Jump to content

Chakakhan

WFG Programming Team
  • Posts

    235
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by Chakakhan

  1. This is a question to the dev leaders and probably of general interest to all new developers. I just finished work on a task and had a few questions about how I completed the it. In this case, I added some code to CInput to handle word delete keystrokes - this was the scope of the task request. Widget code is tricky stuff, so since I was already in there and I had added the logic for discovering word boundaries and was familiar with the code, I also added the code to handle moving around a word at a time. And since I was already in there and doing word stuff, I added the code to handle word selection via. mouse double clicks. Then I noticed that there was code for paste, but no code for copy and cut, so I added them as well. When I tested the cut/paste/copy stuff I discovered a really hard to find bug that made intra-application clipboard operations on Windows fail. So I fixed that bug. My question is, given the above scenario, how would you prefer me to handle such situations? Would you prefer that I would have stopped at some point and created new tasks? If so where? I did finally close the task (well changed it to review) and created a couple of new tasks for CInput because the patch was getting fairly large. Please let me know your preferences on such things. Cheers!
  2. Wow, very nice models!
  3. Makes perfect sense! I was just curious because yesterday on the dev chat they were talking about a code freeze coming up and I was just curious. I was thinking about picking something I could get done before the freeze, but decided it's probably better to work on whats most important and worry about the schedule a little later . Cheers!
  4. Curious what the future release schedule is?
  5. Yes, I keep hitting ESC, so many games use that for the menu popup. Is that one of the user defined keys? If not it probably needs to be.
  6. I created a patch that addresses issue number 788 The goal was to reduce the time between when a move command was issued and the unit responded. Previously all move commands were asynchronously queued up and held until the start of the next turn. Then they were all processed at the start of the next turn. This caused a noticeable delay and made the game feel unresponsive. The problem was compounded when moving formations of units the formation proxy movement was delayed until the next turn, then the unit movement based on that proxy movement ended up being delayed a couple of turns out. The solution was to continue to asynchronously queue up message, but at key points during the turn processing we will now process some of those movement request during the turn where they were queued up. This make the game feel more responsive. Formations feel a little better because we move the formation proxy and let the players re-path based on that movement all in the same turn. This does not increase overall performance and in some cases can make it slightly worse. It only affects the responsiveness, not performance. I added a threshold value that limits the number of "same turn moves" that can be processed during a turn which minimizes excessive same turn re-pathing in massive battles. It also allows us to tune this functionality. This threshold value is in .../data/mods/public/simulation/data/pathfinder.xml, the value is called MaxSameTurnMoves. The default value is 16 based on some testing I did on my slowest machine. If you set this value to 0 the same turn move functionality is completely bypassed. So if you are testing and want to make sure this feature is not involved in an issue, you can turn it off. The code still post all moves asynchronously so that the pathfinder code could potentially run in a separate thread. Thank you to Philip for spoon feeding me the solution. Cheers!
  7. Omri, thanks for the detailed explanation. I can clearly see what you are after now and understand your concerns. I hope you don't feel I was piling on. I am a musician and sometimes my opinions and passion get the best of me . Sebastian, I wasn't trying to distort what you wanted in the score, I was trying to throw in my opinion and ideas as well! The best of musicians are very opinionated, I like what both of you have done thus far - very professional. Sebastian, give it one more go. I understand where Omri is coming from but a little tension between folks on the creative team can lead to great results. Look at The Beatles. Cheers! PS - Omri, I noticed you are from Israel, you familiar with my buddy Amit Poznansky? According to him everyone knows him there!
  8. I just reread my wall of text and I think it sounded like I was ragging on Omri a bit. That was certainly not my intent, I think you have done a fantastic job on the tracks so far. I also thought Sebastians track was awesome and think with some mastering it would fit in with the existing music.
  9. Rob, how is this going? The bits and pieces you have so far are really beautiful. Keep up the good work, I wanna play this bad!
  10. Very nice man! I definitely prefer the first version though. The echo makes it sound like drums being played outdoors in a large canyon - nice effect. Reminded me of times I have sat outside in an area with a similar natural echo delay and tapped out beats. Awesome feeling being "in time with nature". The humming is a "drone" which probably was used quite a bit during that period. Similar to the drone pipes in a set of bag pipes - very Celtic. The recording is done very well and it is recorded at a very "hot" level (this is a good thing as hotter signal means less noise). This is why it seems loud when compared to the other tracks - at the same volume setting it is a bit louder and brighter. The existing tracks are not recorded as hot nor as bright and clear sounding. So the tracks all need to be "mastered" together to make them sound better as a whole. Mastering is the process of setting the levels, the EQ and reverb on the tracks so they play nicely with each other. I am a professional musician and composer and I must agree with Sebastian, I have no clue at this point what what is considered appropriate or believable. I certainly hope we are not going to try to reproduce period music, It would put everyone to sleep. Music has grown considerably in terms of complexity and variety. I think it best to write what ever sounds and feels good and throw in musical references, like Sebastian's use of Dorian mode and the use of the drone and penny whistle to give it the Celtic feel. Me personally I want John Williams dramatic scores that make the hair on the back of my neck stand up, not some naked native scratching on a gourd! I know the native scratching the gourd is more "period", but he smells funny and frankly his music is boring. Not all the time, but sometimes I want that big exciting Hollywood movie track sound. Omri, I think I understand part of what you are trying to do - the blending of live instruments with samples and I think it's a great idea. Just don't be too quick to "poo-poo" a sound just because it is sampled. A well played sample can blow away an average live performance. I have a friend in Israel, Amit Poznansky (www.poznansky.com). He is a great and very popular composer. He has done some amazing tracks where you really can't tell he is using samples. For example check out this: http://www.fablesoun..._CurtainUp!.mp3 Not one single live instrument is used and it sounds amazing to me! So amazing I went out and spent $4000 on that sample set - ouch!!! I love the tracks I have heard so far. Keep up the good work.
  11. Yea, but I love the way they walk! Whoever did those walking animations did a fantastic job! Mythos, I have been playing the latest version today against the AI and I also played a couple of games against my son. The biggest issue for me was hands down the movement delays. It takes nearly a second for a unit to start moving after you click on the destination - it doesn't seem to matter if you are moving a couple of tiles over unobstructed or if you are moving completely across the map. I compound the problem by clicking the destination multiple times and each time you get the one second delay. The pathing in formations is a bit painful. Too much motion and they want to stay exactly in formation until they reach the very end of their path. Probably need to figure out what feels right from a player perspective and figure out how to implement the solution. Philip had a fantastic thread in the forums where he basically blogs while is he is working through solving some of the pathing issues. It was very impressive, he obviously is very sharp. The pathing is probably the most complex part of the gameplay side of things and it is what gives the gamer the feel that he is actually in control or not. I absolutely hated the "Medieval Total War" series because it felt like my units were only taking my commands as suggestions. They were primarily concerned with keeping the formation even when the end of the formation wheeled around and ended up getting chewed off by the enemy as it passed through their lines. There has to be a balance when the formation makes sense and when it needs to be ignored. AoEo does a fantastic job of this. Starcraft 2 and AoEo are both fantastic at the whole pathing responsiveness thing. I loaded up AoM, AoE2 and AoE3 today to do some comparisons. There have been lots of improvements through the years! I would suggest you all give AoEo a try while it is in open beta (open beta registration closes Friday morning) just to get a feel for how responsive and obedient the units feel. It feels good, even in the worst cases such as when you are taking a huge army through a choke point. I looked at the network stuff and I don't think it will be hard to allow re-connections and add an observer mode. The move timing is a challenge, but not as challenging as the pathing . AI needs some work, but it seems like it is coming together pretty well. It seems to do a good job with the big picture stuff overall, just falls down on some of the detail stuff. I have read some forum threads where the developer is planning on improving the Jubot AI and he seems to have a good handle on things. It sounded like he knew where the problems were and had plans on addressing them. From a graphics engine / artwork perspective I think you have a playable game. There is plenty to clean up and improve upon, but they provide a more than adequate platform to play on. Kudos to the artwork team! Very few open source projects I've seen have this level of quality and consistency. This code looks to be in really good shape. Kudos to whoever picked JS over Lua as well! I saw someone was working on getting premake working with VS2010 - that would be nice. So gameplay is key at this point I think. Getting the pathing stuff fixed will make the most difference in gameplay at the moment. First and foremost getting rid of the initial delay will help a lot. Cheers, Kenny Long (AKA Chakakhan)
  12. Just wanted to take a second to say hello and introduce myself. My name is Kenny Long, I am a software developer by day and professional musician by night. I have been programming for about 30 years (yep I'm old - but I did start when I was 14!). I am currently doing .NET development, but spent years writing C/C++ code on Unix platforms. I specialize in massive online transaction processing apps for banks, wall street, etc. I worked on a few games including one commercial game Shadowbane actually made it to release (probably shouldn't have - wasn't ready). I wrote the network layer code and developed a lot of the NPC AI for that. I have periods where I play games pretty much non-stop then periods where I lay off for weeks. I guess I would consider myself an avid gamer. My favorite games are RTS - namely Starcraft and Age*. I am currently playing AoEo in the beta (it is open beta now if you are interested) and enjoying it. Seeing how far you all have come and the current state of the game has inspired me to try to pitch in and help out. I noticed two items on the task list that looked interesting - the network layer mods and the path finding delay problem. I may end up picking up a couple of the smaller ones first though Anyways, hello all and I look forward to more interactions in the future! Cheers, Kenny Long (AKA Chakakhan)
  13. Just wanted to comment on the fantastic new art! I looked at this project a while back and it has really come a long way. Very high quality stuff. You have inspired me to spend some time working on the bug list. Cheers, Kenny Long (AKA Chakakhan)
  14. For those of you who might be interested in checking it out, the beta will be open for signups until Friday morning. I have been in beta for two months and I like it so far. Might come up with some ideas to bring back to this great game! PS - I found out about this game from the forums over there.
  15. The beta is open right now if you want to try it! It is open through Friday morning. I've been playing the beta for the last 2 months. The game play is a lot like it was during the AoE2 era. They added some bits from AoE3 as well. I quite like it overall. The new graphics are crisp and clear and it is quite easy to differentiate between units and even between units that are equipped differently. The dramitic "over the top" animations allowed by the more cartoonish theme make it very obvious what a unit is doing. Every one of my buddies that I played AoE with complained about the game at first, but they have all come to really like it, even the graphics. I'm not a fan boy, just think there might be some interesting things here that might not be obvious on the surface. Take care all!
×
×
  • Create New...