• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

I'm a noob, what the heck is going on with my switch?

CPA

CPA

Elite Member
I have a Netgear FS105 switch (10/100) that I have my computer, my son's computer and sometimes my work computer hooked into. The switch is plugged into a wallplate that travels to our Dlink DIR-655 router, then to our DSL modem, of course.

Right now, only my computer is on and active. I ran a speed test at Speedtest.net, targeted server hosted by Comcast and got 81 ping, 8.06 mbps down (I know it's slow, but it's the best available) and .52 up. This is normal speeds for me.

So, I wanted to see if the switch was impacting my speed at all. Well, it was, BUT in the opposite direction than I expected. Hooking my main computer straight into the wall plate, I got slightly better ping (74, I'm confident that the switch/no switch didn't really have anything to do with this difference), HOWEVER my down speed plummeted to 5.42 Mbps down. I ran the test twice and got similar results.

Why am I getting better speed using the switch? All the switch is supposed to do is parse out bandwidth for multiple users, right? So, how am I getting better speed down than I would getting rid of the switch and going directly to the router?
 
That doesn't make any sense to me.

Try it during the same time of the day, preferably in the AM when most people are at work. During peak times your DL speeds can slow.
 
Danimal1209 said:
That doesn't make any sense to me.

Try it during the same time of the day, preferably in the AM when most people are at work. During peak times your DL speeds can slow.
Click to expand...

I did. The direct test was right after the switch test.

I didn't do it in the AM because my teenage son is on Summer Time right now, meaning he's up all night and sleep in the late morning to early evening. So when I tested it yesterday was the least amount of traffic in my household. Also, the test through the switch gave me results I was expecting. Outside the switch, well, that just doesn't make sense.
 
oh, you should not rely on internet speed tests only. use iperf to test your internal connections between machines
 
CPA said:
I did. The direct test was right after the switch test.

I didn't do it in the AM because my teenage son is on Summer Time right now, meaning he's up all night and sleep in the late morning to early evening. So when I tested it yesterday was the least amount of traffic in my household. Also, the test through the switch gave me results I was expecting. Outside the switch, well, that just doesn't make sense.
Click to expand...

The least amount of traffic in your household doesn't mean much. It's a cable connection, you share bandwidth with the whole local node. If your whole neighborhood is using the internet during prime time, your speeds are affected too.

Bandwidth is also "up to," cable is not a dedicated line and it can fluctuate considerably. I have comcast at home, and on a good day ill get 30/10. On a Saturday morning I get to watch it violently fluctuate between maybe 15 and less than 1.
 
Mushkins said:
The least amount of traffic in your household doesn't mean much. It's a cable connection, you share bandwidth with the whole local node. If your whole neighborhood is using the internet during prime time, your speeds are affected too.

Bandwidth is also "up to," cable is not a dedicated line and it can fluctuate considerably. I have comcast at home, and on a good day ill get 30/10. On a Saturday morning I get to watch it violently fluctuate between maybe 15 and less than 1.
Click to expand...

Understand, but these tests were run within minutes of each other. Basically it was:

Through Switch: 8.06
Through Switch: 8.10
Switch Removed: 5.42
Switch Removed: 5.30
Through Switch: 8.13

the tests take about 20 seconds each. they were run through the same server, one right after another. I can't imagine that it was a result of line congestion/coincidence. Something else has to be going on here.
 
ch33zw1z said:
oh, you should not rely on internet speed tests only. use iperf to test your internal connections between machines
Click to expand...

Will do, however, when I removed the switch only my machine was connected. the other two were offline.
 
CPA said:
Will do, however, when I removed the switch only my machine was connected. the other two were offline.
Click to expand...

you may have to put a machine on hte router's switch, that's where your wall plate runs to correct? Just try to isolate the issue, put a machine off the router and test with iperf.
 
ch33zw1z said:
you may have to put a machine on hte router's switch, that's where your wall plate runs to correct? Just try to isolate the issue, put a machine off the router and test with iperf.
Click to expand...

Correct, the switch connects to a wall plate. Basically, I removed that connection and connected my machine directly into the wallplate, which, of course, goes directly to the router.
 
CPA said:
Correct, the switch connects to a wall plate. Basically, I removed that connection and connected my machine directly into the wallplate, which, of course, goes directly to the router.
Click to expand...

I would setup a machine direct connected to the router (laptop is easiest), and run iperf between the two computers. That connection test will tell you if you're having internal network issues, or its an external issue.

That wall plate port on the router is running 10/100mbit to the router, but the router is gigabit ready.

What is your computer showing as connection speed when connected to the wall plate?
 
ch33zw1z said:
I would setup a machine direct connected to the router (laptop is easiest), and run iperf between the two computers. That connection test will tell you if you're having internal network issues, or its an external issue.

That wall plate port on the router is running 10/100mbit to the router, but the router is gigabit ready.

What is your computer showing as connection speed when connected to the wall plate?
Click to expand...

Just for clarification, the line from my wallplate in the room where the computers are at goes directly to a patch panel where the router is at. The lines going into the router are coming off the patch panel, not another wallplate.

I took a look a iperf and it looks way over my skill level. Ran the speedtests again and direct connection was just as good (actually a little better). So maybe it was an anomaly. I will try a few other times.

Also, can I switch out the wallplate jack to gigabyte or do they only come in 10/100? Sorry for the noobish question.
 
CPA said:
Just for clarification, the line from my wallplate in the room where the computers are at goes directly to a patch panel where the router is at. The lines going into the router are coming off the patch panel, not another wallplate.

I took a look a iperf and it looks way over my skill level. Ran the speedtests again and direct connection was just as good (actually a little better). So maybe it was an anomaly. I will try a few other times.

Also, can I switch out the wallplate jack to gigabyte or do they only come in 10/100? Sorry for the noobish question.
Click to expand...

It has more to do with the cabling than the jack.
 
CPA said:
Just for clarification, the line from my wallplate in the room where the computers are at goes directly to a patch panel where the router is at. The lines going into the router are coming off the patch panel, not another wallplate.

I took a look a iperf and it looks way over my skill level. Ran the speedtests again and direct connection was just as good (actually a little better). So maybe it was an anomaly. I will try a few other times.

Also, can I switch out the wallplate jack to gigabyte or do they only come in 10/100? Sorry for the noobish question.
Click to expand...

iperf is command line, but not too hard. Basic like this:

run iperf -s on one of the pc's

then, from another pc, run iperf -c ip address of the -s pc

this is the default test. you can change many of the options, but the default test is fine. after 10-15 seconds, it will give you a bandwidth result.

I tend to use a couple of old linux laptops I have, you don't need to cd to the directory where iperf is to run it like you do in windows.

Once you get it setup, it's fun to play with. Make a network change and run the test, I find cabling to be a big factor in the test results.
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top