(Continued from The Forest Chronicles)
....Some of them are rumoured to have portals and wormholes within them through which The Watchers can pass undetected. It is curious to note that these portals often appear in pairs.
Mere mortals who have ventured into the heart of the forests and stumbled across these portals have been able to give accounts of their inviting appearance. In my travels too, I have been lucky enough to encounter a few of them and not courageous enough to actually have gone through any of them.
The mortals who were courageous enough to venture into them have not been seen since. There is no trace of them in any of the Universes. This begs the question : is there some Extraversal Space to which they have been transported? What exactly do The Watchers do there?
And why are there pairs of these portals sometimes?
And the most important question of all : Should I go through one of them?
----The Historian
Hello everyone, hope you are doing fine. I haven’t been able to post in a while but am back with some more AI experiments and some writing of my own. I thank you all once again for the overwhelming support and encouragement.
The first 3 images in 16:9 format have been generated with the MSE weighted VQGAN+CLIP notebook by @jotakrevs. The links to the notebooks can be found on any of his recent posts.
For the first time, I tried CLIP guided diffusion. The last 3 images have been generated with it. I can say that prompting it successfully took quite a while. It takes words too literally I guess. For example, if I say dark anywhere in the prompt, the resulting image is so dark that it is difficult to see anything. On the other hand, prompting also feels like you are trying to converse with it. With VQGAN+CLIP, you could get away with grammar not being correct, sometimes even spellings being wrong and so on. Here, I feel I am dealing with an English teacher sometimes. It definitely needs more precise prompts than VQGAN+CLIP.
Another important difference is that with the free version of Google Colab, we can generate only 256 x 256 pixel size images with CLIP guided diffusion. Other AI enhancers will have to be used to enlarge the images for sure. Also, it takes well over an hour to generate a decent image. Ease of use is pretty similar to using the VQGAN+CLIP notebook. Anyone who has used it before should not find any difficulty in using the CLIP guided diffusion notebooks.
Here are the links for the CLIP guided diffusion notebooks, authored by https://twitter.com/RiversHaveWings (The original VQGAN+CLIP was authored by her too) :
https://twitter.com/rivershavewings/status/1419674445044539400?lang=en
- rivershavewings
(thank you @kaliyuga for this link).
This is for generating a 512 x 512 pixel image :
https://twitter.com/RiversHaveWings/status/1437882743426220035?s=20
- RiversHaveWings
Google Colab Pro will be required for this with 16 GB video memory.
Some random science facts related to my writing, many of you might already be aware of these....
What we know about the Universe - all the matter, energy - this includes all the stars and systems and galaxies and clusters, all the electromagnetic waves- all this adds up to only 5% of what is out there.
There is something called Dark Matter - when the gravitational effects of a galaxy is measured, the mass of all the stars in it including all the black holes and all the matter that we know of, is not enough to account for the gravity produced. The remaining matter that we don’t know about is called Dark Matter. It is supposed to account for 85% of mass in the Universe! And we don’t know what it is :)
Wait, there is more! There is something called Dark Energy. The Universe is expanding at a rate not explainable with the current model. The total amount of energy should have caused a slowing in the rate of expansion but the rate of expansion is actually accelerating. This unknown energy is called dark Energy and constitutes 68% of the total Energy-Mass in the Universe.
So Dark Matter constitutes 27% of Energy-Mass and Dark Energy constitutes 68% of Energy-Mass of the Universe. Hence, we don’t even know 95% of the constituent of the Universe!
These figures are accurate, my inferences and writing maybe too enthusiastic and dramatic. Physicists kindly correct me if my inferences are wrong.
There are a few dark forests which you wish you hadn’t seen!
All the images in this post are AI generated using the script mentioned above. The images have been upscaled on https://bigjpg.com/ or using REAL ESRGAN notebook. There are no other edits to the images.
All the writing is mine, inspired by the images generated, my previous posts and my socio-cultural influences.
Thank you everyone and see you in the next one!