Josh

We'll finish the game, funding or not ;-) (it might take awhile though)

I found a nice comparison of two appropriately named images, Eve and then Morning.

You might want to think about choosing a different ticket, that ticket is a wolf in sheep's clothing ;-). Its prerequisites would involve working with a bunch of different files in XML, C++, and JS. I'll write a more technical comment on the ticket itself. Also, 20GB should be more then enough. My SVN checkout is only 6.9GBs right now. If you haven't downloaded the codebase/setup stuff yet this page on the wiki should be helpful. As a little bit of a side note, if you have used/use linux, 0 A.D. can be a little easier to setup there and happily compiles with GCC. Edit: With review the ticket is really two separate issues, one is adding a rating checkbox to the game setup page and the other is making the box automatically configure itself. The first issue has a simple resolution while the second does not. In light of that I'd encourage you to try your hand at the first issue which should only involve some simple GUI XML and JavaScript changes. If you have any questions, please drop by our IRC channel on quakenet.

It's an interesting idea and I would be fascinated to play a mod implementing it.

Actually, that's my issue. I accidentally committed some experimental/broken code which should be fixed now.

I actually rather like your first version. Since this is eye candy and won't be seen very often, the poly count isn't really important.

Should be fixed now, thanks for the report.

Wow! May I say, wheat and barley look massively better.

That's the current idle button in SVN. As to the new fields, I really like them. They look so much more - organic.

This discussion really is a back and forth with little headway on either side, but I'll throw in my thoughts. So, lets reason from a purely historical context. Now, farms were not truly infinite par-se but that was the effective result in the short term historically. Did sieges where the enemy became starved happen? Yes, but why? They starved not because their farms weren't infinite, but because their farms were outside their defences and the enemy destroyed them. If they had payed to defend them they wouldn't have starved. This same metric already applies in 0 A.D. If you want to pay the significant cost to wall in and defend your farms they will be secure, but otherwise they could be destroyed and your civilization would effectively starve. This has happened to me. I don't think any changes need to be made to make it more accurate.

Opps, after a little research it does look like scripting engines are allowed. Thanks for the correction.

It would probably be a good idea to mention at this point that Apple doesn't allow scripting engines into IOS apps and 0AD requires a javascript engine. EDIT: See next post

I used gperftools with runtime CPU profiler linking.

Well, since there's a thread now, here's a bunch of my profiling from before I re-did selection to be fast (GetEntitiesAtPoint): https://googledrive.com/host/0BwIjnJSd7iThRXRiN0ExcVVaeDQ. #1 and #2 are full call-graphs from just sitting in the lobby. #3 and #4 are full call-graphs from a few minutes vs the AI. #5 is a full call-graph of a complete game vs the AI (about an hour). #6 is an extremely detailed (1000s of nodes) version of the previous. #7 focuses on GetEntitiesAtPoint from the previous. #10 is a detailed full call-graph of a few minutes in a debug copy. #8 focuses on GetEntitiesAtPoint from the previous. #9 is an extremely detailed version of the previous. #11 is an insanely detailed call-graph of a few minutes. Some of the SVGs are massively big and may require an external viewer.

I agree that we should just go ahead with the update. Logistically, this patch is way to big to just leave on trac for an extended period of time.

Could you be a little more specific as to your problem? Like are you loading a skirmish, scenario, or random map? Was the map included in the game? Was the map created with the same version?

Sorry that we haven't gotten these snippets published. Even though I'm no historian I'll try to see that all of them are used. Your contributions are definitely valued and I think the lack of criticism is not so much because of low interest but because few team members have the credentials or knowledge to criticize these.

Well, what makes things even more complicated in relation to the pathfinder is that Philip already has nearly finished a new pathfinder so starting afresh would be a waste. But because few of us can understand Philip's pathfinder and Philip seems to have little spare time progress has been very slow.

To put that in perspective, the typical commit rate was somewhere around 0-3 commits per day. You can track every 0AD change made on the timeline.

If we felt particularly crazy, we could use square root tables to eliminate the computational load at the expense of memory.

Unfortunately no. We're actually still working to get some of RedFox's work during his limited work prior to the fundraiser added to the game. Since the fundraiser wasn't quite as successful as we hoped it's going to be hard to find a developer who could work for only about a six month period yet integrate into the team and maintain their work in the future. Fortunately, open-source volunteer development is coming along pretty well with incremental performance improvements being worked on. Sorry I don't have better news.

AMD CodeXL is a nice Linux/Windows debugging tool.

I thought it might be helpful to step in here and mention that there isn't really any reason to setup your own lobby server besides having a private chatroom. The setup is rather complicated and was not intended to be done by an ordinary user.

Tomas, have you had time to work on this any further? Your work was looking very promising and we'd really like have more so we could start potentially adding these as loading backgrounds.

It might be a good idea to look at how the AI assess players to get a better idea of how to calculate scores. That would be because the AI is typically effective at judging who is the strongest player (and hence the player it attacks first).