Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 11 to 20 of 24

Thread: Missing programmers or did something happen?

  1. #11
    Quote Originally Posted by lenlrx View Post
    The current API looks pretty good, compare to few month ago.
    My project is to make a strong SF solo bot,i hope it can be finished
    Good to have you back! Having a strong bot for 1v1 is a good idea!

    API definitely is a lot better now, and I appreciate what Chris has done for us. I was mostly talking about more advanced modules for creating a strong custom bot framework (which is what I'm trying to do).

  2. #12
    Basic Member
    Join Date
    Feb 2014
    Posts
    27
    Quote Originally Posted by Platinum_dota2 View Post
    I think AdamQQQ is still active, but yeah, I think most people got bored. Writing good bots is a lot of effort with almost no reward except maybe learning Lua and scripting, but that's about it. It is really fun at the beginning, but eventually, the excitement mostly goes away!

    We were promised bot vs bot matches around 4-5 months ago (which would've been fun to have), but there is no sign of it as of yet. I thought eventually we'll have sponsored bot tournaments (like Starcraft) but that didn't happen either. We didn't even get something as small as the TI's short film/cosplay contests from Valve, even though making good bots requires 10 times more talent and dedication. There are also some other issues I've bitched about before, but yeah, I get why people are not as passionate as they used to be!
    I agree! I'm always lazy to edit bot talent lines when an in-game upgrade comes along and modifies the hero's talent tree. I imagine how tiring it must be for those who edit all the commands from the beginning.
    And I did not know there were bot tournaments on Starcraft. Does it happen in Starcraft 2?

  3. #13
    Basic Member
    Join Date
    Feb 2014
    Posts
    27
    Quote Originally Posted by Yavimaya View Post
    yea i think it has slowed down because most people have most major elements in game already, valves involvment has slowed down to less than snails pace since they started work on the compendium and i think the license issues that come up seem to have been the nail in the coffin.
    It may be that this is happening too ... For me it is being very difficult because on my team there are always people with behavioral behaviors, so I look forward to updates on these bots.

  4. #14
    Basic Member
    Join Date
    Feb 2014
    Posts
    27
    Quote Originally Posted by DreadedGhoul575 View Post
    I know, it's a real shame. I wish Bot Experiment was updated more frequently, since I tend to use that bot a lot, since it has all the heroes (more variety).
    Me too!!! After this script of this bot was added to Valve's servers, I thought they would have more frequent updates. All we have to do is wait ...

  5. #15
    Basic Member
    Join Date
    Feb 2014
    Posts
    27
    Quote Originally Posted by lenlrx View Post
    The current API looks pretty good, compare to few month ago.
    My project is to make a strong SF solo bot,i hope it can be finished
    Are you the one who programs the 1v1 script in mid against a SF that has in the workshop?

  6. #16
    Basic Member
    Join Date
    Jan 2014
    Posts
    49
    In old days (dota1) instead of several sole developers, there were just one team that worked on AI. so each people focus on sth. but now each developer should write everything (from buying stuff to upgrading, roaming, ganking, last hitting etc.)

    This means LOT's of work. Its quite natural that After some time, many of these developers will lose interest and focus on sth else.

    I know that valve spend LOTs of money on events and esports (even cosplay!) but I think its time to spend a little on AI to.

    You know the current pool on International is 20M+ each year. what if Valve spend 5 percent on these money on a side AI vs AI tournaments in International.

    Or give this money to any AI that could beat (or at least destroy a rax or even a tower!!) the champion of the tournament. (AI vs Human)

    Or even design Just an "Aegis of AI" and give it to best AI of year (Little expense for Valve!!)

    In this fashion there could be chances that AI developers (or even fund raisers!) spend more time on AI.
    Last edited by fandogh; 07-12-2017 at 12:36 AM.

  7. #17
    Basic Member
    Join Date
    May 2016
    Posts
    36
    i totally agree with @fandogh

  8. #18
    Basic Member
    Join Date
    Dec 2016
    Posts
    47
    Quote Originally Posted by vitor_stwr View Post
    Are you the one who programs the 1v1 script in mid against a SF that has in the workshop?
    it should not be me,i just uploaded 4 duplicated bots due to no response of the workshop tool.but that's long long ago.

  9. #19
    Basic Member
    Join Date
    Dec 2016
    Posts
    120
    Quote Originally Posted by fandogh View Post
    In old days (dota1) instead of several sole developers, there were just one team that worked on AI. so each people focus on sth. but now each developer should write everything (from buying stuff to upgrading, roaming, ganking, last hitting etc.)

    This means LOT's of work. Its quite natural that After some time, many of these developers will lose interest and focus on sth else.

    I know that valve spend LOTs of money on events and esports (even cosplay!) but I think its time to spend a little on AI to.

    You know the current pool on International is 20M+ each year. what if Valve spend 5 percent on these money on a side AI vs AI tournaments in International.

    Or give this money to any AI that could beat (or at least destroy a rax or even a tower!!) the champion of the tournament. (AI vs Human)

    Or even design Just an "Aegis of AI" and give it to best AI of year (Little expense for Valve!!)

    In this fashion there could be chances that AI developers (or even fund raisers!) spend more time on AI.
    AGREED

    you mentioned a good point
    we need to work on many facts like last hitting. i started this when bot scripting was provided by vale and i finished this part just a month ago!!!!. as you said, with some minor or major financial incentives AI scripting will be improved but here there are many basic issues that most of developers challenge with them like script debugging, handle fatal errors in game( I didnt find suitable thing for this in dota 2 ) and etc.

  10. #20
    Basic Member
    Join Date
    Nov 2015
    Posts
    108
    Life just got in the way for me; new job, finished my current degree program and moved across country.

    I was hoping to be back in it sooner but the main thing I wanted to focus on was making plug in modules for other bot coders to do necessary things that can be done stand alone like dodging AOE, last hitting, support pulling/stacking, etc. My lvl 1 rosh script is successful with almost any 5 heroes (not that it's worth it usually but I had to write it to prove it to myself, I wanted to see Meepo with solo XP from lvl 1 rosh lol). However the API doesn't really have great tools to do things that way and the bot framework itself doesn't seem to be getting finished. There just isn't enough people around to believe that modules like that would get used.

    I tried at first to get a community code base going like was said about dota 1, but so did a half dozen other people and everyone went in different directions so fast as solos or small groups. I just lost interest. I still check the forums weekly though hoping this project will get some attention from valve and just suddenly be amazing, cause there is always a chance for that.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •