More odd Pixelan behaviour

terrypin wrote on 4/10/2014, 11:14 AM

I applied the Pixelan Chroma warp default to an AVI video clip.

This is a screenshot of the start of the clip in the MEP 2014 monitor full screen:

Or for better quality: https://dl.dropboxusercontent.com/u/4019461/OriginalAVI.jpg

I then took a snapshot of the AVI at that same start point using Export as BMP, expecting it to be identical as usual. But instead it looks significantly different, making it impossible to get a seamless transition to extend the start.

Or for better quality: https://dl.dropboxusercontent.com/u/4019461/ExportedSnapshot.jpg

I suspect this is related to my post earlier, 'Copy/pasting video effect: Bug in Pixelan, MEP or neither?'.

 

--
Terry, East Grinstead, UK

Last changed by terrypin on 4/10/2014, 11:15 AM, changed a total of 2 times.

Terry, East Grinstead, UK. PC: i7 6700K, 4.0 GHz, 32GB with Win 10 pro. Used many earlier versions of MEPP, currently mainly MEPP 2016 & 2017 (Using scores of macro scripts to add functionality, tailored to these versions.)

Comments

browj2 wrote on 4/10/2014, 1:08 PM

Terry,

I didn't think I could reproduce the problem as the images looked so close. I used a video (mts not avi) on which the ChromaWarp default effect is more obvious than yours. I gave it the default ChromaWarp and put the playback marker at the beginning of the clip. I did an export to jpg and another to bmp. I did Alt+prtSc and pasted the MEP screen in Xara and then exported from Xara to a bmp.

They exports all looked the same. I imported the bmp from the MEP export and place it in front of the avi, and then expanded the timeline to single frame. I then compared the first frame of the video and the bmp, and there is a difference. The exported ChromaWarp image is slightly darker and the size is not exactly the same.

So, I did the same thing without the effect. The exported image is also slightly different from the video image. The former is slightly clearer and slightly smaller and the image gets very slightly larger when it hits the first video frame, like a zoom-in, just like the image with the effect.

I then deleted the first frame of the video and compared the second with the bmp. It looked closer but there was still a slight difference. I did a few more frames and it was obvious that they were not used for the export.

Thus since the colour, clarity and size change between the exported image and the video even without the effect, I would say that this is a MEP problem. The use of an effect may multiply the difference, as it seems to have done in your case. In mine, the change is subtle, and because the camera moves on images of leaves and plants, the ChromaWarp effect is constantly in motion. In my case, the difference between the static object and the video is only noticeable because there is no movement of anything for the length of the static image, then all of a sudden there is movement.

Try out the same thing without the effect to see if you get the same result as I do.

Last changed by browj2 on 4/10/2014, 1:08 PM, changed a total of 1 times.

John C.B.

VideoPro X(17U); Video Deluxe 2026 Plus, Movie Studio 2025 Platinum; Music Maker 2026 Premium Edition; Samplitude Pro X8 Suite; see About me for more.

Desktop System - Windows 11 Pro 24H2; MB ROG STRIX B560-A Gaming WiFi; Graphics Card Zotac Gaming NVIDIA GeForce RTX-3060, PS; Power supply EVGA 750W; Intel Core i7-10700K @ 3.80GHz (UHD Graphics 630); RAM 32 GB; OS on Kingston SSD 1TB; secondary WD 2TB; others 1.5TB, 3TB, 500GB, 4TB, 5TB, 6TB, 8TB, 12TB, 14TB; three monitors - HP 25" main, LG 4K 27" second, HP 27" third; Casio WK-225 piano keyboard; M-Audio M-Track USB mixer.

Notebook - Microsoft Surface Pro 4, i5-6300U, 8 GB RAM, 256 SSD, W10 Pro 20H2.

YouTube Channel: @JCBrownVideos

terrypin wrote on 4/11/2014, 3:01 AM

Hi John,

 

Thanks for taking the time to check this.

I've done half a dozen more similar experiments using this test image:


https://dl.dropboxusercontent.com/u/4019461/PixelanTest.jpg

I made various arbitrary changes to the test image using Video effects > Art filter.

But with all of them I found as you did that the snapshot difference was much less obvious than that earlier one.

This therefore looks exactly as you suggest, an MEP problem that happens in this instance to be very pronounced.

The differences in my latest tests were of a similar degree to those I have often found in the past, using exclusively MEP effects. That's an issue I've never really fully understood. Sometimes it's enough to spoil what shold be a seamless transition.

--------------------

See also my additional post in the other thread, reporting that Undo fails to work on a Pixelan effect.

--
Terry, East Grinstead, UK

 

 

Last changed by terrypin on 4/11/2014, 3:01 AM, changed a total of 1 times.

Terry, East Grinstead, UK. PC: i7 6700K, 4.0 GHz, 32GB with Win 10 pro. Used many earlier versions of MEPP, currently mainly MEPP 2016 & 2017 (Using scores of macro scripts to add functionality, tailored to these versions.)

terrypin wrote on 4/11/2014, 9:44 AM

It was important for my project to add a still image of a couple of seconds duration before the start of that video clip. So this problem was a frustrating setback.

I worked around it by viewing the first frame of the video clip full screen, making a careful screenshot, and saving that as my BMP.

Does anyone know why MEP's BMP export doesn't achieve the same result please?

--
Terry, East Grinstead, UK

Last changed by terrypin on 4/11/2014, 9:44 AM, changed a total of 1 times.

Terry, East Grinstead, UK. PC: i7 6700K, 4.0 GHz, 32GB with Win 10 pro. Used many earlier versions of MEPP, currently mainly MEPP 2016 & 2017 (Using scores of macro scripts to add functionality, tailored to these versions.)

johnebaker wrote on 4/11/2014, 1:06 PM

Hi Terry

The 'wrong' image or differences suggest to me that the point in the clip where you placed the timeline cursor was not at a full frame but in between at an intermediate frame .

This would mean the image export has to build up a full frame from for example a previous full frame + intermediates up to the timeline cirsor position.

What I do in these circumstances is aim for a 2 - 3 sec crossfade wherever possible so the change is more subtle.

John

Last changed by johnebaker on 4/11/2014, 1:07 PM, changed a total of 2 times.

VPX 16, Movie Studio 2025, and earlier versions 2015 and 2016, Music Maker Premium 2024.

PC - running Windows 11 23H2 Professional on Intel i7-8700K 3.2 GHz, 16GB RAM, RTX 2060 6GB 192-bit GDDR6, 1 x 1Tb Sabrent NVME SSD (OS and programs), 2 x 4TB (Data) internal HDD + 1TB internal SSD (Work disc), + 6 ext backup HDDs.

Laptop - Lenovo Legion 5i Phantom - running Windows 11 24H2 on Intel Core i7-10750H, 16GB DDR4-SDRAM, 512GB SSD, 43.9 cm screen Full HD 1920 x 1080, Intel UHD 630 iGPU and NVIDIA GeForce RTX 2060 (6GB GDDR6)

Sony FDR-AX53e Video camera, DJI Osmo Action 3 and Sony HDR-AS30V Sports cams.

terrypin wrote on 4/11/2014, 9:30 PM

Hi John,

Thanks, I'd tried a cross fade but that still looked unsatisfactory in this case.

Snapshot was definitely at very start of the video clip.

 

Terry, East Grinstead, UK

Last changed by terrypin on 4/11/2014, 9:30 PM, changed a total of 1 times.

Terry, East Grinstead, UK. PC: i7 6700K, 4.0 GHz, 32GB with Win 10 pro. Used many earlier versions of MEPP, currently mainly MEPP 2016 & 2017 (Using scores of macro scripts to add functionality, tailored to these versions.)