Search the Community

Alright, so I've had this idea for a while now, but have neither the funding nor the time to test it out myself. I thought LTT might find it an interesting video idea. Basically, the idea is to use small cheapo computers to make a bigger one that operates like a human brain. In traditional computers, different parts are specialized for individual functions. GPUs are good at solving many small problems to generate graphics, CPUs are good at solving single big problems, RAM is good at storing small amounts of memory but at a high speed, etc. Everything has an individual purpose. This can sort of be compared to individual parts of the brain, but the way the brain operates is a little different. The brain is made up of neurons that can each act as a "cpu, gpu, ram, and hard drive" all at the same time.... It's not very powerful at any of those things. But thanks to the connections between neurons, they can function as a whole. This leads to less mathematical solving power, but leads to more of the behaviors we attribute to life I thought of a garage way to mimic that way of operating in a computer. Arduinos and Rasberry pis are very similar to neurons in that they're small and weak individually, but are very adaptable and can do a little bit of everything. So I thought "why not connect a couple hundred of them together, get them talking and working together as a democratic unit, and see what happens". I figured it could be tested by throwing some traditional computer tasks at it (like playing video games) to see how it performs (probably terribly compared to a similar amount of money spent on a traditional computer), and then throwing some complex real world tasks at it that traditional computers struggle with (perhaps voice or body language recognition) and seeing if they do any better. I'm not sure how feasible this is, I'm not sure how to get them connected and talking to each other, but it would be interesting to see. LTT could probably take this challenge on and get one built with 100 "neurons" for the same price as a high end gaming PC.

http://www.pcworld.com/article/3005414/computers/intel-plugs-72-core-supercomputing-chip-into-workstation.html Intel's 72-core processor jumps from supercomputers to workstations Intel will ship a workstation with its 72-core Knights Landing supercomputing chip in the first half of next year An engineering sample of Intel's Knights Ferry supercomputing chip. Credit: Intel Agam Shah IDG News Service Nov 16, 2015 7:20 AM Intel wants to change the game in desktop computing with a workstation that packs its upcoming, 72-core supercomputing chip. Intel's workstation will be based on an upcoming Xeon Phi chip code-named Knights Landing, which is being touted as the company's most powerful chip to date. A limited number of workstations will ship in the first half of next year from Intel, which will also control initial distribution. As usage expands, hopefully PC makers and other partners will sign on to sell Xeon Phi desktops, said Charles Wuischpard, general manager of the HPC Platform Group at Intel. Workstations are business desktops typically larger than conventional desktops, with one example being Apple's Mac Pro. The computers are widely used for high-end graphics, film editing and engineering applications. Today's workstations are largely based on Intel's Core desktop or Xeon server chips. Intel's supercomputing workstation will initially be made available to researchers who don't have access to Xeon Phi-based supercomputers for complex scientific calculations. Researchers will be able to write and test code on the computer before deployment on supercomputers. The machine is more of an experiment than an effort to change the way desktops and workstations are designed, Wuischpard said. Coprocessors like Nvidia's Quadro are already being used alongside Intel CPUs in workstations to boost application performance. But Intel's workstation will have only one Knights Landing chip, which will load the OS and increase application performance. The Knights Landing chip can deliver over 3 teraflops of peak performance, which is roughly in the range of some high-performance graphics chips used in the world's fastest supercomputers. Much like graphics chips, Knights Landing is designed for highly parallel computing and varies in design from conventional x86 chips. It mixes conventional x86 CPUs with specialized processing units that help the chip take on heavy workloads. Knights Landing could also bring many innovations to desktops. It includes 16GB of on-package MCDRAM memory, in which modules are stacked and connected through a wire. The memory offers five times more bandwidth than the emerging DDR4 memory. Intel also claims the stacked memory is five times more power efficient and three times denser than GDDR5, which is used on graphics cards. In addition to the OS, programming tools and other software programs will be preloaded on the workstation. The chip itself will be tightly integrated into the system, making it tough to add more memory and components. This workstation could fulfill a long-time dream of Intel to bring a powerful chip to desktops. Intel had such plans with a chip code-named Larrabee, which was unceremoniously dropped in 2010 after design challenges. The Phi chip is a byproduct of Larrabee. Im using pcword.com for title and content, credit to them. any wrong information is not on my side enjoy

Do you guys think people like Carlsen, Kramnik, Aronian, Caruana, etc. could stand up to some of the super computers of today if one was purpose built to process chess? I'm really beginning to doubt whether a person could win considering the speeds of today's computers. The Deep Blue that "beat" Kasparov was made of 120Mhz dedicated chess chips. What if you throw in a few 32 thread 3 Ghz dedicated chip from today, along with some of today's graphics processors? I understand it would take an immense time to create the software and faster hardware would also be available. I also understand that building a dedicated machine like this is not as simple as taking hardware and throwing into a machine then loading software onto it, but still. Use some imagination. I'm just trying to start a conversation here. I think it's an interesting topic to consider.

My new beast Asus x79 sabertooth intel i7 3930k 64g Kingston hyperx memory Sapphire radeon hd 7970 dual-x Asus xonar stx sound cart Asus raidr 240g ssd :D 2kpl 240g kingston hyperx ssd Samsung Spinpoint F3 1 TB Corsair ax850 power Noctua NH-D14 SE2011 + 3kpl 140mm fan Asus 27" pb278q LED displays with PLS :wub: Logitech g15 keyboard Logitech Performance MX Mouse Genelec 6020a speakers It is not winning beauty contest but i am looking pc sceen not my pc :P

Back in fall of 2012 the semiconductor company Adapteva went on kickstarter in search of funds for the Parallella project. The target was to develop an affordable, open, easy to use platform that would allow developers to easily jump on parallel computing. The project got successfully funded and raised 900.000$ from the initial 750.000 asked. 18 months have passed since then and finally backers are starting to receive their perks, and the Parallella board has finally gone into general availability, with international distribution also having recently been put in place. Source: http://www.linux.com/news/hardware/servers/785558-99-parallella-supercomputer-has-successful-launch-after-18-months https://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone The real question now is: Should i get 4 Raspberry Pi and an Arduino board or a Parallella?

Who here wants to see Linus build a super Cluster Computer using 100 Odroid UX4 boards and put it to work? Edit: I really need to proof read...