[Jeff Pulera]

Create NEW SEQUENCE as NTSC DV Widescreen. Drop clip into it. Scale to look correct. Whether you go interlaced or progressive will depend on source clip.

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

Quite simply, you will get best results if you denoise the clips that need it prior to doing any editing. Meaning, apply ONLY the denoiser to the clips, then export using a good intermediate codec like ProRes on Mac, or DNxHD or GoPro Cineform on PC. Now you have a quality source clip that is denoised. Use that new clip to edit and apply your effects to, rather than trying to layer it all together at once.

Denoising takes a LONG time, then if you add the other effects to it, render time increases exponentially in combinations. Faster to do in multiple passes than all at once, and editing is much more responsive without DN applied in Premiere.

Has worked great for me over the years to do it that way. Yes, there is that initial delay to starting the editing process to pre-process the denoising first, but your life will be much more pleasant after that step is done, I promise that.

There have been times when I actually did the Premiere edit first, then denoised original clip in After Effects later when I had time to let that denoiser run overnight, then in Premiere did a REPLACE to swap in the new, denoised clip to replace original source clip without affecting the existing edits! Note this is on wedding ceremonies, so perhaps just two long clips for ceremony, not dozens of clips to deal with. If you attempt the “after the edit” method, do so with a COPY of the Premiere project so you don’t destroy all your work accidentally when trying to replace clips!

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

If you shoot 60fps, and edit in a 60fps sequence, all frames are being “used” already. Therefore, any slowing down of the footage means some frames MUST be duplicated. So if you set clip speed to 50%, each frame will be shown twice. That does not explain tearing, but will get to that later.

So create a 29.97 sequence (30p) and drop that 59.97fps (60p) footage into it. Disregard the request to change frame rate of sequence. And NO NEED to Interpret frame rate – unless you automatically want every single clip to play in slow motion, then you can do that.

When a 60p clip is played in 30p timeline, that timeline is only going to display 30 frames each second, but your clip has 60 per second, right? So what Premiere will do is simply skip every other frame, only playing half the frames, basically treating clip as if it is 30fps. Looks fine. Premiere assumes you want to see normal speed playback so it does this adjustment automatically, allowing you to mix clips of different frame rates together and have them all play normal speed.

Then if you set 50% speed on that clip, that would make a 5-second clip 10 seconds long. You have 300 frames to play with (5x60fps) and 10x30fps = 300, meaning every frame is played once, so you get perfectly smooth slow motion with a new frame played every time, no duplicates. If applying 40% speed to 60fps clip in 24fps timeline sequence, same thing, every frame used once, like the others stated. So then you also export at same frame rate as sequence, either the 29.97 or 23.976 for instance. Clips that do NOT have speed change applied play at “normal” speed, but if you apply the slow speed to select clips, you get that silky slow motion. Good so far?

Now about the tearing – the Warp Stabilizer is a finicky beast. Probably not going to play well with slow motion for various reasons, especially if using 60p clips in 60p timeline, slowing them down (frames duplicated) then also stabilizing, lots going on there!

What I would do if you need to stabilize a clip is this – put 60p clip in 60p timeline, and stabilize. Now EXPORT that clip as 60p using a good intermediate (lossless) codec like DNxHD (.mxf) or GoPro Cineform (.mov). Then, take that new already stabilized 60p clip and put it in the 30p (or 24p) sequence and add your slow motion or 50% or 40% as applicable. Can pretty much guarantee that will bypass the tearing issues you are seeing.

Hope this makes more sense as explained

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

Hi Monty,

First and foremost, please tell us about your computer hardware, as the answer may lie there.

Also, are any effects at all being applied to source, or is this a straight transcode of material?

I’ve never encoded to HEVC before, and am not on my high-performance editing machine right now, but did just attempt on a Core i5 office computer as a test. An 8-second AVCHD clip took about 17 seconds to encode, so about twice real time. Then figure at least 4x slower for 4K since image it is 4x larger than HD, so I guess that makes 8x longer than real time to encode. That still doesn’t explain the 75 hours, unless the computer is severely under-powered or there are maybe some effects filters involved.

Also – I see posts like this all the time -“my render took 3 days”. A word of advice, if you see that happening – ABORT and seek help sooner rather than later, could be an easy solution to reduce render time. That is typically not normal for any kind of render and I can’t imaging having my machine tied up that long.

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

If you can please post a screen grab of the Export Settings panel, that can be quite helpful, as that will show both the Sequence (Source) and Export settings next to one another for checking/comparison of settings used.

I mentioned HDV clips earlier, at the 1440×1080 size with 1.333 PAR, but I should note that if you drag such a clip into a 1920×1080 (1.0) sequence, it WILL fill the screen with no black bars since Premiere is intelligent enough to figure out what’s going on. So if your clips are still having black bars, the PAR of the clip is possibly misrepresented in the video file itself and if that’s the case, you can right-click the clip in Project Bin and use Modify > Interpret and actually set the PAR manually to get better result if needed. Or again, maybe the PAR is wrong in the Sequence.

Make sure the Program window sizing is set to FIT, otherwise that can sometime mess up the display.

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

A square pixel is just that – square – height and width are the same. In your Sequence Settings, check what the Pixel Aspect Ratio (PAR) is set to, should be 1.0 for square pixels, which is what most 720p and 1080p video formats will be using.

1920×1080 only creates an image with 16:9 aspect ratio when pixels are square. If the pixels are rectangular, that will throw off the dimensions.

Remember HDV video, from tape? That actually recorded video at 1440×1080 resolution, however the PAR was 1.333 meaning the pixels were rectangular – wider than tall. The wider pixels made up the difference to still create a 16:9 output. Check out this math: 1440 x 1.333 = 1920. It’s like magic.

So you may have some sort of mismatch going on. Either the footage has an issue (or is being incorrectly interpreted by Premiere for some reason) or the Pixel Aspect Ratio (PAR) in Sequence is not correct.

There was a Panasonic HD format that recorded at 1280×1080, using 1.5 PAR, which would still provide 1920×1080 playback (1280 x 1.5 = 1920). Formats like this are called “thin raster” – skimping out on some scan lines! True 1920×1080 video is “Full Raster” HD, no skipped lines.

Thanks

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

Hi Ken,

Not sure what you mean by previewing before Encore. Are you editing in Premiere? If so, then the proper workflow is to export as MPEG-2 DVD format, using one of the available presets such as NTSC DV Widescreeen Progressive. This special flavor of MPEG-2 is the ONLY thing that can ever be on a Video DVD in the end – if you supply any other format to Encore, it will simply transcode to MPEG-2 anyway.

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

Hi Michael,

Thanks for sharing the error message, but you didn’t really address any of Todd’s very valid questions – help us to help you.

If you can please post a screen shot of the Export Settings panel, that can often be quite helpful. I notice the error says “Accelerated Render” – try turning off GPU Acceleration and export via Software mode and see if that might work.

Besides Todd’s questions, what effects might be applied to that clip that is between 3 and 7 seconds into timeline? Is it a still image or video clip? Details needed

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

I’ve never tried PluralEyes or new Adobe sync feature. Depends on workflow – if I’m doing a 2 or 3 camera event like a wedding ceremony or play or recital, I only have say 2 long tracks to sync together – quickly do it ONCE for the project, and DONE.

I think where PluralEyes is really meant to be useful is for those who say shoot with a DSLR and record all audio to a secondary audio recorder for better quality, so they could have many dozens of video clips with audio, that need to be synced with all the secondary audio clips. That would be a nightmare to do manually!

Right tools for the right job

Thanks

Jeff Pulera
Safe Harbor Computers

[Jeff Pulera]

Hi Mark,

First, create a new unique folder on video hard drive for the footage, like “Card_01” or whatever. From SD card, copy EVERYTHING to that new folder, keeping all folders structure from SD card intact. It is important to do this, rather than just picking certain video files.

Create a NEW FOLDER for each SD card you need to copy over.

Once card is copied over, in Premiere you will use Media Browser to import, rather than File > Import.

Media Browser will use metadata found in other files from SD card to best figure things out – for instance if you made a long recording that camera broke into smaller 2GB or 4GB chunks (spanned clips), Premiere will import so that you get one long clip to edit with, no breaks between the pieces. Just a lot of little benefits like that by using this method.

Thanks

Jeff

Jeff Pulera
Safe Harbor Computers