Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - moody_dorf

Pages: [1]
1
DF Suggestions / [Suggestion] Designation selection by mouse key
« on: July 10, 2011, 11:54:24 am »
That means:
1) mouse key down - start tile selection on point where mouse cursor is;
2) mouse key up - end tile selection on point where mouse cursor is.

2
DF Suggestions / Multi-level designation selection.
« on: July 09, 2011, 09:25:25 am »
Because it is pain in the rear to make stairways without multi-level selection.

3
That is my hope since first playing of this game. I thought it would be added already. I have no idea why there are no priorities yet.   :(

Rather than react with knee-jerk sarcasm, you might try to explain what you mean.

Do you want to have options to lower/raise the priority of different areas of designation, different types of designations, specific dwarf's job priorities?

There are, in fact, already priorities attached to jobs. You may not feel work is correctly prioritized, but you are incorrect in asserting priorities do not exist.

As for "building tasks," do you mean jobs in workshop queues, the priority level for the "construct building task" or [ b ] menu priorities, generally? Care to add some content to your suggestion?
Sorry, I though the man did not understand meaning of priority word, no sarcasm intended.
I added some explanations to the first post. I beg your pardon, I had thought the topic subject is explaining enough.

5
I suggest to:
1. Add option to select priority, when making designation or building task.
2. Make designations and building tasks scheduled according to their priorities.

That is my hope since first playing of this game. I thought it would be added already. I have no idea why there are no priorities yet.   :(

Designation priorities will greatly decrease user's micromanagement and will allow to avoid cave-ins and other related problems.
As a programmer, I can say that designation priorities are easy to implement, and it will not lead to performance penalty.
It will take 2-3 hours to accomplish.
It is actually much easier than begging Toady to implement this.

http://www.bay12forums.com/smf/index.php?topic=58995.0 - here are some illustrations (thanks Vattic for link)

6
DF Suggestions / Re: World generation in not main thread
« on: July 02, 2011, 11:58:05 pm »

Message processing is API issue, Boost is C++ class library, Toady know what these things are. (I hope so)

In current version interruption isn't instant (it is noticeable for large worlds).

The suggested progressbar is to show yearly simulation progress.
Civilization simulation is longest part of world building.
Civilization simulation is processed by years.
In current version, the game window is updated after each year processing

It's reasonable to ask me to use better language, no offence is there. And I assume programming suggestions make no offence for a programmer.

7
DF Suggestions / Re: World generation in not main thread
« on: July 02, 2011, 11:27:18 pm »
Here. I estimate it will take about 10 minutes to read, and will greatly boost the clarity of your posts.

Are meta-suggestions allowed? O.o
Sorry, I am not a native English speaker.
Please point where my speech is hard to understand, I will try to rephrase.

8
DF Suggestions / World generation in not main thread
« on: July 02, 2011, 11:12:18 pm »
It will allow window to process messages (so no hanging while generating large worlds).

Instant interruption (because of instant key presses processing since no need to wait for end of world generation loop).

Then some progressbar about yearly civilization simulation is possible. In large worlds it might take a minute for each year.

It will take not much time to implement it, if you use boost. I estimate 2-4 hours.

Also, Toady, you will earn more money I believe
(users will donate more money, because of increased comfort of program)

9
DF Suggestions / Re: Toady, consider using OpenMP!
« on: July 02, 2011, 10:49:36 pm »
Multithreading an application that is being actively developed is generally a bad idea.
Well this is a publically released program, so some elementary optimisations is good idea (if not nessesity) in this case.
Also, multi-threaded optimisations are better to be made as soon as possible, to prevent larger code refactoring in future.

10
DF Suggestions / Re: Embark search improvements
« on: July 02, 2011, 10:38:59 pm »
To Toady
Embark search is very awkward. There's my suggestions how to impove it:
1. Make embark search interruptable [it's easy!]
2. Make it writing list of found locations [it's about 15-30 minutes of programming i suppose]
3. Player can look through found locations, while search is working [you'll have to make a thread, it isn't difficult]


And please no silly comments from not developers.

No.1 is implemented, Thanks, Toady!

But no. 2, 3 are still not implemented. But I believe it's still good idea to implement them.

I hope Toady will notice this suggestion

11
DF Suggestions / Toady, consider using OpenMP!
« on: July 02, 2011, 10:31:23 pm »
Greetings, Toady.

OpenMP allows to utilize multiprocessing on multi-thread processors (the multicore and/or hyperthreading ones).

It is easy to learn and will not make need of heavy code refactoring (if you use VC++ compiler). No need for manual thread management. Threads are managed by compiler and/or automatically at run-time.
http://en.wikipedia.org/wiki/OpenMP

For instance, it might almost 6 times increase Dwarf Fortress performance on my AMD Phenom 6-core processor.

Here is some code example (taken from Wikipedia).  #pragma omp ... - OpenMP statements
Code: [Select]
#define CHUNKSIZE 1 /*defines the chunk size as 1 contiguous iteration*/
 /*forks off the threads*/
 #pragma omp parallel private(j,k)
 {
  /*Starts the work sharing construct*/
  #pragma omp for schedule(dynamic, CHUNKSIZE)
  for(i = 2; i <= N-1; i++)
     for(j = 2; j <= i; j++)
        for(k = 1; k <= M; k++)
           b[i][j] +=   a[i-1][j]/k + a[i+1][j]/k;
 }

Here take a look on some writings about parallel programming that might help you:
http://software.intel.com/en-us/articles/getting-started-with-parallel-programming-for-multi-core/
http://software.intel.com/en-us/articles/the-secrets-of-parallel-pathfinding-on-modern-computer-hardware/ - A* pathfinding!

You can find more writings at
http://search.intel.com/Default.aspx?q=parallel+programming&submit=+&c=en_US&results=10&offset=0&categories=&filetypes=&q2=&q3=&q4=&method=edit


Thanks for reading, Toady!

12
DF Suggestions / Embark search improvements
« on: August 18, 2010, 11:01:50 pm »
To Toady
Embark search is very awkward. There's my suggestions how to impove it:
1. Make embark search interruptable [it's easy!]
2. Make it writing list of found locations [it's about 15-30 minutes of programming i suppose]
3. Player can look through found locations, while search is working [you'll have to make a thread, it isn't difficult]


And please no silly comments from not developers.

13
DF Suggestions / Add option to clearly dig, leaving no stones
« on: August 18, 2010, 03:26:21 pm »
It will be useful for stockpiles, especially for the large ones.
Since less experienced miners leave less stones than more experienced ones, I think it will be rational if every miner could leave as many stones as unexperienced one, or no stones at all, if he's ordered to do so.

If Toady will add this feature, then there also must be a way to choose which stones must be omitted, and which ones should be leaved (for instance, precious gems and metal ores might be set to be leaved).

Pages: [1]