Silician Sketchbook

It was my first time going to Massey Hall and I got a bit lost as a result. I emerged from the correct exit from Queen Station, but without something as handy as a map, and took the long way around by walking around the block.

Part of the confusion stemmed from having the wrong expectation as to its size. When founded my guesstimate by using Roy Thomson Hall as a reference, which is pretty large and stands alone for the most part. Massey Hall in contrast is much more nondescript and blends in with a cluster of similarly sized brick buildings.

After a bit of thinking, I think I would be partial to going to VGL again were it to come to town. I just wouldn’t pay $80 to do so. And I’d like to hear Kingdom Heart 2′s Passion and maybe some Command and Conquer / Red Alert music, but that’s secondary.

Eighty dollars is a steep price to pay to hear a mix of pre-recorded music and live orchestra. Compound that with the fact that the orchestra rehearsed as a group the day of the performance, and I thought it showed. There were obvious instances where the orchestra was not together either with each other or the pre-recorded bits.

Later, the organizer of the group I “went” with told me something that makes it sound as if the musicians didn’t quite know what they were getting into. I noticed the aftermath, when during one of Martin Leung’s performances one of the cellists got up and left, returning before he finished. It turns out that the piece before, foam that I thought was confetti was sprinkled down on the stage. That cellist found herself constantly pausing to wipe the foam off the cello’s strings, and there’s no telling what effect any liquid exposure could have to a string instrument.

The cellist gave the conductor these dirty looks the whole time and, according to my friend, left to probably chew out the stage manager.

I guess when I hear live orchestra being pitched to me, I immediately set certain standards for things like technical proficiency, co-ordination, and sound balance, and the latter two were sorely missed that night.

The same applies to Martin Leung, as technically accomplished as he is. If you want to get really serious about performance, give the man at least a mic’ed upright piano. A digital keyboard with its flimsy stand, and a typical chair, is not an optimal performance platform. As it was, he was pretty casual on things like hammer octaves, but when things are loud and fast one can pedal things away and the casual fan won’t notice, or just be more impressed with seeing hammer octaves live.

Because I’m a big Hikk-tard, one of the highlights for me was the performance of Hikari known as Simple and Clean for the english version of Kingdom Hearts. The real highlight, though, was the Coronation/Baba Yetu arrangement from Civilization IV. It was far the most inspirational, best integrated with the video game footage since it was all ancient history and epic structure building. Even though the soloist flubbed his part.

After going on YouTube and looking for what a real performance would sound like, I have come to the conclusion that in general it’s a pretty difficult part to pull off. It’s in Swahili, for one, and the sound balance has to be pretty near perfect because it’s not supposed to be too loud. The best performance of this arrangement was from the Hollywood Bowl, and it may have very well been the original game singers present. The next best performance of Baba Yetu isn’t even from VGL (which is kind of sad), but from the Valencia HS choir. The only problem is that the backing orchestra was pretty terrible and on many occasions the interjecting lines come in late.

I think that what with the contest events, video game footage of mostly terrible game play (the Sonic the Hedgehog footage is obviously bad even to those who have never played the games), not so perfect performances, and a bit of rabble rousing from the MC, this was really a kind of fan service for video gamers. VGL caters to the kind that will go to YouTube and watch video game music videos just like how I’d go to YouTube or Stage6 and watch anime music videos.

I’m not putting down those people, but I’m likely to think twice before spending a significant amount of money for something that’s better and free (provided it’s not taken down due to copyright complaint).

MonSTARS vs Dikembe Mutombo. Hilarious.

So many airshots and midair stuffing. I guess the hitboxes are larger, and soldiers are just physically large to begin with, so those shots are a lot easier to make.

No, I’m not even close to being good enough to single-handedly stack a team. I mean the other kind of team stacking, i.e. the selfish and mean-spirited kind.

Upon joining a TF2 pub server where a round is already under way, the first thing I do is look at the scoreboard. If a team is leading, then the decision is easy. If it’s too early or both teams have been deadlocked for a while, then scores are the next best metric. If I can’t join that team because they’ve got more players than the other, I’ll join spectator or otherwise hang around until a slot opens up.

I’m kind of burnt out on TF2 for pub play, unless it just so happens that the server attracts solid players that are comfortable executing strategies with little prior organization. I could play pickups for that, but there are times where no one’s available, plus those are more conducive to clans who want to get some practice in advance of league matches.

I guess long story short is, I despise poor players more than I ever have. I’ve been pretty annoyed in TFC, but never to the point that I will get on my mic and let the team have it for being incompetent. Why?

TF2 is much more team oriented than TFC. Given two pub teams with a mix of average players and stupid ones, a good player can dramatically tip the balance for either side in TFC. Part of it has to do with the fact that medium and heavy classes can take more punishment than their TF2 incarnations. A TFC soldier at full armor can take three rockets and a few shotguns to the face before going down, whereas a TF2 soldier is hard pressed to bear two point blank rockets and still survive.

And even if you were to die in TFC, in ten seconds you’ll be right back into the fight, less if you can jump (rockets, pipes, grenades). In the same amount of time in TF2, you’ll be walking to the fight if you’re on offence, and waiting out the final ten seconds to spawn on defence.

What boggles my mind then is that people don’t hesitate to run out and get slaughtered in TF2, whereas TFC idiots will get all scared and soil their pants while they’re camping out a corner – on offence.

So in summary, good players can be totally nerfed in TF2. With crits, even a gaggle of poor players can dislodge a good player from a location or outright kill him/her. Unable to spawn immediately and get right back in means that the team is at a material skill deficit.

I used to join the losing team, thinking I could turn it around or at least even the playing field some. Most of the time, I realize that losing is terribly un-fun. I can’t compensate for five, six, players who are effectively blind and/or think they’re pro snipers with zero kills (and thus zero skillz), or a team that for whatever reason has not noticed that no one is playing medic when the winning – nay, steamrolling – side is running three or more (class limits kthxbai).

The crux of the matter is, why are these people playing at all? TFC has always been free so I can’t really say anything when I see atrocious players, unless I’ve seen them be atrocious for the better part of 2 years (like [DAD] oh yes I went there).

But TF2? It has to be bought. With money. Come to think of it, I may not even have much of a case. If you want to spend money to show the world just how much you suck at hand-eye co-ordination, or inability to learn after being killed 3 times in a row in the same damn tunnel, then power to you. I guess it’s not much different from alcohol.

Tuning gaming netcode is like performing some mystical ritual. Everyone has their own way to get hitboxes to line up with the models they see on screen. In my experience it’s largely connection dependent, and that ping measurement reported in-game can be misleading.

During undergrad, I had pretty low latency connections through DSL and the university’s own backbone when in residence. Recently, latency performance has improved at home, but I find that I still have to lead my targets even if I do ping 50 ms on a cable modem.

I have yet to try the below chart out, original thread here. I’ve got to focus on studying for the next week. It suggests that the variable cl_interp has no effect. Being new to Source games in general, the initial advice I read was to set cl_interp to match my ping, and set rates according to a handy calculator.

Anyway, chart follows:

cl_interp_ratio 1.0 cl_updaterate 30 interpolation = 0.033
cl_interp_ratio 1.0 cl_updaterate 35 interpolation = 0.029
cl_interp_ratio 1.0 cl_updaterate 40 interpolation = 0.025
cl_interp_ratio 1.0 cl_updaterate 50 interpolation = 0.020
cl_interp_ratio 1.0 cl_updaterate 60 interpolation = 0.017
cl_interp_ratio 1.0 cl_updaterate 66 interpolation = 0.015
cl_interp_ratio 1.0 cl_updaterate 75 interpolation = 0.013
cl_interp_ratio 1.0 cl_updaterate 80 interpolation = 0.013
cl_interp_ratio 1.0 cl_updaterate 100 interpolation = 0.010

cl_interp_ratio 1.5 cl_updaterate 30 interpolation = 0.050
cl_interp_ratio 1.5 cl_updaterate 35 interpolation = 0.043
cl_interp_ratio 1.5 cl_updaterate 40 interpolation = 0.038
cl_interp_ratio 1.5 cl_updaterate 50 interpolation = 0.030
cl_interp_ratio 1.5 cl_updaterate 60 interpolation = 0.025
cl_interp_ratio 1.5 cl_updaterate 66 interpolation = 0.023
cl_interp_ratio 1.5 cl_updaterate 75 interpolation = 0.020
cl_interp_ratio 1.5 cl_updaterate 80 interpolation = 0.019
cl_interp_ratio 1.5 cl_updaterate 100 interpolation = 0.015

cl_interp_ratio 2.0 cl_updaterate 30 interpolation = 0.067
cl_interp_ratio 2.0 cl_updaterate 35 interpolation = 0.057
cl_interp_ratio 2.0 cl_updaterate 40 interpolation = 0.050
cl_interp_ratio 2.0 cl_updaterate 50 interpolation = 0.040
cl_interp_ratio 2.0 cl_updaterate 60 interpolation = 0.033
cl_interp_ratio 2.0 cl_updaterate 66 interpolation = 0.030
cl_interp_ratio 2.0 cl_updaterate 75 interpolation = 0.027
cl_interp_ratio 2.0 cl_updaterate 80 interpolation = 0.025
cl_interp_ratio 2.0 cl_updaterate 100 interpolation = 0.020

I’ve been running ForceWare 169.06 for about a week now, and TF2 no longer crashes.  At all.  Looks like NVIDIA has made significant progress in stabilizing the 8600GT for Vista, although a whole host of users on the download thread have experienced negative results for their own systems.

Frame rate locking did not help in the long run. I just got lucky that time.

The root of the problem seems to be with Windows Vista not playing nice with NVIDIA cards. It’s a problem that has been known about since at least January 2007. Why Vista? I can only speculate that it’s because Vista sucks, but that’s not a new sentiment by any stretch.

NVIDIA has a page listing some Vista hotfixes (note: they’re not for XP…). The first time I tried to play TF2 following hotfix installation, I locked up less than 10 paces outside the spawn doors.

All of this makes me loathe to play TF2, especially for the CEVO tourney. I’ve been lucky thus far, or maybe it’s because 6v6 matches don’t place too much stress on the laptop Vista, but I’m pretty sure that were I to lock up for at least 30 seconds, any momentum the team might have built would be erased in an instant.

Update! ForceWare 169.06 works fine.

Executive Summary

Create an autoexec.cfg in your .../team fortress 2/tf/cfg/ directory, if you haven’t done so already. Force your frame rate to something reasonable. I used fps_max “70″ which is in line with my external monitor’s refresh rate.

Why does this work? I’m not sure (yet), but I suspect that the video card is locking up due to overheating. Reducing the frame rate, as well as graphics quality, may keep it at a sustainable temperature.

Argh

I think the third thing I installed on my pretty midnight blue laptop, after Firefox and video card drivers, was Steam and the Orange Box.

I first noticed odd behaviour while playing Portal. For no apparent reason, my system would simply hang with sound effects playing in a stuttering loop. I could sometimes regain control of the game if I opened Task Manager and then tabbed back in.

Portal lock ups, while irritating, were tolerable because it’s just single-player and I wasn’t completing timed challenges at that point. TF2 is a different story. When you’re a medic about to give invincibility and you lock up, you may eventually get back into the game only to discover that your team lost the round. Playing matches, as opposed to random public servers, makes it worse. People may want to hit you (more) and you feel like hitting something yourself.

I updated the drivers for my GeForce 8600M GT (Dell released a new official driver set yesterday), and generally tried to do things that may fix the stuttering problem.

Downgrading to DirectX 8.1 lessened the severity of the lock up when it occurred, but it didn’t eliminate it. Reducing quality settings gave me a higher frame rate but that was it.

What I found to work, at least for the duration of a match and a couple hours of public server play afterwards, was to force my frame rate to 70 frames per second. The card is capable of delivering in excess of 150 fps with the quality settings turned down, but letting it run unconstrained might be causing it to overheat.

Why do I think it’s overheating? I find that I when I tab out, I need to wait around a bit before being able to maximize the game again. Otherwise, it’ll go to a black screen and may or may not have stuttering sound. Not all is lost, as I can force the Task Manager window to get me out. What might the system be doing during that time? Cooling down, for one.

The game also tends to lock up when there’s a lot of activity in my immediate vicinity. That’s also an inopportune time to take a forced break. I’m usually engaged in a fight and will be greeted by the familiar respawn room upon my return.

This Inspiron 1520 is a fine laptop otherwise, and I would be happy if NVIDIA released a driver that reduced the clock frequency or somehow reduced the load in order to regulate the core GPU temperature. Until then, anything above 60 fps is Good Enough and 70 fps happens to be line with my external monitor’s refresh rate.

This was motivated, coincidentally enough, by the TTC website. Scrolling text effectively forced me to use Internet Explorer to get route information, which is sad.

There are better uses for a JVM than scrolling text. I acknowledge that this is the least of the TTC’s issues, but Javascript and Flash have come of age.

I’m on a new system, with Windows Vista (oh noes!), with the latest version of Firefox, and it froze when trying to open this lame scrolling text applet. A JRE came installed, so it was a bit of a head-scratcher as to just why things weren’t working. Mozilla suggested that I install the latest JRE using XPInstall. I tried, that, but it fails with error code 203, which seems to be a pretty common problem. I wound up using the offline installation executable.

After setting everything up, Java applets no longer crashed Firefox; they just refused to load. I tried disabling and re-enabling the JVM through both Firefox and the Java control panel, to no avail.

What fixed the whole issue was the use of a registry patch. It adds an entry that should be present, but despite numerous versions of Firefox, continues to be forgotten by the installer. If you have problems, merge the registry patch first. It’s the most likely cause.