Thanduel
-
Posts
48 -
Joined
-
Last visited
-
Days Won
2
Posts posted by Thanduel
-
-
So unfortunately when I try to start a game I get a javascript error in the civ menu. I don't have time right now to roll back a couple revisions I will have another go tomorrow.
Type error y.name undefined from gui/common/functions_utility.js line 95
Very exciting way to end the day. : /
-
Well, sadly the application doesn't handle memory very well itself and crashes when I try to profile after loading a map.
If you have a 64-bit OS with more than 4GB of ram, you could perhaps post profile information of an actual game in progress?
Will do later this evening when I get home.
-
1
-
-
Awesome post, you got to doing this before I could!!! The scary thing is that this is before a game has even started.
-
I've noticed in a couple of discussions here that most people are not convinced about memory fragmentation being a significant problem. It may be a good idea to take some fragmentation measurements. I know that in VC++ you can use __heapwalk to walk through the heap and get data on the free/used blocks of memory along with the size of the blocks.
http://msdn.microsoft.com/en-us/library/h0c183dk%28v=vs.71%29.aspx
It might be interesting to do this a timed intervals to get an idea of the rate of fragmentation.
-
1
-
-
Pretty good video on optimisations http://vimeo.com/55639112
-
1
-
-
I appreciate the detailed response, you've evidently spent a lot of time trying to get around temporary object creation. I need to look at the disassembly of the calls to understand what is going on behind the scenes a little better, I'm actually very surprised that the stack copying isn't optimised out but then C++ is a new beast for me. Anyway thanks again for the detailed response.
The question now is what is the way forward with the C++ code base.
-
@redfox, haha I'm not sure I'd call move semantics a hack, I was impressed with the elegance of the solution, and my impression of the design decisions of WG21 were that they were very carefully thought out. In fact the papers are all exceptionally impressive quality.
Another interesting addition in C++11 for the STL is the emplace_back() function for appending to containers:
http://en.cppreferen...or/emplace_back
What this means is that you can construct objects in place without any copies or moving at all when appending to a container. There is an example that demonstrates this at the link above.
@tuan kuranes that is probably a good idea not only will the use of C++11 features add to the performance but it will make a lot of code much cleaner.
-
I really appreciate this discussion, some great resources have been referenced.
To add something into the fray. A lot may be gained by using the new features of C++11. In particular I have in mind rvalue references because they solve a lot of temporary object copying.
http://www.cprogramming.com/c++11/rvalue-references-and-move-semantics-in-c++11.html
-
Indeed I do think it would be silly to use fixed point over floating point to support a bunch of embedded processors representing a tiny fraction of the user base. I'm also well aware that all modern CPU's have FPUs even multiple FPUs integrated. However a large part of the reason for using a fixed point system is performance on any platform. Therefore obviously if we can beat the performance with a floating point system we should use it. What I'm saying though is the performance improvement you get is surprising and might be an indication that there is room for improvement in our fixed point system. There really shouldn't be such a gap even with a FPU.
-
It seems a bit heavy handed to change the entire engine to floating point without doing some significant testing across a number of machines to ascertain whether there is actually any benefit. As you say it is a non-trivial change. I have only given the code a cursory look but perhaps there is room for optimisation of the fixed point library? By the way I'm one of the those embedded devs that likes fixed point, well more like depends on fixed point.
-
I agree with most of what you have to say Fexor, but I must disagree with you about formations. Even in games like warcraft 3, there were bonuses based on abilties that units used. I think formations should be viewed in a similar manner. A group of units will be given a general bonus to armour or attack based on their formation. This is an intuitive side effect of formations. A testudo would for example increase the protection of a group of roman soldiers (a bonus percentage for armour across all attack types) but restrict their ability to fight (a reduction in attack power). This I think is obvious. However, to take away any benefit from using formations other then the natural benefits of having certain units meat shield for other units etc, is to make the feature redundant, that is, just a visual experience rather then something that is actually useful in game.
Personally I think formations as well as the ability to run/charge are features that add to the depth of gameplay in 0ad, as well as the excitement in a battle.
-
I have to say I agree with Spahbod, customisation is very important, pigeon-holing the game into a certain type of gameplay will only ostracise the game from many players and potential contributors.
Personally, I would agree with the game being too slow with too much time spent on the economy followed with fast battles that are not strategically satisfying. I think there is value in the battles being slower or lasting longer. The more time you have in battle the more time you have to think and react, to change your strategy or to respond to an opponents strategy, I've personally always loved the slow epic action of the total war games but obviously the total war franchise is a bit on the extreme side of slow and strategic. I think the formation feature implies gameplay more on this side of the RTS spectrum. I also think that this sort of gameplay puts less emphasis on APM and more on planning and strategy.
What I think would be great to see is the provision of a number of configurable options, say for example a global modifier on building hp, a global modifier on unit hp, a modifier on the rate of building/collecting, and a modifier on the overall gameplay speed. With these the game will likely settle into a popular subset of configurable options. These could then become game-play presets for different game types. However, the only way to find out what works is play to test the options.
A further advantage of having the ability to set health modifiers, is that it provides a way to handicap more experienced players playing against less experienced players.
I think what is clear when reading just these few posts is that choosing a set in stone game-play style for 0ad is going to leave many people dissatisfied.
-
If you want to learn more about JPS then this is a good source http://harablog.word...p-point-search/ . There is a link to a paper on that site as well.
Thanks that is a great link.
-
Thanks, the PDF looks pretty good, with a few holes like a full explanation of the JPS* algorithm. But great that gives me a much better picture of what needs to be done. Indeed I will keep you posted, I can't promise I'll have time but it does look like time will be opening up.
-
I would like to contribute to improving the performance of 0ad, I've been following the game for about two years now but I've never had time. However, I may have some time opening up soon. I'm interested in working on the pathfinder however I expect there is quite a lot to become familiar with before anything meaningful can be contributed. What has been done and what still needs to be done. I see there is a patch on trac, is all the work in the patch documented in Ykrosh's forum posts?
-
I agree that this is a postive change, wherever possible a "wall of text" situation should be avoided. Especially here for tooltips where you will likely need information quickly.
-
Decided to test, I'm running latest svn on kubuntu 12.04, the latest stable nvidia 304 series driver off the xswat ppa with a gtx 460. The issue described above does not appear on my machine using GLSL and associated settings with quality at 10 in Iberian sandbox.
The only bug here is that I now spend too much time distracted staring at pretty graphics and not enough time conquering my enemies
awesome work! -
This may be of interest for 0AD's OGL implementation to replace S3TC compression:
-
I haven't ever used QT on Mac however as far as I know the framework supports Mac. In fact that is one of the frameworks selling points, that is, the fact that the framework supports windows, mac and linux. Their IDE is also the best opensource C/C++ IDE I've used.
I'm not supporting it's use it would be stupid to rewrite anything, I'm just commenting on QT.
-
IMO, I think it needs more orange... First one will always be the best one in my opinion.
I agree with you, the first one looks amazing! In my opinion it isn't that unrealistic, in many places the grass does start to go brown or at least a lighter shade of green especially when you get closer to winter. For example:
http://cranemania.wi...log-may-09.jpg
That is a picture of the South African Highveld which is considered to be a temperate climate.
-
The new version is looking great I especially like the mutliplayer layout great work!
-
You note that the treasure taken from the temple was used to fund the empire. The most noteable thing that the money from campaign was used to fun was the construction of the colloseum. Allong with the money around 12000 jewish slaves were also taken and these were also used for the construction.
[RESOLVED] Javascript Errors In Matches Screen
in Bug reports
Posted
Hi
At latest revision I get
ERROR: SyntaxError: JSON.parse: parsing JSON data in civs/theb.json
ERROR: JavaScript error: gui/common/functions_utility.js line 95
TypeError: y.name is undefined
sortNameIgnoreCase([object Object],[object Object])@gui/common/functions_utility.js:95
initCivNameList()@gui/gamesetup/gamesetup.js:492
initMain()@gui/gamesetup/gamesetup.js:106
onTick()@gui/gamesetup/gamesetup.js:622
__eventhandler28 (tick)([object Object])@setupWindow tick:0
When I enter the single player matches screen.