Quantcast

Download speed issues

classic Classic list List threaded Threaded
38 messages Options
12
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Download speed issues

Mikhail Ramendik
Hello,

I have a Leap 42.2 desktop with a Ralink RT2800 PCI 802.11n Wi-FI
card. I have recently upgraded my broadband download speed to 40 Mbps
and was testing speeds. Unfortunately, I do not have a patch cord that
would reach from the desktop to the modem/router. I use speedtest.net.
Only 2.4 Ghz Wi-Fi is now enabled on the modem (but I can enable 5
GHz).

What I get, in the same geographical position across one internal wall
from the modem/router, is:

- On an Android phone and on a Windows 10 Toshiba laptop, about 35
MBps, which sounds about right for this Internet connection.

- On a Windows 7 Lenovo laptop, initially 20-22 MBps. After updating
the Wi-Fi driver, 24 MBps; disabling MIMO power save, 25 MBps. This
increases to about 35MBps when I connect it to the modem with a patch
cord.

- On the Leap desktop using speedtest-cli, 25-27 Mbit/s.

- On the Leap desktop using Chrome or Firefox, about 12 Mbps!

Upload speeds are pretty much the same everywhere at just under 10
Mbps, which is correct for this Internet connection.

So my questions are:

- Why could browser-based speed be so much lower? Might downloads
using these browsers and other programs be affected? Is there any way
I can fix that?

- Is there any way to improve the speedtest-cli speed, presumably
limited by wi-fi, other than getting a new wi-fi adapter? Might there
be driver tweaks?

- If I do need to get a new adapter, I guess a 5 GHz one would be sure
to get the speed up, as the modem can do 5 GHz, and there are a lot of
cheap RTL8811AU USB dongles, but I wonder if the drivers will be a
hassle?

/sbin/iwconfig output:

lo        no wireless extensions.

wlan0     IEEE 802.11bgn  ESSID:"Ramendik_44"
         Mode:Managed  Frequency:2.422 GHz  Access Point: DC:EE:06:B1:81:F0
         Bit Rate=144.4 Mb/s   Tx-Power=20 dBm
         Retry short limit:7   RTS thr:off   Fragment thr:off
         Power Management:off
         Link Quality=59/70  Signal level=-51 dBm
         Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
         Tx excessive retries:684  Invalid misc:2129   Missed beacon:0

eth0      no wireless extensio


--
Yours, Mikhail Ramendik

Unless explicitly stated, all opinions in my mail are my own and do
not reflect the views of any organization

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Carlos E. R.-2
On 2017-05-18 04:01, Mikhail Ramendik wrote:

> Hello,
>
> I have a Leap 42.2 desktop with a Ralink RT2800 PCI 802.11n Wi-FI
> card. I have recently upgraded my broadband download speed to 40 Mbps
> and was testing speeds. Unfortunately, I do not have a patch cord that
> would reach from the desktop to the modem/router. I use speedtest.net.
> Only 2.4 Ghz Wi-Fi is now enabled on the modem (but I can enable 5
> GHz).
>
> What I get, in the same geographical position across one internal wall
> from the modem/router, is:
>
> - On an Android phone and on a Windows 10 Toshiba laptop, about 35
> MBps, which sounds about right for this Internet connection.
>
> - On a Windows 7 Lenovo laptop, initially 20-22 MBps. After updating
> the Wi-Fi driver, 24 MBps; disabling MIMO power save, 25 MBps. This
> increases to about 35MBps when I connect it to the modem with a patch
> cord.
>
> - On the Leap desktop using speedtest-cli, 25-27 Mbit/s.
>
> - On the Leap desktop using Chrome or Firefox, about 12 Mbps!
Well, this can happen if the speedtest is local to the provider, while
the download is not. Try a download using wget, for instance. You should
also test file downloads on the other machines.

--
Cheers / Saludos,

                Carlos E. R.

  (from 42.2 x86_64 "Malachite" (Minas Tirith))


signature.asc (220 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by Mikhail Ramendik
Do not try speedtest through WiFi.  In addition to poorer bandwidth, the
results may be variable due to interference etc.  On 42.2 & Etherent, I
generally get mid 70s down and 11 up on a 60/10 Mb package.


On 05/17/2017 10:01 PM, Mikhail Ramendik wrote:

> Hello,
>
> I have a Leap 42.2 desktop with a Ralink RT2800 PCI 802.11n Wi-FI
> card. I have recently upgraded my broadband download speed to 40 Mbps
> and was testing speeds. Unfortunately, I do not have a patch cord that
> would reach from the desktop to the modem/router. I use speedtest.net.
> Only 2.4 Ghz Wi-Fi is now enabled on the modem (but I can enable 5
> GHz).
>
> What I get, in the same geographical position across one internal wall
> from the modem/router, is:
>
> - On an Android phone and on a Windows 10 Toshiba laptop, about 35
> MBps, which sounds about right for this Internet connection.
>
> - On a Windows 7 Lenovo laptop, initially 20-22 MBps. After updating
> the Wi-Fi driver, 24 MBps; disabling MIMO power save, 25 MBps. This
> increases to about 35MBps when I connect it to the modem with a patch
> cord.
>
> - On the Leap desktop using speedtest-cli, 25-27 Mbit/s.
>
> - On the Leap desktop using Chrome or Firefox, about 12 Mbps!
>
> Upload speeds are pretty much the same everywhere at just under 10
> Mbps, which is correct for this Internet connection.
>
> So my questions are:
>
> - Why could browser-based speed be so much lower? Might downloads
> using these browsers and other programs be affected? Is there any way
> I can fix that?
>
> - Is there any way to improve the speedtest-cli speed, presumably
> limited by wi-fi, other than getting a new wi-fi adapter? Might there
> be driver tweaks?
>
> - If I do need to get a new adapter, I guess a 5 GHz one would be sure
> to get the speed up, as the modem can do 5 GHz, and there are a lot of
> cheap RTL8811AU USB dongles, but I wonder if the drivers will be a
> hassle?
>
> /sbin/iwconfig output:
>
> lo        no wireless extensions.
>
> wlan0     IEEE 802.11bgn  ESSID:"Ramendik_44"
>          Mode:Managed  Frequency:2.422 GHz  Access Point: DC:EE:06:B1:81:F0
>          Bit Rate=144.4 Mb/s   Tx-Power=20 dBm
>          Retry short limit:7   RTS thr:off   Fragment thr:off
>          Power Management:off
>          Link Quality=59/70  Signal level=-51 dBm
>          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
>          Tx excessive retries:684  Invalid misc:2129   Missed beacon:0
>
> eth0      no wireless extensio
>
>


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Carlos E. R.-2
On 2017-05-18 12:07, James Knott wrote:
> Do not try speedtest through WiFi.  In addition to poorer bandwidth, the
> results may be variable due to interference etc.  On 42.2 & Etherent, I
> generally get mid 70s down and 11 up on a 60/10 Mb package.

At best, try both.

Testing with the cable you see what the external connection can give,
and then if with the WiFi you get less, you know that the culprit is the
WiFi.

--
Cheers / Saludos,

                Carlos E. R.

  (from 42.2 x86_64 "Malachite" (Minas Tirith))


signature.asc (220 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Mikhail Ramendik
I have already tried a fixed connection to a different computer, so it
does seem that Wi-Fi is the culprit for the connection part. I have
manipulated the antennas and am now getting 25 to 30+ (varies) speed
on CLI and 20+ on browsers.

However I can not bring a cable form the modem to this computer, and I
want to speed up the WiFi - what can I do about it? Also, I do have
suspicions about browsers just being slower, but this will need to
wait until I speed the WiFi itself up.

On 18 May 2017 at 11:24, Carlos E. R. <[hidden email]> wrote:

> On 2017-05-18 12:07, James Knott wrote:
>> Do not try speedtest through WiFi.  In addition to poorer bandwidth, the
>> results may be variable due to interference etc.  On 42.2 & Etherent, I
>> generally get mid 70s down and 11 up on a 60/10 Mb package.
>
> At best, try both.
>
> Testing with the cable you see what the external connection can give,
> and then if with the WiFi you get less, you know that the culprit is the
> WiFi.
>
> --
> Cheers / Saludos,
>
>                 Carlos E. R.
>
>   (from 42.2 x86_64 "Malachite" (Minas Tirith))
>



--
Yours, Mikhail Ramendik

Unless explicitly stated, all opinions in my mail are my own and do
not reflect the views of any organization

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

David C. Rankin
In reply to this post by Mikhail Ramendik
On 05/17/2017 09:01 PM, Mikhail Ramendik wrote:
> - On an Android phone and on a Windows 10 Toshiba laptop, about 35
> MBps, which sounds about right for this Internet connection.
>
> - On a Windows 7 Lenovo laptop, initially 20-22 MBps. After updating
> the Wi-Fi driver, 24 MBps; disabling MIMO power save, 25 MBps. This
> increases to about 35MBps when I connect it to the modem with a patch
> cord.

Unless you are actually timing the download (stopwatch and then using the
filesize to calculate the rate) the reported differences in rates are probably
more due to differences in what is reported by the OS than actual differences
in rates. 20-24 isn't bad and it close to what you would expect. I see 2.6 to
3 MB (which is 20.8 - 24 Mb) over a 2.4 GHz wireless (both with Linksys and
Trendnet) regardless whether it is Win10, Win7, SuSE or Arch (same laptop).
Intel Corporation Centrino Ultimate-N 6300 wifi.

--
David C. Rankin, J.D.,P.E.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

John Andersen-2
In reply to this post by Mikhail Ramendik
On 05/18/2017 04:21 AM, Mikhail Ramendik wrote:
> However I can not bring a cable form the modem to this computer, and I
> want to speed up the WiFi - what can I do about it? Also, I do have
> suspicions about browsers just being slower, but this will need to
> wait until I speed the WiFi itself up.

I'm sitting on a laptop that I connect to wifi, and the speed test
is less than half of my rated bandwidth.

If I pick up the little android tablet and do the same test it is
restricted ONLY by my purchased speed rating.

The laptop has 802.11G single band capabilities (2.4GHZ).
The tablet has Dual Band, B/G/N built in.

My AP has MIMO and steerable antennas and dual bands and all sorts
of magic I just barely understand.

It is still true that unless you have a very late model AP, If you have any
single 802.11G device on an 809.11N capable AP, don't
be surprised if many stations get dragged down to G speed or lower.

If you WIFI card can't do B/G/N, drop it and get one that can.
These things are plug replaceable and cheap these days.



--
After all is said and done, more is said than done.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by Mikhail Ramendik
On 05/18/2017 07:21 AM, Mikhail Ramendik wrote:
> However I can not bring a cable form the modem to this computer, and I
> want to speed up the WiFi - what can I do about it? Also, I do have
> suspicions about browsers just being slower, but this will need to
> wait until I speed the WiFi itself up.

It's unlikely you'll ever match the Ethernet speed over WiFi.  Modern
gear usually has 1 Gb Ethernet ports.  WiFi speed depends on the band
used, distance, interference and more.  Further, it's only half duplex.
It can't transmit & receive at the same time, the way Ethernet can.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by John Andersen-2
On 05/18/2017 08:33 PM, John Andersen wrote:
> It is still true that unless you have a very late model AP, If you have any
> single 802.11G device on an 809.11N capable AP, don't
> be surprised if many stations get dragged down to G speed or lower.
>

Not quite.  B causes a significant performance hit to G & N.  This was
because G and N use a different modulation type than B.  This meant that
when they detected a B signal, they'd have to drop into compatibility
mode, where they had to transmit in B a packet to reserve the channel
for the time it takes to send the G or N packet and then transmit the G
or N packet.  On the other hand, since G & N use the same type of
modulation, only the N header has to be transmitted at G spec and then
the rest of the packet can transmit at N speed.  Similar happens on 5
GHz, with A, N & AC.  They all use the same modulation method and just
have to slow down the header to accommodate the lower speed devices.
That said, there's no point in configuring the access point to allow any
slower than you need.  So, if you have no G devices, then don't
configure the access point to allow them.  Hopefully there's no B
devices left in your area.  Since N came out about 5 years ago, there's
not a lot of devices around that can't do it.



--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

John Andersen-2
In reply to this post by James Knott
On May 18, 2017 6:42:09 PM PDT, James Knott <[hidden email]> wrote:

>On 05/18/2017 07:21 AM, Mikhail Ramendik wrote:
>> However I can not bring a cable form the modem to this computer, and
>I
>> want to speed up the WiFi - what can I do about it? Also, I do have
>> suspicions about browsers just being slower, but this will need to
>> wait until I speed the WiFi itself up.
>
>It's unlikely you'll ever match the Ethernet speed over WiFi.  Modern
>gear usually has 1 Gb Ethernet ports.  WiFi speed depends on the band
>used, distance, interference and more.  Further, it's only half duplex.
>
>It can't transmit & receive at the same time, the way Ethernet can.

802.11n uses multiple channels, and can easily achieve 450Mbps, and 802.11ac achieves three times that.

Also true full duplex has been demonstrated and is in the near future. Google it.


--
Sent from my Android phone with K-9 Mail. Please excuse my brevity.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

jdd@dodin.org
In reply to this post by David C. Rankin
Le 19/05/2017 à 02:14, David C. Rankin a écrit :

> Unless you are actually timing the download (stopwatch and then using the
> filesize to calculate the rate) the reported differences in rates are probably

and don't forget the systems are multitask, windows may very well be
downloading the next update on the mean time...

jdd

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by John Andersen-2
On 05/19/2017 12:31 AM, John Andersen wrote:
>> It can't transmit & receive at the same time, the way Ethernet can.
> 802.11n uses multiple channels, and can easily achieve 450Mbps, and 802.11ac achieves three times that.
>
> Also true full duplex has been demonstrated and is in the near future. Google it.

Can you point to any available equipment?  The only way you're likely to
see full duplex is with dual band equipment, with one band used for each
direction.  The issue when using the same band is keeping the
transmitter power  from overwhelming the receiver.  In other areas,
there are radio devices that can run full duplex, by using some method
to keep the transmit power out of the receiver.  A common method is to
use different frequencies, sufficiently far apart that filters can
provide sufficient isolation.  When I did that search, all that turned
up was references to lab experiments, which are quite a way from real world.

Bottom line, any WiFi gear that can be bought today is half duplex, so
part of the time is spent transmitting and part receiving.
Incidentally, with full duplex, both ends have to support it, so that
means both the access point and the computer or other device.

BTW, O'Reilly has some good books on WiFi that cover all the gory
details.  There's one that covers 802.11b & g, with supplements that
cover n and ac.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by John Andersen-2
On 05/19/2017 12:31 AM, John Andersen wrote:
> 802.11n uses multiple channels, and can easily achieve 450Mbps, and 802.11ac achieves three times that.

That is the raw link speed, not actual use, which includes the various
overheads, including being only half duplex.
Still, when testing Internet bandwidth, WiFi introduces variables that
affect throughput.  Ethernet doesn't have that.  If you have a Gb link,
that's what you get, with full duplex.  The only issue would be
overloading a switch, so that packets are dropped.

Now, I have a 60/10 connection, so I may be able to meet that with
WiFi.  But my ISP now offers up to 1 Gb via fibre, in some areas.  With
that, you may see the limitations of WiFi.

Another thing to consider with WiFi is your neighbours.  Unless you are
well away from others, you may have to share the channel with others.
In my home, I can see over a dozen nearby networks.  That means they may
reduce the bandwidth available to me.  There may also be other, non-WiFi
devices that can cause interference.



--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Carlos E. R.-2
On 2017-05-19 15:04, James Knott wrote:

> Now, I have a 60/10 connection, so I may be able to meet that with
> WiFi.  But my ISP now offers up to 1 Gb via fibre, in some areas.  With
> that, you may see the limitations of WiFi.
>
> Another thing to consider with WiFi is your neighbours.  Unless you are
> well away from others, you may have to share the channel with others.
> In my home, I can see over a dozen nearby networks.  That means they may
> reduce the bandwidth available to me.  There may also be other, non-WiFi
> devices that can cause interference.

At the place I'm today, I have 60/60 to internet, but the Wifi does 30
on good days, 10 on bad days, like today. There are 36 neighbours that I
can see. I mean, that the NM sees.

The ISP wanted to double the speed and I said "NO". They "but you will
navigate faster". Me: "not on WiFi". They "But... (whatever)". Me: "I'm
an engineer and I know what I'm talking about. No." So they shut up and
I got a rebate instead. :-)

--
Cheers / Saludos,

                Carlos E. R.

  (from 42.2 x86_64 "Malachite" (Minas Tirith))


signature.asc (220 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
On 05/19/2017 10:44 AM, Carlos E. R. wrote:

> On 2017-05-19 15:04, James Knott wrote:
>
>> Now, I have a 60/10 connection, so I may be able to meet that with
>> WiFi.  But my ISP now offers up to 1 Gb via fibre, in some areas.  With
>> that, you may see the limitations of WiFi.
>>
>> Another thing to consider with WiFi is your neighbours.  Unless you are
>> well away from others, you may have to share the channel with others.
>> In my home, I can see over a dozen nearby networks.  That means they may
>> reduce the bandwidth available to me.  There may also be other, non-WiFi
>> devices that can cause interference.
> At the place I'm today, I have 60/60 to internet, but the Wifi does 30
> on good days, 10 on bad days, like today. There are 36 neighbours that I
> can see. I mean, that the NM sees.
>
> The ISP wanted to double the speed and I said "NO". They "but you will
> navigate faster". Me: "not on WiFi". They "But... (whatever)". Me: "I'm
> an engineer and I know what I'm talking about. No." So they shut up and
> I got a rebate instead. :-)
>

While I usually use WiFi for my network computer, I connect via
Ethernet, if I'm going to do some significant downloads to it.

Incidentally, this shows a minor benefit of Linux over Windows.  When I
connect via Ethernet, I get a different IP address than I do via WiFi.
If I then want to connect to the computer, with SSH for example, with
Linux it doesn't matter which way I'm connected, as either address will
work.  But with Windows, if connected via Ethernet, the WiFi address is
not reachable.  This means, I have to use 2 different addresses/host
names when booted into Windows, compared to it doesn't make any
difference with Linux.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Carlos E. R.-2
On 2017-05-19 17:03, James Knott wrote:
> On 05/19/2017 10:44 AM, Carlos E. R. wrote:


>> The ISP wanted to double the speed and I said "NO". They "but you will
>> navigate faster". Me: "not on WiFi". They "But... (whatever)". Me: "I'm
>> an engineer and I know what I'm talking about. No." So they shut up and
>> I got a rebate instead. :-)
>>
>
> While I usually use WiFi for my network computer, I connect via
> Ethernet, if I'm going to do some significant downloads to it.

Me too, at home. Not here.
The flat has telephone lines on all the rooms, but not ethernet.
I could maybe work it out, but it doesn't entice me to do the job.

--
Cheers / Saludos,

                Carlos E. R.

  (from 42.2 x86_64 "Malachite" (Minas Tirith))


signature.asc (220 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by James Knott
On 05/19/2017 11:03 AM, James Knott wrote:
>> The ISP wanted to double the speed and I said "NO". They "but you will
>> > navigate faster". Me: "not on WiFi". They "But... (whatever)". Me: "I'm
>> > an engineer and I know what I'm talking about. No." So they shut up and
>> > I got a rebate instead. :-)
>> >
> While I usually use WiFi for my network computer, I connect via
> Ethernet, if I'm going to do some significant downloads to it.

That should be notebook computer, not network.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

Mikhail Ramendik
In reply to this post by David C. Rankin
On 19 May 2017 at 01:14, David C. Rankin <[hidden email]> wrote:

>> - On a Windows 7 Lenovo laptop, initially 20-22 MBps. After updating
>> the Wi-Fi driver, 24 MBps; disabling MIMO power save, 25 MBps. This
>> increases to about 35MBps when I connect it to the modem with a patch
>> cord.
>
> Unless you are actually timing the download (stopwatch and then using the
> filesize to calculate the rate) the reported differences in rates are probably
> more due to differences in what is reported by the OS than actual differences
> in rates.

I have enabled the 5 GHz network too and set it up on the Win7 laptop,
instantly bumping the reported speed on that machine to 35-36 MBps
(same as Ethernet). So this is not about OS reporting.

5 GHz has less range, but I would not be worried as both the Win7
laptop and the Linux desktop are not used far from the AP. So now I
just want to move the Linux desktop to 5 GHz, which requires
replacement of my card.

The current card is a 802.11N Ralink PCI device, vintage 2008. I don't
want to spend a lot of money to replace it, so I've sent a question
here about RTL8811AU devices. There are cheap USB dongles on ebay and
I can wait for the delivery from China, so I just need to know if they
will run.

--
Yours, Mikhail Ramendik

Unless explicitly stated, all opinions in my mail are my own and do
not reflect the views of any organization

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

James Knott
In reply to this post by Carlos E. R.-2
On 05/19/2017 11:12 AM, Carlos E. R. wrote:
> Me too, at home. Not here.
> The flat has telephone lines on all the rooms, but not ethernet.
> I could maybe work it out, but it doesn't entice me to do the job.
>

I live in a condo.  When I bought it, it was wired for telephone in 4
rooms, but cable TV only in the living room & master bedroom.  The cable
connection was brought up through the wall between the 2 rooms.  When I
got a cable modem, my ISP ran a coax cable the entire length of my
condo, from the living room to my "office".  At the same time, I had
them pull in a couple of runs of CAT5, so I now have Ethernet in my
living & bedrooms, as well as my office.  The phone cables are  3 pair
CAT3, so I supposed I could have run 10baseT Ethernet along them, but
never bothered doing that.  I now have WiFi that covers my unit, so I'm
not so worried about wired connections.

Incidentally, unlike another major Canadian ISP, mine also provides
IPv6, so I have a /56 prefix all to myself.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Download speed issues

John Andersen-2
In reply to this post by James Knott
On 05/19/2017 06:04 AM, James Knott wrote:

> On 05/19/2017 12:31 AM, John Andersen wrote:
>> 802.11n uses multiple channels, and can easily achieve 450Mbps, and 802.11ac achieves three times that.
>
> That is the raw link speed, not actual use, which includes the various
> overheads, including being only half duplex.
> Still, when testing Internet bandwidth, WiFi introduces variables that
> affect throughput.  Ethernet doesn't have that.  If you have a Gb link,
> that's what you get, with full duplex.  The only issue would be
> overloading a switch, so that packets are dropped.
>
> Now, I have a 60/10 connection, so I may be able to meet that with
> WiFi.  But my ISP now offers up to 1 Gb via fibre, in some areas.  With
> that, you may see the limitations of WiFi.
>
> Another thing to consider with WiFi is your neighbours.  Unless you are
> well away from others, you may have to share the channel with others.
> In my home, I can see over a dozen nearby networks.  That means they may
> reduce the bandwidth available to me.  There may also be other, non-WiFi
> devices that can cause interference.
>
>
>

Agreed on all counts.
My internet connection is pretty much the same bandwidth as yours, so I can't
test higher speeds, but I do get all of 60  down on my multi band devices,
even with background traffic going on.  I haven't run in-house speed tests
yet. I'll have to try that.

I've been over to my neighbors house with my tablet and I can get 120 down
using his wifi, he pays for lots of bandwidth.

I'm not bothered by my neighbors, but they are probably/maybe bothered by me.
My multi-band-multi-channel-multi-Access-point Wifi with MIMO and steerable
antennas and background mesh Ap-to-AP transfer uses a lot of different
channels all the time.  There isn't a dead spot in my house.



--
After all is said and done, more is said than done.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

12
Loading...