Ai Dreams Forum
Software & Hardware => General Hardware Talk => Topic started by: flamehowk on December 24, 2019, 09:39:34 pm
-
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fi.piccy.info%2Fi9%2Fc7cc2d8587aa57d2fbbb9d4199f93fc1%2F1577152891%2F78072%2F1351639%2Ftitulka_800.jpg&hash=38eb82dcdd9fe65153ca49a97128f4fe9a2af356)
Hello.
My name is Vitaliy Monastyrskiy.
I am an inventor, hardware and software engeneer from Ukraine.
I know how to make a programmable neural chip. This will be a field-programmable neural array, onto which it will be possible to program a neural network of any type, with a synchronous or asynchronous way of working. The technology allows you to scale the size of neural networks by increasing the number of chips on the motherboard.
The use of such chips is very large. They can be installed on drones or cars, making portable autopilots. Cards with such chips can be installed in a personal computer so that computer games count on them the behavior of computer opponents (NPCs). Using these chips, it will be possible to create thinking virtual assistants for a smart home, or create a virtual teacher for self-education. Imagine a computer game that you play for only half an hour a day, but for half a year you automatically learn a new language or a new profession ... In the end, such chips will be the first brains in the world for human-like robots.
Now, my project is in the process of developing a prototype. To complete this task I need $ 5000 and 6-8 months of work. And I am looking for funding for my project. Unfortunately crowdfunding is not suitable, because in our country banks do not accept money from crowdfunding platforms. Therefore, I can only collect investments through cryptocurrencies or ... I need a financial partner who will go into my project to finance it, for a share in the business.
I will be glad if the participants in this forum help me with this. Even better if you collectively fund this project. Then we will make a new future together. On its basis, much can be done in the field of AI development.
So I am open to suggestions and hope for your understanding.
Thank you in advance.
-
Why do you need capital? show us a schematic of your design, unless u want to keep it a secret.
I do hardware theory too, ive got a really out there technique which doesn't even use transistors. (So if I manufactured it, I wouldn't need to print semiconductors... ezier.)
If your using transistors, you have to go build your own, unless you have some design which doesn't require miniaturization because it hasn't many components, but that is quite a special case if its any good, you cant actually buy transistors for a cpu, they have to be printed on. Ive gone till now needing no money, and I think its going to stay that way.
Computers, fpgas and hardware neural networks are similar to make, they all end up being a network anyway, all logic is, its not so much of a big deal to me now.
So - if you don't mind - can you explain your hardware technique? And im not like guys anywhere else on the internet, I like zany stuff the best. If I see a 555 timer connected up to and gate ic's it makes me uncontrollably yawn. *bored*
Also.. so you've got a hardware brain planned - what about motor drivers? Ive got one, ill share it with you if you want, and its transistorless too. :)
-
>> Why do you need capital?
To make a prototype. In our country, now there is a war and poverty around. For me now it’s a lot of money.
>> show us a schematic of your design, unless u want to keep it a secret.
Want. This technology costs billions. I am not ready to give it.
>> Ive gone till now needing no money, and I think its going to stay that way.
I'm happy for you.
>> So - if you don't mind - can you explain your hardware technique?
I prefer "zero-disclosure evidence." That is, I want to limit myself to demonstrating the operability of my architecture, rather than revealing its secret.
>> Also .. so you've got a hardware brain planned - what about motor drivers? Ive got one, ill share it with you if you want, and its transistorless too.
Sorry, I don’t understand what kind of "motor drivers" you are talking about. If you mean drivers for stepper motors or servos, then I have my own full box. But now it's not about CNC machines.
-
Vitaliy, what educational background do you have? Is there any of your other work you would like to share with us?
-
Electronics and machines in general have a big shroud of lies hiding their workings, its hard to find help, lectures at university can be tough and hard to understand not because the machine is that complicated, its hard to know who to trust. My best advice if I treated you like myself, is to just keep making random schematics, they wont work at first, but theres always a little truth even to something that is mostly wrong, and youll eventually get there, it seems amazing and mystifying when you first get started, but once you start to work it out and the illusion is removed it can be a little dull to know about, (itll now be easy for you.) but then youll be in a position to get your business running.
-
Welcome Flamehowk.
It would be cool instead of asking directly for money if such a site could exist where people could put in requests for certain equipment or types of materials that could be freely given/donated, traded or sold for a very reasonable cost.
Asking for money without any proof of concept, prototype, model, etc., is like asking a complete stranger for money to support your dreams. THAT is a very hard sell! While the Internet has been a great vehicle in so many ways, it has also been a venomous animal biting some trusting people and taking their money with little or nothing in return.
While I'm not trying to say or imply that your project is one of them, people need more than a few sentences on a screen in order to agree to part with their money.
I do wish you luck and success with your project. Please keep us informed!
-
Vitaliy, what educational background do you have? Is there any of your other work you would like to share with us?
>> Vitaliy, what educational background do you have?
I study every day all my life ... How to measure my education?
>> Is there any of your other work you would like to share with us?
If you mean developments in the field of microelectronics, then I have only three of them:
1) Highly Intelligent Logic (HIL). This is a self-routing CPU architecture that solves the problem of parallel computing.
2) There is also a kernel architecture for the fastest ASIC chip of the Bitcoin miner, I call it SHBA (Serial Hashing Bitcoin Architecture). This is the only one of my OpenSource semiconductor projects, so if you want I can give a link to the presentation video:
https://www.youtube.com/watch?v=HayrYFW0b28&t=6s
3) This project is ProgBrain.
There are also many other developments, but in other areas. Engineering, cryptographic functions, a specialized tool, design of manipulators and much more. But all this has nothing to do with either neural networks or microelectronics.
-
Is that laboratory in the video your company you work in?
-
That video looks like toss to me.
So your technology involves parallel power.
A neural network is a parallel machine, your communicating every pixel on the retina to every cell, and all the cells operate at the same time.
We had this crazy idea on ai dreams not long ago to get rid of all the wires/synapses and replace them with a filter unit in each cell which decide which elements they are communing with, and we actually communicate ultrasonicly through piezoelectric speakers, instead of hard in-place wiring.
That way we get rid of all the wires, which is the main problem with building a hardware neural network.
-
Asking for money without any proof of concept, prototype, model, etc.,
Gentlemen, I ask YOU ALL to make one remark. I am NOT asking for money. No one. I offer cooperation. Investments in exchange for a share of profit or a share in a company.
Therefore, do not be fooled into thinking that I am a beggar. I have had a working experience since 1994, many inventions and developments. I was also a private entrepreneur and had my own small business until the war in 2014 began in my country. I was imprisoned as a political prisoner and I lost everything - money, business, housing.
Now I am free, but I have nothing, and around me there is one poverty. Before, for me to borrow such an amount from my friends was not a problem. Now I can hardly borrow $ 100 when I need it.
Therefore, I am not looking for free bread. I know the price of human labor. If someone is ready to join me and become part of the richest company in the world after in 30 years, we are welcome. But for this you need to do your bit. Since you don’t have my invention, I think that paying a few thousand dollars for the opportunity to get 20% of the future unicorn company is practically nothing. Yes is a risk. But didn’t I risk it when I invested my time and energy in this development instead of just making money?
OK. In the worst case, I can go to work in another country, lose another year of time, but then provide my own development. But in this case, I will not have partners.
Therefore, we can say that at the moment I am selling a stake in my company for start-up capital, which is needed to develop a prototype. I just want to save 1 year of my life, which is not much left. I also want to save my strength, which otherwise I would have to spend on low-intellectual physical labor. That's all.
Technically, I believe that I have the right to make such an offer. And I see no reason to think that I am doing wrong.
About the evidence.
I am ready to receive investments in the amount of $ 500 per month until the end of the development of the prototype, and every month to provide a report in the form of a demonstration of the work of my architecture. Thus, the "risk" lies only on the first payment of $ 500.
I agree that many of you will say - "why don’t you work 1 month at your own expense and not make a presentation of at least the first part of the work?". But I will answer. The only work that I can get right now is stone-cutting in quarries. Labor shift 12 hours. 1 hour walk to work and 1 hour walk from work home. Total 14 hours a day for heavy physical labor. In addition, every 3 days, shifts change from night to day and vice versa.
What do you think - can I do the development? I will tell you ... after this work you come home and sometimes you don’t even have the strength to eat. You just fall onto the bed and die on 8 hours before the alarm rings to come back to life and go to work again.
Therefore, there is no other way. Either I will find investments, or I need to emigrate to another country, where it have high salaries, and after working there for a year or two, earn my starting capital to ensure this development.
Regarding the reality of my model.
My model is fully functional. I can easily appreciate this because I have experience in developing microelectronic devices. But the scheme of her work is my intellectual property and I do not want to disclose it. Of course, if I had a good partner, a brother in spirit, an innovator like me, and I knew that he and I would be developing this architecture together, as well as that he would protect it with me ... then I would reveal to him all the subtleties and nuances. But I do not have such a partner. And disclose the details of this invention = give it to others. I have already given a lot of things to others, and now I don’t even have the money to go to the doctor and patch the holes in my teeth. I say it as it is. Excuse me.
Therefore, I can not offer anything else. The only way.
No jokes. I have left food for 3-4 days. If during this time I do not find investment, I will have to go to the quarries. But if I go there, then I will not work to prove to someone that my architecture is working. Not. I’ll go there to earn money for a passport and visa to Europe, to go there and earn money there. I have no other way.
I am not asking for money. Not. I offer cooperation.
But without your investment, I will only lose one or two years of my life and some of my health. Without my invention, you will lose the opportunity to become part of the future.
However, this applies to those who can invest a similar amount in my project. It is clear that this does not apply to the same poor as I am.
I hope for your understanding.
-
Is that laboratory in the video your company you work in?
This is my project. I am its founder.
But to continue this project, large investments are needed. $ 200,000 for ordering test chips at TSMC. I don’t have that kind of money. We did an ICO, but it did not collect anything. Therefore, this project is frozen until I become rich.
-
We had this crazy idea on ai dreams not long ago to get rid of all the wires/synapses and replace them with a filter unit in each cell which decide which elements they are communing with, and we actually communicate ultrasonicly through piezoelectric speakers, instead of hard in-place wiring.
That way we get rid of all the wires, which is the main problem with building a hardware neural network.
Your piezoelectric elements will not be able to overtake my wires on a microchip.
I decided to go the other way. I partially solved this problem by switching from parallel to serial data transmission. Thus, from one neuron to another (technically) there is only 1 wire for data transfer. Actually more, because we need to provide feedback, programming and control, but formally it is no longer 32 or 64 bits. Such a solution gives us much more field for the placement of neurons.
I also solved other issues related to what you are talking about.
In fact, I have two different solutions. One is slow but more compact. The second is fast, but requires more neurons. In the slow solution, I used the same idea - there will be a filter in each neuron, which from the general data stream will filter out the data that comes from those neurons that are specified in the firmware as “input linksâ€. But this architecture is only suitable for single-chip solutions. A large neural network cannot be built like this. Therefore, there is another model that is based on the capabilities of scaling a neural matrix. But I do not want to tell how I solved this problem. Sorry.
-
What's wrong with PayPal?
I did a bit of research for you, here is how to connect Ukrainian credit card (http://pbank24.com/?p=317&lang=en) to international PayPal service (it took exactly 1 google search and 1 link click).
Is that laboratory in the video your company you work in?
This is my project. I am its founder.
So you put someone else's laboratory in a video of your own project. Well, excuse me for being skeptic.
-
What's wrong with PayPal?
You need to ask PayPal, not me.
In our country, PayPal only allows you to deposit money, but it does not allow them to withdraw. That is - I can put money on PayPal and spend it somewhere on Amazon. But I can’t get paid with Indiegogo ...
So you put someone else's laboratory in a video of your own project. Well, excuse me for being skeptic.
I do not understand what you are talking about. This is my project. This is my architecture. The "lab" you are talking about is my home computer. I made this video.
-
Ok, let's say you are what you are presenting yourself. Then there are several options:
- So you're a computational master. Then programming shouldn't be a problem for you. Learn html, javascript and css, (shouldn't take more than a few weeks), or instead Wordpress (shouldn't take more that a few hours), and then find a honorar job at freelancer sites. Let the customers pay on your friend's foreign paypal account, and he will send you the money after done job.
- Setup Indiegogo with your Paypal account and start the campaign. After successful campaign, send your foreign friend the money over Paypal, and he will withdraw the money and send it back to you.
But you'll need someone you can trust.
-
Ok, let's say you are what you are presenting yourself. there are several options:
Sir, I thank you for trying to teach me how to make money, but you don’t have to think that if I could invent this, then I’m not able to check all the possible options for making money.
I know 10 programming languages. I wrote my first programs back in 1992 at Basic at the Lyceum of Information Technology.
I have already tried to make money on freelance - they do not appreciate the work of specialists.
If I had a “friend†abroad who could accept money, I would have already opened a crowdfunding company and collected as much as possible. But I do not have such friends abroad.
So. Giving smart tips is not difficult. I can also give you smart advice, but I do not. Therefore, if you have a job for a specialist like me - I am here, you can hire me. But I'm afraid you won’t do it. And others do not either. Therefore, all talk of "freelance" and other things is nothing more than a fantasy. I still can’t withdraw the $ 27 that I earned on UpWork in 3 weeks.
-
Some people suck at inventing things. Other people suck at making money. I may suck at both, but I own a PayPal account and a credit card for withdrawing money.
-
Some people suck at inventing things. Other people suck at making money. I may suck at both, but I own a PayPal account and a credit card for withdrawing money.
Congratulations.
-
Some people suck at inventing things. Other people suck at making money. I may suck at both, but I own a PayPal account and a credit card for withdrawing money.
Congratulations.
That was an offer.
But obviously I don't deserve your trust because I try to be honest. Honesty may be a bi**h.
[edit]
The offer stays, you know where to find me. I'll try to do my best.
-
you are not special. you are not a beautiful or unique snowflake. you're the same decaying organic matter as everything else.
-
you are not special. you are not a beautiful or unique snowflake. you're the same decaying organic matter as everything else.
Nonsense. Everyone is unique and special. We may be composed of the same building blocks, but everyone has some super-power they could be proud of. Every living being is worthy of admiration. Each person is an edge of the time that begun with the Big Bang. We are parts of the very Universe experiencing itself. I admire it.
-
This is the golden age of grotesque...
-
We had this crazy idea on ai dreams not long ago to get rid of all the wires/synapses and replace them with a filter unit in each cell which decide which elements they are communing with, and we actually communicate ultrasonicly through piezoelectric speakers, instead of hard in-place wiring.
That way we get rid of all the wires, which is the main problem with building a hardware neural network.
Your piezoelectric elements will not be able to overtake my wires on a microchip.
I decided to go the other way. I partially solved this problem by switching from parallel to serial data transmission. Thus, from one neuron to another (technically) there is only 1 wire for data transfer. Actually more, because we need to provide feedback, programming and control, but formally it is no longer 32 or 64 bits. Such a solution gives us much more field for the placement of neurons.
I also solved other issues related to what you are talking about.
In fact, I have two different solutions. One is slow but more compact. The second is fast, but requires more neurons. In the slow solution, I used the same idea - there will be a filter in each neuron, which from the general data stream will filter out the data that comes from those neurons that are specified in the firmware as “input linksâ€. But this architecture is only suitable for single-chip solutions. A large neural network cannot be built like this. Therefore, there is another model that is based on the capabilities of scaling a neural matrix. But I do not want to tell how I solved this problem. Sorry.
That's the spirit! Sounds like you got some ideas there, keep thinking about it and youll surely get there.
If you use an analogue signal, you only need 1 wire, not the 64 wires 64 bit takes. That's actually correct!
But in neural networks there is MANY wires coming out of a neuron, not just one. they make whats called a dendrite in biology, but artificial networks you can call then fan in's and fan outs. like on integrated circuit I/o.
-
That was an offer.
What is the suggestion? To organize a crowdfunding project on Indiegogo for this invention, with the withdrawal of funds to your PayPal card?
That is, when I look for an offer of $ 500, then you do not trust me. But when do you propose to pour a hundred thousand dollars on your card, should I trust you?
Well, let's say ...
But you, too, must understand that preparing for such a project will itself take time. How will I live all this time so that I can work on the project?
Like it or not, money is needed.
-
That's the spirit! Sounds like you got some ideas there, keep thinking about it and youll surely get there.
If you use an analogue signal, you only need 1 wire, not the 64 wires 64 bit takes. That's actually correct!
But in neural networks there is MANY wires coming out of a neuron, not just one. they make whats called a dendrite in biology, but artificial networks you call then fan in's and fan outs.
No, you were mistaken, NOT an analog signal. Analog computing is still very problematic. It's about digital computing, but sequential ... You can see about it in my video, which is higher in the posts. It talks about a different architecture, but the principles are the same.
In my architecture, it is possible to associate many any neurons with many other neurons. Many inputs - many outputs for each neuron.
-
That's the spirit! Sounds like you got some ideas there, keep thinking about it and youll surely get there.
If you use an analogue signal, you only need 1 wire, not the 64 wires 64 bit takes. That's actually correct!
But in neural networks there is MANY wires coming out of a neuron, not just one. they make whats called a dendrite in biology, but artificial networks you call then fan in's and fan outs.
No, you were mistaken, NOT an analog signal. Analog computing is still very problematic. It's about digital computing, but sequential ... You can see about it in my video, which is higher in the posts. It talks about a different architecture, but the principles are the same.
In my architecture, it is possible to associate many any neurons with many other neurons. Many inputs - many outputs for each neuron.
Entertain me, what problems do analogue computers have?
-
Entertain me, what problems do analogue computers have?
I suggest that you develop an analog CPU instead of a digital one. You will immediately see all the "problems" yourself.
-
That was an offer.
What is the suggestion? To organize a crowdfunding project on Indiegogo for this invention, with the withdrawal of funds to your PayPal card?
That is, when I look for an offer of $ 500, then you do not trust me. But when do you propose to pour a hundred thousand dollars on your card, should I trust you?
Well, let's say ...
But you, too, must understand that preparing for such a project will itself take time. How will I live all this time so that I can work on the project?
Like it or not, money is needed.
My parents support me, but I've got some $7 for spending on my bank account if that can help. $400 is likely coming next summer, after cleaning and booking our apartments. This years salary I spent on a brand new Celeron.
However, I might have some risky job for you, if you are interested. I think it is about two weeks job, and we can split the money after selling the thing out. I need a brand new canvas text editor (nothing fancy) we can use to create content management system based on uploaded SVG-s utilizing this (https://e-teoria.github.io/Orbiteque/) concept. I also have some $80 of unspent Google ads coupon for the advertisement. Think about it, I can send you a private message with details if you're interested. I usually work at a very slow pace, but I could speed it up if it's urgent for you. In two-three weeks we could be up and running, seeing if anyone would like to use it.
I'm sorry, it's a risk, but that's the best I can do right now.
-
I'm sorry, it's a risk, but that's the best I can do right now.
Thank you, but I don’t have time for risk now. You forget that I am no longer young. I have very little time to implement my plan. I can’t spend it on meaningless "risks."
-
Entertain me, what problems do analogue computers have?
I suggest that you develop an analog CPU instead of a digital one. You will immediately see all the "problems" yourself.
Yeh I was just seeing if you knew what they were. I know my tech, I can make adders, dividors and subtractors, I haven't quite got to multiply yet tho.
Analogue computing is excellent, it makes for a lot less parts, 1 wire compared to 64 is a huge reduction, we are talking the difference between a little card and a ping pong table, from what I know.
-
I'd help but I need a kid-friendly example as a viz on image clearly showing (in less than 1 hour to grasp) what it is doing to speed up compute/storage. Lots of people would help you, Someone may steal it but I assure you no one on here if you PM us will steal it.
Do you have basic shelter? If so, then it is more important you get this out there, stolen and breed the idea.
-
I'd help but I need a kid-friendly example as a viz on image clearly showing (in less than 1 hour to grasp) what it is doing to speed up compute/storage. Lots of people would help you, Someone may steal it but I assure you no one on here if you PM us will steal it.
Do you have basic shelter? If so, then it is more important you get this out there, stolen and breed the idea.
I ask you to confirm whether I understand your proposal correctly. If I make a presentation for you, where I explain the functions of my invention (without explaining the secrets of the technology), will you be ready to start financing my project in the indicated amount - $ 500 per month? (Naturally, every month you will receive a report on the work done and we will be able to discuss your share in the future company)
-
If it looks good - to increase AGI/compute, I would want to pay and also attempt to find other backers who can pay more but then you may not want that part to happen. If it's truly good, I want it to come faster, so I'd want to help it+the inventor.
I cant 'guarantee' though unless see it...else I'd be lying if dont pay lol
-
If it looks good - to increase AGI/compute, I would want to pay and also attempt to find other backers who can pay more but then you may not want that part to happen. If it's truly good, I want it to come faster, so I'd want to help it+the inventor.
I cant 'guarantee' though unless see it...else I'd be lying if dont pay lol
Ok I need some time to create a presentation. Maybe a few days. I will try to do as quickly as possible. When ready, I will write about it.
-
Just demonstrate your knowledge of electrical hardware, then maybe youll work out you don't need a large amount of funding anymore - Just go to the electronics store with a few hundred only! :)
-
I think he's not talking about building this and soldering wires and components together in a garage. I believe he is referring to creating a chip using modern equipment, techniques and materials, employing proper environments like a clean room, scopes, magnification and etching areas, for ic's, etc.
Just my thoughts that he might not be interested in hacking some crude prototype. It would likely be far too large without proper equipment.
Don't know for sure as I can't speak for this person.
If what he indicates is factual, the best thing he could and should do, would be to get the heck out of that country and go someplace safer where he could have the opportunity to earn a decent income without risking his health in doing so.
I agree that some personal background info would go a long way with regard to his credentials, previous work, papers, patents (if applicable), work history, degrees (if applicable), etc. Otherwise, he's just a person with a dream. Nothing wrong with that...but it doesn't carry a lot of weight or credibility.
Regardless, I wish him well.
-
Done.
ProgBrain project presentation. (28 minutes).
https://youtu.be/GvDyl2qKvYg (https://youtu.be/GvDyl2qKvYg)
I ask everyone who is willing to participate in the financing of this project to get in touch and start negotiations.
-
Otherwise, he's just a person with a dream.
You underestimate me very much.
-
far be it for a lot us greens to say but
Ah my young padywagon… you are underestimating us.
-
https://en.m.wikipedia.org/wiki/SpiNNaker
:)
-
far be it for a lot us greens to say but
Ah my young padywagon… you are underestimating us.
Behave decency yourself, please. If you can’t, then do not visit this thread - this will help you maintain your dignity.
-
https://en.m.wikipedia.org/wiki/SpiNNaker
:)
Cool link. :)
Great proof that using modern architecture for processing neural networks cannot achieve acceptable compact devices.
-
The compactness is a secondary consideration, getting the tech to work and do something useful is primary.
Many companies have been perusing your proposal for many years, SpiNNaker is just an example.
Whilst I sympathise with your predicament, I doubt the amount of money required to bring such a schema to fruition will be obtained from a forum like this.
Perhaps you should apply to the many Venture Capital sites, set-up exactly for this purpose.
:)
-
Korr has a good point, but there is also something else... How to persuade potential investors that you have a groundbreaking technology without actually exposing it? This question grows to its extents as the amount of investment money grows up.
I wonder what would happen if you had a contact inside, say, IBM? You could finally expose your technology to that one person that is a key for investment action, but could you trust them? I'd go for it, if it was about my invent, but it's your decision.
For confirming your skills, if you don't want to expose your main patent, I think it would help if you had a number of papers published and cited on, say, Academia.edu or a similar site. Some publicity can't hurt, but it would be a project of its own complexity, taking some more years to accomplish.
Anyway, it would be cool to show up some concrete, tangible results on the mothersite.
-
Perhaps you should apply to the many Venture Capital sites, set-up exactly for this purpose.
I've already done it. But so far it does not help ...
-
This question grows to its extents as the amount of investment money grows up.
I do not need to expand investments. The project will begin to make a profit immediately after the completion of the prototype. Firmware based on FPGA may be a sufficient “first generation†product. All the money that will be needed to launch my architecture on ASIC can be earned by selling bitstreams for FPGA.
Now I need very little money. $ 3000 for 6 months or $ 500 per month or $ 125 per week, for work on architecture. When the Verilog code is finished, I will need about $ 1000 more to buy a good FPGA board, because my board is very weak and cheap - it cannot be sold as a commodity.
All. No more investments are needed. These are not millions of dollars. It’s not even hundreds of thousands ... This can be realized with just one financial partner who can allocate $ 500 per month from his salary (I know that in other countries salaries allow this to be done). For such help, I am ready to offer a 20% share of the future company (without the right to manage). I think this is a very good suggestion. But if I do not find such a partner, as I said earlier, I will leave for labor emigration to another country, spend a year or two years of time, but I will earn this money myself. Then I will not owe anything to anyone. But it will waste precious time.
I wonder what would happen if you had a contact inside, say, IBM?
It's impossible. They do not read letters. They do not answer calls. I’ve been trying to get in touch with big companies about my inventions since 2004, when I won the first round of the inventors' competition in my country. Large companies ignore any third-party development. They do not need breakthrough inventions. They need the gradual development of their technologies no more than + 50% per year in order to skim the cream from rich people every year.
If this were not so, then each such company would have done many inventor contests every year, and in our world there would have been thousands of people like Elon Mask, and we would all have flown to the moon in our own cars for a walk with a girl.
For confirming your skills, if you don't want to expose your main patent, I think it would help if you had a number of papers published and cited on, say, Academia.edu or a similar site. Some publicity can't hurt, but it would be a project of its own complexity, taking some more years to accomplish.
It does not make sense. You do not understand. I don’t have time to write articles. I have no money to survive. The day before yesterday I had the last meal in my house. But yesterday, (before the Happy New Year's), one woman bought 1 liter of my vegetarian shampoo from me, which I cook at home manually. This brought me $ 6 and I was able to buy food for several days. But that money is not enough to survive. So I have to look for work in the quarries. In the quarries, shifts of 12 hours 2-3 hours per round trip (on foot). Where can I take the time to write articles?
All these proposals on your part are meaningless. I have already tried everything that was in my power. I have no other ways. But my and your situation are very different. Therefore, you cannot understand my problems.
Therefore, I want to ask you not to give me advice anymore. If you can’t help finance this project, I don’t want to waste time talking meaninglessly. I hope that you will understand me.
Thank you.
-
Just so you know I do hardware theory, heres a Mario brothers type tile based pneumatic machine designer I put together a little while ago.
(https://scontent.fper6-1.fna.fbcdn.net/v/t1.0-9/71273709_1003550249993032_4269233045963276288_n.jpg?_nc_cat=108&_nc_ohc=5efKwiiilXYAQlWLl8SX5xgvUgkAbj_v6dqOhPDQ4SjsyR07J_GXXxxEw&_nc_ht=scontent.fper6-1.fna&oh=087bc925778c9a40dfc05427bd99d875&oe=5E6DFA08)
whats there->
* vias
* resistors
* diodes (In the form of a connected door that hinges open and closed, and Ive only seen them in action once! but they suit my method great.)
* capacitors (where the little mini letters are. the capacitors push up vertically, and split the design into 2 levels, so I describe what halves are conducting to the capacitor)
and no valves/transistors! in my work, just like I said, and that is NOT the industry norm. So my design is all one connected body, so it suits being made out of rubber. (so I can make my rubber monster with it teehee)
But its not tested, its all just in my head still, but I know one thing, that all you need is 2 biscuit layers to do anything, you never have 2 vias on top of each other, and u can lay down any logic function.
Next time I can be bothered, ive got a better way to do it than tile based, and it uses that dot grid paper instead. (do you know what I mean?) It means that the walls of the conductors sit on the lines on the grid paper, instead of taking up whole blocks and its ALOT smaller footprint!
About that video you posted, which had the graph there for how long a cpu takes to build, being 4 years or something, is just a fairy tale, they make one cpu in under 2 hours tops, and its not to 'draw' the design on the silocon, its only to wait for it to dry or set or cure, is the only thing that slows them down. And that's from me thinking for myself! Which of course we all have to do, when ppl are being devious to us and not telling us the whole truth.
A good first machine to make is a pump or even better an oscillator, if you cant make either you cant make anything yet.
-
I do not know what kind of "rubber" and "paper" you write. I do not understand - what does what you wrote have to do with microelectronics.
About that video you posted, which had the graph there for how long a cpu takes to build, being 4 years or something, is just a fairy tale, they make one cpu in under 2 hours tops, and its not to 'draw' the design on the silocon, its only to wait for it to dry or set or cure, is the only thing that slows them down. And that's from me thinking for myself!
I don’t understand what kind of video you are talking about. I have never claimed that manufacturing a processor takes 4 years. At a silicone factory, the manufacture of lithographic masks takes 2-3 months. The manufacture of microchips takes from 2 to 4 months (it depends on what kind of process, because the smaller the process, the more layers you need to perform on a silicon substrate).
Thus, an engineering order at a contract factory (for example TSMC or UMC) takes from 6 to 8 months. Serial order (when you do not need to make new lithographic masks) takes 2-4 months.
But the fact that each processor (I think you are talking about the CPU) is made in 2 hours ... this is definitely a fairy tale, and I don’t know where you got it from.
It is also stupid to say that the processor design does not need to be drawn. In order to make lithographic masks, you need a drawing. This is called topological maps. Such cards need from 100 to 300 pieces for each design. They look like this:
http://i.piccy.info/i9/b63b9b60f983528748003f727f0fe47d/1577873950/30116/1351639/CMOS_NAND_Layout_800.jpg
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fi.piccy.info%2Fi9%2Fb63b9b60f983528748003f727f0fe47d%2F1577873950%2F30116%2F1351639%2FCMOS_NAND_Layout_800.jpg&hash=36d9ad7b572d14bee8b79541e7e002127fe0264b)
This is how topology design looks in a special program for developing topological map design:
http://i.piccy.info/i9/aaf5c618c40b5f72c11f76052f42ae9b/1577874007/58572/1351639/1_800.jpg
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fi.piccy.info%2Fi9%2Faaf5c618c40b5f72c11f76052f42ae9b%2F1577874007%2F58572%2F1351639%2F1_800.jpg&hash=8ed4a06b9f312fc2aa4b500be957482f0c9f1d51)
This figure shows the size of the memory cell on the chip with technology 7 nm (compare this with your "rubber" and "paper"):
http://i.piccy.info/i9/754873f23023d42f35a1f19b02ab412f/1577874057/68006/1351639/2_800.jpg
(https://aidreams.co.uk/forum/proxy.php?request=http%3A%2F%2Fi.piccy.info%2Fi9%2F754873f23023d42f35a1f19b02ab412f%2F1577874057%2F68006%2F1351639%2F2_800.jpg&hash=aa4094a8abcd9f3930e0d5b4984034e93951cad1)
Which of course we all have to do, when ppl are being devious to us and not telling us the whole truth.
You accuse me of YOU cheating. Ok... give evidence that I deceived YOU.
If you say that I am not telling YOU the whole truth, then this is true and I will not tell YOU the whole truth, because this is the subject of my copyright based on intellectual property laws. And if you personally are not comfortable with the fact that I do not want to reveal to you the secrets of my invention, then you can sue me...
A good first machine to make is a pump or even better an oscillator, if you cant make either you cant make anything yet.
Which pump? Air? Which generator? At 220 V? Or at 380 V?
How can you blame me for the fact that I can’t do anything if you don’t know anything about me? My seniority began in 1994. In my life I did a lot of real things, including designing and manufacturing CNC machines, a special tool for processing super-hard materials, and much more. What nonsense are you talking about? What is in your head?
Put your thoughts in order and talk about serious things. If you do not understand microelectronics, then I do not need to talk about the fact that microchips are made of rubber and paper.
-
Im actually only guessing, cause I haven't manufactured a batch of cpus yet.
YOU are too, making computers from scratch is a very rare thing. its not normal! much more special than that.
I meant 2 hours after the lithographic maps are made of course, yes they take longer to make than the finished cpus do. But I bet theres still tricks there to build them faster, especially if they are made out of repeated patterns, did you know disco balls repeat light? so you could disco ball it onto the pcb maybe, like a kaleidoscope and get the job done very quickly.
Whats air power in a rubber machine got to do with microelectronics?
Theres a huge analogy between the two, ohms law works with pneumatics and hydraulics as well as electricity, and theres a few pneumatic computers on the internet.
If things turned out a different way, our neighbourhoods could be powered with air, and no-one even had electricity - and it pretty much would be just as good - but you cant generate light with air, but everything else works fine.
And you seem to know a bit about it, because I never said paper, I just said rubber. :2funny: *twilight zone theme*
-
And you seem to know a bit about it, because I never said paper, I just said rubber. :2funny: *twilight zone theme*
"Next time I can be bothered, ive got a better way to do it than tile based, and it uses that dot grid paper instead."
What do you want from me? This topic is not about your paper and rubber. Thank you for stopping by, but your messages are offtop.
-
I think it's because you mentioned "paper" a couple of times in your article.
For the other readers...the lead of a standard wooden pencil is about 1 millimeter. Now imagine one millionth of that or that a human hair is roughly 75,000 nanometers!
Flamehowk mentioned 7 nm technology so that is extremely small.
-
And you seem to know a bit about it, because I never said paper, I just said rubber. :2funny: *twilight zone theme*
"Next time I can be bothered, ive got a better way to do it than tile based, and it uses that dot grid paper instead."
What do you want from me? This topic is not about your paper and rubber. Thank you for stopping by, but your messages are offtop.
Oh ok I missed that. But what about ohms law, that's got to do with air in a pipe the same as electricity, no? I thought you said you were some old wise dude.
But how are u supposed to design a computer if you don't have grid paper somewhere to design it on? :idiot2:
-
But what about ohms law, that's got to do with air in a pipe the same as electricity, no? I thought you said you were some old wise dude.
You beguiled electrical engineering and aerodynamics.
But how are u supposed to design a computer if you don't have grid paper somewhere to design it on? :idiot2:
I have special programs on my computer.
-
Ok I see that it makes no sense to be on this forum. Therefore, I am leaving the branch. If in the near future there will be a person who is ready to invest in my project, my contacts:
email: flamehowk@gmail.com
telegram: @flamehowk
skype: flamehowk
viber: +380684456284
bye