Moire effect

[Jean-Cyprien]

Hi Igor,

Viewing an AV made with pictures of 1920x1200 pixels, I’ve observed on my screen (1600x1200 pixels) an unpleasant moiree effect – and I’m not the only one who has observed such a bad effect. This problem seems to be more accurate with some videoprojector and large screens. Too bad if it is a big show ! (And it is generally difficult for the author to guess the effect on an other screen than its own).

It seems that in the past, the option called "Mipmapping (anti-shimmering)" was “too strong”, with too blurred images in consequence. And thus, this option is usually avoided.

But I understand it’ll be possible in a near future to adjust the degree of "mipmapping". Could you explain how this new option will work ?

Thank a lot in advance.

Jean-Cyprien

[Igor]

Jean-Cyprien,

You have raised a new interesting theme! So I've moved your post to this new topic from "PicturesToExe Deluxe 6.5 Beta" thread.

The experimental option to adjust degree of "Mipmapping (Anti-shimmering)" already available in version 6.5 Beta 8.

Turn on mipmapping option, save the project and then open .pte project file in a text editor.

Find this text line in a section of necessary object:

MipLevels=1

Add the additional line:

MipLodBias=0

Float values.

0 - no change. Standard mipmapping.

-0.5 - picture more sharp

-1 - picture more sharp

1 - picture more blurred

0.5 - picture more blurred

I recommend try -0.5 which give best result (no moire and almost sharp).

[Jean-Cyprien]

Incredible !

How fast is your answer Igor !! I'll try tomorrow.

Thank you very much !

[Jean-Cyprien]

It works !

But after having turn on the mipmapping option, it’s not necessary to save the project and reopen it : it takes more time than the following process :

Stay in the O&A window, copy the object, paste it in Excel (for example), add the new line with a value between –1 and +1 , copy the new file, return in the O&A window, and paste it.

You can make several return very quickly and easily, and see the result with different values quite immediately.

I’ll test it later.

Thank again ! Jean-Cyprien

[denwell]

Igor

Can we assume that this, most welcome, addition will be available in future betas with a UI control rather than using a text editor?

[Igor]

I think so. We have to decide how to better implement this feature. I'm waiting for results of testing.

[Jean-Cyprien]

Hi,

Here are the results of a few tests I made, to test the effectivness of the new value.

What " I " see is perhaps completly different of what " YOU " see, so my conclusions can't be a certainty.

Thank you. Jean-Cyprien

Here is the link ( the file name is Moire_GB.zip = .exe file and the template)

http://www.sendspace.com/file/wlp414

[fh1805]

Jean-Cyprien,

My friend, that one needs to carry a "Health Warning"!!. Apart from now having a headache (only joking!) I'm not sure I'm any wiser. I think I need to watch it again a few times - but all that shimmering is hard on my eyes. I got the impression that the middle two on the bottom row were always (or almost always) the least shimmering.

regards,

Peter

[Jean-Cyprien]

Ah, Ah, Ah, Peter ! I’m very sorry for your virtual headache ! Ah, ah !

“ I'm not sure I'm any wiser.” But yes, you are, as everybody else could be !

OK, the pattern # 1 and # 2 are very difficult to cure, they are caricature designs, but I think necessary to understand how the tool is working, and try to test its efficiency. They help to show the difference of what I call shimmering and moiré, two different effects (or so it seems) of the same phenomena.

The third pattern is a real one, even if it is a anaglyph. You have the template in hand, so it is easy for you to replace the picture (cheverny.jpg) by one of yours to test by yourself.

Finally, you have the same feeling as me : the positive values are efficient, their adjustment is perceptible, the result is not so hard as with the box “Blur”. (Perhaps a value between +0.3 and +0.6 is the best, for THIS three patterns, but to be examined for other pattern ? ).

Quite a lot of good things !

The differences with the negative values are more difficult to appreciate. For them, it needs a real picture where the “ancient box” “Anti-shimmering” leads to a too strong effect. But unfortunately I have not such a picture in hand.

Thank you, Peter to help us.

regards,

Jean-Cyprien