Why don't I read the articles? - page 4

New comment
 
GKS:

Really cool articles with great joy to read. I would like to continue the started topic of hardware-accelerated optimization.

1 Especially because C++AMP is coming, OpenGL topic is not fully disclosed and is very interesting for acceleration of calculations in Metatrader.

2 I'd love to read an article on "Accelerating cloud network calculations using a graphics card".

Can we expect something similar from you in the near future ? The topic is very interesting and fascinating.

Thank you. I will have to try and understand cloud calculations and their specifics...

2 Yura: Let me keep a secret (but don't tell anyone, OK?): When writing the second article, I was little more familiar with OpenCL than anyone else. The specifics of linking to hardware were being assimilated in parallel with writing the article. Was it really that hard to explain?

 
DmitriyN:
I'm not going to count your money, believe me. You haven't answered the question. Does neuro work in your talented hands or not? Yes or no? In real life.
My answer is more universal: everything has worked, works and will work if used in the area for which the tool is intended. But sometimes you can extend the field of application with skilled modifications. This is why I no longer use standard algorithms of neural networks, but their modified counterparts, customized for specific tasks.
 

I have some theory in the form of a research paper. I would like to write it myself, the only thing is that I don't know the programming language. I can write a pure theory, but no one will be interested in it without a code listing. I want someone to help me.

I have given my opinion about joint writing of articles, I will not repeat it.

If anyone is interested, all the details through the profile.

 
Mathemat:

Thank you. I'll have to try to understand cloud computing and its specifics...

2 Yura: Let me keep a secret (but don't tell anyone, Ok?): When writing the second article, I was little more familiar with OpenCL than anyone else. The specifics of linking to hardware were being assimilated in parallel with writing the article. Was it really that hard to explain?

Everything is much simpler. I have OpenCL on CPU only. I don't see any performance gain; spending money on hardware and time on binding to it makes no sense. When the need arises, i.e. when I have an application task to hand and it cannot be solved without OpenCL, I will definitely get to it. But for the sake of speed of drawing fractals, to measure with someone or something, it's not my domain.
 
GKS:

I have some theory in the form of a research paper. I would like to write it myself, the only thing is that I don't know the programming language. I can write a pure theory, but no one will be interested in it without a code listing. I want someone to help me.

I have given my opinion about joint writing of articles, I will not repeat it.

If anyone is interested, all the details through the profile.

Science is different, so at least specify in what particular area to find like-minded people. I, for example, about most scientific fields know only from the media.
 
Mathemat:

What is missing in them, is the level of presentation too high, etc.

Imho, the level of pitch, among other things, sets the level of your interlocutor. Is there a point in lowering it?

I myself expected there would be more interest -- especially with the interest in the OpenCL thread. But there was nothing from anyone on the subject.

The topic has been hyped up by a few, literally on the fingers of one hand, people. There is no widespread interest. There is interest from the MCs, because if OpenCL is going to be in the cloud -- it's... hard to overestimate, it's essentially adapting the cloud to any computation.

I, personally, have a very specific application task for OpenCL :) . But all in good time.

 
Reshetov:
Science is not the same as science, so at least specify in which particular field so that there are like-minded people. I, for example, know about most scientific fields only from the media.
"Hardware acceleration using GPUs" in general terms.
 
GKS:
"GPU hardware acceleration" in general terms.
GPU hardware acceleration without a single line of code? I think it's unrealistic.
 
GKS:
"GPU hardware acceleration" in general terms.
Well, the beast takes the catcher. Allow me to introduce the author of two articles in your field of science: Mathemat
 
TheXpert:
Hardware acceleration using GPU without a single line of code? I think it's unrealistic.

Who knows? Maybe he's figured out how to race these bees a little differently and the performance will increase to the limit?

1234567891011...18
New comment