If Aliens leave Pioneer alone, it will likely be gravity captured and seed another planet, star, asteroid or black hole. If it's a star, it will vaporize and sterilize. A black hole will compress it into nothing. However, it may "seed" a planetoid. Maybe those tiny bugs will grow into people with the right conditions... er... bug people. But Pioneers did not have a supercomputer on board. They contained a single core custom designed computer. Voyager and numerous satellites contained the RCA CDP1802MPU because it was more "space hardened" which became popular in some early microcomputer kits and scratch computers built by hobbyists.
Now I know that as this is in England and the display is switched off most of the time but that does not mean it has stopped running.
Heater, I like the idea of a supercomputer running with no display. I think it can compute faster internally without going through any added interfaces. I also have in mind a project that can extract this information with no connections, so the supercomputer can keep on computing without obstructions or sponge-off of resources.
Some people claim the definition of supercomputer is being the fastest. That means one, until the next one bumps it off. In doing some research about supercomputers, an interesting fact has arisen. In each year, there are numerous companies, and some individuals, who create "supercomputers" that are clearly not the fastest computer at that time. Some make it to the top ten and some do not. Some make it to the top 500 and some do not. So if ultimate speed at that moment is NOT the criteria for a supercomputer, then what is? How fast must a supercomputer become to be classified as a supercomputer when clearly it cannot reach the top 500? It seems the definition of a supercomputer has different meanings to different sources.
As someone who has worked on building the Jaguar supercomputer mentioned earlier, I guess it's my cue to comment
First, Jaguar has plenty (10's of thousands) of LED's that you can toggle at 10MHz, under software control. Your odds of seeing any of them with the cases shut are near zero. Not to mention that there's no point Meantime, there's also quite a lot of GPIO but it's all on the system control side (mostly in the form of unmentionable 3 letter microcontrollers , not the computation side. The computation side is all Opterons with no attached southbridge. Each pair of Opteron sockets runs standard Linux with some enhancements and a stripped down userspace optimized for running applications.
In any case, the list of supercomputing definitions pointed to is quite out of date. Obviously, the Cray Y-MP was designed as a supercomputer, but is less powerful than the AMD under my desk. There isn't really a good, concise, definition of supercomputer right now, even for those of us in the industry. What we tend to think of as a good definition is, in fact, a general purpose computer that can make the Top 500 list. There are other high-performance computing platforms (including small versions of Jaguar, aka the Cray XT5, which is modular) that really don't meet the definition of supercomputer. For instance, I have one of the XT5's blades downstairs in a tester. I'd hardly call the 8 Opterons on it a supercomputer, even though piling 4800 of those blades together gets me Jaguar.
I like the "supermicrocontroller" name to describe Humanoido's tower. It's phenomenal at doing the things microcontrollers do well: controlling GPIO pins, talking low-level hardware protocols, controlling screens, etc. What it doesn't do well is what traditional supercomputers do well: floating point math, large memory simulations, and parallel programming models. The Prop, awesome as it is, has no floating point and tiny amounts of memory.
If your goal is to create a cool piece of microcontroller hardware that nobody else has and can control massive amounts of I/O, the Prop Tower is pretty darn sweet. However, if your goal is to learn how "mainstream" massively parallel machines are programmed, you're not going to get there with the Prop.
P.S. I've been working to get us to use a Prop in our control paths. No luck yet, but I'm still trying
Gosh, thank you for that insight into the world of todays supercomputers. It's always good to hear from someone with hands on experience.
I'm in total agreement about "supermicrocontroller" let' hope that is the end of the semantic debate that has been rolling along here for too long.
Which brings me back to by original question, what is the ideal application for Humanoido's tower of 40 Props with so much I/O closely coupled to the processors?
I'm in total agreement about "supermicrocontroller" let' hope that is the end of the semantic debate that has been rolling along here for too long.
I like it and appended it to my latest 40 prop project name.
But... as these machines are, some of the things they are less suitable at doing, we will still do them. Well, if you have a small car, it is not the best for long trips, but we still take long trips to reach our final destinations. Maybe we cannot afford the million dollar luxury car which runs at high speed and floats in suspension with comfort. So we make do with what we have, and admire those big machines and then men who build them.
Which brings me back to by original question, what is the ideal application for Humanoido's tower of 40 Props with so much I/O closely coupled to the processors?
Heater, the answer to this question is so simple. To have fun.
I posted a similar question is another thread and got some interesting replies. My BASIC Stamp robot friends think this will blow open the lid on robotics. I learned never to doubt their judgments.
Comments
Humanoido
Now I know that as this is in England and the display is switched off most of the time but that does not mean it has stopped running.
Various definitions of a Supercomputer
First, Jaguar has plenty (10's of thousands) of LED's that you can toggle at 10MHz, under software control. Your odds of seeing any of them with the cases shut are near zero. Not to mention that there's no point
In any case, the list of supercomputing definitions pointed to is quite out of date. Obviously, the Cray Y-MP was designed as a supercomputer, but is less powerful than the AMD under my desk. There isn't really a good, concise, definition of supercomputer right now, even for those of us in the industry. What we tend to think of as a good definition is, in fact, a general purpose computer that can make the Top 500 list. There are other high-performance computing platforms (including small versions of Jaguar, aka the Cray XT5, which is modular) that really don't meet the definition of supercomputer. For instance, I have one of the XT5's blades downstairs in a tester. I'd hardly call the 8 Opterons on it a supercomputer, even though piling 4800 of those blades together gets me Jaguar.
I like the "supermicrocontroller" name to describe Humanoido's tower. It's phenomenal at doing the things microcontrollers do well: controlling GPIO pins, talking low-level hardware protocols, controlling screens, etc. What it doesn't do well is what traditional supercomputers do well: floating point math, large memory simulations, and parallel programming models. The Prop, awesome as it is, has no floating point and tiny amounts of memory.
If your goal is to create a cool piece of microcontroller hardware that nobody else has and can control massive amounts of I/O, the Prop Tower is pretty darn sweet. However, if your goal is to learn how "mainstream" massively parallel machines are programmed, you're not going to get there with the Prop.
P.S. I've been working to get us to use a Prop in our control paths. No luck yet, but I'm still trying
Gosh, thank you for that insight into the world of todays supercomputers. It's always good to hear from someone with hands on experience.
I'm in total agreement about "supermicrocontroller" let' hope that is the end of the semantic debate that has been rolling along here for too long.
Which brings me back to by original question, what is the ideal application for Humanoido's tower of 40 Props with so much I/O closely coupled to the processors?
I like it and appended it to my latest 40 prop project name.
But... as these machines are, some of the things they are less suitable at doing, we will still do them. Well, if you have a small car, it is not the best for long trips, but we still take long trips to reach our final destinations. Maybe we cannot afford the million dollar luxury car which runs at high speed and floats in suspension with comfort. So we make do with what we have, and admire those big machines and then men who build them.
Heater, the answer to this question is so simple. To have fun.
I posted a similar question is another thread and got some interesting replies. My BASIC Stamp robot friends think this will blow open the lid on robotics. I learned never to doubt their judgments.