Mirage Development Diary

Quoting <rsalsman.RemoveThis@gmail.com>:
>On Jan 22, 12:45 am, Krice <pau....RemoveThis@mbnet.fi> wrote:
>>On 22 tammi, 00:59, rsals....RemoveThis@gmail.com wrote:
>>>I'd be curious to know what the "signs" are

The main sign of your roguelike being vapourware is being called "Krice".

>absolutely nothing constructive to say. If you truly want to become
>"brothers" then start acting like one.

"Men who say all men are brothers don't have brothers".

Don't listen to Krice - no-one else does - but the idea of resisting the
feature list that swells without any corresponding swelling in the code is
a valid one.
--
David Damerell <damerell.RemoveThis@chiark.greenend.org.uk> flcl?
Today is Saturday, January - a weekend.
 >> Stay informed about: Mirage Development Diary 

> Don't listen to Krice - no-one else does - but the idea of resisting the
> feature list that swells without any corresponding swelling in the code is
> a valid one.

Thanks for the tip. As far as feature creep goes I don't think I am in
much trouble with that yet. I don't have any solid design document or
feature list yet. I'm really just working on a basic engine that I'll
eventually be able to add to. My feature list is bare bones at the
moment, and right now my focus is map generation.
 >> Stay informed about: Mirage Development Diary 

On 23 Jan, 08:30, David Damerell <damer....DeleteThis@chiark.greenend.org.uk>
wrote:
> I'm not sure that slow projectiles and drifting clouds can't be done in a
> turn-based game...

Of course it can; it just depends on how you divide the time (i.e.
ticks vs. turns). However, will the player notice? Does the player get
to move often enough to actually be able to see and dodge the
projectiles as they travel? OTOH; if they can be dodged, will it
require the player to move around in such a micro-timescale even when
nothing interesting is going on? And most important: will it add to
the fun, or just be confusing with a lot of random projectiles flying
around, where a simple die roll would suffice?

I have some idea for clouds though that I might experiment with in
TSL. I'll see how it turns out.

-the ru
 >> Stay informed about: Mirage Development Diary 

> Well, the other thing I'd say is - if you're going to write an interesting
> series of articles about roguelike development (leaving aside the question
> of whether what you intend is a roguelike), don't post a pointer to a
> Website. Post the articles!

I wasn't entirely sure what the protocol was for posting articles in
this newsgroup. From looking in the past I had seen a sort of split
between posts for specific articles, and posts for just the site (with
links to featured articles). Should I post a separate topic for each
article?
 >> Stay informed about: Mirage Development Diary 

Quoting <rsalsman.TakeThisOut@gmail.com>:
>Thanks for the tip.

Well, the other thing I'd say is - if you're going to write an interesting
series of articles about roguelike development (leaving aside the question
of whether what you intend is a roguelike), don't post a pointer to a
Website. Post the articles!

[Maybe you are going to. Maybe you have started and I'm about to be
embarassed.]

I'm not sure that slow projectiles and drifting clouds can't be done in a
turn-based game...
--
David Damerell <damerell.TakeThisOut@chiark.greenend.org.uk> flcl?
Today is Sunday, January - a weekend.
 >> Stay informed about: Mirage Development Diary 

Quoting the ru <ulf.astrom.DeleteThis@gmail.com>:
>On 23 Jan, 08:30, David Damerell <damer....DeleteThis@chiark.greenend.org.uk>
>>I'm not sure that slow projectiles and drifting clouds can't be done in a
>>turn-based game...
>Of course it can; it just depends on how you divide the time (i.e.
>ticks vs. turns). However, will the player notice? Does the player get
>to move often enough to actually be able to see and dodge the
>projectiles as they travel?

With a given ratio of player speed to projectile speed, it's _easier_ in a
non-real-time game; the realtime one has reflexes to slow down the
response.

>nothing interesting is going on? And most important: will it add to
>the fun, or just be confusing with a lot of random projectiles flying
>around, where a simple die roll would suffice?

A simple dieroll finds it hard to represent deciding to move somewhere you
wouldn't otherwise go.

I did suggest (for DoomRL) that projectiles would sometimes be fired at
your previous location, giving some dodging from lateral movement without
needing slow projectiles at all. I know movement has some dodging effect,
but I don't think that make the cut.

If there's a lot of projectiles to be dodged, making the game realtime
sounds more like an arcade shoot-em-up than a roguelike...
--
David Damerell <damerell.DeleteThis@chiark.greenend.org.uk> flcl?
Today is Sunday, January - a weekend.
 >> Stay informed about: Mirage Development Diary 

David Damerell a écrit :
> I'm not sure that slow projectiles and drifting clouds can't be done in a
> turn-based game...

Dwarf Fortress does it. For clouds and even all projectiles if I'm not
wrong.
 >> Stay informed about: Mirage Development Diary 

Quoting <rsalsman DeleteThis @gmail.com>:
>I wasn't entirely sure what the protocol was for posting articles in
>this newsgroup. From looking in the past I had seen a sort of split
>between posts for specific articles, and posts for just the site (with
>links to featured articles). Should I post a separate topic for each
>article?

You should _definitely not_ post a link to a Web page when you can post an
article for commentary.

I think it might depend. If each article gets a slew of comments, as
Andrew Doull's UnAngband design stuff does sometimes, they might be worth
being threads. If not - make each subsequent one a followup to the first
one. If there's a real consensus that it's not a roguelike, not just Krice
being a dick, give up (but I think frex the real-time relative to
tick-based RLs discussion has not been devoid of interest (then again I
would think that, I've been in it ).
--
David Damerell <damerell DeleteThis @chiark.greenend.org.uk> flcl?
Today is Sunday, January - a weekend.
 >> Stay informed about: Mirage Development Diary 

You should ignore the Krice. It is a small, annoying subspecies of
troll, and unless encountered in packs, not dangerous to those above
first level.

That said, I can explain what you did wrong to attract its attention.
A common problem that roguelike developers (myself included) tend to
fall into is that we are often too ambitious.

In our desire to create something of amazing scope and breadth, we
get caught up in the details of making our code perfect, or our
repertoire of behaviors vast. We redesign from scratch when we
run into a problem or have a better idea, or spend hours and days
and weeks working on things that do not contribute to gameplay.
When this happens, the actual games we are making tend never to
be finished.

I'm comfortable developing such a "programmers' roguelike" - I may
eventually release it, or I may not; that's not too important to me.
To use a martial-arts metaphor, I'm doing katas here, not fighting
tournament bouts nor trying to defend my home or family. I will
release it when I am *proud* of it, and not before.

But for those who want to release something, the most useful advice
to newbies (especially to newbies at programming, whose project-
management or code organization skills may be deficient) is to get
something simple and playable as fast as possible and *then* worry
about adding features.

The specific examples you mentioned (clouds moving across the FOV
in realtime, trees and other plants growing with the seasons, etc)
are nearly canonical examples of "frills" -- that is, fancy and
complicated code with no direct effect on gameplay. There is nothing
wrong with frills really, but you can spend a lifetime working on
frills and not have a playable game. Krices see frills as a sign of
weakness and attack any developer who mentions frills among their
objectives or design points. Their most effective attack is flinging
poop. There's nothing dangerous about the troll-poop; it's annoying
and smelly, sure, but if you're big enough not to drown in it and strong
enough that it doesn't encumber your footing, just ignore it and ignore
the little trolls that fling it.

You make your decision as to what your game will include, and you use
your project management and code organization skills to create the
machinery that makes it happen. If you have a little extra slimy
trollshit left over, then you can always use it to lubricate some
gears.

Bear
 >> Stay informed about: Mirage Development Diary 

On 26 tammi, 21:44, Ray Dillinger <b... DeleteThis @sonic.net> wrote:
> You should ignore the Krice.

So, Ray, did you ever produce anything else than long
posts where you attack personally someone in embarrassing
way, failing to be even remotely funny?
 >> Stay informed about: Mirage Development Diary 

On 26 jan, 20:44, Ray Dillinger <b....RemoveThis@sonic.net> wrote:
> You should ignore the Krice.  It is a small, annoying subspecies of
> troll, and unless encountered in packs, not dangerous to those above
> first level.
>
> That said, I can explain what you did wrong to attract its attention.
> A common problem that roguelike developers (myself included) tend to
> fall into is that we are often too ambitious.  
>
> In our desire to create something of amazing scope and breadth, we
> get caught up in the details of making our code perfect, or our
> repertoire of behaviors vast.  We redesign from scratch when we
> run into a problem or have a better idea, or spend hours and days
> and weeks working on things that do not contribute to gameplay.
> When this happens, the actual games we are making tend never to
> be finished.  
>
> I'm comfortable developing such a "programmers' roguelike" - I may
> eventually release it, or I may not; that's not too important to me.
> To use a martial-arts metaphor, I'm doing katas here, not fighting
> tournament bouts nor trying to defend my home or family.  I will
> release it when I am *proud* of it, and not before.
>
> But for those who want to release something, the most useful advice
> to newbies (especially to newbies at programming, whose project-
> management or code organization skills may be deficient) is to get
> something simple and playable as fast as possible and *then* worry
> about adding features.
>
> The specific examples you mentioned (clouds moving across the FOV
> in realtime, trees and other plants growing with the seasons, etc)
> are nearly canonical examples of "frills" -- that is, fancy and
> complicated code with no direct effect on gameplay.  There is nothing
> wrong with frills really, but you can spend a lifetime working on
> frills and not have a playable game.  Krices see frills as a sign of
> weakness and attack any developer who mentions frills among their
> objectives or design points.  Their most effective attack is flinging
> poop.  There's nothing dangerous about the troll-poop; it's annoying
> and smelly, sure, but if you're big enough not to drown in it and strong
> enough that it doesn't encumber your footing, just ignore it and ignore
> the little trolls that fling it.  
>
> You make your decision as to what your game will include, and you use
> your project management and code organization skills to create the
> machinery that makes it happen.  If you have a little extra slimy
> trollshit left over, then you can always use it to lubricate some
> gears.
>
>                                 Bear

Really a great post Ray, thanks for the advices, I can see in my own
development that I should focus on more important stuff.

> You should ignore the Krice.

I'll follow this advice too from now on
 >> Stay informed about: Mirage Development Diary