CA3LE

God damn!

Advertised 5000/768 -- little better than advertised. :::.. Download Stats ..::: Connection is:: 5446 Kbps about 5.4 Mbps (tested with 2992 KB) Download Speed is:: 665 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 97 times faster than 56K you can download 1MB in 1.54 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=UHPC2341S :::.. Upload Stats ..::: Connection is:: 739 Kbps about 0.7 Mbps (tested with 579 KB) Upload Speed is:: 90 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 13 times faster than 56K you can upload 1MB in 11.38 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=IRAS92F7T

What I read about Videotron (who you have) is that even if you are paying for a dynamic IP it usually remains untouched. Much like mine with Cox Communications here in Phoenix. My IP could change if they wanted to... but has remained the same since at least 2002. Good luck with the shoutcast... fun stuff. Even if you had an IP that changed a lot, you could always go through a static IP service like www.dynIP.com

Well there goes a hundred megs of bandwidth.. ~~~ haha, just kidding test as much as you want

very cool... anyone else?

by the way.. just doing echo 8388608 > /proc/sys/net/core/wmem_max echo 8388608 > /proc/sys/net/core/rmem_max echo "4096 87380 4194304" > /proc/sys/net/ipv4/tcp_rmem echo "4096 65536 4194304" > /proc/sys/net/ipv4/tcp_wmem in /etc/rc.local is good for Linux 2.2.xx but is not as good for 2.4.xx because 2.4.xx has an auto tune feature that resizes the RWIN for each stream depending upon curtain factors..

Van... this is what i just did Okay, this is much better for linux 2.4.XX # increase Linux TCP buffer limits net.core.rmem_max = 8388608 net.core.wmem_max = 8388608 net.core.rmem_default = 65536 net.core.wmem_default = 65536 # increase Linux autotuning TCP buffer limits net.ipv4.tcp_rmem = 4096 87380 8388608 net.ipv4.tcp_wmem = 4096 65536 8388608 net.ipv4.tcp_mem = 8388608 8388608 8388608 Added to /etc/sysctl.conf ifconfig eth0 txqueuelen 1000 Added also will increase TCP throughput by increaseing the size of the interface queue. I guess it's not a good idea unless your on a Gigabit Ethernet connection (my server does have a Gigabit ethernet adapter) but I guess I will have to try this setting out to see how it runs. Let me know if your speeds report better from now on.. even durring peak.

^^ I tried those already, I will look into the others.

that's awesome... what are you talking about.

lol, yea that was when I was erbuilding apache this morn... lol, OOPS!

Well first of all what processor do you have? You have to get the correct one for your processor.

Cobra... look at this man.. someone is stealing your sig business https://testmy.net/forum/index.php?topic=1889.0

OWN3D FireFox M2 - I don't have a 64bit processor so i can't use M3 :::.. Download Stats ..::: Connection is:: 5408 Kbps about 5.4 Mbps (tested with 2992 KB) Download Speed is:: 660 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 97 times faster than 56K you can download 1MB in 1.55 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=1ZQG6MT0A :::.. Upload Stats ..::: Connection is:: 731 Kbps about 0.7 Mbps (tested with 579 KB) Upload Speed is:: 89 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 13 times faster than 56K you can upload 1MB in 11.51 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=JQYKV7I1B

I thought the same at first too.. you will come to notice little things.

At this website... random uncompressable data is sent to you, nothing is done with it... my script times how long it takes to download it and calculates from there... but the data is nothing of importance.

Trust me man, there is a difference behind the facia. Mine runs much smoother that it used to. --- and WAY smoother than IE.

5 more reboots and I think I have the server set to where she is comfy Linux RedHat Enterprise 3 added to /etc/rc.local ifconfig eth0 mtu 1500 echo 256960 >> /proc/sys/net/core/wmem_default echo 256960 >> /proc/sys/net/core/wmem_max echo 256960 >> /proc/sys/net/core/rmem_default echo 256960 >> /proc/sys/net/core/rmem_max echo 1 >> /proc/sys/net/ipv4/tcp_sack echo 0 >> /proc/sys/net/ipv4/tcp_timestamps echo 1 >> /proc/sys/net/ipv4/tcp_window_scaling then reboot and running smooth...

test scores are even consistant from 3MB, 6MB and 12MB tests -- very little difference :::.. Download Stats ..::: Connection is:: 5372 Kbps about 5.4 Mbps (tested with 2992 KB) Download Speed is:: 656 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 96 times faster than 56K you can download 1MB in 1.56 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=EKBAQR8DW :::.. Download Stats ..::: Connection is:: 5475 Kbps about 5.5 Mbps (tested with 5983 KB) Download Speed is:: 668 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 98 times faster than 56K you can download 1MB in 1.53 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=5ENPW9HOY :::.. Download Stats ..::: Connection is:: 5299 Kbps about 5.3 Mbps (tested with 12160 KB) Download Speed is:: 647 KB/s Tested From:: http://www.testmy.net/ Bottom Line:: 95 times faster than 56K you can download 1MB in 1.58 second(s) Validation Link:: https://testmy.net/cgi-bin/get.cgi?Test_ID=MNHZI4WTF

I have made a pretty huge change to the server, I tweaked all the TCP settings on the server and am getting much better performance. It's 7 am and I have yet to fall asleep.. but I think I FINALLY got it tuned well. On my personal test (which were already running good) I am now seeing no little pauses in the progress bar and am getting a much more consistant score varied by very little from test to test. I hope this will resolve the speed degradation during peak hours. Please post in this thread to inform me if you see an increase in accuracy and reliability.

Van ROCKS Micro.. nice avatar -- funny

By the way, for anyone who needs FireFox m1, m2 or m3 they can be found at http://moox.ws/tech/mozilla/

true dat!

I don't mind listing that 386 KB * 0.4 MB

seem better lately?

DUR.. I was missing a line of code, happened with the perl to php conversion.. fixed.. thanks guys