There is one issue that could be radically affecting what we see here - perspective.
Photographs naturally have perspective due to both the basic geometry of near/far objects being observed from a point and due to the focal length of the lens which can exaggerate the phenomenon. And most 3d modeling software will render with perspective as well. It basically means "objects closer to the camera appear larger than objects at a distance". These images definitely have some perspective as can be seen from the fact that the rear seats (especially the 3rd row in the X) appear more reclined than the seats closer to the center of the field of view. That's a typical distortion one would see with a regular perspective projection.
The way in which it can affect these comparison photos is that the Model X has taller front and rear ends. That means the dimension I used to verify that they are at similar scales - overall length - will be affected by the fact that the front and rear edges are closer to the camera in the X picture/rendering. That would mean that a proper scaling would involve figuring out how much larger the X would "appear" due to that perspective magnification and to resize its image to be similarly longer. Unfortunately, it isn't easy to calculate this factor as it depends on how high the "camera" was and what it's focal length was. In terms of a photograph those are physical measurements, in terms of a CGI rendering those values are programmed into the software so it can simulate what the scene would look like for a given camera/eye position.
All in all, I'm willing to bet that there is more room in the X than we see by scaling the images for the overall length of the vehicle. I'm not sure how much of a factor that is, though.
Some alternatives that might be able to correct for this (but we don't have enough information to know how complete this is), would be:
- Scale for the tiles which appear to be scaled differently in the different photos, but perhaps they are modeled identically, but positioned further from the camera in the X picture.
- Scale for an interior detail we know to be the same. The cup-holders, steering wheel, and LCD display are likely to be identical, but are we sure? Also, they'd have to be at the same distance from the camera in both images and we have no idea how high the camera was or was simulated to be.
Even if we did adjust the scales, we'd only be making the images comparable to each other for a given "height". In other words, if we adjust for the cupholders, then only other features at the same height as the cupholders will be accurately comparable. Even if we say that the tiles are both at the same absolute "height" in both pictures, the ride height of the X is taller and the seats probably sit higher so various features will still not be comparable.
In short, I think this effort is going to produce biased results and only big picture answers are useful. It looks pretty certain that the hood length is smaller now, but the difference could be exaggerated. The mirrors probably are further forward, and thus the whole "cab [could be] forward" in the X, but we aren't going to get an accurate measurement of how much from this experiment.
Finally, if these are produced using a CGI of the interior projected onto a real photograph of the body with appropriate layer blending in Photoshop, we have to trust the artist to have gotten a very precise match of the virtual camera parameters from the CGI rendering and the real world camera measurements.
So, all in all, this effort is for entertainment porpoises only. May they make you smile with how they jump through hoops to eat fish...