On networks using NAT64 (and maybe also for DS-Lite, although I'm not so sure there), the IPv6 path is more direct than the IPv4 one since it doesn't need to go through a CG-NAT, which might be at capacity (tracking every TCP connection requires memory) or located farther away than the nearest IPv6 egress router, making routing more indirect.

I believe all three large US mobile carriers use NAT64 at this point exclusively, and CG-NAT is quite common in DOCSIS cable networks.

I am stuck in DS-Lite, and indeed IPv6 is direct while v4 is served via shitty CGNAT

Simpler routing because address blocks are not fragmented. IPv6 does not have a header checksum so that's one less task for routers. Packet fragmentation is no longer done by routers. IPv6 capable gear tends to be newer and higher performance. There's no NAT step or even worse CGNAT.

For those that were wondering: Unlike in IPv4, IPv6 routers never fragment IPv6 packets. Packets exceeding the size of the maximum transmission unit (MTU) of the destination link are dropped and this condition is signaled by a Packet too big ICMPv6 message to the originating node, similarly to the IPv4 method when the Don't Fragment bit is set.

> Packet fragmentation is no longer done by routers.

The overwelming majority of IPv4 packets have the DF bit set. A large number of services drop inbound fragments anyway (if you get more than a couple a minute, it's almost certainly abuse).

> By reducing header complexity, IPv6 contributes to better overall network performance and efficiency. This is especially noticeable in applications sensitive to latency, such as real-time communication and online gaming.

https://www.prefixbroker.com/news/is-ipv6-faster-than-ipv4-a....

This is the theoretical benefit. These have never materialize in real life.

See https://community.cisco.com/kxiwq67737/attachments/kxiwq6773...

Sometimes networks will route IPv6 traffic over different paths compared to IPv4 although that could actually be either an improvement or an regression depending on which gets the better path.

IPv6 was a lot more than a numbering scheme, there are performance (and security?) improvements in a number of aspects of IP networking, like more compression in parts of the messages IIRC.

Gonna have to disagree with that. IPv6 was designed in the 90s and lots of it feels like it is. Icmpv6 is fundamental to the implementation and already responsible for multiple remotely exploitable vulnerabilities. Couple that with almost no one understanding it well besides a true handful of engineers and you have yourself a security nightmare. Proof of this is this very thread with the talk about “trying a local link address”.

Everything about v6 is over engineered with sharp edges.

A possible explanation is the service provider routes IPv4 over their old routers but routes IPv6 over their new routers.