I have built up a show in fullHD (1920x1080 pixel) and of course I used images I made with this resolution.
In one of my slides I added a seconed imagelayer. The image I use just for this layer is 9600x9600 pixels. So it´s five times as big as the final show format.
Now I want to zoom in the seconed imagelayer to 500% so that one pixel of the image is one pixel of the final show. Later on I move and rotate this layer arround by using keyframes. My aim is to keep crisp details if I zoom in.
But if I zoom in everything is totaly blured out?!
It seams to me that Proshowproducer loads the image - shrinks it down to 1920x1080 and if I zoom in just uses the shrinked data to enlarge it again.
So where are all the extra details of the original file gone? It doesn´t matter if I select "Fill Frame, Fit to Frame, ..."
So what´s wrong - anybody who could help me with this?
It´s very important because otherwise every image larger than the show resolution would make no sense for more details ...?
If anybody would like to have the files - please contact me.
ThanX ans sorry for my bad English. I´m German. *:o(
Unfortunately I don't have a good solution for you, but, I have a kind of explanation for what is going on.
This explanation was given to me a couple of years ago, by a Photodex representative, because I was having a similar problem with a very long image:
When you have an image this large, ProShow can't import the full image resolution because it will run out of memory during playback. ... This is related to GPU acceleration. Because of this, ProSow lowers the resolution of the imported image. This isn't usually noticeable at normal zoom levels (e.g. 100-250% Fit To Frame; Fill Frame varies because it might need to apply a large zoom even at 100%), but larger zooms make this more obvious.
Also as mentioned by Photodex representative, it is only the preview in the show itself that is effected, and not the final video output.
The video output will look quite OK.
In my case, the solution that was given to me, was to cut the image into two parts, and to use modifiers to make the two images work seamlessly. Quite a pain, let me tell you.
I don't quite understand what kind of effect you want to make, and why it is so important to calculate pixels.
Sometimes a smaller image will look better than a larger one.
i did´nt render the whole thing out so far because of the blured prv. letzt see what happens. maybe i´m a lucky guy afterwards ...?
i´d like to do some kinetic typography combined with pictures which are fixed to the background. so i have to rotate and scoll a lot.
this is where the oversized image comes in. it would be the easiest way to setup all the stuff on the canvas which is 5 times the show size and then move the whole thing arround.
splitting it into different files is - as you said - very difficult. let´s see ...
the calculation was to buld a canvas five times as big as the show and then zoom in five times. so every pixel on the canvas is one in the show and therefore i have the original quality of my text and images.
i also opened up a support request an photodex. maybee they could say how big the maximum image could bee for ram reasons.
maybee in this case it would be bnetter to do the whole thing in afterfx - sorry for that because i love proshow - its so simple and quick to arrange my things.
I just give it a tray later on. But on the other hand it´s realy realy blurry and there´s no chance to position things up. so what a pitty ...
"ProShow imports images at a lower resolution because the raw bitmap data for very large images (e.g. 9600x9600) would consume too much memory. As a result, the preview within ProShow will look blurry when you use these larges images and zoom to a near-native resolution (e.g. a 1920x1080 area of the original image). When you export to a video file format like MPEG-4 1080p, the original image is used rather than the downscaled version ProShow uses for its preview, so it should be relatively sharp in the final video."
Have you tried with an image of half the size of your original 9600x9600 ? Say around 4000 or 5000 pixels.
Could be better for Producer, and it won't affect your show so much.
It´s a good idea to setup the whole thing with lores files. It´s a little bit pixelated but you see what you get and you can position everything quiet easy.
After the whole process of animating the layer just replace the lores image with the final hirres image (change it in the layer options of the slide) and render it out.
-> The trick is that the positioning is based on percentages - not pixels. So to move a small image by 50% is just the same amount of shifting as 50% effects a huge image. That´s it and so simple?!
Worked for me. Hope anybody else finds this usefull ... let me know.
Sorry but I don't understand the words "lores" and "hirres" ". They are not in English.
I get the general idea of what you want to say, but I'm not sure. Do you mean that you made your animation working with the image with a lower resolution, and then replaced it with the one with the higher resolution, for the final video rendering ?
lo(w)-resolution / hi(gh)-resolution
Sorry about that - I´m an old Amiga-Kid and "lo/hires picture" was a common syntax by using the famous software "Deluxe Paint" I startet my career with. *;o)
- ProShow Hall of Fame
- Posts: 1262
- Joined: Sun Oct 01, 2006 7:53 am
- Location: New Mexico
PHOTODEX GALLERY: http://www.photodex.com/share/hardsoftware
UnEmployed, and getting pretty good at it!
Facebook look for Ben R. Baca.
Who is online
Users browsing this forum: No registered users and 6 guests