Cars 1972 Ferrari 365 GTB/4 Daytona Spyder

Some progress on the dash, lots to be simplified but general shape stands. Tricky detail is that almost all dashboards are restored differently with some having thicker leather or stitching in different places. I'm aiming for some sort of middle ground here.

dash1

dash_pers_top

The edge won't be visible later on, top side will connect to the windshield frame and sides to the inside of the front fender.

dash_edge_delete

dash_WIP5

dash2

With frame...

dash_frame4

With cockpit...

dash front2

With car...

dash_outside2
 
giphy.gif
 
Great to see it in perspective !

Rear axle seems to be a few cm too much forward, but hard to be sure without doors line. I wouldn't go too far with interior, before being as much confident with exterior shape as possible. But this car has rather simple shape, so perhaps it is not the case.

Would never ever rely on those drawings found on internet aka "blueprints" again, learned my lesson with Jaguar. One of many things which was wrong with the blueprints I have had, was that rear end was terribly too short, and rear axle was 3.5cm too much forward.
 
Great to see it in perspective !

Rear axle seems to be a few cm too much forward, but hard to be sure without doors line. I wouldn't go too far with interior, before being as much confident with exterior shape as possible. But this car has rather simple shape, so perhaps it is not the case.

Would never ever rely on those drawings found on internet aka "blueprints" again, learned my lesson with Jaguar. One of many things which was wrong with the blueprints I have had, was that rear end was terribly too short, and rear axle was 3.5cm too much forward.

Yeah I know about the blueprint issue, it's always a gamble and usually gets worse the further back you go. If you look at the blueprints for something like an Audi R8...they're perfect. You can build the car completely from that.

I usually use pictures with favorable angles and try to get as close as possible while eye balling the final shape :)









Based on all that I actually ended up aligning this specific blueprint of the profile and I must say it is a very nice example of an accurate blueprint:



Then again proving our point here is the top view of the same print:



As far as the external shape vs the interior, I finalized all body lines that are in touch with the interior including the door frames, top casing, rocker panels and windshield frame and so far it aligns pretty nicely so I don't expect any major corrections.
 
Great to see it in perspective !

Rear axle seems to be a few cm too much forward, but hard to be sure without doors line. I wouldn't go too far with interior, before being as much confident with exterior shape as possible. But this car has rather simple shape, so perhaps it is not the case.

Would never ever rely on those drawings found on internet aka "blueprints" again, learned my lesson with Jaguar. One of many things which was wrong with the blueprints I have had, was that rear end was terribly too short, and rear axle was 3.5cm too much forward.

One thing to remember, too, is that when you look at a computer screen most things look a bit "wrong" because it's 2D. Our binoculuar vision with horizontal offset makes things in real life look a fair bit wider (depth perception helps, too).

That's why it's never a good idea to design something 3D purely on a computer screen. When the actual product is realized in real life it often will look odd.
 
Blender needs VR support (or any other 3D modelling software) :D

It would help. I'm working on a track right now, and this is extremely big thing for a track. Haven't thought about cars, it must be very significant too. However, those reference pictures and blueprint are 2D too, blueprints are also orthographic. Most confusing thing is the distortion of those reference pictures.
 
Blender needs VR support (or any other 3D modelling software) :D

It would help. I'm working on a track right now, and this is extremely big thing for a track. Haven't thought about cars, it must be very significant too. However, those reference pictures and blueprint are 2D too, blueprints are also orthographic. Most confusing thing is the distortion of those reference pictures.

Yea, reference pictures can be misleading as well due to lens distortions - modeling programs' perspective is actually quite primitive in that most use straight lines for their perspective projection. That doesn't jive well with a lot of photographic reference imagery because it gets lens distortion which curves the perspective projection.


VR in modelling could be very valuable. My folks were in town this weekend so my Dad finally got to try my Rift for the first time. He said he finally understood why he kept flying off Paddock Hill Bend at Brands because he hadn't realized how steep it was before. The 3D binocular vision of the Rift really helps with depth perception and heights like that, which photos and videos always fail to convey (and by extension, 2D views on a monitor screen as well).
 
Last edited:
No 365 Daytona calls in the 70th Anniversary poll so far and probably none when it actually goes live so I guess I'll continue working on this :).

Interesting as this is the LAST Ferrari designed and built under the original management before Fiat 'took over'.

@Ryno917 & @mantasisg

You guys make some great points and most of them are probably spot on but you gotta keep in mind that people have been able to build extremely accurate 3D models - that is when it looks right it is right - for decades with 2D screens and also with blueprints. Car manufacturers design their cars that way, some of the most renowned studios in the world like Polyphony (talking about modeling NOT physics or anything) build their models that way so we should probably take it easy on that front. It is absolutely possible to model accurately with close to no loss using old tech, I know some 3d printer guys who have put literally blueprint to 3D translated designs of car parts (including body panels) through the machine and you wouldn't be able to tell the result from the original part.

Anyway interesting discussion. I'm overwhelmed with classic modeling, seeing all that in VR would prolly kill me :).
 
Real blueprints are one thing, but the simplistic drawings that are circulated online as "blueprints" are often something else entirely.

Glad to see this one back on, though! It's shaping up wonderfully.
 
Yea, reference pictures can be misleading as well due to lens distortions - modeling programs' perspective is actually quite primitive in that most use straight lines for their perspective projection. That doesn't jive well with a lot of photographic reference imagery because it gets lens distortion which curves the perspective projection.

...which can be mitigated by knowing the camera model and adjusting for lens distortion in sth like Photoshop before using the images as reference. IMHO this is an artistic endeavor and you need the passion for it to come out alright. If you don't have that no reference or technical accuracy in the world is gonna help you create sth that watchful observers who know the real thing will say 'feels right'.
 
Well I was looking forward for more updates on the 365 but ok, while we're at this other subject I must agree with the previous post and add that there are more than enough tools for us (3d modelers) to make anything we want, is only a matter of knowing how and what to use.
Sometimes those car drawings can be useful to help get an idea of the overall proportions, or define a certain panel. I even use them on top of real pictures to get the best out of both references, and make my own "blueprints" from them.
In my experience so far, nothing beats camera matching. If it matches the real pics, then you're set :)
 

Latest News

Are you buying car setups?

  • Yes

  • No


Results are only viewable after voting.
Back
Top