News for hackers. Stuff that matters.

ahcox

Re:Netflix was experiencing issues on Tuesday February 03, 2015 @08:47PM (#8994869)
Podcast where Adrian Cockcroft goes into the philosophy and technology behind scaling at Netflix: http://www.se-radio.net/2014/12/episode-216-adrian-cockcroft...
Re:3D GIFs Created with a Simple Visual Effect on Monday February 17, 2014 @09:34AM (#7252224)
They look particularly good expanded out to fill the page:

   http://hoog.li/g?g=http%3A%2F%2Fwww.viralnova.com%2F3d-images%2F&cimw=480

   http://hoog.li/g?g=http://www.mymodernmet.com/profiles/blogs/3d-gifs&cimw=320
Re:GPU Ray Tracing with Rayforce [Nvidia GTC 2013 Poster] on Wednesday March 27, 2013 @10:16PM (#5453070)
There is code for this project available: http://sourceforge.net/projects/rayforce/?source=directory
Re:GPU Ray Tracing with Rayforce [Nvidia GTC 2013 Poster] on Wednesday March 27, 2013 @10:01PM (#5453031)
They are using some sort of graph of connected volumes without hierarchy as their acceleration structure. This could be an interesting alternative to using a tree or grid.

I wish they:

1. Showed a visualisation of their data structure.

2. Gave a straight-up comparison against an alternative like Optix on the same hardware.

3. Characterised the possible construction algorithms to some extent, especially with respect to animated scenes.

It is just a poster, so it will be interesting to read the paper when it arrives.

Re:Hacker News London meetup was excellent on Friday October 19, 2012 @09:28AM (#4673402)
What I don't understand is how a room packed with 600+ people all noisily engaging in social interaction was any more manageable last night.

I found it almost overwhelmingly intense. I did stay to the end though.

Re:The Two-Thousand-Year-Old Computer (BBC iPlayer) on Monday May 14, 2012 @08:51PM (#3974232)
This is no Babbage-style mechanical computer. I found the BBC's title a distraction from the amazing revelation that ancient Greeks were building such complicated mechanisms.
Re:Show HN: my "My Latex resume" project on Monday April 02, 2012 @08:13AM (#3787654)
Perhaps a PDF with an embedded image per page is the way forward.
Re:Show HN: my "My Latex resume" project on Monday April 02, 2012 @08:11AM (#3787640)
I was at an interview recently and was disgusted to see my CV in the hands of the interviewers with the formatting totally messed up and the logo of the recruiters all over it. They had obviously cut and pasted into a word document even though I wouldn't give them this originally.
Re:Show HN: my "My Latex resume" project on Sunday April 01, 2012 @12:23PM (#3784102)
The CurVe class supports multiple versions of a CV with its "flavors" concept: http://text-ex-machina.co.uk/blog/creating-curriculum-vit

I keep all versions of my CV side by side in directories checked in to subversion and copy the closest existing one to make a new version rather than trying to maintain multiple different flavours of CV at the same time.

Re:Show HN: my "My Latex resume" project on Sunday April 01, 2012 @12:07PM (#3784026)
This part of the CV looks like a cross between a covering letter and a profile/summary to me. It seems too long and it repeats information that is present elsewhere in the CV.

I would shorten it and use it to get your main message across to the person scanning it at high speed. Sorry to post my own latex-generated CV twice in one thread, but for ease of reference, this is what I mean: http://www.hoogli.com/Andrew_Cox_cv.pdf

My profile there splits into two parts:

   1. I am ...
   2. I want ...
So: "Strong software engineer with substantial OpenGL and OpenGL ES and general 3D graphics and GPU programming experience, a background in R&D, and currently developing for Android [1]. Looking for challenging short-term contracts that leverage these strengths[2]." That's it.
Re:Show HN: my "My Latex resume" project on Sunday April 01, 2012 @11:56AM (#3783988)
I like the CurVe latex class. It works for me: http://www.hoogli.com/Andrew_Cox_cv.pdf

It is worth going to the author's page about CurVe rather than relying on the ctan package. He has some nice customisations such as the thick horizontal rule that I borrowed, and gives his own CV as an example: http://www.lrde.epita.fr/~didier/software/latex.php#curve

Here is the archive: http://ctan.org/tex-archive/macros/latex/contrib/curve

Yes, you are right. In fact we don't have to infer this. The researchers state it directly in their abstract, "...a novel approach to utilize the CPU resource to facilitate the execution of GPGPU programs..."

Here it is: http://news.ncsu.edu/releases/wmszhougpucpu/