I realize it's a bit grandiose (though probably no more than the whole idea of creating a irl robowaifu in the first place) but I want to begin thinking about how to create a working robowaifu 'brain', and how to create a special operating system to run on her so she will have the best chance of remaining an open, safe & secure platform. OS Language Choice C is by far the single largest source of security holes in software history, so it's out more or less automatically by default. I'm sure that causes many C developers to sneer at the very thought of a non-C-based operating system, but the unavoidable cost of fixing the large numbers of bugs and security holes that are inevitable for a large C project is simply more than can be borne by a small team. There is much else to do besides writing code here, and C hooks can be generated wherever deemed necessary as well. C++ is the best candidate for me personally, since it's the language I know best (I also know C too). It's also basically as low level as C but with far better abstractions and much better type-checking. And just like C, you can inline Assembler code wherever needed in C++. Although poorly-written C++ can be as bad as C code in terms of safety due to the necessity of it being compatible with C, it also has many facilities to not go there for the sane coder who adheres to simple, tried-and-true guidelines. There is also a good C++ project already ongoing that could be used for a clustered unikernel OS approach for speed and safety. This approach could save drastic amounts of time for many reasons, not the least of which is tightly constrained debugging. Every 'process' is literally it's own single-threaded kernel, and mountains of old-style cruft (and thinking) typical with OS development simply vanishes. FORTRAN is a very well-established language for the sciences, but a) there aren't a lot of FORTRAN coders, and b) it's probably not the greatest at being a general-purpose language anyway. I'm sure it could be made to run robotics hardware, but would probably be a challenge to turn into an OS. There are plenty of dujour SJW & coffee languages out there, but quite apart from the rampant faggotry & SJWtardism plainly evident in most of their communities, none of them have the kind of industrial experience and pure backbone that C, C++, or FORTRAN have. D and Ada are special cases and possibly bear due consideration in some year's time, but for now C++ is the obvious choice to me for a Robowaifu OS foundation, Python probably being the best scripting language for it. (1 of 2) >=== -fmt cleanup

(Original thread >>406) Kiwi back from the dead with a thread for the discussion of actuators that move your waifu! Part Two! Let's start with a quick refresher! 1. DC motors, these use a rotating magnetic field created through commutation to rotate a rotor! They're one of the cheapest options and are 30 to 70 percent efficient usually. The bigger they are, the more efficient they tend to be. 2. Brushless motors, these use a controller to induce a rotating magnetic field by turning electromagnets on and off in a sequence. They trend 60 to 95 percent efficiency 3. AC motors, Though there are many different type, they function similarly to brushless motors, they simply rely on the AC electricity to turn their electromagnets on and off to generate their field. Anywhere from 15 to 95 percent efficiency. 4. Stepper motors, brushless motors with ferrous teeth to focus magnetic flux. This allows for incredible control at the cost of greater mass and lower torque at higher speeds. Usually 50 to 80 percent efficient but, this depends on control algorithm/speed/and quality of the stepper. 5. Coiled Nylon Actuators! These things have an efficiency rating so low it's best to just say they aren't efficient. What they are though is dirt cheap and easy as heck to make! Don't even think about them, I did and it was awful. 6. Hydraulics! These rely on the distribution of pressure in a working liquid to move things like pistons. Though popular in large scale industry, their ability to be used in waifu's has yet to be proven. (Boston Dynamics Atlas runs on hydraulics but, it's a power guzzler and heavy) 7. Pneumatics, hydraulics lighter sister! This time the fluid is air! This has the advantage in weight. They aren't capable of the same power loads hydraulics are but, who wants their waifu to bench press a car? 8. Wax motors, hydraulic systems where the working fluid is expanding melted parafin wax! Cheap, low power, efficient, and produce incredible torque! Too bad they're slow and hard to control. 9. Explosion! Yes, you can move things through explosions! Gas engines work through explosions! Artificial muscles can be made by exploding a hydrogen and oxygen mixture in a piston, then using hydrolysis to turn the water back into hydrogen and oxygen. None of this is efficient or practical but, it's vital we keep our minds open. Though there are more actuators, most are derivatives or use these examples to work. Things like pulleys need an actuator to move them. Now, let's share, learn, and get our waifu moving! >--- < add'l, related links from Anon:

Many of the parts needed to build our robowaifus will need to be custom made and they will need to be metal. For parts that have a high tolerance for imperfections a 3d printer can print a mold and then a small scale foundry can be used to cast the piece with metal (probably copper or aluminum). BUT there will be pieces that need a higher degree of precision (such as joints). For these pieces a CNC machine would be useful. CNC machines can widely range in size, price, and accuracy and I would like to find models suitable for our purposes. I know there are CNC machines available that can cut up to copper for under $300, but I don't know if that will be enough for our purposes. (https://www.sainsmart.com/products/sainsmart-genmitsu-cnc-router-pro-diy-kit?variant=15941458296905&currency=USD&utm_campaign=gs-2018-08-06&utm_source=google&utm_medium=smart_campaign) Some CNC machines can be used to engrave printed circuit boards and that may prove useful for our purposes as well. Are there any anons that know more about CNC machines? Anons looking to buy one ask your questions here.

So I'm thinking about design and realize some mistakes I made. There is likely no better short cut to advancing design than these fundamental principles as stated by Elon Musk. 1. Make The Requirements Less Dumb “Step one: Make the requirements less dumb. The requirements are definitely dumb; it does not matter who gave them to you. It’s particularly dangerous when they come from an intelligent person, as you may not question them enough. Everyone’s wrong. No matter who you are, everyone is wrong some of the time. All designs are wrong, it’s just a matter of how wrong,” explains Musk. 2. Try And Delete Part Of The Process “Step two: try very hard to delete the part or process. If parts are not being added back into the design at least 10% of the time, [it means that] not enough parts are being deleted. The bias tends to be very strongly toward ‘let’s add this part or process step in case we need it’. Additionally, each required part and process must come from a name, not a department, as a department cannot be asked why a requirement exists, but a person can,” says Musk. 3. Simplify Or Optimize “Step three: simplify and optimize the design. This is the most common error of a smart engineer — to optimize something that should simply not exist,” according to Musk. He, himself, has been a victim of implementing these steps out of order. He refers to a “mental straightjacket” that happens in traditional schools where you always have to answer the question regardless of whether the premise makes any sense at all. 4. Accelerate Cycle Time “Step four: accelerate cycle time. You’re moving too slowly, go faster! But don’t go faster until you’ve worked on the other three things first,” explains Musk. Here he uses another example of how these steps should occur in order. During a wrongheaded process you should simply stop, not accelerate. He says, “If you’re digging your grave, don’t dig it faster.”

The biggest hurdle to making quick progress in AI is the lack of compute to train our own original models, yet there are millions of gamers with GPUs sitting around barely getting used, potentially an order of magnitude more compute than Google and Amazon combined. I've figured out a way though we can connect hundreds of computers together to train AI models by using gradient accumulation. How it works is by doing several training steps and accumulating the loss of each step, then dividing by the amount of accumulation steps taken before the optimizer step. If you have a batch size of 4 and do 256 training steps before an optimizer step, it's like training with a batch size of 1024. The larger the batch size and gradient accumulation steps are, the faster the model converges and the higher final accuracy it achieves. It's the most effective way to use a limited computing budget: https://www.youtube.com/watch?v=YX8LLYdQ-cA These training steps don't need to be calculated by a single computer but can be distributed across a network. A decent amount of bandwidth will be required to send the gradients each optimizer step and the training data. Deep gradient compression achieves a gradient compression ratio from 270x to 600x without losing accuracy, but it's still going to be using about 0.5 MB download and upload to train something like GPT2-medium each optimizer step, or about 4-6 mbps on a Tesla T4. However, we can reduce this bandwidth by doing several training steps before contributing gradients to the server. Taking 25 would reduce it to about 0.2 mbps. Both slow and fast computers can contribute so long as they have the memory to hold the model. A slower computer might only send one training step whereas a fast one might contribute ten to the accumulated gradient. Some research needs to be done if a variable accumulation step size impacts training, but it could be adjusted as people join and leave the network. All that's needed to do this is a VPS. Contributors wanting anonymity can use proxies or TOR, but project owners will need to use VPNs with sufficient bandwidth and dedicated IPs if they wish that much anonymity. The VPS doesn't need an expensive GPU rental either. The fastest computer in the group could be chosen to calculate the optimizer steps. The server would just need to collect the gradients, decompress them, add them together, compress again and send the accumulated gradient to the computer calculating the optimizer step. Or if the optimizing computer has sufficient bandwidth, it could download all the compressed gradients from the server and calculate the accumulated gradient itself. My internet has 200 mbps download so it could potentially handle up to 1000 computers by keeping the bandwidth to 0.2 mbps. Attacks on the network could be mitigated by analyzing the gradients, discarding nonsensical ones and banning clients that send junk, or possibly by using PGP keys to create a pseudo-anonymous web of trust. Libraries for distributed training implementing DGC already exist, although not as advanced as I'm envisioning yet: https://github.com/synxlin/deep-gradient-compression I think this will also be a good way to get more people involved. Most people don't know enough about AI or robotics enough to help but if they can contribute their GPU to someone's robowaifu AI they like and watch her improve each day they will feel good about it and get more involved. At scale though some care will need to be taken that people don't agree to run dangerous code on their computers, either through a library that constructs the models from instructions or something else. And where the gradients are calculated does not matter. They could come from all kinds of hardware, platforms and software like PyTorch, Tensorflow or mlpack.

This thread is for the sharing of robo waifu banners. As per the rules, fallow these requirements: >banner requirements: File size must be lower than 500 KB and dimensions are 300x100 exactly. Allowed file formats are .jpg, .png, and .gif. >=== -fmt cleanup

Keep your eyes on the prize edition --- Notice: I'm going to have to put a hiatus on this project until at least this Summertime 2024 (probably later). A) There seems to be only a little interest in it right now by anons. B) Right now in my life I have too much else on my plate that needs attention. C) In the meantime, whenever I'm here I'll still be happy to answer any questions anons have for the first 11 classes so far, or for anything in the textbook thread also. I hope to be back with this later, Anons. Cheers. --- • v231125 week-thirteen, lecture: semester-end catchup (N/A) - this week's reading assignments: N/A

How do we eat this elephant, /robowaifu/? This is a yuge task obviously, but OTOH, we all know it's inevitable there will be robowaifus. It's simply a matter of time. For us (and for every other Anon) the only question is will we create them ourselves, or will we have to take what we're handed out by the GlobohomoBotnet(TM)(R)(C)? In the interest of us achieving the former I'll present this checklist from George Pólya. Hopefully it can help us begin to break down the problem into bite-sized chunks and make forward progress. >--- First. UNDERSTANDING THE PROBLEM You have to understand the problem. >What is the unknown? What are the data? What is the condition? Is it possible to satisfy the condition? Is the condition sufficient to determine the unknown? Or is it insufficient? Or redundant? Or contradictory? >Draw a figure. Introduce suitable notation. >Separate the various parts of the condition. Can you write them down? Second.

This is a rather important project for the people involved here. I just had this amazing idea, which allows us to catalogue and make any of the information here searchable in a unique way. It functions very similarly to an image booru, but for markdown formatted text. It embeds links and the like. We can really make this thing our own, and put the entire board into a format that is indestructible. Anyone want to help build this into something great? I'll be working on this all day if you want to view the progress on GitHub. https://github.com/SoaringMoon/WaifuNetwork

Good VR is one important interim goal for a fully-realized RoboWaifu, and it's also much easier to visualize an algorithm graphically at a glance than 'trying to second-guess a mathematical function'. OpenGL is by far the most portable approach to graphics atm.

>"This is an excellent site for starting with OpenGL from scratch. It covers from very introductory topics all the way to advanced topics like Deferred Shading. Liberal usage of accompanying images and code. Strongly recommended."

learnopengl.com/
https://archive.is/BAJ0e

Plus the two books related, but learnopengl.com is particularly well-suited for beginners.

www.openglsuperbible.com/
https://archive.is/NvdZQ

opengl-redbook.com/
https://archive.is/xPxml

www.opengl.org
https://archive.fo/EZA0p