# Speedtests with or without IPS **URL:** \ **Category:** Intrusion Prevention\ **Created:** [12 November 2025 17:19 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251 "2025-11-12T17:19:32Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![bbitsch](https://community.ipfire.org/user_avatar/community.ipfire.org/bbitsch/32/7420_2.png) [@bbitsch](https://community.ipfire.org/u/bbitsch)\ **Post date:** [12 November 2025 17:19 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/1 "2025-11-12T17:19:32Z")
I got a new router from my ISP to reach the new 1000/50 MBit/s speed. Switching to Bridge mode was a bit ‘diificult’. But this is another story. 😉 I checked the speed and got 100Mbit/s only. Some further investigation showed - speedtest-cli with IPS : 100MBit - speedtest-cli without IPS: 400MBit - iperf3 with IPS: 100MBit - iperf3 wthout IPS: 950 MBit All measurements are done from IPFire commandline. Conclusion: - speedtest-cli doesn’t mmeasure exactly, iperf3 is preferable - IPS throttles down to 100MBit. Why? ---
**Author:** ![bonnietwin](https://community.ipfire.org/user_avatar/community.ipfire.org/bonnietwin/32/12350_2.png) [@bonnietwin](https://community.ipfire.org/u/bonnietwin)\ **Post date:** [12 November 2025 17:27 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/2 "2025-11-12T17:27:33Z")
> [@bbitsch](#): > > IPS throttles down to 100MBit. Why? Am I correct in remembering that you have an APU system. If yes, then that system is not capable of running at the full bandwidth with IPS running. If I am remembering wrong, please provide info on which system you do have IPFire on. I used to have the APU4 mini system and that maxed out at around 200MBit with the IPS running while my line capability was 1GBit. I now have the new mini appliance and that runs at around 1GBit with the IPS enabled. This performance effect is shown in the benchmarking that was done back with CU158 & CU159 [https://www.ipfire.org/docs/devel/benchmarks#results](https://www.ipfire.org/docs/devel/benchmarks#results) ---
**Author:** ![bbitsch](https://community.ipfire.org/user_avatar/community.ipfire.org/bbitsch/32/7420_2.png) [@bbitsch](https://community.ipfire.org/u/bbitsch)\ **Post date:** [12 November 2025 17:37 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/3 "2025-11-12T17:37:40Z")
Yes, you are right. I use an APU system. Remains the factor 2 in speed measurement. ---
**Author:** ![bonnietwin](https://community.ipfire.org/user_avatar/community.ipfire.org/bonnietwin/32/12350_2.png) [@bonnietwin](https://community.ipfire.org/u/bonnietwin)\ **Post date:** [12 November 2025 17:44 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/4 "2025-11-12T17:44:44Z")
> [@bbitsch](#): > > Yes, you are right. I use an APU system. > Remains the factor 2 in speed measurement. My measurement was not done via iperf3 or via fireperf, it was done via speedtest so it might well be too optimistic. The fireperf based benchmark in that link gave `52/125 Mbit/s TX/RX` with CU158 and `170/140 Mbit/s TX/RX (+118/+15)` with CU159. I think that is in the same magnitude as you are showing. That benchmark is based on a 2019 mini appliance system ( APU4 I believe). Also did you do the test with complete minimum/default setup as described in that benchmark link. ---
**Author:** ![bbitsch](https://community.ipfire.org/user_avatar/community.ipfire.org/bbitsch/32/7420_2.png) [@bbitsch](https://community.ipfire.org/u/bbitsch)\ **Post date:** [12 November 2025 17:55 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/5 "2025-11-12T17:55:05Z")
Because I was interested in real time performance, I did the tests against a speedtest server in the internet. My results show commisioned speed with iperf3. This speed was measured by a Vodafone technician at my router (Fritz!Box), also. ---
**Author:** ![flinx](https://community.ipfire.org/user_avatar/community.ipfire.org/flinx/32/12572_2.png) [@flinx](https://community.ipfire.org/u/flinx)\ **Post date:** [12 November 2025 19:59 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/6 "2025-11-12T19:59:51Z")
Double-check that the green ECO mode 100 Mbit only LAN 4 Port on Fritz!box is not in use. ---
**Author:** ![bbitsch](https://community.ipfire.org/user_avatar/community.ipfire.org/bbitsch/32/7420_2.png) [@bbitsch](https://community.ipfire.org/u/bbitsch)\ **Post date:** [12 November 2025 22:37 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/7 "2025-11-12T22:37:23Z")
@flinx , welcome to the community. And thanks for your advice. But this wasn’t/isn’t the problem. Internet is bridged to LAN2 and this interface runs at 1000MBit. ---
**Author:** ![ms](https://community.ipfire.org/user_avatar/community.ipfire.org/ms/32/16615_2.png) [@ms](https://community.ipfire.org/u/ms)\ **Post date:** [13 November 2025 12:29 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/8 "2025-11-13T12:29:05Z")
With IPS the APU boards won’t go above 100 Mbit/s in a speedtest. Otherwise you should hit the 1000 Mbit/s. It depends what you want to test. In a real-world situation you don’t want to run IPFire without the IPS. ---
**Author:** ![bbitsch](https://community.ipfire.org/user_avatar/community.ipfire.org/bbitsch/32/7420_2.png) [@bbitsch](https://community.ipfire.org/u/bbitsch)\ **Post date:** [13 November 2025 12:44 UTC](https://community.ipfire.org/t/speedtests-with-or-without-ips/15251/9 "2025-11-13T12:44:13Z")
Means, with IPS I should look for a faster device. A path we know from windows. ☹