Talk:Miniature faking

Comments
I don't know if these external links are worth mentioning, but photographer Keith Loutit has some incredible examples of Tilt-shift Miniature Faking on his websites: He says he aims to make Sydney look like "The Model City". --Epynephrin (talk) 17:56, 8 January 2009 (UTC)
 * Vimeo
 * KeithLoutit.com

It is somewhat unclear, from both the title and content, what this article is about. Is it miniature faking in general, “tilt-shift miniature faking”, or combination thereof? If the first or last, I think this article must remain separate from the Tilt-shift photography article; if the two articles are merged, the non-photographic material will probably need to be removed. As I've indicated on the deletion talk page, I've recommended keeping this article but clarifying its meaning and cleaning it up.

As noted, the article lacks reliable sources for any of its statements. I've added tags for some that are questionable:


 * 1) The opening sentence talks of “manipulating” a photograph, which would seem to imply post-exposure processing. But it isn't clear whether this is also intended to include use of tilt or swing when the photograph is taken.
 * 2) The focus isn't “distorted”; a scene is selectively blurred, either by postprocessing or by rotating the plane of focus using tilt.
 * 3) A macro lens, though convenient, isn't needed for close-up photography, and the DoF is the same regardless of the type of lens used (for an asymmetrical lens, the calculation is slightly different, but the DoF is ultimately the same).
 * 4) Use of tilt can give a very small area of sharpness, but the effect is quite different from the shallow DoF that obtains from close-up photography. How well tilt can “simulate” close-up shallow DoF is perhaps subjective, but the statement should be qualified. It's probably fine to acknowledge general perceptions, but when making factual statements, principles of optics must be respected.
 * 5) The article makes several references to “tilt-shift” and “tilt-shifting” as if this were some sort of technique, when in fact there is no such thing (cite a reliable source, however, and I will stand corrected). Tilt and shift are separate movements, used for different purposes, and it's not at all clear how shift applies to miniature faking.
 * 6) The article states that a horizontal subject is a better candidate for tilt-shift miniature faking than a vertical subject, but cites no authority, gives no explanation, and shows no example.
 * 7) The second paragraph under Techniques begins, “It is perhaps improper to refer to this technique as ‘tilt-shifting’.” This statement is absolutely correct, but it is not clear why one would even want to simulate “tilt-shifting” if the original objective is to simulate a miniature scene.

The material here needs cleanup whether the article remains separate or is merged. Some of the changes, such as close-up photography for macro lens, are trivial; others may be somewhat more controversial. Some statements are simply wrong, and without clarification or citation of reliable sources, may need to be removed. JeffConrad (talk) 02:07, 12 September 2009 (UTC)

Ummm ... the edit comment for the article itself should have read, “Added tags”. JeffConrad (talk) 02:11, 12 September 2009 (UTC)

The consensus among those proposing to keep this article seems to support moving to Miniature faking to better reflect its content, as well eliminate some possible ambiguity. Unless there's an objection, I'll make this move. If people insist, we can keep Tilt-shift miniature faking as a redirect, though I think it's a bit misleading. I'll also try to address some of the issues that I've raised above, dealing primarily with optical techniques. I've have little experience with blurring using digital postprocessing, so for the most part, I'll leave that to someone else. Presumably, the material on digital postprocessing will expand to include other techniques. JeffConrad (talk) 20:12, 18 September 2009 (UTC)

Edit and page move of 22 September 2009
I've made the page move and made a first attempt at cleaning up the article.



I've tried to briefly describe the differences among close-up images of miniatures, simulation using lens tilt, and simulation using digital postprocessing. It's proven more difficult that I expected, simply because I could not find anything that would constitute a reliable source under Wikipedia's normal criteria. I don't think the principles involved are in dispute, but they are nonetheless not widely understood, and there are few references specific enough to be meaningful. Moreover, most of the few solid references on the Scheimpflug principle address maximizing DoF rather than minimizing it. The main references (essentially, Harold Merklinger) are in the Scheimpflug principle article, so I haven't included them here.

I've included descriptions of some of the more important principles to support some of the statements that follow. I've tried to keep this material to the minimum necessary, to avoid making the article too technical. A few diagrams might provide better illustration, but I'm assuming that would be more than the typical reader would want (it's also a fair amount of work).

I've removed all references to “tilt-shift” except for the link to Tilt-shift photography, because  There is no such technique, and Most of the captions using it were for digitally manipulated images that did not use lens tilt at all. 

I've changed the order of the images, putting the best simulation at the top, and the others in the section that discusses them. Two of the images really don't provide convincing miniature simulation, and I've tried to explain why they don't and the image moved to the beginning of the article does. GIR actually called the situation quite well in the edit of 20 October 2007.</li> <li>I've restored mention of Photoshop because it's the predominant image editor, and quite honestly, it's the only image editor with which I'm familiar. I think that without at least some mention of a specific technique, the discussion of digital postprocessing consists only of meaningless generalities. Perhaps someone more familiar with other image editors can expand that discussion.</li> <li>Some tutorials on miniature faking would undoubtedly assist in understanding the techniques discussed here. Many are available on the web, some good, many less good. I've not included any because I don't have any great ideas for selecting those to include. Absent good criteria, including some but not others encourages capricious practice by Wikipedia editors and at the same time invites the link spam that Mfield has been fight for the last couple of years.<li> <li>I've left the tag because there still aren't any references. Quite honestly, I'm not sure any exist. There are some that cover specific techniques; for example, Martin Evening's Photoshop CS4 for Photographers covers the Lens Blur tool at some length. But I don't think it covers miniature faking. There are plenty of unreliable sources, but I can't see how including them does the article anything but a disservice.</li> </ol>

In any event, I think the article is a bit better and more accurate than it was. See what others think. JeffConrad (talk) 09:06, 22 September 2009 (UTC)

Also removed while we're at it. — Nahum Reduta &#91;talk&#124;contribs&#93; 05:55, 29 July 2010 (UTC)

Images
As I indicated, I don't think the image of the image of the building is especially notable, but it is mentioned in the text as an example of a simulation that isn't really convincing. If it's removed, at the very least the text needs to be changed so that it doesn't refer to a nonexistent image.

To me, the Jodhpur images seem reasonably convincing, even though the blurring is a bit exaggerated. The other images reflect an attempt to make do with what's available. I think the ideas in the article would be much better illustrated by more carefully selected images, with perhaps some taken specifically for this purpose. A really good illustration might show an image taken from a fairly low angle to the ground (like that of the large, low building), and include
 * An unmanipulated version, preferably with the entire scene within the DoF.
 * An unmanipulated version, using the lens's maximum aperture to give the smallest possible DoF.
 * A digitally blurred version using a simple technique such as a bidirectional gradient.
 * A digitally blurred version using an edited depth map that eliminates sharpness gradients on objects obviously at the same distance from the camera.
 * An image of the same scene from the same location using lens tilt and the lens's maximum aperture; depending on the scene, this might or might not differ significantly from the digitally blurred version.
 * An actual image of a scale model that at least has some conceptual resemblance to the full-size scene.

This might be too many images, and might run afoul of WP:NOTHOWTO. It would also pushing it with regard to WP:OR, but absent a good source, I'm not sure there's a viable alternative. And I don't think including such a group of images here would be any worse than citing a web site that would not really qualify as reliable under WP:SOURCES. Perhaps we don't need all the examples I've suggested, but I think adding even a few would be an improvement. JeffConrad (talk) 01:03, 6 November 2009 (UTC)

Online Tilt-shifting software
http://labs.artandmobile.com/tiltshift/

86.31.24.91 (talk) 09:18, 5 April 2010 (UTC)

opening sequence in gulliver's travels
I just watched the film Gulliver's Travels 2010 and added the note in "usage in film/television" Rokman (talk) 19:58, 29 December 2010 (UTC)

Image missing under techniques
Under the techniques section it references a picture of jodhpur which is no longer there. Should the reference be deleted or can we bring the image back? TiddiesTiddiesTiddies (talk) 15:56, 6 January 2022 (UTC)