Three questions:
Is there a way to _reduce_ the number of connections? I set it for "at least" 4, but most of the time it was carrying 8-10. Doesn't having more connections make the DL/UL times much slower? e.g. I'm on DSL, and when I was used Scour [RIP ] I could get files @ 14k/s and sometimes as fast as 50K/s, but with Gnutella they go at a maximum 3K/s, and most of the uploads from my computer averaged about 1 k/s.
I just left Gnutella on as a server for a few hours, and every now and then would get 7-8 upload requests from the "Host" for the same file, each seeming to start with a different amount of the file already downloaded. Is there some variant of Gn which splits the DL requests and then reassembles them? (a bit like a resume DL feature for files it hasn't started to DL!?) If so, where can I get it?
In the 3 hrs I ran it as a server there were 142 requests for uploads, and 128 of them ended in "error" messages. I admit I killed about 6, because they were going 0.2k/s, but that's still less than a 6% success rate. Is this a typical ratio of UL requests to successes for Gn?
Inkwire
Is there a way to _reduce_ the number of connections? I set it for "at least" 4, but most of the time it was carrying 8-10. Doesn't having more connections make the DL/UL times much slower? e.g. I'm on DSL, and when I was used Scour [RIP ] I could get files @ 14k/s and sometimes as fast as 50K/s, but with Gnutella they go at a maximum 3K/s, and most of the uploads from my computer averaged about 1 k/s.
I just left Gnutella on as a server for a few hours, and every now and then would get 7-8 upload requests from the "Host" for the same file, each seeming to start with a different amount of the file already downloaded. Is there some variant of Gn which splits the DL requests and then reassembles them? (a bit like a resume DL feature for files it hasn't started to DL!?) If so, where can I get it?
In the 3 hrs I ran it as a server there were 142 requests for uploads, and 128 of them ended in "error" messages. I admit I killed about 6, because they were going 0.2k/s, but that's still less than a 6% success rate. Is this a typical ratio of UL requests to successes for Gn?
Inkwire