What's new

What New Charts Features Would You Like?

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

thiggins

Mr. Easy
Staff member
What can we do to make our Charts more useful to you?
 
DD-WRT Supported, or others...

Would love to have an indicator of the status of support for the 3rd party firmwares. Also an indicator of the status of support (Infancy, Maturing, Stable, etc...) and how long it's been supported.

If the physical ports are gigabit and support Jumbo Frames (and what size).

If the unit supports simultaneous bands or not...
 
The diagram on the open air tests is vital to understand what a-f means, but often I can't remember where they are in relation to the test wifi router.

Could a coronal plane diagram of the home be drawn so we can more easily reference the location of the tests? Can it show relative thickness of the walls and floors? Can it be drawn to scale, since wifi reception is distance dependent? Sample enclosed.

The image can be a pop up near the charts, and anywhere that the testing procedure is done.

The next item is integrate the typical throughputs of different 11x flavors as best speeds from location a-f. So this becomes the 'benchmark' to beat.
 

Attachments

  • house-smalls.jpg
    house-smalls.jpg
    20.3 KB · Views: 612
Last edited:
I personally think it's funny that you broke apart wired and wireless routers into their own charts. A router is a router, and wireless is simply a feature or an option, much like you have checkboxes for VPN and QoS. Wireless should just be a checkbox.
 
I personally think it's funny that you broke apart wired and wireless routers into their own charts. A router is a router, and wireless is simply a feature or an option, much like you have checkboxes for VPN and QoS. Wireless should just be a checkbox.
Good point. Perhaps the charts are mis-named. They really show routing and wireless speed vs. range, which do require separate charts.
 
Oh, ok, your right. I thought that a router would either be on one chart or the other, but looking now, I see that the wireless routers are also listed in the WAN/LAN charts.

Along that train of thought, I guess that is more of a user issue than a technical issue.

I consider stability one of the most important aspects of a router. I'm sure we have all experienced the router that requires a daily or weekly power cycle in order to function correctly.

This may be a hard thing to find in a review unless the stability is so bad that it locks up during a test more than once, since I don't imagine reviewers take each router home and run them for a month or so. Maybe having a way for users to rate their experience, and rank them by average user perceived stability. Not sure if you already have some sort of user rating system in place, as I usually just come here for benchmarks; get what I came for and move on.

.... in fact, I see you have another thread for this sort of thing....
 
Last edited:
I consider stability one of the most important aspects of a router. I'm sure we have all experienced the router that requires a daily or weekly power cycle in order to function correctly.

This may be a hard thing to find in a review unless the stability is so bad that it locks up during a test more than once, since I don't imagine reviewers take each router home and run them for a month or so. Maybe having a way for users to rate their experience, and rank them by average user perceived stability. Not sure if you already have some sort of user rating system in place, as I usually just come here for benchmarks; get what I came for and move on.
You are correct. I only live with these products for a short time, so can't provide reliability testing.

That's why I've added the ability for readers to rate and comment on reviewed products. Just go to any product review.

I'll soon be adding charts or pages that show the top rated products
by users.
 
Idea to make the router chart more useful. Please review.

I recently had the opportunity to use the router chart, take your advice in how to buy a router 2014 and then go buy a new router. So far so good. Installed the shiny new router... fiddled with the features, impressed my friends and myself on the value I believe you helped achieve with the charts.

If you were wondering what I got.. Linksys ea2700 to add or replace a couple of well used WRT54G and GS devices running DD-WRT. Mostly because I thought it was time to have a Gigabit LAN to speed access to and from the NAS. Faster wi-fi was nice to have but not really a focus of the selection process.

Once I got over the fiddling it was time use the new LAN. It did not take long to experience something I did not expect. While on a skype video call, someone started up a PS3 MMPORG and the audio on the skype call started to sound rough now and then, also on occasion videos I was trying to watch might stutter.

We have a cable modem with 10Mbps/1.5 Mbps rock solid service. So I went hunting for potential causes and it did not take long to determine that one of the new ea2700s I was using as the gateway router to the internet was running between 15% and 25% packet discards and 10-13 ms range for jitter for audio/video when under heavy video load. This seemed unusually high but then again its not something I have been tracking for the last few years as the user experience has been complaint free with the WRT54's other than I wanted more LAN speed internally.

So I replaced the EA2700 with one of the old WRT54GS V7 running dd-wrt and ran the test again- it showed 1%-2.7% packet discards and 6-6.7 ms jitter. Then I dug up a WRT54G V5 with the OEM/Linksys firmware stuck it and reran the heavy video load test and got 4-6.5% packet discards with 6.4-6.9 ms jitter.

Both the new and the old products behave nicely under light load as when you are watching a 160x120 web show with much lower jitter readings and 0 or near zero discards.

Just to make sure I was not nuts I borrowed an Asus -rtn66u and reran the heavy video tests and also found higher jitter readings.

So that brings me to the point... Shouldn't the charts have some sort of line items that would classify potential performance with respect to jitter and packet discards etc. I was very surprised to learn that I now need to decide what is more important... Gigabit speed and faster wifi or ... lower jitter and lower packet discards.

Since this experience I have learned that my experience might be called bufferbloat based on what others on the web are writing. Do you agree?.. is there someway to rate/classify routers on their ability to handle both light traffic and heavy loads without making a dogs breakfast out of the data stream? Max bitrate/throughput of routers as in the charts does not seem to address the experience I had. Quality of throughput seems to be the issue.
 
Interesting idea. How are you measuring jitter and packet discards?
 
I wonder if you can quickly test jitter and packet discards by hammering the router with pings of very large sizes and very low wait times.
Ex; ping 192.168.1.1 -t -l 60000 -w 1 > pingtest.txt
At the same time have a significant but not unreasonable amount of data passing through the router. This ought to get the data you need.


What about VPN throughput?

FAT/NTFS speed from onboard USB as well?

A lot of this data is already collected from what I can see, and shouldn't add much to the testing process if anything.

edit:
apparently the ping test looks like it works, you can easily simulate load and record dropped packets/jitter quite easily.
Just pinging an IP beyond the WAN side of the firewall does the trick.
 
Last edited:
Add Lan to Lan speeds

Testing and reportin Lan port to Lan port speeds would be helpful in determining maximum internal network file transfer capability.

The site it a GREAT help in making a purchase selection and general info on networking. Good work!
 
LAN to LAN throughput is wire-speed, handled entirely by the switch chip. Routing code is not involved. There is no significant difference to be measured in today's switch chips.

Testing it would more reflect the capabilities of the computers doing the testing.
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top