ADSL Ping results

[Kalvaer]

I know there have to be some technical people here and would like to ask if this is the SAT3 pipe or what is causing this.

I asked a friend of mine in Texas to do a trace route to my PC while I did one to him. Both of us hit a problem at the same place.... TELKOM.

His words to me after he did the trace were
Pete: "Everything is perfect until I hit Telkom... who is that"
Me: " hmmm My ISP and ADSL provider
Pete: "BWHAHAHAHA... Unlucky for you that sucks bro"

Friends Trace route to me:

1 7 ms 7 ms 8 ms 10.99.224.1
2 7 ms 11 ms 8 ms pos2-0.crchtxcw-ubr1.stx.rr.com [24.170.32.97]
3 8 ms 7 ms 7 ms pos5-2.crchtxhe-rtr1.stx.rr.com [24.170.32.45]
4 7 ms 8 ms 7 ms pos3-0.crchtxhe-rtr2.stx.rr.com [24.170.32.126]
5 12 ms 14 ms 31 ms son2-0-3.hstqtxl3-rtr1.texas.rr.com [24.93.34.58]
6 14 ms 15 ms 15 ms pop1-hou-P4-0.atdn.net [66.185.133.157]
7 41 ms 15 ms 16 ms bb1-hou-P2-0.atdn.net [66.185.150.148]
8 13 ms 14 ms 14 ms bb1-dls-P6-0.atdn.net [66.185.152.132]
9 20 ms 19 ms 30 ms pop2-dls-P0-0.atdn.net [66.185.133.97]
10 54 ms 52 ms 52 ms GlobalCrossing.atdn.net [66.185.134.58]
11 134 ms 135 ms 169 ms so6-0-0-2488M.ar2.AMS2.gblx.net [67.17.65.234]
12 132 ms 132 ms 132 ms Telkom-SA-Limited.so-0-2-0.ar2.AMS2.gblx.net [64.210.21.46]
13 134 ms 134 ms 137 ms ny-ip-dir-globalc-pos-6-0.telkom-ipnet.co.za [196.43.9.54]
14 136 ms 145 ms 136 ms ny-ip-dir-stacker-ge-3-0-0-101.telkom-ipnet.co.za [196.43.9.194]
15 140 ms 152 ms 147 ms ny-ip-dir-globalc.telkom-ipnet.co.za [196.43.9.250]
16 453 ms 453 ms 453 ms rrba-ip-lir-1-pos-6-0.telkom-ipnet.co.za [196.43.9.113]
17 464 ms 453 ms 460 ms rndf-ip-er-2-ge-2-0-0.telkom-ipnet.co.za [196.43.11.158]
18 451 ms 452 ms 452 ms rndf-ip-esr-1-fe-1-0-4.telkom-ipnet.co.za [196.43.10.6]
19 470 ms 473 ms 472 ms rndf-146-16-230.telkomadsl.co.za [XXX.XXX.XXX.XXX]

My Trace route to him (first hop is my firewall, last one was his firewall):
1 * * * Request timed out.
2 20 ms 19 ms 19 ms rndf-146-0-01.telkomadsl.co.za [165.146.0.1]
3 21 ms 19 ms 21 ms rndf-ip-er-2-fe-12-0-1-1.telkom-ipnet.co.za [196.43.10.181]
4 354 ms 354 ms 354 ms ny-ip-dir-globalc-pos-7-0.telkom-ipnet.co.za [196.43.9.149]
5 343 ms 347 ms 363 ms so-2-2-1-ecr1.nyk.cw.net [195.2.5.85]
6 352 ms 338 ms 392 ms so-3-0-0-dcr1.nyk.cw.net [195.2.3.13]
7 398 ms 355 ms 499 ms 65.59.192.13
8 450 ms 344 ms 337 ms ae-1-55.bbr1.NewYork1.Level3.net [4.68.97.129]
9 379 ms 379 ms 380 ms as-0-0.bbr2.Dallas1.Level3.net [64.159.0.137]
10 395 ms 393 ms 387 ms ge-6-0-0-56.gar1.Dallas1.Level3.net [4.68.122.162]
11 400 ms 385 ms 388 ms ROADRUNNER.gar1.Dallas1.Level3.net [4.78.232.6]
12 457 ms 462 ms 477 ms son2-0-0.hstqtxl3-rtr1.texas.rr.com [24.93.34.30]
13 462 ms 468 ms 464 ms pos2-0.crchtxhe-rtr2.stx.rr.com [24.93.34.57]
14 479 ms 480 ms 467 ms pos3-0.crchtxhe-rtr1.stx.rr.com [24.170.32.125]
15 464 ms 468 ms 464 ms pos1-0.crchtxcw-ubr1.stx.rr.com [24.170.32.46]
16 468 ms 468 ms 481 ms pos1-0.crchtxcw-ubr2.stx.rr.com [24.170.32.98]
17 * * * Request timed out.

Both of us have a 300ms increase after hitting the telkom "hop" between South Africa and New York, What causes this, is it the distance?

 

[stoke]

Well - my theory ...
Welcome to port shaping and loooooong distance - what you see there is just after the last point in south africa, also the shaping point.

Here-s mine to the 2nd last point - I use AYCE which is a SAIX line...

2 18 ms 19 ms 19 ms rndf-146-0-01.telkomadsl.co.za [165.146.0.1]
3 20 ms 19 ms 19 ms rndf-ip-er-2-fe-12-0-0.telkom-ipnet.co.za [196.43.10.5]
4 337 ms 340 ms 330 ms ny-ip-dir-globalc-pos-7-0.telkom-ipnet.co.za [196.43.9.149]
5 339 ms 330 ms 350 ms so-2-2-1-ecr1.nyk.cw.net [195.2.5.85]
6 348 ms 330 ms 330 ms so-3-0-0-dcr1.nyk.cw.net [195.2.3.13]
7 331 ms 330 ms 330 ms 65.59.192.13
8 332 ms 330 ms 330 ms ae-1-51.bbr1.NewYork1.Level3.net [4.68.97.1]
9 371 ms 410 ms 380 ms as-0-0.bbr2.Dallas1.Level3.net [64.159.0.137]
10 372 ms 380 ms 380 ms ge-7-0-0-53.gar1.Dallas1.Level3.net [4.68.122.66]
11 380 ms 380 ms 380 ms 67.72.0.46
12 450 ms 450 ms 450 ms son2-0-0.hstqtxl3-rtr1.texas.rr.com [24.93.34.30]
13 461 ms 460 ms 460 ms pos2-0.crchtxhe-rtr2.stx.rr.com [24.93.34.57]
14 458 ms 460 ms 460 ms pos3-0.crchtxhe-rtr1.stx.rr.com [24.170.32.125]
15 478 ms 460 ms 460 ms pos1-0.crchtxcw-ubr1.stx.rr.com [24.170.32.46]
16 460 ms 460 ms 460 ms pos1-0.crchtxcw-ubr2.stx.rr.com [24.170.32.98]

Methinks it's more shaping that long distance. But - proof is hard to come by.

 

[Kalvaer]

Is there no way at all to get this proof though? I'll ask my friend to tracert a site in the UK as the distance should be the same and see if he gets the 300ms lag there as well

Also does this mean they are not only doing it to us in South Africa, but people "routing" into South Africa, Are they trying to destroy us by driving away potential international business

I've heard "rumors" that Telkom is actually routing people via A satalite link rather than the SAT3 pipe? Due to the fact that you get anything from 300ms to 500 ms lag from Sat links, But that I have no proof on that either?

 

[stoke]

Teklom will route using the least cost possible. They have full authority to use any means available to them. If sat is cheaper, then you can bet that we're running on the SATTELITE - link.

Here's a trace on my work's network - using IS - they have assured us that it is not sattelite.

2 20 ms 20 ms 20 ms 168.209.117.81
3 30 ms 20 ms 20 ms core2b-rba-gi8-0-0.rtr.isnet.net [196.26.0.9]
4 261 ms 290 ms 281 ms 168.209.0.86
5 300 ms 261 ms 270 ms core1b-ny.nmszone.is.co.za [168.209.244.3]
6 301 ms 280 ms 291 ms POS1-2.IG2.NYC4.ALTER.NET [208.192.177.205]
7 280 ms 281 ms * 0.so-0-0-0.XL1.NYC4.ALTER.NET [152.63.22.14]
8 261 ms 280 ms 271 ms 0.so-6-0-0.BR1.NYC4.ALTER.NET [152.63.21.77]
9 281 ms 290 ms 291 ms so-8-1.car1.NewYork1.Level3.net [209.244.160.181]
10 291 ms 280 ms 281 ms ae-1-55.bbr1.NewYork1.Level3.net [4.68.97.129]
11 341 ms * 331 ms ae-0-0.bbr2.Dallas1.Level3.net [64.159.1.110]
12 361 ms 330 ms 320 ms ge-6-0-0-56.gar1.Dallas1.Level3.net [4.68.122.162]
13 360 ms 331 ms 320 ms ROADRUNNER.gar1.Level3.net [67.72.0.34]
14 330 ms 321 ms 360 ms son2-0-0.hstqtxl3-rtr1.texas.rr.com [24.93.34.30]
15 340 ms 331 ms 360 ms pos2-0.crchtxhe-rtr2.stx.rr.com [24.93.34.57]
16 380 ms 331 ms 340 ms pos3-0.crchtxhe-rtr1.stx.rr.com [24.170.32.125]
17 330 ms 330 ms 341 ms pos1-0.crchtxcw-ubr1.stx.rr.com [24.170.32.46]
18 351 ms 340 ms 321 ms pos1-0.crchtxcw-ubr2.stx.rr.com [24.170.32.98]

Note how this trace jumps up before it's even left SA. So methinks it's definately a traffic control mechanism. Remeber that ICMP packets (ping) will always get lowest priority.

 

[[)roi(]]

My take on the subject is rather port shaping + long distance, plus a bad or no ratio guarantee. If it were satellite you would expect the delay to be significantly higher, from my experience satellite latencies are closer to 1s or more...
Basically it's the monopoly doing anything they want to...

 

[Tinuva]

5 300 ms 261 ms 270 ms core1b-ny.nmszone.is.co.za [168.209.244.3]

Uh notice that is not in south africa.

-ny = New York

thus ping still gets high as it left South Africa. Do not make wrong assumptions just because it has a .co.za at the end...

same with:
4 261 ms 290 ms 281 ms 168.209.0.86

 

[Kalvaer]

Uh notice that is not in south africa.

-ny = New York

thus ping still gets high as it left South Africa. Do not make wrong assumptions just because it has a .co.za at the end...

same with:
4 261 ms 290 ms 281 ms 168.209.0.86

granted the hop stated is not NY, the one before that is not not a "SA" IP either...

BUT!

The ping you get from that is AFTER you leave SA which means you had the bad "latency" before you had that hop...

IE:

Driving from point A to point B takes 5 min. From point B to C takes 30 min.. Therfore the drive from A to C is 35 min... each time you go from one hop to the next you have to take into account how much time it took you to get from point A to B, before you get to C.

So if there is a 300ms difference from B to C, there is a bottle neck or a latency problem at B.. IE TELKOM!!!!!

My reasoning for suspecting satalite is due to a certain person I know who works for a HUGE international Pharmacutecial (spelling?) company who I am afraid I can not mention due to his position at the company and what he has told me about what his company had to go through to force Telkom to remove them from their Satalite link. The ping reply from a Satalite depends on how far it is from "earth" If you have a Satalite in shallow orbit you can get a 300ms reply, if its in outer orbit (like the GPS Sats) its 500ms or greater... I have no clue what Sats Telkom has access to and what they use... though I would think they would use low orbit sats due to the latency achieved

This is of couse all speculation based on word of mouth, and is as such in admisssable.. BUT (again) you can not disprove the fact that each time you hop from a telkom based IP, to an outside location. you get a 300ms difference in time... I'll send the Trace route from my buddy in Texas in the morning and see if the distance to london causes the same difference

 

[Tinuva]

Ok I think you got it wrong, but are close to it. A->B 5min and B->C 30min. Now you say that point B is the problem. I dont see how the point is the problem if the road inbetween B and C is very long and narrow. Its a Highway like 8am traffic between Pretoria and Johannesburg, too many traffic cars and not enough lanes. Thus the problem is not point B, but the road between point B and C. Dont blame a router when the problem is the amount of traffic telkom allow over the SAT3 cable.

 

[stoke]

Tinuva - more speculation.
On the return route - when the US guy was tracing to SA, the problem started at the same point. Kinda obliterates your theory.

 

[Kalvaer]

Ok I think you got it wrong, but are close to it. A->B 5min and B->C 30min. Now you say that point B is the problem. I dont see how the point is the problem if the road inbetween B and C is very long and narrow. Its a Highway like 8am traffic between Pretoria and Johannesburg, too many traffic cars and not enough lanes. Thus the problem is not point B, but the road between point B and C. Dont blame a router when the problem is the amount of traffic telkom allow over the SAT3 cable.

Granted yes... I should say then that the problem starts somewhere between B and C.

Last time I checked though (at work now with out ADSL so cant check), if you do a trace to a UK address you end up hitting the same router where the problem of the 300ms ping starts. From logical deduction, that tells me that the router in question is the router used on our side of the SAT3 cable (unless we are being routed via the USA to get to the UK), therefore the problem can not be the cable inbetween B and C.. but B itself

Soon as my buddy in the US wakes up Ill get him to do a trace for us from Texas to the UK, Granted the distance is smaller to the UK but it should be able to give a rough guide as to if the ocean trip is whats causing large pings.. If there is the same +/-300ms ping then granted its the long distance, if its not the same.. the distance can be ruled out.

I'll also ask him to do a trace to Moscow, as there the distance is very close to the distance needed to travel in a direct line from him to SA, even if there are more than likely going to be many more hops, or if there are any other faulty/bad routers along the way.

Could prove myself completely wrong.. Though knowing Telkom I still think its their router, That or somebody forgot to feed the hamster that drives it this week

 

[Tinuva]

Tinuva - more speculation.
On the return route - when the US guy was tracing to SA, the problem started at the same point. Kinda obliterates your theory.

I dont know about you but I think I can read.

First of all those 2 routes do not even follow the exact same path in both directions.

Second of all:

Friends Trace route to me:
14 136 ms 145 ms 136 ms ny-ip-dir-stacker-ge-3-0-0-101.telkom-ipnet.co.za [196.43.9.194]
15 140 ms 152 ms 147 ms ny-ip-dir-globalc.telkom-ipnet.co.za [196.43.9.250]
16 453 ms 453 ms 453 ms rrba-ip-lir-1-pos-6-0.telkom-ipnet.co.za [196.43.9.113]

My Trace route to him:
4 354 ms 354 ms 354 ms ny-ip-dir-globalc-pos-7-0.telkom-ipnet.co.za [196.43.9.149]
5 343 ms 347 ms 363 ms so-2-2-1-ecr1.nyk.cw.net [195.2.5.85]
6 352 ms 338 ms 392 ms so-3-0-0-dcr1.nyk.cw.net [195.2.3.13]

Take notice, its not the same IP that the lag starts!
Lag starts before the hop. Thus its still between point B and C and not B or C.

Im telling you now, its not the router, its the cable between the 2 routers.

 

[stoke]

Sry - seems I can't read.

 

[Kalvaer]

Im telling you now, its not the router, its the cable between the 2 routers.

I really hope its not.. There aint much we can do about the cable...

Still though I agree the IP address are not the same. It would be virtually impossible to get the same route from both directions. Who can tell what DNS table each router has and where its going to send you next. Of course both those IP address are for the same subnet, and both owned by Telkom, And Yes both are somewhere between the cable.

One thing in comman that I found with my testing last night with Pete... he always got bad pings on his 4 last hop to me. I always got bad pings on my 4th hop from me.

I did ask from the beginning if it was the cable/distance etc, and it might very well be the distance... Anyway to find out if it is in fact the distance between the cable though? or a Satalite link?

 

[Kalvaer]

Tinuva, Just checking again... I can see where you say it is between B or C, the large ping is always between New York and South Africa, that much we agree on right?
So things we know:

1. Stoke's pings from his two ISP he also got about a 300ms difference between SA and NY,
2. His work ISP also had the 300ms ping when leaving SA (been told its not Satalite
3. My ISP and his both had bad pings on the 4th hop (and both the same IP addy)
4. My friend also got bad pings inbetween NY and SA (4th hop from me)

These can all be caused by a few things:

• Long Distance
• Bad Router (or group of routers in the same subnet)
• Routing via Satalite

The bad router could also very well be a set of rules set up on that whole subnet, owned by Telkom or "Shaping" as Stoke mentioned?
Also why is there the same 300ms ping time when traveling from SA-NY, as there is from SA-London, if it was distance it would be related some how
Distance from SA to NY (Straight line): +/- 12 500km
Distance from SA to London (Straight line) +/- 9600km
Would there not be say a smaller ping from SA London due to the distance?

Now as well am I right in saying the SAT3 Cable is Fiber optic?

If so:
Speed = Distance/Time

Speed of Light in Glass +/- 1.9696 x 10 to the power of 8, m/s
Distance to NY in Meters: 125 000 000 m

So therefore:

Time for light to travel in a straight line from SA to New York: 0.0635 Secs (or 60ms)

Let assume that for the routers to read this data and transfer it is the same as our second hop between two routers of +/- 10 ms.

Add that to the above, for both routers (even if that is WAY to much) gives us 80ms, And for any errors, not really a straight line, etc... lets double it!

Thats 160ms for the two routers to send data. Where is the extra 140ms coming from?

 

[Tinuva]

a.Long Distance
- a definite yes

b.Bad Router (or group of routers in the same subnet)
- dont think so, most major isps fix this the same day

c.Routing via Satalite
- also dont think so, satellite links give 600ms++ pings

d.shaping (you didnt mention this)
- also yes

the best best best latency ive seen from SA to International was 199ms directly on teh Internet Solutions backbone.

With my unshaped account I once got 210ms to google, when traffic was offline and just cambe back, as soon as rest of SA realised this, pings went back up to 280ms

 

[Kalvaer]

With my unshaped account I once got 210ms to google, when traffic was offline and just cambe back, as soon as rest of SA realised this, pings went back up to 280ms

199 ms to 280 ms round trip would seem right if you do the calculations of the speed the fibre can handle..

300ms between one router and another does not... Of course you did say your account is unshaped... I know mine is shaped, so Stoke could be right then that it is distance and port shaping, so option B of the router being the problem taking into regards that the router is slowing down or "shaping" certain traffic is valid... right?

 

[Tinuva]

According to me shaping traffic is a feature, not a router problem.

According to telkom ICMP packets get about 1% traffic on the sat3 cable for the shaped accounts. Yes with my shaped telkom account (when i had one) was about 340 - 400ms to google.

When a router slows you down its not a problem but what you haved payed for. You could always go buy diginet have have close to that 200ms pings to google. I know back when uunet adsl still were cool in beta, pings was 230-240ms to google...which means even unshaped telkom accounts get bad bandwidth. Im still sure the router is not broken, however it might be your problem, as you see it as the bad point in your connection

 

[Kalvaer]

I agree 100 % that I'm getting what I paid for as I asked originally with the first post I was asking what actually caused the large ping at a specific point.

The router might not be broken but it is the "cause" of the large ping and would seem to be due to shaping of the line correct?, so the "problem" is in fact at B.. and therefore not entirely due to the distance between B and C?

What is completly wrong though is that it would seem that the router is not only shaping outgoing traffic, but also incoming from the USA... Is that what the people on the other side of that router are paying for when they try to reach sombody on the otherside that is paying for a shaped service? ADSL is not known for its great latency, and I never expected it to be great, just wanted to know what causes the HUGE difference, At Telkoms current pricing I have to happy with what I have.

Guess my question has been answered though, thanks But I'll add the tracert from the USA to Moscow and London when I get them later as well for interests sake

EDIT:
Seems I cant read either LOL.. Didn't see that you had put in an option D Tinuva for shaping... So you had answered earlier that it was distance and shaping... though also if you read my post above that, you will see that I also mentioned

The bad router could also very well be a set of rules set up on that whole subnet, owned by Telkom or "Shaping" as Stoke mentioned?

Meaning that my statement for "Bad Router" meant either faulty router or a Router that Shaped the traffic

Seems none of us can read

 

[stoke]

Methinks that a policy change occured during 2002, september methinks. Back then ADSL was sub 250 to Denmark. It waz lovely, and I made a lotta friends there playing online games. Them days are over.

 

[Kalvaer]

LOL aint that the truth Stoke... all my friends in the US where all meet through online gaming, Some of whom I've gone over and visited... Of course my reasoning for not playing anymore is being stupid enough to decide to study a BSC degree in Physics part time.. which means No free time.

anyway..

The Tracerts from Texas to London (Oxford University):

Tracing route to www.ox.ac.uk [163.1.0.45]
over a maximum of 30 hops:

1 9 ms 7 ms 7 ms XXX.XXX.XXX.XXX
2 6 ms 18 ms 21 ms pos2-0.crchtxcw-ubr1.stx.rr.com [24.170.32.97]
3 7 ms 8 ms 7 ms pos5-2.crchtxhe-rtr1.stx.rr.com [24.170.32.45]
4 8 ms 8 ms 8 ms pos3-0.crchtxhe-rtr2.stx.rr.com [24.170.32.126]
5 13 ms 13 ms 13 ms son2-0-3.hstqtxl3-rtr1.texas.rr.com [24.93.34.58]
6 27 ms 17 ms 30 ms son2-0-0.dllatxl3-rtr1.texas.rr.com [24.93.34.29]
7 28 ms 17 ms 17 ms so-2-1-0-0.gar1.Dallas1.Level3.net [67.72.0.33]
8 17 ms 20 ms 19 ms ae-1-56.bbr2.Dallas1.Level3.net [4.68.122.161]
9 53 ms 53 ms 51 ms ae-0-0.bbr2.Washington1.Level3.net [64.159.0.230]
10 124 ms 126 ms 124 ms as-0-0.bbr1.London2.Level3.net [4.68.128.101]
11 125 ms 124 ms 124 ms ge-3-0-0.gar1.London2.Level3.net [4.68.124.62]
12 126 ms 124 ms 126 ms 195.50.116.202
13 123 ms 126 ms 124 ms po2-0.lond-scr4.ja.net [146.97.35.221]
14 125 ms 126 ms * po1-0.read-scr.ja.net [146.97.33.26]
15 126 ms 127 ms 127 ms po2-0.oxford-bar.ja.net [146.97.35.162]
16 128 ms 128 ms 126 ms 163.1.0.45

The Tracert to Novosibirsk State University in Russia (Which by the way is about 2000 km further than Moscow in сивезя... Thats if I can still remember how to spell in russian)

Tracing route to aurora.nsu.ru [193.124.215.195]
over a maximum of 30 hops:

1 8 ms 7 ms 8 ms XXX.XXX.XXX.XXX
2 7 ms 7 ms 7 ms pos2-0.crchtxcw-ubr1.stx.rr.com [24.170.32.97]
3 8 ms 7 ms 7 ms pos5-2.crchtxhe-rtr1.stx.rr.com [24.170.32.45]
4 8 ms 7 ms 8 ms pos3-0.crchtxhe-rtr2.stx.rr.com [24.170.32.126]
5 12 ms 13 ms 13 ms son2-0-3.hstqtxl3-rtr1.texas.rr.com [24.93.34.58]
6 18 ms 27 ms 17 ms son2-0-0.dllatxl3-rtr1.texas.rr.com [24.93.34.29]
7 22 ms 20 ms 19 ms so-5-2-0-0.gar1.Dallas1.Level3.net [4.78.232.5]
8 17 ms 18 ms 17 ms ae-1-54.bbr2.Dallas1.Level3.net [4.68.122.97]
9 51 ms 52 ms 64 ms as-4-0.bbr1.Washington1.Level3.net [64.159.0.138]
10 51 ms 50 ms 51 ms ge-7-1.hsa2.Washington1.Level3.net [4.68.121.74]
11 52 ms 51 ms 78 ms ALLINSON-COM.hsa2.Level3.net [63.215.131.22]
12 55 ms 56 ms 91 ms so-7-0-0-dcr1.nyk.cw.net [195.2.3.2]
13 127 ms 128 ms 126 ms so-2-0-0-dcr1.tsd.cw.net [195.2.10.114]
14 134 ms 134 ms 135 ms so-0-0-0-dcr2.amd.cw.net [195.2.10.145]
15 132 ms 131 ms 132 ms so-5-0-0-bcr1.amd.cw.net [195.2.10.29]
16 143 ms 133 ms 134 ms so-3-0-0-zcr1.amt.cw.net [208.173.211.238]
17 138 ms 136 ms 133 ms ge-0-3-0-zar1.amt.cw.net [208.173.220.131]
18 144 ms 134 ms 135 ms ano1.amt.cw.net [208.173.220.98]
19 183 ms 171 ms 171 ms MSK-M9-RBNet-7.RBNet.ru [195.209.14.181]
20 219 ms 218 ms 218 ms NSK-RBNet-3.RBNet.ru [195.209.14.74]
21 215 ms 215 ms 217 ms 217.79.60.46
22 * * * Request timed out.
23 * * * Request timed out.
24 * * * Request timed out.
25 227 ms 227 ms 228 ms NSC-fo-NSU.nsu.ru [193.124.208.61]
26 232 ms 233 ms 224 ms aurora.nsu.ru [193.124.215.195]

Is it just me then or SA the only place that seems to be "shaping" incoming International traffic..
Also from those specs and please rip them apart and show me if I'm wrong the trans-Atlantic trip for a IP packet is 70ms.. NOT 300ms.. and not even the 200ms that you are getting Tinuva, so even on your unshaped account.. you are being RIPPED OFF!!!!