So, what are you working on right now?

[Sotha]

Oh, and nbohr, about that screenie you showed from Knighton Manor.

 

Remember that the player can move in the game. Moving about makes the edge obvious.

[Tels]

Yep.. Now I will reveal why I do not make so high detail maps.

 

I confess I do not see any differences between the screenies here. All of them look really good to me, and none of the screenies is better in my eyes than the other.

 

Therefore, why bother to see trouble implementing a decal, which is invisible in the scene?

 

But yep, this is just me rambling. I'm prolly blind or something.

Keep up the good work!

 

The difference is definitely there, maybe your monitor is weak I have trouble spotting it on my LCD, tho I know where it is. It is however, a lot more clear on a CRT.

 

Btw, the difference is supposed to be subtle, if the player can't spot it, it is well executed. The difference only registers subconsciously and make the scene look "less flat". Fidcals rooms are always top-notch to begin with, so the effect is less than if you would apply this to say f.i. doom stock textures and doom stock walls in an room lit by a single light source, where you have large flat sections of a wall that are all colored flatly.

[Fidcal]

This shading also looks particularly good in ambient-only light which normally looks flat. It also increases the proportion of soft shadows to hard, cast shadows which is always good imo.

 

I'm thinking what Sotha means is not that he can't see any difference between the before and after shots (which incidentally are not the very best examples of this grime useage) but that he doesn't prefer one to the other so why go to all that trouble (and it is a lot of trouble I admit.)

 

Another point to consider is when used even if it is not so effective or even negative in some parts of a room to another but once you've started in one area you have to do all of it or at least certain complete areas or the contrast is obvious.

 

BTW, does anybody else use leaning left and right now and again to get better depth perception when there is doubt?

[Aprilsister]

Btw, the difference is supposed to be subtle, if the player can't spot it, it is well executed

 

This may well be true but as a player I will vote for Sotha's philosophy of concentrating on making the gameplay fun rather than tweaking things endlessly. TDM looks great to me. If FM builders are up to it and want to make some eyecandy though I'd rather see a new painting or tapestry than an overly detailed handrail built with hours of brush work in other words.

Edited November 24, 2010 by Aprilsister

[Bikerdude]

nbohr1more : Wow - I was looking at the floor - didn't even see that vertical line for a while. Yes, that might have been helped with a bit of grime.

 

Ah i see what you mean, ok in that case then I suggest a lighter or thinner edged grime decal...?

Edited November 24, 2010 by Bikerdude

[Sotha]

I'm thinking what Sotha means is not that he can't see any difference between the before and after shots (which incidentally are not the very best examples of this grime useage) but that he doesn't prefer one to the other so why go to all that trouble (and it is a lot of trouble I admit.)

Yes, exactly. It is a trouble/benefit ratio -thing and I do not see it worth the trouble.

 

BTW, does anybody else use leaning left and right now and again to get better depth perception when there is doubt?

Well, I do sometimes. It is very rare because usually I get the correct impression of my surroundings simply by moving. More commonly I do left/right leaning to see better objects obstructed by other objects.

 

This may well be true but as a player I will vote for Sotha's philosophy of concentrating on making the gameplay fun rather than tweaking things endlessly. TDM looks great to me. If FM builders are up to it and want to make some eyecandy though I'd rather see a new painting or tapestry than an overly detailed handrail built with hours of brush work in other words.

 

Exactly. There are brilliant mappers, who are capable of creating fantastic scenes, but sometimes with the expense of gameplay. All these mappers need to do is to take one unit of mapping energy and place it in gameplay rather than appearance. Perfect missions would be generated..

 

But of course, the main motivation to map is because it is fun. And people feel entertained differently, if someone really enjoys honing the appearance to perfection, it is certainly their right.

[nbohr1more]

I have yet to see a mission that I've found to be awful in either appearance or game-play but I suppose the converse could be true as well. A mission author who makes fun maps could go back and tweak and refine the appearance to improve the overall experience to perfection. Perhaps if you wouldn't mind, someone super fastidious about details (like Fidcal) could go over your maps and touch-up things according to your artistic taste. That would save you the tedium of that kind of minutia and would make your missions more appealing to folks who (wrongly) judge by screen-shots... (Again, I still think your missions look fantastic so this is all relative of course ).

[Berny]

WOW!

 

 

 

 

This looks fantastic!!!

 

 

 

 

 

 

nbohr1more : Wow - I was looking at the floor - didn't even see that vertical line for a while. Yes, that might have been helped with a bit of grime.

 

Here are a couple of my own before (without grime) and after. I certainly think the 'grime' shader works well as either dirt, shadow, simulated ambient occlusion, patina, whatever.

 

 

Edited November 24, 2010 by Berny

[Fidcal]

There is room for all kinds of FM. I have frequently promoted effort value, perhaps more on the private forums than publicly. Chalice of Kings was good value - basic build in 27 days plus maybe 14 tweaking - whereas Heart was poor value. Heart probably took ten times the effort but it's not ten times better than Chalice. But we need all types.

 

But this is going off-topic and is kind of assuming that everybody believes that using grime corners is almost unnoticeable and adds months to a build whereas most players think it looks great and it only adds a week or so to a big FM.

 

To me, not using grime is like saying let's not bother about texture quality or model quality or animation quality when something 10% less will do. It enhances the textures, adds character, dramatically improves the lighting and shadow, softens corners, and transforms rooms visually in many ways. I'll have to see if I can find a better before/after example.

[nbohr1more]

I am grateful for any mission made through any design philosophy that the mapper desires.

 

I am just a devil that wishes that (for example) Bikerdude would just completely unauthorized go out and spiff-up Sotha's maps with a little extra sparkle and then just hand it back and go "what do you think?"

 

Then have Sotha tinker with the AI patrols in a Bikerdude map as retribution.

 

I think that the contrasts in design style sometime produce something that is better than the singular experience. I just have this John Lennon\Paul McCartney, Roger Waters\David Gilmore mentality...

 

Again, I must stress that I love missions from both authors...

[AluminumHaste]

I appreciate any map that someone puts out regardless of technique used for imagery.

Whatever the author decides is what I will take, the more the merrier.

[Melan]

Heart has probably been the single best TDM mission I have played, particularly commendable for its masterful use of environmental interaction. So what if it took some more time to build?

[ungoliant]

Oh I agree, reality often is less than stellar compared to TDM

The only complaint I have with those comparison shots (and I have to say they look lovely), is that Ambient Occlusion does not have any effect on an anything other than acute or right angles, like corners and such. The outside edge of a corner will be fully lit so AO will not be applied. so there should not be any AO on pillars, since there's no occlusion going on.

 

yep, grime is good, but maybe it shouldn't be spammed on every square surface. for issues like protruding edges like the pillars, grime could still be used, but maybe just use the texture tool to pull the edge of the grime away from protruding edges, and leave the grime on the recessed edges. takes a bit more work, but i think it would look better.

 

maybe just using RGB keywords in the material definition for the more distant parts of the wall to dim the texture would be an easier solution than projecting that hard-to-align ambient texture...

 

Maybe the new distance-based stuff in 1.03 could be exploited to manipulate RGB brightness levels based on "distance from the light source AND the player" for mapper specific areas of interest... (or maybe they have a better built-in solution anyway... )

 

mmm, as far as i know, using the RGB keywords affects the entire texture, not portions of it, and is not distance based. also, if you really wanted to do that for multiple surfaces in a released map, you would end up with material files with like 50 "duplicates" of the same shader, except with different rgb values... probably would be easier to work with the lighting instead.

 

Haven't gotten that far in my own map yet, but I've had ideas run through my head about binding an additional very soft ambient light with a falloff that stretches all the way across a room, that would supplement a non-moveable light source. If extinguishable or could frob on/off, the bound ambient light could turn on/off with the light. maybe that'd help?

[nbohr1more]

Yeah the RGB keyword method would lead to Material definition bloat unless a programmatic method were used to say "affect the RGB if this this and that"... A bit tricky. (shaderParm can be distanced checked so it's possible..)

 

What you are describing with having an Ambient tied to a real light is just about the same concept that I was talking about for having Six projected ambient lights (with envshot textures) accompany every light to have realistic faked radiance contributors.

Edited November 24, 2010 by nbohr1more

[Fidcal]

Yes, there may be some specific applications which are debatable but that hasn't been specified so far.

 

Couple more before/after grimes. If you can display these in tabs or a picture viewer and flip between them it is easy to see the improvements...

 

 

[AluminumHaste]

Yes, there may be some specific applications which are debatable but that hasn't been specified so far.

 

Couple more before/after grimes. If you can display these in tabs or a picture viewer and flip between them it is easy to see the improvements...

 

 

 

Now you're getting it, those look fantastic and a much closer approximation of real AO .

 

EDIT: I just wanted to add that I have a perfect real life example of AO in the office here at work, I just need to take a pic and upload it to show you guys what I mean.

[ungoliant]

whoooaaa, those latest after-shots are seriously groovy

[nbohr1more]

Argggh!

 

The wait for this mission keeps getting tougher with every screen-shot

[ungoliant]

I probably have no idea what i'm talking about, but i wonder.... if we ever get gradient alpha when open source is released, if you could export an entire room to a model using the export script, make all the surfaces completely transparent, then bake in AO with gradient transparency, etc, and import the whole thing back on top of the original scene and put noshadows, noimpact, etc on it? i'm sure I described that entire process incorrectly, but you get the idea of what I mean. If it worked it would completely eliminate the necessity of grime altogether.

[Melan]

Couple more before/after grimes. If you can display these in tabs or a picture viewer and flip between them it is easy to see the improvements...

Note how the grime decals not just add a patina-like effect, but also smoothen out the edges of architecture - see particularly the first two shots.

Edited November 24, 2010 by Melan

[Fidcal]

Yes, in the library the edges of those pillars is softened especially when there is shadow behind.

[STiFU]

Love your grime-work for the most part. I am just having trouble with grime applied to convex edges, e.g. the vertical edges of the pillars. Those edges shouldn't be darkened, unless it's supposed to be dirt instead of lighting variation, which would still be rather weird. I bet it would improve the scene greatly, if the grime was only applied to concave edges.

 

EDIT: I believe I also noted this about Heart once.

[ungoliant]

yeah, there were some big blocks on the players height that could be heavily illuminated with the player lantern in heart, looked awfully funny with the protruding convex edges being pitch black under a full light.

Edited November 24, 2010 by ungoliant

[AluminumHaste]

Note: the first shot is of an unfinished area. The rest are also subject to change; for a start, those candlelights are getting desaturated in the final version.

 

So I was just taking some screenshots of your v1 of Return to the City, could this be part of the reason you had performance issues? There must be a hundred AI in that part of the map.

 

[Melan]

It may have contributed, but it was not the most serious problem - after all, a lot of sections performed well even on relatively weak configs. The area has been nuked, however.