TOS Enterprise WIP

Discussion in 'Fan Art' started by blssdwlf, Apr 24, 2010.

  1. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    And that's how I know your analysis is flawed :techman:

    Perspective. It causes the widest part to be obscured. Did you even notice in Workbee's screenshot that the port saucer nav light was not at the widest visual point but behind and narrower than the widest point? That's going on for the whole ship.

    For those who may not be familiar with what perspective can do to obscure and obfuscate measurements of the widest point on curved objects, I've included a graphic below.

    [​IMG]
     
  2. Mytran

    Mytran Commodore Commodore

    Joined:
    Aug 5, 2009
    Location:
    North Wales
    A picture really is worth a thousand words, especially when we're dealing with really awkward compound curved shapes! Thanks for this :)
     
  3. Robert Comsol

    Robert Comsol Commodore Commodore

    Joined:
    Sep 10, 2012
    Location:
    USS Berlin
    Indeed it is. The two top graphics apparently resemble the shape of the torpedo bay when viewed from the stern / behind.

    Therefore I have linked front / bow view pictures and according to the lower two graphics (apparently topedo bay front / bow views) these clearly illustrate, like the screencap link Workbee provided, that the furthest objects appear narrower than they are when compared to objects in the foreground.

    Therefore the width of the dorsal is either measurable in relation to the torpedo bay launchers or the widest part of the dorsal is further away - and would therefore look much narrower than it actually is in relation to the launchers. Thus we'd be looking at 59% + but not less.

    [​IMG]

    Bob
     
    Last edited: Dec 14, 2013
  4. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    @Mytran - Thanks :)

    Actually, it's the 3D shapes and perspective that is throwing you a curve.

    The dorsal's shape makes it's maximum width visible from both front and back. The torpedo bay's shape OTOH makes it's maximum width obscured from both front and back. This will make simple attempts at a visual measurement inaccurate as the dorsal will be at it's max width while the torpedo bay is not. Thus 59% is not correct and it is less.

    Again, another picture hopefully worth even more than a 1,000 words :)

    Click to enlarge.
    [​IMG]

    As I've said before, you need to identify the true width of the torpedo bay either through an orthographic view of the TMP Enterprise or by a direct measurement. Your current methodology is flawed and inaccurate.
     
  5. kennysmith

    kennysmith Lieutenant Commander Red Shirt

    Joined:
    Jan 7, 2013
    Location:
    sacramento ca
    the question i have for you?, why are you so set on the refit of the movie star ship any way, don't you have any info of the tv show of the 1966s star ship any way?.
    did you ever look into the role playing game blue prints of the star ship from FASA. this is what i am looking for. i am trying to do a clean up of the game blue prints. so far the only thing i can tell you is this i just need the spaceing of the ribs of the ship. in meters, like mm, and m.

     
  6. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    Sorry Kennysmith I cannot help you. The refit Enterprise is relevant to my work in this thread and more importantly, this is my thread and NOT yours. I strongly recommend that you start your own thread is in this section and request the help that you need. Thanks!
     
  7. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    One more thing - did you bother to check this image? It is less than 59%.

    [​IMG]
     
  8. Albertese

    Albertese Commodore Commodore

    Joined:
    May 3, 2003
    Location:
    Portland, OR
    Sorry, Robert Comsol, I gotta go with blssdwlf on this one.

    --Alex
     
  9. Workbee

    Workbee Commander Red Shirt

    Joined:
    Sep 9, 2004
    Could this effect be minimized by finding a head on shot where the camera is pulled back far from the model? It seems to me that as the distance increases, the parallax (I hope is the right word) will become more like parallel lines. While still flawed, might provide an improvement in measure over a closer shot.

    Of course, even if this were an acceptable direction, finding such a shot is difficult. The first one I though of off the top of my head here does not quite seem perfectly centered. Plus I am not sure how to differentiate between shots where the camera has been pulled back from ones that were optically reduced in size. So in essence, I am providing an idea that still may not produce acceptable data, and may not even be actionable. :rolleyes: (rolleyes directed at myself). But I figure it couldn't hurt to throw out there.
     
  10. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    @Workbee - Thanks for looking but unfortunately it's not far enough back and/or with a high enough focal length. Theoretically we would need to find something like a photo from very far away with a telephoto lens set to the highest/longest focal length (like well over 500mm) to minimize the perspective.

    Because you can see the interior facing sides of the nacelles in your references screenshot it indicates that the VFX shot was made much closer and with a wider angle (shorter focal length) lens. It's compounded also by the small visible area where a couple of pixel difference could mean a big variation in the ratio.

    I've probably looked through every publicly available photo or video of the filmed TMP Enterprise and I haven't found the magic photo. I think the only way to be sure is to either:
    1. Measure it directly from the filming model
    2. Measure it from an orthographic rendering from the CGI TMP DE model
    3. Measure it from an extremely low distortion photo (long distance, shot with telephoto) of the filming model
     
  11. Workbee

    Workbee Commander Red Shirt

    Joined:
    Sep 9, 2004
    ^Yes, that makes sense about seeing the inside of the nacelles. Good catch.

    IIRC, somewhere there was a series of a lot of B&W pictures taken of the model just before filming it in TMP. I have looked and have been unable to find them again. I imagine you are already familiar with those pictures, but just in case maybe there might be a good picture? Though as I recall, most of those were up close to capture specific details -- so the opposite of what we would need. What would be ideal is a good picture straight down the neck before the saucer was attached. Assuming such a stage existed.

    I think someone will just have to track down the actual model with a set of calipers.
     
  12. QuinnTV

    QuinnTV Lieutenant Commander Red Shirt

    Joined:
    Jul 17, 2002
    Location:
    Philadelphia, PA
  13. Maurice

    Maurice Vice Admiral Admiral

    Joined:
    Oct 17, 2005
    Location:
    Walking distance from Starfleet HQ
  14. Robert Comsol

    Robert Comsol Commodore Commodore

    Joined:
    Sep 10, 2012
    Location:
    USS Berlin
    No need to be sorry. This is not about supporting a rationalization theory (I got one for later today regarding the interpretation of the TWOK torpedo bay footage).

    I would just like to remind that the issue that started this discussion was not "how to exactly measure the max. width of the connecting dorsal based on pictures" but "would the dorsal's width in an orthographic drawing be 48% of the torpedo bay launcher's width or rather 59%?"

    Let's have again a look at the latest wireframe graphics published.

    On the left side we have exclusively stern views of the dorsal while the two on the upper right are bow views. Unfortunately the width of the torpedo bay launcher as a reference point is 67 mm (on my screen) on the left and 6% less on the right side.

    All the three on the left are orthographic renderings while the three on the right are simulations of how they'd appear in pictures and the like with the inevitable distortions you get because objects further from the viewer will appear smaller / narrower than objects closer to the viewer (especially with a diagonal connecting dorsal in contrast to the "straight" conning tower of a submarine).

    However, since we are looking at a basic "ovoid" dorsal shape the edge of the dorsal's bow is too narrow to "hide" the wider dorsal areas behind.

    On my screen this orthographic rendering on the left shows the maximal width of the dorsal to be 37 mm, the "picture view simulation" on the top right takes the distortion into account and therefore yields only 29 mm (- 6% = 27 mm) where the dorsal rests on the torpedo bay because the maximal width of the dorsal is further from the viewer than the launchers of the torpedo bay - it therefore appears narrower than it is in relation to the observable width of the launchers.

    Conclusion: The max. width of the connecting dorsal part (that rests on top of the torpedo bay) in pictures appears narrower than it would be in real life compared to the observable real life width of the torpedo launcher (reference point for measuring)

    [​IMG]

    That the dorsal's width in the above picture near the torpedo bay roof measures "only" 56.7% (and not 59%) therefore doesn't come as a big surprise because the widest part of the dorsal is not near the bow and atop the torpedo bay launchers but further behind and at a greater distance from the viewer and/or camera.

    Bottom line: Regardless whether the max. dorsal width is 59% (my estimate) the width of the torpedo bay launcher or 56.7% it most definitely is not 48% as the CGI TMP DE model orthographic rendering wants to make us believe.

    [​IMG]

    @ QuinnTV & Maurice

    Great pictures gentlemen, thanks! The one Maurice provided will help me immensely trying to reconstruct a decent top view of the torpedo bay.

    But here is one from cloudster.com (hopefully back online soon again) that should be helpful, because it's a straight stern view of the TMP Enterprise's connecting dorsal.

    Of course, since the bottom part of the dorsal (the black "photon exhaust" area) is closer to the viewer than the (diagonal) dorsal's upper part one could decide to take the picture with a grain of salt.

    OTOH the upper part of the dorsal is rather close to the impulse engines and almost at the same viewing distance, so the width of the upper dorsal could be estimated or calculated by using the width of the impulse engines as a reference.
    IMHO it nevertheless gives us a good impression that the TMP Enterprise's dorsal is wider than previously assumed.

    Bob
     
  15. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    Bob, since there are other pictures that show a less than 59% ratio that would tell you that the methodology you used to arrive at 59% has a flaw in it.

    My wireframes are there to illustrate the complexities of these shapes and perspective distortion. I hope you took the time to understand why the visual measurement is difficult with just screenshots with short focal length and short distances.

    I don't think you realize why I picked the max width of the torpedo bay and the lowest part of the dorsal.

    The max width of the dorsal will generally be visible while the max width of the torpedo bay will generally be obscured due to perspective. The lowest portion of the dorsal happens to be closest to the torpedo bay and thus closest in relation since a frontal view will make the top portion of the dorsal appear wider since it is closer to the camera.

    This shows that the ratio is smaller than 59% and that depending on the focal length of the camera and the perspective distortion that number will fluctuate. 56% is just from this screenshot. If you do some more comparisons you'll see that number go up and down depending on the focal length and distance. Just a random sampling there is a photo from Mark Dickson/Mike Emery that is at 51.2% and I can find one in excess of 60%.

    Since that number can fluctuate then the methodology of attempting to measure it with the forward or stern perspective shots will be inherently inaccurate.
     
  16. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    Yes - I have those photos from cloudster and they're great reference shots. Unfortunately they all come with some form of distortion :(

    Thank you QuinnTV! There are a couple of tell-tale signs that it will be problematic to analyze though. It's shot from a close distance (you can see the warp pylons are not "edge on" towards the camera and angled outwards) and the shadow and graininess will make it a little challenging as to what is visible and what is not on the torpedo bay or dorsal.

    Thanks Maurice. I'll try and put something together for it.

    @Bob and anyone else - if you use this photo, keep in mind that on the torpedo bay, past the connection to the secondary hull the roof of the bay tapers down so from this angle it will appear that the bay is widening towards the bow but it in actuality is not. (The roof of the bay curves up in the middle too.)
     
  17. BK613

    BK613 Fleet Captain Fleet Captain

    Joined:
    Sep 3, 2008
    Location:
    BK613
    This page has some images that might also be useful (also has links to two other pages of photos):

    http://www.modelermagic.com/?p=16470

    There are still images up of the studio model at Christie's as well:

    http://www.mutara.net/Christies/EnterpriseA.html

    Fair warning: most of these images are huge (which is why I didn't link to any directly)

    Hope they help. My first instinct tho is that the neck, like an airplane wing, looks thinner than it actually is...
     
  18. Shaw

    Shaw Commodore Commodore

    Joined:
    Feb 21, 2007
    Location:
    Twin Cities
    Maybe this is a dumb question... but why don't you guys use the actual measurements of the studio model rather than attempting to get this info from images? Based on that I get 57.6%, but I'm not sure if there is a reason for "re-inventing the wheel" in this case that I am missing.
     
  19. Mytran

    Mytran Commodore Commodore

    Joined:
    Aug 5, 2009
    Location:
    North Wales
    This may sound equally dumb, but is there a definitive source for the official measurements?
     
  20. blssdwlf

    blssdwlf Commodore Commodore

    Joined:
    Feb 26, 2010
    Does anyone know who bought the big filming model?

    AFAIK, there are several possible sources:
    1. the big filming model
    2. the small filming model used for TUC?
    3. the CG TMP DE model