The problem occurs regardless of the final format. Previously, selecting an Intel graphics card resolved the problem of improper interleaving of the material. Now, unfortunately, 2 cards spit out the material in the same way. Interestingly, Mercali at 50i frames the video. It is not properly stabilized and is shaking. In the old version of X12, the problem does not occur at all. Do you have any idea how to solve it?

@Pawel-Rybinski

hi.

I can't really see what is going on in one video as at present it is only showing on YouTube at 320p so not enough detail to make out anything, but on the second video the stabilization is not shaking but has the 'jellow' effect which is a warping effect as the camera angle changes as you pullback from the shot. To stop that the camera would need to rise as you step back to keep the camera angle the same or you should use a zoom lens and stand still while zooming out.

The parallax view changes within the 2D image with the change of angle. It is something extremely difficult to correct for frame by frame.

You could try changing the camera setting to something else other than universal to see if it can be corrected better but it is better not to change camera angles when shooting. It is not really the fault of the effect but this version needs setting up more carefully and fine tuning.

Personally I would have taken that shot walking towards it as it is easier to keep the camera angle the same than when going backwards. Then in the editor I would have reversed the clip. Professionally these shots are easy to take with the aid of a steady-cam harness or hand held steady-cam mount and plenty of practice. But using something like a mirror-less or DSLR camera without one is very difficult indeed.

Ray.

@Pawel-Rybinski

hi.

I can't really see what is going on in one video as at present it is only showing on YouTube at 320p so not enough detail to make out anything, but on the second video the stabilization is not shaking but has the 'jellow' effect which is a warping effect as the camera angle changes as you pullback from the shot. To stop that the camera would need to rise as you step back to keep the camera angle the same or you should use a zoom lens and stand still while zooming out.

The parallax view changes within the 2D image with the change of angle. It is something extremely difficult to correct for frame by frame.

You could try changing the camera setting to something else other than universal to see if it can be corrected better but it is better not to change camera angles when shooting. It is not really the fault of the effect but this version needs setting up more carefully and fine tuning.

Personally I would have taken that shot walking towards it as it is easier to keep the camera angle the same than when going backwards. Then in the editor I would have reversed the clip. Professionally these shots are easy to take with the aid of a steady-cam harness or hand held steady-cam mount and plenty of practice. But using something like a mirror-less or DSLR camera without one is very difficult indeed.

Ray.

Ok, I'll look for it. I try to show the shots without people.

@Pawel-Rybinski

hi.

I can't really see what is going on in one video as at present it is only showing on YouTube at 320p so not enough detail to make out anything, but on the second video the stabilization is not shaking but has the 'jellow' effect which is a warping effect as the camera angle changes as you pullback from the shot. To stop that the camera would need to rise as you step back to keep the camera angle the same or you should use a zoom lens and stand still while zooming out.

The parallax view changes within the 2D image with the change of angle. It is something extremely difficult to correct for frame by frame.

You could try changing the camera setting to something else other than universal to see if it can be corrected better but it is better not to change camera angles when shooting. It is not really the fault of the effect but this version needs setting up more carefully and fine tuning.

Personally I would have taken that shot walking towards it as it is easier to keep the camera angle the same than when going backwards. Then in the editor I would have reversed the clip. Professionally these shots are easy to take with the aid of a steady-cam harness or hand held steady-cam mount and plenty of practice. But using something like a mirror-less or DSLR camera without one is very difficult indeed.

Ray.

Pls try this

@Pawel-Rybinski

OK.

That is what you see happening in the editor which may drop frames if the computer can't cope with displaying the project effects real time. What does the exported video look like?

Ray.

The video material exported to x12 with the use of Mercali V5 looks perfect. If you want to compare, it will show the export of the same file to x12 and x13 on another video.

@Pawel-Rybinski

Sorry. I'm not sure if we are talking about the same thing at the moment.

What does the rendered finished file look like when exported / rendered, from VXP13 using Mercali v5?

So is the complaint abut how the files are playing within VPX 13 or how they appear in a finished video?

If it is about how smoothly the video is showing in VPX 13 then it is down to the additional strain the program is putting on the processing parts of your computer system. This is a normal symbiotic relationship between newer versions of programs the their operating requirements. Although the running specs may be similar a newer program will always demand more processing power than its predecessor. The increases in performance promised often rely on newer or higher performing components. The promised 'improvements' won't come from the program alone unless the system it was placed on was above the minimum requirements for the old version. This is also true for any change in plugins or offered third party effects. There is no doubt in my mind that Mercali V5 needs more processing power than the earlier offering.

If on the other hand the exported / rendered project does not look OK, then there may be alterations that can be made to improve the results but expect slower than promised export times.

@CubeAce

Ray, What does the rendered finished file look like when exported / rendered, from VXP13 using Mercali v5?

It's shown at the end of the video above.

There is no doubt in my mind that Mercali V5 needs more processing power than the earlier offering.

V5 is being used in both cases.

@Pawel-Rybinski

I agree there's something seriously wrong there. It looks to me like the original video is not being processed properly by VPX and that is translating into a bad Mercalli export. The CPU, when you're running the Mercalli Preview, is 56% so not working hard. Mind you, 6gb for 21 seconds is a pretty insane bitrate.

Is the jerkiness apparent when you step through the video frame by frame?

You could try changing the settings for the graphics cards and CPU on the Settings>Device Options tab (Import/Processing/Export).

Re your YT video, I suggest you slow down your mousing and keystrokes; it's great seeing what you're doing but it's very hard to keep up because you're going so fast. Also, I tried to download it (to make it easier to view) but I couldn't with 4k Video Downloader which I normally don't have any trouble with.

@AAProds @Pawel-Rybinski

Hi Al.

The video above is not the one I think I saw earlier.

The main difference I see between the two is the amount of crop used to stabilize the outputs.

Could it be the clip needs to be manually zoomed in more with the new plugin to allow for the difference?

Ray.

@CubeAce @Pawel-Rybinski @johnebaker

Ray, as I mentioned above, the jerkiness appears to me to be visible before Mercalli is applied, when the play marker is dragged along the timeline.

As for zooming in more before applying Mercalli, I would have thought Mercalli would be restricted to using what is actually visible at 100% on the Preview monitor as opposed to the video hanging off the off the edges after a zoom-in (as you would see if you went to 75%).

@Pawel-Rybinski

Just for a giggle, I'd try setting Bottom Field First and see if that improves things. I acknowledge that Mediainfo does report TFF but you never know.

@AAProds

Hi Al.

I was mainly saying to manually set the zoom in feature in the plugin itself a little more as the old plugin seemed to zoom in more automatically. If you are referring to combing effects being visible could that be down to frame speed selection within the project (compared to the frame speed the camera shot at) as well as which field is selected first?

My other thought about the combing is the video is very jerky at those points. What if the refresh rate between fields at that point is not enough or the frame rate too slow for the amount of camera movement? Some in camera stabilisation systems rely on pixel shift within the sensor. It may be interesting to know which camera took the video.

Ray.

In interlaced materials, there are lost frames and incorrect stabilization of the material. This applies to 50i materials. The next material is the comparison of materials after the conversion of X12 vs. X13. Everyone will be able to see what I'm struggling with since the release of the latest version of Magix. I'm just glad that the Magix support promised me to extend the update service until they fix the errors with the program. Unfortunately I still have to work on the X12 version ...

Wait pls for 4K vesion

Exporting to mpeg-4 gives the same result. I noticed when I already uploaded the video. 😳

@Pawel-Rybinski @AAProds @johnebaker

Hi.

Looking at the ProDad site for Mercali V5 and I found this that may be relevant even though technically for a Vegas product the programs share similarities and may be in the other 100 pages of problems reported or so it may have relevant information on the ProDad site.

Ray.

@Pawel-Rybinski

Hi

Looking at the video clip above you are not comparing like for like video export formats.

The screenshot below shows you are comparing a .m2ts (AVCHD) from VPX 12 and an MP4 (h.264 or h.265) from VPX 13

If you look at my reply to your initial question here, I clearly stated that AVCHD (.m2ts) is an Interlaced video format, and h.264 or h.265, which the MP4 file is most likely encoded with, is a Progressive video format.

This is most likely causing the jerkiness and 'tearing' shown in the video clip at the end where you compare the 2 videos. The Mercalli stabilisation has possibly exaggerated the difference between the Interlaced and Progressive formats.

The jerkiness is known to occur when converting Interlaced video to Progressive video, how much depends on the rate of movement in the scene.

John EB

@Pawel-Rybinski @johnebaker @CubeAce

Pawe, to do a valid comparison, as John alludes to, you should export your X12 example to MP4 so that it is a valid comparison with your X13 export. Alternatively, export the X13 version into M2TS. However, I don't think it will reveal a lot; the job of a video editor is to export faithfully the source, with conversion Interlaced to Progressive being it's bread and butter. VPX 13 should be doing a far better job at MP4 than it is here. That MP4 from X13 is almost unwatchable.

I still think that the issue is that VPX13 isn't interpreting the original M2TS properly in the first place, even before Mercalli is applied.

Could you put 00019.MTS onto Google Drive for us to have a look at?

As I wrote earlier, the problem does not concern the format itself, because I dealt with it by outputting the material to 50p. After stabilization, the X13 makes me jelly in the image and despite the 50p setting in the output video, it does not give me the same results as the X12. I gave you a link to the original file and what I redirected so you can check it out for yourself. I don't know what it's caused by.

https://www.filemail.com/d/jkzxbosemamdhdw

@Pawel-Rybinski

Thanks. I see the jello.

Exactly

@AAProds

Hi Alwyn

I must be snow blind - at what time in the video are you seeing the jello?

John EB

All the material is shaking sideways ... you can see it. Compare the result files. From the beginning on the grass and then the roof of the building was torn off. It shows. The top frame frames a different frame than the bottom one and the whole image is shaking.

@johnebaker

John, frame 201 area: the top and bottom are moving separately. It's awful.

I just did a 50P export using Mercalli 5 and MEP 2021 and it's smooth as silk.

Unfortunately, I don't have Mercalli on my MEP2022 machine.

I checked my daughter's computer and the problem is the same. This is clearly the fault of the X13 version. The computer is on AMD Ryzen 5 3500 16 GB of RAM and the problem is occurring.