What's happening to my image size?

[K1200]

I've prepared the attached example in an attempt to unravel a problem with FileMaker's capability to drag and drop images from one container field to another. The drag/drop part never fails -- it's the image size that suffers. It gets downsized by some mechanism that I can't figure out. I've tried different combinations of Graphics Format -- Enlarge and Reduce -- to no avail. FileMaker seems to apply some unseen sizing restriction.

Has anyone else noticed this? Is there a way to preserve what the user originally drops or pastes into a container.

Thanks in advance for any help.

[Lee Smith]

See if this recent Thread helps Link

[K1200]

Thanks for the response. Although the link covers the subject from a different angle, it confirms that there's a degree of mystery to exactly how images are handled in containers. It sent me back to my test database to try additional cases. What I found is both encouraging and disconcerting. At this point, my hypothesis is as follows:

For bitmap images placed in containers, there are two cases:

If you specifically request that the image be stored (via a Set Field), FileMaker faithfully stores all image attributes such that any subsequent Restore (via a Set Field) brings the image back 100% of its original size. (see Load and Store buttons in attached file)

But left to its own devices (as when dropping an image into a container) FileMaker processes it into some internal format that reduces its raw image size by about two-thirds AND limits the image’s display size to 75% to 99% of original size, scaled according to the image’s original dimensions. ( A 1680 x 1050 test image is reduced 25% to 1260 x 780 ; a 400 x 300 image is reduced 1% to 396 x 297)

The good news is that if the Compacted Image is exported, FileMaker somehow reproduces the full 100% version in the resulting file.

The downside, of course, is that to the user, the image appears shrunken -- yet it isn’t ... not really ... but what you see isn’t what you get ... which is confusing.

I’ve attached an improved ContainerTest file if anyone wants to explore this further. At this point, I’m resolved to accept that it’s just the way it works -- which is easier to do now that I’ve examined things in more detail. Maybe this will help others.

ContainerTest.fp7.ZIP