Author Topic: Expert's Analysis Examined  (Read 8849 times)

0 Members and 1 Guest are viewing this topic.

Offline tomi

  • Hero Dronie #2
  • ********
  • Posts: 668
  • Karma: +37/-26
Expert's Analysis Examined
« on: June 10, 2008, 06:48:19 PM »
Looking at the example that was recently posted on this web page by someone named "SecretWeb"..
http://www.abovetopsecret.com/forum/thread289007/pg316

We have here an obvious mistake in perception. Take the Raj pic15 and enlarge the drone considerably and study the arm with these "misaligned" bolts. 

What you will see is quite amazing.  And for the sake of discovering this, I think he has in a way really contributed to the amazing complexity of this drone. 

The writing along the arm and the perspective of the arms termination show that this "arm" wing or whatever it is, is actually shaped in a similiar way to an expanding airplane wing. It obviously has a curvature and this allows the ever so slight extended right width.  The whole wing or arm has a wider right side than it does left from where you would normally spot as the center length. 

You can see how it curves a bit lower on the circular base then it extends itself a little wider to the right all along it's length and actually this extra width provides a lift effect.  Very amazing!  Thanks whoever you are. It's not a CGI mistake, it's aerodynamics in action.

Offline Nemo492

  • Administrator
  • Hero Dronie #2
  • **********
  • Posts: 769
  • Karma: +30/-1
    • Ovnis-USA
Re: Expert's Analysis Examined
« Reply #1 on: June 10, 2008, 07:05:47 PM »
SecretWeb = Wayne.
Some people have reported some problems while visiting ATS. So, here is the Post :



"Possible Causes for the offsetting

1 â??He eyeballed the positions and thought his eye was â??so goodâ?? he didnâ??t need to use proper alignment tools (or didnâ??t know how to.) (This is more common than youâ??d realise amongst modellers of all levels.) I wouldnâ??t give this the best chance of being the reason and included for completeness, although itâ??s obviously a possibility.

2 â?? He forgot to group all the geometry parts and missed some while move / scale / rotate.

Explanation of â??groupingâ??

Models often are comprised of multiple parts and as such the ability to group these into a temporary unit is found in nearly all major modelling applications. This allows the modeller to transform (a fancy name for moving, scaling or rotating) the model without fear of parts ending up out of alignment or being left behind by accident.

Chances are that if this is the reason then he tried to put them right or simply didnâ??t spot it until it was too late. He simply didnâ??t get it as spot on as if aligned using the proper solution.

3 â?? A conscious decision as he thought it would add a â??less CGIâ?? subtle effect instead of perfect perfection. This is misguided as if aliens are anything like our engineers they donâ??t like â??unpredictableâ?? or â??randomâ??. So the chances of this one are remote as a reason but possible none the less.

4 â?? If the craft was parented to a null object to make it easier to move about (or he could have animated the shots and taken single frames to composite into different shots later), unless the centre pivot was in the right place it could cause behaviour such as this when it was moved scaled or rotated.

Explanation of a â??nullâ??

Think of a null object as a non renderable (i.e. not visible in renders) simple cube (although other shapes can and are used) that us used to link the geometry, lights, bones or camera etc to.

5 â?? Not being â??groupedâ?? (or grouped with not all the parts in the â??droneâ?? in the group) chances are that he has accidently rotated the model by a degree or two. This explains why some â??screwsâ?? are further out the further towards the edge they go. Out of all the reasons, this would be my bet. â??User errorâ??.

Those are the reasons off the top of my head that are probably causes of the misaligned screws. This misalignment obviously produced some problems with shadows being deeper and slightly larger around the worst of the misalignments. Its worth remembering that a 'center pivot' is not always in the center of an object and that they'interact at times when mutiple grouping or linking is used if not correctly handled.

Wayne... "
http://ovnis-usa.com
The only motivation for the DRT is finding the truth.

Offline 10538

  • Administrator
  • Hero Dronie #2
  • **********
  • Posts: 826
  • Karma: +33/-0
Re: Expert's Analysis Examined
« Reply #2 on: June 10, 2008, 07:12:19 PM »
Wayne is claiming there is a misalignment of the panda group.  It could be an optical illusion from the reflected light off the pandas and also the shadowed part of them not sticking out.  There is also the odd perspective.  It's a 3d object in a 2d representation and camera focal length plays a part in how we see it.  IMO, inconclusive.

Offline tomi

  • Hero Dronie #2
  • ********
  • Posts: 668
  • Karma: +37/-26
Re: Expert's Analysis Examined
« Reply #3 on: June 10, 2008, 07:54:28 PM »
When you say misalignment do you mean off-center?  In my post below I see this too, but just as an airplane wing can extend itself, can't this be to an extension of the width of the right side of the length of the arm, of course it would have to grow a bit and make it's own curvature as required, but I do not see this as an exceptional explaination for the type of units we are dealing with here. 

Just wondering if I'm missing something else he is refering to?

Offline spf33

  • Administrator
  • Major Dronie
  • **********
  • Posts: 216
  • Karma: +29/-0
Re: Expert's Analysis Examined
« Reply #4 on: June 10, 2008, 08:40:47 PM »
i see the effect that wayne is talking about.

i just disagree with his conclusion, i'm thinking the apparent off centering
is due to perspective foreshortening combined with blooming, combined with motion blur.  after all we're only talking about differences here of under 10 pixels, hardly hoax busting territory for me.

more investigations into correct lighting angle, surface material and motion
blur are needed.  motion blur especially as that will stretch out pixels and
details.



« Last Edit: June 10, 2008, 08:48:40 PM by spf33 »

Offline tomi

  • Hero Dronie #2
  • ********
  • Posts: 668
  • Karma: +37/-26
Re: Expert's Analysis Examined
« Reply #5 on: June 10, 2008, 08:58:46 PM »
That is a fine example study and I agree with your concept of the perspective foreshortening.  I still think I see a bending and it becomes more obvious with the top section of the arm and the second section at the point where it joins with the two units between them.  The top section looks as if it bends slightly independently from the second.  Like it can also extend it's widith.  Maybe that is what those connectors are for, to allow separate slight curvature.  Also it is possible that the little edges that are so well shown in the illustration you provided allow some kind of extension in width, they might retract or extend a bit too.

majicbar

  • Guest
Re: Expert's Analysis Examined
« Reply #6 on: June 10, 2008, 09:13:40 PM »
Excuse me. It seems we are forgeting that these are JPEG images, not true photographs. The JPEG system for all its value is not used by NASA or other others where "absolute" photograph integrity is called for. NASA, actually JPL, will post its photographic documents in TIFF files. JPEG uses compresion algorithims to compress the data that is presented and will sometimes slightly move the items in an image as it does this. In LANDSAT imagry this movement could misplace items greater than acceptable for mapping purposes and thus for precision JPEG would not be usable. TIFF will maintain the placement of individual pixels to the proper placement and is therefor the prefered platform for image compression where geometric measurements and integrity are important.

http://en.wikipedia.org/wiki/JPEG

« Last Edit: June 10, 2008, 09:37:43 PM by majicbar »

Offline spf33

  • Administrator
  • Major Dronie
  • **********
  • Posts: 216
  • Karma: +29/-0
Re: Expert's Analysis Examined
« Reply #7 on: June 10, 2008, 09:29:45 PM »
It seems we are forgeting that these are JPEG images...

yes, and thanks for the reminder.

jpeg compression artifact introduction:



lens barrel distortions:



blooming:



chromatic abberations:



motion blurred pixels:



Tutorial.... Image Distortion
« Last Edit: June 10, 2008, 10:06:21 PM by spf33 »

Offline mgrandin

  • Full Dronie
  • ***
  • Posts: 29
  • Karma: +9/-0
Bolts on my ceiling fan
« Reply #8 on: June 11, 2008, 04:42:56 PM »


The effect Tomi may tried to explain may be showed on my ceiling fan,  according to attached picture.  The bolts are symmetrically placed,  but due to slightly concave shape of blade facing downwards,  the bolts appear somewhat misplaced according to drone picture. 

« Last Edit: June 11, 2008, 05:01:27 PM by Nemo492 »

Offline Nemo492

  • Administrator
  • Hero Dronie #2
  • **********
  • Posts: 769
  • Karma: +30/-1
    • Ovnis-USA
Re: Expert's Analysis Examined
« Reply #9 on: June 11, 2008, 05:10:37 PM »
Wayne is claiming there is a misalignment of the panda group.

He was SO confident about this one... Just needed a ceiling fan to look at.
Well done, Mgrandin.
http://ovnis-usa.com
The only motivation for the DRT is finding the truth.

Offline 10538

  • Administrator
  • Hero Dronie #2
  • **********
  • Posts: 826
  • Karma: +33/-0
Re: Bolts on my ceiling fan
« Reply #10 on: June 11, 2008, 05:47:39 PM »
The effect Tomi may tried to explain may be showed on my ceiling fan,  according to attached picture.  The bolts are symmetrically placed,  but due to slightly concave shape of blade facing downwards,  the bolts appear somewhat misplaced according to drone picture. 

Very impressive, MGrandin.  Basically confirms what I was thinking.  It's all about camera perspective and optical illusion.  Also confirms that just because one is handy with cgi software does not make one a good photo analyst.


Offline tomi

  • Hero Dronie #2
  • ********
  • Posts: 668
  • Karma: +37/-26
Re: Bolts on my ceiling fan
« Reply #11 on: June 11, 2008, 06:06:27 PM »
Also confirms that just because one is handy with cgi software does not make one a good photo analyst.

Obviously !   ;)