CyberLink Community Forum
where the experts meet
| Advanced Search >
Hello,

this is really hell to edit every single audiofile extra. The Normalize function is uselless in case of trying to get f.e. car noises out in dozens of scenes but not out of all scenes.

Missing:

1) EQ presets which can be saved like "Phone" effect, Notch, base Locut, base Hicut
2) main master EQ
3) Master track in MIX

Can someone direct me where I can put this as a feature request?
Quote



The answer is correct, HEVC is problem ... just about uploading the AVC.



Final video here (AVC 5,7K) -
Quote


Hm, now Render without LUTs out a nd still the same errors, looks like render problem



The answer is correct, HEVC is problem ... just about uploading the AVC.
Quote Hello,

are those visible mistakes related to using 2 luts in one video or to HEVC render?



Thank you.

M.


Hm, now Render without LUTs out a nd still the same errors, looks like render problem
Hello,

are those visible mistakes related to using 2 luts in one video or to HEVC render?



Thank you.

M.
Quote Sure. Just use the Scale setting in the regular PiP Designer.


Anyway, I gave up. The reason is - After I go to PiP designer and I put the basic keyframe to initiate scale setting to be applied, it creates visible seam/suture on the backside of the clip regardless that I don't change any scale parameter yet.

Furthermore - after I change the scale settings it somehow ignores the FOV anyway. So it makes me confused and I decided, that I have to go for standard FOV because I spent already hours on this thing without reliable and replicable result.

Guess: I'm not sure if some transtition effect apllied between clips where I'm trying to get the FOV wider can affect this behavior as well, but I guess not. I recognised the seam/suture firstly after I aplied the Interfere2 transition between those clips). But it could be there already, I don't know really. Maybe it is same problem as with the motion tracker ...
Quote Yes. The only thing you have absolute control over in a 360 video is the starting orientation and FOV. If you want to present your viewers with a noninteractive experience, switch to standard (non-360) mode and you can fully control all content aspects.


Yes in overcapture it is working just fine ... but I was struggling how to bring it to life in classical 360, that´s impossible
Quote

You should be able to track in 360. Take a look at this post for more details.


Yes, I found https://www.cyberlink.com/learning/video/820/360-object-tracking-with-powerdirector-16

... but it seems that in case of 360 vids it is working somehow weird - maybe because of the fact, that the user can still orient himself freely using the mouse ... so in fact we are just trying to do the motion tracking for passive users who let the video go as it is without interaction, correct?
Quote Sure. Just use the Scale setting in the regular PiP Designer.


Thanks great ... nice would be if target tracking could work somehow as well but it is not working in 360 mode (I think - but I'm learning).
Hello,

I'm not able to set wider view for default starting 360 view - there are settings in "Tools -> 360 Start view settings" - for all axis rotations but not for the default "zoom" or FOV parameter? Is there perhaps some possibility to wider the FOV as a starting parameter for viewing the 360 vids?

Or other workaround for "how to anchor the wider FOV view in 360 vids"?

Thanks in advance.
Quote I've just created an H.265 profile for the full 5.7K (5760x2880) resolution, and it's now in the shared OneDrive folder. Since your nVidia card can use HEVC for up to 8K clips, you should be able to drastically cut down on your producing time.



Using your 1:34 5.2K test clip, I produced to this 5.7K profile in only 2:22.



Great I will do, thank you for your help.

So If I get it correctly - I just have to copy the content of custom profile.ini under new custom profiles created under H.265? No further changes required?
Quote

I'm not aware of any experimental NVENC features to help with H.264 resolution limit. Switch to H.265 in similar profile.ini jury-rig and then NVENC will be supported at that resolution.

If you need H.264 at that resolution, NVENC will not be utilized, so yes, patience or more CPU capability.

Jeff


I´m not sure if H.265 and 360 vid is ok? If yes, I´ll jump into it immediately ...
Quote

Probably because the profile.ini simply jury-rigged the 5.7K capability that was never intended at this release. Basically, NVENC does not support resolutions higher than 4096 in H.264 with any current hardware, hence your observation.

Jeff


O ou ... so perhaps the right time to test some experimental NVENC features or to armor myself with patience, correct?
Hello,

just hopefully one last trouble ...

It ignores hardware acceleration (as I can see in taskmgr), I´m unable to set hardware encoder when using this 5.7 k custom profile. And 162 hours for 360° vid 45 min is siplme too much and not necessary (I got GTX 1070 8G and 48 G RAM and SSDs).

And I´m unable to check/uncheck the hardware acceleration in production mode of PD when using this custom profile.

There are Encoder type and Encoder mode parameters (within the profile.ini file ) I don´t know what the value should be? Please help ...
Quote Hi,

I think I have the 360 5.7K IN & OUTput working from PD17 and think this warrants a new thread.



Great it works ... https://youtu.be/dOJJIHRKTuc ... although I'm still unable to get proper sound when editing, the result works. So the workaround is to use the proxies in separate program to find out the scenes I/O.
Quote

I'm afraid I don't quite understand your setup. In the other thread, I thought you were converting your 5.7K clips to MagicYUV, and if you do that there shouldn't be any need for shadow files in PD.

Try lowering the timeline preview resolution to HD when working with MagicYUV files, and if you also/instead use shadow files, you have to wait until all the yellow icons turn green before doing any serious editing. This pic shows that only one clip has its shadow files finished:



Yes, this I know already, unfortunately does not work. So I´ll try to remove the shadow files and use only the preview quality.
Hello,

although I have proxies created (720 p) I have to open this proxies (shadow files) during editing in a normal media player separed next to PD to hear sound properly and to be able to navigate throught my footage ...

Why? Because during playback in PD I hear only first word smoothly and then "ke ke k kskskakkskek ekekekekskak kakak eee a a a a s t t t t u u u i i i i o o o f ff h h h h saw s s s s s s " ... so very jerky and definitely not clear sound out of playback within my purchased PD17 at all.

I have VR gear (predator 15, 8 G video RAM on GTX 1070. 48 G RAM and SSD discs only) so really curious what is wrong with it? Please help, this is useless.
Quote

That video has been flattened before you imported it. You need to use different settings in Fusion Studio - make sure you choose 360, and not Overcapture.


Hello, no, it has not been flattened, it works just fine - the mistake was in the steps order in PD ... Before in my failed attempts - I always started PD directly to 360 mode, that was the mistake. Because it started to work after I just:

1) changed the ratio in the upper bar back to 16:9
2) then changed back to 360
3) changed to 360 in view and switched to clip mode ... so timelapses and the vids too from fusion are working perfectly.

As a small bonus for members here, I finaly got the proper sequence of commandline commands to get the 5.7 K export outof gopro fusion studio instead of just 5.2 K. For better results use the 2.0 firmware, but I' m on old 1.8.

So the exact sequence on a comp with Win 10 is:

FusionStudio_x64.exe --front e:\fusion_stitchdir\GF027014.MP4 --back e:\fusion_stitchdir\GB027014.mp4 --width 5760 --videoCodec 1 --pc 1 --stabilization 2 --iq 3 --denoise 128 --globalContrast 128 --sharpness 128 --temperature 128 --synchro --wave e:\fusion_stitchdir\GB027014.WAV --amb 1 -l 4 --output E:\27014.mov

You can then play with the other parameters - there is definitely not good description of the cmdline fusionstudio commands (something is if you try just the command with -h for help) but this what I pasted above is the starting point and then you can play around (the 128 is the middle value between 0 - 255) - I' m realy curious how my first PD export will look like. Very time expensive attemps
Quote

That video has been flattened before you imported it. You need to use different settings in Fusion Studio - make sure you choose 360, and not Overcapture.


Btw: Why to use for export from gopro fusion studio the Cineform and not the Prores ? Might that be the reason, why my export was flattened?
Quote

That video has been flattened before you imported it. You need to use different settings in Fusion Studio - make sure you choose 360, and not Overcapture.


Overcapture export looks at the beginning like the small planet. This export is different, but I understand what you mean. I will try.

And thank you very much!
Go to:   
Powered by JForum 2.1.8 © JForum Team