Jump to content

sniperton

Members
  • Content Count

    783
  • Joined

  • Last visited

Community Reputation

261 Excellent

1 Follower

About sniperton

  • Rank
    Member

Profile Information

  • Gender
    Male
  • Location
    Contra-Aquincum

Recent Profile Visitors

1062 profile views
  1. Are you sure? Each time I restarted the game with a modified clouds_samples value I could see its effect both visually and on my FPS counter. Apart from that, I fully agree with the rest you said. @rowdyb00tI still have to test v10, but concerning the two videos above, the ground seems to me too colourless on #1, and too colourful on #2.
  2. It's probably because there's no such thing as a "pure AI", what people encounter is an AI in a given mission situation, where mission design (good or bad) and mission logic (appropriate or inappropriate) plays a huge role. The AI is less prone to do silly things in human-made and human-tested missions and scripted campaigns than in auto-generated career missions. Even among career missions, there might be huge differences depending on which map, which airfield, which squadron, which type of mission is concerned actually. Ground attacker units seem to be more affected. I'm some 30 missions into the Kuban campaign from Geleznik airfield in a fighter unit, frequently tasked with air cover for attacking Sturmoviks, and haven't seen yet what you describe. One time two A-20s collided when the formation made a sharp turn, but it was probably due to mission design (waypoints placed too close to each other).
  3. I'd add the FPS impact too. If you have a better GPU, you probably won't notice the difference, but rendering clouds in BoX may take up half of your GPU power.
  4. The stock gpresets file is packed in the data\scripts.gtp archive and you overide it with this new instance. Before each update it's wise to remove all mod content, apply the patch, restart the PC, restart the game, exit the game, and re-enable mods again. If you see flickering on far-away clouds, try to reduce "horizon draw distance" to 70 km. Dithering on trees while close to the ground is an old game issue, it comes and goes, I haven't seen it recently, but I guess this is what you mean:
  5. He's probably using the mod with the default value of 64 for clouds_samples in the gpresets.cfg file. Possibly you should clearly state in the opening post what base settings this mod is designed for. For me it gives best experience on "ultra" with clouds set to "high". Anything lower than 128 in gpresets results in wobbling edges. Same as banding, which is unavoidable as I understand, but more prominent with low gpresets values. Anyway, thank you for this great mod, it's really a game-changer. I wish you could correct the lights, the scenery is sometimes too green.
  6. That's all true, guys, I didn't say there's nothing more to do here, on the contrary. I too am very much annoyed when the side gunner of a shot-down bomber snipes me through the thick smoke with a perfect 90 degree deflection shot. I just wanted to point out that according to my test experiences the AI flyer does have blind spots, and those do depend on the aircraft's geometry. Furthermore, very much depends on the priority level of the AI's waypoints (set by the mission designer). If the waypoint priority is set high enough, the AI doesn't evade you until your first rounds impact.
  7. FYI the AI has blind zones as proven and discussed here (roll down for the next 15-20 comments):
  8. I had this months ago after a game update. I could fix it by - disabling all mods in JSGME and turning mods off ingame; - deleting the update folder and let the Launcher analyze my game files and reinstall the last patch. As I see you have BoX from Steam, maybe Steam install didn't properly remove all leftovers from previous sessions, and hence the glitch. Search for "clean reinstall when using Steam". Just my two cents
  9. Just wanted to point out that low or missing AI scores are more a gameplay issue than a historical inaccuracy. Agreed.
  10. I agree that AI stats need some boost to keep pace with the player, but I think the problem is not the AI, but the player. We are top aces compared to what was possible in reality. I'm in the 9th mission into the Kuban campaign with a Yak1B/Yak7 unit, one mission failed, but I have 12 ground kills from 4 ground attack sorties and 8 air kills from 4 escort sorties. And add to that that I'm a very bad pilot. Hungary in 1944 was a combat-rich environment, but the average figure is 1 kill per 10 sorties for ace pilots with over 20 kills. Now factor in pilots with lesser abilities, and it results in a guesstimate that whole squadrons had to fly 2 to 3 sorties in order to score one single victory. From this point of view, the performance of the AI is historically correct.
  11. Although Fraktur or Sütterlin was widely used in print, Antiqua was preferred wherever readability was important, e.g. in technical documents and on cockpit instruments. Moreover, pilot's notes are expected to be type-written, and most typewriters had Antiqua letters. Handwriting is a different story, it would be hardly legible for today's people.
  12. Thank you. I really appreciate your advice. In your opinion, can I use it as an argument when I reclaim my money back from 1CGS?
  13. One can enjoy a war movie on a small TV, but one cannot effectively play a war game on a small TV. One way to go is to scale up the contact, another way to go is to scale up the screen. You propose the latter (do you?), but not everyone is ready or willing or able to run the game on a 50+ inch display in 4K. So what? I repeat: We.Are.Seeing.Everything.Much.Smaller.Than.In.Real.Life. We.Are.Seeing.Everything.Much.Smaller.Than.In.Real.Life. IMO current scaling should be kept within gun-range (<400m) to preserve consistency with the piper. Beyond that mark, however, target size shouldn't shrink in a linear way, but apparent size should be observed instead, that is, a target at 600m should be slightly smaller than a target at 400m, but not as much smaller as it is now.
  14. You're right, I could have been more precise. I try again. We see what realistic spotting does to airplanes in the range beyond 1 km (unzoomed). Some argue it's realistic, but I'm here for flying, not for some pixel-hunting game. Now the problem is that whatever you do to overcome this problem, you need to substitute those "realistic" tiny pixels with a more apparent "placeholder". This placeholder image shouldn't be necessarily an icon, but it should behave as an icon in order to do the trick ("the main goal should be to get it more or less even for everybody regardless of their display and settings"). I guess the root of our problems is that the game is scaling down objects relative to the real world and our FOV, and not relative to our screen size. FOV change/zoom is there to compensate for this, but zoom comes at the price of losing SA. Ideally, there would be no zoom, but objects would appear as big as in RL regardless of screen size.* Anyway, I cross my fingers that the devs can find the sweet spot. * A 10 m object watched from 1000 m has the same apparent size (in degrees) as an 1 cm object watched from 100 cm. With displays smaller than 50" we can't have that mark even with full zoom.
  15. This translates to "icons for all' provided the contact is close enough. I'm for it, to be clear, but it's a sacrilege for the hard-core people.
×
×
  • Create New...