foxhill

37 ms·

331.
by foxhill 12y ago
there is also pocl ( http://portablecl.org/ ), which aims to one day target GPUs (although currently it does not). the issue with free OpenCL implementations on the GPU is that the runtime will be closely coupled with the dev
332.
by foxhill 12y ago
so, we have a political system whose goal is to maximise it's economic output - or at least this is one goal. but what about its people? aren't the government supposed to work for the people? at what point in history has there bee
333.
by foxhill 12y ago
OpenCL Pro Tip: the __restrict keyword behaves as normal in OpenCL C, so adding it to kernel code may not move read only data into the constant cache. to achieve this, try using the __constant address space.
334.
by foxhill 12y ago
some people are born without the ability to walk. others, are born without the gift of hearing. i was born with an incredibly low "cringe" threshold. reading this "exchange" almost put me in a coma.
335.
by foxhill 12y ago
given that OS X ships with OpenCL, which already runs on nvidia hardware, and is open, and portable, i don' t understand why anyone would want to switch to CUDA.
336.
by foxhill 12y ago
marketing.
337.
by foxhill 12y ago
> "Because humans have unlimited storage capacity..." this is patently untrue, unless human brains defy the known laws of physics. this makes it hard to believe everything that he mentions.
338.
by foxhill 12y ago
DNS is for humans to have memorable addresses, and can be trivially side-stepped.
339.
by foxhill 12y ago
whilst performance does look staggeringly good, it is entirely negated by the pitiful rationalisations made against not being open source. given everything that's happened regarding privacy recently, why on earth should anyone trust th
340.
by foxhill 12y ago
i just ran it on my mac successfully with wine..!
341.
by foxhill 12y ago
for even simple programs, this would produce a very large output. we've become accustomed to the idea that C is a low level language, but actually it's somewhat high level. optimisations are in your code everywhere , and if a C
342.
by foxhill 12y ago
actually, realloc infrequently moves data around, especially if it's large. quite often the returned pointer value is the same. but yes, even if it has been free'd, the page in which it resided is likely still valid, and free do
343.
by foxhill 12y ago
well, this would mean you'd get feedback every time you declare a variable without giving it an initial value? (even if the value is never read from before being written to - i.e is entirely safe to do?) or how about when you write a f
344.
by foxhill 12y ago
the compiler will rely on undefined behaviour all the time for any given set of optimisations, a lot of them are harmless. the compiler can't know if a use of undefined behaviour is intentional and safe, or unintentional and dangerous.
345.
by foxhill 12y ago
specifically, the idea that UI elements should behave as if they were tangible objects isn't a new idea. this is what apple have been doing for.. quite a while, and honestly I didn't think this was breakthrough stuff - elastic bou
346.
by foxhill 12y ago
um.. what do you mean, exactly? all types in C are fixed bitwidth. but the uintx_t types make it obvious what what programmer wants.
347.
by foxhill 12y ago
the stuff he talks about applies to C and C++ - i think we can forgive that faux pas :)
348.
by foxhill 12y ago
really, the issue is, if you want a fixed bitwidth data type, use the types in unistd.h - uint64_t or uint32_t.
349.
by foxhill 12y ago
a single test case hardly means "beating" FLAC. indeed, if you just wanted to encode a sine wave at one frequency for 30 seconds, your encoder would encode this as a couple of bytes - for any general compression codec there will b
350.
by foxhill 12y ago
> Most of the time is spent on generating the random string when i took out string generation, i went from 1.2 GH/s to 1.8GH/s (although, i'm assuming my string generation was much simpler) > I very quickly realised loo
351.
by foxhill 12y ago
i was intrigued by this, so i done it in C. on my 4-core MBP (2.6ghz ivy bridge) i can manage ~1.8 billion hashes per second. i could parallelize with OpenCL, but i think this is enough. after a few minutes, i get ARbyhlf as a valid name
352.
by foxhill 12y ago
DirectX has it's own problems.
353.
by foxhill 12y ago
> Performance largely myth > legacy code using OpenCL is not like OpenMP, you don't just add a few pragmas and you're set. C code needs to be rewritten for OpenCL. this is largely copy and paste, due to the syntax being so
354.
by foxhill 12y ago
> Templates are an essential tool to write type-independent algorithms. They enable meta-programming, an invaluable tool to provide flexible yet efficient active libraries to users. They allow automated kernel-space exploration. So templ
355.
by foxhill 12y ago
templates are not something that (outside of simple uses) you'd want to use in your kernels. regardless, nvidia should be pushing their improvements through to OpenCL by exposing extensions. they might get adopted into the core profile
356.
by foxhill 12y ago
at the kernel level, fortran isn't really that different to C. it has a power operator. there is no for loop. you certainly will not be reading from files inside a kernel, so.. why do you want to use fortran? i can't tell you whom
357.
by foxhill 12y ago
> OpenGL 5 will happen in the next two years (and GLES 4) and it will take the wind out of all these alt-language sails. 2 years? in two years time nvidia will have their next line of GPUs out (pascal), intel will have launched knights l
358.
by foxhill 12y ago
> The only place where this is likely true is for their Intel graphics drivers false. i cannot cite a source for this, but i know they have direct access to current nvidia and AMD driver source trees, which they themselves extend/mo
359.
by foxhill 12y ago
> There's a limit to how much performance a general purpose graphics abstraction can enable. that remains to be seen, especially given how similar all the GPUs are today. if it weren't for the longs peak shambles, we might not
360.
by foxhill 12y ago
> But yes, OpenCL with its basic C API is way behind what CUDA offers in terms of language support. strange, language support is really the only thing that CUDA doesn't have over OpenCL. there are C++ (and python, Java, various othe

More ›