WiFi Speed Observation <RESOLVED>

[linxuser]

My workaround for this is going from my new laptop via wire to my old laptop that runs xp and connects via it's WiFi (shared).
It's 3 to 2 times faster. Varies 'course. I've observed this kind of result since 6.06.

I've tried 9.10, 10.04 and 10.10 and imo 8.04 is without many newer bugs. Also I'm sold on KDE3.5 . I just like it.
Currently running dual 8.04 kbunt/Win7 with other installs on other partitions. (9.10 ubunt-studio, 10.04 kbunt-trinity, and Mint 9) They all share the same /var/cache/apt/archives only separate ones for 8.04, 9.10, and 10.04. They're linked to a /home/.cached-packages/8.04 or ..packages/9.10 .../10.04 and without having to reformat my /home and all files saved in software sources I no longer need to wait while it downloads everything during reinstall.

After only 4 years I figure this out. I think of all the time I've wasted over the years. Course it's all been dsl or slower, so no hurt on the net throughput. But even...

[Chopstick]

Re: WiFi Speed issue

I'm not sure I understand what the problem is...

But I don't know much about 8.04 or KDE 3.5 anyway, but I think you should be more specific, if you want anyone to help you...

[askrieger]

Re: WiFi Speed issue

(0) Personally, I don't use wireless, so this may be nonsense, but:

(1) I have noticed that many users at this site get better results with "wicd" than with "knetworkmanager". Some get decent results with "gnome network manager".

(2) WPA is much more secure and more polished than WEP.

(3) I have cat 5 cable strung all over the basement ceiling emerging from holes drilled in the floor in discreet locations, so that desktops and even laptops can be run from (more or less) convenient locations. I don't advice this if you rent your domicile.

[Chopstick]

Re: WiFi Speed issue

WICD: I can second that, especially with earlier releases of kubuntu I experienced a lot of problems with KnetworkManager, but now (since 9.10 I'd say) it's actually quite OK. Prior to Karmic I used WICD and found it to be very reliable and easy to use.

WPA is better than WEP but there are often quite confusing options w.r.t. TKIP/PSK/WPA1/2 etc. in WiFi routers, and they do not necessarily match the options in WICD or any other network manager. Just try any see what works... it may take a while.

[linxuser]

Re: WiFi Speed Observation (not issue)

Sorry about wrong statement. I've uninstalled all network-managers and use cli manual method of editing /etc/resolv.conf and /etc/network/interfaces . The manager packages seem to not work the way I'm thinking they would.
I've just noticed with my wifi interface (wlan0) active I get about 2/3 to 1/2 of KBps throughput, compared to close to 300 KBps.
(I may be saying that wrong with the whole kBps vs. kbps thing) But through my computer's ethernet interface (eth0) connecting through my other computer's ethernet, and that other computer's wifi being the access to internet. And it being an XP machine, sharing it's internet. I get throughput of 300 as compared with 150-200 kBps or kbps.
I've posted this observation in the past when I only had one dual boot machine. And that's all I'm saying.

Does anyone else see this, when using kubuntu's wifi interface or am I alone on this. I've noticed it on 2 previous toshiba laptops and my current dell one. It seems it might be just how it is. Any comments?

askrieger, I noticed you run wire. Do you see the same kind of thing?

No slight against kubuntu or linux intended. The 150-200 speed is good enough. The whole system even being viable is an amazing effort on everyones part. I'm just curious is all, and inquiring via search engines doesn't cure this curiosity.

Salutations all:

[GreyGeek]

Re: WiFi Speed issue

...
I have cat 5 cable strung all over the basement ceiling emerging from holes drilled in the floor in discreet locations, so that desktops and even laptops can be run from (more or less) convenient locations. I don't advice this if you rent your domicile.

My computers were in my office and that was where I was spending most of my day, banging on the keyboard at various Linux sites. One day my wife said "I'd appreciate it if you would come in to the living room and do your computer stuff there so we could talk from time to time". You can't say no to a request like that. I got 100' of cat5, drilled a hole up near the top of the wall, in the corner, and ran the cat5 out into the hall. I kept it tucked into the corner of the ceiling and hallway with J nails. In the dining room I drilled another hole through the wall into the living room, and tacked the cable across the room, turned and went across the fireplace, and down the fireplace to my recliner, where I coiled about 15 feet of it.

Then, my wife wanted to turn her laptop on in the Living room but with the wireless router back in the office and too many rooms and walls (with foil lined insulation) between her signal was weak enough to make her bandwidth marginal. So, I re-routed the cable-tv RG59U around to the living room and up through the floor behind the HDTV cabinet. I put the wireless router and wireless print server on top of the HDTV cabinet and the printer sets on the ledge of the fireplace, at the end of a usb cable. Now, the cat5 cable takes the internet connect BACK to my office, but I never use it there. It's a lot more fun out here with my wife. We share stuff we see on the web, like this one:
Embrace Life:
http://www.huffingtonpost.com/2010/0..._n_471799.html

[Snowhog]

Re: WiFi Speed Observation (not issue)

That is a very well conceived and expressed video. I really enjoyed it. Smartly done.

[askrieger]

Re: WiFi Speed Observation (not issue)

@linxuser: I suspect that the differences that you are seeing are EXTERNAL to your home. I just ran 6 different internet bandwidth calculator websites. Google "internet speed calculator". The download calculations varied between 21 Megabits per second (note NO ABBREVIATIONS, no divide by 8 or 10 for bits to bytes) and 9.3 Megabits per second. Uploads can be as much as a factor of five or ten slower. The most important difference was the location of the test server. One site had four test servers located in different US timezones and allowed you to choose the nearest one. The results differed by about a factor of two.

My internet connection is "fiber to the pole", coax to the cable modem, cat 5 to the router/switch, and cat 5 from there to the computers. I'm getting cherry juice on the laptop (presently in the kitchen, but some times in the room the grandchildren use when they visit), as I type this. (talk about sticky keys).

@GG: I love that video.

[linxuser]

Re: WiFi Speed Observation (not issue)

Nix this!

<X>
Ok. I'm seeing the difference in throughput from same site(s) using apt-mirror and the same when using synaptic. My workaround going through the winDoze's share works. I don't know why. And enough of that. Thank's everyone, for you comments.
</X>

Today 2010-08-20
I'm seeing equal throughputs. So that means the variation is due to reception problems with wifi. My other laptop with XP is on the other side of the room. For some reason this laptop (I'm runnin my Kubunt 10.04) is getting the same throughput today.

@linxuser: I suspect that the differences that you are seeing are EXTERNAL to your home.