[Turambar]

are those alpha blended?

you could get much deeper reds with alpha blending, as it is now, its a little pinkish.
there's also a dim white area on the outer edge of the beam, and it kinda detracts from the effect

[Herra Tohtori]

True alpha blending doesn't work with beams... Well, not very well, to say the least. These ones have a black background and FS_Open uses the black as pseudo-alpha channel. I can post screenies of alpha-blended beams (which I did try first) if you wish to check it out, but you'll risk severe retinal hemorrhage if you look at them too long.

I'm working on the whitish stuff to reduce it. Also, in the green beams the core is way too thick and bright, the edges lose most of their detail... But as I said, I'm working on it.

And what comes to colour tone, I tried to make the beams consistent with the tones on the beamglow animations.

[Turambar]

well, can i see the awful horrible blended ones?

im probably gonna give it a shot myself eventually, and i want to see what it looks like.

[Herra Tohtori]

Well, here is one example what happens.

[attachment deleted by admin]

[Turambar]

eew

someone get a coder on this, stat!

well, it looks like the transparency actually works right, but the colors and brightness are off.

[Herra Tohtori]

I think Taylor already knows about that because he gave me a lot of advice when I started making these beams. Anyway, I have some questions regarding two Beam Section values. +Tile Factor and +Translation.

I'd like to know on what these two Beam Section entries actually do. The wiki entries are somewhat lacking when it comes to documenting at least these two:

+Tile Factor:
Implemented before FS_Open 3.6.5
Defines the beam texture tileing factor. Second number defines the tileing type. Type 0 tiles the graphic in the beam as many times as defined by the 1st number. Type 1 uses the 1st value as the lenght of the tile.
Syntax: Value, Value

Okay. While trying to figure out suitable table values for my beams I have become slightly frustrated in this documentation, because it doesn't actually offer any information. The second value is either one or two, and it defines how the first value should be interpreted. And if the second value is "1", the first value should define the length of the beam texture.

The problem is this: The length of the texture is definitely not as many metres as I put into the field. It's much, much more... Most beams are about ~50 metres wide. That kind of beam looks (usually) rather good with Tiling Factor values from about 15 to 30. It doesn't make any sense at all. So it doesn't use metres as units, that's for sure. But using aspect ratio doesn't feel like working either.

What I would like is that someone who knows what the code does with these values would check and update the wiki as to what units does the beam system use as "texture length", or does it perhaps even take a fixed aspect ratio multiplier (?) or something instead of actual length...

What I would like is that all same-coloured beams would have the same aspect-ratio in-game. But I cannot achieve this because thetable entry documentation is what it is and to be honest, I'm done with trial and error. If I don't know what to expect I can never get suitable values for all the beams, and I can't really go and read the code to find out what the hell it's supposed to do so...

Another one:

+Translation:
Implemented before FS_Open 3.6.5
Defines the speed of translation of the beam, a positive value moves from the impact point to the firing point, while a negative value moves away from the firing point towards the impact point.
Syntax: Value

Value? Value? What value? How big should the value be? Metres per second? Or should the values be positive/negative integers?

Metres per second didn't seem to work that well. I tried to make them go forward two times their width per second, and they became a fuzzy line, which makes me think that the speed was tad bit too high... But any other units just don't make any practical sense, so I'd like to know what units does it use.


Don't get me wrong now, the wiki is excellent tool, but it's just plain irritating to try and interpret some ultimately ambiguous definitions for some table values.

So, thanks if someone takes the time to find answers to these questions.

[Turambar]

look out Herra

your next post is 1000

make it a good one!

[Herra Tohtori]

Hey, thanks wor warning, I'll make it good.

...Oh bugger.

[brandx0]

Best thousandth post ever...

[Herra Tohtori]

There's nothing special in numbers.


By the way, if someone is interested in slightly less hideous Bakha...



With 1024x1024 texture, glow- and shinemaps. It's simply made on top of the old texture because I haven't figured out how to do UV mapping, so some parts may be a little screwed up but in my opinion it's better than the hideously low-res retail map, so it'll do for me until someone decides to to a proper HTL Bakha. If someone else wants to use this until that, it's fine by me, just tell me if anyone wants this, and I'll post it.

And, I hope I'll have some answers on the mysterious beam table entries sooner or later...

[WeatherOp]

Those Shivan beams are much better than the current ones.

[WMCoolmon]

Kudos to whoever it was (taylor, I think) who made lighting apply to decals. They look much better now.

[Mongoose]

That Bakha map is almost infinitely better than the current low-res monstrosity.  Kudos.

[Herra Tohtori]

Bakha maps, enjoy.

If someone really wants the TGA files, be my guest and PM me. This one was surprisingly easy to pull off. You gotta love what kind of a surface you can get in Gimp with noise->gaussian blur with value 2... It creates nice amount of detail and structure to completely blurred surface. The hull plating was almost all that I had to do to it after all. Glows turned out rather nice, and speculars were (imo) spot-on at first try, although some may dislike the metallic look on the plates. But I figured that they probably are metal, as they look like exposed armour plating.


But, the beam table trouble is still there. Is still don't know for sure what the tile factor value does and that causes inconsistensies in the beams, which is something I do not want to happen. So, no properly configured beams after someone who knows the code takes a look at the values and what the code does with them. Pretty please.

[DaBrain]

I already reworked the Bakha maps, but I'll take a loook at your maps when I'm back home.

The look nice on the screenshots, but it's really hard to tell if they're better, cause the pics are pretty small.

[Herra Tohtori]

The pics are thumbnail links to 1280x1024 shots...

There are some parts where the armour plating is not spot on in relation topolygons in the model, that's because I simpley drew the lines on top of the old texture which had the gaps between the plates about a mile wide so I had to kinda guesstimate where the actual poly edges were in the model and so on. But in fact it looks kinda good to my eye in-game.

Though, if you have already reworked them, I don't doubt your job will most likely surpass mine in quality. Could you post some screens too?

[DaBrain]

I made them in like ... ten minutes or less. Therefore, my maps do not looks very great.
On the other hand, the texture isn't that hard to upgrade.

Afaik UnknownTarget is working on a HTL version of it, that will use new maps, so it's only temporary anyway...

[Turambar]

i dunno, last time UT worked on a vasudan ship, he kinda chickened out halfway through

*hacking cough* PTAH *cough*

[DaBrain]

A shot of my version:

[Herra Tohtori]

Cool. The hull plating lines looks better there than on my version IMO. And the cockpit also seems to be more seamlessly attached to the hull.

Wanna try my glow map on it?