Jump to content

Taz1004

Members
  • Posts

    2479
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by Taz1004

  1. Included lua edit that optimizes Clipmaps for all terrains. This should save couple more hundred MB of VRAM again at no quality loss. Since it's lua edit, it's very small download. DOWNLOAD If you are willing to sacrifice some quality, then reduce the value at line 153 to 512. Should save another couple hundred MB at very slight reduction in quality on distant mountains. I actually didn't notice the difference until took screenshots and flip them back and forth. Any edit you do to this file affects ALL terrains. This was tested on PG, Caucasus, Syria, and Mariana Islands. null
  2. This was my first paid campaign. I first played Operation Piercing Fury which as we know is open source. Even tho OPF was free and I enjoyed it, many parts of it is broken since it hasn't been updated for a long time. So I had to do a lot of modifications in Mission Editor. I didn't like that part as it took out all the surprise element of missions. So I decided to try a paid campaign hoping that everything works without me having to fix anything. I enjoyed about half of the missions in this campaign. The other half, frustration. I wouldn't necessarily fault Baltic on it as most if it probably is on ED's AI. I liked the story, voice acting, procedures and documents. With that being said, I don't think I would ever buy another campaign. Not just from Baltic but from anyone. Bugs are still present (regardless who's fault it is) and not being able to fix it myself actually added to the frustration. And the missions are spoiled anyway from playing multiple times due to the bugs so my original goal for purchasing a campaign was not met. Only suggestion I have is to keep the missions simple and less dependent on wingmen. Thanks to everyone who worked on this campaign.
  3. Anvils not launching seems to be the source of this problem too. But they still don't launch even when I use CAT1. And still don't get how one F5 takes out two hornets consistently.
  4. I was trying to troubleshoot the exterior light being too bright (which was brought up many times before) and noticed that the size of the light depends on the game resolution. Below is screenshot taken at two different resolutions and then blown up to match in photoshop. You can see the light appears larger in 1280x720. This does not happen on Viper or A10C-2 as shown further below. Screenshots are taken in 2D to compare better but the problem is much worse in VR. Because size of the light on Hornet depends on the resolution of your VR mirror. Not resolution of your headset. So if you have small VR mirror, you get huge Christmas lights in VR. Light rendering on Hornet needs to be world space like Viper. Not screen space. I believe this is what's also happening to Supercarrier IFLOLS. Viper - Size of the light remains same between two resolutions.
      • 8
      • Like
      • Thanks
  5. Tried again and same thing. One Iranian F5 wasted both Sledges and came after me. Another thing is that I watched RedKite and Iankypilot's video and they're getting confirmation from Anvil 12, 13, and 14 on bomb hits. I'm not getting that. I just get Flip saying "bug out to 230" and then nothing. No mods or Vaicom.
  6. I hit my targets in formation, egress to 230 and nothing further happens. I saw something got shot down to my right thinking it was the F5's but mission debriefing shows it was actually one F5 that shot down both Hornets. Is this supposed to happen?
  7. By the way. I think terrain noise should be turned off in FLIR. At least the very fine TV static looking noise. Noise is supposed to simulate variation in grass color etc. Grass temperature doesn't change if it's slightly darker grass or lighter grass. Most real FLIR doesn't show these type of noise at close up. This is what I was initially trying to do.
  8. Track is attached. There are more extreme cases but I chose this spot to keep both trucks in close proximity to reduce outside factors. And I didn't compare between different type of surface like grass vs tarmac in case it might be the temperature of the surface heating up the unit (I don't think this is simulated in DCS but still). These are both on grass. And I know these two parts of grass use different noise image. And the difference becomes more apparent if I alter the terrain noise image. And I used Hornet this time thinking maybe it's auto exposure of A10C-2's TGP messing with it. So this one I tried on Hornet with ALG turned off. And same result. The difference will be much more noticeable if you take screenshots and compare. Test_TerrainFLIR_1.trk The question is... why is the terrain the units are sitting on affecting the IR signature of the unit. Even when it's same type of surface.
  9. I was trying to adjust how the terrain appears in FLIR and discovered that the terrain texture affects the unit's IR signature? Below is same truck placed on different type of grass not too far apart as you can see. Clear sky, no fog. And you can see the one is brighter than the other. I masked them out to compare. They also look different on the road, grass, dirt... So what may look right on one surface may not look right on another. Or what might look right on PG might not on Caucasus etc. Below, the two flipping images are taken from same track file except that I took out the noise from the terrain on one. Only thing changed here is "landnoise.tif.dds" in Caucasus. Nothing on the truck's texture or FLIR texture. But that terrain noise texture is affecting the truck's IR. null Part of the problem with FLIR system might lie on the terrain. Not the unit's FLIR files. Or how the FLIR is rendered on the terrain. I couldn't look at how FLIR textures are rendered but it may be rendering as multiplier or transparency?
  10. Just by remote chance somebody internally at ED insists that there is benefit to 32bpp normal maps, referencing videos like below, I'd like to make one things clear. One thing common about these videos showing benefit of 32bpp normal maps is that they almost always show it very close up on highly reflective surfaces like a chrome or mirror object. Because that is the only way to see the difference. On real world game scenarios like DCS where nearly everything is matte and weathered, you will not be able to tell the difference between 8bpp and 32bpp. Below is another video that starts out by showing the benefits of 32bpp normal maps on chrome material but towards the end at 11:49, he shows 8bpp is just as good for matte materials. And as he said in 12:22, if you bake in 32bpp out of 3D software and then reduce it to 8bpp in Photoshop, it works well even on chrome materials. This may be viewed as another step in the pipeline but all this can be automated with batch, command line, and even scripting. And if somebody does make a chrome plane, it still should be 16bpp as the video shows. Not 32bpp.
  11. Thanks. Original post was about 3 years ago when A10C-2 was released. My hope at the time was ED would be more efficient with textures going forward at the least. But I saw that it was getting worse with every new content such as Shark3, Apache, and new FLIR.
  12. Did you get that backwards? And you don't have to own the modules to view them in Encyclopedia.
  13. By a lot? I get better performance on Syria in MT. By restarting DCS, I do get lower VRAM on same mission than playing multiple missions in a row. But as long as it's not maxed out, I don't see much difference in FPS. When you bring up Viggen or Mirage F1 in Encyclopedia, does it show up right away or does it take long time?
  14. I can only speculate from my desk. But if it's 7.8 without and 8.1 with, then it sounds like the mission you're running is maxing out your 8GB VRAM either way. If the mission requires 10GB VRAM and this mod saves 1GB, you're still gonna max out. I suggest testing on lighter mission. Also, make sure you compare under same conditions. If you start another mission or browse encyclopedia and then run your test, it'll report higher VRAM.
  15. This is another one I've been using which I haven't shared yet. Some may find it useful. By default, on screen text is too large for me in VR. Often overlays on my cockpit if the message is long. This mod reduces the font. Also makes comm menu to be smaller. They look too small in the comparison below but they look right in VR. Reduced Text Font With comm menu ADJUSTMENTS: Onscreen message font sizes are adjusted by line 148, 255, 340, 462 of gameMessages.dlg. Default font size is 20. Comm menu fonts are adjusted by line 49 of CommandMenu.lua NOTES: Passes IC Even with this mod installed, it doesn't change the font size in 2D. This only affects VR. INSTALL: Files go into main DCS install path. Mod Manager recommended. DOWNLOAD
  16. What's the total size of DCS main install folder before and after.
  17. Yea, I try not to touch textures for complex shaders. Damage map being another one. You'll find out which ones they are as you do more as you're already finding out. And whenever I see regular image files like tga, jpg, png, bmp, I do try to convert them to dds. Those regular image files do not support mips and they shouldn't be used for diffuse. Converting them usually work as long as you use same file name. And you have to create mip levels using tools menu in Visual Studio.
  18. Done. Thanks!
  19. Files are total 16GB. It saves you 24GB. So that's still 8GB saving even if you don't overwrite. But this is why I said: I highly recommend putting your Mod Manager repository on separate drive than your DCS install drive. Even if it's on a USB drive.
  20. Optimized liveries uploaded. Liveries alone would save about 10GB of drive space. There were some crazy large livery files. I don't do multiplayer so I'm not sure but I'm guessing if players join with these huge liveries on multiplayer server, I can see how it could pile up on VRAM. Especially with memory leak.
  21. Whoever did these South Carolina National Guard Liveries for Apache had them optimized exactly the way I would. So there is someone at ED who agrees with me.
  22. It was on Kegety's thread. You can download below. Search for "Taz" on where to adjust. DCS will recompile each time you re-install this. Doesn't take more than 30 seconds for me. Taz1004_Haze.zip
  23. It's just haze which I already shared. I just put it as separate mod so I can test other mods without having to disable it as that requires recompiling. Man, you guys are after the crumbs.
  24. Just my various preferences. Labels, haze, shadow, countermeasure programs etc.
  25. Anything that's not cockpit, terrain or liveries. Basically ground units, ships, infantry, weapons... I'm working on liveries next. They are messy.
×
×
  • Create New...