3/31/2010

03-31-10 - Video Annoyances

Oo lord working with video is annoying.

I noticed over at Media Xiph.org they have some sample videos in Y4M 444. 444 means it has full res chroma. I have previously tested against x264 using some Y4M 420 sequences like "parkrun", but the problem with that is that it has pre-killed chroma, so it actually is biased against me because I don't kill chroma in the same way they do, so I'm sort of taking a chroma killing penalty twice. What I want is full res color data so that x264 and I can both kill chroma in our own ways and then compare to the original.

So I get the full res 444 Y4M data. Then the problems begin. Most Windows video tools won't real Y4M because it's not packaged in an AVI with a codec for DirectShow and all that. But the linuxy tools that are starting to dominate do read Y4M. *BUT* my favorite tools, x264 and mplayer/mencoder both don't read Y4M 444, they only read 420. Now the linuxy thing to do would be to pipe the video through y4mscaler, but then I'm killing the chroma. That's fine to do the x264 compress, but I need to get back to RGB full res for my compress.

So now FFMPEG does read Y4M 444. But for some fucking reason the ffmpeg on my machine refuses to show any command line help. All the things I'm supposed to be able to do to make it tell me what codecs it has and so on don't work. Yay. So I get to randomly search the web and try things in the dark until it works.


ffmpeg -i t:\park_joy_444_720p.y4m -vcodec rawvideo -pix_fmt bgr24 r:\park_joy_444_720p_rawvideo.avi

So we have raw video out, yay! That's awfully big so we want to compress it with something lossless. The only lossless codec that I have found which really works in all players and supports plain RGB right is Lagarith. (other people have had success with HuffYUV but I have trouble making it work with RGB and it randomly crashes on me sometimes).

BTW if you want 420 subsampled colorspace YUV, x264 is a great option for lossless video, and it's easy, just compress with -q0. In fact this is a good way to convert a 420 Y4M or YUV file into an MP4 which can then be loaded by Windows video tools. HOWEVER you may need to specify low profile and no bframes, and even then some Windows video tools will still have problems with x264 output. Also - Y4M is a raw YUV file with a header. A .YUV file has no header and you have to tell it the size. The easiest way is to name the file "blah_widthxheight.yuv" and then give it to x264, which will parse the file name for width & height.

Unfortunately we cannot go straight to Lagarith in ffmpeg because FFMPEG is some linuxy thing that uses hard-coded "lib" system rather than using the Windows pluggable codec thing. So even though I have Lagarith on my machine and all the Windowsy video products can encode to it, ffmpeg can't. (I guess on Linux ffmpeg actually loads libs using the purported magic Linux lib system, but on Windows they are hard-compiled into FFMPEG; that is advantageous in that it means ffmpeg actually fucking works, unlike the apps that rely on DirectShow codecs to be set up right which is a fragile disaster, but it's disadvantageous in that you rely on the whims of ffmpeg builders to put in the right codecs). So anyway to transcode I have to use VirtualDub. Vdub is a pretty nice app for simple AVI transcoding, it's basically just an interface to the DirectShow filters.

(BTW I guess I could do this directly in ffmpeg by going throught AVISynth since there is an ffmpeg-avisynth connection and then avisynth can get you to anything in DirectShow, but lord help me if I have to get into that snake pit).

I set up my config for transcoding in the VirtualDub GUI and then I save the settings out to a "VCF" file and then I can run it from the command line like this :


c:\progs\virtualdub>vdub /s r:\vdub_lagarith.vcf /p r:\park_joy_444_720p_rawvideo.avi r:\park_joy_444_720p_lagarith.avi /r

Okay, so now we have the lagarith RGB video. That is all I need to be an input to my coder, but to compress with x264 we need to go back through some hoops. The most reliable way I've found is to use MPlayer to turn that into Y4M :


mplayer -benchmark -ao null -vo yuv4mpeg:file=park_joy_444_720p_lagarith.y4m park_joy_444_720p_lagarith.avi

And then finally we can run x264 on the Y4M :


x264_2pass park_joy_444_720p_lagarith.y4m park_joy_444_720p_x264.mp4 --bitrate 10500 --preset slow

x264_2pass.bat :
set x264args=%3 %4 %5 %6 %7 %8 %9 %10 %11 %12 %13 %14 %15 %16 %17 %18 %19 %20 %21 %22 %23 %24 %25 %26 %27 %28 %29
x264 -o %2 %1 -p 1 -I 9999 %x264args%
x264 -o %2 %1 -p 3 -I 9999 %x264args%
REM x264 -o %2 %1 -p 3 -I 9999 %x264args%

(note I specifically do not want to start from the original Y4M and just use y4mscaler or something to get to x264, I want to start from the same RGB lagarith video that my coder will start with, since that's the most unbiased source type; eg. neither of us code in RGB and neither of us are tuned for RGB, so it's not biased one way or the other).

Yay ! Oh, and BTW x264 seems to often severely miss it's bitrate, usually over. That bitrate 10500 is my attempt to make it hit 12500. Depending on the video you'll have to look at the output size and fiddle with that to make it hit the right size.

One annoying niggle is that somebody along the way seems to fuck up framerate a bit. On park_joy , which is at 50 fps, it seems to be perfect. However on the 29.97 fps videos (30000/1001), somebody is doing the math wrong and getting off sync, so on long videos it will have a frame slip, so that if you compare the original to the final output, you'll find a frame where they get out of sync by one frame and you have to step ahead in the source to get them back in perfect sync. I haven't tracked down who the culprit is or how to fix it, so I just live with manually stepping a bit to find sync when comparing. I also haven't found an app that can just change the time code on a video without touching the frame data, that is take a 29.97 fps video and make it claim to be 30 fps, but show the exact same frames, eg. change duration but keep the frame count the same.

03-31-10 - More Porsche 997.1 Owners Notes

The front bumper has two radiators off to the lower sides in air scoops. These scoops fill up with leaves and pebbles and junk from the road. Some people have done a DIY to install mesh screens over these holes (the GT3 has the screens standard). I might do that someday, but for now I just stuck a vacuum cleaner nozzle in the scoops. Make sure you dig the nozzle way back in to the far corner, because there's a little pit back there that is just full of pebbles.

Around the luggage compartment under the hood are water drain channels. These are pretty shallow and if they have gunk in them the water will overflow and can get into the luggage. Make sure you keep leaves and junk out of them. On each side of the battery are little water drain holes to get water away from the battery. Again leaves and junk can easily block these holes which then causes the "brake booster" to suck in water. Check them every so often to make sure they are not clogged.

The oil pressure gauge seems worthless at first (because the oil temp is easier to read and they are semi-redundant), but actually it's the best way to check on the health of your engine and know when it is ready to thrash. When you start it up, it should read around max, 5 bar. Once the oil gets loosened up and is flowing right, it should settle in to about 1-2 bar at idle and 3-4 bar at high RPM. If it doesn't get up to 3-4 bar once warm at high RPM, you have an oil leak. My next oil change is going to be to 5w50.

There are a few known design flaws with the car (aside from the RMS/IMS elephant in the room) :

They seem to be prone to premature serpentine belt failure. This is not a huge problem, just something to be aware of. Changing the serpentine belt is very very easy on this car, like most of the minor maintenance. All you need is a 24 mm wrench. Take a photo of the belt routing before you remove it so that you can replace it the same way. The tensioner is self-adjusting, not manually tensioned, so you don't have to worry about tightening it right when you put it back on. One reason for failure seems to be that the pulleys can wear grooves . I would replace every 30-40k miles instead of the recommended 60k.

Another is that shift cables can break , particularly under heavy use. This is also pretty cheap to replace, though not a DIY, and obviously inconvenient if it happens out on the road. Generally the transmission is very "notchy" and heavy, which is cool and by design, it is meant to be able to handle a lot of torque to quickly shift and get power through. The syncros in the transmission that help you rev match are pretty minimal and hard, that is you need to rev match pretty well or they will balk, unlike a lot of comfort cars where the syncros are very forgiving and let you shift even with a bad rev match. The problem is if you are trying to jam the transmission without rev matching well enough, it puts heavy loads on those cables and they can't take it. Even if they don't break they are prone to get out of adjustment, particularly the 1st/2nd gear clutch control.

3/30/2010

03-30-10 - The Worst

One of the weirdest/worst types of people are the guys who wear really wacky clothes, but then are just boring as hell. Like the accountant who toils away in the cube farm at your company, but wears purple sparkley dress shoes every day. Or maybe he has a whole collection of weird shoes. Often these guys focus on some particular weird accessory, like maybe rainbow suspenders, or bow ties, or hats, or whatever. They are into rockabilly or swing music or some other type of music that for a second you think "oh that's kind of cool" but then you realize it's completely uncool and uninteresting. (another variant of this loser is the guy who is constantly wearing clothing related to their hobby or interest, like wearing their swing dancing outfits to work, or wearing their league bowling shirt or whatever; oo how interesting, I'm just dying to ask you about your hobby which you are so obviously advertising). Their wacky clothes scream "talk to me!" but generally they are quiet and introverted. Obviously they are desperate for human interaction and to be interesting, but their actual personality can't make that happen, so they try to use the wacky clothing items to spark some conversation. Of course I categorically refuse to help them out by asking about their interest or wacky clothing item.

Sometimes I find myself dangerously close to this precipice. On the one hand, I think everyone is fucking boring and retarded, so I have no interest in actually talking to people and being outgoing. On the other hand, I think that just wearing what you're "supposed to" is fucking ridiculous and boring, so I am tempted to get wacky and break the rules of decorum to show that I am not going to play society's stupid conformist fashion game. The result is that I become the boring quiet guy in wacky clothes. That must be avoided.

Recently I've become much more aware of how childish and pouty I can be. When something doesn't go my way, I go into a quiet huff. Partly that's processing and thinking about the next course of action, but it's also excessive dwelling. It's also a subconscious way of showing the world that I have been bothered and you better not do that any more; I actually make a pouty face and glower, it's very obvious if I have a camera pointed at myself (which would be an awesome tool BTW). For some reason I am much more able to step outside my own consciousness and observe myself as I do this recently and realize "wow you are being really pouty".

Playing DDR the other night I realized I have a bad habit of clinging to easy things I am good at instead of pushing myself by doing hard things I'm not so good at. Years ago when I played DDR more seriously, I played almost the whole time on light mode, I moved up to the hardest songs like Paranoia, but would practice them over and over until I could get a 100% perfect, but I stayed on light mode. A few times I tried normal mode and it was really hard and I moved back to just perfecting light mode. I would have gotten much better much faster if I had just pushed myself more to move up in levels, even though it felt hard and frustrating. I realize now that I did the same thing when I was playing poker. I stayed at lower levels where I could completely dominate and make massive profit rates, where I was confident that I was the best at the table, rather than move up to levels where I might struggle a bit.

3/29/2010

03-29-10 - WA Roads for Biking and Driving

I've been doing a lot of scouting and joy riding recently.

Vashon - went to scope out the bike riding - mm meh, it really doesn't look like a great place to ride; for one thing when you get off the ferry you are immediately faced with a monster hill. I hate hills right at the start of rides when I'm not warm. Then once you get to the top, the rest of the island is pretty flat and pastoral. The shoulders are not great and there's a lot of fast traffic. Most of the route slips that I've seen take you on a loop around the island, around the west side and then back on Vashon Hwy on the east. That's fucking retarded, Vashon Hwy is awful to ride on, just do an out and back on the west side. There is one real nice bit of twisties for driving around Cedarhurst but otherwise pretty meh.

West Snoqualmie Valley Road in the Duvall - Carnation area. Way too much traffic for fast driving. Narrow road and no shoulder - does not look like a nice place to bike at all. Scenery is pretty good though - you get views of the Cascades and pastoral river valley farming stuff. It's an enjoyable sunday drive.

A bit further north of there you can do High Bridge Road and Woods Creek Road. Both pretty nice drives as the traffic is a lot lower; these are popular with local motorcyclists, so they may slow you down, and cops do watch them. Again not great bike rides, though Woods Creek might be okay despite lack of shoulder because it is pretty low traffic.

Snohomish is a pretty abysmal tourist trap, but there are still some decent sights there. There seems to be a hot air balloon launching every time I go by it.

A bit further north I hear Menzel Lake and Jordan Road are good, but haven't been up there yet. Also Mountain Loop Hwy up to Silverton - Bedal looks cool, but I think it's pretty rough dirt road through the highest part.

Renton - Maple Valley trail. This is a "rails to trails" conversion, so it's all flat, separated from the road. On a nice weekend it might get unbearably crowded near Renton, but the traffic thins out and it gets more scenic as you get closer to Maple Valley.

Green River Gorge Road / Green Valley Road - this is some pretty rural Washington. Not great for driving though, around the Gorge Road area the pavement is in really abysmal condition, so you have to slow to watch for killer pot holes, and then in the Valley area it's residential and trafficky so you have to / should take it slow. Gorge Road area seems promising for biking, though there are quite a few sport bikes and ricers and such joy riding out here so it may be a bit dangerous. The gorge itself is pretty fucking rad, really deep shear cliffs down to rushing river, but it's hard to access (the small shitty state parks in the area don't provide any access to the good parts of the river); I hear that the locals know the access paths and guard their secrets.

There are some very promising driving areas further afield (mainly in the Mt Rainier - Mt St Helens area) but that will have to wait for summer.

It's sort of weird that it's easier to get away from people in San Francisco than it is here. Part of the issue is that SF was really amazing about saving wild land right near the city (the whole of Marin and the Peninsula). But I think another part of the problem is the Cascades. It's nice that we have the Cascades right here, but there are basically no roads in them, and even very few decent hiking trails at low elevations. So essentially we are trapped in a little strip of land on the west side of the Cascades and almost every bit of it is developed and suburbanized.

03-29-10 - DDR Pads

We fetched out the DDR pads and played a bit. Not a bad game, but I wish there more games to play on the pads. Has anybody made a bunch of mini games that are playable on the DDR pads? (any game that just uses arrows is playable, but that equals roughly 0% of games). Maybe PS2 games are hopeless, but I could also play PC games. Are there any PC games that just use the arrow keys and nothing else? (one I can think is Daleks) (also Boulder Dash)

Simon Says. Just a simpler version of "hit these buttons". It plays a sequence, you repeat it, it adds to the sequence, you repeat it, ad infinitum. Each direction you hit makes a note and the sequence winds up playing a song.

Track & Field. Various competition games that basically consist of slamming on left-right really fast and then press up to jump, etc. permutations thereof. Discus throw could require you to do up-left-down-right-up-left-down-right and so on so there are various patterns to practice, and hurdles would be fun.

Katamari Damacy type game (or Monkey Ball if you prefer) (or Marble Madness). Anyway the point is you steer some ball around by tapping the different arrows.


I tried to get some PC games working on the HTPC so we could play on the TV. Jeebus what a clusterfuck. Fucking video games are such a disaster. I've written this many times before and nobody is paying attention, but here goes again : all video games (and 3d apps in general) must launch initially in windowed mode and must be able to run at whatever the desktop resolution is. (you can launch in maximized windowed mode if you like but you better respond to Windows-M minimization requests and you better fucking let me Alt-Tab to other apps). Furthermore, you better fucking have keyboard controls at least on your menus (arrows and enter please), and you better fucking respect "Escape" and "Alt-X" or "Alt-F4" or whatever.

Basically any jump to "full screen" or another resolution is extremely intrusive and should never be done except when the user specifically requests it. I'll give you a few reasons why : 1. On the HTPC I remote-desktop in to set things up. Jumping to full screen fucks the remote desktop in various ways. 2. I always want to be running in my LCD's native res. My desktop is always at the LCD's native res. You should never just jump me to a res that's different. 3. On the HTPC I use analog TV out. It has to be calibrated for each res. When you jump to some other res, the signal is all stretched and offset and I have to go back to the ATI control panel and adjust it. 4. You don't know what resolutions are good for my card/display - I DO, and I chose them on the desktop, so just leave it alone. 5. "Full screen" is no different than a maximized window on most modern 3d cards. The habit of jumping to full screen for performance is archaic. (yes you might save one blit, but leave that as an option for enthusiasts, not the default). 6. I may well want to run your game and check my email at the same time. Let me. You do not control my experience, I do.

Also, gamepad support on the PC is just such fucking balls. Maybe if I had an MS branded gamepad and only played new games built on XInput it would be okay, but old games using DirectInput and random old gamepads = massive fail. Good luck even getting it detected and selected as the controller, then the sticks will be all crazy out of wack with fucked up deadzones and drifts, and then when you fix that you find the button mapping is all boofood and you have to go into config and map all the buttons to something sensible. Half an hour of setup to get a game to work.

Installing games on the HTPC in general is probably a bad idea, because they are generally so fucked up that I risk destabilizing the system, and I don't want to have to deal with setting it up again cuz it's working pretty well right now. This is why I never play video games. I spent an hour last night fucking around installing things and trying to configure them right.

3/23/2010

03-23-10 - Pinnacle

Bike Snob NYC is really the pinnacle of internet comedy.

I believe there are two major forms of comedy that have been primarily developed by the internet; obviously these existed before, but they have been honed and perfected and seen their hayday in recent years. One is the sardonic / snarky condescension comedy, done so well by Surly Gourmand , but this also goes back to stuff like "Old Man Murray" ; basically in this form you humorously point out how retarded everyone else is. The other major internet comedy form is the "meme" ; this comedy form primarily lives on web forums, 4chan and /b/ being the primary nexus, but most major forums create their own memes. In this form you create a neologism or repurpose a word into a reference, and then it is used repeatedly in humorous ways. The meme is really only super funny if you know the history of where it came from.

Bike Snob combines the snark and the meme to optimum effect. The really amazing thing is that he also pulls off the snark without bile, and he creates all his own memes rather than just stealing them from the net. When you first read Bike Snob you don't really get it, but after following for a few weeks, you see the creation of memes, then the development of their humor value, and then they become funny just any time he drops them.

Linkage :

Quantum Diaries Survivor is a pretty damn good physics blog.

I can never get enough of Denny regrade photos

Scans of Vintage Bicycle Advertisements - Diablo Scott - Picasa Web Albums

Name Your Porsche - in case you thought your Porsche wasn't douchey enough already, you can name it "Bruce". "Bruce" !!

McLaren Automotive officially launches itself and MP4-12C supercar - yeah this thing is pretty fucking hot. The funny thing is that for the masses of technology they put into these super high end cars, they get microscopic returns. It's technology porn really, it's technology for its own sake, to drool over, not to do anything. If you look at the C&D Lightning Lap results for example, the WRX and the Speed 3 are great values at about $25k for a 3:15 lap, and the closer you get to the top the more ridiculous your cost per second gets. Obviously you're paying for looks and feel and so on, the focus on power and speed is all wrong.

fastestlaps.com View topic - airdrag... factory claims vs car magazine tested numbers.. really sweet page of photos and results from the german magazine "Sport Auto". (note that the photos are fakes though still very interesting - the actual runs for Cd are done with much higher air speed and with the wings up, the photos are taken at low speed air flow where you get those nice laminar lines; the real high speed flows have a lot of turbulence wakes; if more photos like this exist, let me know). The key number is not "Cd" , the coefficient of drag, but rather "CdA" , and of course downforce is very important too. The GT3 has very impressive aerodynamics.

Compact Crank Overload - all about the "compact double". I highly recommend compact doubles for all bikes. Sadly my Litespeed is on a 130 BCD crank and I don't want to have to replace the whole crank so I'm fucked.

All About Motor Oil - in case the last one I posted wasn't enough

Aldo's Pic of the Day - classic cycling photos.

3/22/2010

03-22-10 - Pleased

I replaced my engine air filter over the weekend, and it made me feel oh so pleased with myself. Oh look at me, I'm such a blue collar manly man, I'm all covered in engine grease and doing my own maintenance. Of course it's a very easy job, it should take 15 minutes or so, it took me an hour because I'm anal and I cleaned everything out (cleaned the throttle body too which I highly recommend, you should also clean the MAF on newer cars).

It also felt nice to give back a bit to the car. The car gives me so much, it gives me adrenaline rushes and erections and warm butt massages. If I just take all that pleasure and don't give back, I'm treating the car like a cheap lay. You have to show your car some commitment, do nice things for it, give it baths, buy it air filters, it shows you really care.

Working on cars would be pretty fun if you had your own garage with a lift and a full set of tools, including pneumatic drivers and impact sockets and all that shit. I'm kind of tempted to get into it, but it really doesn't make much sense. I could easily do my own oil change, it's quite easy to get to the oil drain bolt and the filter on my car, but then I'd still have to deal with taking the oil to a disposal site and all that shite. I changed my own spark plugs on my old pontiac long ago (which was easy and worth it), and I changed the pads and rotors on my Prelude with the help of my uncle (which was a huge pain in the ass and not worth it). It all would be pleasant and easy if things worked out the way they are supposed to, but there's always some fucking problem where you don't have a certain tool you need, or you bought the wrong part, or some bolt is frozen, or some screw is stripped, and you have to run to the auto parts shop or get a mechanic to help with something and it kills all your theoretical savings. Really you do it yourself not to save money, but for the pleasure of touching your machine, and also because you know that when you do it you take your time and do it right, whereas mechanics rush and fuck things up and don't take care; if some part doesn't go back together quite right the mechanic just hammers it in and then zip ties it down.

We live cheek by jowl with our neighbors. It's strange, we're in an old neighborhood, all the houses are from around 1910, and I've seen photos from Vintage Seattle of when the area was being built (bleh I can't find the photo I'm think of, but this is a decent demonstration ) - there's tons of empty space all around, but they build the houses right next to each other. I'm not sure why they did that exactly back in the day; it wasn't because of the cost of land, it must have been for some practical reason, like maybe running water or sewer lines was really hard, so you saved that work by putting things together? Dunno.

Our neighbors are mostly very quiet, in particular the directly adjacent neighbors are wonderfully quiet, so the proximity is not a problem. But lately one of the slightly farther neighbors has started to violate the unwritten code of neighbor conduct : thou shalt pretend that each other does not exist. You're supposed to close your ears when you hear neighbors talking; if they are outside their house doing something, you pretend you don't see it and you certainly don't go over and talk to them. (note : there is an exception to the "I can't see you" rule, which is when you are on the twilight hour promenade, in that case all neighbors may exchange in cordialities). One neighbor has crossed the line. When I go out on the balcony, sometimes they call out "hello" from their balcony. Fuck you, I won't say hello, I'm trying to pretend you don't exist! The other day N sneezed and neighbor yelled out "bless you". What !? That sneeze is not yours to bless. Fie!

There are also some neighborhood kids who have been playing baseball and other such hard-edged activities right next to my car. I don't have a garage, just a driveway, so I am not protected from the dings of mini basketballs. I feel like it's way too douchey to go out and say "don't play next to my Porsche", but I know it's just a matter of time before I get a dent from those fuckers. The car has got extremely thin body panels to save weight; the hood in particular is so thin you can dent it just by closing it if you aren't very gentle. Maybe I should get a Pit Bull and put it on a leash next to the car and blow a dog whistle any time a child comes near it.

Some of the neighbors are semi-hippyish. I love people who are anti-corporate, people who grow their own veg, people who are tolerant of different lifestyles, people who are into drugs and music, but the fact is most hippies are fucking dicks. Hippies/liberals are the worst kind of people - they act super chill and want everyone to get along - but only if everyone goes along with their idea of how things should be. They're extremely prejudiced and judgemental, if you wear "uptight" clothes they assume you're a square who will harsh their vibe. They're identical to "family values" conservatives in that respect who call for government to stay out of our personal lives, but what they really mean by that is they want government to enforce *their* choice of lifetyle. Hippies espouse "tolerance" but by that they mean tolerance for sexual variations and races, not tolerance for christians and capitalists. I guess extremists on the left and right are both rotten and inconsistent, however let's not get carried away with moral equivalence, someone who wants to force peace and love on everyone is not the same as someone who wants to force christianity and income inequality on everyone. Anyhoo, I can see the hippie neighbors scowl at me because I'm obviously a capitalist pig, and I'm part of the gentrifying yuppie force that is invading their old neighborhood.

3/20/2010

03-20-10 - Offenses

If you're releasing a fucking sample mp3 track from your album, name it "artist - title.mp3" , don't fucking name it "02.mp3" you fucking tardball musicians. I guess a lot of people aren't even aware of the actual name of their music files any more because they use fucking iTunes or some shit that isn't file based. Fucking iTunes not only fucks me up if I try to use it, but even when I don't use it it fucks me up.

Google Maps print display needs more white background. The colors should only be for outlines and road, not the big colored fields they use for things like parks, water, and city blocks. I think they have single handedly made $100 M or so for HP.

When you're going straight on a road that has two lanes in each direction and you come to a red light and you are the first car there - get in the fucking left lane. I am constantly confounded by coming up to red lights where I want to turn right and some solitary cocksucker is sitting there in the right lane blocking me. Similarly if you come up to a red light on a one lane road that has a bit of space, shade to the left a bit so that people have room to turn right around you. This is just being considerate, that is, thinking about how your actions affect others. All the so-called "nice" people rarely give much thought to being actually considerate. Speaking of which :

When I go biking, obviously the worst people are the ones who try to run in to you, or throw bottles at you, or honk at you to scare you, but those are pretty rare. Far more common is the "nice" fuckers who get it all wrong. Maybe the most annoying of all are the people who wave at you to go when it's their turn. I'll come up to a stop sign, and I come to a complete stop because there's a car already there in the perpendicular direction. The best thing the car could do is just fucking go quickly when it's their turn, that way I can go behind them smoothly. Instead they just don't fucking go, I look straight at them like "WTF?" and they wave at me to go. Fuck you you fucking stupid wanker, go when it's your turn. (the most extreme variation of this is people who do not even have a stop sign, I'm stopped with my foot down at a perpendicular stop sign and they come to a stop for me to go; you fucking retard, just go so I can go behind you). A similar stupid "nice" person move is the fucking perpetual hover; they want to pass but don't want to cut too close to you, so they just drive right behind you for like a mile. That's really fucking annoying because I have to be super aware of them all the time to protect myself, just make your fucking pass; usually these people are the retards who refuse to cross the yellow line ever.

People who do their own major home improvement projects are fucking cockmunchers. This old guy that lives next to us is doing a total rebuild of his house. Bang bang bang every day. I think he's been doing it for three years or so; certainly in the 6 months that I've lived here he's made no apparent progress. I think he puts up one board each day. Hire some fucking people and get it finished. Or you should have to pay your neighbors $100 a day for each day that you do construction on your house.

Carry some fucking cash people. If you don't have cash and there's a long line for some cheap service - get the fuck out of line and go home. No you don't get to buy a magazine with your credit card while we all stand in line behind you and tap our feet. And don't blame the fucking convenience store or movie theater for having a slow Visa network connection - you know perfectly well that a lot of places have slow card machines, that's why you carry some fucking cash. It takes two minutes to grab some cash at an ATM, and then you have it for weeks. It saves *you* time, and it's considerate. Oh and when you do go to the ATM, get at least $200. What are you a fucking child that can't carry cash because it makes you spend it too fast? It's also fucking retarded all these people who carry mace and flashlights and leathermen so they are "ready for an emergency" but don't carry cash; cash is the #1 most important survival tool.

Why the fuck are you people on cell phones all the time? Drivers, pedestrians, people in stores, I see them all the time just going through life seemingly constantly with a cell phone next to their head. What the fuck are you talking about? I'm way more interesting than you and I don't have anything to say, I just can't imagine what these people are blabbering on about all day every day.

Don't fucking speed in residential neighborhoods you cocks. Any time there is anything blocking your vision, such as parked cars, you should assume there is a child right behind it who might jump out in front of you. You should only speed in areas with clear visibility or where you are reasonably certain noone is going to jump out (eg. on freeways there are concrete pillars for overpasses that obscure vision, but I'm pretty sure noone is behind them). There's another Porsche in my neighborhood who goes ripping down the narrow streets here and it pisses me off especially because of the association with me. There are also a lot of intersections around here in the residential back streets that have no signage at all - no stop sign or anything, just a 4 way intersection. People go blazing through these intersections which is just so ridiculously stupid.

There are many fucking annoyances with C that I have written about previously. Most of my major annoyances are with the fact that it does not provide you mechanisms to protect yourself as a coder (for example writing "overload" next to a virtual function to make sure it overloads something, which protects you from changing the signature in the parent and not doing the same change to all children). One bug I frequently contend with is accidentally using the same variable name in an inner scope. eg. I have some global named "foo" and then I make an "int foo" inside a function, and then get confused about which one I'm dealing with. I wish I could make that a warning, and have a keyword, maybe reuse "overload" to indicate when I want to do it intentionally.

3/19/2010

03-19-10 - Porsche 997 Owner Notes

Some things I discovered after purchase that I wish I'd known before and are hard to find out :

The car doesn't really like to be cold. For one thing it's sold on "summer tires", and below 45 F or so you really need to be on "winter tires". (winter tires does not mean "snow tires" ; the difference between summer and winter tires is the rubber compound; you want tire rubber to have just the right softness in the operating temperature range; winter tires are chemically created to be soft at low temperature, but you can't use them year round because they would get too soft in heat; there's also a difference in the depth and style of tread, but the rubber compound is the bigger issue; summer tires become hard like plastic below 40 F). But even beyond that, the engine is designed to be track friendly, which means its operating temperature range is very high. It's hard to make an engine that works well both at very low and very high temperatures; one issue is the motor oil in the engine; no oil can handle a temperature range from 0 F (cold winter start) all the way up to 300 F (heavy track use). Track racers use 15W50 , winter street users use 0W40 ; the engine's not really happy until it's over at least 175 F. This means if you want to go racing in the winter you need to warm it up a long while (5-10 minutes ; just watch your oil temp gauge).

Some Porsche engines eat oil, some don't. It's sort of a random chance whether you get an oil eater or not. Not really a huge deal either way, it just means adding a liter or two between oil changes (oh, BTW, the manufacurer recommended 10k or 15k miles between oil changes is bullshit - it's part of what causes blown engines; use 5k intervals!). What *is* annoying is the fucking electronic oil measurement. It refuses to give a reading unless the oil has descended into the oil pan, so you can't do it when the engine has been run at all, you have to let the car sit for 10-30 minutes to get a reading. But you also don't get an accurate reading if the car has sat for a long time, so you can't just do it in the morning after the car sits overnight. Also the car has to be level to get a reading. Very annoying. Each electronic oil tick is half a quart. You should try to keep it between the bottom and middle tick mark - not at the top, keeping it fully topped up will encourage it to burn oil faster. (Porsche officially says that burning up to 1 quart per 400 miles is "by design" ; most cars, like mine, seem to burn about 1 quart per few thousand or so, while a few cars don't burn any at all).

The rear parking sensor thing beeps. It's a bit annoying because it starts beeping when you are like 10 feet away from the thing behind you. I'm always like "really? you're beeping already?". I just discovered recently that it will actually go to a solid tone after beeping and you still have about 6 inches behind you at that point. The parking sensor also can't detect thin or very low objects, so it's better to just use your eyes and ignore it (preferrably buy a car without that option).

The PCM computer thing is really awful for playing music. It does play MP3 CD's and it recognizes the directory structure and will show you folder names, that is handy. There is NO aux input for ipods or whatever. That is fucking retarded. You can get an aftermarket kit to plug in your ipod for $500 or so. The PCM audio can't do simple things like turn off the radio but leave the CD player on (eg. when you eject CDs it starts playing radio). You can't turn on & off the audio separately from the Nav/etc. It also can't pause CDs from the normal controls; I just discovered the other day that you can in fact pause CDs from the multi-function steering wheel by pressing on the volume wheel.

The Bose stereo is pretty fucked up; first of all you must turn off the "adaptive surround" or whatever it's called where they try to create surround sound from stereo. Then you have to tweak all the treble and bass settings drastically to try to get something decent out of it. One quirk is that there are separate settings for FM and CD mode, so you have to be playing in that mode to tweak it, and you have to do all the tweaks twice.

The rear subwoofer and the rear seats are both easily removed if you want to save weight for track days or make more room for cargo or whatever.

The default alignment on the car is very "mild" ; that is, it's very stable, keeps the car straight on the freeway when you let go of the wheel, and resists turning, making the car understeer slightly. This is nice for highway driving and many people will be happy with it. If you want more aggressive turn-in, the easiest way to fix that is just to get a more aggressive alignment; the biggest difference comes from getting more negative camber up front, but you can also get less toe in. You can just ask for the "rest of world performance alignment".

Something about the front alignment (maybe it's the caster ?) means that when you turn all the way to lock, you actually are up on the edges of the tire. Obviously you don't have great control when you're on the edges of the tire and it can feel squirelly, especially if the tires are cold and stiff. They can "scrub" or "crab" and make some crunchy sounds. It's not bad once the tires are warm, but you still probably shouldn't slam on the gas at full lock. It's not a great car for gymkana unless you do some suspension mods. (BTW getting a "performance alignment", as you should do, will mostly fix this)

It's hard to get the transmission back into 1st gear once you've been up to high speeds (after you slow back down). You can solve this by double clutching : put it in neutral, let out the clutch, put the clutch back in, put it in 1st. This is "by design" ; many Porsche drivers just keep the car in 2nd once it gets moving. You can't really use 1st gear once you get moving, which sort of sucks for low speed corners (such as in autocross) because you get into super low revs in 2nd gear.

Almost every part of the car is just held together by little plastic tabs, you know those bits that click together. This is kind of handy because it means everything is very easy to take apart and put back together, but it also means that it's not super solid feeling. The cabin has a lot of squeaks and rattles. You can fix these pretty easily by just popping out the offending piece, putting a bit of foam or felt tape under it and popping it back in. But trying to do that on every piece would drive you mad.

I often think that there's some horrible rattle from a loose piece of the car. In fact every time it has just been something I put in the car, a key chain, a quarter in the change tray, etc. The car gets a lot of vibration from the road because the suspension is pretty stiff, so anything you carry in the car will bounce around quite a lot.

The electricity is on inside the car all the time, even when it's shut off and the doors are locked. This means your cabin accessories plugged into the various 12V DC ports will stay on. Personally I find that annoying, others may like it. It also means the car will drain the battery if it sits a while; people who keep their Porsches in garages all the time usually have to buy a battery tender.

If the battery goes completely dead it's a pain to deal with cuz the hood is power operated. You have to first put jumpers on the fuse panel in the driver's footwell, then use the hood open switch on the key, not the one in the car. (there is also an emergency manual hood open wire, but it requires taking apart the right wheel well, so the fuse panel method is preferred when possible)

Like most cars these days, you can disable the seatbelt warning chime by plugging the seat belt in and out 15 times quickly (this makes the computer think it's broken and thus disabled it). Or you can buy a Durametric (Professional) cable that will let you toggle all the option codes in the chip the way a dealer can. (it's OBDII which I guess is a standard interface that lots of cars are on these days, you can get a generic OBDII device for cheap that can at least read engine failure codes and clear service lights, but you can't toggle options with a generic device).

The Porsche as a daily driver is mostly great. The main problem with it is not the harsh ride or the low cargo space, it's that it wants you to really drive it. It begs you to hammer the throttle and swing through the curves. That's great, right? Well mostly yes, but not always. Some days you just want to turn off your brain for the commute and get to work without incident. Some days it's easier to deal with the traffic and annoyance if you have a sedative in the form of a non-performant car. I found that when I was driving the Nissan Versa rental car - you can't go fast even if you want to, and that actually is very relaxing. The Porsche is like a girlfriend who wants to have nasty hard sex every day. Sounds great, right? Well, after the 20th day in a row of fucking, you kind of just want to watch TV and be left alone, but she's still jumping on you and whipping you with her hair; at that moment you kind of wish you had a fat lazy girlfriend who just eat chips with you and sits on the couch.

3/18/2010

03-18-10 - Physics

Gravity is a force that acts proportionally to the two masses. (let's just assume classical Newtonian gravity is in fact the way the universe works)

People outside of science often want to know "but why?" or "how exactly? what is the mechanism? what carries the force?" . At first this seems like a reasonable question, you don't just want to have these rules, you want to know where they come from, what they mean exactly. But if you think a bit more, it should be clear that these questions are absurd.

Let's say you know the fundamental physical laws. These are expressed as mathematical rules that tell you the behavior of objects. Say for example we lived in a world with only Newtonian dynamics and gravity and that is all the laws. Someone asks "but what *is* gravity exactly?". I ask : How could you ever know? What could there ever be that "is" gravity? If something was faccilitating the force of gravity, there would have to be some description of that thing, some new law to describe it. That would mean some new rule to describe this thing that carried gravity. Then you would ask "well where does this rule for the carrier of gravity come from?" and you would need a new rule. Say you said "gravity is carried by the exchange of gravitons" ; then of course they could ask "why is there a graviton, what makes gravitons exactly, why do they couple in this way?" etc.

The fundamental physical laws cannot be explained by anything else.

That's almost a tautology because that's what I mean by "fundamental" - you take all the behavior of the universe, every little thing, like "I pushed on this rock with 10 pounds of force and it went 5 meters per second". You strip away every single law that can be explained with some other law. You strip and strip and finally you are left with a few laws that cannot be explained by anything else. These are the fundamental laws and there is no "why" or "how" for them. In fact the whole human question of "how" is imprecise; what we really should say is "what simpler physical law can explain this phenomenon?". And at some point there is no more answer to that.

Of course this is assuming that there *is* a fundamental physical law. Most physicists assume that to be true without questioning it, but I wrote here at cbloom.com long ago that in fact the set of physical laws might well be infinite - that is, maybe we will find some day that the electrical and gravitational force can be explained in terms of some new law which also adds some new behaviors at very small scale (if it didn't add new behaviors it would simply be a new expression of the same law and not count), and then maybe that new law is explained in terms of another new law which also adds new behaviors, etc. ad infinitum - a russian doll of physcial laws that never ends. This is possible, and furthermore I contend that it is irrelevant.

There is a certain human need to know "why" the physical laws are as they are, or to know the "absolute" "fundamental" laws - but I don't believe there's really much merit to that at all. What if they do finally work out string theory, and it explains all known phenomena for a while, but then we find that there is a small error in the mass of the Higgs Boson on the order of ten to the minus one billion, which tells us there must be some other physical law that we don't yet know. The fact that string theory then is only a very good model of the universe and not the "absolute law" of the universe changes nothing except our own silly human emotions in response to it (and surely crackpots would rise up and say that since it's not "100% right" then there must be angels and thetans at work).

What if we found laws that explained all phenomena that we know of perfectly. We might well think those laws are the "absolute fundamental" laws of the universe. But how would we ever know? Maybe there are other phenomena that can't be explained by those laws that we simply haven't seen yet. Maybe those other phenomena could *never* be seen! (for example there may be another entire set of particles and processes which have zero coupling to our known matter). The existance of this unexplained phenomena does not reduce the merit of the laws you know, even though they are now "not complete" or "don't describe all of nature".

It's funny to think about how our intuition of "mass" was screwed up by the fact that we evolved on the earth in a high gravity environment where we inherently think of mass as "weight" - eg. something is heavy. There's this thing which I will call K. It's the coefficient of inertia, it's how hard something is to move when you apply a certain force to it. F = K A if you will. Imagine we grew up in outer space with lots of large electrical charges around. If we apply an electric field to two charges of different K, one moves fast and one moves slow, the difference is the constant K. It's a very funny thing that this K, this resistance to changes of motion, is also the coupling to the gravitational field.

3/17/2010

03-17-10 - Porsche 911 Buying Guide

This one goes out to all the internet searchers in the world. I figure I should brain dump to the wild electronic blue.

ADDENDUM : this is really a 997.1 buying guide , not an old-911 buying guide (there are plenty of those around if that's what you want, look elsewhere) ; first a buyer has to learn is that Porsche people don't ever call it a 911, you have to refer to the specific generation you are talking about, and when people say "911" they usually mean the very early models, pre-964.

Buy used. If you have any doubts, remember that Porsche makes $28247 per vehicle which is almost all the difference between new and 1 year old car prices. The cars have a pretty huge price cliff when they transition from "almost new" ( under 10k miles ) to "yuck the seat is soaked with someone else's ass sweat" ( over 10k miles ) ; you will get the best value by buying the most recent year you can get, but with higher than average miles. Also, high mileage on these cars is not actually a terrible thing; the absolutely worst thing you can do for these engines is to let them sit in a garage for months and then a nice day comes around and you go out and thrash it, which is what a lot of the low mile cars do. Also at high mileage it's more likely that a major engine problem (RMS/IMS) would have shown up by now. Also, Porsche buyers are nuts about minor cosmetic damage; that means you should of course buy a car that has some minor cosmetic flaws like paint scratches and bumper rash, and you should get a big discount for it.

Naming : "911" is the line of rear engine cars; it also refers to one specific generation of the line. Future types of 911 are either called the "2001 911" or by their numerical code name, eg. "996". Sometimes people will also just say "the coupe" which is understood to mean "current year 911 carrera".

History : Porsche repeatedly tried to kill the 911 and get traction with mid-engined (914) or front-engined (944,928) cars. Before 1999 the 911 was air-cooled (really "oil cooled" , which it still is, but now the oil is additionally cooled by water, while then the oil was just cooled by air). The old 911 is the longest production running car after the Beetle, which is interesting since it is just a Beetle (this claim is pretty bogus). Most Porsche nuts idolize the early cars which is why you need to ignore the opinion of Porsche nuts. Older Porsches were also more rare and held value better; it was a weird cult car; the modern 996+ car is not like that any more at all, it is a mass market easy to drive car with huge depreciation; some of the old Porsche nuts don't understand or refuse to recognize that change.

1999 introduced the 996 (actually 98 but you will never see a 98). The current car (2010) is largely the same as that car - same platform, same basic engine. All modern 911's are ECU controlled with air sensors and drive-by-wire electronic throttle and all that kind of shit that makes them just like a BMW or something, but they do a good job of feeling simple and pure and manual. In general the advantage of the 911 over competing cars is just that it is an incredibly well sorted car which is well designed for people who like to drive. The power is not amazing by modern standards, but the way it puts down the power is fantastic - it has less drivetrain loss than average, less throttle delay, you sit low, it's just the right kind of stiff, the ergonomics are great for real driving. They have also kept the weight pretty low by modern standards - you can get a base 911 for under 3000 pounds with all the AC and carpets intact , most competitors are around 3500.

The lineup of variants is basically the same for all the years (though they're fickle about which variant is available in what year), so first I'll go over the variants :

Basically you pick each part of the car and put them together, so you pick :
roof type : {Carrera/Cabriolet/Targa}
+ engine type : {S , not S }
+ drive type : {2wd or 4wd}
Put them together, so eg. Carrera 4wd S = C4S

Note : the Turbo, GT3, and GT2 are a separate car that is not a variant of this; I will address them at the end.

Carrera = Coupe - this is the good one. Lighter, more rigid.
Cabriolet = Convertible. This is for very fat bankers. If you want a convertible car there are many better choices.
Targa = glass top ; at some point the meaning of "Targa" changed from a T-top type of thing to the new variant which is a huge glass sliding roof. The glass roof is heavy and it's weight in a bad place - up high - which hurts handling, but I must say it is absolutely glorious inside there; it's all the advantages of a convertible with none of the disadvantages. It is a painfully uncool fat banker's car, but it's also really fucking great to be in. If you don't care about what other people think of you (which you clearly don't if you are considering a 911), this is a great car, though they are a bit rare thus hard to find cheap used. The rear glass also opens like a hatch back, which is especially great if you delete the rear seats (which you should).

911 "base" (Carrera/Cabriolet/Targa) - don't buy this, it's for retards and fat (but poor) bankers.

911 "S" - the "S" is a huge upgrade over the base for not much money, it's obligatory. You get a bigger engine, bigger brakes, bigger rims, better suspension, standard PASM and other sport options, better air intake, better exhaust, it's just upgrades all around, you really cannot pass up the "S".

"2" vs "4" (referred to as C2S or C4S for example for the Carrera 2WD S) - this was the hardest buying decision for me. The "4" gets a few bonuses, like pre-pressurized brake lines and larger brakes from the turbo, as well as the wide body and larger tires. The "4" also gets 100-150 more pounds of weight (though most of that is in the front which is not a terrible thing), and a bit of understeer and a slightly more numb feeling. In the real world, the 4 is actually faster in almost all scenarios in the hands of 99.99% of drivers, however the 2 *feels* faster because it is lighter and there's less drivetrain loss and lag - you get the engine in the back directly driving the wheels in the back which makes it super responsive. If you do get a "4" you should go for the 2009+ variant. I think 90% of the fat banker buyers should be getting the 4, because it is more practical and safer for the shitty driver; serious enthusiasts probably should get the 2 for the tighter feel and wacky tail-out fun times.

Now we'll go over the years. The 996/997 has been basically the same but tweaked over the years and there are some major differences :

The 996 generally (1998-2004) is the ugly stepchild of recent Porsches. It's very ugly, the front end is identical to the old Boxster front end, the interior feels to me just like a Dodge Stealth, and the steering feel and throttle and brakes and everything just feel a lot worse than the 997 (even though they are fundamentally the same).

996.1 : 1998-2001 : 3.4L engine (300 HP) : do not buy these. They are known for engine failures (RMS is just a small leak problem while IMS is a major engine failure) and the 2002 has a lot of upgrades.

996.2 : 2002-2004 : 3.6L (320 HP) : adds variable valve timing, better aero and stiffer chassis, lots of little improvements over 996.1 , mostly sorts out the engine failures. This car is much hated, though it's actually not a bad car, so it should be available for under $30k , and if you do a bunch of mods it could be a nice car. I still don't recommend it, because if you are thinking of a 996 you can get a GT3 or Turbo for so cheap (see later).

997.1 : 2006-2008 : 3.8L (S) (360 HP) : lots of improvements over the 996.2 - bigger engine, bigger rims, brakes, handling more sorted, clutch feels a lot better, better steering feel. Adds PASM adjustable sport suspension (you are getting an S of course so this is standard) which is very good. Oversteer is extremely well sorted now through suspension and alignment and tire sizing, if anything the cars slightly understeer out of the factory. PSM (electronic traction control) is very good - keeps you going straight without interfering too much.

997.2 : 2009-2010 : 3.8L DFI (385 HP) ; big engine improvement, moves to direct injection which gives more power, more economy, more torque. Fixes the IMS problems by having no IMS. The biggest improvements however are in the automatic (PDK) which gets the 7-speed double clutch from the Turbo, and the C4S which gets the electronic 4WD from the Turbo (PTM) (the old 4S had mechanical viscous clutch 4WD which is less effective as a 4WD but maybe is better at making the car feel like a RWD). If you're considering a C4S, especially an automatic, the 997.2 is a big win over the 997.1 ; some indications that the early engines are not sorted (eating a lot of oil); they also have tuned the suspension to be a bit more dead out of the factory, so the 997.2 is even less tail-happy and a bit more numb than the 997.1 , but you can easily have this fixed aftermarket if you want more snap. Also if you want an automatic (PDK), you must get a 997.2 with SC (Sports Chrono), the earlier automatics suck (tiptronic), and normally SC is worthless, but with PDK it is a must. Another huge change for 997.2 is that an LSD (Limitted Slip Differential) is now standard (on S/PASM cars anyway, which you are of course getting), so that's a nice bonus. Right now these cars are hard to find cheap used, however you can still find unsold 2009's at dealers and they are offering big deals on them.

2011-2012 will bring a major overhaul and is currently codenamed the 991.

Buying options on the 997 : (a general note : options may cost a huge amount when you buy the car new, but on used cars they are worth almost $0 ; do not pay more for a car because of options ; if someone tries to tell you the car is worth $5k more because it has PCCB, tell him you're sorry he wasted his money on options that are worth nothing in resale).

Sports Chrono : must get this with PDK on the 997.2 (it gives you a faster-shifting more aggressive mode for the automatic; must have; also gives launch control; those cars also have the ability to switch the 4WD to 2WD on C4S models which is pretty cool). On other cars all this does is give you a "sport" button which remaps the throttle to be a steeper curve. Basically it just makes e-gas have a *= 2 factor on it, so that 100% trottle comes on quicker. That's retarded and actually hurts lap times because it gives you less fine throttle control.

PASM : this will be standard (you are buying an S right) and it's good.

19" rims : this will be standard (you are buying an S right) but it's not good. 18" rims are preferred - they are much lighter, let you buy cheaper tires, give a softer more comfortable ride, and are at worst neutral on performance (and might help).

Bose stereo : terrible, avoid. Adds weight and might make the audio quality worse. The car is so loud the stereo quality doesn't matter anyway.

Nav & PCM : terrible, avoid. Much improvement in 997.2 (PCM v3) , so if you need electronics get a 997.2 ; otherwise just use an aftermarket nav system, or pull over and use your iPhone.

Parking assist, dimming rear view mirror, headlight washers, TPMS, etc. etc. - none of this is good, avoid it all. Even if you think you want it, it seriously does not work well, avoid. Like it's not just overpriced, but having it is worse than not having it.

X51 package : performance package that boosts HP about +25 ; this is a very nice kit (exhaust, headers, intake, valve flaps, cylinder heads), which you should definitely get on a used car if you can find it cheap. Definitely not worth the $$ on a new car (but you aren't buying new cuz that's ridiculous with a Porsche). The OEM X51 kit is way way better than any aftermarket 3rd party intake/exhaust mods. (ADDENDUM : x51 is really a very good deal in a used car and you should get it if possible; it also includes a lot of upgrades to make the engine more reliable and handle track G's better, such as a better oil pan and extra oil scavenge pump, better cooling, better cylinder heads, etc. The value is really not about the small power boost, it's that the car can handle heavy use without blowing up.)

Tiptronic (automatic) : avoid because you're not a fat boring lazy banker. Also because the PDK in the newer cars makes this look like stone age dogshit, so these cars will suffer badly in the resale market. (If you do want an automatic, that's a good reason to get a 997.2 instead of a 997.1).

Ceramic brakes (PCCB) : this is a pretty "meh" option ; definitely don't pay more for it. In theory it's cool because they basically don't wear or fade with use, and for a street car they might be a cool thing, but they do crack under very high heat track use, and when they crack they cost $10k to replace, so maybe not an awesome thing for a car you will track (and if you don't track, then normal pads are fine).

Okay, now on to the GT3, GT2 and Turbo . These are significantly different than the other cars, even though they obviously look very similar. They have different suspension, engines, intake & oil coling, and body panels. The Turbo generally has the latest technology, which then moves down to the base cars in the next model. Most significantly and what sets these apart as a family from the other 911 cars is that they are actually built on a different engine. All the 996/997 GT3/GT2/Turbo cars are based on the same engine : the Porsche GT1 race engine from 1996-1999. This engine was brought to the retail 911 for the 996 GT3 in 2003-2004 ; even though it is a 3.6L flat 6 like the normal 911, it is a totally different engine that revs higher, can take more heat, and has a true dry sump for oil cooling which lets it sustain over 1 lateral g without drying out. (technically it's the M64/GT1 engine, while the main line of 911's has an M96/M97 engine). There's not as big of a difference between 996 and 997 variants of these cars as there is with the base cars, because these cars have stayed with the same GT1-based engine the whole time (eg. a 996 GT3 vs. a 997 GT3 is not a huge difference, just body style, some suspension degrades, addition of electronic nannies, stuff like that). The main difference in 996-997 on these cars is the body style and the tech doodads. The GT named cars are quite clear - they are named by the race class they are designed for ; GT1 is the highest race class, then GT2, then GT3. (* addendum : the 997.2 Turbo is now the DFI "9A1" engine as the 997.2 line, just with a Turbo stuck on it; the 997 GT3 is still the M64/GT1 engine, and the 996 and 997.1 Turbo is also an M64/GT1 engine).

Turbo : (often called the TT, as in "996 TT") a bit like a C4S with the better engine & a turbo stuck on it. Also the latest tech doodads to control all that power, fancy brakes and suspension and all that. Cabin is just like a C4S. Obviously an awesome machine, but I really don't like this car; it's not ideal for the track, and the problem is that it's not fun until you are over 100 miles an hour, which makes it pretty impractical in the US. This car is made for blazing on the autobahn. On the plus side, 996 Turbos can be had for quite cheap now.

GT3 : RWD, naturally aspirated, a bit like a stripped C2S with X51 but really a whole other beast. Stripped of weight and unnecessary doodads like computers. The lower tighter suspension makes this not a very practical street car. Also all that power in the rear-engine RWD layout makes this car much more tail happy than the C2S (which is tweaked to soften up the oversteer tendencies). The 997 GT3 has a lot of the modern tech doodads (PSM, PASM) which make it more liveable on the street. The 996 GT3 is the last really raw non-teched-out Porsche, and they can be had used for quite cheap, this is my recommendation for a real raw track car. vs a C2S the car has a much lower rotating weight due to light weight fly wheel, lighter rims and brakes. This is a great car, and a tempting buy because it basically takes a 911 and does all the mods to it that you wish you could do if you're a serious driver, but if you are buying a car that's mainly for the track and a bit uncomfortable on the road, there are better options (like a Cayman or Lotus if you just want fun, or a Nissan GTR if you really want speed, or a Corvette for RWD madness, or lots of other things that are all much cheaper). The only advantage of the 911 over its rivals is its practicality, and the GT3 throws that away.

GT2 : this is a GT3 with a turbo stuck on it, or a Turbo that's been converted to RWD and stripped of weight and made stiffer. This is a ridiculous dangerous car and you should not buy one. This is the one car that really carries on the 911 legacy of being an out of control killer .

Of course there are various other variants (GT3 RS, Cup, etc. etc.) but you aren't going to buy them so whatever. You should be able to get a 996 Turbo or GT3 in good shape for $50-60k now.

Some good guides :

Deutsch Nine history of the models
PorscheDoc's 996 buying guide
Grant Neal's 997 in detail

If you are shopping, do not trust any dealer or private guy about what variant or options a car has. Ask them for a photocopy of the first page of the service book, which will have the VIN and option codes. Then enter them here :

VIN code guide
VIN Decoder
Option code Guide
Option code decoder

ADDENDUM :

Most people will recommend a PPI ("Pre-Purchase Inspection") when buying a used Porsche. I got a PPI and it was pretty worthless, because Porsche mechanics use it as a way to make free money from people who don't know Porsches, and they don't really care if you get a good car or not. So, I'll tell you what you should look for. I do recommend that you get a PPI, but rather than letting the mechanic just check the car out for you, ask for this specific information :

1. DME readout of CELs and overrevs. They scan this with an OBDII reader and should get a page of information. Have them send you an exact copy of what they get from reader, not just their opinion on it. What you want to see is no overrevs in range 4 or 5 (range 1-2 are okay, range 3 is iffy). (you can also buy your own OBD reader and do this yourself)

2. Compression test - this is the only way to tell if there's damage inside the engine. You want the compression number on each cylinder; this should only cost about $100, any more and they are ripping you off. In particular if the engine has been run too hot or oil-starved, the cylinder liners can be gauged or warped out of round, and then they will not seal properly and you'll see reduced compression on that cylinder.

3. Check for oil weep on the engine, particularly in the front. Obviously a seller can hide this by cleaning the engine, but they usually don't. Obvious oil weep near the engine-tranny mounting is a sure sign of RMS leak.

4. Check for bent suspension parts, cracked coil packs, and rust on the exhaust. These are all easily visible on the bottom of the car when its lifted.

ADDENDUM 3-1-2011 :

if for some reason I was buying a new 997 right now it would be a GTS for sure. The GTS is a 997.2 (9A1 DFI), it's RWD but with the wide body from the 4WD car, has the x51 Powerkit option standard for +20 hp, alcantara interior bits, some aero body bits, a nice rear-seat delete option like the GT3. In non-America it also comes with sunroof delete, but in the US that's not an option. Basically the GTS is a good set of options, and you actually get them at a discount, unlike the retarded "Sport Classic" or "Speedster" which are just 997's with a ridiculous markup for no reason. Personally I wish they would have done a bit more weight reduction for the GTS, but instead of a real lightweight enthusiast car, what they did was make the GTS the car that all 997's should have been all along. At 408 HP it's actually got the power to match its rivals. (the only thing not to like about the GTS are the absurd center-lock wheels, but you can opt for normal lugs).

3/16/2010

03-16-10 - Video Reviews

Watching a bit of "The Inbetweneers" (BBC). Meh, it's passable so far as TV comedy goes (way better than Modern Family for example, but not as good as The League). But I'm really sick of the whole nerdy loser boys in high school genre that was made so big by Judd Apatow and that fat guy who's always in his movies and Michael Cera and all that. Uh, yeah we get it, high school boys are horny and terrible with girls and they get drunk and puke cuz they drank too much ha ha ha. And then they learn lessons about life aww look how fast they grow up. I find the whole need to relive childhood or laugh at our childhood very juvenile and tedious.

"Banlieue 13" (District 13/ District B13 - very bad translated titles) is fucking great. It's silly and frenetic and imaginative, everything that action movies should be. It's a bit strange that it completely flopped in the US, since Luc Besson is pretty well known and it's a fucking great movie, but maybe it's just a bit too French ; you won't really get how close to reality it is unless you've followed a little bit of French news in the last 10 years, and then it's got the French goofiness with silly cars and parkour and banging soundtrack. It's the best Parkour in a movie for sure

Feynman's "Messenger Lectures" - ( "The Character of Physical Law" ) ; this is available from Bill Gates but that's a fucking stupid ploy to make you use stupid Silverlight , just download the torrents. God these are so good. You don't really need any strong physics or math background, these are aimed at college students, but Feynman always aims high and doesn't talk down, so you do have to be smart to follow. It's so satisfying and invigorating to watch something that's actually intellectually stimulating, it makes my brain start firing. I wish I could watch only things like this all the time.

High Stakes Poker is back on now and as usual is amazing. It sucks they ditched Aj Benza in the booth though, obviously he was just a talking bag of puss, but Gabe can't work solo, commentary is always a team; Gabe Kaplan alone is almost like John Madden alone - not good.

Full Tilt "Durrr" Million Dollar Challenge is also interesting. The first 6 episodes are a bit boring because he's playing against pretty weak opponents, but in episode 7 Zigmund sits in and they play some insane PLO. It's also funny to see that Durrr was playing these games all day long while Isildur was tearing him up every night; Durrr wound up losing around $3M to Isildur very quickly. It's very interesting to watch Durrr play, partly because he gives tons of action which makes things interesting, but also because he plays so weird that it creates all kinds of new situations and thought patterns. Some of the things that he does are just so exploitable, but people are afraid to do the right thing against him. For example, Durrr will often bluff when it's clear that the opponent has a very very strong hand; people think "he knows I have a strong hand, so he must not be bluffing" and fold, when really they need to just open their game way up against him. Similarly Durrr rarely folds to 3-bets. The correct counter play to that is to 3-bet a lot of good hands, and then be very happy to play big pots with anything decent post flop. The problem is people are so afraid of variance and big pots that they try to pot control and wind up giving Durrr the pot.

3/12/2010

03-12-10 - Hot Douche Gasses

Aftermarket exhaust products for cars really make me angry.

The car company engineers do a really amazing job of making modern cars high performance and also low pollution. Modern catalytic converters and variable valve timing and computer controlled feedback loops with air sensors are real technological marvels. And yet some redneck with a blow torch thinks he can improve it by cutting some holes or putting on fatter pipes.

First of all, "freer flowing" or "reduced backpressure" is rarely a good thing to do to your car. Putting on fatter pipes rarely helps performance much, and if it does help it probably helps only at high RPM at the cost of performance at low RPM. The reasons are rather complicated, but basically the combustion chamber is designed to have just the right amount of exhaust exit flow. What that should be exactly depends on the car and the RPM, but typically you want the flow of exhaust to create a vacuum to help evict the cylinder after combustion; some engines also want the exhaust to stop moving and create a sort of wall as the exhaust valve closes and new air and fuel comes in. Changing the diameter of pipes can screw up these pressures, and the result is often torque loss at low RPM. At high RPM it's a simpler situation because the cylinder is exploding so many times and making so much gas you just want to get it out as quickly as possible.

( see for example 1 or 2 )

The latest trend in car mods is to do a full or partial bypass, just cutting a hole before the cats that dumps exhaust straight out the back of the car, sometimes on a switch, sometimes not. Even if it did help, it would still be dickish to ignore your fucking up the air quality just because you need 5 more horse power. Some of the modders are just confused because cats actually did used to hurt performance back in 1950 or whenever they first came out, but modern engines and cats work marvelously (very large high reving engines like race cars can still see a small benefit from not using them).

But worst of all its just super douchey to make your car louder. I personally love to hear my engine, and actually I wouldn't mind it being a bit louder, but I want it loud just for me in the cabin, not for the whole rest of the world (and I'd like it on a switch too, so I can have quiet mode and loud mode). People who amp their exhaust are obviously doing it intentionally to be obnoxious. Hey pedestrians walking on the road, listen to this! Yeah fuck you for trying to have a conversation, look over here at my great fucking car! My penis might be tiny but my exhaust is huge, oh yeah!

In other douchey car news, my car has xenon head lights and I hate it. I'm kind of tempted to "smoke" the headlights just to reduce their brightness a bit. I can see people in front of me get annoyed when I pull up behind them. Of course I'm quite low so my sin is only one tenth the sin of a high-carriage car with xenons, but I still hate to be that guy. (In other "that guy" news, I used the carpool lane through Bellevue for the first time. I have no regrets).

ADDENDUM : another awesome example is what people do for air intakes on the 911. There are a few varieties of stupidity here.

One popular mod is like Fabspeed smooth red tube . The OEM air intake tube has a few very clever features; for one it is baffled to reduce noise and drag (contrary to Fab's claims, rough surfaces actually make less air resistance than smooth in many cases; see eg. the golf ball; the rough surface will create turbulence while the smooth surface will build up a big sticky boundary layer; of course the reality and details are incredibly complex and it's hard to say what's better).

But the main feature is the little tube that comes off the OEM intake. That little tube is a Helmholtz Resonator, which vibrates at a certain frequency as air passes over it (just like blowing over the mouth of a bottle). It's tuned so that that resonance cancels out the vibration resonance of the air coming in at low RPM. At high RPM they go out of tune and no longer cancel. The result is the car is nice and quiet at low RPM and then howls at high RPM. (see here for example)

The main thing that the air intake mods do is just break those clever features. Of course they could get the same effect by just putting a piece of duct tape over the resonator tube. And hell while you're at it stick a playing card in your air intake so you get some extra noise.

To some extent the whole "cold air intake" thing is another left-over from the 50's modder days when manufacturers actually got things wrong. Back then you had carburated cars taking big gulps of hot air from inside the engine bay. If you just stuck on a tube that ran the air intake out to the outside of the car, it would let it breathe cold air, which increases engine efficiency (cold air is denser in oxygen mainly). Modern cars all route cold air to the air intake very efficiently, either using their own tubes or simply through controlling the air flow patterns when the car moves. Modern cars are not designed to breathe correctly when they are sitting still (eg. on a dyno) which is part of how these nonsense tuners can claim gains on the dyno (the main way is just faking the results, eg. running the "before" test on a cold engine and the "after" when it's all warmed up).

Anyway, it's a cheap way to get a little sound, which is what buyers really want. The funny thing is they can't admit that, everyone has to pretend it's about performance. Both buyers and sellers happilly go along with the fraud, sellers post fake dynos and buyers claim they "feel a difference". (kind of like how ballet is really about seeing some hot half-naked people, but everyone has to pretend it's about the story).

Another popular mod is a BMC or K&N cotton oiled air filter. These filters do seem to in fact let in a tiny bit more air (maybe 1% more) which might increase power, but they also let in more particles, which might hurt engine life. It's almost a tautology that more air flow = less filtration, unless you actually increase the surface area of the filter. (see here for example). There's also some indication that those filters let in more air when clean, but less air when dirty (compared to a standard OEM paper filter). Many ricers want to use BMC or K&N filters because "racers use them". It's a completely different situation for a few reasons - 1. race engines have a lifetime of less than 1000 hours, so longevity isn't really a big concern, 2. they actually need as much air as they can get, most road engines are not air limitted (or rather, they are, but the problem is not availability of air at the intake, it's driving it into the cylinders, and the solution to that is forced induction, not filters), and perhaps most importantly 3. the racers are sponsored by those filter companies.

03-12-10 - Friday Linkage

YouTube - SeatofEmpire's Channel - cool documentary about Seattle as the manifestation of American capitalist-military exploitation. This is just previews but they are fun and well made.

YouTube - Heel & Toe - best video I've seen of heel & toe technique. For racing, through a corner you want to go from braking hard directly into gassing hard with no coast time in between. It's actually much more important to do this in a turbo car or any car with a high narrow power band - in that case you need to actually shift while you go through the corner so that you keep the car in the power band, you can't wait and shift right before you power out at the apex because the RPM will have dipped too low.

YouTube - minutegongcoughs's Channel - good old music

Wheel Alignment A Short Course
Caster, Camber, Toe
The new 911 has some weird shit going on with the alignment to keep the tail from sliding out. I had no idea what camber or toe were or how they affected vehicle dynamics; crazy interesting stuff. Some people who race 911's make the alignment more neutral so that they oversteer more.

John Sizemore - Weird_Weird_Science on Dailymotion - awesome awesome videos of extreme zooms on materials all the way down to the atomic scale. How do they do this?.

Alan Turing - Wikipedia, the free encyclopedia
Turing biography
Of course we all know Turing from the Turing Machine and the Turing Test and his thoughts on AI and Computability, but he also did work on chemical pattern formation, abstract mathematics, and code breaking. I had no idea about his personal life story, though, which is quite shocking.

NOVA Sputnik Declassified A Tainted Legacy PBS - speaking of scientist biographies

Popular Science - Google Books Old Popular Science magazine is fucking awesome. Just the other day I was oiling up my bicycle and was wondering why we don't oil up cars the same way (oiling all the hinges and gears and such) - well of course in the past they did!

Pelican Technical Bulletin All About Motor Oil
A little bit archaic but pretty interesting.

Brian Beckman's Physics of Racing Series - really this is "Physics of Driving" ; it's mainly a review of the basic forces involved in cars. Pretty good stuff. I guess BB is at MS Research and may have contributed to "Forza 2" , though if it's anything like the typical MS Research video game "contribution" it means he sent them a link to his articles and they ignored it. (BTW just found this nice series of videos on basic car physics made by a dry narrator in a Seven).

Aircrew Survival Equipmentman 2 - Aviation theories and other practices - what to do when your plane goes down. Includes how to maintain or fabricate your parachute.

3/10/2010

03-10-10 - Distortion Measure

What are the things we might put in an ideal distortion measure? This is going to be rather stream of conscious rambling, so beware. Our goal is to make output that "looks like" the input, and also that just looks "good". Most of what I talk about will assume that you are running "D" on something like 4x4 or 8x8 blocks and comparing it to "D" on other blocks, but of course it could be run on a gaussian windowed patch, just some way of localizing distortion on a region.

I'm going to ignore the more "macroscopic" issues of which frame is more important than another frame, or even which object within a frame is more important - those are very important issues I'm sure, but they can be added on later, and are beyond the scope of current research anyway. I want to talk about the microscopic local distortion rating D. The key thing is that the numerical value of D assigns a way to "score" one distortion against another. This not only lets you choose the way your error looks on a given block (choosing the one with lowest score obviously), it also determines how your bits are allocated around the frame in an R/D framework (bits will go to places that D says are more important).

It should be intuitively obvious that just using D = SSD or SAD is very crude and badly broken. One pixel step of numerical error clearly has very different importance depending on where it is. How might we do better ?

1. Value Error. Obviously the plain old metric of "output value - input value" is useful even just as a sanity check and regularizer ; it's the background distortion metric that you will then add your other biasing factors to. All other things being equal you do want output pixels to exactly match input pixels. But even here there's a funny issue of what measure you use. Probably something in the L-N norms, (L1 = SAD, L2 = SSD). The standard old world metric is L2, because if you optimize for D = L2, then you will minimize your MSE and maximize your PSNR, which is the goal of old literature.

The L-N norms behave differently in the way they rate one error vs another. The higher N is, the more importance it puts on the largest error. eg. L-infinity only cares about the largest error. L-2 cares more about big errors than small ones. That is, L2 makes it better to change 100->99 than 1->0. Obviously you could also do hybrid things like use L1 and then add a penalty term for the largest error if you think minimizing the maximum error is important. I believe that *where* the error occurs is more important than what its value is, as we will discuss later.

2. DC Preservation. Changes in DC are very noticeable. Particularly in video, the eye is usually tracking mainly one or two foreground objects; what the means is that most of the frame we are only seeing with our secondary vision (I don't know a good term for this, it's not exactly peripheral vision since it's right in front of you, but it's not what your brain is focused on, so you see it at way lower detail). All this stuff that we see with secondary vision we are only seeing the gross properties of it, and one of those is the DC. Another issue is that if a bunch of blocks in the source have the same DC, and you change one of them in the output, that is sorely noticeable.

I'm not sure if it's most important to preserve the median or the mean or what exactly. Usually people preserve the mean, but there are certainly cases where that can screw you up. eg. if you have a big field of {80} with a single pixel spike on it, you want to preserve that background {80} everywhere no matter what the spike does in the output. eg. {80,80,255,80,80} -> {80,80,240,80,80} is better than making it go -> {83,83,240,83,83} even though the latter has better mean preservation.

3. Edge Preservation. Hard edges, especially long straight lines or smooth curves, are very visible to humans and any artifact in them stands out. The importance of edges varies though; it has something to do with the length of the edge (longer edges are more major visual features) and with the contrast range of the region around the edge : eg. an edge that separates two very smooth sections is super visible, but an edge that's one of many in a bunch of detail is less important (preserving the detail there is important, but the exact shape of the edge is not). eg. a patch of grass or leaves might have tons of edges, but their exact shape is not crucial. An image of hardwood floor has tons of long straight parallel edges and preserving those exactly is very important. The border between objects is typically very important.

Obviously there's the issue of keeping the edges that were in the original and also the issue of not making new edges that weren't in the original. eg. introducing edges at block boundaries or from ringing artifacts or whatever. As with edge preservation, the badness of these introduces edges depends on the neighborhood - it's much worse to make them in a smooth patch than once that's already noisy. (in fact in a noisy patch, ringing artifacts are sort of what you want, which is why JPEG can look better than naive wavelet coders on noisy data).

4. Smooth -> Smooth (and Flat -> Flat). Changing smooth input to not smooth is very bad. Old coders failed hard on this by making block boundaries. Most new coders now handle this easily inherently either because they are wavelet or use unblocking or something. There are still some tricky cases though, such as if you have a smooth ramp with a bit of gaussian noise speckle added to it. Visually the eye still sees this as "smooth ramp" (in fact if you squint your eyes the noise speckly goes away completely). It's very important for the output to preserve this underlying smooth ramp; many good modern coders see the noise speckle as "detail" that should be preserved and wind up screwing up the smooth ramp.

5. Detail/Energy Preservation. The eye is very sensitive to whether a region is "noisy" or "detailed", much more so than exactly what that detail is. Some of the JPEG style "threshold of visibility" stuff is misleading because it makes you think the eye is not sensitive to high frequency shapes - true, but you do see that there's "something" there. The usual solution to this is to try to preserve the amount of high frequency energy in a block.

There are various sub-cases of this. There's true noise (or real life video that's very similar to true noise) in which case the exact pixel values don't matter much at all as long as the frequency spectrum and distribution of the noise is reproduced. There's detail that is pretty close to noise, like tree leaves, grass, water, where again the exact pixels are not very important as long as the character of the source is preserved. Then there's "false noise" ; things like human hair or burlap or bricks can look a lot like noise to naive analysis metrics, but are in fact patterned texture in which case messing up the pattern is very visible.

There are two issues here - obviously there's trying to match the source, but there's also the issue of matching your neighbors. If you have a bunch of neighboring source blocks with a certain amount of energy, you want to reproduce that same patch in the output - you don't want to have single blocks with very different energy, because they will stand out. Block energy is almost like DC level in this way.

6. Dynamic range / sdev Preservation. Of course related to previous metrics, but you can definitely see when the dynamic range of a region changes. On an edge it's very easy to see if a high contrast edge becomes lower contrast. Also in noise/detail areas the main things you notice are the DC, the amount of noise, and the range of the noise. One reason its so visible is because of optical fusion and affects on DC brightness. That is, if you remove the bright specks from a background it makes the whole region look darker. Because of gamma correction, {80,120} is not the same brightness as {100,100}. Now theoretically you could do gamma-corrected DC preservation checks, but there are difficulties in trying to be gamma correct in your error metrics since the gamma remapping sort of does what you want in terms of making changes of dark values relatively more important; maybe you could do gamma-correct DC preservation and then scale it back using gamma to correct for that.

It's unclear to me whether the important thing is the absolute [low,high] range, or the statistical width [mean-sdev,mean+sdev]. Another option would be to sort the values from lowest to highest and look at the distribution; the middle is the median, then you have the low and high tails on each side; you sort of want to preserve the shape of that distribution. For example the input might have the high values in a kind of gaussian falloff tail with most values near median and fewer as it gets higher; then the output should have a similar distribution, but exactly matching the high value is not important. The same block might have all of its low values at exactly 0 ; in that case the output should also have those values at exactly 0.


Whatever all the final factors are, you are left with how to scale them and combine them. There are two issues on scaling : power and coefficient. Basically you're going to combine the sub-distortions something like this :


D = Sum_n { Cn * Dn^Pn }

Dn = distortion sub type n
Cn = coefficient n
Pn = power n

The power Pn lets you change the units that Dn are measured in; it lets you change how large values of Dn contribute vs. how small values contribute. The cofficient Cn obviously just overall scales the importance of each Dn vs. the other D's.

It's actually not that hard to come up with a good set of candidate distortion terms like I did above, the problem is once you have them (the various Dn) - what are the Cn and Pn to combine them?

3/08/2010

03-08-10 - Unrighteous Indignation

One of the things that makes me angriest in the world are people who think they're saintly and have the rules and righteousness behind them, but are just fucking wrong. There are a few varieties of this, one are just people who are purely incorrect and have the facts wrong, the other is people who choose to dickishly try to enfore the strict letter of the law on one behavior (that they don't like), but hypocritically are slack about it in other ways (that they do like), and really even if they are in fact always perfect rule abiders themselves, to not be flexible about when a rule is actually important or not is fucking dickish.

Riding a bike in the city you run into this kind of thing constantly. Every pedestrian and driver thinks they are an expert about bicycle traffic laws and they are eager to inform you with their incorrect and dickish "knowledge". It's a very Seattle/Scandinavian way of being an ignorant dick - to cloak your sour misanthropy in condescension and enforcement of rules. (I'd much rather be yelled at by a New York stereotype who says "hey! wassa matta you, I'm walkin' here!" or whatever)

About a week ago we were riding along. We stopped at a stop sign and waited for the cars to pass. On the other side of the road I noticed a woman standing at the corner talking on her cell phone, just sort of ambling, certainly not making a move to cross. So we take off and ride through the intersection. As we're passing, the woman says "you know pedestrians still have the right of way".

Eh... as usual it took me a minute to register; I thought she was just talking on her cell phone call, but about half way down the block I realized it was directed at us. It's such badly placed hate; if you want to yell out hate at cyclists you should at least pick a time when the cyclists are doing something remotely wrong, there are plenty of dick cyclists out there, I'm sure you won't have to wait long to take out your sour vile rotten soul on them. And, you know, learn to fucking be a pedestrian, and get off your damn cell phone too.

There are frequent difficulties riding down the road. Many cars seem to think that cyclists are legally obligated to make way, or to ride as far to the right as possible. This is not correct. Cyclists are legally required to ride as far to the right as is *safe* for them to do so, except when turning. Generally to be safe, you should ride about three feet to the left of a parked car. People in parked cars do not watch out as they should, and as a cyclist you need to pre-reserve your swerve space, since you can not know when you will need it.

A large amount of the problem with this is that drivers are just so fucking stupid about what rules are important to follow exactly and which are okay to be flexible on. Often I will be riding along and some car will come up behind me and just stay behind me - even though the other side of the road is completely empty and all he has to do is pull out a little bit and go around. People in most of the world have no problem with this and do it constantly, but here in the US there is this bizarre unwillingness to cross the yellow line; oh god forbid I pull out across the yellow line a little bit, it is holy and inviolable. Hell, a lot of the times it's actually a road with two or more lanes going the same direction, and the driver won't even change into the left lane. Nah, I want to be in the right lane, fuck this fucking cyclist cock blocking me, I couldn't possibly be bothered to just move to the left lane. Often roads have a wide turn lane down the middle that noone is using ; hello, the fucking turn lane is an ideal way to pass safely around cyclists, you only need to go about halfway out into it, people going the other direction can also come out halfway into it, everyone is happy, fucking don't be a retard.

The other day N and I were riding along on a narrow busy multi-lane street which is nevertheless a "bike corridor" here in hated bad-street Seattle. This street has parked cars, narrow lanes, and basically no space for bikes, so the only safe way to ride is to take a lane, which is not a big deal for sane people because cars have another lane going the same direction to just go around you. So since we're taking the lane we ride side by side, which makes it clearer that we are just taking the lane. (one of the most important aspects of riding safely as a cyclist is to make it very clear to cars what you are doing - you should not make timid or sudden moves, when you are taking a lane or turning left, it's good to clearly telegraph your intentions, then take the whole lane, or pull out all the way to the left; if someone is opening their door and you have to swerve around, swerve way around so you're more visible, etc.). Anyway, some self-righteous ignorant cock of a driver chose this moment to pull over and roll down his window and yell at us about how it was illegal to ride two abreast. Umm, first of all, no it's not, it's illegal to ride *more* than two abreast. Second of all, your situational awareness and ideas about what rules are important and when is completely fucked; even *if* that was the law, it would be a retarded time to yell at us, since we couldn't really get out of the way anyway even if we rode single file (and if you tried to get in a tight lane with us riding single file you would be a dangerous cock).

Anyway this all happened a while ago and I wasn't going to even write about it because it just makes me sad how fucking stupid and mean people are. It's not just that they're stupid, it's that they are almost intentionally stupid in a self-righteous selfish way; like they choose willfully to not actually know the law, or to not think about the other person's risks and rewards in a certain situation, they only want to follow some dumb rule without thinking situationally, and they want to be fucking right and lord their rightness over others. But I was researching again (in vain) to find a lawyer that can handle Oregon speeding tickets, and I stumbled across so many of this type of comment in web forums and blogs :


"You will lose your license for a while. Hopefully"

"You could lose your license for that and I hope you do."

"HOw do you figure going 102-106 mph isn�t reckless? Are you trying to kill somebody or are you just too immature to deserve a license?"

etc. you see this kind of thing posted all over the internet - people being holier than thou rule-touting fuckers. Do you all have no concept of what is actually dangerous? Going 100 on a freeway is really not dangerous at all (assuming low traffic, and assuming your car is in good kip - you check your tires often, have good brakes, and you have a car that is stable and maneuverable at speed). I dunno what freeways these people are driving on, but the freeways I drive on don't have any pedestrians, or hard turns, or oncoming traffic or parked cars. Going just the speed limit is way more dangerous if you are talking on your cell phone and drinking your big gulp. Going at the speed limit in a busy cyclist/pedestrian area is way more dangerous.

In related news, the Washington State House failed to pass a proper tough cell phone law. Talking on a cell phone is currently a secondary offense which basically means there is no law at all. Earlier the Senate passed a better version that at least makes it a primary offense, but IMO is actually still not tough enough as it allows hands-free talking and it still allows dialing by hand. (for those that don't know, driving while talking on a hands-free is roughly equivalent to driving while exhausted or driving drunk in terms of the affect on reaction time, braking and obstacle avoidance).

While I'm on the subject of car safety, I'll repeat my call for all interactive car computers to be banned, as well as Xenon headlights (damn blinding shit), also window tint (dumb fuckers can't see at night). It should also be illegal for children to bother their mother while she's driving; people need to pay damn attention to the road. I also had an idea on cell phones the other day after yet another incident where some dumb cell-phone talking pedestrian tried to walk in front my car : cell phones are just dangerous in any kind of use with movement, not just driving. Maybe they should just put speed sensors inside all cell-phones and put them on a hard switch to turn off when they are in motion, so you have to stop and stand still (or pull over your car) to talk. Oh, and headlights and bumpers should be mandated to all be at the same height. Like fine if you want to drive a retarded SUV, go for it, but your headlights and bumpers are still going to be one foot above the pavement, not aimed directly at my brain.

I also randomly stumbled across this Bizarre WA Supreme Court Moving Violation Ruling in which they seem to basically rule that because the trooper did not see the defendant commit the crime, he couldn't issue a moving violation. What !? So if someone runs their car into a lamp post and tests positive for alcohol, you can't give them a DUI because you didn't actually see them driving? This seems to set a completely retarded precedent. Says the Supreme Court :


"Negligent driving in the second degree is a moving violation. 
For the infraction to be valid, the movement must have been made in the officer's presence."

more story here or WA supreme court blog

old rants