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 >
Just an update after a few more projects: it looks like the issue was somehow caused by audio sample rate mismatch. I haven't gotten any pops or clicks between clips even without fading/crossfading the boundaries as long as I export at ~the same sample rate (the original Audacity project is always 44100 Hz, and the WAV profile I've been using with success in PowerDirector mentions 44 kHz; they might've just rounded for the description, or the 0.2% mismatch isn't enough to cause noticeable issues).

I have not tried adjusting the video export profile to directly produce the MP4. It might be possible, though, if anyone is having this same issue. In my process flow, I do two exports anyways (the second just to add an end screen rather than worry about scaling and moving each element). If the resulting audio has any issues that would require a re-export, audio only takes a few seconds while a video export would be another half an hour, and since the second video pass doesn't require cutting the audio, there is no risk of new clicks being added.

On another note, since I'm doing it this way, I've also started leaving all of the breaths in for the original edit, and just edit them out using audacity while the video exports. I find the audacity envelope tool to be much better for that, and if that 0.2% mismatch ends up letting a pop through, I can just get rid of that in the process.
Quote I'd recommend working with an existing project and making changes now so you'll be prepared for the next session.

Sure, change the audio bitrate in Audacity and see if that helps.

In my experience, this issue was usually intermittent and would often show up in a different place if I produce the project again, so it was really hard to know for sure if things were going to work properly or not.


I get your point, and agree that it is inconsistent regarding how which cuts are affected on export, but the reason it's an issue for me is that each of my projects has dozens if not hundreds of cuts, so I'm confident there will be clicks regardless. For instance, the video the waveforms in the images were from had 10-20 audible clicks and 3-5 REALLY bad ones for both the MP4 and WAV files.

The projects are all in my free time regardless, so it's not like I'll be worrying about missing a deadline or anything. It's just a matter of whenever I get it ready to export (it's already recorded).
Quote There have been issues like this with various PD versions over the years. The first thing I'd do is to make sure you have the final update for PD16, which is build 3424 from 2018. You can download the patch from here.

If the issue persists, try using Handbrake or similar free 3rd-party app to convert the audio to WAV instead of MP3 on all your existing clips. Ideally you'd also be able to change the recording format on your device so you wouldn't have to convert each clip going forward.

In the past when I had this problem, I was able to use audio fade-outs and then unlinked the audio from my clips to extend one past the start/end of the other clip so that the fades were inaudible, but the fact that you can't do that implies that maybe something else is going on.

Another option is to change the output audio profile and see if producing with LPCM or Dolby Digital or AAC (any format other than the one you're currently using) makes a difference.

Last option is to see if using the GPU for producing is somehow causing the issue. Disable hardware decoding from Preferences | Hardware Acceleration and uncheck the Fast video render technology box before you produce.


That was a very quick reply. Thank you.

Unfortunately, I am on build 3424, and I record my mic with Audacity and always export it as a WAV already.

I'll try changing the output audio profile for my next project and seeing if disabling hardware decoding helps if that doesn't... I'm not very hopeful, though... I'll let you know if any of those fix it. Otherwise, I'll just start exporting before silencing breaths and handle both the breaths and clicks in audacity after.

It's really weird that it adds a click in the waveform, though... Could it be an issue with the audacity sample rate or something? I don't know how PowerDirector handles audio.
Yes, it sounds like an issue of not splitting at a zero crossing, but I don't think it is.

My projects tend to have a lot of audio cuts. Hearing pops/clicks due to mismatched waveforms would not be strange. It would be nice if that were fixed on export, but it's a complicated problem to solve in the background, although I thought it was strange that the pops occured at different splits and with a different amplitude every new export. I figured PD was just (unsuccessfully) trying to fix it.

Under the assumption that it was the zero crossing issue, I faded each clip to -inf both in and out. There were still pops. So I Made sure they were at -inf for at least a full frame for each clip start and end. Inexplicably still pops on export. I extended each and crossfaded each clip. It seemed to help, but there were still some feint pops on export. Making sure to apply a very aggressive noise gate to the audio before exporting as a WAV to use in PowerDirector on top of all of the other steps seemed to fix it.

I'm starting a new job (to clarify, editing videos is my hobby, not my job. The new job is just restricting the time I have available for the hobby), and won't have time for all of that work for small clips, so I decided to just try it with the noise gate, and the result was very bad. So I opened the MP4 in Audacity just to see how the waveform looked at a particularly bad boundary, and decided to also produce a WAV with PowerDirector to see if it would be better.

The waveforms were shocking. It was not a matter of mismatched waveforms. PowerDirector was ADDING clicks. In the attached images, the upper muted track is from the MP4, and the lower track is the WAV. As you can see, the area surrounding the click is very quiet, but there is an abrupt, click at the cut, often a half square wave. As you can see, the amplitude and length of the click is not at all consistent.

Is there a way to fix this, or will I need to export any track with cuts, import it to Audacity, go through it to remove any audible clicks, re-import it, and then produce the video file?

EDIT: It looks like the issue was caused by a mismatch between source and export/production audio sample rate. See my May 25th, 2023 response for more info.
Rick,

I have both 13 and 14 installed and haven't had any stability issues. My "solution" is to do all of the editing in 14 since it's so much better in that regard. I then create the .MP4 file, but also create a .WAV audio file with the same project. With both of those, I open up PD13, drop them both at the begining, mute the .MP4, and produce the finished .MP4 using the same profile settings that were used to make the intermediate .MP4. It only takes a couple of minutes longer, give or take your system's load time, since the second production is done with SVRT. That way you get all of the benefits of PD14 editing without the nasty audio problems.

Hopefully they hammer that problem out before 15. Ideally, they'll patch 14 because I really would rather not upgrade unless there are significant improvements... The entire reason I chose PowerDirector was due to its one-time fee. Yes, Cyberlink has the best forum for help and support that I've ever seen, but the subscription based editors seem to have more tutorials and even quite a bit of open-source resources and plugins.
So, I was capturing some webcam footage to complement a couple of videos, and the capture was scrunched to the top left corner while also turning the light parts bright red and basically blacking everything else out. Have you guys seen this before?

The preview was normal, and the capture is usually fine. The only difference is that I had PD open on a second monitor, and my primary monitor had a fullscreen program with a significantly smaller aspect ration, so I expect that was what happened.

Since the footage isn't critical to the videos, this is more of an "is there a known way to recover the files" post rather than a request to help troubleshoot or figure out a way to recover them. That's also the reason that I've only attached an image of one of the videos rather than a video itself.


TL;DR
Has anyone seen this happen before? If so, is there a way to fix it? Otherwise, don't worry about it.
Yeah, if it IS a 6700, that one would be the way to go, but without confirmation about what kind of graphics card the Dell has, I can't really say for sure. I wouldn't be surprised if it just relied on the 4k "capable" integrated graphics on the 6700.

It's also worth noting that the HP has two empty memory slots, so if you decide you want more RAM, you can just slap it in.


Edit: stevek's suggestion could be better.
Depending on what that i7 is, that HP one with the 960 could be pretty good.

P.S. Flight sims are games too
Oh yeah... I forgot there were off-the-shelf machines that the companies market as gaming computers, I thought Alienware laptops were all highly customizable. I could be wrong, though.

If you don't have a friend who can help, do that.
I agree that getting a new computer entirely would be the best course of action.

While I agree that getting a custom gaming rig or workstation is usually the easiest and often most reliable option, a chunk of that price is often put towards purely aesthetic uses, such as fancy lighting, and color coordination along with a snazzy case. They may also only offer components with features that you'll never use, such as fan control and in-depth temperature monitoring.

My suggestion is to find someone you know who's tech savvy. Ask for help picking out the parts given your workload, and they might even be willing to put it together for you for free. It's not difficult, so it wouldn't surprise me. The hardest part is figuring out what components you'll want to use.

If PowerDirector is the only taxing program you'll be running, the system requirements for it would be the only important ones. I would personally probably take a couple steps up from the recommended specs, though.

Back when I was putting together my old desktop, AMD/ATI cards were the best choice (for your wallet, at least) for low-mid range cards. Is that still the case, or would an older series NVIDIA card be preferable? With regards to NVIDIA, if you can wait a bit to buy a card, I would. Pascal just came out this past Friday and I have pretty strong feeling that older series cards will take a heavy price hit.

Does anyone know how Skylake CPUs are with PowerDirector? Can you do hardware acceleration with their surprisingly powerful integrated graphics? If so, that would DEFINITELY be the way to go if you're trying to keep the price down. If you did that, you wouldn't need a graphics card.
Yeah, I was referring to all companies, not just Cyberlink. I haven't had any problems with them.
The original audio was recorded as mono, so theoretically, the channels are identical. I originally did it with just the right channel as you did with yours to replicate the "experiment" as best I could, and it still had the "static." Since they were different bitrates, and different programs that produced them, I expected the small amount of desired audio to remain as it did.

As I said, it IS a tiny, almost completely unnoticeable amount of noise relative to the important audio, but it's perceptible enough to potentially drive away potential viewers. Do you hear it now, or am I just crazy?

Considering it only occurs when other sounds are also occuring, I think it may be a result of residual noise that wasn't removed during noise removal of the original recording. It seems rather strange, though, that the program would amplify that noise, doesn't it? How would that even happen? Could it be aliasing or something similar?
Number 6 and of stevek's response were my favorite. I think it's probably mostly a problem with the background processes and other programs running, as has been stated. While the system itself seems like it would be straining pretty hard running such a complex program, the original poster is only having problems while saving. Since you're waiting two and a half hours to save, there are a lot of things that are building up in your RAM. I'm not sure how PD handles saving a project, but from what I've seen in my experience, it causes a burst in RAM usage. If it runs out of RAM to use, it will crash.

There IS a way to allocate hard drive space for a sort of pseudo-RAM, but I'm not familiar enough with it to explain it over a forum, and I would definitely not suggest a novice to attempt it without clear and simple instructions. It's significantly slower, especially if you're using HDDs, which it appears you are, but it will prevent crashing. If anyone else knows what I'm talking about and would be comfortable walking MAURO through it, that could be a quick fix.

Biggest tip I can give, though, is relevant to every program in existence:
Save early, save often.


It may just be a result of being part of the younger generation, but I don't get why people would go to customer support over an active forum when it's available for software and technical problems. Sure, they're "trained professionals," but too often their "training" consists of being friendly and pointing the customer to the FAQ. A lot of companies even monitor their forums and will respond in said forums.
Weren't thought the only cards with those driver issues pre-Kepler? I guess PepsiMan's experience disproves that, though.

Quote: Why can you not go into the BIOS and disable integrated video? Then your video card is only one available?

Quote: You can't do that on a laptop. The nvidia card goes "out" to the internal LCD screen via the intel chip.

As JL_JL said, this isn't always the case. An example that comes to mind stemmed from an issue with GTA V. I'm pretty sure that it's since been patched, but at launch, if the client detected integrated graphics on your system at all, which included any Optimus system such as almost every gaming laptop, the client wouldn't let the game run. The most common suggested fix to this was to disable the integrated graphics through the BIOS.

Just a side note to defend gaming laptops: I think everyone knows that desktops are better than laptops in every way I can think of except for one, but that one is often critical. As an engineering student, I needed to get a performance laptop to run taxing programs like simulations wherever I go, be it at home, in class, in lab, or on vacation, so I spent a bit more on a graphics card rather than buy an entirely separate rig. Portability can be very important. (To avoid any risk of derailing this thread, I'll say now that I won't reply to a comment that is only about this last paragraph.)
Sorry to everyone for the consecutive posts. Last one before someone else responds.

Quote:
I've done some preliminary subtraction of the right channel of PD14a.m4a with the right channel of PD13.m4a and it's near silence, suggestion the two audio files are nearly identical and the reason several have posted they hear no static anomolies.

Is this what you got? To me, the voice is basically gone, but the noise is still there, and almost completely burries the remains of the vocals.
Here's the DxDiag file with everythingI generally have connected during editing, with the exception of headphones and monitors.

I've been waiting for Pascal to launch to put together a decent desktop, so if that would be a fix, that'd be nice. I wouldn't be surprised if part of it was a result of my old hardware.
Checked through YouTube on a different computer, same problem, so I highly doubt it's a 3rd party codec issue during playback.
I'll generate the diagnostics file in a minute.

Quote: You can attach a 3-5 second sample of the video file. The forum should accept it if it doesn;t start acting up again.

It's more of an issue with my internet speeds than with the forum, and the video wouldn't benefit it.


Quote: Have you tried an audio file of mp3 format in the project and what are the results when you 'produce' that?
When you say 'During playback...' are you saying that when you see and hear it during 'preview' in the editing room?

Yes, I've tried using mp3 files. It has the same grumble, but with a noticeably poorer quality overall.
By "during playback," I indeed meant during the preview in the editing room, which is to be expected as it should have been playing back the original audio, but may have indicated that it wasn't PowerDirector misreading the file.


Quote: Not sure what you mean here, can you share how you changed input codecs, I assume audio codecs associated with your source video file?

Yes, I tried exporting the Audacity project using varous codecs and containers, such as various wav formats, mp3, and tweaked things like the sample rate.


Quote: I'm not sure I understand what you mean by adding artifacts.

It was just a hypothesis. I was suggesting that it may have produced the same way that PD13 did, checked the bitrate, and just added filler to end up with a file that has the desired bitrate. I tried outputting the audio clip with a bitrate of 95kbps, and if I remember correctly, the artifacts weren't there, but the quality was abysmal.


Quote: I too found no discernable difference in the audio channels you posted. The playback effect you note might be a side effect of installed third party codec packages if you have installed some.

That could potentially be the case, and I'll need to try it on a different computer. Would that affect playback through streaming the videos from YouTube, though? I suppose it could... I'm really hoping that's the case.



It may be one of those "once you hear it, you can never un-hear it" sorts of things. It's pretty faint, and sounds quite a bit like someone moving around in an old leather/pleather chair (I actually thought I heard in a PD13 render of a clip, but it turned out to be me shifting in my computer chair). I can imagine it getting really annoying in the same way that a dripping faucet, or the hum of a cheap power supply would.
Back in December, I upgraded to PowerDirector 14 Ultimate using PowerDirector 13 as a base. I ran into this issue then, but just went back to using 13 thinking it was either an issue that would be patched, or an issue with Windows 7. I've upgraded to Windows 10, downloaded the latest build, and the issue still persisted. Today, I uninstalled both 14 and 13, then reinstalled them, and the issue still persisted.

During playback, the audio sounds fine. However, when I produce an mp4 video, the resulting audio has a fairly large amount of static whenever there is audio. I've tried using quieter and more dynamically diverse tracks to no avail, so it definitely isn't clipping. The strangest part is that PowerDirector 13 doesn't have this problem, and PowerDirector. I've attached the original WAV file produced by Audacity, and the M4A files from PowerDirector 13 and PowerDirector 14 of a sample to show this issue. I would have attached MP4s, but the file sizes would have been bulkier than necessary. I also produced a file with PowerDirector 14 using the file produced with PowerDirector 13, which had the same issue, but the forum will only let me upload three files. All of the M4A files were produced with a bitrate of 384kbps selected.

I've tried using various input codecs, containers, and formats, including using a WAV file produced by PD14, but none of them solved the issue. I then tried using the file produced by PD13 as the file for PD14 to use since that would be as close to the output format as possible, but the PD14 output still had the same artifacts, so I think it's pretty safe to say that the input file format is not the problem.

While examining the resulting files, I noticed that the bitrate of the PD14 files were 385kbps while the bitrate of the PD13 file was 97kbps. After seeing that, I thought the issue could have potentially been that PD14 was locking the output bitrate by adding artifacts while PD13 allowed the bitrate to drop below the designated bitrate if need be. I couldn't figure out how to allow PD14 to allow a variable audio bitrate, so this is still a possible solution.

I've tried multiple profile settings for various codecs using an mp4 container, but none of them fixed the problem at all. For the H.264 codec, I also tweaked various settings in the .ini file to no avail. I've tried changing the audio to be a voice track. I've tried all of the TrueTheater Surround settings. I can't get anything to work.

I'll probably start editing in 14, produce the video, produce a WAV file, then put them together in PD13 until someone smarter than me can figure it out. I like all of the new features 14 brings, and I really want to make use of the program I bought... That's really a long process to go through just to get acceptable audio quality, though...

Power Director Ultimate 14.0.2820.0
Long project, huh, gregston?

So, Dafydd (or anyone else who is knowledgeable about the subject), can you also give a brief explanation of what each of the settings actually does, or point me to a place that does? I'm not asking for a crash course, just what the reference IDs are actually referencing. For instance, For the "Pattern Setting" I'd be looking for something like "GOP Pattern" rather than an explanation of what a GOP is. It would also be nice to know what each integer setting does if at all possible.

Specifically, I'd like to know if and how you can toggle between open and closed GOP, but knowing the rest of them may be helpful as well.

Thanks
Go to:   
Powered by JForum 2.1.8 © JForum Team