DoW2live: What do you use as a guideline for what units are included in the game?
Egon Spengler: Looking at existing races, we looked at combination of "is this race missing something?" - in the case of Eldar, they were missing some heavy ranged firepower, especially against vehicles in T2, so that helped decide on the Wraithguard unit. With Orks, units already had distinct roles fleshed out, so we picked something that was fluffy and fun, which is why we went with the Weirdboy.

DL: People complain that Razorbacks and Wartrukks are limited in usefulness - has anything been changed with pathfinding, and why doesn't Chaos have a transport vehicle?
ES: We didn't want to hear any METAL BAWKSES jokes. At all. As far as the Razorback goes, damage has been buffed to make it more useful. We're always looking to improve pathfinding, but a lot of it comes down to the size of the map and the pass-throughs on it, and often pathfinding complaints are best addressed via map-edits.

DL: Do you use win/loss stats to look out for OP commanders (such as RA w/tunnels)? ATM only Eldar have 3 viable commanders - other races tend to be locked to a single commander to be competitive.
ES: Absolutely. One of the problems we have is that in team games, stats are fairly close to 50/50 W/L, particularly in 3v3. A lot of the imbalances (in particular with RA) only become egrigious in 1v1. Changing things for a 1v1 player can have a negative effect for team players, so we need to be careful with what we do. With the RA, it's not just his tunnel - Corrosive Devourer has recieved a nerf to help bring him in line, because he is clearly winning more than he should.

DL: Can Wraithguard and Brightlances garrison buildings?
ES: The BL can - all platforms can - and WG cannot, since they're too big. They can't go into webways either.

DL: Is MS Certification required on small balance patches that don't have any other changes?
ES: MS Cert isn't required for everything we change - if we just change the data, such as anything in the attributes, we don't need to go through MS, we can do that on our own, which we have in the past. With code changes, if they're limited in scope we have a much faster cert process for that. If you look at the number of patches/updates we've released over the last year, MS Cert isn't slowing us down any. We've had more DoW2 patches than the total number of patches for all of the previous DoW games. Cert doesn't get in our way in terms of quantity, but it means we have to bundle up our fixes to go out in a certain way - if it requires 1 line of code, it has to go through MS, but if it's 1000 lines of data, we can do it ourselves.

DL: Can we expect more frequent balance updates that only make very small changes in order to fine-tune units?
ES: We'd like to see most of our balance changes come down in size - it's unlikely that we'll do anything the size of TioW again. Whenever we release an expac, we're always going to see new units, and new units always cause new balance issues which need to be addressed with fairly significant updates. We'd like to see a similar frequency of patches that we had early last year, but with smaller changes.

DL: Even though FFA is unranked, will you release a matchmaking system for it to make it easier to play online?
ES: Currently it's custom game only - we'll see how popular it turns out to be. If we get a lot of people playing it, we may see improvements to the matchmaking and number of maps.

DL: Will you let us bind our own keys without having to use a mod?
ES: This will be possible with the first patch of Chaos Rising, which will be "some time" after release, in the form of an editable .txt file.

DL: Are there any plans to include/improve different region-based 'gateways' for automatch to prefer matching against local players?
ES: The problem with putting up a firm 'fence' system like that is that if there aren't enough players in a given area - and this will be true in the smaller regions - then matchmaking may never complete. We do prefer matching players with lower ping, but if you're searching for a game and the only candidates who happen to be on are far away, it will pick one of those over not choosing anyone at all and failing to automatch. However, we are looking at improving that algorithm in the future.