Announcement: Our new CyberLink Feedback Forum has arrived! Please transfer to our new forum to provide your feedback or to start a new discussion. The content on this CyberLink Community forum is now read only, but will continue to be available as a user resource. Thanks!
CyberLink Community Forum
where the experts meet
| Advanced Search >
Problem with Profile Analyzer
optodata
Senior Contributor Location: California, USA Joined: Sep 16, 2011 16:04 Messages: 8630 Offline
[Post New]
I've gotten used to clicking on the Profile Analyzer whenever I use clips from a new source so that I can get the most optimized results. However, when I tried that with 4K HEVC clips from my new Blackvue dash cam, I got the Production unsuccessful message:



There was no problem with the CES logger registry key, and I couldn't produce to this profile at all using QuickSync or NVENC or even with the CPU by itself. I could produce to the default HEVC 4K 30p profile without any issue, so I extracted those settings and compared them to the new profile the Profile Analyzer had made.

The differences are shown in orange:



As you can see from the notes in red, after a bunch of testing I found that the problem wasn't any of the video settings at all - it was the <Audio SamplesPerSec> setting of 32. If I set that to 16 or lower, the video produces normally, but if it's 17 or higher the error occurs.

I'd like to see if other people get the same Best Matched profile when using the Profile Analyzer on this same clip, and also whether changing <Audio SamplesPerSec> to 1 - 16 allows PD to produce. Both the clip and the original Best Matched profile.ini file are in this OneDrive folder.

YouTube/optodata


DS365 | Win11 Pro | Ryzen 9 3950X | RTX 4070 Ti | 32GB RAM | 10TB SSDs | 5K+4K HDR monitors

Canon Vixia GX10 (4K 60p) | HF G30 (HD 60p) | Yi Action+ 4K | 360Fly 4K 360°
ynotfish
Senior Contributor Location: N.S.W. Australia Joined: May 08, 2009 02:06 Messages: 9977 Offline
[Post New]
G'day optodata -

Man - it must have taken some digging to determine that the issue was related to the "Audio Samples per sec"!

I can confirm what you've stated. Best Matched Format for the Blackvue dashcam clip sets that value at 32 & production fails. When it's edited to 16 or below, production is successful.

Scrolling through the other profiles in my Profile.ini, I found that most have "<audio samplespersec="">1", but some have "<audio samplespersec="">64" (they're AVC H.264 MP4 with LPCM audio).

I produced your Blackvue clip successfully using those "64" profiles.

Anyway - it's pretty obvious it's an issue if PDR sets [Edit: a misplaced apostrophe! surprised] best matched profile & it fails to produce! undecided

Cheers - Tony</audio></audio>

This message was edited 1 time. Last update was at Sep 11. 2020 16:14


Visit PDtoots. PowerDirector Tutorials, tips, free resources & more. Subscribe!
Full linked Tutorial Catalog
PDtoots happily supports fellow PowerDirector users!
optodata
Senior Contributor Location: California, USA Joined: Sep 16, 2011 16:04 Messages: 8630 Offline
[Post New]
Thanks very much for checking Tony, and for doing your own testing literally above and beyond my "16 or less" finding

I've reported the issue on ticket #CS002219471, although I imagine the people at CL are busy readying everything for the expected launch of PD19. I'll report back once I hear from them.
optodata
Senior Contributor Location: California, USA Joined: Sep 16, 2011 16:04 Messages: 8630 Offline
[Post New]
UPDATE: Tech support has confirmed the issue and they've forwarded it on to engineering. They also recommend using a default profile if anyone else runs across this issue.
Shakeyleg [Avatar]
Newbie Joined: Sep 17, 2019 20:56 Messages: 23 Offline
[Post New]
Quote Thanks very much for checking Tony, and for doing your own testing literally above and beyond my "16 or less" finding

I've reported the issue on ticket #CS002219471, although I imagine the people at CL are busy readying everything for the expected launch of PD19. I'll report back once I hear from them.



Hello Optodata! I saw you mentioned the launch of PD19 and thought about all of the new graphics cards that are coming out at this time. My computer really struggles with the 3840X2160 videos I am editing. I film concerts and at times have 6 different streams I am editing with on the timeline and often have to uncheck half of them just to be able to see my edits on the timeline. (Even at High Preview Resolution which isn't great) I know you have great knowledge with the technical aspects of PD so I was wondering if you might glance at my specs and tell me if you think one of the new GPUs that are coming out would fix my problem or if I need a whole new computer? I am running PD 365 with 2 SSD drives and 1 NVMe M.2 SSD.

Motherboard - B450 Tomahawk (MS-7C02)
Processor - AMD Ryzen 7 2700X eight-core Processor 3700 Mhz, 6 cores, 12 logical Processors
GPU - Radeon RX570 - 8GB GDDR5 Memory
RAM - 32GB

I guess I am hoping you will say one of the new $499 GPUs might fix my problems!

So thankful for all of your input on here. I also just saw your tutorial on moving a PIP object along a map. Good stuff!
optodata
Senior Contributor Location: California, USA Joined: Sep 16, 2011 16:04 Messages: 8630 Offline
[Post New]
Your question isn't really related to this topic, but briefly your CPU seems fine.

I don't have any experience with AMD GPUs so I don't have any idea of whether you'd benefit from a new card or not. I do know that typically, the CPU is responsible for most of the workload when editing, so using shadow files when you have 6 concurrent 4K videos on the timeline is pretty much required. A new GPU would likely only make a difference when producing.

Another option would be to convert your clips into a lossless, unencoded format like MagicYUV which can be read and displayed much quicker than encoded clips can. The main downside is that the file sizes can easily be 10-15x larger, but your NVMe drive will easily be able to handle all that data, even six 4K streams at once (possibly even in Full HD preview resolution).

You may find this thread useful: https://forum.cyberlink.com/forum/posts/list/79660.page

YouTube/optodata


DS365 | Win11 Pro | Ryzen 9 3950X | RTX 4070 Ti | 32GB RAM | 10TB SSDs | 5K+4K HDR monitors

Canon Vixia GX10 (4K 60p) | HF G30 (HD 60p) | Yi Action+ 4K | 360Fly 4K 360°
Shakeyleg [Avatar]
Newbie Joined: Sep 17, 2019 20:56 Messages: 23 Offline
[Post New]
Quote Your question isn't really related to this topic, but briefly your CPU seems fine.

I don't have any experience with AMD GPUs so I don't have any idea of whether you'd benefit from a new card or not. I do know that typically, the CPU is responsible for most of the workload when editing, so using shadow files when you have 6 concurrent 4K videos on the timeline is pretty much required. A new GPU would likely only make a difference when producing.

Another option would be to convert your clips into a lossless, unencoded format like MagicYUV which can be read and displayed much quicker than encoded clips can. The main downside is that the file sizes can easily be 10-15x larger, but your NVMe drive will easily be able to handle all that data, even six 4K streams at once (possibly even in Full HD preview resolution).

You may find this thread useful: https://forum.cyberlink.com/forum/posts/list/79660.page


Great advice sir as always. I haven't had much luck with shadow files and see some say turn them on and some say turn them off. I will turn them on and give that a try again. I was hoping it was a hardware issue as that would be a simple fix but if shadow files help that is much cheaper and easier.
optodata
Senior Contributor Location: California, USA Joined: Sep 16, 2011 16:04 Messages: 8630 Offline
[Post New]
The issue is still present in the latest PD365 19.1.2321.0 update, so the workaround of using a default output profile still applies.

This message was edited 2 times. Last update was at Nov 30. 2020 21:23

Powered by JForum 2.1.8 © JForum Team