Syndrome Posted December 5, 2019 Share Posted December 5, 2019 (edited) Was watching how fast the newest DCS plane can move its stabs and it made me wonder what it is about the F-16 that feels a bit sluggish sometimes, and it turns out the Viper's stabs are moving about twice as slow as they do IRL. Going from level to full stick back takes 0.6 seconds and nearly 2 seconds make a full sweep stabilator sweep cycle. Whereas the real Viper takes 0.3 seconds for a full stick back deflection and about 1.3 seconds to make a full stabilator sweep. You'll notice that the real stabilators are moving so fast that they snap crisply at the end of the range, where the game ones slow to a stop and then return as if pivoting in molasses. I am guessing this might be what some people are referring to when they say that the flight model of the F-16 takes too long to reach 9g in a turn and feels delayed and somewhat disconnected from inputs. For clarity, this has nothing to do with input lag or curves. I had my curves set to a step-wise binary deflection so there was zero input delay. Eg all y values above 50 counted as 100 on the "curve" and all values below 50 counted as zero. EDIT: Updated the video with stopwatch timers to show the quantitative difference. Measured in 60ths of a second. Eg 0:22 = 0.37 seconds, and 0:38 = 0.63 seconds. If you slow down the playback through Youtube settings, you'll see that the real F-16 maintains the same speed of rotation throughout the range of motion, whereas the DCS F-16 stabs slow down as if fighting a spring that provides more resistance at the top of the motion creating the illusion of a "pause" but the last couple of degrees of motion takes a huge fraction of the time and is still moving a few pixels per fraction of a second right up until the 0:38 marker. If it was higher resolution, you'd see it moving past that up to 0:40, but the youtube version is downsampled so I just showed the timer marking showing motion that was easily visible to all. I am also including images of the binary stepwise curves used so you can the slow down at the end isn't from input lag on the joystick. Edited December 6, 2019 by Syndrome Link to comment Share on other sites More sharing options...
Hummingbird Posted December 5, 2019 Share Posted December 5, 2019 I noticed the same, and I'm suspecting it's born from an error in coding the DFLCS control laws. Link to comment Share on other sites More sharing options...
ED Team cofcorpse Posted December 5, 2019 ED Team Share Posted December 5, 2019 I noticed the same, and I'm suspecting it's born from an error in coding the DFLCS control laws. Do you know the error or is it just hypothesis? In general, there could be several reasons of such behaviour, if it really exists. I think, we will check and make corrections if needed. Thank you for report! Link to comment Share on other sites More sharing options...
sobek Posted December 5, 2019 Share Posted December 5, 2019 Watching the video my impression is that the actuation rate is rather correct, it's just that the pause between switching from full pos to full neg is longer than IRL. Good, fast, cheap. Choose any two. Come let's eat grandpa! Use punctuation, save lives! Link to comment Share on other sites More sharing options...
Syndrome Posted December 5, 2019 Author Share Posted December 5, 2019 Watching the video my impression is that the actuation rate is rather correct, it's just that the pause between switching from full pos to full neg is longer than IRL. There is no pause between switching. The switch in input was nearly instantaneous because the curves were set to "binary" either 100 or 0 with a single point of neutral deadzone in the center of the stick @ 50. The delay you are noticing is the built in lag caused by the game slowing the stabilators down to half their real life speed so that when it reverses direction it looks like a "pause". This is an optical illusion caused by the slower movement of the stabilators. Try it yourself by the setting the curves to a binary step wise function and going through the full motion, there is no pause. It's impossible to recreate the real life snapping at the end of the movement range because the stabilators are moving twice as slow. Link to comment Share on other sites More sharing options...
Syndrome Posted December 6, 2019 Author Share Posted December 6, 2019 (edited) Do you know the error or is it just hypothesis? In general, there could be several reasons of such behaviour, if it really exists. I think, we will check and make corrections if needed. Thank you for report! Thank you for checking this out. Much appreciated. I updated the video with stopwatch timers to show the quantitative difference in rotation speed. Measured in 60ths of a second. Eg 0:22 = 0.37 seconds, and 0:38 = 0.63 seconds. If you slow down the playback through Youtube settings, you'll see that the real F-16 maintains the same speed of rotation throughout the range of motion, whereas the DCS F-16 stabs slow down as if fighting a spring that provides more resistance at the top of the motion creating the illusion of a "pause" but the last couple of degrees of motion takes a huge fraction of the time and is still moving a few pixels per fraction of a second right up until the 0:38 marker. If it was higher resolution, you'd see it moving past that up to 0:40, but the youtube version is downsampled so I just showed the timer marking showing motion that should be easily visible to all. Edited December 6, 2019 by Syndrome Link to comment Share on other sites More sharing options...
sobek Posted December 7, 2019 Share Posted December 7, 2019 the DCS F-16 stabs slow down as if fighting a spring that provides more resistance at the top of the motion creating the illusion of a "pause" but the last couple of degrees of motion takes a huge fraction of the time and is still moving a few pixels per fraction of a second right up until the 0:38 marker. Fair point. Good, fast, cheap. Choose any two. Come let's eat grandpa! Use punctuation, save lives! Link to comment Share on other sites More sharing options...
Recommended Posts