Joystick "Slope" settings in FSUIPC4

[chinookmark]

I've been playing around with the Slope settings in the joystick calibration menu of FSUIPC4, and I think I've stumbled on a problem.

The positive slope values respond fine, but the negative values seem to be reversed. A slope setting of -2 has the most extreme effect on the controls, while a setting of -15 has the mildest effect. Even stranger, a setting of -1 responds more like the positive side of the slope curve. At first I thought it was my imagination, but the input and output values confirmed this during some semi-empirical testing.

Could you tell me if I have this installed incorrectly, or perhaps I'm interpreting this incorrectly? Thanks

FSX/Acceleration

FSUIPC4 v4.40

Vista 64

Saitek X52 pro and pro pedals

[JSkorna]

Hi,

Pete will chime in here, but if he programmed things according to mathematical rules, -2 is greater than -15 so it should have a larger effect. Sometimes we all forget what negative numbers really mean!!

Thanks!

[Pete Dowson]

I

The positive slope values respond fine, but the negative values seem to be reversed. A slope setting of -2 has the most extreme effect on the controls, while a setting of -15 has the mildest effect. Even stranger, a setting of -1 responds more like the positive side of the slope curve. At first I thought it was my imagination, but the input and output values confirmed this during some semi-empirical testing.

Could you tell me if I have this installed incorrectly, or perhaps I'm interpreting this incorrectly? Thanks

Why are numbers bothering you? Use the slider on the right and view the fraphic, not the number! Slide it up and down. At 15 (with the slider at minimum, lowest position) you have the FLATTEST centre response. i.e very slow centre response. As you slide up the curve flattens, at 0 is it straight -- a 45% slope from one corner to the other, a perfectly linear response.

The negative values then make the centre stepper and steeper, nearer vertical as it gets to the top of the slider at -15. At that setting the central response is to fast and the response at the extremes slow. The -ve values are almost the exact inverse of the +ve values.

for a control with no centre, like the single throttle, only the upper half of the S-curve is used, but the principle is the same -- the steepness is at the low end, from idle.

Please, try to understand it by the shape of the curve rather than by a nominal and otherwise meaningless reference number!

Pete

[chinookmark]

That's what I'm sayingthe numbers don't match the picture. I want my joystick to be more responsive around the center, so I use a negative value. I move the slider to -2, and the picture shows just a small curve, but the aircraft are hypersenstive, and the numbers confirm this. If I set the slider all the way to -15, the picture shows a dramatic curve, but the aircraft responds much more mellow, closer to default response, and the numbers also jive with this. I'm not concerned with the setting number, ie -2 or -15, it just seems the response is opposite what the picture shows. I'm happy with a setting of -12, now that I know what it's doing.

It was frustrating at first, because I was looking for a small adjustment from default. One click (-1) made little difference, but one more click (-2) made the aircraft almost unflyable. It wasn't until I experimented so I could tell you the exact problem I was having, that I found I needed a lot of clicks (-12 to -15) to get the small change I was looking for.

But I also discovered one click (-1) has the OPPOSITE response, and deadens the stick near center. Again, I'm not concerned with the setting number, just using them as reference points. And I'm not complaining --I've found a setting that works for me, just bringing your attention to something that doesn't make sense to a less computer saavy user like myself. :)

[Pete Dowson]

I want my joystick to be more responsive around the center, so I use a negative value. I move the slider to -2, and the picture shows just a small curve, but the aircraft are hypersenstive, and the numbers confirm this.

Ah, so it isn't the graphics you are saying show anything wrong, but only the results? I didn't understand that from your original post.

The actual look-up tables for the slopes are generated once, when the program first loads, then used by the index. Maybe the indexing for the -ve numbers is wrong. I'll check it out here. If it is wrong it has been wrong for about 8 years.

I suspect no one before you has ever wanted to make the response more severe than default. Most folks want more precision, for delicate adjustments and precise flying, near the centre, so the usual curves are those with flattened centres.

One more question before I go checking this myself. You DO have the null zone, in FS, set to minimum, don't you? Otherwise the middle part of the curve, the bit you are talking about, isn't really having any effect in any case, and it's the shape of the tops and tails which you are experiencing. That will be quite misleading. For all FSUIPC calibrating operations, if you are assigning in FS, you need minimum dead zone and maximum sensitivity, else it is mostly a waste of time. ;-)

Regards

Pete

[Pete Dowson]

I'll check it out here. If it is wrong it has been wrong for about 8 years.

You are right! It IS wrong, and has been wrong for, well, for all of FSUIPC's 9 year life, or at least since I added Slopes which was many years ago!

The bitmaps for the display are generated okay, as are the look-up tables for the conversion of IN values to the OUT values, but the indexing of the one is done differently to the indexing to the other -- the negative values are indexing backwards! Huh, shows how many folks don't use that area! ;-)

This is a very good catch. I'm fixing it now. It will mean, unfortunately, that anyone who is already using the -ve slopes will have to re-select the 'correct' one now. If you want to test the interim version with this fixed, please email me at petedowson@btconnect.com.

Thanks & Regards

Pete

[chinookmark]

Thanks Pete! Sorry for taking so long to get back.

I know it sounds odd to want a more "severe" response near center, but I have two reasons: One, a trick I learned flying RC helicopters, is that if it's numb near the center, while precision hovering is a little easier, it is also easier to overcorrect when moving the cyclic farther and digging into the aggressive part of the curve. The second reason is that my X52 seems to have an odd response itself. Comparing the joystick to the visual cyclic/yoke in the sim, moving the joystick has a very small effect until the extreme edges of joystick travel, when the sim cyclic jumps to catch up. It's like it has a lot of built in smoothing, and using the aggressive slope in FSUIPC counters this.

[headley]

I'm fixing it now.

Can you tell us when the fix will be available?

Will it be an update or interim version or are you looking for a new "full" version?

Edit : OK please ignore, i was to quick and not read the pdf providing the changes. Thanks Pete :)

[Pete Dowson]

I'm fixing it now.

Can you tell us when the fix will be available?

Will it be an update or interim version or are you looking for a new "full" version?

Edit : OK please ignore, i was to quick and not read the pdf providing the changes. Thanks Pete :)

Groan!

Pete