Vodacom HSUPA tested

[rpm]

Vodacom HSUPA tested

Vodacom recently launched HSUPA – this is what you can expect

 

[Tns]

i wonder how much they going to try and the user pay if you want speed you must pay

 

[flenters]

International latency was impressively low for a wireless service with an average ping time to Google of below 300ms and below 400ms to BBC. Locally ping times were around the 100ms mark.

And that is why you cant use UPA for gaming, as said in another thread increasing the uplink won’t improve latency. Don’t even get me started on packet loss on a radio interface...

 

[saffakanera]

I think when they talked about "Internet infrastructure" they werent talking about you playing games on a Friday night.

 

[AirWolf]

i wonder how much they going to try and the user pay if you want speed you must pay

Something wrong there somewhere

For the time being it is a paid for service, but with improvements in the future such as the landing of more cables then the next speed up the rung will be the paid for service and this will be the standard service

 

[Pitbull]

I would love to see a WoW players input on this latency promises.

Anyone of you guys play WoW, if so pls post the Latency for the EU servers

Much appreciated

 

[Rynomster]

Latency doesn't seem much better.. I'm getting pretty much the same latencies with HSDPA.. But what a rip off..

 

[Pitbull]

Latency doesn't seem much better.. I'm getting pretty much the same latencies with HSDPA.. But what a rip off..

I can see the speed helping for things like E-mails and so forth. But to try and say speed = Latency got me a little confused

But thanx for the response

 

[Basilrsa]

Come on Vodacom

All this hype about 3.6 HSPDA and increased upload speeds.
I ran the trial period on the increased speed offering.
Guess what, I am back to 1.8
I felt the gain was not all that great, so maybe 7.2(when they can get backhaul from Telkom) wil be worth waiting for.
Vodacom should offer this service to all their clients on packages of 1gb per month or greater at no extra cost.
That would make them a leader instead of a follower.
They always seem to be trailing MTN.

 

[vodacom3g]

I can see the speed helping for things like E-mails and so forth. But to try and say speed = Latency got me a little confused

But thanx for the response

Latency IS related to speed.

If you do 100Km/h or 200Km/h on your Honda over the same stretch of road do you do it in the same time?

But how much time will you save if you travel from JHB to CPT and back but you could only go faster on that one section of the road?

Geddit?

In the case of UPA, there is a substantial increase in speed (about a factor 3) in the RADIO UPLINK. So it must stand to reason that latency on that link will improve. But it's only one link in a long list of links that give you your gaming response.

 

[Pitbull]

Latency IS related to speed.

If you do 100Km/h or 200Km/h on your Honda over the same stretch of road do you do it in the same time?

But how much time will you save if you travel from JHB to CPT and back but you could only go faster on that one section of the road?

Geddit?

In the case of UPA, there is a substantial increase in speed (about a factor 3) in the RADIO UPLINK. So it must stand to reason that latency on that link will improve. But it's only one link in a long list of links that give you your gaming response.

Hi V3G

I hear what you are saying. But still speed does not = Latency.

With your example as stated above that is one Honda not 2 different ones and it's the same road.

Now if you have to test HSUPA latency against the like of iBurst as an example. That is a completely scenario aint it ?

Thus was the reason i needed someone to post a latency for WoW on an EU server. Thus I can personaly see if there is any latency benefit with regards to the players in the field.

I can go and say yes, I had a latency yesterday of 400ms and today i have one of 350ms. This is purely because of weather and so forth. Saying that latency improved because of me having a quicker speed has nothing to do with it. Same scenario and my current Upload speed has increased. Thus still doesn't mean my latency would improved. There is to many variables.

I would like to see the latency people get on HSUPA vs another provider without the technology. Be it Telkom, Iburst, MTN which ever.

I'm not saying you wouldn't see a latency increase if you are using the same network, I am saying someone without HSUPA on a different network can get the same if not lower pings.

Hope it makes sense

 

[flenters]

"Latency IS related to speed."
but i get better latency on 384kbps dsl, not speed but the physical layer

 

[Pitbull]

"Latency IS related to speed."
but i get better latency on 384kbps dsl, not speed but the physical layer

Correct

 

[vodacom3g]

All this hype about 3.6 HSPDA and increased upload speeds.
I ran the trial period on the increased speed offering.
Guess what, I am back to 1.8
I felt the gain was not all that great, so maybe 7.2(when they can get backhaul from Telkom) wil be worth waiting for.
Vodacom should offer this service to all their clients on packages of 1gb per month or greater at no extra cost.
That would make them a leader instead of a follower.
They always seem to be trailing MTN.

Providing the service as part of the bigger bundles is a viable option and one the P&S guys do look at continuously.

How do you get the 'trailing MTN' bit? Last time I looked the VC-3.6 service was available on every tower around the country while forumites over in the MTN sub-forum were hard-pressed to find a few 3.6 towers and no 7.2 at all. Seemed (at launch time) the only place MTN 3.6 and 7.2 was available was in the Sunday paper ads.

Might be a bit better now but I still know of only 1 MTN-7.2 tower in Cape Town (Harrington st).

 

[HavocXphere]

In the case of UPA, there is a substantial increase in speed (about a factor 3) in the RADIO UPLINK. So it must stand to reason that latency on that link will improve. But it's only one link in a long list of links that give you your gaming response.

By speed, do you mean the physical radio waves travel faster or do you just mean more kbps?

Sub 300 pings are certainly impressive though...

 

[vodacom3g]

Hi V3G

I hear what you are saying. But still speed does not = Latency.

With your example as stated above that is one Honda not 2 different ones and it's the same road.

Now if you have to test HSUPA latency against the like of iBurst as an example. That is a completely scenario aint it ?

If it's the same bike on the same road and you increase speed (bits/s), time (latency) must decrease.

But it seems you're trying to get a feel for how much closer to other type of services, it will be with the addition of UPA. And doing real-world gaming tests (over a longish period) will be the best bet.

 

[vodacom3g]

By speed, do you mean the physical radio waves travel faster or do you just mean more kbps?

Sub 300 pings are certainly impressive though...

The day I get radio waves to travel faster, I'm taking the whole myADSL forum on a 2 month piss-up.

No, I meant the speed at which you clock bits in and out the system, i.e. the bit-rate.

 

[Pitbull]

If it's the same bike on the same road and you increase speed (bits/s), time (latency) must decrease.

But it seems you're trying to get a feel for how much closer to other type of services, it will be with the addition of UPA. And doing real-world gaming tests (over a longish period) will be the best bet.

Things like reliability, KM/L you get per bike and so forth ... hehe get what I'm saying

But yes you have it smack on the money. We need real time game testing, would you or RPM be able to put this to the test against the other wireless providers ?

Let's say

iBurst - 3G
MTN - HSDPA (1.8)
Vodacom - (3.6+HSUPA)

I would be interrested to see those results

I think that would be interresting to see. Lets leave Telkom as that is fixed line. Maybe chuck Neotel in the mix if someone here has access.

This would give us gamers a clear idea of which of these services would be a close contender for Telkom in general. Also maybe do the tests at different locations and so forth.

Pls it would save us gamers a hell of alot of frustration

 

[HavocXphere]

The day I get radio waves to travel faster, I'm taking the whole myADSL forum on a 2 month piss-up.

/me Makes mental note to include extra trickiness in future questions to ensure said piss-up happens.

/me mumbles something about distorted spacetime & blackholes.

No, I meant the speed at which you clock bits in and out the system, i.e. the bit-rate.

Kinda makes sense now. Higher bit-rate = less time wasted until next opportunity to send info = less delay = lower ping.

But, the effect is tiny: For 384k vs 4mbps the effect is 0.00230471 ms.

So I still don't get how a higher speed / bitrate equates (materially) to lower latency.

If it's the same bike on the same road and you increase speed (bits/s), time (latency) must decrease.

Thats the thing. You are increasing bitrate not speed. i.e. the ability to send many bikes down the same road in quick succession does not mean the first bike gets there any faster. And latency is all about how fast the bike can get to london and back.

 

[lilDeath]

International latency was impressively low for a wireless service with an average ping time to Google of below 300ms and below 400ms to BBC. Locally ping times were around the 100ms mark.

And that is why you cant use UPA for gaming, as said in another thread increasing the uplink won’t improve latency. Don’t even get me started on packet loss on a radio interface...

If that is the latency we can expect from such a high-speed service, they can keep it. I'm getting better on my 'sweet-ass' 384 ADSL.

VC and MTN should rather ensure that they have proper coverage in specifics suburbs in JHB, like Bryanston, instead of punting yet another 'awesome' service.

All just smoke and mirrors imo... need more $$$

meh