Archive for category Recumbent Bicycling
The retina-burn white reflective tag under the black hand strap is actually a foreshortened view of the Arkel logo.
They’re longer and taller than the old packs, which isn’t entirely a Good Thing: the inside bag gently kisses the pavement during steeply banked high speed turns. The main compartment is slightly narrower, so I bent the license plates (which used to fit neatly on the bottom) to form a hard floor with a low lip on the inside edge. That, in combination with tightening the pack’s internal strap, prevents the foam-core bottom panel from drooping; maybe the edge won’t hit the pavement quite so often.
They also ride much higher on the racks. To install the packs, I had to unbolt the seat to raise it upward, slide the packs underneath, twiddle the clamps onto the rack rods, then reinstall the seat. Those puppies are not getting loose without tools and a struggle!
Because they’re longer, the right pack collided with the HT mount behind the seat:
So I moved the mount up to the middle crossbar:
I’m not entirely happy with that arrangement, as the holder sits snug against the rear packs. So far, I rarely need those in addition to the RT-40s, as each underseat bag can swallow an upright gallon milk jug or two Butternut squashes in addition to all the other stuff I normally carry.
The array of reflective patches and piping and pull tabs probably makes me look (more) like a low-flying UFO at night, but that’s fine with me: the more it resembles a UFO, the less hassle I get.
avconv incantation required to put text on frames extracted from a video file looks like this (it’s all on one line, so you’ll need some side scrolling action):
avconv -ss 00:11:47 -i /mnt/backup/Video/2014-09-08/MAH00070.MP4 -t 1 -f image2 -q 1 -vf "drawtext=fontfile=/usr/share/fonts/truetype/ttf-dejavu/DejaVuSansMono.ttf : text='2014-09-08 10\:58\:47' : fontcolor=white : fontsize=60 : box=1 : email@example.com : x=1200 : y=30" MAH00070-001147-%03d.jpg
-ss 00:11:47 sets the starting time relative to the beginning of the file, so it’s an offset that, when added to the file start time in the Exif metadata, produces the actual time-of-day. The extracted frames begin at the closest “seek point”, which I presume will be pretty close to the specified second. The
-accurate_seek option may be relevant. Verifying all that could be tricky.
-t 1 specifies the duration. Each second produces 60 frames, numbered from
060 in the output filename, as defined by the
%03d in the output filename format string.
-vf "drawtext=" gibberish does the actual text overlay, with all the parameters tucked inside the double quotes.
You must escape all colons in the
text string (as
'10\:58\:47', note the single quotes), because unescaped colons separate the
fontsize seems to be in pixels with an upper limit of 72.
boxcolor rectangle just barely covers the characters; there’s no way to enlarge it just a few more pixels to make a nice frame. The fraction at the end of
firstname.lastname@example.org string produces 70% opacity.
I manually added the actual starting time (10:47) to the offset time for each segment (previewed with
vlc), jammed that into the
avconv command, and extracted some interesting frames from a recent ride…
I get plenty of clearance while approaching an intersection, which is pleasant:
Absorbed in something on the passenger seat while I’m trackstanding the ‘bent and watching the brake lights:
The turn signal goes on just after acceleration commences:
Because I never pass on the right, I didn’t participate in a classic right hook:
The traffic signal goes yellow as I cross the walk ladder, with the tail of the SUV visible beyond the crosswalk on the right. The green-to-yellow transition takes 10 frames = 1/6 second and this image shows the half-intensity point of both incandescent bulbs:
The rest of the ride seemed less eventful.
Frankly, that’s way too much handwork for the results in the upper-right corner. I think a better way starts with extracting unannotated frames from the video, then slapping timestamps on them using ImageMagick, calculating and feeding it the appropriate values for each frame.
Putting the annotation up in the sky seems better than near the bottom corners, if only because images of the pavement might actually be useful. The timestamp needs the frame number and I think splitting it into two shorter sections (date and time) in the left and right upper corners might work better.
The signs at every Dutchess Rail Trail grade crossing and access point seem unambiguous:
More specific signs appear at random intervals along the trail:
You can’t see it, but every sign includes an invisible asterisk introducing the invisible clause “Except Cops”:
Back when the Dutchess County deputy sheriffs rode huge ATVs that occupied nearly the entire paved trail and bulldozed everybody out of their way, I had the temerity to ask why they weren’t riding bikes. The deputy sheriff told me, rather condescendingly, that they had to be prepared for anything and that there had already been incidents.
These little ATVs aren’t quite so imposing and, more likely, also fit on the new bridges and between the bollards, which may explain everything.
I’ve seen what might be their best use case, although ambulances can attract your attention without an ATV escort:
Straight up, I have no objection to police patrols on the rail trail.
I do object to the official mindset that simply adds an invisible exception to any inconvenient rule.
As I see it, the root cause of the militarized police and extralegal government activities we’ve seen across the country in recent years boils down to “That law / regulation / rule does not apply to us, because we are the government.”
I can ride the length of the DCRT and back in about two hours, averaging 12 mph, without getting particularly sweaty in the process; the track in that link shows a three hour ride that includes the HVRT and a Walkway scrum, plus the ride from and to home. A police ATV can’t go much faster than that on the trail, even with lights and sirens, because oblivious pedestrians keep getting in the way.
If an officer on a bike can’t keep up with me, then something has gone badly wrong with the job requirements for becoming a deputy sheriff.
As far as “being prepared for anything” goes, the cargo capacity of those little ATVs rules out a bunch of hardware that fit in the big ones: anything seems an elastic concept. A bike can carry enough equipment for many incidents; my tool kit weighs more than some bike frames, the packs have plenty of room to spare, and there’s always the trailer option. I doubt genuine Mil-Spec assault rifles would come in handy on the rail trail.
It’s also not clear why an officer on a bike can’t call for the same backup as an officer on an ATV: those buggies lack fancy VHF antennas, so they’re using a hand-held radio or phone. The 5 W amateur radio on my bike, through a mobile VHF antenna on a crappy ground system, can easily reach local amateur radio repeaters and APRS nodes. Many pedestrians seem absorbed with their phones, so getting microwaves into and out of the trail doesn’t pose much of a problem.
Cops-on-bikes present a much less aggressive aspect than cops-on-ATVs who ignore the rules that apply to the rest of us.
They could do it differently, as the department has both bikes and ATVs.
While installing new underseat packs (about which, more later) on my Tour Easy, I discovered a bight of PTT cable had been touching the top of the chain:
The gentle ripples to the right of the worn-through section seem particularly nice; you couldn’t do that deliberately if you had to.
This section of cable should have been taped to the upper frame bars. It’s hidden under the seat, just in front of the rear fender, and between the under-seat packs, so it’s basically invisible from any angle.
Soooo, that probably explains a bit of the intermittent trouble I’d been having with the PTT switch, although most of it came from the corroded switch contacts.
Rather than replace the whole cable, I cut out the eroded section, spliced the conductors, and taped it firmly back on the tubes.
A pocket camera can’t do justice to the Bald Eagles, just before dusk and halfway across the Hudson River from our river cruise yacht:
We got a closer look at the pair of eagles who once graced the original Grand Central Station and are now standing guard at St Basil Academy in Garrison.
This one glanced away from the entrance, perhaps to keep an eye on us:
Another watches over an interior road:
They’re two tons of cast iron apiece and, should any of you want a restoration project, I’m sure the good folks at St Basil’s could work something out.
We saw those during the Cycling the Hudson Valley tour: riding during the day, camping and touristing in the evening.
Several years ago we encountered a Penn Station eagle at the Washington Zoo:
Fly away, young Valkyrie, fly away …
It had to happen:
Given that the Sony HDR-AS30V has a 170° FOV (probably diagonally corner-to-corner), the honeybee must be an inch or two from the lens. Subsequent frames show it circling around in front of the fairing and continuing about its business.
Faced with the utter confusion caused by trying to figure out the interactions of the Sony HDR-AS30V time settings, I set it to local time, Time Zone GMT+0, DST off, enabled the GPS receiver, disabled the auto-off battery saver, and parked it outside with a nice view of the sky.
About ten minutes later it achieved GPS lock, which suggested the time wasn’t terribly wrong. After locking, the camera set itself to the current local time +1 hr, TZ = GMT-5, DST off; it apparently runs an embedded Linux distro, so it can convert from geographic location to time zone easily enough. It cannot know about DST, so I set DST on, the time jumped -1 hr, and it’s now set to local time, TZ = GMT-5, DST on.
A short movie produces this metadata:
exiftool MAH00046.MP4 | grep -i date File Modification Date/Time : 2014:09:02 17:58:49-04:00 File Access Date/Time : 2014:09:02 16:59:51-04:00 File Inode Change Date/Time : 2014:09:02 16:59:51-04:00 Create Date : 2014:09:02 20:58:41 Modify Date : 2014:09:02 20:58:48 Track Create Date : 2014:09:02 20:58:41 Track Modify Date : 2014:09:02 20:58:48 Media Create Date : 2014:09:02 20:58:41 Media Modify Date : 2014:09:02 20:58:48 exiftool MAH00046.MP4 | grep -i zone Time Zone : -04:00 ll MAH00046.* -rwxr-xr-x 1 ed ed 18M 2014-09-02 17:58 MAH00046.MP4 -rwxr-xr-x 1 ed ed 11K 2014-09-02 17:58 MAH00046.THM ll -c MAH00046.* -rwxr-xr-x 1 ed ed 18M 2014-09-02 16:59 MAH00046.MP4 -rwxr-xr-x 1 ed ed 11K 2014-09-02 16:59 MAH00046.THM ll -u MAH00046.* -rwxr-xr-x 1 ed ed 18M 2014-09-02 16:59 MAH00046.MP4 -rwxr-xr-x 1 ed ed 11K 2014-09-02 16:59 MAH00046.THM date Tue Sep 2 17:13:44 EDT 2014 date -u Tue Sep 2 21:13:47 UTC 2014
The camera quickly achieves GPS lock with DST on (so it’s still able to use its recent GPS ephemeris data) and doesn’t change any of the current time, Time Zone, or DST values (so it’s happy with all that).
Based on that short experiment:
- Internal Exif timestamp values are UTC
- Time Zone Exif value is (Actual Time Zone + DST)
- File modification is (local ending time +1 hour)
Some poking about shows that I misunderstood the “create” time, which actually holds the time when the file metadata changed, meaning there’s no way to tell when the file was created. I did know that the “access” time tracks the last time the file was opened: reading the Exif data will update the access time.
The +1 hr offset in the modification timestamp will (probably) vanish when DST goes off in the fall.
The Exif data contains the correct UTC times for the video’s Create and Modify dates, as well as the net Time Zone offset, which means I (well, a script using exiftool) can calculate the (UTC or local) time corresponding to the beginning & end of the video file. Indeed, exiftool can whack the file’s modification time directly from the Exif data:
exiftool '-FileModifyDate<ModifyDate' MAH00046.MP4 1 image files updated
Extracting still images proceeds from a time relative to the start, so the number of frames plus the calculated start time gives the actual time for that frame, modulo converting frames into hh:mm:ss.ff format and getting the addition correct.
Perhaps it makes more sense to set the modification time to the starting time of the video, thus simplifying the calculations:
exiftool '-FileModifyDate<CreateDate' MAH00046.MP4 1 image files updated
Both of those times are the UTC values, so a further adjustment based on the time zone would be in order. It’s not clear whether exiftool can perform that calculation internally or if it must be a two-step process.
A cheat may be in order: the thumbnail file modification timestamp matches the starting time of the corresponding video file. The script can base its calculations on the thumbnail timestamp. Of course, both of those are subject to the FAT-vs-DST problem.
More pondering is obviously in order.
For what it’s worth, the raw GPS log data from the camera lives in the
PRIVATE/SONY/GPS/ directory, with a
YYMMDDnn.LOG file name. It relentlessly accumulates two records every second:
@Sonygps/ver5.0/wgs-84/20140902205841.000/ @Sonygpsoption/0/20140902205842.000/20140902205842.000/ $GPGGA,205842.000,4139.5196,N,7352.4515,W,1,0,,,M,,M,,*5B $GPRMC,205842.000,A,4139.5196,N,7352.4515,W,8.89,,020914,,,A*5E $GPGGA,205843.000,4139.5177,N,7352.4501,W,1,0,,,M,,M,,*50 $GPRMC,205843.000,A,4139.5177,N,7352.4501,W,8.86,,020914,,,A*5A $GPGGA,205844.000,4139.5160,N,7352.4488,W,1,0,,,M,,M,,*51 $GPRMC,205844.000,A,4139.5160,N,7352.4488,W,8.54,,020914,,,A*54 $GPGGA,205845.000,4139.5148,N,7352.4476,W,1,0,,,M,,M,,*5B $GPRMC,205845.000,A,4139.5148,N,7352.4476,W,7.77,,020914,,,A*50 $GPGGA,205846.000,4139.5143,N,7352.4467,W,1,0,,,M,,M,,*53 $GPRMC,205846.000,A,4139.5143,N,7352.4467,W,6.61,,020914,,,A*5E $GPGGA,205847.000,4139.5145,N,7352.4461,W,1,0,,,M,,M,,*52 $GPRMC,205847.000,A,4139.5145,N,7352.4461,W,5.17,,020914,,,A*5D $GPGGA,205848.000,4139.5156,N,7352.4458,W,1,0,,,M,,M,,*55 $GPRMC,205848.000,A,4139.5156,N,7352.4458,W,3.31,,020914,,,A*58
The battery life is so terrible with the GPS on that I don’t plan to use it much at all, but at least now I know how to set the clock’s time…