Misplaced Pages

User talk:Cuddlyable3

Article snapshot taken from Wikipedia with creative commons attribution-sharealike license. Give it a read and then ask your questions in the chat. We can research this topic together.

This is an old revision of this page, as edited by Thegreenj (talk | contribs) at 20:44, 11 May 2007 (FPC). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Revision as of 20:44, 11 May 2007 by Thegreenj (talk | contribs) (FPC)(diff) ← Previous revision | Latest revision (diff) | Newer revision → (diff)

Deletion of Image:Carburetor2.png

An image or media file that you uploaded or altered, Image:Carburetor2.png, has been listed at Misplaced Pages:Images and media for deletion. Please look there to see why this is (you may have to search for the title of the image to find its entry), if you are interested in it not being deleted. It has been nominated for deletion because it has been obseleted by a more recent version of the diagram. Thank you. WikipedianProlific 10:16, 18 March 2007 (UTC)

FPC

A Koch curve has an infinitely repeating self-similarity when it is magnified.
Anti-aliased example

Your animation Image:Kochsim.gif has been nominated for Featured Picture. Beacause it has recieved some complaints over size and aliasing, I wonder if you might be able to upload a larger, anti-aliased version. It certatinly is interesting, and I would love to see a better version. J Are you green? 21:08, 6 May 2007 (UTC)

I have added notes to the image description that may interest you.Cuddlyable3 19:17, 7 May 2007 (UTC)

It really is an interesting illustration. Do you think that you could redo it as, perhaps, a 400 by 200 pixel animation in greyscale with antialiasing? I love the idea, and I would absolutely support a newer version. J Are you green? 20:07, 7 May 2007 (UTC)
Increasing the pixel resolution is easy and just makes the file bigger. Rendering in a greyscale however would need some arbitrary process which goes beyond what the Koch curve defines. Aliasing is the result of sampling in space or time (see my image description notes) so there are several possible sources to consider. Strictly speaking, we should not see a 2-D line at all, nor the structure of the fully developed Koch curve. For a beautiful image, search out (Google) the sphereflake! Cuddlyable3 08:18, 8 May 2007 (UTC)
That's OK - I just thought it might have a chance if you could do that. J Are you green? 00:39, 9 May 2007 (UTC)

To do anti-aliasing, just render it 3x as big as the final image, and shrink it down (e.g. with bi-cubic). Of course to do "perfect" anti-aliasing you'd need an infinitely large initial rendering, but it doesn't need to be perfect. A separate comment, there's too much white space as the bottom. —Pengo 15:23, 9 May 2007 (UTC)

Do you have the means to do this and see the result? The code to draw the Koch curve is rather simple and I can help you with that if you wish. However you could also take the existing image (or just one frame of it) and reduce its size to 67 x 34 pixels; that simple exercise might save you some time and possible disapointment. As to the white space, you are right that it could be reduced. Cuddlyable3 19:29, 9 May 2007 (UTC)
I reduced one frame - looks tiny but antialiased to me... If you upoload a new version of the Koch curve that is identical to this one except that it is rendered at, perhaps, 900 by 450 pixels, I can shrink it down for you to 300 by 150 pixels and get antialiasing as a side-effect, as Pengo suggested. J Are you green? 20:31, 10 May 2007 (UTC)
J, please post your reduced frame here if you can, so we can all see it. Since the object is scaling invariant we don't need to push especially large files through the Wiki server, do we?
I note that the antialiasing process Pengo describes if done on a 2-colour (monochrome) image generates a 16-colour (greyscale) image. This is because one filters by taking 3x3 blocks of pixels, using 3 different coefficients for center, mid-side and corner.
However I think a quest for an "antialiased" Koch curve by increasing pixel resolution will only lead to huge image files (slow to load) and no new aesthetic delight, until one has magnified it so much that the finite iteration limit of the curve computation becomes visible. At that stage you are just seeing a monochrome line figure, which is where it all started. Cuddlyable3 07:48, 11 May 2007 (UTC)
OK; here you go. It obviously is tiny, which is why I am asking you to render the original at 900 by 450 pixels. As for file size, relax. Your GIF is currently 4 KB; I cannot see a 900 by 450 version being more than 85 KB, still a really small file. If you upload a large verision over the current one, I'll downsample it for you. J Are you green? 20:08, 11 May 2007 (UTC)
Oh, and as for the resolution and limitations, its not really how much deatil is really there (especially for something like this where antialiasing will destroy that ultrafine detail) as how easy it is on the eye. To be honest, a 200 by 100 pixel image looks tiny on my screen (about 2 by 4 cm). I really wouldn't mind the lack detail so much as to have a larger, anti-aliased image. By the way, downsampling probably will destroy any visible limitations of the "finite iteration limit," so I wouldn't worry about that too much. J Are you green? 20:44, 11 May 2007 (UTC)