MichaelTan
Arch-Supremacy Member
- Joined
- Oct 28, 2000
- Messages
- 13,137
- Reaction score
- 23
https://www.smallnetbuilder.com/wireless/wireless-reviews/33079-ignition-design-labs-portal-reviewed
Cons:
1. Wired router throughput may be iffy for gigabit-grade services
2. Storage sharing is very slow
3. Features still a work in progress
I'm not so concerned with 3 since i am sure more still will come out with firmware updates but can you do a test for 1 and 2? My main devices at home are still wired up so this is a concern.
I've worked on puzzling out the smallnetbuilder article with Portal engineers and discussions have led to the following points:
1) The 5G testing was done on a 6-7 year old qualcomm USB stick, one of the first available, which didn't support some of the latest strategies available on the latest chipsets. It might not reflect real world scenarios where your devices are actually less than 2 years old.
2) The Wi-Fi landscape has shifted significantly since SNB's Wi-Fi test process was last changed only a year ago. Portal is one of the parties who worked with SNB to change some testing procedures to reflect new trends which cutting edge router/ap like Portal Wifi have epitomized. Therefore, SNB will debut a new revision to their testbed and retest the Portal wifi router this month, and it's expected to improve in most areas.
https://www.smallnetbuilder.com/wir...mallnetbuilder-s-wireless-testbed-revision-10
3) Although SNB's testing has changed, some points still have to be noted - which will affect SNB's test results in the real world. Chamber testing is SNB's method, but real world testing is ... uhh... more reflective of real world usage.
First, SNB's test procedure allows for 40Mhz wide channels at 2.4GHz - this allows for great performance increase in a chamber, but in the real world this is unrealistic. It's difficult to find a use case where HT20 didn't supply enough bandwidth for the application to work flawlessly. Even with proper high quality video streaming uses HT20. But on the downside, HT40 2.4GHz will be highly affected by interference, including but not limited to 2.4GHz bluetooth interference which has been affecting many devices, which drop when bluetooth comes on. Other interference is from your neighbours' tons of wifi. Result is your HT40 WIFI will drop in the real world or not even carry data properly. To be optimal today, 2.4GHz wifi, if you do use it, should be HT20 and the testing should reflect as such.
Secondly, SNB still refuses (until now) to introduce interference testing. With wifi traffic signal generators easily available and controllable, interference testing should be now (ironically) uniformly applied to all tested products, to make it more in sync with the real world situation but until now they have resisted that kind of testing. More importantly, now that Mesh networks are becoming popular in homes, it becomes even more important to introduce interference testing since more mesh APs mean more potential for interference, and there is such thing as BAD mesh if your router/ap is not capable of avoiding interference from nearby mesh installations, or even from your own units. The Portal router's main trick is to give you the fastest WiFi possible using a variety of interference avoiding strategies, like using DFS, and also real time channel interference scanning with uninterrupted channel switching without a disconnect (which most channel switching competitors have to do).
Portal believes that testing to find the max performance possible from a router is valuable. However, it is more valuable to test how the router's performance degrades with loading, distance, and interference - which is `real world'.
Hence, until then, take SNB's testing procedures with a pinch of salt. What you should really consider, is how properly configured Portal routers perform in the real world, standalone or with Mesh.
What I personally look forward to in SNB's new testing procedures is a proper testing of mesh networks, and finally it will reveal the difference between Portal's mesh networks and its ability to sustain bandwidth over multiple hops, vs some of the other mesh solutions which degrade when more hops are added.
Last edited: