Ping times vs my Spark Connection

Forums › Vetta and Quic › Ping times vs my Spark Connection

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#319924 14-Jun-2025 17:56

Have been hearing about Quic a little and have been looking into whether I should consider switching over from Spark. As part of this was doing a few rudimentary pings and comparing it against a couple of public Quic based ping servers that I came across in some threads.

For the most part the times seem roughly on par but a few of them, which I think are Akamai based sites, showed as about 14ms on the Quic based server but show as 2ms locally. Just wondering if this difference would really show itself in real world performance and whether it's due to something like Spark having an Akamai delivery node within it's network vs Quic only peering to one?

saf

221 posts

Master Geek
+1 received by user: 533

ID Verified

Trusted

Vetta Group

Subscriber

#3384064 14-Jun-2025 18:14

It would depend on where you are in the country, however my guess would be that as Quic peer with Akamai in Auckland, my assumption is you’re a bit further afield from Auckland, and Spark host an Akamai cache closer to you.

That said, based on the content Akamai deliver being predominantly download and streaming content, I would say it’s highly unlikely you would ever notice a ~12ms difference to Akamai day-to-day.

Quic sits on the AS64073 network which engages heavily in peering in Auckland, Wellington, Christchurch, as well as outside NZ.

My views are as unique as a unicorn riding a unicycle. They do not reflect the opinions of my employer, my cat, or the sentient coffee machine in the break room.

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3384065 14-Jun-2025 18:19

Sorry, yes, location would probably have helped as I'm in Kāpiti so that sounds about right

michaelmurfy

meow
13579 posts

Uber Geek
+1 received by user: 10910

Moderator

ID Verified

Trusted

Lifetime subscriber

#3384158 15-Jun-2025 03:05

@r2b2 it also can depend on the DNS servers configured on the endpoint.

If you run a test from the Paraparaumu hosts here: https://bgp.he.net/AS64073#_traceroute you'll be running it via one of my probes - All my RIPE / GLOBALPING probes are also configured to use Quic's DNS servers.

Lastly, going to https://www.quic.nz/status/ you'll see a Speedtest at the bottom you can try 😊

Michael Murphy | https://murfy.nz
Referral Links: Quic Broadband (use R122101E7CV7Q for free setup)

Are you happy with what you get from Geekzone? Please consider supporting us by subscribing.
Opinions are my own and not the views of my employer.

Lias

5655 posts

Uber Geek
+1 received by user: 3978

ID Verified

Trusted

Lifetime subscriber

#3384297 15-Jun-2025 11:48

r2b2:

... whether I should consider switching over from Spark..

Yes, yes you should :-)

I'm a geek, a gamer, a dad, a Quic user, and an IT Professional. I have a full rack home lab, size 15 feet, an epic beard and Asperger's. I'm a bit of a Cypherpunk, who believes information wants to be free and the Net interprets censorship as damage and routes around it. If you use my Quic signup you can also use the code R570394EKGIZ8 for free setup. Opinions are my own and not the views of my employer.

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3384318 15-Jun-2025 14:21

michaelmurfy:

@r2b2 it also can depend on the DNS servers configured on the endpoint.

If you run a test from the Paraparaumu hosts here: https://bgp.he.net/AS64073#_traceroute you'll be running it via one of my probes - All my RIPE / GLOBALPING probes are also configured to use Quic's DNS servers.

Lastly, going to https://www.quic.nz/status/ you'll see a Speedtest at the bottom you can try 😊

Thanks - I think that's the one I was testing with initially based on some other posts.

I have done a few Speedtests via the website and have been finding some strange results with download speeds. e.g. this is one example from my local Speedtest
---
Server: 5x5 Telecom - Los Angeles, CA (id: 21001)
        ISP: Spark New Zealand
Idle Latency:   159.84 ms   (jitter: 0.25ms, low: 159.69ms, high: 160.33ms)
   Download:   954.26 Mbps (data used: 1.6 GB)
               319.05 ms   (jitter: 76.10ms, low: 160.83ms, high: 476.76ms)
     Upload:   114.14 Mbps (data used: 200.2 MB)
               246.11 ms   (jitter: 69.37ms, low: 163.71ms, high: 333.44ms)
Packet Loss:     0.0%
Result URL: https://www.speedtest.net/result/c/4518f480-e54d-4d7d-8eed-95e39c3dc640

---

When run via the Quic status page I get

---
Download Speed: 398.34 Mbps
Download Latency: 240.702 ms (150.627 - 787.838)
Download Jitter: 69.588 ms
Upload Speed: 538.81 Mbps
Upload Latency: 227.043 ms (149.499 - 781.511)
Upload Jitter: 68.331 ms
Ping Latency: 150.441 ms (150.241 - 151.669)
Ping Jitter: 0.739 ms
Server Name: 5x5 Telecom (21001)
Server Location: Los Angeles, CA, United States
---

My download (which is on a Fibre Max plan) seems crazily higher so wondering if my test is somehow showing a conflated speed or if the Quic Speedtest is perhaps not working to the best of it's ability at the moment?

MaxineN

Max
2049 posts

Uber Geek
+1 received by user: 1662

ID Verified

Trusted

Subscriber

#3384498 15-Jun-2025 20:06

That certainly can be replicated.

To reach 5x5 we go via HE.

Fairly direct.

https://www.speedtest.net/result/c/52956c0c-3f9e-4fd2-8554-20f6a9d74829

Kind of pants to be honest.

Hivelocity however... 😉 via HE too (although to be fair, v4 vs v6 testing)

Just via V4.

Seems HE v4 might be a bit sluggish but this may not affect day to day... who knows, this isn't really scientific. Also using Vetta's resolvers.

Ramblings from a mysterious lady who's into tech. Warning I may often create zingers.

Quic Broadband

Move to New Zealand's best fibre broadband service (affiliate link). Free setup code: R587125ERQ6VE. Note that to use Quic Broadband you must be comfortable with configuring your own router.

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3384682 16-Jun-2025 13:37

Thanks. It's good to see that a reasonably high download speed can be achieved.

It's a little difficult for me to ignore the high Spark download speed results compared with the test results available via the status page but perhaps I'm starting to become a bit too obsessed with Speedtest results and should start thinking about the other advantages Quic brings to the table 😅

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3385015 17-Jun-2025 09:18

michaelmurfy:

@r2b2 it also can depend on the DNS servers configured on the endpoint.

If you run a test from the Paraparaumu hosts here: https://bgp.he.net/AS64073#_traceroute you'll be running it via one of my probes - All my RIPE / GLOBALPING probes are also configured to use Quic's DNS servers.

Lastly, going to https://www.quic.nz/status/ you'll see a Speedtest at the bottom you can try 😊

I've been doing a bit of looking at a few traceroutes and have noticed that google.com seems to resolve to US based servers. Is that what actually happens in practice or is it just a byproduct of how the probes work? As far as I can tell (and maybe I'm wrong so happy to be corrected) it looks like AS64073 peers to Edge-IX in Sydney along with the Google AS15169 so I'd have thought it would go to Sydney rather than US?

saf

221 posts

Master Geek
+1 received by user: 533

ID Verified

Trusted

Vetta Group

Subscriber

#3385016 17-Jun-2025 09:53

r2b2:

I've been doing a bit of looking at a few traceroutes and have noticed that google.com seems to resolve to US based servers. Is that what actually happens in practice or is it just a byproduct of how the probes work? As far as I can tell (and maybe I'm wrong so happy to be corrected) it looks like AS64073 peers to Edge-IX in Sydney along with the Google AS15169 so I'd have thought it would go to Sydney rather than US?

Some Google services are served out of USA, and the vast majority of their IPs will be geolocated there.

However, we do peer both over multiple IX's, and privately, with Google in Sydney. You could try a trace to the likes of 8.8.8.8 to see the route across the ditch, for example.

 Host                                                                                                               Loss%   Snt   Last   Avg  Best  Wrst StDev
 1. snip                                                                                                            0.0%     8    0.7   0.7   0.6   0.8   0.1
 2. bng1-chc2.vetta.net                                                                                             0.0%     8    5.6   5.7   5.0   7.2   0.7
 3. core1-chc2.vetta.net                                                                                            0.0%     8    6.2   6.1   5.0   7.1   0.6
 4. core1-akl2.vetta.net                                                                                            0.0%     8   18.8  18.3  17.3  19.4   0.6
 5. (waiting for reply)
 6. 72.14.194.162                                                                                                   0.0%     8   41.7  41.7  40.8  42.5   0.7
 7. 192.178.98.29                                                                                                   0.0%     8   41.7  42.0  41.1  42.6   0.5
 8. 142.250.234.213                                                                                                 0.0%     8   41.9  42.5  41.8  44.2   0.8
 9. dns.google                                                                                                      0.0%     7   41.0  42.2  41.0  43.4   0.8

My views are as unique as a unicorn riding a unicycle. They do not reflect the opinions of my employer, my cat, or the sentient coffee machine in the break room.

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3385029 17-Jun-2025 11:16

Thanks that's good to know. One "final" question, is it possible to have a Quic connection setup on a second ONT port so I can get hooked up and make sure things are looking good before ditching Spark?

michaelmurfy

meow
13579 posts

Uber Geek
+1 received by user: 10910

Moderator

ID Verified

Trusted

Lifetime subscriber

#3385032 17-Jun-2025 11:37

r2b2:

Thanks that's good to know. One "final" question, is it possible to have a Quic connection setup on a second ONT port so I can get hooked up and make sure things are looking good before ditching Spark?

~~Easy - when you sign up just pop in the comments "2nd port activation".~~

Edit: my information is outdated. Ignore me. See below vvv

Mighty Ape

Shop now at Mighty Ape (affiliate link).

saf

221 posts

Master Geek
+1 received by user: 533

ID Verified

Trusted

Vetta Group

Subscriber

#3385041 17-Jun-2025 11:47

r2b2:

Thanks that's good to know. One "final" question, is it possible to have a Quic connection setup on a second ONT port so I can get hooked up and make sure things are looking good before ditching Spark?

Yep, just select the "Additional ONT Port" option under Connection Type on the order form. :-)

My views are as unique as a unicorn riding a unicycle. They do not reflect the opinions of my employer, my cat, or the sentient coffee machine in the break room.

r2b2

586 posts

Ultimate Geek
+1 received by user: 109

#3385042 17-Jun-2025 11:53

Easy as - best thing to do is to try it out in (hopefully not literal) anger and see how it goes.

Thanks all for your help.