solarexplorer

11 ms·

61.
by solarexplorer 11y ago
The separate CPU is much faster, because it has an entire die for itself and because it is a high-volume off-the-shelf chip. An embedded CPU/FPGA will suffer much more when it executes the serial part of the application on the embedded
62.
by solarexplorer 11y ago
Also see https://chessprogramming.wikispaces.com/Timeline for a computer chess timeline
63.
by solarexplorer 11y ago
What you describe really sounds like functional programming.
64.
by solarexplorer 11y ago
They don't...
65.
by solarexplorer 11y ago
AFAIK end-to-end encryption can only be used with Android clients. http://www.heise.de/ct/artikel/Keeping-Tabs-on-WhatsApp-s-En... tl;dr end-to-end encryption in WhatsApp is not really useful (yet)
66.
by solarexplorer 11y ago
Yes Clojure, sorry about the typo! The difference is that with Haskell or Erlang you get hard guarantees about side effects, but access to imperative libraries is very difficult. So you are kind of locked in a small ecosystem. With Clojure,
67.
by solarexplorer 11y ago
I keep forgetting about C# and F#, but with Microsoft embracing open source they may become actually interesting for people outside the windows world.
68.
by solarexplorer 11y ago
Other languages like Clojure, Python and even Javascript are catching up to Go and are adding decent support for concurrency too. But I would definitely prefer a functional language for a project that involves a lot of concurrency. "no
69.
by solarexplorer 11y ago
I would have piped this through awk, but this is probably better done with macros. Thanks for taking the time to elaborate. I will definitely give macros a try.
70.
by solarexplorer 11y ago
Isn't :%s/.)/hello)/ simpler? Maybe that's because I know nothing about macros. Thanks for the explanation.
71.
by solarexplorer 11y ago
Performance increase over a JVM _interpreter_. They admit this is slower than compiled code. It would be interesting to see data on the complete Cost/Performance/Power/Memory tradeoff relative to a state-of-the-art JIT compil
72.
by solarexplorer 11y ago
Signal processing. FPGAs have high speed links and can perform thousands of operations in parallel. When you have a huge amount of data to process in real time, an FPGA (or ASIC) is often your only choice.
73.
by solarexplorer 11y ago
ES6 may make the language itself more complex, but I would argue that it does make your code simpler and more concise. Destructuring and arrow functions may seem gimmicks, but they do reduce the visual complexity of the source code and allo
74.
by solarexplorer 11y ago
Well, Apple makes most money with hardware. Unlike Google who makes most money with ads. So at least in theory Apple can more easily invest in privacy. EDIT: Check their last quarter's results, Apple makes only 6% of their revenue with
75.
by solarexplorer 11y ago
I use the (soon to be) language standard as a reference: http://wiki.ecmascript.org/doku.php?id=harmony:specification...
76.
by solarexplorer 11y ago
I remember reading it here: http://www.esquire.com/news-politics/a16681/elon-musk-interv...
77.
by solarexplorer 11y ago
AFAIK Loongson just licensed the patent necessary to implement the MIPS ISA but the implementation was done by chineese academia without any intervention from MIPS. You cannont judge the quality of MIPS processors in general by looking only
78.
by solarexplorer 11y ago
I disagree. This comparison is extremely unfair because they used a chineese clone instead of the real CPU/compiler from MIPS...
79.
by solarexplorer 11y ago
You are right, I would haved loved this as a student too. But today there are real open source alternatives. Especially in an academic environment it is essential to be able to modify and share source code without restrictions so you can ex
80.
by solarexplorer 11y ago
But the reason they break code compatibility or implement older versions of the architecture is that parts of the ISA are patented by MIPS. Other ISAs like OpenRISC, SPARC and RISC-V don't have these limitations.
81.
by solarexplorer 12y ago
Thanks for the link, that was a great read! I agree that the branch (target) prediction will fail horribly on a naive interpreter that calls different branch targets from the same static instruction over and over again. That alone explains
82.
by solarexplorer 12y ago
Yes. I think it will be worth the time to rethink it for Swift, because Swift is much better suited for this style of programming than Objective-C++.
83.
by solarexplorer 12y ago
ComponentsKit is Objective-C++ only (no Swift).
84.
by solarexplorer 12y ago
Here is a photo of the thing in the living room of his parents: http://www.computerhope.com/jargon/z/z1.jpg
85.
by solarexplorer 12y ago
"Triumph des Willens" was released in 1935, two years after the Nazis grabbed power.
86.
by solarexplorer 12y ago
When you want to minimize transfers between disk and ram and your OS does paging, then it makes sense to put unfrequently used data (like error messages) together in a page with other unfrequently used data. It's mostly what you would
87.
by solarexplorer 12y ago
Well, aligning for words or cache lines is good. Aligning for pages not so much. Of course this has been known for ages, but sometimes it is hard to avoid, e.g. inside the Linux Kernel: http://yarchive.net/comp/cache_th
88.
by solarexplorer 12y ago
Not really. See Launch Images in iOS: https://developer.apple.com/library/ios/documentation/UserEx...
89.
by solarexplorer 12y ago
Yes, just like the DPS interpreter translates (high-level) DPS into (low-level) pixels. Looks pretty similar to me. (but then I am not overly familiar with DPS or ReactJS) EDIT: To clarify, I am sure that the current UIKit also tries to do
90.
by solarexplorer 12y ago
Am I getting this right: they use something like Display Postscript[1] instead of the virtual DOM? Are we back in the 80ties? [1] http://en.wikipedia.org/wiki/Display_PostScript

More ›