Corel Pinnacle Studio 21 Ultimate

VidThreeNorth · Aug 24, 2018

On Windows 8.1 64-bit Asus M32BF AMD A8-6500

Project:
'20180714 Toronto JCCC KyoWa Taiko "Shunan"'
""

Recorded with Canon R40 Full HD, (the software works the same for videos taken with most common cameras)
Output: MP4, 24mbps 30 fps
- Monopod

NOTE: Some setting might be wrong -- I did not write them down.
Used in camera stabilization, standard mode

Sound Auto Level
Mode [P? or Sports?]
Auto Focus, auto exposure (no in camera compensations used)

My choice of editing software is limited because I need 3D coverage but I cannot spend a lot. Until recently I have used Magix Movie Edit Pro and Pinnacle Studio Ultimate. Magix stopped supporting 3D in MEP probably because they want people to buy Vegas, so it is imperative that I get more accustomed to Pinnacle. My earliest tests with Pinnacle for 3D failed due to a bug in earlier versions. I have not tested it for 3D lately.

This is the first project that I completed using Pinnacle Studio 21 Ultimate. As I mentioned elsewhere, my first test was a failure because Pinnacle did not support "user defined aspect ratios" and I could not output a 4:3 screen format for 720P (2D).

This project was simple. It could have benefited by re-stabilization, but overall it was steady enough to release as-is. "Automatic" video correction has lighted the image and might have made other minor changes.

Source File specs:
The source file was 1,819,587,843 bytes for a length of 10:10.
Original audio bitrate was 254 kbps.

The biggest problem I ran into was that the default 1080P output estimated a file size over 5GB, which I would not be able to store locally. I decided to reduce the output to 720P, which resulted in a file 653,512,022 bytes (7 min, 29 sec) at 30 fps at Video 11421 kbps, audio 191 kbps (2 channel stereo 48 kHz), total 11613 kbps.

The reduced audio bitrate noticeably degrades sound quality. I will have to check and see if I can set this to the source quality in the future. The difference in total file size is negligible.

Camera Work:

This was not my best work. As has been typical lately I did not have much prep time. My camera choice was either Sony (CX240) or Canon (R40 or R70). The Sony actually has better sound, but the Canon general have better image quality. I was not sure if there was going be be outdoor folk dancing (weather was not certain), but I chose Canon based on that chance. Low light sometimes favours the Sony.

Background:

I bought a pair of Canon R40s back around 2015 for 3D work. I have been using one of that pair for 2D work. Recently I bought an R70 for 2D work because I was worried that I might break of the the R40's. I am not sure if the R70 can be used as a matched camera with an R40, so the idea of continuing to use an R40 until one breaks and swapping in the R70 for 3D as a method of stepped upgrading is not one I am currently pursuing.

Known Issues:

The R40 lens is an issue. It is labelled an "HD" lens, and I believe that that was its intention. In particular, I think that Canon did design the lens for 1280 x 720, but made sure its performance was "over spec" so that it could cover Full HD 1920 x 1080 eventually. But at Full HD it is a bit compromised. It is noticeably a bit soft overall and in some situations chromatic abberation is quite bad. But in well lighted situations, it give a very noticeably better image than the Sony. As far as I know, the R70 (and even the latest R80) lens has not been improved.

Beyond that, I did not have a chance to check the performance schedule, and I have not had any recent experience with presentations by this organization. So I did not have a "feel" or any kind of recording plan for this show. As expected, this lack of preparation does show up in this video. But it is not so bad that the results embarrass me.

I ended up in a location that was further away than I like, which resulted in my having to chose between small performers on the stage or gambling with panning and zooming to cover the performers. I chose to pan and zoom, but with moderation. It could have turned out better, but not really by that much.

Frames:

"02-Src-19h40m19s544-C1.jpg"
This is a frame from the source video clip. It is not a match of the following clips, but it is close. The drums in particular can be compared against the output file frames.

"01-1080-13h11m31s492-C1.jpg"
This frame is from a test output at Full HD. The "automatic" video adjustments in Pinnacle did a nice job overall, but the output file size was too big for me.

"01-0720-13h28m13s905-C1.jpg"
This is a frame from the file I uploaded to YouTube. As usual, I uploaded an "over spec" file to YouTube which forces a YouTube re-conversion. I think that YouTube probably has the best codecs, so the result is probably better than if I code it to spec myself. I could be wrong about that, but it is my current practice.

[2018-08-27 22:23]
correction:
"Match01-Src-19h40m40s966-C1.jpg"
This is the frame from the original source file that corresponds to the Pinnacle processed frames I posted before. Usually when I post a correction I removed the previous images, but in this case, I think I will just leave it. Having two sample frames out of a 7+ min video can be good. At the least, one can get an idea of the random changes in image noise.

VidThreeNorth · Jan 24, 2019

Sundown

I finished this clip ahead of others that I had wanted to complete before it. The reason was that it was the easiest to edit, and I need to clear off disc space. In fact, I am thinking about cancelling some projects and moving them off my computer. So while I am happy with the result and having gotten a project done, I am also annoyed. Life's like that. . . .

The biggest change in my plan was that I had hoped to use different music. I had a specific song in mind, but to use it would have meant making the effort to contact the related artists and waiting for approvals, and again, that would take a lot of time. So I substituted a pair of royalty free tunes from YouTube's library.

I was fortunate to find a pair of tunes that work well together, but are very different, which highlights the differences between the source videos.

Two Views of the Same Event

The point of this video is not obvious from watching it. These are two videos of the same event -- the sundown on Jan. 20, 2019. The first clip is a time-lapse facing away from the sun, recording the "golden hour" light on the distant buildings. Clouds and brief snow flurries are being pushed by strong winds, away from the camera. Briefly a small flight of geese flickers throughout the frames at about the 1 min, 20 sec. part. At 1 frame = 1 sec., they are only there for a couple of frames. Literally, if you blink you will miss them. The camera was my Git2 with 4.35mm F2.8 rectilinear lens and V 1.8 1229 firmware. This is the first daylight clip I have released using this firmware and these settings.
[2019-01-26: Some additional information.]

After starting the time-lapse, I went to a second location and recorded the sundown with my SJCAM SJ5000plus using version 3.3.6 firmware. This camera is also capable of time-lapse, but I could not remain at that location long enough to record it that way. It was too cold. I only had time to record perhaps 10 - 15 minutes before I left the area. So I recorded it at a normal 30 fps. Since it was recorded in a normal fashion, I have a sound track of traffic and wind which I mixed in with the second music recording.
[2019-01-26: Some additional information.]

I recorded the second clip with the camera mounted on a plastic hand-pod which I rarely use. I even held the pod down to keep it steady. But the wind was so strong at times that the camera vibrated, which showed in the recording.

Corel Pinnacle Studio 21 Ultimate:

It might seem like I am approaching this program in "baby steps", but in past years I have had so many problems that I do not feel like making a big jump resulting in a failed project. This project required inputting the Git2's anamorphic 16:9 2880 x 2160 video file and converting it to a true 16:9 output format (2560 x 1440). The Git2 file is otherwise a normal H.264 variable bitrate file, but as is typical for a time-lapse file, with no audio track. Likewise, the SJ5000plus has a slightly unusual file. It is "2304 x 1296 [16:9] @ 30 fps" constant bit-rate. So both the screen resolution and the "constant bit-rate" codec are unusual. Pinnacle Studio had no problem working with either file.

The second thing that was unusual was that I used the stabilizer on only the second file (SJ5000plus). I have used the stabilizer before, but until now, I have only had one clip per project and so everything was stabilized. In this case I can only report that the result looks good. I presume that the Git2 file was not touched by the stabilizer.

I had experimented with adding music to a video before, but this is the first actual project where I have done it in Pinnacle. I can report that I encountered no problem and the result is good.

I trimmed both clips within Pinnacle and had no editing problems (the last version of Pinnacle created "random" cuts in clips that should not have been cut when I tried to do this).

I considered the idea of making other adjustments to the second clip -- correcting the barrel distortion, altering the exposure, and adding saturation to match the first clip. In the end, I decided it did not need these changes because it was appropriate to have a completely different look from the first clip.

Finally, I decided to output to 2560 x 1440 which is less than the resolution of the Git2 but higher than the resolution of the SJ5000plus. Since the stabilizing would probably reduce detail a bit, I felt that I did not want to enlarge it too much. This was also another reason not to try correcting the barrel distortion. The 2560 x 1440 format is NOT a pre-set in Pinnacle. Oddly Pinnacle has a 2560 x 1280 pre-set format. I entered the 2560 x 1440 format and it worked without problems. There is no "live preview" of the frames that are made.

The resulting file that I uploaded to YouTube was about 2GB. It should look quite good in Full HD, but I would not recommend viewing it UHD. If you have a 2560 screen, I think it should look very good.

"20190120 Toronto Sundown01", posted 2019-01-24
""

[2019-01-29 More about the recording.]

Git2 clip:
Windows 8.1:
"2019_0120_155559_001.MP4"
size 1,229,272,010 bytes
Created January 20 2019 15:55:58 [started]
Modified January 20 2019 18:25:40 [ended]
- recorded time ~2.5 hours
Length 05:00
Video:
Data rate 32,763 kbps
Total bitrate 32,763 kbps

SJ5000plus Clip:
"SJCM0002.mp4"
size 922,943,488 bytes
Created January 20 2019 16:41:54
Modified January 20 2019 16:46:54
Length 05:00

From this clip, I removed roughly the first 30 seconds, and used about 1:30, and trimmed the last three minutes. I wanted to feature the birds flying starting roughly at the last minute of the final edit. But I wanted the birds gone for a while before the end of the clip. They are a feature of the clip, but they do not "define" the clip.

According to the Toronto Star, Sunset was 17:13. "Sundown" would be just before sunset. I should check with an astronomer, but I recall being told that it takes "seconds, not minutes". If you really want to feature an actual sunset, then think "slow motion" not "time-lapse".

I used the first three minutes from the Git2 time-lapse, which works out to about 1.5 hours. Starting from about 15:55, that covers to around 17:25. So I definitely covered "sundown", and "sunset". I also definitely covered the SJ5000plus clip.

It does not surprise me that I did not cover the actual "Sundown" with the SJ5000plus clip. I was about a half hour early. I had not checked the weather report before I went out to make the recording (a mistake in my preparation). Due to the local terrain and the fact that the sun was hidden by clouds, and also the trees, I had to guess where the sun was. I do not mind the "mistake". The cloud cover I got was probably better than the sky I would have captured if I had be "astronomically corrected".

[2019-03-02 9:09]
P1140005-2d-rsz1640-C1
I am cleaning off disc space and I will be removing this "scouting" image that I took while preparing for this video.

For this two camera project I did scout the site more than once. This image was one that I had in mind when I did the camera work. I knew that I would not get this particular cloud set. I thought I would probably try a couple of times and use the best pair. But I was happy with my first try. I think the weather on the day of this picture had been pretty good. The weather on the day I recorded the real takes was very different. The clouds on Jan. 20 were different, but I felt they were equally as good as the day of this picture.

I have partially corrected the perspective for this upload version, and reduced the size, but that is all.

VidThreeNorth · Mar 6, 2020

I have just uploaded what should be my last version (1.2) of "Autumn". The file version uncovered yet another bug, but I don't know what caused it and so far, I cannot correct it. I will not say what it is. If you don't notice it, then fine. I do not know if I will try to correct it again. I am still working with version 21 Ultimate, which was the 2017 version -- three years old. If I update it, perhaps the next version will fix the problem. At this point, Pinnacle is the only program I can afford that claims to support 3D. However, I never did complete a 3D video with this program, and the version that failed did so many years ago. I should try to make a new 3D test video to see if it is still worth upgrading. This is very frustrating. Anyway, aside from the one bug, this is pretty close to the video I wanted.

"[UHD] 20191020 Toronto Autumn v1.2"
""

AlanKlein · Mar 6, 2020

Lots of hard work. Good job. Video is a long learning curve.

VidThreeNorth · Mar 7, 2020

Much thanks! "Hard work" is not a problem for me. It's the lack of time. Well, ok, money too.

Speaking of time and money, I am thinking about getting a new computer, which means I need to clean up disc space more completely, which means I am looking over older projects. I will probably re-do a couple of old projects before I pack them away -- things I am confident that I can improve. That should be nice. Well, we'll see. . . .