Hiveware for Gaming
info@hiveware.com
 

With the advent of IPv6, computer gaming can move away from its addiction to controllers. IPv6 introduces “swarm communication”. The Hiveware engine and thereby Hiveware for Gaming is essentially swarm communication. Here is a reprint of an interview with Geoff Huston from games.on.net how expounds on that topic:

With_the_advent_1_Geoff_Huston



Geoff, can you tell us a bit about yourself, your history with the development of the internet in Australia, and what you're up to now?

I stumbled across networking while working at the Australian National University in the 1980's, and had the opportunity to construct the first national Internet in Australia in the very early 1990's: the Australian Academic and Research Network (AARNet). I've been working in the Internet pretty exclusively since then, including a stint at Telstra working on Telstra's Internet Services and now I work for the Asia Pacific Network Information Centre (APNIC), the folk who manage Internet Protocol addresses in the Asia Pacific Region. I've spent a lot of time on IP technology, and have chaired a number of working groups in the Internet Engineering Task Force and also spent some years on the Internet Architecture Board. So I guess I'm pretty familiar with many aspects of the technology of IP!

Can you give a quick explanation of IPv6 for our readers?

"I had the opportunity to construct the first national Internet in Australia in the very early 1990's"

IP is a connectionless datagram protocol. By this we mean that "conversations" in an IP network are divided up into a sequence of data units, or "packets", and each packet is marked with a header to indicate which machine sent the packet into the network and which machine is the intended destination.The packet is then passed through the various switching units in the network along a path to the intended destination. The network is not necessarily a reliable network, in that individual packets may be discarded by the network en route, particularly when there is local overload along the packet's route. For all this to work every connected machine in the Internet requires a unique address - addresses that are in fact single strings of bits, or ones and zeros.

In the currently used version of the Internet Protocol addresses are constructed as strings of 32 bits. As the Internet grows and more and more devices connect to the Internet the pressures on an address pool of some 4 billion unique address values are becoming untenable, and in response to those pressures we've devised a new version of the Internet Protocol. Most of the protocol is unaltered, but one major change in version 6 of the protocol has been the alteration of the address fields, which have been extended from 32 bits in length to 128 bits in length.

IPv6 is IP, with bigger addresses!

One of the biggest bugbears for Australian gamers is the high ping when playing international games. Does IPv6 do anything to help with this?

The delay in sending a packet from one part of the world to another is constrained by a couple of basic physical properties. One is the speed of propagation of a photon through fibre optic cable, which is a little under two thirds of the speed of light in a vacuum, and the second is the circumference of the planet and the distribution of continents on the planet's surface. Unfortunately IPv6 does not make light go any faster, nor can it shrink the planet or redistribute the continents any faster than the existing geological processes. So no, ping times will remain much as they are today.

With_the_advent_2_Geoff_Huston
Traditional MMO's - a thing of the past?

But there are other benefits to speed that gamers will notice - you mentioned the ability to directly interact without the need for a network controller?

One important aspect of many games is multi-player environments. With IPv4 and Network Address Translators (NATs) used by home gateways the concept of allowing one game player to interact directly with another game player is virtually impossible, and most of the multi-player games relay on controllers to act as message relays to allow one player to interact with another.

IPv6 changes all this. With IPv6 the home gateway NATs no longer exist and it is possible to pass traffic directly from one game player to another. This concept of "swarm communication" has not been exploited in IPv4 simply because of the widespread use of NATs. It will be interesting to watch the evolution of game models in a world of IPv6, where it is possible to operate game environments without the necessity to synchronise a game state against a game controller.

Many online games such as MMO's, require a central connection to a server to ensure that players remain connected and to stop cheating, hacking, etc. With IPv6 removing the need for network controllers, will this allow us also to create more distributed online worlds - or is that simply something where we'll have to wait and see what develops out of the tech?

I think it is a wait and see proposition. Some MMO games now rely heavily on central controllers to coordinate the game state and this will not change any time soon. However there is now the potential for self-clustering games where the interaction between players is not centrally managed, and this form of game structure is facilitated by IPv6, whereas it has been extremely challenging to set up in IPv4.

Do game developers need to introduce any specific technology to take advantage of the benefits of IPv6, or will it happen naturally?

"the absence of NATs allows direct peer-to-peer communication, allowing concepts such as self-sustaining swarms of players who directly communicate with each other"

IPv6 is sufficiently close to IPv4 that existing models of game interaction will just work. A carefully coded game will work just as well over IPv6 as it does over IPv4, so nothing need change. But IPv6 can do so much more, and thats where things get interesting! IPv6 has better support for multicast, allowing a single sender to send a single packet that is received by all multicast recipients. The network itself can perform the necessary packet replication to ensure that everyone who is a multicast member will receive a copy of the packet. And, as I've already noted, the absence of NATs allows direct peer-to-peer communication, allowing concepts such as self-sustaining swarms of players who directly communicate with each other.

You mentioned some developers were being conservative about what the network is capable of supporting. Are there any concerns about IPv6 that are yet unaddressed?

Obviously the major concern here is the timing of its deployment. IPv6 is not a widely used protocol today, and there is still a fair way to go before a sizeable proportion of users are connected in IPv6 as well as IPv4. In terms of its technical capabilities IPv6 is perfectly capable of doing anything that can be done in Ipv4, and some more as well. From that respect thigs are looking good.

How is IPv6 progressing, and when do you anticipate we will have fully transitioned away from IPv4?

That is interesting. Microsoft and Apple have been shipping operating systems for some years that has IPv6 already built into the software, and now up to one half of the world's computers have an IPv6 protocol engine sitting in the system, but most computers never use it. The issue now is that very few ISPs provide IPv6 as part of their services.

Here in Australia, Internode is the first of the major commercial ISPs who are offering all their customers the ability to turn on IPv6. We are still waiting for the other ISPs to follow their lead, and time is running out for the Internet. Already in the Asia Pacific region we've run out of our stocks of IPv4 addresses, and an ISP who wants more IPv4 addresses to meet anticipated growth simply cannot obtain any - we've run out! So its very definitely time to change gear and start deploying networks and services using IPv6!

How long will this take? It would be good if this happened in a few months! Of course thats completely unrealistic, and a timespan of some five years or so looks more realistic at the moment. It might be many more years before IPv4 completely disappears, but it is likely that the norm for new connections to the Internet using IPv6 is something I expect to see in Australia in some 2-to-3 years from now, if not sooner.

Should gamers be looking at upgrading their own home hardware, or is IPv6 support already well-established in the existing equipment?

IPv6 is not a hardware upgrade to your computer. If you are running an operating system that was released any time in the past 5 years or so the chances are very high that you already have IPv6 on your system. What you may need to change is your DSL modem. Internode have tested modems that support IPv6 and IPv4 concurrently. More information is available at this address.

Thanks very much to Geoff for taking the time to chat to us!

IPv6 is available to all Internode ADSL2+ and NBN customers now. Visit http://www.internode.on.net/ipv6 for more information.

 

 
  - Nov 22, 2023, you can now follow development of Hiveware's built-in apps. Just go to top Hiveware domains, then find and click on (DEV). This will show you a pdf of and history of these projects development from a GUI perspective.

- June 15, 2021, Presented CableLabs with Hiveware Inc and Microsoft findings that their DOCSIS 3.1 gateway modem specifications have not led to ISP venders implementing IPv6 end point to end point Reachability. Local Reachability succeeds, but both Intra-ISP and Inter-ISP cable modem Reachability fail.

- Sept 15, 2020, Determined that ISPs that offer Ipv6 like Cox and Comcast, are not inter-connectable. See my explanation, which means Microsoft's socket library, Winsock2, is not to blame.

- May 18, 2020, Hiveware Ipv6-Ipv6/Ipv4-Ipv4 connectability succeeded Debug and Release. This breaks the stranglehold NAT has on Hiveware residential deployability (but only for intra-ISP comms for now, fx, XfinityWifi does not work where the problem lies with either Microsoft, Xfinity or Cox).

- March 17, 2020 opens Hiveware for Ipv4Ipv6Comms initial hive offering until June 19th, 2020.

- March 16, 2020, Hiveware for MyFiles private Digital Asset App Offering closed and March 17th, 2020, Hiveware for MyFiles public Digital Asset App Offering opens and will close again on June 19th, 2020.

- March 16, 2020, Hiveware BigBang Test 2-PC Basic succeeded again, but this time using Ipv6. This is the '1' of the decentralized '3-2-1 persistence' model.

- March 17, 2019, Hiveware for MyFiles public ICO began and ended June 16th, 2019

- December 17, 2019, Hiveware for MyFiles private Digital Asset App Offering began and closes March 16th, 2020.

- January 17, 2019, Hiveware BigBang Test 2-PC Basic succeeded. This is the '1' of the decentralized '3-2-1 persistence' model.

- October 1, 2018, Hiveware LittleBang preview running again, this time using production engine code

- August 17, 2018, Hiveware for MyFiles private ICO will begin

- July 17, 2018, Hiveware ICO ended. SoftCap not reached.

- Jun 3, 2018, first to file for Securities Act of 1933 compliance regarding HVW-generating dapp ownership ICO sale

- May 11, 2018, Microsoft delivers native MFC (C++) on ARM64, opening mobile devices and market up to Hiveware code

- April 17, 2018, Hiveware ICO began

- April 13, 2018, white paper published

- Dec 27, 2017, Hiveware engine (4th rewrite) POC done

 
 
  Site Map