Genesis (take #1)

The NonStampCollector has the always found the ability hide some hilarious concepts and intriguing commentary in his videos. Below you will find a few of them, helping to uncover some of the more hilarious aspects of the genesis stories, and showing that even Australian people has the ability to come up with better ideas than god did.

0 Comments

Down and to the Right

Today in class we discussed the aspects about software and hardware programming that i hate to admit are important, and for that i am sad. The good news is that this lecture is a one off and wont, according to my professor, ever happen again. Here are the common pitfalls of computer architecture:

  1. ignoring cost when developing software or hardware
  2. ignoring what your customers are looking for
  3. ignoring trends in technology
  4. focusing on Local optimizations
  5. over-optimizing and creating complexity
  6. faulty performance measures

Unfortunately, I cant really argue against any of these items. There is a certain benefit to the “Well, It Works!” mentality – at least they reach production from time to time, usually ahead of time. It is with great regret that i have to admit so much. It is a real shame.

Optimizing is still my drive, and doing things the best way is always going to be something i can drive for, i just have to accept that the best way is subjective to the goals of whatever project I am part of. Lucky for me, game development requires optimization in spades =D

At least from here on out VHDL is going to be taking the reigns and everything we code is going to be focused on speed.

We looked at a graph, plotting the cost/performance versus performance graph. This graph plotted three technologies (A, B, C), each overlapping and shaped like a flat bottomed U and broken into three distinct parts.

  1. Under-Utilization on the left, with quickly decreasing cost for any performance gain
  2. Optimal-Utilization at the bottom, where cost and benefit are matched
  3. Over-utilization on the right, with sharply increasing cost for any performance boost.

Cost can come from anywhere, including but not limited to books, effort, hardware or tools. A laundry list of anything you can imagine that would impact your ability to improve a product. My focus has always been on optimization and understanding the underlying technology can only help. Writing my first CPP & ASM application was a lot of fun, but there is an added layer of complication in that it was MASM ASM and blah blah blah – all kinds of crazy platform dependencies. Hopefully this class will make me happy, i want to over-optimize the shit out of this mother board I got.

0 Comments

Treatise on Morality

0 Comments

Homeopathic Remedies brought to the internets

[05:15] <Oelund> Forum is down?
[05:15] <@maggie> yes
[05:15] <@maggie> host is down
[05:18] <Oelund> Lets try some homeopathic stuff to get it back online.
[05:18] <Oelund> Lets try putting some treebark in the harddrive… It can’t hurt, it’s natural.
[05:20] <@maggie> that’s not homeopathy
[05:20] <@maggie> it’s herbalism
[05:21] <Oelund> you got me there.
[05:21] <@maggie> homeopathy would be putting a drop of internets in a cup of water and haking it and then a drop of that into more water and then on like that until there are no molecules of internets left
[05:21] <@maggie> *shaking
[05:21] <@maggie> and then declaring it super-strong medicine
[05:22] <@maggie> and then pouring it over the server
[05:24] <@maggie> science!
[05:24] <Oelund> what if that drop of mixture you take always happen to be 100% internet? it could happen.
[05:24] <Oelund> Changes might be a billion to one.
[05:25] <Oelund> but still, in my experience a billion to one changes crops out nine times out of ten.
[05:25] <@maggie> yes, it could happen, but the chances would be far far slimmer than a billion to one
[05:25] <@maggie> no, they crop up one time in a billion.  by definition.
[05:26] <Oelund> thats because you don’t factor in luck
[05:26] <@maggie> the chances of picking up 100% internet the first time might be a billion to one
[05:26] <@maggie> but the chances of picking up 100% internet the second time would be a billion * a billion to one
[05:27] <Oelund> no, you would be able to do it nine times in a row.
[05:27] <Oelund> see statement above
[05:28] <@maggie> statistic chances don’t change just because you’re the unluckiest fucker on the planet
[05:28] <@maggie> the stats stay the same and you just beat them all
[05:30] <Oelund> still better than playing the lottery.
[05:31] <@maggie> lol
[05:31] <@maggie> well anyway our homeopathic conversation fixed the forums
[05:32] <Oelund> thats because we watered it up with soo much bullshit that there is not a single molecules of homeopathic conversation left.

0 Comments

FAS 1.0e Incoming

0 Comments

HL2 Ragdoll fix

Source programming is hard enough without the abundance of crazy bugs already existing in the code that exist without your having ever done anything. One such bug is found when you try to join a game already in progress, and its a real piece of work. Your client spawns ~ 40-400 ragdolls before even opening up the MOTD notice, leading to a full stop of all possible interaction. Its a real irritating situation and it was bringing Firearms: Source to a halt server joining. .

Was.

Continue Reading »

0 Comments