Comments 1 - 18 of 18 Search these comments
The Machine’s design includes other novel features such as optical fiber instead of copper wiring for moving data around. HP’s simulations suggest that a server built to The Machine’s blueprint could be six times more powerful than an equivalent conventional design, while using just 1.25 percent of the energy and being around 10 percent the size.
The next computing breakthrough most likely comes from either
1) Google, or
2) an unheard-of company
printer ink.
I read once that printer ink (sold in $36 cartridges) is something like 5000 times more expensive per ounce than the world's finest champagnes!
The next computing breakthrough
What we really need is an NP Computer.
Has you seen "The Travelling Salesman"?
printer ink.
I read once that printer ink (sold in $36 cartridges) is something like 5000 times more expensive per ounce than the world's finest champagnes!
Yep. It is highly lucrative.
What we really need is an NP Computer.
Many complex problems have much simpler approximations. We want a computer that think like life.
In the end, would you prefer the "best" solution in 100 years or a 99.9% optimal one in 5 seconds?
Cryptography based on "asymmetric" functions or any form of "computability" is not sustainable.
It sounds like they've developed a new type of vaporware, by announcing non-existent software for hardware that doesn't exist yet either:
It reminds me of a Beatles song: "I got no car and it's breaking my heart, but I've found a driver and that's a start." I hope it works out OK but meanwhile I'll stick with the Beatles version.
It sounds like they've developed a new type of vaporware, by announcing non-existent software for hardware that doesn't exist yet either
HP has nothing to lose: they're pretty much down to selling expensive ink.
That's why they're dangerous.
The best HP computer to date is still the trusty 12C.
Funny you mention that.
My college roommate and I both took an intro computing course (around 1980).
Our professor was very old school and taught the course using punch cards fed into a gigantic IBM 390 at computer center at the edge of campus.
However, he was very flexible in how we presented our assignments, since the course was more about numerical analysis than programming.
So, while I and the rest of the class carried around cartons of punch cards to the center, and waited hours for it to spit out watermelon paper that was the results, my friend did the entire coursework in his bed, programming on his HP and storing the results on the tiny memory strips that it used. I forget how he printed out the results, but I believe it had some sort of transfer or print cable.
I'll take my optimal local police force every time...
In the end, would you prefer the "best" solution in 100 years or a 99.9% optimal one in 5 seconds?
So, while I and the rest of the class carried around cartons of punch cards to the center, and waited hours for it to spit out watermelon paper that was the results, my friend did the entire coursework in his bed, programming on his HP and storing the results on the tiny memory strips that it used. I forget how he printed out the results, but I believe it had some sort of transfer or print cable.
Some earlier HP calculators came with printers.
Here is an online museum of HP calculators:
My college roommate and I both took an intro computing course (around 1980).
Our professor was very old school and taught the course using punch cards fed into a gigantic IBM 390 at computer center at the edge of campus.
However, he was very flexible in how we presented our assignments, since the course was more about numerical analysis than programming.
So, while I and the rest of the class carried around cartons of punch cards to the center, and waited hours for it to spit out watermelon paper that was the results, my friend did the entire coursework in his bed, programming on his HP and storing the results on the tiny memory strips that it used. I forget how he printed out the results, but I believe it had some sort of transfer or print cable
Sounds more like an urban legend to me. The 390's didn't even come out until the 90's and I don't think the 12c was around till 82/83 or so and was pretty damn expensive like 400-500 in today's dollars. I did moonlighting as a night operator in various data centers from 78-85 and only saw one punch card machine the whole time. That one only ran really old jobs no one had bothered to put on disk about once a week or so. I bought a 12c circa 1988 or so when the prices came down some. It didn't have memory strips or an external interface. Took me quite a while to really get the hang of RPN. If I remember right you were limited to about 100 lines of code. You could have multiple programs but the total number of lines for all of them couldn't be more than 100. You just went to program mode then jumped to the first line of the program you wanted to run. Crude but effective
Makes a nice story though.
Sounds more like an urban legend to me.
I did moonlighting as a night operator in various data centers from 78-85 and only saw one punch card machine the whole time.
Did you visit the Princeton Computing center in 1980?
This was way before I went there, but they were still using those punch card keyboards as late as 1979-80 (although it seemed to be mostly my class..another professor was teaching a similar course but using interactive terminals and pascal!):
https://twitter.com/muddlibrary/status/535433896289398784
The 390's didn't
A 360 then? We used JCL, and FORTRAN on the WATFOR compiler.
Maybe his calculator was the Ti-59:
http://en.wikipedia.org/wiki/TI-59_/_TI-58
Note the strip in upper image:
http://en.wikipedia.org/wiki/TI-59_/_TI-58#mediaviewer/File:TI-59.jpg
Revolutionary Linux++ from HP allows them to create 'data center in a box'!!
http://www.technologyreview.com/news/533066/hp-will-release-a-revolutionary-new-operating-system-in-2015/