Does the receiver have to wait for the Window to fill before sending an ACK?

Regards,
Rygel

Wendy wrote:

No.

Regards,

Marco.

In article <d5db4614.0412010710.445c39a9@posting.google.com>, rygel_16
@hotmail.com says...
No. But typically, Delayed-ACK is in play so it will ack every two
packets (or when the delayed-ack timer goes off)


--

hsb

"Somehow I imagined this experience would be more rewarding" Calvin
*************** USE ROT13 TO SEE MY EMAIL ADDRESS ****************
************************************************** ******************
Due to the volume of email that I receive, I may not not be able to
reply to emails sent to my account. Please post a followup instead.
************************************************** ******************

Wendy wrote:
Most operating systems send an ACK for every 2nd segment they receive or
after 200ms if there are no more segments.

In article <41b33897$1_3@news.bluewin.ch>,
=?ISO-8859-1?Q?Ignaz_Kr=E4henmann?= <ikraehenmann@bluewin.ch> wrote:
:Wendy wrote:
:> Does the receiver have to wait for the Window to fill before sending an ACK?

:Most operating systems send an ACK for every 2nd segment they receive or
:after 200ms if there are no more segments.

Hansang mentioned 'Delayed ACK' in his posting, which is the same
feature as the 200ms timer that Ignaz mentions.

As a small clarification: 200 ms is a common maximum timeout value.
The actual timeout used is theoretically uniformly randomly distributed
between 1 ms and an implimentation-dependant maximum value. The largest
timeout allowed by the standards is 1/2 second (500 ms.)

In practice, the times are often not uniformly randomly distributed.
For example, recently on SGI's IRIX, I noted timeouts in the range of
176 ms to 214 ms, but 201 to 203 ms were noticably more common than
other values. I also noticed that on IRIX, all the connections with a
given remote machine ran off of the -same- timer, resulting in those
connections operating in lock-step, rather than being independant
for each connection.
--
WW{Backus,Church,Dijkstra,Knuth,Hollerith,Turing,v onNeumann}D ?