jaeger

Registered Users
  • Content count

    1,174
  • Joined

  • Last visited

Community Reputation

0 Green Tag

About jaeger

  • Rank
    Banned
  • Birthday
  1. Note the tracked S35 marked with yellow. I have observed this tank for around half an hour and there were so many impacts close to it that the bomb blast decals seen in the screenshot do not account for it in the least. Several 250 kg bombs hit <5 m next to it without taking it out.
  2. This happened after playing around 45 minues.
  3. The culling rate is not static so you will not notice it in irrelevant scenes with high frame rates.This can be seen/abused easily by panning a tank cannon back and forth with the most narrow FOV. Enemy units that are completely hidden by bushes can be spotted easily this way and an more and more players are definitely abusing this. This has even been talked about several times on open channels. This is a completely unacceptable in my opinion and not even offline games of lowest quality use lazy view frustum culling as a mean to save frame time.
  4. The low update rate for view frustum culling makes it possible to see enemy units behind foliage. Many players are abusing this.
  5. Moving the offsite server location from Dallas to Boston (this is not a problem nowadays) and using a true cold-potatoe carrier would chop the latencies for the many European customers down to 1/2. And then there is UDP of course...
  6. The difference with and without UI is 0.8 ms in a scene that takes 49.6 ms to render. I have measured this 10 minutes ago. You cannot differ these 0.8 ms so this is irrelevant. And this should not be done because it breaks a lot of gameplay.
  7. Yeah, you will improve the frame rate from 20.0 Hz to 20.1 Hz. You would be cut off from a good junk of game-play for no good reason.
  8. This scene is irrelevant. The UI doesn't make a difference in critical scenes.
  9. The problem is that while frame rates improve by 20% in a 10 ms scene where most players would do such tests (and they mostly are not even comparable), the difference in scenes which do matter (>30 ms) will be absolutely irrelevant because 90% of the frame time is spend on the CPU.So you will get crappy filtering for no relevant benefit.
  10. A full GPU load is not unusual and your measurement software is not reliable anyway.Those artifacts are a sign that your GPU or GPU memory gets too hot. Make sure that you do not use software that modifies the standard fan settings like rivatuner and so on.
  11. Driver options like different filtering methods do NOT improve the frame rate in scenes where you get less than 30 Hz unless you are using a low end GPU (Geforce 8600 or Radeon 2600) and high screen resolutions (>1.5 MP). People are VERY easily tricked to see improvements (in scenes which matter) even though there aren't any. -> Placebo
  12. You are using the lowest detail settings of the game already?
  13. Using vsync reduces the load on the GPU in most cases. Make sure to force vertical sync in setting in the driver.
  14. They are saying that the trees look more realistic in 1.30 and cost less performance.
  15. Did you notice a significant improvement when changing the game settings from max quality to max performance?