P-47 Progress Thread - Page 4
Goal amount for this year: 8000 USD, Received: 2,185.00 USD (27%)
Welcome Back, Unregistered     Today's Date:       Current Time:

Please Support the SOH "2023 Charter Membership Drive"

All online donations at this time are through PayPal

Snailmail donations can still be sent, see this thread

Alternate Donations

Accepted Payment Types

1. Originally Posted by sdsbolt
i suspect the SDK is not correct. I think frame 100 is use for full flap which is 40 degrees in your case. ...
I just realised that there is a flap gauge ingame: When pressing F5, the HUD displays the % of flap extension!

This leads me to believe that the SDK is right but that the info only applies to pure rotational animation names (stock application of "l_flap", in this case). When using key frames ("l_flap_key") in the aircraft.cfg, the key frames are read as a percentage of flap extension in the HUD, keyframe 0 being 0% flaps and key frame 100 giving 100% flaps. This accounts for the altered flight behaviour, as key frame 68 equals 68% of flap extension, while the mesh only shows 10% of flap rotation. I don't know how to correct this, but Daniel might have a point in suggesting using the scalars. This, unfortunately, is beyond my knowledge.

I'll review my animations but in the mean time, I am open to suggestions to solve this problem.

Right now, this is the current state of affairs:
key frame 68 = 10° of physical rotation of flap mesh = 68% of flap extension ingame -> should be equal to 10° of flap (flight behaviour)
key frame 93 = 20° of physical rotation of flap mesh = 93% of flap extension ingame -> should be equal to 20° of flap (flight behaviour)
key frame 97 = 30° of physical rotation of flap mesh = 97% of flap extension ingame -> should be equal to 30° of flap (flight behaviour)
key frame 100 = 40° of physical rotation of flap mesh = 100% of flap extension ingame -> should be equal to 40° of flap (flight behaviour)

Maybe there's something I can do with the naming of parts or the hierarchy...

2. I'm probably just displaying my ignorance here, but perhaps having a dummy part using the l_flap tag could be added in a way that is tied to the keyframe somehow?

3. That gave me an an idea, Daniel.

And I think I might have been an *ss : I realised that I got here by using the key frame number distribution taken from the first P-47 model I created . That was key frame 0, 25, 50 and 100 to give 0°, 20°, 30° and 40° of flap extension. With this model I started of using the same 25 key frames distribution and added some extra key frames to keep the geometry of the system intact. But there's absolutely no reason to use this distribution of 25 frames. I can also animate the system with key frame 10 giving 10° flap extension, key frame 20 to give 20° etc. The speed at which the mesh moves between the 4 flap settings may differ but that might be acceptable.

I'll try this and report back later!

4. I'm certainly no genius... .
The basic thought of dividing 100% of flap extension into four settings with 25 key frames each was alright after all, I just had to start with the 10°, 20°, 30° and 40° rotation of the flap mesh and then match the mechanics of the extension system to those positions - not the other way around, duh... Inverse kinematics, anyone?

I say: "Done!"

5. Yep. That's exactly how I remember it being done. Any of the animations in gmax must be 0-100.

Flight model, the airfile, has flap entries in it, for the total degrees, how many positions, and the drag effect at 100% (40 degrees).

6. Glad you got it figured. I know I'll enjoy watching the intricacies of the mechanism in motion!

7. Originally Posted by crossram
That's exactly how I remember it being done. Any of the animations in gmax must be 0-100.
For external animations, yes, mostly. In the cockpit you need to check the SDK for the many exceptions to that rule.

Flaps can depend on the game engine using l_flap or r_flap but this only suits flaps that pivot on a fixed axis and full flaps is at keyframe 100. Frosty's P-47 flap animation will use l_flap_key etc and each keyframe corresponds to one degree of flap deployment according to the SDK so 40° will be at keyframe 40. The actual flap positions are then defined in the config, so I'm not sure the actual P-47 operating procedure can be followed in CFS3?

Animations are fun and really satisfying when you get them right and these look really good!

8. Aw, really Tom?

And I thought I had figured it out! Where in the SDK did you find that each keyframe represents 1° of flap? Wouldn't that also mean that it is possible to create 100° of flap? Then again, I can still use the original aircraft.cfg settings of 10, 20, 30 and 40. So my first (or was it second?) thought of using key frame 10, 20, 30 and 40 wasn't so bad after all .

Anyway, this time I can adjust the animations more easily by moving the key frames to their correct positions (from 25, 50, 75 and 100 to 10, 20, 30 and 40)

PS, has something happened to the forum software? I tried to use the 'quote' button, but no quote appears in the reply field? Something similar happens when I click the 'Preview'-button: the preview does appear but the reply field becomes blank so I can't edit the message again before clicking 'Submit Reply'...

9. Yes, I’m having the same bother quoting, usually when the site’s busy.

The flap keyframe info is in the table of keyframe animation names. You could I s’pose animate 100 degrees of flap - the Spit does almost 90 and in a hard up/down option after all. Oh, if l_flap​ and such are used, all you do is align the flap’s pivot so the X axis is along the hinge line and the built-in animations do the rest with limits from the config, so ignore the waffle about 100 keyframes for that. It’s been a heavy week.

10. Quote not working, I guess.

After all these years, remembering some of this stuff.

11. ## Boost?

I kinda recall, when I made the Ta152 H1 (beta), in it's limit, I modeled in boost. Off the top of my head, don't remember all the details, all in the airfile, I think.

Don't have any files/programs on this PC. And, unfortunately don't remember if it was included in the last upload. I do remember, it did work, though.

Mesh is looking great! It all looks great.

12. Just wanna throw in my "wow" for this amazing project...
cheers

13. Just to let you all know I'm still working at it; Playing around with pylons and loadouts:

14. Good work Joost. Now with social distancing you will be able to spend more time on it!

15. Beautiful work Joost!

16. Originally Posted by mongoose
Good work Joost. Now with social distancing you will be able to spend more time on it!
You (and I) wish!
I'm still expected to appear 5 days a week at work! But with teamsports and all other social gatherings dead in the water, the evenings do offer more opportunities for modeling.
The main parts are nearly done, so it now is mostly a matter of finetuning and adding eyecandy.

17. Glad to see you decrypted the Pylon Code. Binary is such fun to get your head around!

18. Really looking great ! Can't wait for this one.

19. ## err

Prop hub ain't turning...

20. I thought about that but then wondered if it was due to the specific RPM?

21. Originally Posted by crossram
Prop hub ain't turning...
Hi Ted,

That's what you get when looking at a still from the sim... .
But I understand your remark. It is caused by the way the two parts are modelled: The prop disc is a semi-transparent 2D plane painted to simulate spinning blades (as with all/most CFS3 props - but I tried to give it some 3D-ness when looking from aside) and the hub is a true 3D model. I can assure you that it is turning ingame but not with a speed that turns it into a similar semi-transparent disc. I tried modelling it like that but it looked totally unrealistic. The 'sides' of the hub where the blades are attached, will unfortunately be visible but the solution seen most is to texture that area the colour of the prop blades (black).

Until now, that really was an acceptable compromise but you just gave me an idea that might be worth investigating: CFS3 prop models have 3 stages: still, slow and blurred. Trouble is that the transition from one to the other is rather sudden. This makes any changes in the 3D hub model between the stages also quite obvious, a bit like LOD popping. But I could try making the hub slightly transparent at the blurred stage... The blade attachment points will still be visible but perhaps less conspicuous than they are now.

In the mean time I suggest looking at some internet piccies of spinning P-47 props; you will see that depending on the shutter speed, the hub is often better visible than the blades (speed is higher at the edges of a spinning circle than at the center). Oh, and YouTube doesn't count! The FPS does really weird things to spinning props or wheels.

22. Nice Joost.

I'd love to see an improvement on spinning props in cfs3

23. Wow, nice!
Neat to see it parked on your PSP

24. Originally Posted by Frosty
... But I could try making the hub slightly transparent at the blurred stage... The blade attachment points will still be visible but perhaps less conspicuous than they are now.
Originally Posted by sdsbolt
Nice Joost.

I'd love to see an improvement on spinning props in cfs3

Yeah, well, that didn't work...
In fact, it was worse that the current setup. It gave sort of Z-order problems with the hub disappearing on my PSP etc. I think I'll leave it as-is, but the spinning prop itself will get some 3d-ness.

But just for fun and as a diversion, I have textured the wheels. Bump mapping is awesome!! It's so cool to have '3D' treads without having to model them. It really brings this model to life. Can't wait to start using that on the rest of the external model! Thanks so much, Ankor!! BTW, it also works for weapons etc. so now I have real iron bombs

I have also finished modeling and animating the oil cooler doors (+ interior), the intercooler exhaust doors (+ interior), most of the landing light and today, I finally modeled the dorsal fin. I now have a fuselage with a dorsal fin and one without. This offers the possibility of doing a D-30 and a D-40 (or even a D-30 retrofitted with a dorsal fin). Cockpits will differ accordingly.

I'll keep you posted.

25. Yea Joost the bump mapping is a great addition, opens up a lot of neat options!