Jump to content

rafi

Established Members
  • Posts

    8,960
  • Joined

  • Last visited

  • Days Won

    148

Everything posted by rafi

  1. are you sure you have you upload rate reasonably limited
  2. problem #1: download speed limit bug (already reported...) example: max DL speed - ~180K. UL limit: 12K (tested with OpenOffice) calc_overhead= true max peers 200/1000 per-torrent/global 1 torrent DL - 100(1600)/10(700) s/p - ~ 130-140KB (below but close to the maximum possible) when we just add one more torrents - it drops: #1 - ~90K 90(1600)/16(700) #2 - ~20K 80(1600)/20(600) stopping #2 - and #1 goes back to ~130-140K. The thing is - we should not see any effect of #2 on #1... Edit: problem #2: setup double-click to perform download command - select "all feeds" - doble-click//download some releases. Once in a while - the selection-highlight from all-feeds - vanishes and the view is being totally clears.
  3. Neronut: it marks releases published in the last 24 hours. This was available in the previous RSS style view. BTW, I found three issues on this preliminary test-build, that does seem to fix/improve the issue with obeying of the upload limit : * all download limits do not work properly * RSS view clears sometimes when selecting "all feeds" and double-clicking a release (to download). The focus on "all feed" - vanishes. * handling/downloading from more than 150-200 peers (2 torrents) - feels like degrading the speed performance.
  4. btw, alus, how should the new uTP effect the net.calc_overhead calculation ? UDP does not have embedded overhead in the protocol itself, BUT if you have none-dataload-related-complete-UDP-messages shouldn't they also be counted as overhead ?
  5. ...which is in contradiction to the purpose of this change...
  6. no, just luckier (or un-luckier ?) ... @alus: 1) anyways - did this help at all ? 2) As many others here noted, I also noticed (and posted) speed differences seen in external measurement utilities and uTorrent. Did your investigation reveal any explanation/fix to this ? 3) I'm not bugging you about the RSS much... .. but I'm just curious if enhancements are planned for this release?
  7. crash dump 1.8.1 + utp on: http://www.mediafire.com/?sharekey=bd914c1eeae18c0ad2db6fb9a8902bda
  8. Look at one of my previous posts. I think we can see that same thing at the netmeter screen-shot I posted. edit: http://forum.utorrent.com/viewtopic.php?pid=380524#p380524
  9. Seems that when uTP is selected, the UL is maxed to beyond my defined max, and kills my poor 20K UL connection in spite of the uTP theory... Keeping to the defined limits - IS important until this release is stable, otherwise it may become useless for low UL bandwidth connections. Also changing of the advanced setup seems to have no effect on the UL limit control until you restart uTorrent. Not sure if calc_overhead has any effect here... Downgrading to previous release for now...
  10. ok. See that you don't limit your upload too much. I think the overhead in this version is higher and it does not show always the correct speed (use external tool). I also have net.calc-overhead enabled.
  11. the reason might be that this torrent has few but fast seeds, and a good seeds/peers ratio. Not sure about your other torrents.
  12. I have UL limit set to 12, and see this behavior (with about 1000 seeds and 1000 peers, most with 80-200 msec latency): http://img300.imageshack.us/img300/6350/30234694tv7.png @Asagrim: try this: http://borft.student.utwente.nl/~adrian/torrentphp/torrent.php/OOo_3.0.0_Win32Intel_install_en-US.exe.torrent it used to utilize 100% bandwidth with previous versions.
  13. dump : http://www.mediafire.com/file/u3ljqykuyjl/13582-utorrent.8fd3.dmp just froze to death... XP SP2
  14. Thanks. Works great. It will be nice if you complement this with download completion messages as well. Any other changes planed in 1.9 to RSS from this list ? http://forum.utorrent.com/viewtopic.php?pid=352646#p352646 Since I have 2 ISP connections now, with one - capping me and one not... I think I'll try to test tonight if uTP effects the capping. Should it ? edit: I'm just curious - why is the uTP "flag" in the peers' list not a "real" flag in the flags' column ? if it should be analyzed, or counted, maybe a separate column will serve best (as I would like to have for the countrys'-flags...)
  15. does this control work with 1.8.1 ? what does "0" value do i?
  16. alus: the new min rate inactivity disconnect seems to not be in 1.9. is it intentional ?
  17. DHT "update" seems to never stop. sames goes for tracker updating
  18. How about a change-log with the rest of the changes (not related to the "main change" ) ? edit: found this about uTP : http://forum.bittorrent.org/viewtopic.php?pid=391#p391
  19. hey, any news/activity on 1.9 beta ?
  20. the question was why should it behave this why ? sort is sort (for each column per it's type), why should columns' types effect it's operation at all ?
  21. looks fine to me, and works here... so it picks it up and just don't start DLing it, ha ? I suggest you open you own thread in the troubleshooting section.
  22. @zorrax: put a screenshort of your favorite dialog (DON'T mark "don't start download automatically"...)
  23. I see two notations for V 1.8.1 in the peers' list: 1) "uTorrent/1.8.1.0" 2) "uTorrent 1.8.1" what does it mean ? btw, in 1 - most are with IE & I flags only.
×
×
  • Create New...