Archive for April, 2006

Chicks n’ Power Tools

Saturday, April 29th, 2006

You may be thinking, “One has nothing to do with the other.” That’s my experience, and I realize it’s a generalization…

Laura and I borrowed the neighbor’s power washer and went to town on the back deck, deck stairs, back porch, various outdoor furniture, doormats and astro-turf that lives in the porch.

Let me tell you, I had to fight Laura off to keep using it! This is the bona-fide first power tool Laura has ever shown interest in. If we didn’t have great neighbors who would loan theirs to us, Laura would already be on a mission to figure out which one was the best mix of price and performance. Heck, it might even already be in the garage!

Laura’s response to this observation was of course, “It’s a cleaning power tool.”

“Oh.”

Midnight Comedy

Tuesday, April 25th, 2006

Our house was built in 2000 and in six years time deeds have switched
hands four times. We believe people have left because our house is
haunted. The smoke detectors are el-cheapo units presumably put in place
by the builder at lowest possible cost, and they tend to set themselves
off when we leave the house open (which we do a lot in the spring and
fall). My suspicion is the detectors are not very good at discriminating
between smoke and humid air. When we leave the house open at night during
the time when the dew point drops, sometimes they go off…

This has caused us a few nights headache. My typical remedy is to take
them down until morning, then blow them out with compressed air and put
them back. I usually only do that if they go off more than once at night.

So one night last week, at almost midnight exactly, they go off. Beep.
BEEEP. BEEEEP! I just roll over and look up at the annoying beeping thing
on our ceiling. At this point my bolt-from-bed-in-fear-of-fire reactions
have dulled a bit, but I’m awake anyway.

Meanwhile, Laura is laying in bed, not fully awake yet, flapping the
covers while lying in the bed. Yep, madly flapping the covers. I ask her
what she’s doing, and she sasses that she’s fanning the detector. “From
there?” I ask incredulously, wondering how flapping the covers six feet
below a smoke detector, possibly not the one setting off the whole racket
(since they are all linked), will do any good. Agitated and frustrated at
the situation (and at me now), she tells me to shut up.

I roll over and go back to sleep, thankful it’s stopped.

In the morning we have a good laugh about the effectiveness of flapping
the bed covers at the smoke detector. We’ve come to hate those smoke
detectors, but at least we got a little laugh out of them recently!

MythTV High CPU Usage Resolved for the Time-Being

Monday, April 17th, 2006

I used to run my mythbackend/frontend on an Athlon XP @1733Mhz. It was mostly able to keep up with 1080i using Bob deinterlacing, and could easily do 720p.

With a recent switch of hardware to an Athlon 64 @1800Mhz on an Asus A8V motherboard, the hardware has been using 92% or more CPU to play 1080i using Bob deinterlacing, and 60-65% CPU to play 720p…

Given the old 32 bit hardware was doing the same job almost as well, I found the performance unacceptable. Because the high CPU usage for 1080i with Bob deinterlacing was causing glitching issues, I switched over to linear blend deinterlacing and saw a drop to about 70% CPU for 1080i. While this wasn’t the ideal solution, it at least kept me from going nuts for a while.

Today a user on irc.freenode.net #mythtv-users suggested I boot the machine using the “noapic” kernel option. What this does is disable the APIC interrupt routing. Immediately, I noticed a difference. Playing 720p only used 35-40% CPU and playing 1080i with Bob deinterlacing only used 55-60% CPU.

It may not be the long-term solution, as I believe there are benefits to having a working APIC configuration, but at least for now it has made using the mythfrontend much smoother and less temperamental.

So if you’re seeing overly high CPU usage during playback, check to see if APIC is being used, and if so, change your kernel command line parameters to include “noapic”. You can do this by editing the appropriate line(s) in /etc/lilo.conf and running lilo, or editing the /etc/grub.conf or /boot/grub/menu.lst file and adding the parameter to the appropriate line(s).

Hearing Loss Test Results

Friday, April 14th, 2006

I received a call from the physician’s assistant at Johns Hopkins yesteday morning…

My MRI came back normal in all respects, except possibly some cyst material near my right TM joint (the back of the jaw). Since I don’t have any TMJ pain or clicking this probably isn’t something to be concerned about.

The tests for bacterial infections known to cause hearing loss came back negative, as did indications for an auto-immune condition that could be playing a part. My Lyme disease test results did not yet come back.

The only thing of minor concern the results showed was a little bit of hyperthyroidism. Their normal range is .5-4.5, and mine came back .46. The physician’s assistant indicated this could easily be a lab error, and didn’t mention anything about this slight deviation from “normal” causing any issues with hearing, but she wanted me to follow-up with a primary-care physician to retest and track this if necessary.

Some of you have wondered what causes this condition. Unfortunately, there are a lot of theories, but the doctors don’t really know at this point. The physician I am seeing at Johns Hopkins has indicated a correlation with viral infections. On a business trip the week before this occurred, I had thought I was food-poisoned, but in retrospect, the lingering GI issues may have been an indication I had a viral infection. It would seem to fit the correlation they have noticed. I get the impression it is not easy or possible to test for an arbitrary viral infection that might cause this, since people react so differently to viral infections, and the virus could be any of a huge number of viruses.

Here’s the article I found when searching for symptoms that matched my condition last Friday morning.

I’ve not had any real hearing improvement since Sunday, but according to the physician’s assistant, there is no trend they have found in subjective patient reports of hearing improvement. Some people get it up front, others on the backend of the steroids course, others after they’re off the steroids. Keep your fingers crossed with me!

Sudden Sensorineural Hearing Loss

Tuesday, April 11th, 2006

For those of you that may not have known, I experienced a severe loss of hearing overnight last Thursday… (continued after jump)

In the late evening I noticed a slight ringing in my right ear, which became a loud distracting ringing accompanied by some hearing loss by bed time. When I woke Friday morning, the ringing was “over-amped” to the point it felt (rather than sounded) like ringing, and I couldn’t hear much but loud reverberations anytime there was a high pitched noise like the beep of a vehicle backing up.

I did some searching on the internet and found a paper on something called Sudden Sensorineural Hearing Loss. Thinking this to be a match for what I had, I tried to get seen by a local ENT without much success. Laura’s primary care doctor would not give out the name of the ENT they refer to without me being seen, despite the fact that our insurance lets us consult with a specialist regardless of whether we’ve seen a primary care physician first. After some pushing, I got an appointment with Laura’s primary care physician at 10:45a Friday. After some harried phone calls with my mom (a nurse practitioner), I found out she had gotten in touch with the physician’s assistant at Johns Hopkins who is running a study on this particular condition. She made it clear I should be seen today. So with that in mind I started calling emergency rooms in the Richmond area, trying to find one with an ENT doctor on-call. I settled on VCU/MCV, thinking that because they did in fact have an ENT doctor on-call and it was a teaching hospital, I would get similar concern as Johns Hopkins was showing for the seriousness of my situation.

To make a long story short, I waited 90 minutes and saw an ER doctor who could not get the ENT doctor on-call to come down and examine me. The ER doctor examined my ear with an otoscope and spoke on the phone to the ENT doctor on-call for a few minutes, then prescribed me a short course of high-dose steroids, along with an anti-viral drug.

According to the physician’s assistant at Johns Hopkins, getting on the high dose steroids immediately turns out to be the most important factor in hearing recovery for the study they are performing. According to her I really should have had a full workup (bloodwork, MRI, hearing test) on Friday, but starting the steroids immediately and getting to the rest on Monday was the next best thing. So with that in mind, she scheduled appointments for me on Monday and Laura and I made plans to drive to Maryland, stay the night with my parents, and go in Monday for my diagnostics.

Monday morning at Johns Hopkins, I first had a hearing test. My left ear results came out very well. My word discrimination was 100%, with a “pure tone” average level of 11db. What this means is that the level at which I could hear tones at 1000hz, 2000hz, 4000hz, and 8000hz was averaged, and came out to a relatively quiet 11db for the left ear, which is functioning normally. In contrast, my right ear word discrimination was 72% with a pure tone average level of 45db. This was enough to be considered “moderate” hearing loss, but not enough to be eligible for the study Johns Hopkins is running.

After an ENT-neurological exam, some examination with an otoscope, and consultation with the physician at Johns Hopkins experienced with this condition, I was prescribed another 10 days of the high-dose steroids, as their study shows that to be the most important factor in regaining hearing. The physician indicated my prognosis was good, given that I’ve already seen some progress over how my hearing was on Friday morning, so I remain hopeful for a full recovery of my hearing.

While we were at the Johns Hopkins Outpatient Center, I also had blood drawn and an MRI. We decided that the wait-time on the MRI was probably going to prevent me from getting to the bloodwork before they closed, so while in line for the MRI we went down to get my blood drawn. I believe they filled 4 tubes, but they still needed to stick me twice - the first time they either didn’t get a vein, or it wasn’t producing. My veins tend to roll and be hard to get ahold of with a needle.

The MRI without contrast went smoothly, but not so smoothly when the time came to inject the medication for the contrast imaging. Again, my veins were hard to get ahold of, and I got stuck 4 times to get the injection, with a painful little side-trip of the medication injected into my skin instead of my vein. Let me tell you, your arm begins to ache in a hurry when that stuff goes in your skin!

The cool thing about the MRI is there was a very nice guy doing a lot of the work, and I asked him if I could have a copy of the results. He burned a CD for me to look at that includes a bunch of data files and a Windows-based program for viewing and compositing into 3D views. I will try to export some screenshots in the near future in case anyone is interested.

Looking forward, I’m very hopeful I get my hearing back to what it was prior. Having the moderate amount of loss I do is disconcerting. I’m sure I will adapt and be able to better understand speech, but high-frequency noises (in most music, as an example) make my right ear feel like someone is ringing a very loud bell right next to it. It’s extremely distracting and will probably prevent me from enjoying audio/multimedia, which would be a big disappointment. For now, I’m not going to focus on that, but instead just keep my fingers crossed and think positively!

I’ll be following up with another hearing test and examination by the physician at Johns Hopkins in two weeks.

More then…

GTK Mac OS X Rounding the Corner?

Saturday, April 8th, 2006

I recently got word that GTK compiled natively against Aqua may be starting to round the corner to usability. I’m looking to spend some time contributing in the near future, but this may take a backseat to some MythTV modifications I want to make.

Check out the status of GTK Mac OS X for yourself.

SSHKeychain Universal Build

Saturday, April 8th, 2006

Need a universal binary for SSHKeychain.app? I built one using sources pulled from the trunk of the read only svn mirror on 20060407. My casual testing has shown it to work as expected on my intel mini. Please let me know if you find issues with it.

Here it is: SSHKeychain-universal.dmg.bz2

MythFrontend for Mac OS X PPC/Intel

Saturday, April 8th, 2006

I needed to upgrade my MythTV backend to the release-19-fixes branch from SVN in order to solve an issue with timeouts to the MySQL server (see: mythtv-users thread). As part of the upgrade, I went ahead and built .19-fixes (checked out 20060407) frontends for Mac OS X PPC and Mac OS X Intel (MMX enabled).

Here are the links:

And for those running older backends, I still have the binaries I built for .19 and .18.1 sitting around: