dd-wrt: More cpu-load than tofu11?

Discussion in 'DD-WRT Firmware' started by Top_Gun, Jan 8, 2006.

  1. Top_Gun

    Top_Gun Network Guru Member


    I have replaced my Netgear WGR-614v4 two days ago with the Linksys. There were no unsolvable issues with this unit, maybe I just got a little bored by the piece of hardware ;)

    So far, I have testet the latest original firmware, dd-wrt v23 and tofu11.

    I do like the pptp-server in dd-wrt v23, because it would allow me to set up a service-vpn to my father's box appx. 30km away.

    However, I found that the use of Emule and Azureus rendered the unit too unstable. I had to reset the box about once per hour, even with 4096 connections, TCP set down to 300s and UDP set to 45s.

    I tried the two little startupscripts I could find in this forum, nothing helped.

    Yesterday evening, I switched to tofu11 (stock, no modifications), and the router works without a problem. No pptp, but I can live with openvpn - though I prefer not to install too much software on my machines.

    What I discovered: The router's workload with dd-wrt is about a magnitude higher than with tofu11. While my box showed 0.3 to 0.9 as standard workload with dd-wrt, it is down to 0.0x with tofu11.

    Is the workload of 0.3-0.9 normal with dd-wrt, or did I configure something unusual?

    I used the following services with dd-wrt:

    static DHCP (why Linksys does not offer this standardfeature is beyond me)
    some port-forwarding (5x)
    uPNP: Switched it off as azureus claimed that there were known problems with the WRT54G (probably with the stock firmware)

    WLAN is configured to WPA2/AES+TKIP, however the filesharingbox is connected with copper.

    The router is identically configured with tofu11, except for the non-existant pptp-server.

    Top Gun
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice