🔒 Locked

Game algorithm contest!

Started by uppcercut Jan 21, 2004 at 9:12 PM 84 replies 5k views

Original Post

Jan 21, 2004 at 9:12 PM

The Omega Architects are having are contest to see how can come up with the best new Omega Architecture. Prizes include swag/code/membership to the Architects/etc. So if you want to enter, or want to learn how to write code so good it actually increases the speed of your cpu, head over to.. to this place

Sneftel

296

Sneftel

Jan 21, 2004 at 9:20 PM

Man, that has to be the most incredible programming site I''ve ever seen! I''ve already begun implementing the Omega rendering architecture into my game engine, and it''s already starting to speed up my AI code!

"Sneftel is correct, if rather vulgar." --Flarelocke

pinacolada

112

pinacolada

Jan 21, 2004 at 9:39 PM

holy crap! those papers are amazing! You guys have really opened my eyes to some out-of-the-box thinking!

Actually, I was fooling around with the TOZA algorithm.. and I found out something really neat! It seems like the TOZA compression algorithm is not limited by the sparseness/noisiness of the data, because I found out that you can run it TWICE and get the same compression ratio!! That''s a 65025-fold difference in file size!!! I tried to see if you could run it 3 times, but something went wrong with my computer.

joelmartinez

Jan 22, 2004 at 12:52 AM

ok, I''m obviously missing something at a very fundamental level ... of what papers does pinacolada speak? you mean the description of the toza algorithm? that''s hardly a paper.

And while I''m on the subject of the TOZA algo, again, I must be missing something, how is that compression? did I understand it correctly in saying that for every bit you will have a tza chunk of all the values that it''s not? wouldn''t that inflate the data instead of compressing it?

Joel Martinez
http://www.codecube.net/

uppcercut

Author

uppcercut

Jan 22, 2004 at 1:53 AM

i believe pinacolada stated our mission and purpose very well, thinking "outside the box". we confine ourself with definitions, such as compression, effeciency, etc. we help broaden horizons with new thought patterns relating to computer science and game related algorithms.

for example, toza does accomplish compression. very well in fact. if we define compression as "making something smaller", then we can infer that after toza compressing something, the free space on the hard drive gets smaller. so we fit within our nomenclature. i hope this helps, i would hate to be misunderstood. oh, and Sneftel, we''re working on an omega A* algo, that would benefit your already benefiting AI code greatly i think.

[omega]Uppercut

Fruny

203

Fruny

Jan 22, 2004 at 2:02 AM

quote:
Original post by uppcercut
oh, and Sneftel, we''re working on an omega A* algo, that would benefit your already benefiting AI code greatly i think.

Is it based on the drunkard''s walk?

“Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.” — Brian W. Kernighan (C programming language co-inventor)

joelmartinez

Jan 22, 2004 at 2:23 AM

quote:
Original post by uppcercut
i believe pinacolada stated our mission and purpose very well, thinking "outside the box". we confine ourself with definitions, such as compression, effeciency, etc. we help broaden horizons with new thought patterns relating to computer science and game related algorithms.

true true true, I was in no way trying to put down the concept behind this endeavor ... I love the concept of breaking "tradition" to find other (and potentially better) ways of doing things.

quote:

for example, toza does accomplish compression. very well in fact. if we define compression as "making something smaller", then we can infer that after toza compressing something, the free space on the hard drive gets smaller. so we fit within our nomenclature.

aaaaahhhhh ... that truly does answer my question. In fact, I read the above paragraph several times before my brain realized (accepted?) the fact that you weren''t trying to use "compresion" in the traditional sense.

keep on keepin on

Joel Martinez
http://www.codecube.net/

uppcercut

Author

uppcercut

Jan 22, 2004 at 7:01 PM

to joelmartinez, it brings us all great joy that another accolyte has realized the intrinsic workings of toza.

to Fruny, although the drunkard''s walk has many omega qualities, the algo were we specifically designing is slightly different. one version works more or less exactly as the normal A*, though instead of keeping track of nodes we have visited, and those that are to be considered (very time consuming), we just allocate nodes then search through the entire contents of RAM looking for them later, this way we might bump into other relevant data we might want to process.

to pinacolada, the toza ratios are amazing, we have a ''super'' toza that works with 64 bits(a long long), instead of a measily 8 bits, though i have not been able to test it yet, as a toza compressed byte works out to be several thousand terabytes. one day.

[omega]Uppercut

PlayGGY

Jan 22, 2004 at 7:12 PM

HAHAHAHAHA!!!!!!!!!!!!!!!!!!

[edited by - PlayGGY on January 22, 2004 8:13:55 PM]

[edited by - PlayGGY on January 22, 2004 8:15:57 PM]

[edited by - PlayGGY on January 22, 2004 8:17:55 PM]

cowsarenotevil

478

cowsarenotevil

Jan 22, 2004 at 7:21 PM

By the way, you forgot to photoshop out the total CPU usage at the bottom for your TORA algorithm.

uppcercut

Author

uppcercut

Jan 22, 2004 at 11:16 PM

holy cow, cowsarenotevil, i cant believe i let that one slip by me. and im curious why PlayGGY edited his ha''s 3 times. my guess is that it started out as haha, then he realized that for emphasis hahahahaaa was edited in. then after yet another glance, an epiphany graced him and the final HHAHAAHAHA that we see was formed.

[omega]Uppercut

owl

267

owl

Jan 22, 2004 at 11:30 PM

Wow, I wonder how many weight I can loose with this system... The woman of the picture looks very happy.

SarsDP

Jan 22, 2004 at 11:47 PM

I don''t think ill ever understand how that program brought a file size from 8meg - 2.2gig and windows thought there was more room on the drive

IT DOESN''T MAKE SENSE

okonomiyaki

Jan 23, 2004 at 9:35 AM

I don't freaking get it. You guys are scams (of some sort). Every idea you have either makes things much slower or take up more memory. Are you trying to inhibit computer science development?

edit: misspelled 'things'

[edited by - okonomiyaki on January 23, 2004 10:35:50 AM]

Cosmic One

Jan 23, 2004 at 9:54 AM

I''m afraid I''m still not understanding TOZA architecture... If anybody can explain to me how writing 255 bytes of information rather than 1 byte of information is space saving and efficient then I''d love to hear it. Right now it seems as though their TOZA compression is expanding the space required at a ratio of 255 to 1, which is just ludicrous.

Anyways, dazed and confused (as usual)

Cosmic One

Drath

Jan 23, 2004 at 10:12 AM

I thought the double memcpy concept was rather interesting in the toga article. Dunno why I hadn''t thought of that before...

PurpleAmethyst

Jan 23, 2004 at 10:13 AM

This is complete bollox. a guru wouldn't have such a shit site either

[edited by - pkelly83 on January 23, 2004 11:17:40 AM]

pan narrans

114

pan narrans

Jan 23, 2004 at 10:16 AM

*tee-hee*

pan narrans | My Website | Study + Hard Work + Loud Profanity = Good Code

OrangyTang

333

OrangyTang

Jan 23, 2004 at 10:17 AM

Are you perhaps related to the Timecube guy?

rypyr

Jan 23, 2004 at 10:24 AM

For those of you who don''t get it...

WHOOOOSHHHH!!!!

Hear that sound? It''s something going over your head...

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.