New resource for sizecoding

category: code [glöplog]
I am curious what you mean by "update it to 2016" though

erm, exactly what i said in the sentence: update the article about bytetro coding that already exists in in4k to 2016 based on info that you been putting into sizecoding.org.

damn, is my written english that bad?! :D

i don't see sizecoding.org as a any sort of slight against in4k whatsoever. :)
added on the 2016-08-22 20:15:42 by psenough psenough
Yes, I'm damn sure that "it" referred to "in4k" in ps' quote.

FWIW I wanted to add articles and tutorials on www.256b.com, but I didn't feel entitled to enroll more people and turn the site into a mix between Pouet.net and Sizecoding.org as the site wasn't really mine; Insomniac/Neon gave me ftp+db access because he didn't have the time to maintain it anymore.
added on the 2016-08-22 21:19:16 by p01 p01
Thank you folks for in4k.github.io and sizecoding.org \o/
added on the 2016-08-22 21:24:54 by p01 p01
update the article about bytetro coding that already exists in in4k to 2016 based on info that you been putting into sizecoding.org

Oh, okay, I misunderstood. Thanks for clarifying :-)

damn, is my written english that bad?! :D

No, my reading comprehension is that bad ;)
added on the 2016-08-22 21:46:21 by trixter trixter
No, my reading comprehension is that bad ;)

That's ok, I also thought "it" referred to "info that you have been putting into sizecoding.org". The word usually refers to the last subject written, so when we read "it" we popped "sizecoding.org" off the stack. :) There should be a word for "it, I mean the next-to-last thing I said".
added on the 2016-08-22 22:56:47 by phoenix phoenix
BB Image great initiative! BB Image
added on the 2016-08-23 16:12:16 by stfsux stfsux
@phoenix: Try German, or any other other language with gendered nouns :-) Having items hashed into three buckets (instead of the single universal “it” bucket) makes collisions much less likely.
added on the 2016-08-24 00:33:29 by Sesse Sesse
We felt that focusing sizecoding.org to x86+256b and smaller gave us the opportunity to hone in on a very specific area. In our eyes, it really is a different discipline than 4k coding (the most obvious differences being no opportunity for compression, no compilers, no shaders or GPU acceleration). If you're wondering why we stood it up instead of contributing to in4k, that's why. It was because we felt there wasn't that much overlap, and was definitely not meant as a slight or anything.

I do feel the same way. I think at this point its more a strength to have a direct separation between the two fields.

I also feel that an important difference are their operating systems. In DOS, <1k intro coding dont need consider Portable Executable or a MZ executable programs. That is have no need for headers. There is no dynamic link library hell, hashing and/or ordinal lookups etc. Win32/win64 APIs are useless in DOS, not all x86 interrupts can be called in Windows. On the software side of things both Operating Systems are mostly obscured from eachother. This is one of a few things that contributes to the reason why I think its ideal to not mix them up on a wiki-page or whatever. For some programmers focusing on tinytro coding below 1k for DOS specifically dont need all the resources windows tinytro programming need. (There might ofcourse be cases where the machinecode and/or assembly code have a few tricks both platforms can use - but then this has to be specified somewhere and not be related to any OSes on any page) Some whole branches (atleast software-related) are removed in DOS tinytro-coding (specifically for files like COM that are loaded at a pre-set address 0100h). Just running the machinecode and most if not all is programmed in assembly language. No inline-assembly, like phoenix said, compression is something thats just not possible to do in these tiny programs. Atleast not with current compression-technology.

Im sure you allready knew all of this, but a resource site should imho draw a separation between tinytros for DOS and stuff in Windows thats been compressed with Crinkler, or Windows intros in general. They are just too far apart from eachother - atleast in the present moment as I see it.

For me and maybe for others as well coding tinytros for dos is alot like weaving code and data, or more like a puzzle game, because the different sizes of the puzzle pieces does not always fit the required size of the whole image. This might sound a bit strange, but the way to do "compression" is to actually do it manually, that is by finding other means by other opcodes and operands (some people like to call it tricks) to make the code smaller. This is probably a more direct manual way to compress code instead of using a compressor.

I dont want to interfere in any discussion about where to put <1k intro coding material on the web, for me it really does not matter, as long as its separated im fine with it.
And.... its really really late here, so Zzzzz
added on the 2016-08-24 03:59:57 by rudi rudi
trixter: i suggest you ensure that loading sizecoding.org behaves the same as loading www.sizecoding.org
added on the 2016-08-24 04:55:58 by psenough psenough
I'm hosting multiple virtual domains on a single IP address via CNAME aliases. If you are an apache mod_rewrite guru, I welcome the help.
added on the 2016-08-24 20:48:50 by trixter trixter

i would believe the common way to solve it would be to have the dns settings on your provider as:
A @ youripaddress
CNAME www @

and the apache config file with:
<VirtualHost youripaddress:80>
ServerName www.sizecoding.org
DocumentRoot /www/localpathto/sizecoding.org

<VirtualHost youridaddress:80>
ServerName sizecoding.org
Redirect permanent / http://www.sizecoding.org/

don't blame me if everything explodes though :D

i keep telling people i'm not a sysadmin, don't know why people keep asking me to figure these things at work :p
added on the 2016-08-24 23:52:36 by psenough psenough
Neat resource, looked through it a bit, but I doubt I'll ever produce a 256-sized prod. Nice work. :)

@ psenough, ARGH! I really dislike sites these days that redirect to their "www" version, because why? What's the point? "www" is a subdomain made special for historical reasons only. It's far cleaner to omit it unless you need it to identify a URL on physical goods that for brevity you've removed the "http[s]://" from.
(and yes, I know, Pouet does it too... but that doesn't mean I have to like it, or think it's less silly. :P)
Putting "www" at the start makes the url look nice and symmetrical.
added on the 2016-08-25 00:55:39 by Gargaj Gargaj
well, you can always do it the other way around if you prefer :p
have the main without the www, and the one with www pointing to the bare domain
i just like that both end up on the same place because it's easy to make the mistake of not knowing which one is the main.
added on the 2016-08-25 01:37:49 by psenough psenough
@psenough. Why does my iPad need a dress?
added on the 2016-08-25 06:02:57 by kuiash kuiash
added on the 2016-08-25 09:52:00 by xTr1m xTr1m
"www" is a subdomain made special for historical reasons only.

Actually, "www" is a hostname, not a subdomain. Entering just a domain and expecting it to go somewhere is considered bad form (despite the fact that many people do it and expect it to work).

I will re-investigate rewrite rules for www.sizecoding.org. In the meantime, y'all should contribute some tips and tricks to it (the ones you don't mind giving up, of course :)
added on the 2016-08-26 03:05:43 by trixter trixter
don't blame me if everything explodes though :D

Caused an infinite loop, but I think I've fixed it with a ServerAlias directive. Test and let me know if otherwise.
added on the 2016-08-26 03:18:28 by trixter trixter
Works fine, I get a 301 redirect to the www variant.
Not sure why I didn't think about this before, but 301/302 redirects can be done through html meta tags as well... not exactly a robust solution though. :)
added on the 2016-08-26 11:49:53 by VileR VileR
Well, if you already set up a subdomain to serve a HTML file with a redirect, then you could just instead make it send a HTTP header too. ;)
yeah, both urls seem to be working :)
added on the 2016-08-26 15:18:32 by psenough psenough
Thanks the guys who made this. This is a must have for anybody who want to do some intro democoding.
added on the 2016-11-29 19:12:07 by Tigrou Tigrou