palish

I've been a happy ViEmu customer for... Hmm... Three years? Maybe a little longer. Anyway, it's just perfect.

After working for them over the summer of my Junior year, it seemed like a good idea to continue working for them rather than attend my Senior year. Thankfully, I was correct in that assessment. Otherwise I would have wound up with no career and no diploma... :)

By rigorously separating the concerns into a clean object model, LMAX achieved a level of performance which might correctly be labeled "miraculous". http://screencast.com/t/g67kFj8nRue

It's written in pure Java. Amazingly, I haven't found anything that's achieved better performance thus far. It's very elegant.

  At 09:59 PM 1/27/2008, you wrote:
  >
  >Hey John,
  >
  >How does someone like yourself become a better engineer?  I'm
  >nineteen, and I dropped out of high school during my senior year to
  >work in game development.  Your work is inspiring on many levels.
  >Were you simply born with the ability to learn very quickly, or was
  >it the result of hard work?  I'd like to be the best engineer I can,
  >but I'm not sure how to keep improving.  What are your thoughts?

  Using your time effectively is very important, and there is often a 
  non-linear relationship between the amount of time you can stay 
  focused and the amount that you can learn or accomplish.  It is often 
  possible to get more done in a highly focused 12 hour stretch than in 
  a normal 40 hour work week that is interspersed with email, chat, and 
  other distractions.  Someone that can be completely obsessive about 
  something does have an advantage, but the same questions about focus 
  apply for any amount of time you choose to devote to an 
  undertaking.  Most people work at only a fraction of their potential.

  John Carmack

When I someday succeed in my life's work (to build a game studio), I'll owe it entirely to him.

On fractals -- I wish it had a clear definition. It's too subjective. For example http://en.wikipedia.org/wiki/Fractal. By the popular definition, it seems fine for non-linear growth in the number of shapes vs the level. (Though not especially interesting.) But I agree more with your interpretation.

It is obvious how to extend the drawing "to infinity".

I agree that some of them aren't (no recursion to infinity) but others are.

Sorry; this is a tangent. I'll blog about it or something.

As a game developer, I don't want my goal (to make fun games) to be confused with Zynga's agenda (to extract money).

Really? Awesome. How?

That is a beautifully accurate description of modern software development. I hope you get quoted.

Is there any way you could give your first-reaction answer for "what is the worst thing about working at Google?"

Let's say you have some floating point numbers, like...

  float A = 42.90F;
  float B = 00.01F;
  float C = /* ... */;
  float D = /* ... */;

In fact, if you do (A + B + B + B ...) very much, then you will wind up exacerbating the problem --- it can be a source of bugs if you repeatedly increment a counter by small floating point values. For example:

  const int Iterations = 10;
  for ( int i = 0; i < Iterations; ++i )
    A += B;
  printf( "%2.4f\n", A );

Astute readers probably would note that 00.01F can't be exactly represented in base 2, and wonder whether something like 00.0125F would still suffer from this problem. Alas, yes:

  A = 42.9000F 
  B = 00.0125F

  8    iterations ==  43.0000 
  80   iterations ==  43.9001
  800  iterations ==  52.9006
  8000 iterations == 142.8820  /* we would expect 142.9 */

  float A1 = 42.9000F;
  float B1 = 00.0125F;

  float A2 = 42.9000F;
  float B2 = 00.0125F;

  const int Iterations = 4000;
  for ( int i = 0; i < Iterations; ++i )
  {
    A1 += B1;
    A2 += B2;
  }
  float A = A1 + A2 - 42.9000F;
  printf( "%2.4f\n", A );

You may be tempted to try various things to get around this fact... for example, what about counting from zero rather than from 42.9000? (Answer: the error can be even worse, which is a topic worthy of its own separate post) ... but you will quickly discover that this is a fundamental issue. The order of operations matters.

Most people, by definition, are closer to average or dumb. This war unfortunately requires convincing large numbers that SOPA is a really, really bad idea.

With Dropbox, 100GB would cost me $199/year.

I've been with Dropbox forever (I have account #315) but man, it is tempting to snag 80GB for the price of half a tank of gas.

Convenience is obviously the primary loss. But I'm a nerd. It would likely be easy to figure out some way of auto-syncing my whole music folder.

My other primary usage for Dropbox (and incidentally the motivation for its inception) is as a way to transfer files to other people. There's nothing like being able to host your own .html file within your own dropbox with literally zero effort. That said, a hybrid of the two seems best -- ridiculously cheap bulk storage, combined with the convenience of Dropbox when needed.

EDIT: I just saw your edit. That's what I'm doing --- shm_open(), ftruncate(), then mmap(). A little annoying that it doesn't automatically kill the shared memory when the program terminates, but still...

Would someone please help me out and point me at a way of efficiently exposing a read-only block of memory to Python?

EDIT: Looks like "MemoryView" introduced in 2.7 is the answer.

Would you post more info about previous attempts at this? I was about to create precisely that. Literally today.

But if it's fundamentally unworkable...

EDIT: Eh, apologies if this sounded like chastising -- I didn't mean to. As a developer who's been trapped in "Windows-mindset" for many years, I wanted to try to inspire other Windows devs to try to use *nix-based solutions even if their only option is Windows development. Cygwin is in a very good spot right now -- it's achieved so much acceptance that even the most hardened institutions now allow it to be installed.

  grep -R properites .

This isn't true at all.

At what point are the equations "accurate"? What margin of error is acceptable?

We still have a long way to go before we can claim accurate lighting equations.

Did you find it difficult?

> You could parachute him into an island full of cannibals and come back in 5 years and he'd be the king. If you're Sam Altman, you don't have to be profitable to convey to investors that you'll succeed with or without them. (He wasn't, and he did.)

More than that, I can't seem to inspire confidence in people. I don't know the cause, but it's proving lethal to my ambitions. As far as I can tell, it's the same situation Jobs found himself in:

> "You know, I’ve got a plan that could rescue Apple. I can’t say any more than that it’s the perfect product and the perfect strategy for Apple. But nobody there will listen to me.” [1995]

I don't know. I'm just enjoying life and pursuing my research. But it feels unsettling coasting like this. I have Carmack's technical skill; if only I could rally like-minded capable people who wanted to build the next Id or Epic, then we'd stand a fair shot at success, at influencing our industry and the world. We wouldn't be just another puppet for publishers. We'd have a shot at making an impact: the next Half-Life or Final Fantasy or World of Warcraft. The formula is good people + shared goal + time + rapid iteration + don't get distracted. It works, it's proven, and I've seen it happen firsthand at S2. We can achieve that same success.

But why should anyone take me seriously? I probably wouldn't, if I weren't me.

- a deep knowledge of the details. I share Robert Morris' passion for understanding precisely what is going on at every level. My insight into the low-level details of next-gen CPUs came from reading the Intel architecture manuals several times over.

- an ability to achieve high-quality results within a very short period of time, regardless of the project or goal. If you show me a problem, I'll show you a solution by the end of the day, probably with a rough prototype. By a week, I'll have discovered which solution/direction is the best long-term strategy, and will have a working production version 1.

- I have good judgement. This turns out to be rare.

Specific technologies:

- a master of C/C++. I've been coding since age 12, and I'm now 23, so that's easily 30,000+ hours I've spent practicing. I've been writing C/C++ professionally daily since my first gamedev job in 2007. Python is my language of choice; Lua a close second, primarily when performance is of concern (such as in gamedev). But a language is merely a means to an end, and I work with all of them.

- several years of hobby experience writing Rails websites/Javascript. I write Lisp interpreters for fun. I want to write an Arc compiler someday.

- extensive sysadmin knowledge, both Unix and Windows. FreeBSD is my preferred OS for server-side deployment.

- master of realtime 3D graphics; both OpenGL and D3D. Given time and freedom, I can build you a renderer of the finest quality (equal or higher than Rage).

- artistic talent and taste. Several years of 3D Studio Max and Maya modeling and animation experience. I build next-gen character art (high-poly sculpt in ZBrush, then decimate / build low-poly shell in 3D Coat; project detail; touch up textures via Photoshop -> rig in 3DSMax -> animate -> import into game).

Projects:

- in highschool, wrote a game engine roughly equivalent to Quake.

- several game engines / game ideas.

- a full game editor, using C#/.NET WinForms.

- next-gen realtime graphics experiments, including realtime skin rendering (the same technique was recently shown by Epic in their "Good Samaritan" demo).

- several Rails websites. Made a system for schools to administer tests, with the intent of surpassing Blackboard. (JavaScript-based; this was back before "Ajax" was a fancy buzzword.) Tried to apply to YC with it. pg gave valuable advice (selling to schools is like selling to government; quality doesn't always win). Was invited to Hacker School though, which was fun (seeing Anybots' Dexter robot in person walk around on two legs was just awesome).

Experience:

- Simutronics, HeroEngine. Graphics programmer (3 years)

- S2 Games, Heroes of Newerth. Engine programmer (1 year)

My thirst for knowledge is unquenchable in five lifetimes, and I strive for beauty and elegance in everything I pursue.

If anyone feels these claims seem too bold to be true, I'd love the opportunity to work with you to prove myself while serving your interests.

Maybe it was true, but by telling me that, she accidentally laid the foundation of the rest of my life: I started distancing myself from my classmates, and eventually my peers.

The result? I now have a skillset roughly equivalent to Carmack's, but no chance to capitalize on it. No one believes nor cares. (http://dl.dropbox.com/u/315/portfolio/pre-textured.jpg ; http://dl.dropbox.com/u/315/portfolio/face.png)

By setting myself apart, I likely ruined myself without even realizing it. It turns out that my skills don't matter. A Carmack without a company is no Carmack at all. And now I've no ability with people, as he does.