This one here.
    > > > http://www.pingplotter.com/downloads/pngplt_1.exe
    > > > is one of them.

    > >
    > > I hate to break it to you, but I get nearly identical results with that
    > > program:

    >
    > Hmmm.......interesting.
    > That tells me that the isp has probably fixed
    > the TOS(type of service) setting that your computer
    > attaches to packets on the internet.
    > Probably has something to do with the routing overhead
    > of the cable system.
    >
    > > Maybe you should sign up with Comcast if these results are such a
    > > surprise to you. ;)

    >
    > Not a chance in hell. :)
    > BTW, no cable company serves are area.
    > They think we are not worth their time.
    > And they can shove their cable up their A$$ for all I care. :-o
    >
    >
     
  6. N9WOS

    N9WOS Guest

    > Even if they
    > did, the fact that the internet has some level of "QoS" (quality of

    service)
    > implemented in any but the most specific of circumstances is pure folly.
    > There's definitely no cross provider QoS occuring as of yet.


    I will only say this.
    Things on the internet change fast.
    Back in 2000, they was talking about how to standardize
    QoS systems via the TOS byte in the header.

    A moderate amount of time ago, I upgraded to XP.
    XP, by default disables the setting/changing of the ToS field.
    Same as win 2000.
    Unlike win 95 and earlier that allow changing of that field.

    ATT would not allow me to log on with the old dial up software.
    I had to upload a new version.
    The new version has QoS scheduler enabled by default.
    And one of the default settings on the QoS scheduler is
    that it enables marking of the ToS field.
    If you turn that QoS scheduler off (by uninstalling it),
    ATT will NOT allow me to establish a connection.

    It will not allow you to change the QoS settings.
    If you change anything on the QoS scheduler by editing the registry,
    the connection will only last a few minutes, before it will kick you off.
    And when you look at the registry, you will find that the ATT program
    reset the QoS settings back to default right when it kicked you off.
    QoS over the internet is something ATT has instated by force.
     
  7. Quick

    Quick Guest

    In line comments...

    "N9WOS" <n9wos@nobug.worldnet.att.net> wrote
    > > Even if they
    > > did, the fact that the internet has some level of "QoS" (quality of

    > service)
    > > implemented in any but the most specific of circumstances is pure folly.
    > > There's definitely no cross provider QoS occuring as of yet.

    >
    > I will only say this.
    > Things on the internet change fast.
    > Back in 2000, they was talking about how to standardize
    > QoS systems via the TOS byte in the header.


    Yes... TOS has been replaced by diffserv in the standards
    (although there are some "soft" provisions for backward
    compatibility).
    Quality of service requires end-to-end implementation in
    each node along the way (and even nodes not along the
    way) and generally remains something that exists in private
    networks.

    > A moderate amount of time ago, I upgraded to XP.
    > XP, by default disables the setting/changing of the ToS field.
    > Same as win 2000.


    Many disable this service to recoup the (most likely) wasted
    resources allocated towards it.

    > Unlike win 95 and earlier that allow changing of that field.
    >
    > ATT would not allow me to log on with the old dial up software.


    Dial up? To a modem? If so you haven't entered the packet
    realm until the other side of the modem.

    > I had to upload a new version.
    > The new version has QoS scheduler enabled by default.
    > And one of the default settings on the QoS scheduler is
    > that it enables marking of the ToS field.
    > If you turn that QoS scheduler off (by uninstalling it),
    > ATT will NOT allow me to establish a connection.


    Yes, ATT may have implemented QoS on *their* network.
    It is unlikely that your ToS will have much effect once you
    exit their network onto the internet. Do you know what the
    TOS setting are?

    > It will not allow you to change the QoS settings.
    > If you change anything on the QoS scheduler by editing the registry,
    > the connection will only last a few minutes, before it will kick you off.
    > And when you look at the registry, you will find that the ATT program
    > reset the QoS settings back to default right when it kicked you off.


    Was the connection dropped locally or from the remote end?

    > QoS over the internet is something ATT has instated by force.


    Ahhhh, errrr, ummm, my guess is that: QoS over *ATT's* network
    is something ATT has instated by force.

    -Quick
     
  8. N9WOS

    N9WOS Guest


    > Quality of service requires end-to-end implementation in
    > each node along the way (and even nodes not along the
    > way) and generally remains something that exists in private
    > networks.


    As far as I know, QoS based on the Tos header is a point application.
    As long as the other routers that don't support QoS don't change the
    ToS byte, then the routers that support QoS will not be affected by them.
    Just the routers that don't support QoS will not benefit by it.

    > > ATT would not allow me to log on with the old dial up software.

    >
    > Dial up? To a modem? If so you haven't entered the packet
    > realm until the other side of the modem.


    Where did you hear that at?
    The data is already down to the packet stage on a modem.
    Point to point dial up, point to point cable, and point to point fiber
    IP connections use the same end to end packet transmission.

    If you are thinking about the transmission used on Ethernet and
    subscriber cable systems.
    That is a multiple access system of a whole other type.
    they depend on collision detection.

    > Yes, ATT may have implemented QoS on *their* network.
    > It is unlikely that your ToS will have much effect once you
    > exit their network onto the internet.


    If the other operators don't implement QoS systems, then you
    are 100% correct.

    >Do you know what the TOS setting are?


    How would I?
    The ToS setting on the packet flag will depend on what
    the program sets it at.
    Now if I used a packet snifter program to capture the packets
    when they leave for the internet, then I could tell what the program was
    sending them out as.

    >
    > > It will not allow you to change the QoS settings.
    > > If you change anything on the QoS scheduler by editing the registry,
    > > the connection will only last a few minutes, before it will kick you

    off.
    > > And when you look at the registry, you will find that the ATT program
    > > reset the QoS settings back to default right when it kicked you off.

    >
    > Was the connection dropped locally or from the remote end?


    Yes.
    The access node drops the connection like a rotten sack of potatoes.

    > > QoS over the internet is something ATT has instated by force.

    >
    > Ahhhh, errrr, ummm, my guess is that: QoS over *ATT's* network
    > is something ATT has instated by force.


    All right!!!!!!!!
    A bad choice of words.
     
  9. Quick

    Quick Guest

    "N9WOS" <n9wos@nobug.worldnet.att.net> wrote
    >
    > > Quality of service requires end-to-end implementation in
    > > each node along the way (and even nodes not along the
    > > way) and generally remains something that exists in private
    > > networks.

    >
    > As far as I know, QoS based on the Tos header is a point application.
    > As long as the other routers that don't support QoS don't change the
    > ToS byte, then the routers that support QoS will not be affected by them.
    > Just the routers that don't support QoS will not benefit by it.


    Choice of words again... Its not individual routers that benefit from
    TOS. It's the network as a whole (or hole as the case may be) and
    the apps running over it that benefit. For example you would probably
    want your VoIP packets to have precedence over your FTP packets.
    This works on a network engineered throughout, including the
    applications running over it. If they implemented this on the
    internet then everybody (and/or apps) would simply set their
    TOS to high priority and you have degenerated to having no
    TOS at all. So TOS doesn't really work unless you control
    everything and its implemented from end to end. The treatment
    is fixed at the routers and it is the applications that determine
    the precedence of their streams of packets. Without coordination
    among well behaved applications it doesn't work. Note that TOS
    doesn't really give you quality of service. For that you need
    RSVP (again, requires implementation from one end to the other).

    > >Do you know what the TOS setting are?

    >
    > How would I?
    > The ToS setting on the packet flag will depend on what
    > the program sets it at. Now if I used a packet snifter
    > program to capture the packets when they leave for the
    > internet, then I could tell what the program was sending
    > them out as.


    Which program? The dialer? or the particular application you are running
    over your connection (browser, telnet, ftp, etc.)? My guess is that *all*
    your traffic (and probably all ATT users with similar subscriptions) goes
    at the same TOS regardless if the particular application you are running.

    -Quick
     
  10. Did you actually ask about ISDN? It does not have the distance limits
    of DSL. Some circuits are many miles!
    Tim
    On Fri, 12 Mar 2004 18:23:56 GMT, "N9WOS"
    <n9wos@nobug.worldnet.att.net> wrote:

    > > > They are slow, in comparison to what they charge per month.
    >> >
    >> > I expect fiber to the home!
    >> > 10Mbps, to 100Mbps.
    >> > And at prices less than $100/month

    >>
    >> Yeah, and I want a DS1 to my house for $200 per month. And I'm five miles
    >> from my CO. (Seriously. I live in a rural area.)

    >
    >I wish I could get any broad band to my house for any price.
    >Satellite is not for me.
    >The lag time on the signal just throws everything to heck.
    >Try to play mech warier 3 on a sat connection and you will be scrap in no
    >time.
    >A dial up user eat you for supper.
    >
    >20 miles is how far I am from the nearest switching office.
    >I could never get ISDN or DSL.
    >Cable system stop about five miles from this place.
    >The people in the valley have been crying for years, but that doesn't help.
    >FTTH opens up the possibility.
    >Current systems will support up to a 50 mile run at full speed.
    >It will be the first broad band technology that can
    >make it to this area, and free us from the 28K(on a good day) speeds.
    >
    >> FTTH is coming, but you're not getting 100Mbps for $100/month. I'll bet

    >the
    >> house I don't own that it won't be that cheap. :) (My landlord might have
    >> a problem with me betting her house, but she'll get over it :)

    >
    >For $70 a month for a 10Mpbs line,
    >you get local and long distance telephone service,
    >a basic package of cable TV channels.
    >a 2Mbps synchronous data stream for your IP usage.
    >
    >I think the $150 a month 100Mpbs package will give you
    >a couple phone numbers instead of one,
    >A premium channel package,
    >and a higher allotment of bandwidth for your IP usage.
    >
    >> > Some trials have set the 10Mbps subscription to $70 a month.
    >> > And I suspect that the price will go down when it becomes
    >> > widely available.

    >>
    >> I agree with that. In addition, some municipalities are starting to run

    >fiber,
    >> and they may be offering it to residents at an inexpensive price.
    >>
    >> > I think that the 100Mbps subscription will fall below $100 in time.

    >>
    >> Well, yes, *in time* is the key phrase here.
    >>
    >> But in the current environment, I think your statement is incorrect.

    >
    >The current environment is always changing.
    >For FTTH and any other broad band systems to thrive,
    >the data/bandwidth prices will have to come down.
    >The more fiber that is installed, and the better management of
    >the loading of the fiber, and the cheaper the prices will be.
    >prices will have to drop way below the multi thousand price tag
    >for true bandwidths over 1Mpbs.
    >
    >The thing that FTTH home will allow is high speed local access.
    >(ie) To other people on the same local fiber network.
    >Local shops running web pages for local content.
    >Local media providers streaming IP for audio and video
    >They will be able to interact at full 2Mbps+ speeds.
    >The line to the outside world will be restricted to
    >sub T1 speeds, and additional loading problems.
    >But as the networks grow, the people within the
    >local fiber blanket will grow.
    >
    >Long distance global IP will always be expensive, and slow,
    >but local high speed capability should be relatively cheep, and fast.
    >
     
  11. N9WOS

    N9WOS Guest

    > Did you actually ask about ISDN? It does not have the distance limits
    > of DSL. Some circuits are many miles!


    Yes, you can get an ISDN line out here.
    But speed is inversely proportional to distance.
    They calculated it up, and I would only get about
    40Kbps useable speed at my location.

    There has been one business out here that has
    been thinking about subscribing to a microwave link
    for data service.
    On the tallest hill around here, you can see the blinking lights
    from a microwave tower about 5 miles away,
    on one of the hill tops near the state park.
    If it wasn't so g*^*^*(^(&)(* expensive, I would think about it too.
     
  12. N9WOS

    N9WOS Guest

    > Which program? The dialer? or the particular application you are
    running
    > over your connection (browser, telnet, ftp, etc.)? My guess is that *all*
    > your traffic (and probably all ATT users with similar subscriptions) goes
    > at the same TOS regardless if the particular application you are running.
    >


    I cranked up the packet sniffer program.
    It can check what is coming in, but not going out.
    At least I will be able to check what is coming in.

    When I ping a router with a set ToS, I seem to be getting
    the same Tos back.
    There is a few ATT routers (that I have seen so far)
    that return a 192 in the ToS field irrelevant of the original field.
    (ie) System management packet with a priority at level 7 out of 8 levels,
    with no other priorities set.

    I also detected a ping from another att user,
    or the ATT security system is just doing it's normal rounds
    and rattling doors again.

    I have got to admit, there doesn't seem to be anyone really implementing
    QoS,
    On all the shout cast servers, and real audio servers I can get to(below
    28K)
    I haven't seen one of them send out a packet labeled as time sensitive.

    As for as general pinging to see if any other internet companies have
    there system set up to support QoS
    (ie) they return with the ToS field set to a system management level.

    I found one by a few quick tracerts is.
    212.179.191.101 Bezeqint.net

    All thier routers i seen, are sending back level 7 packets.

    My head hurts right now, from looking at all those raw packets,
    so I think I will avoid this subject for a bit, until my headache goes away.
    :-(
     
  13. Quick

    Quick Guest

    "N9WOS" <n9wos@nobug.worldnet.att.net> wrote
    > > Which program? The dialer? or the particular application you are

    > running
    > > over your connection (browser, telnet, ftp, etc.)? My guess is that

    *all*
    > > your traffic (and probably all ATT users with similar subscriptions)

    goes
    > > at the same TOS regardless if the particular application you are

    running.
    > >

    >
    > I cranked up the packet sniffer program.
    > It can check what is coming in, but not going out.
    > At least I will be able to check what is coming in.
    >
    > When I ping a router with a set ToS, I seem to be getting
    > the same Tos back.
    > There is a few ATT routers (that I have seen so far)
    > that return a 192 in the ToS field irrelevant of the original field.
    > (ie) System management packet with a priority at level 7 out of 8 levels,
    > with no other priorities set.
    >
    > I also detected a ping from another att user,
    > or the ATT security system is just doing it's normal rounds
    > and rattling doors again.
    >
    > I have got to admit, there doesn't seem to be anyone really implementing
    > QoS,
    > On all the shout cast servers, and real audio servers I can get to(below
    > 28K)
    > I haven't seen one of them send out a packet labeled as time sensitive.
    >
    > As for as general pinging to see if any other internet companies have
    > there system set up to support QoS
    > (ie) they return with the ToS field set to a system management level.
    >
    > I found one by a few quick tracerts is.
    > 212.179.191.101 Bezeqint.net
    >
    > All thier routers i seen, are sending back level 7 packets.
    >
    > My head hurts right now, from looking at all those raw packets,
    > so I think I will avoid this subject for a bit, until my headache goes

    away.
    > :-(


    Yup. Everybody will have some default value to originate packets with.
    Some may even have different values for different services. Note that
    Ping is usually implemented using ICMP which is a part of IP. On the
    remote end its not going up to an application but gets turned around/
    re-originated at the IP layer. Makes sense that the ToS field might be
    set to a system management level.

    Its one thing for applications to use the TOS field. You can look for
    applications that might use something other than the system default
    TOS. The default will be system specific and could be different from
    one to another of the same system if its configurable. For example
    I would expect all systems running a particular windows OS to have
    the same default which applications could override with some other
    value. I think the only way to tell would be to find packets from two
    applications on the same host with different TOS values.

    Now you need the network part. It may be impossible from your
    end to determine if the network implements TOS. If implemented
    routers should look at the TOS field of packets they are store-and-
    forwarding and treat them differently under certain conditions.
    They might do something like use different priority queues for
    different TOS values or allocate more bandwidth to some TOS
    values and less to others. Treatment of packets in different queues
    will only differ when there is contention for resources on the router.
    In a greatly simplified example a router could check the high priority
    queue twice a often as the low priority queue. If the router has plenty
    of resources and horsepower, packets going through either queue
    will not be delayed. In more intense cases the same router may delay
    packets in the low priority queue or even discard them to provide
    better service to the high priority queue. On the other hand if there
    is contention for resources among packets with the same TOS then
    TOS will have no effect (and not help).

    I expect that few applications bother with TOS. I expect that few
    internet routers bother with TOS (I don't know). Even if they did
    it ony works if everyone (applications) is coordinated and "good
    citizens".

    TOS and diffserv work on private networks. If you are inspired
    to investigate further I, for one, would be interested in what you
    find.

    sorry about your headache -:)
    -Quick
     
  14. N9WOS

    N9WOS Guest


    > I expect that few applications bother with TOS. I expect that few
    > internet routers bother with TOS (I don't know). Even if they did
    > it ony works if everyone (applications) is coordinated and "good
    > citizens".


    I have been thinking about it, and I can see why ATT, and some others
    would support QoS, while other's would not.

    ATT carries a lot of voice traffic on their network.
    They branded it as an intelligent network, and
    it was designed to handle both data and voice from the start.

    So they have QoS support on their routers.
    That way, during peak times, the system still can handle
    the primary voice traffic without delay, but the secondary
    WWW traffic would be secondary.

    So, having the ToS field set to 0 would not bother them a bit
    on WWW traffic.
    It is there for the programs to use, but if they don't, it is no
    skin of ATT's back.

    But some carriers only carry WWW traffic, and very little, or no VoIP.
    So they would have little use to implement QoS.

    And a side note.
    I have found out that I can tell if a server has the capability to
    set the ToS on outgoing packets.
    If you ping it with a 192 ToS packet and it will only return
    a packet with the default 0 ToS setting, then the owner
    has not removed the restriction on ToS setting in the software.
    (ie)It can not turn the ping back around with the ToS intact.
    All the internet routers I have pinged so far will return the
    same level packet or one that is set to 192.
     

Welcome to VerizonForums!

Unfortunately you can't reply until you log in or sign up.


Forgot your password?