dragandj

52 ms·

541.
by dragandj 8y ago
I am living in Serbia and, although you are right about Mehtmet-Pasha Sokolovic (Sokollu), it was certainly not the norm. It is difficult to judge different times from the modern perspective, but the actual term that is used for the period
542.
by dragandj 8y ago
Sure, but they were not mercenaries. They did not enroll, but were taken from their families in occupied lands as small children, raised and educated as warriors, and the only career option they had, and family they knew of is the army.
543.
by dragandj 8y ago
Janissaries were slave-warriors rather than mercenaries, though.
544.
by dragandj 8y ago
What have they received in return? Real United States Dollars or ETH, BTC, or, even worse, some shitcoins?
545.
by dragandj 9y ago
The headquarters are in Germany, but tiny Slovakia, for example, has huge car production facilities.
546.
by dragandj 9y ago
Not only that, but writers around the world still use text to express their thoughts. And, 5000 years ago, they started with drawings. But, smart as they are, they realized it sucks, and that for anything beyond simple concepts, text is d
547.
by dragandj 9y ago
Yet most of the apps in the AppStore are absolute crap. Add to that the useless search. OTOH, many students or open source developers may want to try offering free or cheap apps of ok quality, but they are disincentivized by the cost.
548.
Out of the Box Support for TensorFlow and PyTorch for Clojure in Neanderthal(dragan.rocks)
3 points by dragandj 9y ago|0 comments
549.
by dragandj 9y ago
It's only optional though. You can use (list 1 2 3), (vector 1 2 3), (hash-map 1 2 3 4), and (fn [x] (inc x)).
550.
by dragandj 9y ago
No one is forcing anyone to upgrade. What worked on Java 8 3 years ago, will work on Java 8 in 17 years. If your team have extra requirements (audit, certification etc.), it seems it sells enterprise software. All those things cost money, b
551.
by dragandj 9y ago
There is no OpenCL in the story there. AMD created HIP as a direct copy of CUDA. It then gets compiled to work on AMD hardware, or can be translated to CUDA. AMD also supports OpenCL (which I prefer to both CUDA and HIP), but it's no
552.
by dragandj 9y ago
OTOH, even Intel CPUs support OpenCL, so practically there aren't such many cases where you can't work with a "real" OpenCL implementation.
553.
by dragandj 9y ago
That prediction could be right even if that hypothesis was wrong. Hypothesis 2: City X different than Troy existed is equally well made more plausible by digging a city up in Asia Minor.
554.
by dragandj 9y ago
ClojureCL is driver-agnostic. As long as the OpenCL platform implementation (POCL) is OpenCL-compliant, ClojureCL should work (but I have not tried). Regarding profiling/benchmarking, you can measure wall-clock time very precisely with
555.
by dragandj 9y ago
Add to that that TensorFlow was practically a latecomer, not the first to market.
556.
by dragandj 9y ago
Library source code: https://github.com/uncomplicate/clojurecuda Documentation: https://clojurecuda.uncomplicate.org OpenCL alternative: https://github.com/uncomplicate/clojurecl
557.
Interactive GPU Programming, Part 3: CUDA Context Shenanigans(dragan.rocks)
8 points by dragandj 9y ago|5 comments
558.
by dragandj 9y ago
Seems quite logical to me: completely open borders make H1B obsolete.
559.
by dragandj 9y ago
There is a Bande Desinee with a plot just like that: https://www.bdgest.com/preview-1673-BD-nous-les-morts-les-en...
560.
by dragandj 9y ago
You can, and yes, it is.
561.
by dragandj 9y ago
His logical reasoning skills seemed to be quite all right for Google when he passed their (challenging) hiring process. How the hell he managed to lose that while working there? As for his writing ability, it may not be at the level of Mark
562.
by dragandj 9y ago
Except that no one is going to sell you anything for such toilet paper money, banks do not offer any credits, and everything is calculated in USD or EUR that become really, really scarce. Basically, all "official" commerce that us
563.
by dragandj 9y ago
Yes. OpenCL supports that. However, I am using it for computation and Machine Learning algorithms, so I haven't tried that myself and can't give you detailed pointers.
564.
by dragandj 9y ago
For equivalent code, you will not experience any noticeable difference compared to C. It is nothing like Aparapi.
565.
by dragandj 9y ago
The source repository for the lib(s): https://github.com/uncomplicate/clojurecl https://github.com/uncomplicate Part 1: Hello CUDA at http://dragan.rocks/articles/18/Interacti
566.
by dragandj 9y ago
Are you sure that this particular translation is "nonsense"? To me, it seems almost perfect. The only mistake in the French version is the gender of the teacher, which is a mistake that many humans can and do make, in addition to
567.
by dragandj 9y ago
We are talking about time-series databases here. I guess they are more optimized for the specific workload than a generic RDBMS.
568.
by dragandj 9y ago
The language and basic platform is not a problem. OpenCL was and is OK. However, the libraries are far and between. CUDA offers cuBLAS, cuFFT, cuDNN, cuSolve, etc. For OpenCL, even the decent BLAS library (CLBlast) had to be written by a
569.
by dragandj 9y ago
FWIW, all my libraries work with both CUDA and OpenCL. While I agree with your sentiment, unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by
570.
by dragandj 9y ago
Note the word Interactive . That's not something you'll get in the plain C tutorial.

More ›