View Full Version : Ipv6
What do you think? Do you think BI should make OFP2 support ipv6?
IMHO it would be a good thing because there are more adresses and therefore you could for example host behind a Router which does NAT or ICS without having to forward ports as there are so many adressess that you don't need to NAT for ipv6.
Also I've heard that its routing is better meaning that the packets maybe would be transfered faster (=less lag)
Another pro is that it should not be hard to make ipv6 support.
A big con is that not many use ipv6 yet and it is hard to predict how long it will take until most people have it.
Aug 15 2003, 22:48
by reading your post I got an idea of what Ipv6 is....but i'm not totally sure....could you post a comprehensive explanations of what Ipv6 is? cause not everyone here is a geek or know everything about the new technologies out there http://forums.bistudio.com/oldsmileys/smile_o.gif
Yeah, no problem and I must admit that there are things which I don't know either.
A normal ipv4 adress has 32bits and looks like this:
An ipv6 adress has 128bits and looks like this:
theoretically like this but the above is a short form which you can use
So with ipv4 we have 4294967296 which seems like a lot but in fact 4 billion IP adresses is not enough and the world is running out of them so ipv6 was created with 3,4028236692093846346337460743177e+38 adresses which is quite a lot.
For example I myself have 79228162514264337593543950336 ipv6 ip adresses available for my personal use.
Because there are so many there is no need for nat and for example in my home network every linux or winxp comp has his own global ipv6 adress but only private 192.168.* local adresses with only a single global adress on the router.
Now, secondly the part where I don't know so much yet:
ipv6 also has some other improvements over ipv4 concerning the routing but I have forgotten what I read about this and anyway it is not important as long as you don't have native ipv6. At the moment 90% of all ipv6 connections are established through ipv6-over-ipv4 tunnels which makes them perform much worse than ipv4.
Aug 19 2003, 05:26
Here in the US there is not a big push to move on IPv6, because we still have a glut of IPv4, but overseas, especially asia, there is a serious dearth of open addresses. A lot of the unix IPv6 work was started and still runs out of Japan (Kame project) for example.
Because many if not most games are advertised by IP address, this is a function that 'may' need to be looked at. The real question is, will IPv6 have much if any end-user deployment by 2008? OFP-1 came out in 2000ish, and sequel OFP is looking for '04(?). So put the commercial 'life' at 3-4 years, and BIS will likely base tech designs around that.
Now, if you play strictly DirectPlay, you could take a DNS NAME like 'server.network.com', feed it into DirectX, and let DirectX deal with IPv4 or v6. If you do it your self... that's a whole different story.
With NAT's popularity, what is likely to happen is that local end-networks will continue as private IPv4, and backbones will migrate to full IPv6 routing, sometime.
This means, your best option as an end user is to use a server locator like gamespy or use full DNS names for your games. That's what you are doing already when you play on a Linux server.
Hehe, I think that BI has learned and won't use DirectPlay in OFP2. They did a good job on their own multiplayer code and I am quite sure that they will do again.
I also just had a look at ipv6-patches for some Unix programs and it really seems that you do not have to change much (I guess that doing ipv6 for windows software is not harder than for unix)
Quote[/b] ]Hehe, I think that BI has learned and won't use DirectPlay in OFP2.
I have programmed DirectPlay and I don't get the point of it. It's messy and impossible to get anything working behind a firewall or nat router without opening and directing ranges of ports.
Some classic TCP/IP socket programming is just as easy (easier), cleaner and the final client works behind a firewall or nat router.
However, I hope BIS changes the netcode engine from the current sync model to a client/server model for OFP2.
Powered by vBulletin® Version 4.2.0 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.