Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[darius]

Simple suggestion: add "center on" option. So it would centre the gameview on that dwarf. If you need memory offsets:http://www.dwarffortresswiki.net/index.php/Memory_hacking/v0.28.181.40d16

[chmod]

Simple suggestion: add "center on" option. So it would centre the gameview on that dwarf. If you need memory offsets:http://www.dwarffortresswiki.net/index.php/Memory_hacking/v0.28.181.40d16

Not a bad idea, but first requires getting locations per dwarf, and then would require inputing commands to DF which is a precedent I'm not sure I'd like to set.

[darius]

Well yeah it does kinda violate your program tidiness (i wish my programs were that cool )

[chmod]

Well yeah it does kinda violate your program tidiness (i wish my programs were that cool )

Hehe, some of my programming rules:
 * Do this simplest thing that could possibly work
 * Don't repeat yourself
 * Debugging is harder than programming, and if you spend 100% of your brain power creating the code, then by definition you are not smart enough to debug it.

[njero]

Hmmm, 0.4.0 and 0.4.1 aren't working for me on XP. I'd open an issue and upload the .log files, but even with -trace on I'm not getting much. Pretty uninformative repetition steps too, I double-click the program, it windows thinks for a second, and then I get the classic "DwarfTherapist.exe has encountered a problem and needs to close.  We are sorry for the inconvenience."

Spoiler (click to show/hide)

[20:49:54.319] [Info] Dwarf Therapist 0.4.0 starting normally.
[20:49:54.319] [Info] MINIMUM LOG LEVEL SET TO: TRACE
[20:49:54.319] [trace] loading translations
[20:49:54.319] [trace] english translation loaded

Anything I can do to get you some more detail? I'm really bummed, your recent improvements have been fascinating!

[Need_More_War!]

Thats strange, I downloaded it just now and it works fine for me on XP

[Saber Cherry]

* Debugging is harder than programming, and if you spend 100% of your brain power creating the code, then by definition you are not smart enough to debug it.

That's a very interesting philosophy.  I have a similar problem - I created an algorithm that does something very efficiently, but I'm not smart enough to analyze it.  I can prove empirically, through brute force, that (for sufficiently small input sets) it is superior to the previous best published algorithm...  and I can show that for any given input of any size, it will outperform the prior best algorithm.  But at some fundamental level I don't really understand how it works well enough to even explain why it works, let alone bound its asymptotic performance.

It seems I was too clever while making it, and in the future I will heed your advice.

In dwarf terms, I dug too greedily...  and too deep.

[Taritus]

Off on a tangent, but what is your algorithm for?

[chmod]

Hmmm, 0.4.0 and 0.4.1 aren't working for me on XP. I'd open an issue and upload the .log files, but even with -trace on I'm not getting much. Pretty uninformative repetition steps too, I double-click the program, it windows thinks for a second, and then I get the classic "DwarfTherapist.exe has encountered a problem and needs to close.  We are sorry for the inconvenience."

Spoiler (click to show/hide)

[20:49:54.319] [Info] Dwarf Therapist 0.4.0 starting normally.
[20:49:54.319] [Info] MINIMUM LOG LEVEL SET TO: TRACE
[20:49:54.319] [trace] loading translations
[20:49:54.319] [trace] english translation loaded

Anything I can do to get you some more detail? I'm really bummed, your recent improvements have been fascinating!

Do you by chance have an AMD processor? I think this is a confirmed fixed bug. I got more aggressive with compiler optimization at the 4.0 point and set a requirement for SSE2 (which I thought all modern processors supported) but apparently Athlon's don't get it

It will be fixed in the forthcoming 0.4.2 if this is the case.

[chmod]

* Debugging is harder than programming, and if you spend 100% of your brain power creating the code, then by definition you are not smart enough to debug it.

That's a very interesting philosophy.  I have a similar problem - I created an algorithm that does something very efficiently, but I'm not smart enough to analyze it.  I can prove empirically, through brute force, that (for sufficiently small input sets) it is superior to the previous best published algorithm...  and I can show that for any given input of any size, it will outperform the prior best algorithm.  But at some fundamental level I don't really understand how it works well enough to even explain why it works, let alone bound its asymptotic performance.

It seems I was too clever while making it, and in the future I will heed your advice.

In dwarf terms, I dug too greedily...  and too deep.

Yea I want to know what your algo was for too

And that's not my advice. That was the ubiquitous and quotable Brain Kernighan (The "K" in the original K&R C)

Non-paraphrased:
"Everyone knows that debugging is twice as hard as writing a program in the first place. So if you're as clever as you can be when you write it, how will you ever debug it?"

Regarding your algorithm, you have either touched on a new transcendental truth, or you're not as smart as you think you are

[Impaler[WrG]]

Regarding your algorithm, you have either touched on a new transcendental truth, or you're not as smart as you think you are

The former though probably less likely should still be investigated, I mean penicillin WAS discovered on a moldy petri dish that was about to be thrown  away after all.  I'd still recommend you look into your algorithm further, perhaps recruiting some help if necessary.

and then would require inputing commands to DF which is a precedent I'm not sure I'd like to set.

How would this be precedent when the whole point of the program is to edit dwarf labor settings?  Is it because labor settings are just arrays of booleans and are easy to switch on/off and centering the view is different?

[chmod]

How would this be precedent when the whole point of the program is to edit dwarf labor settings?  Is it because labor settings are just arrays of booleans and are easy to switch on/off and centering the view is different?

I mean I don't want to end up writing functions that directly mess with the user interface of DF. Right now, yes, DT basically is a boolean-flipper, but it's main goal is a to be a decision support tool. Zooming itself isn't the problem really, but it's a "gateway" feature. Like pot. Start small, then we're strung out on bedroom designation macros, and a jump to next sock command. It's the avalanche of features that come after opening that floodgate that I'm scared of.

[Saber Cherry]

Off on a tangent, but what is your algorithm for?

Unidirectional ring leader election / maxima finding.  It's a problem commonly encountered in ring networks, but is adequately solved, so a better algorithm would have only theoretical significance.

I don't think I encountered any transcendental truths, but I did burn a lot of CPU cycles trying to find an upper bound by brute force (about 3 weeks x 2 cores).

Thanks for the encouragement on the algorithm...  perhaps I'll quit my DF habit long enough to give a proof another try.

[darius]

Then if no "center on" another suggestion:
My research shows that 576 bytes into the creature is his age (6 bytes or so) is his age. This would be a nice little thing to have in DTherapist

[chmod]

Then if no "center on" another suggestion:
My research shows that 576 bytes into the creature is his age (6 bytes or so) is his age. This would be a nice little thing to have in DTherapist

is that 576 decimal or hex? And are you referring to windows 40d16? I agree that could be useful to folks.