<?xml version="1.0" encoding="utf-8"?>
<feed version="0.3"
	xmlns="http://purl.org/atom/ns#"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xml:lang="en">
	<title>captain&apos;s log</title>
	<link rel="alternate" type="text/html" href="http://b-cage.net/log/" />
	<modified>2006-12-09T02:08:12Z</modified>
	<tagline>Dingens - Made in Germany</tagline>
	<id>tag:b-cage.net,2006:/log//1</id>
	<generator url="http://www.movabletype.org/" version="3.16">Movable Type</generator>
	<copyright>Copyright (c) 2006, b.gr</copyright>
	<author>
		<name>b.gr</name>
		
		<email>b.gr@b-cage.net</email>
	</author>

	<entry>
		<title>ACE arrived</title>
		<link rel="alternate" type="text/html" href="http://b-cage.net/log/archives/2006/ace-arrived/index.php" />
		<issued>2006-02-25T22:09:19Z</issued>
		<id>tag:b-cage.net,2006:/log//1.59</id>
		<created>2006-02-25T22:09:19Z</created>
		<modified>2006-12-09T02:08:12Z</modified>
		<summary type="text/plain"><![CDATA[
&hellip;about one month ago.
]]></summary>
		<dc:subject>Misc</dc:subject>
		<content type="text/html" mode="escaped" xml:lang="en" xml:base="http://b-cage.net/log/">
			<![CDATA[<p>
&hellip;about one month ago. I still have no clue why, though :] Anyway, here are a few pictures of The ACE Cube:
</p>
<div><div>
<script type="text/javascript" src="http://www.flickr.com/badge_code_v2.gne?count=1&display=random&size=t&layout=x&source=user_set&user=14848076%40N00&set=72057594070234160&context=in%2Fset-72057594070234160%2F"></script>
</div>
<a href="http://www.flickr.com">www.<strong style="color:#3993ff">flick<span style="color:#ff1c92">r</span></strong>.com</a>
</div>]]>
			
		</content>
	</entry>

	<entry>
		<title>New Versions</title>
		<link rel="alternate" type="text/html" href="http://b-cage.net/log/archives/2005/new-versions/index.php" />
		<issued>2005-11-30T05:31:42Z</issued>
		<id>tag:b-cage.net,2005:/log//1.58</id>
		<created>2005-11-30T05:31:42Z</created>
		<modified>2006-12-09T02:05:13Z</modified>
		<summary type="text/plain">Some thoughts on recent new releases of the applications I use often. And instead of concentrating on the goodness I am in the mood to point out the annoying parts.</summary>
		<dc:subject>Misc</dc:subject>
		<content type="text/html" mode="escaped" xml:lang="en" xml:base="http://b-cage.net/log/">
			<![CDATA[<h3><a href="http://www.winzip.com/" rel="external">WinZip 10.0 (6685)</a></h3>
<p>
	Updating WinZip always results in a reset to default options.
</p>

<h3><a href="http://www.25hoursaday.com/weblog/PermaLink.aspx?guid=4e452051-706b-4464-92b8-52221fca8077" rel="external">RSSBandit v1.3.0.38</a></h3>
<p>
	Eventually no custom build to enable my bron (Björns River Of News, not
	released yet) theme anymore :] The new notification icons look a bit
	funky, though: the working one seems to notify unread items, while the
	one used when unread items are available looks a bit weird. The notification
	icon status updates are still a bit off, too.
</p>

<h3><a href="http://weblogs.mozillazine.org/mitchell/archives/2005/11/firefox_15_better_than_ever_1.html" rel="external">Firefox 1.5</a></h3>
<p>
	<a class="l" href="/log/imgs/fx-1.png"><img src="/log/imgs/fx-1t.png" /></a>
	That was the first thing I saw.
</p>
<p>
	Twice. Good to be able to identify the extensions in question on a glance
	to contact the author.
</p>
<p>
	The next step was to check the <strong>JS Console</strong> to see
	which extensions failed and wondered who came up with the idea to flood
	it even more with useless errors: CSS parsing errors do not have much
	in common with JavaScript errors. Finding the errors you are interested
	in will be even harder now with all this internal browser and CSS errors
	from random pages junk&hellip; Tab sorting, on the other hand, is cool.
</p>

<h3><a href="http://wtl.sourceforge.net/" rel="external">WTL 7.5</a></h3>
<p>
	Cool :]
</p>]]>
			
		</content>
	</entry>

	<entry>
		<title>I got a new Mouse</title>
		<link rel="alternate" type="text/html" href="http://b-cage.net/log/archives/2005/i-got-a-new-mouse/index.php" />
		<issued>2005-11-20T00:13:09Z</issued>
		<id>tag:b-cage.net,2005:/log//1.57</id>
		<created>2005-11-20T00:13:09Z</created>
		<modified>2006-12-09T02:05:38Z</modified>
		<summary type="text/plain">About hardware configuration software</summary>
		<dc:subject>Users are...</dc:subject>
		<content type="text/html" mode="escaped" xml:lang="en" xml:base="http://b-cage.net/log/">
			<![CDATA[<p>
	&hellip;and it (the <a href="http://www.logitech.com/index.cfm/products/details/DE/EN,CRID=2142,CONTENTID=10121" rel="external">Logitech MX518</a>) rocks :]
</p>
<p>
	The problem I have is the configuration software that accompanied it.
	Why? Because it limits the configuration of the individual buttons:
	certain settings are only available for the buttons that were design to
	trigger the functionality. The button that defaults to "browsers history
	forward" cannot be set to "browser history back". Same goes for the other
	buttons: the default configuration is only available for the that button,
	there is no way to switch the functionality.
</p>
<p>
	Additionally, the UI shows "Standardtaste" ("Default button") - this might
	be a localization thing only, though - for the default configuration. What's
	the standard for a (other than the left and right and maybe the middle)
	button? With 8 buttons and a flyer for the manual only users will tend to
	forget (or better: not even know) the default&hellip;
</p>
<p>
	Lesson learned today: users do not expect what you assume.
</p>
<p>
	<strong>Update:</strong> The new version of the Set Point
	software fixes this.
</p>]]>
			
		</content>
	</entry>

	<entry>
		<title>C#, MDX and Colo(u)rs</title>
		<link rel="alternate" type="text/html" href="http://b-cage.net/log/archives/2005/c-mdx-and-colours/index.php" />
		<issued>2005-11-09T07:58:17Z</issued>
		<id>tag:b-cage.net,2005:/log//1.56</id>
		<created>2005-11-09T07:58:17Z</created>
		<modified>2006-12-09T02:05:49Z</modified>
		<summary type="text/plain">A small adventure in the managed world of DirectX to (inaccurately) simulate the perception of the desktop with color vision deficiencies.</summary>
		<dc:subject>.NET</dc:subject>
		<content type="text/html" mode="escaped" xml:lang="en" xml:base="http://b-cage.net/log/">
			<![CDATA[<p>
Reading about <a href="http://www.michelf.com/projects/sim-daltonism/" rel="external">SimDaltonism</a> made me remember my affords from October 2003: back then I used a color lookup table to do the translation and so the speed was not really satisfying. But now with Managed DirectX, pixel shaders and a <a href="http://colorlab.wickline.org/colorblind/colorlab/" rel="external">nice algorithm</a> at hand I revived the project to see if it can be made perform any better than a few frames per second. And yes, it is possible: around 100 FPS for a 640x480 screen capture simulation on my ATI 9600 Pro.
</p>
<p>
The hardest part of this project was to squeeze the algorithm into the 64 arithmetic instruction slots of a version 2.0 pixel shader. But a few optimizations later and shortcuts later the compile finally did not end in a "<samp>error X5608: Compiled shader code uses too many arithmetic instruction slots (69). Max. allowed by the target (ps_2_0) is 64.</samp>"
</p>
<p>
The quick prototype used the MDX <code>SampleFramework</code> class put for the final application I wanted no to use it (why? I have no idea, file size wouldn't be the reason: 204 kb with and 52 kb without do not really make a difference; maybe because I did not need the high-speed message loop&hellip;). So I had to implement all the boring device householding myself which led to some weird exception when minimizing the application. I had to cancel the devices resizing event in order to get rid of this because otherwise the device wants to recreate(?) itself which will fail for a zero-size window. Why this is necessary I do not know and the documentation does not mention this neither (at least I did not find it). This was an annoying and time costly experience :]
</p>
<p>
For the screen capturing I use an optimized version of <code>Fusion8Design.Win32.Desktop.GetBitmap()</code> by <a href="http://blogs.geekdojo.net/brian" rel="external">Brian Scott</a> of Cropper fame which uses one static bitmap only. Also, it captures the requested part of the screen only and not the desktop area. This reduced the memory footprint significantly - ok, I might have missed the one supposed dispose on the returned bitmap, but creating two bitmaps per frame is rather suboptimal&hellip;
</p>
<p>
<strong>&hellip;and now for something completely different:</strong> <a href="http://blogs.msdn.com/johnrdurant/default.aspx" rel="external">John R. Durant</a> is right about 
<a href="http://blogs.msdn.com/johnrdurant/archive/2005/11/07/vsto2005launch.aspx" rel="external">Metallica's St. Anger</a>: since I heard the title track for the first time it felt like the old (Ride the Lightning, Master Of Puppets and &hellip;And Justice For All) Metallica again. Powerful drums, nice riffs and cool lyrics (ok, not even a single Kirk solo is sort of confusing, but it <em>works</em> - and no one could have been more surprised by this than me). To me, the black album was of a disappointment back then in 1991 and I still think it is one of their weaker albums.
</p>
<p>
Note to self: I either need to find a desktop blogging application that fits my needs or create one; drafting posts in OneNote and converting it to HTML I accept is not very pleasing&hellip;
</p>
]]>
			
		</content>
	</entry>

	<entry>
		<title>Fonts for Programming</title>
		<link rel="alternate" type="text/html" href="http://b-cage.net/log/archives/2005/fonts-for-programming/index.php" />
		<issued>2005-11-05T22:36:35Z</issued>
		<id>tag:b-cage.net,2005:/log//1.55</id>
		<created>2005-11-05T22:36:35Z</created>
		<modified>2006-12-09T02:05:56Z</modified>
		<summary type="text/plain"><![CDATA[About The Right Font&trade; for programming.]]></summary>
		<dc:subject>Misc</dc:subject>
		<content type="text/html" mode="escaped" xml:lang="en" xml:base="http://b-cage.net/log/">
			<![CDATA[<p>
<a href="http://untidy.net/blog/2005/10/27/programming-fonts/" rel="external">Simon says</a> (a pun a day keeps the devil away) to <a href="http://www.donationcoder.com/Software/Jibz/Dina/" rel="external">test Dina</a> as a font for your editor(s). Well, I did and I like it. But (there's always a but if you believe that Courier New is the best suitable font for programming [and I tried a lot of monospace fonts]) there are few things I do find irretating:
</p>
<ol>
<li>Different widths for normal and bold styles (this is only an issue with <a href="http://www.pnotepad.org/" rel="external">PN2</a> and commenting out code as I am too lazy to change its default style settings)</li>
<li>The '+' (plus) is too small (not the same width as the '-' (minus)) and sort of a less dark '*' (asterix)</li>
<li>Sometimes I confuse the '0' (zero) with the '8' (eight)</li>
</ol>
<p>
Anyway, until I get a LCD to get the maximum out of <a href="http://www.microsoft.com/typography/ClearTypeInfo.mspx" rel="external">ClearType</a> and test <a href="http://www.microsoft.com/resources/design/ClearType.html" rel="external">Consolas</a> I assume I will to stick to Dina as my editor font of choice for now.
</p>
<p>
<strong>Update 2005-11-18:</strong> Fixed link to Dina font.
</p>]]>
			
		</content>
	</entry>

</feed>