In reply to a post by netizen_uk:

We have installed Business VOOM-500 (plus a /28 subnet) since MAY.

As was said by another poster it's the /28. My business service has been in modem mode for nearly 2 years and is fine. The GRE tunnel used to supply the /28 appears beyond the capabilities of either the modem, the kit at the other end or both.

Hi,

I've had Voom 500 for about two weeks now and i'm not sure if i have a speed issue.

My daily averages using speedtest.net to virgin servers was about 200Mbps down. Upload was always around 35Mbps. If i did speedtests to no virgin servers i would average around 330Mbps. Test were made up of 16 tests throughout the day.

I also tested using fast.com which they recommended and i got about 30Mbps more than what the Virgin servers reported on Speedtest.net.

However when downloading on Steam i can hit 64.7MB/s, when downloading from torrent sites (Legal downloads) i can get 40MB/s to 50MB/s which would mean i am getting better than the speedtests.

I also have five Static IPs. My setup is the Hiltron in Modem mode to a Draytek 2960 and then into my switch which then gets distributed around the house.

I'll start doing some testing using nperf.com to see what averages i get.

Thanks

Ross

But when all said and done the speeds are usually from X to Y depending upon numbers of users at any one time so will be 'up to' whether we like it or not.

In reply to a post by Rossmartin79:

Hi,

I've had Voom 500 for about two weeks now and i'm not sure if i have a speed issue.

My daily averages using speedtest.net to virgin servers was about 200Mbps down. Upload was always around 35Mbps. If i did speedtests to no virgin servers i would average around 330Mbps. Test were made up of 16 tests throughout the day.

I also tested using fast.com which they recommended and i got about 30Mbps more than what the Virgin servers reported on Speedtest.net.

However when downloading on Steam i can hit 64.7MB/s, when downloading from torrent sites (Legal downloads) i can get 40MB/s to 50MB/s which would mean i am getting better than the speedtests.

I also have five Static IPs. My setup is the Hiltron in Modem mode to a Draytek 2960 and then into my switch which then gets distributed around the house.

I'll start doing some testing using nperf.com to see what averages i get.

Thanks

Ross

From Speedtest.net, what tests did you run? Single-Thread Mode or Multi? It is mainly the Single that has speed problems.
I agree with type of downloads like torrents however this represents MULTI, and not Single.
The speed we should be focusing (when a subnet is present) is the single-thread one.

Also in Speedtest.net I have found that the "best" speeds only come from "31173 Services AB" test server.
Please test and report back if possible.

Thank you.

Post deleted by Rossmartin79

Hi,

So i was running multi thread as that was the default however i have run some more tests.

Speedtest.net - Four tests to Virgin servers in Bromley, London, Luton and Brentford in single thread mode
Average Ping: 18ms
Average Download: 40.34Mbps
Average Upload: 5.66Mbps

Speedtest.net - Four tests to Virgin servers in Bromley, London, Luton and Brentford in multi thread mode
Average Ping: 19ms
Average Download: 153.65Mbps
Average Upload: 11.71Mbps

nperf.com- Four tests to Erith 10Gb server
Average Ping: 22ms
Average Download: 405.45Mbps
Average Upload: 24.16Mbps

Speedtest.net- Four tests to 31173 Services AB in single thread mode
Average Ping: 23ms
Average Download: 479.39Mbps
Average Upload: 1.67Mbps

Speedtest.net- Four tests to 31173 Services AB in multi thread mode
Average Ping: 36ms
Average Download: 459.02Mbps
Average Upload: 7.37Mbps

It does look like the speed tests in Single Thread mode over Virgin server is particularly poor!

Thanks

Ross

I'd guess it has something to do with this too.

My standard 350mb package gets ~385mb multi and just now 370+ single threaded.

https://www.speedtest.net/result/8411257309.png

Edited by chriscdotcodotuk (Fri 12-Jul-19 13:14:01)

Just an update with another four tests done so i have averaged the speeds:

Speedtest.net - Eight tests to Virgin servers in Bromley, London, Luton and Brentford in single thread mode
Average Ping: 20ms
Average Download: 51.81Mbps
Average Upload: 19.52Mbps

Speedtest.net - Eight tests to Virgin servers in Bromley, London, Luton and Brentford in multi thread mode
Average Ping: 19ms
Average Download: 166.585Mbps
Average Upload: 23.79Mbps

nperf.com- Eight tests to Erith 10Gb server
Average Ping: 22ms
Average Download: 436.63Mbps
Average Upload: 28.78Mbps

I'm going to escalate to Virgin to see what they say!

In reply to a post by Rossmartin79:

Just an update with another four tests done so i have averaged the speeds:

Speedtest.net - Eight tests to Virgin servers in Bromley, London, Luton and Brentford in single thread mode
Average Ping: 20ms
Average Download: 51.81Mbps
Average Upload: 19.52Mbps

Speedtest.net - Eight tests to Virgin servers in Bromley, London, Luton and Brentford in multi thread mode
Average Ping: 19ms
Average Download: 166.585Mbps
Average Upload: 23.79Mbps

nperf.com- Eight tests to Erith 10Gb server
Average Ping: 22ms
Average Download: 436.63Mbps
Average Upload: 28.78Mbps

I'm going to escalate to Virgin to see what they say!

Thank you very much for your detailed tests.
What matters (and I emphasise this) is the Single-Thread performance (when a subnet / GRE tunnel exists).
As i said before, if you want to transfer a single file (say via FTP or SCP) using a single thread, you will utilise single-thread speeds.
In my case (new line) I wanted to transfer large Virtual Machines (many Gigabytes) from a DataCenter server that has Gigabit Internet. As I said to the technical support it will be MUCH quicker if I go down to Central London, put the VMs on disks and bring them to the office!

PS. It is really pointless to test with VM servers. Even their technical support they don't recommend them (because the servers are busy).

It appears that they created a product (service) that cannot operate as thought initially and advertised afterwards. That's FINE. People do mistakes.

The big question is WHAT will be the solution?
If the amount of users affected is big then they will either:
1) Have to notify them that they will cease the service (with compensation) or
2) Create a NEW PRODUCT (service) which will replace the VoomXYZ with a DIA service (dedicated line) that will be capped to the original limits the customer had with the Voom Service. Note that DIA lines have no issues with subnets.

it goes without saying that the price should stay the same.

Shall we gather signatures and send them to Richard?
(I have put the smiley face but I am in fact serious about it!!!)
Line installed in May19, the line has been classified as "non performing", the installation itself as a "failed one" - despite two visits- however the billing/charging part works flawlessly!
And NO, they will not discuss about compensation for as long as the technical issue is open.
This is NOT Business customer service and I'm not being funny about it!

Thoughts?

Edited by deleted (Sat 13-Jul-19 21:37:40)

is this based on multi or single speed tests the reason I as this is, I've been on VOOM 500 since early March when it was first released, it was a rocky start when I upgraded as the firmware had issues that they rolled out, having taken 3 weeks to get this resolved, here we are 3 and a half months later and speeds have more or less halved, I average around 280mbps to 310mbps and this is done on fast.com, if I do the same tests on speedtest.com on multi connections I see the same results however doing it on a single connection I see speeds well below there minimum speed of 250mbps on VOOM 500.