His success was unforeseeable.

2014 saw Facebook's $2.3 billion purchase of Oculus VR.

19-year-old Palmer Luckey founded Oculus. He quit journalism school. His parents worried about his college dropout.

Facebook bought Oculus VR in less than 2 years.

Palmer Luckey started Anduril Industries. Palmer has raised $385 million with Anduril.

The Oculus journey began in a trailer

Palmer Luckey, 19, owned the trailer.

Luckey had his trailer customized. The trailer had all six of Luckey's screens. In the trailer's remaining area, Luckey conducted hardware tests.

At 16, he became obsessed with virtual reality. Virtual reality was rare at the time.

Luckey didn't know about VR when he started.

Previously, he liked "portabilizing" mods. Hacking ancient game consoles into handhelds.

In his city, fewer portabilizers actively traded.

Luckey started "ModRetro" for other portabilizers. Luckey was exposed to VR headsets online.

Luckey:

“Man, ModRetro days were the best.”

Palmer Luckey used VR headsets for three years. His design had 50 prototypes.

Luckey used to work at the Long Beach Sailing Center for minimum salary, servicing diesel engines and cleaning boats.

Luckey worked in a USC Institute for Creative Technologies mixed reality lab in July 2011. (ICT).

Luckey cleaned the lab, did reports, and helped other students with VR projects.

Luckey's lab job was dull.

Luckey chose to work in the lab because he wanted to engage with like-minded folks.

By 2012, Luckey had a prototype he hoped to share globally. He made cheaper headsets than others.

Luckey wanted to sell an easy-to-assemble virtual reality kit on Kickstarter.

He realized he needed a corporation to do these sales legally. He started looking for names. "Virtuality," "virtual," and "VR" are all taken.

Hence, Oculus.

If Luckey sold a hundred prototypes, he would be thrilled since it would boost his future possibilities.

John Carmack, legendary game designer

Carmack has liked sci-fi and fantasy since infancy.

Carmack loved imagining intricate gaming worlds.

His interest in programming and computer science grew with age.

He liked graphics. He liked how mismatching 0 and 1 might create new colors and visuals.

Carmack played computer games as a teen. He created Shadowforge in high school.

He founded Id software in 1991. When Carmack created id software, console games were the best-sellers.

Old computer games have weak graphics. John Carmack and id software developed "adaptive tile refresh."

This technique smoothed PC game scrolling. id software launched 3-D, Quake, and Doom using "adaptive tile refresh."

These games made John Carmack a gaming star. Later, he sold Id software to ZeniMax Media.

How Palmer Luckey met Carmack

In 2011, Carmack was thinking a lot about 3-D space and virtual reality.

He was underwhelmed by the greatest HMD on the market. Because of their flimsiness and latency.

His disappointment was partly due to the view (FOV). Best HMD had 40-degree field of view.

Poor. The best VR headset is useless with a 40-degree FOV.

Carmack intended to show the press Doom 3 in VR. He explored VR headsets and internet groups for this reason.

Carmack identified a VR enthusiast in the comments section of "LEEP on the Cheap." "PalmerTech" was the name.

Carmack approached PalmerTech about his prototype. He told Luckey about his VR demos, so he wanted to see his prototype.

Carmack got a Rift prototype. Here's his May 17 tweet.

John Carmack tweeted an evaluation of the Luckey prototype.

Dan Newell, a Valve engineer, and Mick Hocking, a Sony senior director, pre-ordered Oculus Rift prototypes with Carmack's help.

Everyone praised Luckey after Carmack demoed Rift.

Palmer Luckey received a job offer from Sony.

• It was a full-time position at Sony Computer Europe.

• He would run Sony’s R&D lab.

• The salary would be $70k.

Who is Brendan Iribe?

Brendan Iribe started early with Startups. In 2004, he and Mike Antonov founded Scaleform.

Scaleform created high-performance middleware. This package allows 3D Flash games.

In 2011, Iribe sold Scaleform to Autodesk for $36 million.

How Brendan Iribe discovered Palmer Luckey.

Brendan Iribe's friend Laurent Scallie.

Laurent told Iribe about a potential opportunity.

Laurent promised Iribe VR will work this time. Laurent introduced Iribe to Luckey.

Iribe was doubtful after hearing Laurent's statements. He doubted Laurent's VR claims.

But since Laurent took the name John Carmack, Iribe thought he should look at Luckey Innovation. Iribe was hooked on virtual reality after reading Palmer Luckey stories.

He asked Scallie about Palmer Luckey.

Iribe convinced Luckey to start Oculus with him

First meeting between Palmer Luckey and Iribe.

The Iribe team wanted Luckey to feel comfortable.

Iribe sought to convince Luckey that launching a company was easy. Iribe told Luckey anyone could start a business.

Luckey told Iribe's staff he was homeschooled from childhood. Luckey took self-study courses.

Luckey had planned to launch a Kickstarter campaign and sell kits for his prototype. Many companies offered him jobs, nevertheless.

He's considering Sony's offer.

Iribe advised Luckey to stay independent and not join a firm. Iribe asked Luckey how he could raise his child better. No one sees your baby like you do?

Iribe's team pushed Luckey to stay independent and establish a software ecosystem around his device.

After conversing with Iribe, Luckey rejected every job offer and merger option.

Iribe convinced Luckey to provide an SDK for Oculus developers.

After a few months. Brendan Iribe co-founded Oculus with Palmer Luckey. Luckey trusted Iribe and his crew, so he started a corporation with him.

Crowdfunding

Brendan Iribe and Palmer Luckey launched a Kickstarter.

Gabe Newell endorsed Palmer's Kickstarter video.

Gabe Newell wants folks to trust Palmer Luckey since he's doing something fascinating and answering tough questions.

Mark Bolas and David Helgason backed Palmer Luckey's VR Kickstarter video.

Luckey introduced Oculus Rift during the Kickstarter campaign. He introduced virtual reality during press conferences.

Oculus' Kickstarter effort was a success. Palmer Luckey felt he could raise $250,000.

Oculus raised $2.4 million through Kickstarter. Palmer Luckey's virtual reality vision was well-received.

Mark Zuckerberg's Oculus discovery

Brendan Iribe and Palmer Luckey hired the right personnel after a successful Kickstarter campaign.

Oculus needs a lot of money for engineers and hardware. They needed investors' money.

Series A raised $16M.

Next, Andreessen Horowitz partner Brain Cho approached Iribe.

Cho told Iribe that Andreessen Horowitz could invest in Oculus Series B if the company solved motion sickness.

Mark Andreessen was Iribe's dream client.

Marc Andreessen and his partners gave Oculus $75 million.

Andreessen introduced Iribe to Zukerberg. Iribe and Zukerberg discussed the future of games and virtual reality by phone.

Facebook's Oculus demo

Iribe showed Zuckerberg Oculus.

Mark was hooked after using Oculus. The headset impressed him.

The whole Facebook crew who saw the demo said only one thing.

“Holy Crap!”

This surprised them all.

Mark Zuckerberg was impressed by the team's response. Mark Zuckerberg met the Oculus team five days after the demo.

First meeting Palmer Luckey.

Palmer Luckey is one of Mark's biggest supporters and loves Facebook.

Oculus Acquisition

Zuckerberg wanted Oculus.

Brendan Iribe had requested for $4 billion, but Mark wasn't interested.

Facebook bought Oculus for $2.3 billion after months of drama.

After selling his company, how does Palmer view money?

Palmer loves the freedom money gives him. Money frees him from small worries.

Money has allowed him to pursue things he wouldn't have otherwise.

“If I didn’t have money I wouldn’t have a collection of vintage military vehicles…You can have nice hobbies that keep you relaxed when you have money.”

He didn't start Oculus to generate money. His virtual reality passion spanned years.

He didn't have to lie about how virtual reality will transform everything until he needed funding.

The company's success was an unexpected bonus. He was merely passionate about a good cause.

After Oculus' $2.3 billion exit, what changed?

Palmer didn't mind being rich. He did similar things.

After Facebook bought Oculus, he moved to Silicon Valley and lived in a 12-person shared house due to high rents.

Palmer might have afforded a big mansion, but he prefers stability and doing things because he wants to, not because he has to.

“Taco Bell is never tasted so good as when you know you could afford to never eat taco bell again.”

Palmer's leadership shifted.

Palmer changed his leadership after selling Oculus.

When he launched his second company, he couldn't work on his passions.

“When you start a tech company you do it because you want to work on a technology, that is why you are interested in that space in the first place. As the company has grown, he has realized that if he is still doing optical design in the company it’s because he is being negligent about the hiring process.”

Once his startup grows, the founder's responsibilities shift. He must recruit better firm managers.

Recruiting talented people becomes the top priority. The founder must convince others of their influence.

A book that helped me write this:

The History of the Future: Oculus, Facebook, and the Revolution That Swept Virtual Reality — Blake Harris

*This post is a summary. Read the full article here.

Writing for readers:

14 years of daily writing.

I post practically everything on social media. I authored hundreds of articles, thousands of tweets, and numerous volumes to almost no one.

Tens of thousands of readers regularly praise me.

I despised writing. I'm stuck now.

I've learned what readers like and what doesn't.

Here are some essential guidelines for writing with impact:

Readers won't understand your work if you can't.

Though obvious, this slipped me up. Share your truths.

Stories engage human brains.

Showing the journey of a person from worm to butterfly inspires the human spirit.

Overthinking hinders powerful writing.

The best ideas come from inner understanding in between thoughts.

Avoid writing to find it. Write.

Writing a masterpiece isn't motivating.

Write for five minutes to simplify. Step-by-step, entertaining, easy steps.

Good writing requires a willingness to make mistakes.

So write loads of garbage that you can edit into a good piece.

Courageous writing.

A courageous story will move readers. Personal experience is best.

Go where few dare.

Templates, outlines, and boundaries help.

Limitations enhance writing.

Excellent writing is straightforward and readable, removing all the unnecessary fat.

Use five words instead of nine.

Use ordinary words instead of uncommon ones.

Readers desire relatability.

Too much perfection will turn it off.

Write to solve an issue if you can't think of anything to write.

Instead, read to inspire. Best authors read.

Every tweet, thread, and novel must have a central idea.

What's its point?

This can make writing confusing.

️ Don't direct your reader.

Readers quit reading. Demonstrate, describe, and relate.

Even if no one responds, have fun. If you hate writing it, the reader will too.

> Say Hello To The Lean-Hedge Model

People think starting a business requires significant debt and investment. Like Shark Tank, you need a world-changing idea. I'm not saying to avoid investors or brilliant ideas.

Investing is essential to build a genuinely profitable company. Think Apple or Starbucks.

Entrepreneurship is risky because many people go bankrupt from debt. As starters, we shouldn't do it. Instead, use lean-hedge.

Simply defined, you construct a cash-flow business to hedge against long-term investment-heavy business expenses.

What the “fx!$rench-toast” is the lean-hedge model?

When you start a business, your money should move down, down, down, then up when it becomes profitable.

Many people don't survive the business's initial losses and debt. What if, we created a cash-flow business BEFORE we started our Starbucks to hedge against its initial expenses?

Lean-hedge has two sections. Start a cash-flow business. A cash-flow business takes minimal investment and usually involves sweat and time.

Let’s take a look at some examples:

A Translation company

Personal portfolio website (you make a site then you do cold e-mail marketing)

FREELANCE (UpWork, Fiverr).

Educational business.

Infomarketing. (You design a knowledge-based product. You sell the info).

Online fitness/diet/health coaching ($50-$300/month, calls, training plan)

Amazon e-book publishing. (Medium writers do this)

YouTube, cash-flow channel

A web development agency (I'm a dev, but if you're not, a graphic design agency, etc.) (Sell your time.)

Digital Marketing

Online paralegal (A million lawyers work in the U.S).

Some dropshipping (Organic Tik Tok dropshipping, where you create content to drive traffic to your shopify store instead of spend money on ads).

(Disclaimer: My first two cash-flow enterprises, which were language teaching, failed terribly. My translation firm is now booming because B2B e-mail marketing is easy.)

Crossover occurs. Your long-term business starts earning more money than your cash flow business.

My cash-flow business (freelancing, translation) makes $7k+/month.

I’ve decided to start a slightly more investment-heavy digital marketing agency

Here are the anticipated business's time- and money-intensive investments:

1. ($$$) Top Front-End designer's Figma/UI-UX design (in negotiation)

2. (Time): A little copywriting (I will do this myself)

3. ($$) Creating an animated webpage with HTML (in negotiation)

4. Backend Development (Duration) (I'll carry out this myself using Laravel.)

5. Logo Design ($$)

6. Logo Intro Video for $

7. Video Intro (I’ll edit this myself with Premiere Pro)

etc.

Then evaluate product, place, price, and promotion. Consider promotion and pricing.

The lean-hedge model's point is:

Don't gamble. Avoid debt. First create a cash-flow project, then grow it steadily.

Check read my previous posts on “Nightmare Mode” (which teaches you how to make work as interesting as video games) and Why most people can't escape a 9-5 to learn how to develop a cash-flow business.

Note-taking is a vital skill. But it's mostly learned.

Recently, innovative note-taking apps have flooded the market.

In the next few years, Evernote and Craft will be important digital note-taking companies.

Evernote is a 2008 note-taking program. It can capture ideas, track tasks, and organize information on numerous platforms.

It's one of the only note-taking app that lets users input text, audio, photos, and videos. It's great for collecting research notes, brainstorming, and remaining organized.

Craft is a popular note-taking app.

Craft is a more concentrated note-taking application than Evernote. It organizes notes into subjects, tags, and relationships, making it ideal for technical or research notes.

Craft's search engine makes it easy to find what you need.

Both Evernote and Craft are likely to be the major players in digital note-taking in the years to come.

Their concentration on gathering and organizing information lets users generate notes quickly and simply. Multimedia elements and a strong search engine make them the note-taking apps of the future.

Evernote and Craft are great note-taking tools for staying organized and tracking ideas and projects.

With their focus on acquiring and organizing information, they'll dominate digital note-taking in 2023.

Pros

• Concentrate on gathering and compiling information

• special features including a strong search engine and multimedia components

• Possibility of subject, tag, and relationship structuring

• enables users to incorporate multimedia elements

• Excellent tool for maintaining organization, arranging research notes, and brainstorming

Cons

• Software may be difficult for folks who are not tech-savvy to utilize.

• Limited assistance for hardware running an outdated operating system

• Subscriptions could be pricey.

• Data loss risk because of security issues

Evernote and Craft both have downsides.

1. The risk of data loss as a result of security flaws and software defects comes first.

2. Additionally, their subscription fees could be high, and they might restrict support for hardware that isn't running the newest operating systems.

3. Finally, folks who need to be tech-savvy may find the software difficult.

Evernote versus. Productivity Titans Evernote will make Notion more useful. medium.com

Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.

Move people with tiny words.

Apple fanboy here.

• Macs are awesome.

• Their iPhones rock.

• $19 cloths are great.

• $999 stands are amazing.

I love Apple's microcopy even more.

It's like the marketing goddess bit into the Apple logo and blessed the world with microcopy.

I took on a 30-day micro-stalking mission.

Every time I caught myself wasting time on YouTube, I had to visit Apple’s website to learn the secrets of the marketing goddess herself.

We've learned. Golden apples are calling.

Cut the friction

Benefit-first, not commitment-first.

Brands lose customers through friction.

Most brands don't think like customers.

• Brands want sales.

• Brands want newsletter signups.

Here's their microcopy:

• “Buy it now.”

• “Sign up for our newsletter.”

Both are difficult. They ask for big commitments.

People are simple creatures. Want pleasure without commitment.

Apple nails this.

So, instead of highlighting the commitment, they highlight the benefit of the commitment.

Saving on the latest iPhone sounds easier than buying it. Everyone saves, but not everyone buys.

A subtle change in framing reduces friction.

Apple eliminates customer objections to reduce friction.

Less customer friction means simpler processes.

Apple's copy expertly reassures customers about shipping fees and not being home. Apple assures customers that returning faulty products is easy.

Apple knows that talking to a real person is the best way to reduce friction and improve their copy.

Always rhyme

Learn about fine rhyme.

Poets make things beautiful with rhyme.

Copywriters use rhyme to stand out.

Apple’s copywriters have mastered the art of corporate rhyme.

Two techniques are used.

1. Perfect rhyme

Here, rhymes are identical.

2. Imperfect rhyme

Here, rhyming sounds vary.

Apple prioritizes meaning over rhyme.

Apple never forces rhymes that don't fit.

It fits so well that the copy seems accidental.

Add alliteration

Alliteration always entertains.

Alliteration repeats initial sounds in nearby words.

Apple's copy uses alliteration like no other brand I've seen to create a rhyming effect or make the text more fun to read.

For example, in the sentence "Sam saw seven swans swimming," the initial "s" sound is repeated five times. This creates a pleasing rhythm.

Microcopy overuse is like pouring ketchup on a Michelin-star meal.

Alliteration creates a memorable phrase in copywriting. It's subtler than rhyme, and most people wouldn't notice; it simply resonates.

I love how Apple uses alliteration and contrast between "wonders" and "ease".

Assonance, or repeating vowels, isn't Apple's thing.

You ≠ Hero, Customer = Hero

Your brand shouldn't be the hero.

Because they'll be using your product or service, your customer should be the hero of your copywriting. With your help, they should feel like they can achieve their goals.

I love how Apple emphasizes what you can do with the machine in this microcopy.

It's divine how they position their tools as sidekicks to help below.

This one takes the cake:

Dialogue-style writing

Conversational copy engages.

Excellent copy Like sharing gum with a friend.

This helps build audience trust.

Apple does this by using natural connecting words like "so" and phrases like "But that's not all."

Snowclone-proof

The mother of all microcopy techniques.

A snowclone uses an existing phrase or sentence to create a new one. The new phrase or sentence uses the same structure but different words.

It’s usually a well know saying like:

To be or not to be.

This becomes a formula:

To _ or not to _.

Copywriters fill in the blanks with cause-related words. Example:

To click or not to click.

Apple turns "survival of the fittest" into "arrival of the fittest."

It's unexpected and surprises the reader.

So this was fun.

But my fun has just begun.

Microcopy is 21st-century poetry.

I came as an Apple fanboy.

I leave as an Apple fanatic.

Now I’m off to find an apple tree.

Cause you know how it goes.

(Apples, trees, etc.)

This post is a summary. Original post available here.