This was fun indeed, thanks @betazoid! And the bilingualism wasnât even such a big issue imo.
The technical issues were definitely partly due to Jitsi, but also due to sometimes less than perfect internet connection on peopleâs end. Thatâs not something a server can fix and just an unfortunate consequence of people scattered across the planetâŠ
@XavAL Iâm so sorry, but I was very new to Twitch and apparently needed to check a setting to make sure the stream was saved, which I didnât Iâll make sure this works next time.
I have 2 videos indeed , one for the jitsi part, another for the zoom part.
But they are 6GB and 22GB respectively, so I have to learn how to edit them and above all, make them smaller!
Also my scheduler started recording when the jitsi session was already running, so I donât know how much is definitely lost. Sorry. I was at work when everything startedâŠ
Thanks to Anna @betazoid for organizing this meeting.
Thanks to you afre @afre for this evaluation and translation.
Thank you all for sharing this moment.
Certainly the problem of languages is (somewhat) surmountable, but I was in some cases in the incomprehension
It is very difficult in such a short time (and I had not prepared anything) to share the purposes and functionalities of a tool like Local Adjustments - I have highlighted some key points:
deltaE and scope
working in full image and move the RT-spot
Log encoding and Ciecam
Wavelets
Tone mapping (Mantiuk) and ânormalize luminanceâ
other tools would have deserved a review like âLA Denoiseâ, etc., but Andy @Andy_Astbury1 made some very good videos
I had great difficulties (Internet, my old computer, âjitsiâ), but when we switched to âzoomâ, no more problems
And it is a pity that some participants lost the line, I think to Wayne @Wayne_Sutton) or could not connect as Xavier @XavAL
First I thought log encoding should not be among the local adjustments, it should be global. But then I realized that it does not hurt that it can be a local adjustment. Probably, with this tool scope should be set to at least 80. For this photo, I think I set it to 95 or so and placed the central point on the mountains. That way I could increase the contrast and at the same time save more of the highlights in the sky (less contrast in the sky than in the mountains).
For increasing the saturation, I created a new local adjustment, placed the center also on the mountains. I set the scope again to a rather high value. I used color & light for increasing the saturation but since I placed the center on the gound, I did not create a color cast in the sky.
So this is a very minimal edit with just two local adjustments and no global adjustments.
Probably, it could be even more minimal, using exclusively log encoding.
Thanks to everybody for participating yesterday. It was not perfect as far as the video conferencing technology and the language/translation difficulties are concerned but I hope it was better than nothing.
Just a remark about âglobalâ / âlocalâ. There are only advantages to using âlocalâ (in this case full image - you can also use several spot with transitions), because you get
usage of deltaE (except if scope=100)
the possibility to use âexcluding spotsâ to exclude one or more parts of the image from the global processing
the notion of objects - you can associate for each type of spot (normal, excluding, full image) all the other tools of âLAâ (graduated filter, denoiseâŠ)
I used Shotcut to edit the videos which started being H.264-avc1
The settings where:
codec: H.264-nvenc
Quality based VBR
Quality: 55% (vglobal_quality=23)
GOP: 240
B frames: 3
Sound at 22050Hz and 128Kb/s
Donât know if itâs of any use to you, thoughâŠ
I have to add that the blockiness of the mouse picture background was there in the original, caused by the settings in the screen capture app (I think). So all in all, thereâs not much lost from the original video.
@XavAL I use Shotcut. Always interested on what others do in terms of settings. I made a stealth edit on my previous post that you may have missed: how did you record the streams?
I just finished watching the Zoom video. Many thanks for pulling all of this together and making it available for everyone. Really enjoyed the discussions and seeing @Andy_Astbury1 and @jdc working through examples of RT.