Hello John!
Your explanation is probably correct, but I am very surprised. Firstly: the image of photo F is in the computer's memory (digital image), so its duplication by the method (e.g. Ctrl-Alt-J) should not result in a loss of quality.
Second: the photocopying analogy is not good as there is no scanning process here. In the case of a photocopier, the main reason for reducing quality is the scanning process.
Third: I do not understand when the program performs (and why) compression / decompression during the implementation of the described operations.
John, I don't dispute your explanation, I just don't understand it.
I'm sorry, but I can't take your click and drag advice because in my project the following images F, F1, F2 ... are modified in such a way that the last Fi frame is the first Fi + 1 frame. Especially:
a) all Fi (i = 1, ..., 13) are 2s long (50 frames)
b) Fi's photo is the background for Ai's animation.
c) the last frame (fiftieth) of the Fi photo is the background for the Ai + 1 animation in the Fi + 1 photo.

The details of my project are irrelevant (I'm only writing about it because you were kindly John kind to respond to my complaint) because the described Magix "bug" occurs for a regular F photo, without any additional operations.

I don't understand your explanation John, because I don't know enough about how Magix works, but I can confirm that what happens to photos in the Ctrl-Alt-J duplicate process - actually resembles the effect of multiple scanning-printing.
Best wishes
Richard

@Turkana

Hi Richard,

John EB's explanation is spot on, except that to make a copy on the timeline, he omitted the Ctrl part. Hold down Ctrl+drag the object with the left mouse button. This is a copy of the original that is in the memory. The shortcut D does the same thing, but then you have to move it.

In your method, the copy does not come from the memory, You are exporting, most likely in a lower resolution than the original, importing the newly exported version, exporting that, importing it, exporting, importing, etc. Each generation will have a lower quality than the previous. My original photo is 3264x1836. The export is 1920x1080. I now have a lower resolution image. This is like John EB says, photocopying a photocopy of a photocopy....gets worse. No bug here, completely predictable.

Why would you do this anyways? You have the original, you can duplicate as I pointed out. Or, you can simply import it over and over. Same difference.

John CB

@Turkana

all Fi (i = 1, ..., 13) are 2s long (50 frames)

If you are just putting more F1s beside each other to extend them showing on the timeline, you can set an increased time by either setting the default length in Settings>Program> Video/Audio Tab, or by changing the image length when on the timeline:

or by dragging the lower right handle (or left) on the the end of the image object to stretch it:

Put the mouse pointer on that corner and drag it right/left to increase the length/time of the image.

@AAProds

Hi Al,

I hope that there is more to it than that. Otherwise, it's just simply extend the duration of the photo.

@Turkana

So, what exactly are you trying to do (without the F1, F2...F50 stuff? I could understand if you were then changing each one of the same replicated image, but if not, then it does not make sense.

If you just want that image to be the duration of all of the other photos, then do as @AAProds indicated or right-click, change photo length and make it 100s or whatever.

John CB

@browj2

Thanks for the heads up 👍 - post has been corrected.

John EB

😂🤦‍♂️🤷‍♂️ This post intrigued me, so I just tried it to see & maybe to work out 'why' Turkana was doing it this way,

I copied a single frame from an MP4, (I also tried it on a jpg image) the 1st copy made was very slightly brighter with a BMP,JPG & PNG than the orig MP4 frame, then each copy of the copy had that added brightness, so the last was however many of the copied times brighter 🤦‍♂️ but no change in pic size. & this is a still image copy/format of a video format,I wouldn't expect it to be exactly the same.

In my game, joinery, if you want repeated accurate copies you use the orig as a template & only copy from that one, because otherwise you get this stacking of inconsistences if you copy from a copy..,

& why not as has been mentioned Ctrl+drag to copy, or drag the copied image out to any length you want & cut into whatever pieces then add the effects to each, 🤦‍♂️🤷‍♂️

Mum why does it hurt more every time I bang my head on the table ? 🤔🤣🤣🤦‍♂️

@Turkana

Hi Richard,

I am still a little befuddled about what you are doing. If I understand correctly, you are exported the composite image, not just the original image, then using the composite image as the source for the next image with a different mask. Is this correct?

Are you applying a mask to the masked area or are you masking a different area of the image each time?

I would like to know more. Screen shots would help.

For multiple masks and masks that change shape/size/position, I would use Xara opened from MEP/VPX.

John CB

@Turkana

Hi

Like @browj2 I am trying to understand what you are doing, I have an idea from your description and agree with John CB about using Xara, Pohoto and Graphics Designer.

Is there an example on Youtube/Vimeo etc that shows what you are trying to achieve?

John EB

Hi John CB
Yes! I export a composite image: a photo with an animation overlaid. This composite image is the background for the next animation 13 times.
I mask a different area of ​​the image each time.
I can see John EB is showing a polite interest as well. So please: Here is the result of my work and that was the cause of all the confusion I caused:
https://www.dropbox.com/s/i2yutv4wk8npdb1/Book.mp4?dl=0
I did it on three paths yesterday.
On the first track of the timeline: a photo of the ancient "book" (photo F) in .jpg format
On the second: a mask in the shape of 12 steps of 8000x2000 pixels in .png format
On the third track of the timeline: ancient marble Greek text in .png format
The whole animation boils down to one operation of moving the mask along the x-axis.
But this was only made up yesterday.
A few days ago, when I came to this animation in my work on the film, I automatically and without hesitation used a rectangular Matte_15_00001.png mask from Magix's Templates, which I reduced to the size of a poem from the presented "book". The following lines in the "book" were revealed by moving the mask first along the x axis, and then moving to the next line along the y axis. And so 13 times. But there was a problem with this mask, because when you moved on to the next line, the line above was hidden. That's where the idea came from to export the last frame of each line to a .jpg file and then use it as a background in the revealing operation of the next line in a "book". And so 13x.
After the animation was over, I noticed a Magix "bug" that I described on the Forum, which means that subsequent Ctrl+Alt+J gave worse and worse photos.
Jon EB immediately explained the reason, but I still thought it was Magix's fault, but when John CB endorsed it, I thought it was unlikely that everyone was drunk and only I was sober. I started to think - which happens to me less and less - but ... I understood the essence of Ctrl+Alt+J export.
Thanks to the discussion at the Forum, I devoted more time to my animation and the task was completed elegantly - at least technically.
Thanks again to John EB and John CB and to all those at the Forum for their interest in my case.
Regards -
Richard
PS
While working with masks, I would like to show off my last idea of ​​drawing a circle. I do this with one mask and only the rotation operation (no shift, no reflection and no magnification). Masks are math and that's probably why I enjoy it.
I know I'm talking to experts, so words are unnecessary and just look at the drawing: 
https://www.dropbox.com/s/lo1fitnljstv2xj/Ring_RL.jpg?dl=0
The black circles are a mask that revolves around a point on a red (exposed) circle (in everyday language - on the edge of a drawn circle).
I know that my method has disadvantages: 1) the drawing speed is not constant, 2) for thick lines there is a problem with the beginning and end, but I am happy when what I calculate gives not only theoretical effects.

@Turkana Hi, can you screen capture the MEP screen- timeline, thanks, you record it at 1920x1080 you can post it directly on here, I'd like to see how you positioned each event,

@Turkana

Hi Richard

The book animation is very realistic - a good solution for the Greek text - elegant and using the minimum necessary to achieve the effect.

The circle effect problem with thicker lines is an issue. I used a similar technique for road maps revealing a route and, if zoomed in too close, the route line would reveal in 'unusual' ways where there were sharp turns, I never found a satisfatory solution with out using multiple different shaped masks.

John EB

@Turkana

Hi Richard,

Here is another method. See the text in the video and the timeline. I did not have to export/import even once.

BTW, you can upload images and videos directly into the message using the up arrow (upload file) button.

John CB

@browj2 Hi, this is what i came up with last night,

The text is a b/w image so i had to add alpha to remove the black background, the masks are white with alpha added, the top one masks all but the bit i want to reveal, the one below does the reveal, copy n paste then adjust size/position of only the masks

@browj2 Looking at yours tho maybe it's the same, mines just all on the same tracks 🤷‍♂️😂👍

Hello John CB,
All clear. Cool method.
PS
Thanks for the advice "upload images and videos directly". I have not yet read about the possibilities of the Forum.
I am about to test how it works:

The final result ...

Unfortunately, the movie will not be available, because it does not want to be uploaded using the "upload" method 😕 Maybe another time 🙂

Ryszard

@Turkana Hi, that's a busy mask, is really wide that travels across the screen?

I have to reduce the video size to 1920 x 1080 to upload here, i'm not sure what the size limit is I've never found that info,

Hello Gid,
In my project, the average length of a line of Greek letters is approximately 420 pixels, which times 13 lines equals 5460 pixels.
So if I only want to use one mask, it has to be that long.
But that doesn't seem to explain the uploud problem, as I tried to upload a rendered movie 1920x1080.

Sincerely -

Ryszard

@Turkana Timing of a mask that long to reveal in the right order would be fun, either it has to be shaped exactly right or lots of keyframes, your text was similar length on each line tho, lots of keyframes in the way i did it if you add them up on each event tho,

Hmm if i can't get a video to upload it's because it 4k or too big, that video i posted was a Magix MP4 AVC/AAC 1920x1080 rendered at default settings, 40MB in size, so don't know why you can't upload. 🤷‍♂️

@Ehemaliger User

Hi

. . . . Timing of a mask that long to reveal in the right order would be fun, either it has to be shaped exactly right or lots of keyframes, . . .

As we cannot see how many keyframes there are from the images of the timeline above, as the mask steps are constant width and height, the minimum number of keyframes needed in this example would be four - two for the left/right movement, and two for up/down to take care of the small vertical movement as the mask appears to be at a slight angle. The steps take care of the line by line reveal.

John EB

@johnebaker Yep if you wanted just one simply linear slide to one side, but as i wrote 'text was similar length on each line' so each step on the mask would be the same, if one line of txt was short, maybe just a word or two on a line then there'd be a delay in the reveal unless the mask was stepped correctly,

@johnebaker What i'm trying to say is if you look at my video, the second line down has only Vegas Pro 19 centred, i had to alter the mask reveal so that each line 'typed-on' in a consistent manner, if that makes sense, Turkana's text was all the same so the mask could be stepped regularly & so you're right only simple keyframes needed