XPost: alt.comp.os.windows-11   
   From: dan@djph.net   
      
   On 2025-09-10, Daniel70 wrote:   
   > On 10/09/2025 2:30 am, Carlos E.R. wrote:   
   >> On 2025-09-09 16:04, Daniel70 wrote:   
   >>> On 9/09/2025 10:41 pm, Carlos E.R. wrote:   
   >>>> On 2025-09-09 14:34, Daniel70 wrote:   
   >>>>> On 9/09/2025 10:13 pm, Dan Purgert wrote:   
   >>>>>> On 2025-09-09, Daniel70 wrote:   
   >>>>   
   >>>>>>> YEAP!! Undergerstumble!! But did my data travel via Undersea   
   >>>>>>> Co-ax or via Low Earth orbit Satellite or via High Earth orbit   
   >>>>>>> Satellite ..... or via one of those reflector Panels   
   >>>>>>> (supposedly) left on The Moon by the Apollo Astronauts??   
   >>>>>>   
   >>>>>> It doesn't matter - throughput is not latency.   
   >>>>>   
   >>>>> But I'm not really interested in throughput OR latency but how long   
   >>>>> it takes to get from A (my place) to B (test site) .... and then   
   >>>>> how long it takes to get from B (test site) to A (my place).   
   >>>>   
   >>>> That's latency.   
   >>>>   
   >>> Ah!! O.K., my mistake.   
   >>>   
   >>> I thought 'Latency' was how long it took something to get up'n'going!!   
   >>> i.e. overcoming "inertia".   
   >>   
   >> I asked chatgpt to make sure. :-)   
   >>   
   >> Q: In the scenario of measuring internet connection speed, what would be   
   >> the latency?   
   >>   
   >> A: In the context of measuring **internet connection speed**,   
   >> **latency** refers to the **time it takes for a data packet to travel   
   >> from your device to a remote server and back**. It's usually measured in   
   >> **milliseconds (ms)** and is also known as **ping**.   
   >   
   > So a "one-way" journey.   
      
   No, "There and back again".   
      
   For example, a 22.1 millisecond ping to google.com means that   
      
     1. We send an ICMP Echo request to Google at time = 0   
     2. Google got it at time = x (we will assume x=11, knowing "22.1"   
         was the full RTT)   
     3. Google created an ICMP Reply at time = x+y (we will assume y=0.1,   
         and add that to "x=11" from step 2)   
     4. We get the Reply at 22.1 milliseconds   
      
   Or, if we're doing "human scale" time.   
      
   I ask Voyager 1 for some data at time = 0   
   The spacecraft receives the message some 20-24 hours later.   
   The spacecraft "instantly" responds   
   I receive the message 20-24 hours later.   
      
   Total latency (Round-Trip Time) of my request -- 40 to 48 hours.   
      
   >   
   >> ### Breakdown:   
   >>   
   >> * **Latency = Round-trip time (RTT)**   
   >   
   > But wait, "Latency" is now Return Trip Time!!   
      
   No, not "Return", "Round" Trip (aka "There and Back Again").   
      
   >>   
   >> * **100 Mbps** is how much data can be received per second.   
   >   
   > ... but one-way, only!!   
      
   Link data rate and RTT are not exactly "coupled" to one another.   
      
   You can have a very high-capacity (data / payload) link that has high   
   latency -- for example, a station wagon full of tapes hurtling down the   
   freeway.   
      
   You can have a very low-capacity (data / payload) link that has low   
   latency -- for example a 56k modem that is 22 milliseconds away from   
   google.   
      
      
   >   
   >> * **20 ms** latency means it takes 20 milliseconds to send a small   
   >> request to the server and get a reply.   
   >   
   > So is that "Round Trip" time??   
      
   yes.   
      
   --   
   |_|O|_|   
   |_|_|O| Github: https://github.com/dpurgert   
   |O|O|O| PGP: DDAB 23FB 19FA 7D85 1CC1  E067 6D65 70E5 4CE7 2860   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)