Meant to be Seen

Hi guys!

Dave Seigle, President of In-Three, sent me this program to distribute to our members. It is called the Depth Machine.



Based on your screen size and resolution, it will calculate your perceived depth inside and outside the screen. When you fill in the horizontal width, don't go by the diagonal width of your monitor, go by the actual horizontal measurement.

Share your results! Does the theory translate to the 3D experience?

Regards,
Neil

hi! can you provide software for 2d->3d conversion?

hmm I having some problems with the screen dimmensions... Due I live in Germany I'm not comfort with imperial units...

My Zalman has a horizontal resolution of 1680, one pixel are 0.282mm. So the screen is 473,76mm width. I also meashured it, nearly the same value.

In the help of the Depth Machine is sayed, that you need the width in feet.
1 foot = 304,8 mm

So my screen is 1,55 feet with...

That value seems to me to be a bit low... compared to the 40 entered there in default... are my calculations right?

Also something about the values: (Refering to the picture in first post)

When the pictures (left and right) are seperated by 0 pixles it appears an screen depth or 100%.
When the pictures are seperated by 5 pixles and the left pic is on the left side the picture is sensed to be at seven times the distance between you and the screen away.
When the pictures are seperated by 5 pixles and the left pic is on the right side the picture is sensed to be at the half distance between you and the screen.

Is that right or I am wrong?

yuriythebest wrote:hi! can you provide software for 2d->3d conversion?

Some are in Alternative 2D/3D Conversion (NOT Photoforms) forum

The Depth Machine had a bug fix today. An updated version is online now. Thanks In-Three!

News is on the front page.

Regards,
Neil

I tryed it, it works well.

I adjusted the convegence settings in Team fortress 2 so that the two picture of the gun are approx. 6,3 cm divided. (I used a ruler) With glasses on It appears realy to be at the half way from the screen to me.

Realy a great too if you want to create serious content!

Do you know how to use it to predict convergence or pop-out adjustments? If you can explain, that would be helpful.

Regards,
Neil

Neil wrote:Do you know how to use it to predict convergence or pop-out adjustments? If you can explain, that would be helpful.

Regards,
Neil

mm I think that will have to vary from game to game as objects can be of a different scale

I've only done some ruler probes.

I sayed thet my gun should pop out by 50% so I configured the convegence so that the gun's pictures were 6,3cm divided.

Theoretical you can adjust it if you know the distance (in depth) of two objects ingame and the distance of you to your screen.

Lets say you have ingame a hall that is 2m deep. you're sitting 60cm away from you screen. So you need a depth of 333,33%, when you say that your character is at screen depth. So you configure the convegence so that the nearest point of your gun has no seperation (screen depth). Now get a ruler and took of your glasses. You change now the seperation so long that the two pictures of the back wall form the hall (Search for marcant objects like edges or lights) are seperated by your calculeted value (4,44 cm at 22" / 1,833 feet).

Now you should sense the hall to be 2m deep. When you like the pop-out effect you can adyust you gun to be at 75% (25% out of screen). Then your back wall is at 308%, but it gets more difficult to configure becaus you have to cange seperation and convegence to match both values.

One more hint: Thake a clear ruler so you can see the pictire under it.

[EDIT: Sorry, I didn't realise 'convergence' had several meanings in 3D (how confusing)]

It's worth remembering that any time we're talking about convergence, we're talking about an incorrect stereo pair.

We really need to have a thread about that...