I zoom in and out all the time, using the screen zooming when in the phone or the pad when using the laptop.
The other option is to record or take a screenshot of the screen and then open it and zoom in and out and again take a screenshot, make a GIF or another video
Do you mean the touchpad of the laptop itself…and doing that with ‘gestures’?
Yes, sorry, i meant the touchpad indeed.
Using the “2 fingers” gestures for zooming in and out, yes
Ah OK, now I understand how to do it.
Thank you. 
You’re very welcome
The same as i do it on my phone screen basically
Well, I would say I do it manually in the editing.
To avoid confusion with some things @David and @John78 were just talking about, you can manually zoom an scroll in a video window while you are capturing the footage and thereby create such effects. I did something like that when I started out, but that’s not how I’m doing it now.
These days I always capture the full screen VHTV footage and then do all the formatting later in the editing. When you have learned how to do that in your video editing software it is just so much more effective, precise and easier, that the “manual manual” way is really not an option anymore.
With a GIF I fisrt review the recorded footage to find a moment that I think might make a good GIF and decide in what aspect ratio I want to frame that. For Shotcut I’ve made templates in aspect ratios 5:2, 16:9, 4:3, 1:1, 3:4, 9:16 and 2:5. So in the case of the Rollie GIF above I imported the recorded footage into my 3:4 aspect ratio template, meaning the GIF could be, for example 300 pixels wide and 400 pixels high. This one is 450x600.
Then in this 3:4 project I can cut the clip and zoom it however I want it to fill that 3:4 “window”. And when I use so-called key frames I can go to any moment in the clip, and define how I want it to be zoomed in at that moment. I can also define if I want the “camera movement” between these key frames to be a linear, or smooth or pseudo-handheld motion. For a GIF such as this I’ll just define some of these key frames, watch a preview, make some adjustments, watch a preview again and so on.
In Shotcut there’s also an option to do automatic object tracking, and I’ve played around with that, but to get that to work the way you want it to may take much more time than just doing it manually. At least for the VHTV stuff that I’ve been doing. If you want to stabilize some shaky handheld footage that might be another thing.
It is all a bit more complicated than what I explained here, but that’s the general gist of it. And if anybody needs specific help with how to do this in Shotcut, I can be of assistance.
hi,
i have done a zoom in/out once with player and OBS-studio. i just used media-player classic as source and recorded the part i wanted.
in MPC-HC you cam zoom with the numpad. 9-in, 1-out, 5-back to original,
ctrl/arrows for postioning up, down, left, right, and 1,3,7,9 diagonal.
very easy to do, when it not necessary to zoom and postion at the same time.
best of course when the action is in the middle and no postioning is needed.
btw…shotcut is quite good. but i still have not found a good coding-setup for correct quality/size combination.
because of that, i still make slideshows with irfan-view and for video-cutting without new encoding, i use MediaCoder. you can export the part needed (video and audio) with parameters ‘copy video’ and ‘copy audio’ without loss of any quality and no file size increasing. fastest of all.
i did not find that in shotcut. 
Meaning you have difficulties finding the right video quality settings to aim for a certain file size with your video. I can sing a song about that, I struggled with keeping my 20 minute videos within the VHTV forum file size limit.
When aiming for a certain file size, in the Codec tab, for the Rate control option, you need to use Constant Bitrate or Average Bitrate. While it’s an easy choice for shorter videos, Quality-based VBR is an incalculable gamble because the resulting file size very much depends on the quality or “information density” of your source material. Constant Bitrate is a very inefficient use of your available file space for the best video quality. The best results you’ll get with Average Bitrate and a dual pass. Though it will take longer to encode.
Encode one video this way. Or a few more. You know your combined bitrate (that is your video stream bitrate plus the audio stream bitrate), you know your video lenght in seconds, you know your file size. This gives you an equation which you can use to calculate the combined bitrate for your next video and target file size. Btw, in a pinch, don’t waste too much bitrate on audio, with the poor quality of VHTV audio 128 kbps should do. 96 kbps mono is fine.
To stay under our file size limit of 97,7 Mb, with some safety margin, I arrived at this rule of thumb:
Max. combined average bitrate in Mbps = 800 / video length in seconds.
Maybe that helps.
i tried a lot combinations…nothing reached the original-stream-recorded quality, even not a slideshow.
that i’ve not tried.
i will today. but not now…a few hours ago i did experiments with short testfile and watched results in player while streaming two cams and all at a time…the grafik gave up 
and i had to restart. 
This may sound funny, but I think this actually makes sense because the VHTV video streams are of piss poor quality already. Meaning they are highly compressed and not a very accurate representation of the image data that the camera sensor originally delivered. For a very accurate representation of a video you need a certain, high bitrate.
Now, if you want to make a new video using VHTV footage, and you don’t want to loose any more video quality than the lots of quality that have already been lost in the first compression, you need to create a highly accurate representation of that very inaccurate representation that is the original VHTV stream, and thus you actually need a higher bitrate. Does that make sense.
Making a perfect copy of a turd is more difficult than making the original turd.
It also makes sense that Shotcut should be unable to re-use the original encoding, since pretty much every single modification you could make to that video, every filter you could apply, would negate that encoding.
I’m not even sure if the kind of editing you do in MediaCoder supports single frame accurate cuts. Can you actually join several clips together with that software, or can you only export single clips? If the latter, I think I’m doing the same thing with the trimming function of my Mac’s Quicktime player, when sometimes I roughly shorten too long footage recordings before actually using them in the editing.
for that it seems not to be the right thing. i allways got an error in a few tests. but i did not test much with that. i just use it for cut out.
a result is just posted in ‘san&ledian’. a part out of 280 mb. exported in 2 seconds. identical quality.
i will also dig out ‘VirtualDub’, that i used years (and even more years) ago. i guess there was also an option for ‘use original’ and when i remember correct, it can combine without encoding, when the parts have the same parameters, from one record for example. (what MediaCoder also should be able to do)
with 2500K and dual pass i get a pretty identical quality and filesize (24.1MB->22,9MB). so that’s good for combining the clips.
left original, right new.
That’s one of the common misconceptions. Hardware acceleration or hardware encoding sounds like it’s the super cool thing that you should always use. It isn’t. It is all optimized for speed.
Which is fine when you’re working with good quality footage and normal target bitrates. But software encoding can be more thorough. If you have to fight for every little bit of video quality because you have only very little bitrate available, like with our filesize limit, or because your footage isn’t that good to begin with, software encoding may give you better results.
But that comes comes at the cost of longer encoding times and also decoding times. Meaning it may take a bit more processing power to play back your video file. But with the tiny little files we’re doing here that is of absolutely no concern. There’s a few more things that may give you a bit better quality and increase encoding time significantly.
If you scroll up to post 46 of this topic, there’s a video that took my computer, if I remember correctly, more than 90 minutes to encode.
as i have just seen a 100mb 30 sec vid (.mov), that was announced as compressed, i like to post the shotcut parameters for video, that has the quality, that the streams delivers (2500 kb/s average)
when we see one, who struggles with it (like me before), we can post a link to this.
first line for 1920x1080 as it comes from the stream.
second line for 1280x720
In the 1080p file you have a framerate of 30 fps, the VHTV streams however only have 25 fps. Choosing a higher framerate in your export settings will do nothing to increase the video quality, only the file size. Also I think every frame in your video wil have to be created by interpolation.
When you see video files in the forum that have 30 or 60 fps that is probably because most video capturing tools capture the display on your computer. When I use my Mac’s native Screenshot.app I get 60 fps .mov files that are much too big.
The Video DownloadHelper browser plug-in will latch onto the stream by itself and, as far as I can tell, will generate files that only contain the original raw stream data as it arrives at your computer. Which is what you want for editing. You don’t even have to watch what you are recording, you can start a recording and let it run in the background. You can run multiple recordings at once, and they even record faster than in real time. It is a bit difficult though to time the recordings as you will see if start using it.
yes…i see. it’s not really what i wanted to show 
i will change.
that is what i also took as ‘that’s what comes’ example for me.
I was thinking, maybe you want to look into this conversation I have been having at the Shotcut forum about export settings. For example for a long time I completely ignored the settings in the “Other” tab (Sonstiges), which are not completely irrelevant.
This began as a little editing exercise, trying to make a GIF as a really short summary of a sex act. For a GIF it was quite a bit of work to get it this short. Some of the clips are sped up significantly. Once I had a few zooming motions in there, it occured to me that maybe it could create a certain magnetic pull, suck the viewer in, so to say, if I expanded that to almost all the shots. Or is it too much? What do you guys think?
BTW, although there’s not too much hot sexual action in that realm, Agnes has kind of become my main muse at the moment. I really like her.
Since you asked for feedback: I think the gif is fine. The only thing I would have done differently:
I would have left the “voyeur-house.tv” out of the sequences where it’s already there, so that there is no duplication or overlapping in the course of zooming.
Then I’d have to add the logo as a filter to all the clips individually, or arrange the clips in separate tracks with or without the logo, instead of just adding the filter once to the overall output. It’s really not worth the effort, I think. The timestamp and logo in the footage are just nuisances I’ll have to live with.
Well, there are some things you can do to remove the logo and timestamp if you really want to. The simpler tools just blur them and turn them into a smudge. I have some ideas how I could get better results, but it would be painstaking work. If the position of the stamps would be the same on all VHTV cams I might even be tempted to create a set of filters for that purpose. Other programs may have better tools for this than those that I am aware of in Shotcut.
