Signal Aspect Visibility

Several commercial routes have recently touted methods that have improved signal aspect visibility from just a few tenths of a mile to more than a mile. I've seen one such method and it really does work. A 2 color aspect can be discerned through the cab windshield at about 9/10's of a mile down straight track against a dark background.What I would like to know from those who have mastered these techniques and have hooked up with the for sale people: When will this item be found in the public file bin, like all those wonderful track systems?? Or when will the recipe be explained so those of us with the talent can create the textures ourselves?Regards,
Hank

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: Signal Aspect Visibility, Claude350, Jul-18-04, 06:04 PM, (1) RE: Signal Aspect Visibility, Signalmaster, Jul-18-04, 06:59 PM, (2) RE: Signal Aspect Visibility, ctillier, Jul-18-04, 07:44 PM, (3) RE: Signal Aspect Visibility, Coonskin, Jul-18-04, 10:02 PM, (4) RE: Signal Aspect Visibility, Claude350, Jul-18-04, 11:36 PM, (5) RE: Signal Aspect Visibility, CVL_Post, Jul-19-04, 01:59 PM, (6) RE: Signal Aspect Visibility, DTA, Jul-19-04, 06:10 PM, (7) RE: Signal Aspect Visibility, Mark_Blum, Jul-19-04, 10:01 PM, (8) RE: Signal Aspect Visibility, Hack, Jul-19-04, 10:47 PM, (9) RE: Signal Aspect Visibility, sniper297, Jul-19-04, 11:16 PM, (10) RE: Signal Aspect Visibility, GreatNortherner, Jul-20-04, 05:11 AM, (11) RE: Signal Aspect Visibility, Signalmaster, Jul-20-04, 06:43 AM, (12) RE: Signal Aspect Visibility, ctillier, Jul-20-04, 08:22 AM, (13) RE: Signal Aspect Visibility, sniper297, Jul-20-04, 09:38 AM, (14) RE: Signal Aspect Visibility, Hack, Jul-20-04, 11:25 AM, (15) Elementary, Watson!, sniper297, Jul-20-04, 01:35 PM, (16) RE: Elementary, Watson!, Hack, Jul-20-04, 02:41 PM, (17) RE: Signal Aspect Visibility, fixin2, Jul-20-04, 02:42 PM, (18) RE: Signal Aspect Visibility, sniper297, Jul-20-04, 03:57 PM, (19) RE: Signal Aspect Visibility, jon130, Jul-20-04, 05:00 PM, (20) RE: Signal Aspect Visibility, Mark_Blum, Jul-20-04, 05:26 PM, (21) RE: Signal Aspect Visibility, sniper297, Jul-20-04, 06:10 PM, (22) RE: Signal Aspect Visibility, cn_bayview, Jul-20-04, 06:59 PM, (23) RE: Signal Aspect Visibility, jdilworth, Jul-20-04, 07:15 PM, (24) RE: Signal Aspect Visibility, sniper297, Jul-20-04, 07:34 PM, (25) RE: Signal Aspect Visibility, CVL_Post, Jul-20-04, 09:40 PM, (28) RE: Signal Aspect Visibility, CVL_Post, Jul-20-04, 07:41 PM, (26) RE: Signal Aspect Visibility, EugenR, Aug-08-04, 05:18 PM, (31) RE: Signal Aspect Visibility, CVL_Post, Jul-20-04, 08:31 PM, (27) RE: Signal Aspect Visibility, sniper297, Jul-20-04, 10:55 PM, (29) RE: Signal Aspect Visibility, sniper297, Jul-21-04, 06:20 PM, (30)

Conferences | Topics | Previous Topic | Next Topic

Well Hank I think you may be on the dead end of this one.I would love to see your work with this technique installed.But I don't think the payware folks will devulge their secrets on this one.If they do I'd love that.Your signals with this new technique will give the payware folks a run for their money.I hope something will come thru for you.Claude Debnam NS Piedmont Division Raleigh East

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

ClaudeWell, maybe not. It's not the first time the question has been asked just what can be proprietary and what cannot. I'm surely not against payware, I've supported many payware ventures, been pleased and sometimes disappointed.I am far from a copyright expert, but it is my understanding the manipulating a Kuju/MS proprietary object like an ACE file, it can't be owned. I can't "own" the signal scripts I've written to create new signal functions not provided in the off the shelf MSTS. All I can hope for is that who ever uses the rework script gives credit where credit is due. Why wouldn't the same apply to a 'doctored' ACE file?With payware, aren't we just compensating the authors for the time, effort and mass production of the product and not the contents thereof? Copyright protects my signal objects from being used in a pay mode without my permission.Hank

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Since when did MSTS have any secrets?While not everybody knows every trick in the book, that's not the same as something being "secret"... as soon as Cascades ships, this stuff will be reverse-engineered faster than you can say it.I don't know if you were around when MSTS first came out, when Train Artisan announced a "secret" technology for how to fix the blurry textures problem (mip bias -3). That secret technology is now common knowledge, and the same fate awaits this distant signal thing.cheers

--
Clem

http://www.tillier.net/speedworks/index.html

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hank:Fret not. I'm sure the new technique will be quickly assimulated and understood once said product is in hand. Like you say, aside from models and perhaps original textures, there's not much about this sim that isn't available for reuse to any/all with the skills to open it up. That's a good thing, IMHO. We all benefit. You're right again in that for this payware vendor, I'm only trying to be compensated for the huge amount of work it is to create a route from scratch. No "techniques" are copywrited, only the original objects/textures have a form of copywrite on them. As far as you giving us "payware guys a run for our money", your work has always been top notch and held in high esteem by this payware guy, long before I was a payware guy. Thank you for all your work you've accomplished to date within the signal art.Andre Ming
Visit VSC's temporary web page at: http://www.vscalecreations.comVisit VSC's Forum at: http://www.3DTrains.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

I sure hope you can get it Hank.I'd love to see your signals with that fix.Got any new signals in the works.Anymore work being done on the D type signals that was posted alittle while back?Claude Debnam NS Piedmont Division Raleigh East

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jul-19-04 AT 02:02 PM (EST)

Hi Hank,I've been busy with making Dutch payware Signals about a year ago and i've found a methode to increase visibility, but i'm not sure if it's the methode you looking for, you might allready used it.I've made a light-texture with a corona, which means the light texture is increased in size and so gives a better visibility.
I've used the PSP-sun option to create the corona.
Don't vorget to edit the light-radius in the sigcfg.dat so the center is the same size as the old light-texture.
What i mean is at the right side off the screen-shot i've attached.But you made me think off the light-texture, so i used ReMipIt on a number-light-texture and visibility or i might better say readability was increased, i also notices that ReMipIt hasn't hardly any effect on the visibility but only on the readability.
At the center off the screenshot i've put a example off a number 12 at a distance off about 150 meters.On the left on the screenshot you can see the visibility-distance i can reach with the corona at night, it's about 500meters, which is much better than the standaard MSTS-signals!But you can't see the light anymore, i think because off the jpg-compression, it was about one pixel!I hope this will help you a bit,greetz from Dereck The Netherlands

Greetings from Dereck van Dijk, The Netherlands.
I've made some nice Dutch Payware-signals, see at www.msts.nl and click on "Dick's seinen"

Attachments

http://www.train-sim.com/dcforum/User_files/40fc196e48fa47fd.jpg

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Some of the Protrain series used it, but basically it doesn't work, because you can see yellow and red signals in the distance, but cannot tell which track it is on, and green remains dull. Often you slow up for no reason. The cabview systems for signals is a far better option.David

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hank:I'd have a hard time giving this away if I were a payware vendor, but I sure would love to see it in your hands. It's only a matter of time. I have a theory, but I'm not a model maker. Perhaps someone will comment.I think they make the light very large, so it can be seen at a distance. Something to do with LODs? Then, as you get closer, some dark rings start to appear and block out the big red, green, or yellow disc, working from the outside in, to make it smaller, until you get up close. Larger and larger obliterating rings appear as you get closer, and when you're close enough to see details, the huge bright light you saw at a distance is blocked out except for a proper center. Heck - maybe it's alpha'd out using those viewing distance adjustments. I don't know if any of this is possible, but it wouldn't suprise me.Maybe this reply is worthless, but I thought I'd throw it out for comments (as he ducks under the desk).Mark
A Southern Railway Fan

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi Hank,Scott Miller developed the signal images for the Surfliner, which he'll be releasing to the public shortly after Surfliner goes live. His method involves the texture itself, and not the model, so I suspect anyone will be able to use these images by simply copying them to the texture folder (assuming he's using default names, which I'm not sure of off hand). Cheers!
Marc - 3DTrains
http://www.3dtrains.comPro Series SuperPacks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

That was my first guess, so I tried using remipit (made by the same Scott Miller) on siglight.ace, no effect. I also tried wierd stuff like making it totally transparent with the alpha channel (no lights at all!), cranking up the gamma (lose all the fresnel detail without any improvement in distance), changing the texture modes in SFM to full bright, cranking the LOD to 8000 meters, even hacking the colors in the sigcfg.dat to make them brighter.Near as I can tell the color plus brightness are defined in the sigcfg.dat file, and that color is overlayed on whatever texture is defined for "signal" in the shape file. I suspect ol' Scotty has been holding out on us, came up with some kind of super remipit that cranks the visible distance way beyond the remipit that's in the file library.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

For all those that want to increase signal visibilty quite simply:
just increase each signal (aspect's?) radius in the sigcfg.dat (backup first!)Be aware, though, that it looks quite ugly when you come up close to a signal - they have kind of huge lamps, then. But if you don't overdo it (just experiment a bit) and don't mind your signals looking not exactly real you can easily improve gameplay a lot by getting signal visibilty up to nearly a mile. If I remember correctly, I've set mine to radius 0.35, which is quite large.Not the best solution, but for actually driving trains (and not willing to use the unrealistic track monitor) I found this to work quite well.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hey, Sniper!Just a bit off my topic, how did you get the cabsigs in your GP38? I tried to add one to the GP38 in the CE but it didn't work.I had the cabsignal ACE, used the add display dialog, set the parameters, moved the display to where I wanted it, saved the work. Fired it up and it crashed. I've looked for a tutorial, but can't find one.Hank

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Guessing from the suspect involved, I'd wager half a six-pack that the trick is to make the image larger with each mip level...cheers

--
Clem

http://www.tillier.net/speedworks/index.html

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

That might work, but then the light itself wouldn't be larger because the light size is defined in the sigcfg.dat as radius? Can't even try it tho, since the remipit I have don't have that feature. Larger radius does work very well, altho it looks like *&%$#! up close. The signal set I have is one that Hank sent me in bits and pieces via email, and looking at those in SFM the LOD is one and only one, 2000 meters. Changing that so the LOD increments at 250 meter intervals, for example, could work if the signal lens were disproportionatly large at 10 polys, then smaller at 20 polys, finally the actual size at the full poly count. I just now looked at one of the BNSF signals in SFM, it's 102 polys at 30 meters, 42 at 50, 22 at 150, and only 4 polys at 500. Haven't experimented with boosting those yet since I just now thought of looking at a default signal shape!Hank, I don't think there is a cab tutorial anywhere, to me it's the most difficult editor in MSTS. As you recall, back in the early days there were reskins, then activities, then routes, then new train models, and new cabs came last. Pretty much confirms my impression, the cab editor took the longest to figure out. The only thing I can suggest is don't do it the way I did it, trying to jam a clock and speed limit ahead display on top of the aspect display, that sometimes works and sometimes one of the digital displays grows shy and hides underneath the aspect ace file, and no amount of coaxing can induce it to come out and play. The other thing I've found is aliasing, if the CVF you're editing is all self contained it works, if some of the controls or displays are aliased to a different cabview it don't work. Opening the CVF file in wordpad and copying and pasting the;CabSignalDisplay (
Type ( ASPECT_DISPLAY CAB_SIGNAL_DISPLAY )
Position ( 0 0 64 112 )
Graphic ( aspect.ace )
)Sometimes works, sometimes don't, but the main thing is to add whatever new displays you want to an unaliased cabview, then edit in wordpad later to do the aliasing and delete all the ace files from that cabview folder. Cab editor don't like aliasing for some reason. In fact as near as I can tell it don't like me either.I have the same problem with signal scripts, the speedposts suck because they "capture" any parallel track within 30 meters, and I'm trying to control AI traffic speed without affecting the player track. Simple solution for that would be to clone some signals and make the least restrictive indication approach, with 5mph, 15mph, and 30mph variations, so it would affect only the AI track. So I cloned the searchlight dwarf, carefully copying and pasting the datablocks, incrementing numbers where necessary, in both files, doublechecked, crossed my eyes and dotted the Ts, and the end result is send/don't send before I even place one on the track.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Scott manipulated the mip levels in the ACE file by hand. As mentioned, he'll be releasing this sometime after Surfliner goes live, so I'm not sure why you guys are trying to recreate what's already coming.Cheers!
Marc - 3DTrains
http://www.3dtrains.comPro Series SuperPacks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

"I'm not sure why you guys are trying to recreate what's already coming.", I should think the answer to that is obvious - impatience. You've been here from the beginning too, Marc, so I don't have to tell you the whole sad story - Buggy game with unsupported editors and tools, tech docs translated from Japanese by a Frenchman into what he conceived to be English, 8 months waiting for a patch which didn't fix anything, no more patches after that, hints, announcement, and cancellation of MSTS 2, and now PI Engineering and Train Artisan are working on new and better games."he'll be releasing this sometime after Surfliner goes live", MSTS is 3 years old and its days are numbered. If one of the aforementioned games hits the shelves "sometime after Surfliner goes live", then Scott's new magic will be completely meaningless to a lot of people, myself included. Kinda like the guy who waited until the year 2000 to unveil his new design for an 8 Track player, seems kind of pointless to me.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Understood. However, my point was that if you know it'll be released, then it would be more efficient, IMHO, to spend the time on other areas that would be within (relatively) easier reach. Scott manipulated the ACE format by hand to achieve the desired results - something that requires an extensive knowledge in programming and image formats. Until the ACE files are made public, LODs in the models are our best option.Cheers!
Marc - 3DTrains
http://www.3dtrains.comPro Series SuperPacks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Guess they are looking at the problems of inventing the wheel that _insert inventers name here_ is rumored to of invented but has not bestowed upon us underlings in the world. So, presently the product/technique is currently unavailable. What is wrong with MSTS brightest users discussing a problem and probable solutions? Warren

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hee-hee, well, Marc didn't really say there was anything wrong with it, he just asked why. Marc's point is trying to invent something that's already invented is pointless, my point is holding back anything for even a few days at this late stage is pointless. Nobody's pointing fingers, we're just trying to make our own points, get the point?Anyway, the point (there I go again!) is that only Scott "Igor" Miller has the whole technique, and apparently it requires specialized skills, so there's no point in sharing the method itself, since not everyone could duplicate it. We just gotta wait until the ace file itself is made available, then everyone can use that.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

What about the free download of Sigview from MLT available here http://www.mapleleaftracks.com/news.html under the BNSF free download.Jonl

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Well, JonSpeechless for a moment. Perhaps you've said more in your third post than I've said in a couple hundred over the years. Not being too interested in BNSF, I never would have found it!I guess MLT was, or is going to get a chuckle out of this one.Thanks!Mark
A Southern Railway Fan

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Ha! Well, let's hope that's an early beta.It works, the two signals under the arrow are actually more visible than they are with this jpeg compression, but look at the near signal on the left! Didn't Scotty have this same problem with the cornrow objects? Somehow he finally got the alphas to play nice, but I forget how.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jul-20-04 AT 07:02 PM (EST)

I got it to work on KHP2 perfectly and on the CSX M&M- I believe it has the same signals as MP3.1 and FBL V3, so adding it to those routes should be a snap! Here is a shot from KHP2...Looking across the Kicking Horse river, about half way to the Field station. Closer.Close up.JPG compression made the lights a little harder to see in the pictures, but in the game its great. It took me a few minutes per file route just changing the #'s. I havn't had to experiment much.M&M looks about right at 1.04, but I'm going to try 1.03 or 1.02
Rob

TH&B Van #68
www.greatlakesrail.com

Check out my train photos!http://www.railpictures.net/showphotos.php?userid=472

Attachments

http://www.train-sim.com/dcforum/User_files/40fdb0490fa6379c.jpg

http://www.train-sim.com/dcforum/User_files/40fdb05d0fe8651b.jpg

http://www.train-sim.com/dcforum/User_files/40fdb09710693395.jpg

http://www.train-sim.com/dcforum/User_files/40fdb1e113e476ec.jpg

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Oh yeah, one thing I forgot to mention in the readme. Even though the ACE file is just a transparent image file, the simulator engine treates it like an alpha transparent image. You won't be able to alter the alpha sortings for these signal lenses, so you will have to make sure you have no other alphas around. Sniper, in your case, you would have to have trees using the -trans switch versus the -alpha.Regards,
Jason@MLT

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hmmm, nope, that ain't it. stpine1.ace, stpine2.ace, and spitpine2.ace are all 43kB, from TGAtool2 Export as ace, -trans, compress DXT, patch #1 makeace. I usually do that anyway if it don't affect the quality, since the 1 bit alpha compresses to 43kB for 256x256, and 171kB for 512x512. The 8 bit alphas don't compress anywhere near as tight, so I use those only when I have to. These being forest regions might make a difference, possibly this new autogantry forest idea that Chris Schwan came up with might work. Forest regions tend to not work and play well with others when multiple alphas start stacking, even if they're all 1 bit (-trans) alphas. I'll try sticking a single object that's not a forest region in between, altho I hate to think of doing that at 80 some odd junctions.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi Jason,It's a great idea this improvement in visibility I've made some tests and i believe the secret is the edited SigLight.ace, i've tried to edit it with TgaTool2 and recompressed it, but the effect was gone, so the secret is not only the blowed-up light-texture
I think the SigLight.ace has edited mip-levels, but the opposite what ReMipIt does, it looks like a extra blurry-texture! Does anyone knows what ReMipIt does whit a ace-file??Jason, can you reviel the secret off the ace-file to us please? Thanks for answering and Greets from Dereck

Greetings from Dereck van Dijk, The Netherlands.
I've made some nice Dutch Payware-signals, see at www.msts.nl and click on "Dick's seinen"

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi sniper,This solution looks like i descibed earlier in this post:
Blow up the light-texture and edit the lightradius in the sigcfg.dat!
But this light-texture is realy blowed up to 512x512!
I used 256x256 so i have no problem that it sticks besides the signal-backgroundplates but i've less visibility, although Dutch-Signals has bigger background-plates I wonder if the light.ace is specialy edited, besides the alfa, would ReMipIt has any effect? It didn't had any effect on my light-textures i've made.But it's a great solution!Greetz from Dereck

Greetings from Dereck van Dijk, The Netherlands.
I've made some nice Dutch Payware-signals, see at www.msts.nl and click on "Dick's seinen"

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi,
Ich have installed the Siglight.ace of MLT. But the transparent Region is for me Signals of the Swiss Railways too big.Signals of Swiss Railroad:
http//www.the-train.de
Download >Objekte> Streckenobjekte> Signale&StreckenTafelnWhere I can get a siglight.ace with better visibility witth only 256x256bit?
Can you sent me your file?
ch.signal@bluewin.chCan you explain how I can make my own siglight.ace with better Visibility?EugenR

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi sniper,Same problem with the standard route "Maria Pass" Any suggestions??greetz from Dereck

Greetings from Dereck van Dijk, The Netherlands.
I've made some nice Dutch Payware-signals, see at www.msts.nl and click on "Dick's seinen"

Attachments

http://www.train-sim.com/dcforum/User_files/40fdc7764fe51ffc.jpg

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hmm, well, I haven't seen Scott since he moved out of Chicago and became one of them downstaters, but as I recall anything having to do with makeace destroys the mipmapping, so reducing the size wouldn't work. Since this one was remipped manually running remipit on it wouldn't work either. Looks like them track textures you have in that screenshot could use remipit too, that's really UGLY!! This is obviously a totally different animal, I just tried compressing the PO&N track texture (3Dtracks.zip), and I get the same effect you have there, have to makeace uncompressed and run remipit, then leave it uncompressed. This siglight.ace is compressed, so it's definitely different from normal remipit.Anyway, the problem you have there is probably the same, forest regions. The good news is the ace with the alpha is only visible from a cone roughly 160 degrees on the lit side of the signal, the bad news is you gotta put real trees between the signal and any forest region that would be in that same sight line.I've been fooling around with this:http://www.train-sim.com/dcforum/DCForumID15/3860.htmlAnd if you overdo it, the autogantry forest does drop framerates compared to forest regions. The answer seems to be a balance between forest regions and real trees, and what I did was cloned the USdecidclump1 and 2 for tree clump objects:And since they're not forest regions the visible alpha doesn't show up. Again, the problem is any forest region in line of sight of any signal needs to be blocked to eliminate the visible alpha, so you'll have to place a lot of them on any route that has a lot of forest regions. Anybody wants to try these, here's the shape files and ref text, you can steal the textures out of PO&N 8... what? WHAT?! You don't have the Port Ogden and Northern?! HERETIC!! INFIDEL!! UNBELIEVER!!!!!!!Oh all right, I'll throw in the textures too (grumble grumble);http://www.train-sim.com/dcforum/User_files/40fde6a12ee1c947.ziphttp://www.train-sim.com/dcforum/User_files/40fde6ba2f33ce89.zipDownload those two, copy into the route you want to use them on, open the route's ref file in wordpad and scroll to the bottom, open pineclump.ref, copy the text out of that and paste it under the last datablock in your ref file. The four PONpineclumps will be in the "Vegetation Group - Tree" class in the object menu. Depending on the route you may have to delete or move forest regions farther from the track to make room for the pine tree clumps. I left the shape files uncompressed, so you can open the .s files in wordpad and do an EDIT FIND "ace", replace the 3 ace file names with whatever suits your fancy.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Update on this, it seems to affect only forest regions, I don't see the visible alpha against any other background. Bad news is with the tree clumps the alpha becomes visible between the trees if there's a forest region behind them, so it has to be completely blocked. Most route developers are aware of that little quirk about forest regions, when there are multiple forest regions you get ghost trees visible thru other trees, and even the GP38 and Dash 9 hoods turn semi transparent and you can see forest region trees thru the hoods and fuel tanks. Placing a single object that is not a forest region, even a small bush, between the track and the multiple stacked forest regions cures the problem. That don't work for this ace file. One of the screwball ideas I tried was cloning an object and painting it with a completely transparent texture, and placing the "invisoshield" behind the signal, but that don't work either. Sometimes you can unscrew alpha stacking problems by throwing another alpha into the mix, but not this time.Looking at the task ahead of me, my first thought was going back to the dark spacer signals on the AI tracks like PO&N8 has, but then I came up with a better idea. The only lights that really need to be visible from a distance are the ones the player goes by, so:SIMISA@@@@@@@@@@JINX0G0t______LightTextures ( 2
LightTex ( "BNSF ltex" "SigLight.ace" 0 0 1 1 )
LightTex ( "Scott" "smille.ace" 0 0 1 1 )
)After that I went thru the different signals, using SignalLightTex ( "BNSF ltex" )for all the AI only signals, andSignalLightTex ( "Scott" )for the player trailing point, facing point, and distant switch signals. Then I renamed this new file to "smille.ace", copied the default siglight.ace from USA2 into the textures folder, now I have only 3 signal sets to worry about blocking.That cuts the number of tree clumps and forest region relocations at least in half, so I may be able to finish this abomination before the new Trainmaster comes out and makes it irrelevant. I hope.Anyway, for the wierd colors here's what some experimenting has turned up;
LightsTab ( 4
LightsTabEntry (
"Red Light"
colour ( 255 230 40 40 )
)
LightsTabEntry (
"Amber Light"
colour ( 255 255 210 10 )
)
LightsTabEntry (
"Green Light"
colour ( 255 32 196 32 )
)
LightsTabEntry (
"White Light"
colour ( 255 211 228 254 )
)
)Further adjustment is probably necessary, but the lights in the distance don't look quite as neon flourescent with those values.

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | How To... | File Library ] Copyright © 2004 by Nels Anderson. All Rights Reserved.