Overclock.net banner

[ARStechnica] Microsoft makes play for next wave of computing with quantum toolkit[Update]

2K views 30 replies 13 participants last post by  HowHardCanItBe 
#1 ·
https://arstechnica.co.uk/gadgets/2017/09/microsoft-quantum-computing-toolkit/
Quote:
Later this year, Microsoft will release a new quantum computing programming language, with full Visual Studio integration, along with a quantum computing simulator. With these, developers will be able to both develop and debug quantum programs implementing quantum algorithms.
I may learn my first programming language after all. Looks like I am going to teach myself " borrowing elements from C#, Python, and F#. " in order to prepare.
Update:

Download here
Quote:
Get to know Q#, the brand-new quantum-focused programming language. Fully integrated with Visual Studio, enterprise-grade development tools give you the fastest path to quantum programming
 
#3 ·
Haha, they missed mobile, they aren't going to miss this one.
wink.gif
thumb.gif
 
#4 ·
32GB of ram to work with 32 qubit simulation.
Quote:
Originally Posted by caenlen View Post

https://arstechnica.co.uk/gadgets/2017/09/microsoft-quantum-computing-toolkit/
I may learn my first programming language after all. Looks like I am going to teach myself " borrowing elements from C#, Python, and F#. " in order to prepare.
I'm not going to dissuade you, but realize that this is really, really far from being useful.

You would be much better served learning another language now that has a practical application. Your knowledge will transfer to the programming of quantum computers when they are ready for mainstream in 300 years.

If you are really hyped - it would be useful to learn about the logic behind quantum computing. Still, my bet would be on just learning c# web development or something like that now and applying it later.
Quote:
Originally Posted by remz1337 View Post

Visual Studio integration! awesome
I think it just means syntax and code highlighting. Which is cool I guess.
 
#6 ·
I don't know where you're getting the idea that quantum computing is a far-off technology, there are in-production quantum CPUs right now. Both IBM and D-Wave produce them; miniaturization is inevitable, and I would bet we are closer to a 15-20 year timeframe at most for this to be practical. While that does make this still a little "useless" in terms of practicality, it will be something good to understand going forward.
 
#7 ·
Quote:
Originally Posted by williamdabastrd View Post

I don't know where you're getting the idea that quantum computing is a far-off technology, there are in-production quantum CPUs right now. Both IBM and D-Wave produce them; miniaturization is inevitable, and I would bet we are closer to a 15-20 year timeframe at most for this to be practical. While that does make this still a little "useless" in terms of practicality, it will be something good to understand going forward.
In production quantum annealing "CPU"s (not a CPU, more of a co-processor than anything).
Very different from what "real" quantum processor is when typically thought about. The current stuff is meant for 1 type of problem and that one only, and it cannot do other things nor does it use other quantum mechanics that would be needed to make true, end to end quantum computing a thing. By that I mean the setup of the problem to be solved is done "by hand", the communication into the quantum processor is entirely old school bit mechanics, and all data outside of the problem during the solving is still old bit stuff, all busses are the typical stuff we use in other computers.

here, read this article to get a feel of the differenes between a quantum annealing processor and other, large scale (universal) quantum computing:
https://medium.com/quantum-bits/what-s-the-difference-between-quantum-annealing-and-universal-gate-quantum-computers-c5e5099175a1
 
#8 ·
Quote:
Originally Posted by _02 View Post

32GB of ram to work with 32 qubit simulation.
I'm not going to dissuade you, but realize that this is really, really far from being useful.

You would be much better served learning another language now that has a practical application. Your knowledge will transfer to the programming of quantum computers when they are ready for mainstream in 300 years.

If you are really hyped - it would be useful to learn about the logic behind quantum computing. Still, my bet would be on just learning c# web development or something like that now and applying it later.
I think it just means syntax and code highlighting. Which is cool I guess.
I think you underestimate how fast quantum is going to be moving. It is cold war 2.0 mentality at the moment to reach the first AI and quantum computers, things happen with that kind of mentality and money flow.
 
#9 ·
Quote:
Originally Posted by caenlen View Post

I think you underestimate how fast quantum is going to be moving. It is cold war 2.0 mentality at the moment to reach the first AI and quantum computers, things happen with that kind of mentality and money flow.
Quote:
Originally Posted by williamdabastrd View Post

I don't know where you're getting the idea that quantum computing is a far-off technology, there are in-production quantum CPUs right now. Both IBM and D-Wave produce them; miniaturization is inevitable, and I would bet we are closer to a 15-20 year timeframe at most for this to be practical. While that does make this still a little "useless" in terms of practicality, it will be something good to understand going forward.
You guys are misinterpreting what I said.

I'm talking about learning quantum computing languages provided by microsoft, the topic posted by OP. Not hardware evolution.

All technology moves fast, and it seems a waste to learn something like this now that might be applied in 10+ years. Especially when the fundamentals of that can be learned and applied with current technology. Ask anyone who works in technology now if they'd invest in learning a development platform now, that will be released in 10+ years.

But if it really floats your boat, go for it =)
 
#10 ·
Quote:
Originally Posted by _02 View Post

All technology moves fast, and it seems a waste to learn something like this now that might be applied in 10+ years. Especially when the fundamentals of that can be learned and applied with current technology. Ask anyone who works in technology now if they'd invest in learning a development platform now, that will be released in 10+ years.
10+ years?

50 qubits is where a quantum computer becomes faster than any classic computer (at specific problems):
https://www.pcworld.com/article/3176853/hardware/ibms-new-q-program-to-include-a-50-qubit-quantum-computer.html
https://arstechnica.co.uk/gadgets/2017/03/ibm-q-50-qubit-quantum-computer/
 
#11 ·
Quote:
Originally Posted by Asmodian View Post

10+ years?

50 qubits is where a quantum computer becomes faster than any classic computer (at specific problems):
https://www.pcworld.com/article/3176853/hardware/ibms-new-q-program-to-include-a-50-qubit-quantum-computer.html
https://arstechnica.co.uk/gadgets/2017/03/ibm-q-50-qubit-quantum-computer/
I was quoting the 10+ year stat from above. The linked article states
Quote:
... quantum computing future is, fortunately, still likely to be many years off.
I'm just advocating to learn programming if you are interested in implementing this. You aren't going to learn programming on a bleeding edge quantum simulation with its own form of logic, but you can bend already gained programming knowledge to utilize the quantum paradigm. The tutorials and language are being released as a kind of wrapper for people who already understand programming to begin wrapping their heads around a non binary quantum system and how to create logic for it.

Do as you will, but trust me, you're better served going into that understanding functions, parameters, iteration, etc etc etc.

/drum /soapbox /deadhorse

OT - it is pretty cool that you can debug the quantum logic. I will admit that tickled my fancy a bit.
 
#12 ·
Fair enough, most quantum programming will probably be exactly the same as classical programming, calling functions is calling functions after all, but it is happening way faster than I had expected.

We might see consumer use of quantum computers (via the cloud) in 5 years! I would have called you crazy if you suggested we might see this in 10 years 5 years ago.
 
#13 ·
Quote:
Originally Posted by Asmodian View Post

Fair enough, most quantum programming will probably be exactly the same as classical programming, calling functions is calling functions after all, but it is happening way faster than I had expected.

We might see consumer use of quantum computers (via the cloud) in 5 years! I would have called you crazy if you suggested we might see this in 10 years 5 years ago.
My only argument is that if there are new programming langauges being designed, yes experienced people will still have an edge, but learning the new material parts like quantum gates, well I have seen things happen weirder before, but sometimes the newbie comes to look at things from a unique perspective when learning something over an expert... maybe if I fool around with this toolset in a year or two I will see an opportunity for something small and can get cracking on it... I dunno its w.e

Just liked if I had learned Swift when it first came out, I probably would have had a better chance of getting a job, as most programmers already have their coding certs and jobs... does any of this make sense?
 
#15 ·
If any new developers wanna learn the magnitude of this start learning ec++ mess with low level. Forget lame scripting languages. Bitshifting, twos compliment, etc then realize the power of this. Amazing what microsoft is doing here. Any news on when IBM chips will hit mainstream?
 
#16 ·
Quote:
Originally Posted by Asmodian View Post

Fair enough, most quantum programming will probably be exactly the same as classical programming, calling functions is calling functions after all, but it is happening way faster than I had expected.

We might see consumer use of quantum computers (via the cloud) in 5 years! I would have called you crazy if you suggested we might see this in 10 years 5 years ago.
Im actually surprised we are moving so fast into quantum compuiters and never moved on to photonic computers. Intel and IBM had been researching photon circuits and seemed to be doing breakthroughs, but I guess they just couldnt get the emitters and receivers small enough (into the couple nm size range) to be worthwhile. The interestring thing of photon computer chips would be that unlike silicon transistors with copper interconnects whre you have to route around all the pathways and layers, with photon based circuits the layers could pass tyhrough each other and interact in ways that would allow a massive jump in complextity and capability of computer chips.

I did read a cool article last year or something though that someone found a way to use nanometer size LEDs to output entangled photons for quantum computer use. Would be an interesting hybrid of photonic and quantum computer systems.
 
#17 ·
Quote:
Originally Posted by EniGma1987 View Post

Im actually surprised we are moving so fast into quantum compuiters and never moved on to photonic computers. Intel and IBM had been researching photon circuits and seemed to be doing breakthroughs, but I guess they just couldnt get the emitters and receivers small enough (into the couple nm size range) to be worthwhile. The interestring thing of photon computer chips would be that unlike silicon transistors with copper interconnects whre you have to route around all the pathways and layers, with photon based circuits the layers could pass tyhrough each other and interact in ways that would allow a massive jump in complextity and capability of computer chips.

I did read a cool article last year or something though that someone found a way to use nanometer size LEDs to output entangled photons for quantum computer use. Would be an interesting hybrid of photonic and quantum computer systems.
If you could link the article to what your references with photon stuff I would like to read it, never heard of that before.
 
#18 ·
I would also be surprised if a quantum computer device would reach first commercialization to the common day PC's.

My bet would be photonic/optical cpu's etc would be the biggest leap to come in decades, first as perhaps some ring like bus and later as a volatile memory cache/buffer mechanism.

It reminded me of this news i saw a while back on guru3d about researchers at MIT with a neat video of the cpu running a (very basic) 3d render image of a pot, it could probably run pong!
biggrin.gif

 
#19 ·
thought you were kidding, yep they made a pot, lol. I am hoping Putin's rhetoric on the first country to get AI will rule the world line will start a new race for tech... because well its true what he said.
 
#20 ·
Quote:
Originally Posted by _02 View Post

You guys are misinterpreting what I said.

I'm talking about learning quantum computing languages provided by microsoft, the topic posted by OP. Not hardware evolution.

All technology moves fast, and it seems a waste to learn something like this now that might be applied in 10+ years. Especially when the fundamentals of that can be learned and applied with current technology. Ask anyone who works in technology now if they'd invest in learning a development platform now, that will be released in 10+ years.

But if it really floats your boat, go for it =)
Agree. Most likely will take a few months to learn the language as with any programming language, however by the time the juicy stuff starts happening, the language will have moved on and what was learned early on wont be of much use. Still will be interesting to have a play when it is released to see how it is.
 
#21 ·
Quote:
Originally Posted by caenlen View Post

If you could link the article to what your references with photon stuff I would like to read it, never heard of that before.
http://www.eenewseurope.com/news/pyramidal-micro-leds-output-entangled-photons-quantum-computing
https://www.sciencealert.com/researchers-just-developed-the-most-complex-quantum-computer-chip-so-far-out-of-light

The original thought a decade ago was that we would be able to reach photon CPUs within 1-2 decades and quantum computing was still 50+ years off. So light based CPUs would be the stopgap breakthrough between our current computing and quantum. However it seems that quantum computing is making massive strides and is far closer than ever thought, and most of the research on photon CPUs has actually merged with quantum computing. Some of the breakthroughs in the photonic CPU stuff has actually had the purpose of being useful for the quantum CPUs. So now the work being done is to use the photonic processors in quantum entanglement and processing.

This is also an interesting read:
http://news.berkeley.edu/2015/12/23/electronic-photonic-microprocessor-chip/
 
#25 ·
Quote:
Originally Posted by spinFX View Post

Agree. Most likely will take a few months to learn the language as with any programming language, however by the time the juicy stuff starts happening, the language will have moved on and what was learned early on wont be of much use. Still will be interesting to have a play when it is released to see how it is.
Ahahahahahaha A few months? Your not just learning new syntax, your having to undue and base two experience from your brain and learning base three (quantum ternary state) now. There is a huge difference. Not only that, even base two languages or script kiddie languages (interpreted languages ie python, java-script) take years to master correctly to the point you can easily manipulate object orientation.

EDIT: This also is Microsofts "topological qubit" is what will make quantum computing mainstream. If you have cash buy stock now. M$ just took the next leap.
 
#26 ·
Quote:
Originally Posted by Dimaggio1103 View Post

Ahahahahahaha A few months? Your not just learning new syntax, your having to undue and base two experience from your brain and learning base three (quantum ternary state) now. There is a huge difference. Not only that, even base two languages or script kiddie languages (interpreted languages ie python, java-script) take years to master correctly to the point you can easily manipulate object orientation.

EDIT: This also is Microsofts "topological qubit" is what will make quantum computing mainstream. If you have cash buy stock now. M$ just took the next leap.
Microsoft will finally be worth some money =p

I think you're overestimating effort. If you already understand how to write production code, picking up a new paradigm or language should take you weeks, not years. There is a big difference between obtaining mastery and competency. I may have met a handful of people in my life that qualify as masters of a language or even a framework within a language.

Not to downplay the magnitude of a new paradigm, but saying it will take 1+ years to implement something in this is a bit crazy (to me)

A new user actually stands to be at a benefit as they have less to unlearn - but I would still not recommend cutting your teeth on something so new, with so little support and existing documentation.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top