<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Disqus - Latest Comments for adurdin</title><link>http://disqus.com/by/adurdin/</link><description></description><atom:link href="http://disqus.com/adurdin/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Tue, 08 May 2018 21:07:58 -0000</lastBuildDate><item><title>Re: \n: Thief II: The Vocal Age This follows on from my...</title><link>http://backslashn.com/post/173632242812#comment-3892100899</link><description>&lt;p&gt;I really don’t think it’s robust enough to use in a mission (though I’d be happy to be proven wrong!). Even aside from the times when it doesn’t play the correct animation, the assertion failure errors—which you can only ignore in Dromed, they’ll crash the game—seem insurmountable. Although *if* you don’t use any Conversation-driven voice lines, you might be able to avoid that. But I think it might still happen if two guards with animated mouths are talking at the same time, I’m not sure.&lt;/p&gt;&lt;p&gt;If you just want to change facial textures and don’t need the mouth animations for speech, just use Renderer &amp;gt; Mesh Textures (assuming your map is for Thief 2), and with a couple lines of squirrel scripts you can make them change whenever you need. Since the neutral/smiling/wincing etc. in the Face Textures property was never hooked up to AI behaviours, you’re not missing out on anything by not using that property—and you’re avoiding potential crashes.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Tue, 08 May 2018 21:07:58 -0000</pubDate></item><item><title>Re: Radiator Blog: Let's Play: the first section of Anomalous Materials from Half-Life 1</title><link>http://www.blog.radiator.debacle.us/2013/04/lets-play-first-section-of-anomalous.html#comment-879996356</link><description>&lt;p&gt;The primacy of architectural geometry over set dressing?&lt;/p&gt;&lt;p&gt;You're right that high budget games usually put a lot of effort into set dressing, but I'm not really sure what you're saying beyond that.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Mon, 29 Apr 2013 19:35:44 -0000</pubDate></item><item><title>Re: \n: System Shock 2: Verisimilitude</title><link>http://backslashn.com/post/44238785850#comment-816447171</link><description>&lt;p&gt;I agree that the Von Braun as a whole doesn’t work. As you point out, the decks don’t line up, there’s bits of level sticking in all directions—and there’s really no way to align the shape of the space that you see in side with the exterior view of the vessel in the opening cutscene.&lt;/p&gt;&lt;p&gt;But it feels plausible while playing—whether it would withstand any attempted analysis as a real starship is of course irrelevant.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Fri, 01 Mar 2013 08:38:59 -0000</pubDate></item><item><title>Re: \n: Game developers: stop using shitty SSAO that...</title><link>http://backslashn.com/post/37712343299#comment-737848402</link><description>&lt;p&gt;Whoops, I completely misinterpreted your original comment. I thought you were making a snarky comment, and replied similarly. Sorry.&lt;/p&gt;&lt;p&gt;I'm definitely with you on the relative value of photorealism.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Sat, 15 Dec 2012 17:30:22 -0000</pubDate></item><item><title>Re: \n: Game developers: stop using shitty SSAO that...</title><link>http://backslashn.com/post/37712343299#comment-737186498</link><description>&lt;p&gt;Jason and Evan, I'm so proud of you both! The most I had hoped for was to inspire people to  improve their SSAO implementations to reduce the incidence of prominent, ugly artifacts. Perhaps also to look for more perceptually correct models instead of ones that overemphasise contrast in proximity. But that you would abandon it entirely in favour of inferior or impractical solutions? Gentlemen, I admire your courage!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Sat, 15 Dec 2012 06:12:22 -0000</pubDate></item><item><title>Re: http://backslashn.com/post/11371816533</title><link>http://backslashn.com/post/11371816533#comment-333713249</link><description>&lt;p&gt;No. The two games are not formally related, although some have speculated that Shadow of the Colossus is a sequel to ICO (based rather weakly on the evidence that the two characters are boy and girl, and that the boy grows horns in SotC—despite not losing them in ICO). &lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Thu, 13 Oct 2011 13:00:49 -0000</pubDate></item><item><title>Re: Old Sketchbook Designs #2 – Lighthouses</title><link>http://www.mainlyaboutgames.co.uk/2011/07/old-sketchbook-designs-2-%e2%80%93-lighthouses/#comment-251935967</link><description>&lt;p&gt;This sounds rather like a twin-stick equivalent of Flight Control.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Thu, 14 Jul 2011 07:52:51 -0000</pubDate></item><item><title>Re: Stay Hungry, Stay Foolish - Immutability and Blocks,
  Lambdas and Closures | …on Coding, Mac OS X |
  Algorithm.com.au</title><link>http://algorithm.com.au/blog/files/immutability-blocks-lambdas-closures.html#comment-177292472</link><description>&lt;p&gt;Mutability of names (the "variables" in Python) is the issue in your example—not mutability of values that Manuel discusses.; you don't need to create a deep copy of the 'do', 're', 'mi' objects for your lambda to work on.&lt;/p&gt;&lt;p&gt;The behaviour you expect from the example would require copying the names in the scope when the closure is defined. This sounds reasonable, but would be mean closures would behave radically differently from other functions in Python as regards name resolution. Name resolution in Python is always dynamic: a name is only resolved when the expression using it is evaluated. And so function bodies never resolve the names referred to within it when the function is created, but only when it is executed. Changing this behaviour just for closures would be even more confusing—and, well, every function in Python is really a closure over the module's "global" scope, so it would be a radical change to the entire language.&lt;/p&gt;&lt;p&gt;The other half of the issue is that for loops do not create their own scope—only function and class definitions do that. If on every iteration a for loop created its own scope for its body every , and bound the loop name 'm' in that scope, then your example would work exactly as you expected. This would be semantically equivalent to:&lt;/p&gt;&lt;p&gt;    def _loop_body(m):&lt;br&gt;        funcList.append(lambda: callback(m))&lt;br&gt;    &lt;br&gt;    for _tmp in ('do', 're', 'mi'):&lt;br&gt;        _loop_body(_tmp)&lt;/p&gt;&lt;p&gt;This is basically the same semantics that Ruby has, using .each and a block. This has the side-effect that the loop variable is no longer available after the loop has exited. It would also mean that the loop could not rebind—i.e. assign a new value to—any name in the enclosing scope, as assignment in Python always binds the name in the current scope (with the sole exception being when the execrable global keyword is used). So this example of a not uncommon pattern would no longer work:&lt;/p&gt;&lt;p&gt;    # hypothetically creating a new scope every iteration&lt;br&gt;    found = False&lt;br&gt;    for needle in haystack:&lt;br&gt;        if test(needle):&lt;br&gt;            found = True  # rebinding name in outer scope&lt;br&gt;            break&lt;br&gt;    if found:&lt;br&gt;        print needle  # name 'needle' not in this scope&lt;/p&gt;&lt;p&gt;One final note: Guido's remark that he regrets including lambda in Python are about the syntactic construct; he is not expressing regret over allowing closures.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Sun, 03 Apr 2011 05:54:02 -0000</pubDate></item><item><title>Re: 5 Mistakes Big Copyright Holders Constantly Make</title><link>http://www.plagiarismtoday.com/2011/03/21/5-mistakes-big-copyright-holders-constantly-make/#comment-169503164</link><description>&lt;p&gt;"rather than giving permission for such reuses, they enforce their copyright in piecemeal, erratic ways. This creates an atmosphere of confusion and one where uses of a work are not treated equally. Both sides would benefit from a clear guide on these issues, but rightsholders don’t want to give up any rights at all, even temporarily, as they fear creating bigger problems."&lt;/p&gt;&lt;p&gt;This is exactly the problem that Creative Commons licences are designed to solve: clear and simple information on how end users may redistribute and adapt content. But big content is still primarily considering middlemen as their customers—TV stations, CD and DVD retailers, digital music stores—and not considering the end user. So it's not too surprising that they keep making the mistake of failing to meet the needs and desires of the end users.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Tue, 22 Mar 2011 07:21:48 -0000</pubDate></item><item><title>Re: Mini Ninjas</title><link>http://www.mainlyaboutgames.co.uk/2011/03/mini-ninjas/#comment-163173416</link><description>&lt;p&gt;I loved the environments. The art was cute, the sound was very good. The environments were very varied throughout the levels as well.&lt;/p&gt;&lt;p&gt;Unlike you, I actually enjoyed the largeness of the levels, despite them being mostly empty. But then I'm very much an "explorer" in games, and enjoy exploration as an end in itself.&lt;/p&gt;&lt;p&gt;The boss fights were pretty tedious, and I eventually stopped playing before completing the game on (I think) the second-last boss.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Wed, 09 Mar 2011 16:30:21 -0000</pubDate></item><item><title>Re: Heavy Rain, Metro 2033, and being tired</title><link>http://www.mainlyaboutgames.co.uk/2011/02/heavy-rain-metro-2033-and-being-tired/#comment-154571266</link><description>&lt;p&gt;“barring any unfortunate difficulty spikes involving hordes of melee-attacking enemies.” “Melee + fast + low = First Person Shitness.”&lt;/p&gt;&lt;p&gt;The worst spike of that is in the middle of Chapter 4. If you haven’t got to it yet, you’ll know it when you see it. The monster enemy designs are one of Metro 2033’s biggest flaws in my estimation.&lt;/p&gt;&lt;p&gt;My two favourite parts for shootery were the first and last levels of Chapter 4. I gather you must have already done the “front lines” level—I’m surprised if you thought that was “crushingly boring”.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Wed, 23 Feb 2011 08:41:04 -0000</pubDate></item><item><title>Re: The silent protagonist is wrong</title><link>http://backslashn.com/post/3400897097#comment-153165703</link><description>&lt;p&gt;Chris, you're right in that I'm only considering games that are trying to tell a story about the protagonist. I don't think Mario falls into that category. I want to be clear—I am not meaning to write off games that let the player mold the character of their avatar to a high degree—Alpha Protocol and New Vegas being two recent examples of this type—I was thinking solely about those with a linear narrative to impart.&lt;/p&gt;&lt;p&gt;Far Cry 2, Half-Life 2 and Bioshock are all in this category. And they each have a different approach. HL2 I have covered above; although I'll note that it provides no opportunities for meaningful player expression.&lt;/p&gt;&lt;p&gt;Far Cry 2 gives you a choice of several interchangeable avatars, but then puts you in a world where you know nobody and nobody knows you—so apart from being a mercenary, your avatar’s character is a blank slate. And the rest of the game is (to my reading) about the role you play as a mercenary in the conflict, and for the most part it leaves you alone to work out your own motivation and control your avatar’s actions accordingly. But it won’t let you decide that you don't want to play the mercenary role; that is fixed.&lt;/p&gt;&lt;p&gt;Bioshock gives your avatar both an identity and a history—which is crucial to the plot—but again, ascribes no particular traits to the character. (Does Jack not speak in the game? I thought he did.) But the rest of the game is trying to tell a particular story. Most of the game is just mechanical, with no room to explore either your avatar’s or your own character—except perhaps the rescue/harvest dilemma.&lt;/p&gt;&lt;p&gt;I'm trying to think of a game that exemplifies my point. Mass Effect 2 would be a perfect example, assuming the game responded to the character of Shepard from the first one instead of merely certain events. I don't believe you start out being either renegade or paragon, for example, even if you maxed out one or the other in the first game—but correct me if I'm wrong.&lt;/p&gt;&lt;p&gt;But otherwise Mass Effect 2 is a good example, moreso if you haven't played the original, or are not importing a save. Shepard has a lot of history, both her reputation as a soldier, and past relationships and romances with other characters: Tali, Wrex, Liara, Garrus. When you meet these characters, they respond to your avatar based on their shared  history, of which you the player may be ignorant (as I was in the case of Liara; in ME1 I had little to do with her, but I lost my savegame before playing ME2 and had to go with the canonical Liara-as-love-interest). So the preexisting relationship between the characters informs your responses, but does not impose upon your actions, and your motivations and actions still determine the future course of these relationships: the player’s agency is still king.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Sun, 20 Feb 2011 17:49:08 -0000</pubDate></item><item><title>Re: Free Metro Can’t Stop Making Money
	
	
		
		
	
	
	
	
	
	
		
	
	
	
	
		
		
	
	
	
	
	
	
	
	
	
	
	
	
	
	
		
	
	
	
	
	
	
	
	
	
	
	
	
	
	| 
	paidContent:UK</title><link>http://paidcontent.co.uk/article/419-free-metro-cant-stop-making-money/#comment-137789602</link><description>&lt;p&gt;Your bolding of “it made the best part of £30m” is disingenous, as it draws undue attention that phrase as if it were factual, when in fact the whole sentence makes clear that it is just a guess or estimate.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Mon, 31 Jan 2011 16:01:08 -0000</pubDate></item><item><title>Re: Python Exception Handling Techniques - Doug Hellmann</title><link>http://www.doughellmann.com/articles/Python-Exception-Handling/index.html#comment-11450762</link><description>&lt;p&gt;Well, looks like &lt;/p&gt;&lt;pre&gt; didn't work there, and the indentation was completely lost.  Shouldn't be too hard to figure out I hope.&lt;/pre&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Fri, 19 Jun 2009 11:05:38 -0000</pubDate></item><item><title>Re: Python Exception Handling Techniques - Doug Hellmann</title><link>http://www.doughellmann.com/articles/Python-Exception-Handling/index.html#comment-11450690</link><description>&lt;p&gt;When writing library code, I often want to catch numerous exceptions and wrap them in a error from the library.  Then consumers of the library can manage different classes of exceptions more readily:&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;pre&gt;class DatabaseOpenError(StandardError): pass&lt;br&gt;&lt;br&gt;def open_db(db_filename):&lt;br&gt;    try:&lt;br&gt;        db_file = _internal_open_db_file(db_filename)&lt;br&gt;    except IOError:&lt;br&gt;        cls, exc, tb = sys.exc_info() &lt;br&gt;        raise DatabaseOpenError("Could not open database"), None, tb &lt;br&gt;&lt;/pre&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;The traceback the caller can get from the DatabaseOpenError then traces all the way back to the original IOError in the depths of _internal_open_db_file(), so no information is hidden.&lt;/p&gt;&lt;p&gt;Normally, I'd also store the original exception instance in an attribute of the DatabaseOpenError, but I omitted that for brevity.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Fri, 19 Jun 2009 11:03:43 -0000</pubDate></item><item><title>Re: Resurrecting MissingDrawer plugin for TextMate</title><link>http://jannisleidel.com/2008/02/missingdrawer/#comment-13856264</link><description>&lt;p&gt;BUG: COMMAND-R toggles the folder list instead of just revealing the current file in it (it does do the reveal, but always toggles the list).&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Wed, 27 Feb 2008 05:33:00 -0000</pubDate></item><item><title>Re: OS X icon transparency and Tasks bundle file icon</title><link>http://thepugautomatic.com/2007/08/os-x-icons/#comment-8325237</link><description>&lt;p&gt;I use the excellent &lt;a href="http://iconfactory.com/software/iconbuilder/" rel="nofollow noopener" target="_blank" title="http://iconfactory.com/software/iconbuilder/"&gt;IconBuilder&lt;/a&gt; plugin for Photoshop to create icons.&lt;/p&gt;&lt;p&gt;And Icon Composer that comes free with Xcode (installed in /Developer/Applications/Utilities) will preserve transparency if you drag   drop a Photoshop image onto it, creating a suitable .icns file.&lt;/p&gt;&lt;p&gt;Then the handy freeware app called &lt;a href="http://www.icons.cx/goodies/" rel="nofollow noopener" target="_blank" title="http://www.icons.cx/goodies/"&gt;icns2icon&lt;/a&gt; can will apply the icon itself to the .icns file, so you can then copy   paste it accordingly.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">adurdin</dc:creator><pubDate>Thu, 16 Aug 2007 13:33:41 -0000</pubDate></item></channel></rss>