WIP: F-14 B/D
-
…Although everything does eventually get converted to quads, working in triangles makes it very hard to maintain a smooth shape for the rendering engine to shade.
You mean: “everything does eventually get converted to triangles” right? BMS doesn’t have any adaptive tessellation, does it?
Stringray: To smooth things out I usually try to select a patch of vertexes then w -> smooth. That might help with the ripples. Are you using the edge-split modifier? If so, are there any edges unintentionally marked sharp?
Also, might be a good idea to keep a super high-res mesh around in case BMS adds normal mapping at a later date (again, they don’t do that yet, do they?). Then the normal map could be baked from the high res mesh, then applied to the lower res one. Though I wonder if panel seams might be the best case for using normal maps in flight sims. Might even be able to get away with just using a monochrome bump map for those.
-
Yes! Exactly. It’s difficult to do manually, so you may end up having to more or less redo it, but you want an even grid of quads. Although everything does eventually get converted to quads, working in triangles makes it very hard to maintain a smooth shape for the rendering engine to shade. It’s especially noticeable with cylindrical shapes like a fuselage. Even the tiniest misalignment of a vertex results in the wrinkled look of misaligned normals! (I’ve learned this by screwing it up, myself.)
Obviously a simple model, but check out the clean grid-like quads of this model. It becomes more difficult to do when you have to consider things like the gunport and formation lights, but it’s still possible.
https://www.turbosquid.com/3d-models/f-14a-tomcat-wolfpack-max/779951I can see in this screenshot, for example, that there are visible seams between your faces around the gunport causing you additional problems. You’ll need to weld those vertices together.
https://www.flickr.com/photos/96028865@N04/32052152034/Again, I would recommend redoing that whole section of the mesh. It’s good that you haven’t started unwrapping/texturing yet, as this will be much harder to stomach once you’ve begun that process. The hardest part will be joining it to the canopy frame without distorting the topology too much. I recommend taking a cross-section near the intakes, and gradually extruding forward, refining the shape (as little changes as needed to match the basic profile of your current nose,) and keeping everything in quads.
It’s probably my bad because I’m too open and transparent with my WIP that sometimes people get the impression it’s the final product or that I missed something. I get what you’re saying, thanks for the detailed description. I am aware of the issue and will take care of it. Like I said it isn’t difficult at all especially when you know some shortcuts in Blender when it comes to aligning vertices, smoothing out lines and sections on a fuselage and so on so it’s really not a big deal.
Thing is there are many things unfinished or I haven’t decided on them yet. Like the seams you’re referring to. Up until recently I worked with a mirror. Now that doesn’t work when you have an asymmetric part like the gunport and cooling duct on the starboard side. So I disconnected that part and made it a separate object so it wouldn’t get mirrored hence the seam - can’t weld one vertice to another one that doesn’t exist right?
At some point I will put something up and invite each and every observer to tell me what they’re noticing or any inconsistency or issue however miniscule it might be that appears. That point is not now.
Don’t get me wrong I appreciate the input but now that I see in detail what you mean I realize lots of these things are on the todo list already or simply not an issue like a seam that’s not really there.
-
You mean: “everything does eventually get converted to triangles” right? BMS doesn’t have any adaptive tessellation, does it?
Stringray: To smooth things out I usually try to select a patch of vertexes then w -> smooth. That might help with the ripples. Are you using the edge-split modifier? If so, are there any edges unintentionally marked sharp?
Also, might be a good idea to keep a super high-res mesh around in case BMS adds normal mapping at a later date (again, they don’t do that yet, do they?). Then the normal map could be baked from the high res mesh, then applied to the lower res one. Though I wonder if panel seams might be the best case for using normal maps in flight sims. Might even be able to get away with just using a monochrome bump map for those.
Thanks man, you actually told me that before :D. Thing is I had this issue and solved it already on the wings and the glovebox. The way I did it was to basically create a multitude of horizontal lines usually along the x axis (straight or curve) and have them trace the shape I’m trying to model. Then I align the vertices along the lines and the result will always be aligned quads after aligning the vertices along corresponding vertical lines along the y axis. That approach never failed me and always takes care of ALL misaligned or bad topology.
I’ll do that on each and every surface before I move to Max anyway but now would be just too early to do it cuz half the geometry (exaggerating but you never know) might get changed anyway.
-
Kneel, Baby.
Tomcat does it seamlessly :D. There’s about 10% left in terms of room to collapse, have to check ref pics to see how far down it actually goes when attached to the catapult.
Started finalization of geometry and integration of all moving surfaces. Only group done so far is the landing gear doors. The surfaces have been completely smoothed out and welded, extruded in closed position to ensure fit and then disconnected as separate objects and assigned rotation axis and limits. It’s the only way I can open and close them at this point to finish the work on the model without losing my zero point positions every time I open something. Plus it’s a nice exercise in terms of 3D think :).
Eagle Eyes among you will notice the closed ‘cove door’ wrapping itself around the flap hinges on the main wing. All surfaces are at max so Flaps (main and aux, they’re two separate parts) at 35deg, spoilers at 55deg, slats at 17deg and airbrakes at 60deg. The eyebrow door has been connected to the flap but not yet made movable. I’ll build the wing insides and add some more detail and take some closeups of that next.
-
-
-
IDon’t get me wrong I appreciate the input but now that I see in detail what you mean I realize lots of these things are on the todo list already or simply not an issue like a seam that’s not really there.
No worries, just some input!
You mean: “everything does eventually get converted to triangles” right?
Yes.
-
No worries, just some input!
Yes.
All good, I’ll for sure get back to you once I refine the front fuselage section. The tricky thing is as you pointed out stuff like the gunport, cooling ducts (NACA) and anti collision lights. If you have any input as to how to cleverly arrange the geometry in order to avoid the distortions when rendering it’d be much appreciated! The port side as you can see in the render is not really an issue but starboard is very messy :).
-
Finally getting to ‘recycle’ some of the parts I modeled a loooong while ago such as the spoiler hinges and some other parts. Flap hinges will follow once I got the eyebrow door separated.
I modeled the wing in such a way that all three ‘Tomcat specialties’ will be possible for animation
-
the cove door, rotates up to facilitate airflow between main wing and flap when flaps are fully down, in that case the spoilers actually move down up to 4 degrees to meet the cove door and create a seamless surface between wing and flap
-
the eyebrow door, also a sealing surface, when flaps are up it rotates upward to meet the spoilers and fixed parts of the wing as to close the gap between the hard spoiler edge and the downward sloping forward flap area, when flaps deploy it rotates fully down to build the counterpart to the spoiler/cove door assembly and create a seamless surface along the forward flap edges
-
inside spoiler acting as airbrake, in RL when the airbrake is deployed spoilers 3 and 4 (inside) go up to their max 55deg position, therefore the spoilers are all independently movable but paired in terms of 3/4 and 1/2
Lots of details and fine tuning to do but now that I have the rotation limits set actually modeling as in smoothing out and arranging geometry is simply a question of time :D.
-
-
-
A flap, well actually two flaps - aux and main. They mostly move together but not always.
Above we now have two separate flaps, two separate eyebrow doors, flap hinges for the hydraulic actuators and the base attachment points. Alignment and complete geometry unfinished. I guess in the end I’ll have two merged flaps, separate objects for hinges, eyebrow doors and holding structures to keep tris count down.
Main wing slat area adjusted to RL shape. I’ll add some minimal structure to the slat itself and the hinges to keep the balance between realistic look and minimum geometry added.
And corrected shape with accurate colors, the real paint scheme has a small grey stripe on top as to not have the ‘danger red’ shimmer through when the slat is fully retracted.
The correct animation of the slat will be postponed to 3ds because I’ll need one translation and one transformation DOF (as Pumpyhead has remarked like 30 pages back) in order to mimic the movement forward and down. Right now the slat is rotated which in RL it is not. In RL it rolls off a rail on circular hinges thereby moving it down and forward. This would be too complex to implement in Blender without using an armature which would be pointless.
-
Absolutelly
-
@SEG:
Absolutelly
Believe me, that’s my favorite pic and probably most used. Hail to Thorsten at M.A.T.S.!
-
In case you would integrate full moving surfaces laws :
The two outboard flaps sections are the main flaps utilized during both modes of operation (10°). The inboard section (auxiliary flap) is commanded only during takeoff or landing.
The slats consist of two sections per wing moving at the same time.
For roll control below 57 deg, the F-14 uses spoilers in conjunction with its all-moving, swept tailplanes, which are operated differentially; above 57-deg sweep, the tailplanes operate alone.
For unswept, low-speed combat maneuvering (20° sweep), the outer 2 sections of trailing edge flaps can be deployed at 10 deg and the nearly full-span leading-edge slats are drooped to 8.5 deg.
Again and again thank’s for you Huuuuuuuge work -
@SEG:
In case you would integrate full moving surfaces laws :
The two outboard flaps sections…slats consist of two sections per wing…moving at the same time.
Again and again thank’s for you Huuuuuuuge workEverything you describe can be implemented except the systems overlay in terms of spoilers for example. AFAIK it would not be possible to assign a dual function to the inside spoilers for example. Acting in certain conditions along the envelope as spoilers and then as a fixed aux part of the airbrake when that is deployed. Apart from that it’s all possible including the dual flap depending on the conditions you mentioned. By the way the spoilers are operable throughout the entire wing sweep regime as they add roll control when wings are fully swept since the tailplanes cannot operate fully as you stated. Just saying they’re always active in some sort of capacity.
As far as the physically split parts of slats and flaps I’m aware of the construction BUT it would be unfeasible to model that. I have textures to give it a realistic look and you wouldn’t be able to see a mesh separation anyway as they always move at the same time.
Thanks for the detailed explanation!
-
stingray I believe we told you already… you are sick m8… :lol:
Don’t change your prescription… you are on the right direction… :lol:
So you are done already or more details need to go in before you switch to 3ds max and start Falcon working on this puppy?
Really are there any other details to go in? :lol:RESPECT man!!!
-
stingray I believe we told you already… you are sick m8… [emoji38]
Don’t change your prescription… you are on the right direction… [emoji38]
So you are done already or more details need to go in before you switch to 3ds max and start Falcon working on this puppy?
Really are there any other details to go in? [emoji38]RESPECT man!!!
Good questions, I guest at this point it’s a really short list:
- intake ramps and connection to air outlet
- mesh optimization
- separating and naming some objects (important for 3ds as that info is also imported)
- refueling probe and doors
- tail hook detail and assembly
- AC and position lights on the main wings
- crew steps and ladder
- RIO ejection seat rails
- simplistic actuators for all moving parts (landing gear doors, airbrake, flaps, etc)
- landing gear details and some simplification
The tex mapping can also be exported to 3ds not sure where I’ll do that yet.
Sent from my STV100-4 using Tapatalk
-
For the textures and materials to work on may I suggest substance painter…
It has a magnificent database to create the basic materials. Best I’ve seen for metals.
Also it’s magical for the wearing and dirty parts…U can paint on the 3d model no need to back and forth with Photoshop.
It has layers and everything needed.
Sure it ain’t free #cough# #cough# but it’s exceptional.
About texture mapping I don’t know if Blender uv mapping is compatible with 3ds max.
You can do a small separate part in blender and test it in 3ds max.
-
For the textures and materials to work on may I suggest substance painter…
It has a magnificent database to create the basic materials. Best I’ve seen for metals.
Also it’s magical for the wearing and dirty parts…U can paint on the 3d model no need to back and forth with Photoshop.
It has layers and everything needed.
Sure it ain’t free #cough# #cough# but it’s exceptional.
About texture mapping I don’t know if Blender uv mapping is compatible with 3ds max.
You can do a small separate part in blender and test it in 3ds max.
I actually have a painter like that (may even be substance painter) that works as the one you describe in which you ‘paint’ the material on the 3D surface in PS and it automates the tex mapping. Those programs are f***ing awesome!! Not sure how accessible or applicable they are here.
As for the Blender UV mapping I’ve tested it already and it transfers everything lossless if you apply some tricks when exporting from Blender.
The only thing that 3ds is really necessary for is the DOFs and material types for export to BMS AFAIK.
-
Acting in certain conditions along the envelope as spoilers and then as a fixed aux part of the airbrake when that is deployed.
This should be possible as well if you are talking about spoilers as airbrakes on landing. I believe it is DOF 7. I will have to check one of the heavies that have it (DC-10, E-3, E-8, KC-135).