Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 31 Mar 2003
Offline Last Active Jul 19 2014 10:42 AM

Topics I've Started

Gamedev medical advice thread

14 November 2011 - 02:09 PM

So, I have a medical question I cannot get a good answer to by means of furious googling. And since we all know game programmers represent the top of the line in medical advice, ive got something for you to chew on:

I broke a CFL bulb, and im wondering what its effect on me might be. Yes, I know, if you see a CFL bulb break, ventilate, stay away for a bit, and nothing much to worry about.

However, consider the following scenario: I bought two CFL's; one had a cracked layer of glass, but as far as I could tell only in the outer cosmetic layer. The coiled up glass containing the mercury seemed fine. The light seemed fine on turning it on, so I figured the inner glass must be intact.

A few days down the road, I notice the light color of the cracked bulb is noticably off (shifted towards higher color temps - the mercury is there to modulate it to visible light..), and the overall light production about half of the other bulb. Upon closer inspection the outer glass seems not only cracked, but noticably dislodged from the base (seems like a production error).

What triggered me to be reminded of the dubious CFL situation was a strange rash on my arm (I am normally not a rashy person). Further I have a variety of symtoms (dry cough with some bloody pleghm, diahrea without the usual cramps that accompany it, and a general sense of weakness and lack of focus); though nothing crazy and I had attributed it to viral infection earlier.

Considering I have been spending about 36 hours out of the 72 since installing the lamp on a mattress directly at ground level (where mercury vapour accumulates), with closed windows and a tight fitting door, and no other means of circulation, my question to you is:

Am I backwards rationalizing my rather a-specific symptoms, or should I go see a real doctor and get me some DMSA ASAP? Its a tricky question since most information out there seems heavily biased; either crackpot the one way, or the typical more subtle propaganda the other way that only tells you about the best case scenario (break lamp, clean according to guidelines, no biggie). The total amount of mercury ingested varies by many orders of magnitude (from the one decent study ive seen), depending on air circulation and how close to the ground you are, and the important factors all seem to be working against me. Even if only a few percent of the few mg of mercury escaped, I wouldnt be surprised if I took a substantial hit of that. That would be bad, mkay.

Any obvious steps I could take? Simple household tests for mercury in carpet or hairs (or complex ones, I have access to pretty much any chemical or instrument I can imagine)? I certainly saved the lamp to sue the shit out of the assholes that made this thing if all my hair falls out over the next week.

Or goes the fact that my lamp is still burniing at all prove that there can not have been significant gas exchange between the inner tube and environment? Perhaps over pressure during heating while on allowed gasses to escape, but the under pressure while off was not big enough to let non-noble gasses in and burn the phosporus inside?

On a related rant, I feel like the eco-terrorists really won from me this time, in a roundabout way. All else being equal, I tend to assume all health scares are at least three orders of magnitude overblown, because of the whole health-hazard-industrial-complex, but I hadnt realized they are spinning the other way with regard to this issue. I had never stopped to think that if you break a CFL near your face while breathing in, you are a dead man; the only thing that makes these things safe, in a statistical sense, is that 'under reasonable assumptions', you are only going to breath in a tiny fraction of the total contained vapor. Im not buying any more of these things, thats for sure, and why people are allowed to keep them in close proximity to children but are not allowed to just rape them outright is a mystery to me. Oh yeah and to any eco-terrorists out there: thanks for banning incandescent bulbs; at least ill have some sort of last laugh while I run the electricity my overpriced and broken CFL saves through my electric heater to achieve the same purpose, minus the light.

Linux is a LIE

05 March 2011 - 10:04 AM

As an introduction, ive always been an exclusively windows user. Not for any particular reason; im just not into system administration. The less time I spend on it the better, so that pretty much makes one use a single operating system. Which due to historical reasons, happened to be windows.

Now, that has worked just fine for me up until now, but at my new job I find that many people are using unix stuff, so im going to have to bite the bullet and learn a new paradigm that I dont care about sooner or later, I figured. I was actually kinda looking forward to it. Even though I dont care for knowing all the innards of my operating system, id been told this was optional nowadays with linux; that it has a high 'it just works' factor. Especially respected distributions like Ubuntu.

Lies, I tell you. Installation is smooth; just a few clicks, no problems there. But the first problem turns up soon enough: my linksys wireless network isnt working. Why? Cause they dont make any linux drivers. Well then thats linksys fault you say? Well great, but I couldnt care less whos fault it is: its not working. Some googling informs me that one can make the windows XP driver work on linux. You just need half a day of hacking around the innards of linux; that is, if you are an experienced linux user, which im not. But the most damning thing is of course: if this driver magic is technically possible, then why isnt there some tool included with linux to facilitate the process? Bottom line: it doesnt work, and its not going to work given the amount of time im willing to spend on it.

An hour or so later, after I try to reboot into linux after some windows work, it has already completely died. If any of my hardware broke, windows didnt notice. I didnt change a single setting, didnt download a single driver, exited linux properly. Yet it fails to boot, without any visible excuse. Well, it dumps a ton of jargon on my head, which, if I knew what it meant, might suggest a single-second solution. But I dont, and more importantly, I dont care to find out. Ive got actual work to do. But hey, at least this screen of death isnt blue; I guess that counts for something. For the record: i havnt had windows pull such crap on me since windows 98, and if it did, it was always clearly my own fault, messing around with system drivers or something.

So, to recap: Linux is a lie. In particular the notion that it can even lick windows' boots when it comes to 'just working'. Yeah, im looking at you, GDnet lounge linux fanbois. You had me fooled. I guess im going to shell out for snow leopard and do my unix stuff on that.

Flame on!

Strange computer behavior

08 November 2010 - 03:11 AM

Not sure if its against any (informal) rule to use the lounge as tech support, but ive got an interesting issue that I havnt been able to crack, and seeing the youtube-comment like level of the average message board, I thought id pitch it at the people here, who actually tend to have a clue how a computer works its magic.

Ive got an i3(330) sony laptop, running up-to-date win7, SSD, latest bios, no tweaks. Ever since I got it, I noticed it every now and then my mouse cursor would freeze for a second or so. Didnt happen often; new operating system, no crashes; I didnt think much of it.

But its quite a peculiar freeze; its not just the mousecursor, but seems like a brief complete system-wide anyeurism. For instance, when listning to music, the output goes all bonkers while its freezing, as if the DA converter is unexpectedly choking for fresh input. Keystrokes entered during freeze do appear after unfreezing, but the display is completely frozen, including aforementioned mousecursor and blinking text cursor.

My flash player just crashed, and since then (havnt rebooted yet), its happening almost every minute instead of once a day. An unhandled exception in visual studio is a guaranteed way of triggering it. But its not as if maxing out processor or memory load seems to matter a thing.

I found one report from a guy who had the same issue with his sony i3(330), tech support obviously completely clueless as to what to do; he got a new sony i3(350): same thing.

Could this be sony-bloatware related? Faulty memory? Other sputtering hardware? A win7 compatibility issue? The consistency of the unhandled exception thing seems the most peculiar and informative piece of information, but im not sure how to interpret it.

I must say I can only guess as to what could cause this; is there even any software that could completely hyjack a processor for such a time? Doesnt modern hardware automatically pass control to the OS-tread periodically (with a highly sub-second period)? If other laptops with similar hardware do the same thing it cant be a sony thing. My guess is crappy first gen mobos, and there not being much I can do about it.

Please dont say: 'let sony take a look at it'. Im not going to miss my connection to the world for three weeks for a new system install with more bloatware.

View Library

18 January 2009 - 05:26 AM

I was writing some code that involves a lot of lazy filtering of data. As i was writing this, i started to feel like i was reinventing the wheel, and indeed i was. The concept of a 'view' container adaptor as a predicate being applied to a range isnt a new one. So, i want to be able to write stuff like:
View lazy_set_intersection = filter_view(left, predicate_contains(right));
Where view is a container adaptor which iterates over the entries in left, skipping all entries for which the predicate doesnt evaluate to true. So i decided to quit my reinvention of the wheel. I figured that surely, boost must have a solid implementation of something like this included? I find it hard to believe, but it seems the answer is 'no'. All this stuff is included in boost/fusion for friggin metadata containers, but for plain old data containers, it isnt? wtf? STL? No luck either it seems. The best ive found so far is VTL, the View Template Library. It hasnt been updated since 2000, and generally, im not too convinced its all that optimal in terms of boost/stl compliance. Surely i must be missing something here?

Foam party

13 December 2008 - 06:10 AM

Im organizing a party with a bunch of people: one idea i had was a foam party, but not with soap-foam, but the foam flakes often used in packaging. I think i got the idea from this very lounge, from capn_midnight specifically. The idea was to go to a store that disposes of this stuff in bulk, and load a few cubic meters of the stuff into a van. Hilarity ensues. Does anyone have experience with this? What kind of products are typically protected by these things? (i wouldnt know what store to call really) How much time should we allocate for cleanup, or is that so astronomical that we shouldnt bother?