It was too wet and windy today for my weekly house photos, but it occurred to me that I
should get some photos of the building site before they poured the slab. Over to take a
look, and to my surprise found some people laying out the slab structure:
Chris Bahlo and I discussed it at breakfast and tried to guess what it would say without any
context. Clearly the bloke on the right is putting himself in the scene for some reason,
and it seems that the bloke on the left is looking at him and wondering if he's completely
sane. And is the woman in the middle blind, or is she leading a dog (or both)? About the
only thing that is clear from the background is that they're on a road about 100 m from a
pedestrian crossing.
In fact, it's part of a state election campaign brochure:
The caption to this particular one is “Committed to improving pedestrian safety”. How? By
allowing them to cross roads outside designated crossing areas? How does the photo relate
to that problem?
More to the point, though: what party
does Geoff Howard belong to?
There's nothing in the brochure to indicate the party. We discussed and came to the
conclusion that he belongs to
the Liberal Party of
Australia (maybe because of his surname). But why didn't he state it? Is he ashamed of
his party? In fact, it turns out that he belongs to
the Australian Labor
Party. You'd think it wasn't important.
I was somewhat ambivalent when I discovered that we had a “nature strip” in front of our
house, and that we weren't allowed to tidy up the mess. But at this time of year a
surprising number of wildflowers have shown up. There's another strip on the other side of
the road, and that's where I found this flower:
What's that? Nothing in /var/log/messages. The other X server was still running,
but server 0 was using lots of CPU time and was otherwise unresponsive:
Over to the site again today to take a closer look at the slab. Yesterday I had got the
impression that they had spread the polystyrene foam over the entire surface of the house,
but that made no sense, and it wasn't the case. There are channels of differing thicknesses
around the whole area:
They go down to the clay 60 cm below, so they're probably to stabilize the slab on the soft
topsoil. The first one is marked on the slab plan, but the second one isn't.
CJ along this evening to help bring Chris Bahlo's crush (cage for horses) over to our
place. Chris had said that it was too heavy for less than 4 people, but the real problem
turned out to be that it was jammed under another part of her equipment. No big deal, but
somehow we're accumulating more and more stuff in Kleins Road.
I've finally decided on an exposure meter. After some backwarding and forwarding, decided
on the Sekonic
L-308S after all. The cheapest offer was from
eGlobaL [sic] Digital Cameras, only AU $184 with postage. The cheapest price
I could find on eBay was US $167.90, which
corresponds to AU $198.40, and reputable US sellers are offering it for round US $235 plus
postage.
eGlobaL shut down in 2018 rather than fight changes in the Australian GST legislation.
So I went ahead and fought my way
through Yet Another Broken Website to finalize the purchase.
Most web sites are too leet to accept phone numbers in standard formats (like (03) 5346 1730
in Australia). Instead they want all punctuation removed, like 0353461730. Clearly the
programmers can't do it themselves and don't realize that this punctuation contains valuable
information for error detection.
But that didn't work either here: they want the leading 0 removed as well!
That's not a suggestion, it's a requirement. Finally I had it all. Submit, go to
PayPal, pay. Return to complete transaction.
“Your request failed: Internal Error (Code: 10001)”. OK, try again. PayPal refused: sum
already paid. But the web site didn't want to know. What happened
to ACID?
How do you contact these people? They don't have a phone number! Sent in a support
request, to which as of present I have had no response. In the meantime gave in and tried
their online chat, which doesn't start until 11 am. Finally I got on to them and they
apologized, confirmed that the item was paid, gave me an order number, and promised to mark
the shipment urgent.
Checked my account a few hours later. Nothing. Chat again. Once again they had to confirm
that the sum had been paid, which was more difficult this time. Promised me email.
Silence.
That must be one of the worst experiences I have ever had online. It's clear that they're
honest enough, but their web software is dangerously buggy, and the trouble is worth more
than the few dollars I save. Let's see what happens tomorrow.
Another GPU hang today! What's causing it? This system has been running unchanged since
February, and it's been up for 139 days. The hang only affects one of the two X servers, so it doesn't seem to be hardware. It's not logged
anywhere except on the console, not helped by whining flash plugins:
When I first saw the problem, the error message was still on the screen, but later it had
long been flushed by these useless error messages.
The resultant image is strange. The root window (with an xearth display) seems normal, and I can switch to and from it, but the server
is no longer responding:
It's no longer in production. How old? Could be decades. It's nicely built, only 600 g in
weight, and specially designed for close-up work, and it's idea for the kind of wildflower
photos I've been trying lately. Why is nothing like this being produced any more?
Out to Kleins' Paddock to take photos of these tiny orchid-like flowers:
They're about 1 cm across, and there are still focus issues, not the fault of the tripod.
The second image was taken with my
10 dioptre close-up lens, which shows
significant chromatic aberration. Here first without, and then with:
My tripod came from the USA, and was supposed
to have been sent by USPS. Was it? Hard to
say. Each address label has been pasted onto the previous one:
But I have tracking information. What does that tell me? It starts with eBay's tracking information, which is backwards, of course, and in such a tiny window that you
can only see two hops at a time. Here's the start:
I suppose that's the state of the web for big companies.
Back to eBay, where I found that it took a total of 3 days being shunted
around Erlanger KY before
being sent on to LA, presumably Los
Angeles, though with a code 3352, my own post code. And it took another 4 days to get
there. One week and counting.
But that was 5 days ago. What then? There was another
link to a company called wnDirect, who gave
me further information: In transit, tomorrow at
5:32 GMT. Clearly a time zone issue, in
particular a misunderstanding of the difference between GMT
and UTC, but when I looked at it, the time
was really “yesterday” GMT, in fact round 22:54 on 2 November.
Further investigation showed that the time was
really AEST, UTC+11:00, and yes, indeed, the
parcel had arrived in Wendouree at that
time. Why are web sites all so useless?
Why is the mother (AuSome Zarzuela) in a crush? Because she rejected the filly (provisional
name Narrawin Esmeralda). Chris has been
breeding horses for decades, and though she had heard of such a thing, she had never
experienced it. A good thing we brought the crush over here on Sunday.
Unlike last month's foal, this one looks healthy and strong, but she still needed bottle
feeding:
And this is presumably the offset for the (steel) house frame, with the “permanent temporary
pole” for the electricity, and also conduit for the phone line I fear we're going to need:
After yesterday's fun with eGlobaL, it seemed to be prudent to check whether my order number
was now visible. No.
Waited until 11:00 to “chat” with somebody (they only seem to have ids CS-JL and CS-JH), and
they discovered that they had associated it with a different email address. Finally got
that and discovered yes, processed, sent out yesterday with tracking number which tells me
only that it came from Hong Kong—not
what I had expected. The price is well below the customs threshold, but there's nothing on
their web site that indicates that customs might be a problem. Still, things seem to have
sorted themselves out.
“Storm water” is a rather strange term used in Australia for “rainwater”. We're dependent
on rainwater for our fresh water supply. It's collected from the roof, piped underground
and back up to tanks almost as high as the roof. JG King should have done that, but
the Ballarat office made me an offer I
could refuse, so now Mari Hendriks and Warrick Pitcher will do it. It has to happen right
at the start, which doesn't make much sense, and Mari got bent out of shape about it, but
we're contractually obliged to do it. So tomorrow's the day.
How much piping? Going over the not-very-detailed plan, it seems to be about 84.4 m. Mari
will bring more, but it'll be interesting to see how things pan out.
Yvonne wanted to print a document today, something she does
so seldom that she needed my help. And in the process discovered that I hadn't configured
the printers since last month's system upgrade, five weeks ago. Note to self: check /etc/printcap and that the
spool directories exist.
Off to Geelong in the afternoon for my
six-monthly periodontal check. While there went looking for Asian foodstuffs. No, Chung
Vinh Phat still didn't have any ketupat,
or at least they didn't understand the name and description, and I had forgotten to bring
the empty packet of the last ones. But I had come prepared, with a list of shops
in Belmont. Most seemed to be
Indian, but one was called Indo-Asian groceries, which I thought might be broader in scope, but it proves to be
run by M. Singh and P. Kaur. They did have some other stuff, interestingly
including Masa harina, something to
remember when we run out, but once again “ketupat” brought a blank stare.
Picked up some ghee and looked around the area. The other shops that I have found on the
web in that area were all Indian, but there were enough of them that a random search sounded
like a good idea. Sure enough, just round the corner in the Belmont Centreway found a
little place titled “Gourmet Asian Grocery”. In there, and heard the proprietor (Jason Yap)
speaking English. He's Chinese, but Chinese speak English with different accents, and this
one was clearly from the Malaysian area. Ketupat? Yes, pronounced correctly (not an easy
thing to explain), and of course he had them. Not the same kind; we'll see what they're
like. But clearly a place to remember.
In the middle of the night I woke up and it occurred to me that we needed to think about
garden irrigation. That's basically sprinklers fed off bore water, requiring underground
piping and cables for the relays. Last time David Yeardley came over and dug trenches for me, and it didn't take too long. But why do
that? Tomorrow they're digging trenches for stormwater, so they could throw the other stuff
in at the same time. While in town Yvonne picked up most of
the stuff we need, but she couldn't find the control cable anywhere. Finally at the
Ballarat Pump Shop they dragged out a couple of partial rolls of cable and sold it to her:
“There must be 100 m there”.
It's damned expensive stuff, and looking at the rolls I wasn't convinced that it was 100 m.
And in any case, we really needed about 80 m in one run. Yvonne spoke to Warrick Pitcher,
who gave her the name of UPI (Ultimate Pumps and Irrigation), who should have it. Yvonne
back into town and finally came up with the real thing. Looking at the diameters, I'm even
more convinced that the 2 rolls from Ballarat Pump Shop are nowhere near 100 m:
Today Warrick Pitcher and Mari Hendriks came to dig the trenches. But they weren't the
first. I arrived just before 8:00 to find somebody (I think his name was Keith) unloading
the frames for the house—right on top of where they wanted to dig the trenches. A little
issue of site supervision?
The problem was that the frame packs are heavy, and Keith couldn't lift them back on the
truck to put them somewhere else. Warrick and Mari arrived soon after, and Warrick came to
the rescue with his excavator, moving most of them out of the way.
I went off to breakfast and to talk to Craig, who was cleaning out the mess in the garage.
I should have taken a photo of what it looked like before, but afterwards it was
transformed, and the skip was full:
That's an old bed on top, one that we bought in March 1975. How this stuff brings back the
memories.
When I got back on site, they hadn't got much further, but I got the information that Duncan
(we think—he didn't introduce himself) had been there and taken a look. They got going
pretty fast after that, though:
They were finished by 16:00, rather later than expected, but Mari had time to come and look
at a blocked drain in Kleins Road. The blockage wasn't where I expected it, and since there
are no inspection openings where they should be, it looks like it could be quite a problem.
The extension tubes for my camera arrived today, but I didn't have time to do much beyond
confirm that they worked. But they came with a strange device:
Things are not going well with Zarzuela and Esmeralda. Zarzuela has almost no milk, and she
still rejects Esmeralda as vehemently as ever. Chris bought some milk substitute, and
they're feeding her with that. She also put her in proximity with Guardaraya, the mare who
lost her foal in
September. That looks like it might work
Off today with my camera, my mantis tripod, the new extension tubes, the dogs and Yvonne to Kleins Paddock to try Yet Again to get usable photos of this
tiny wildflower:
It wasn't easy. The mantis isn't ideal for this sort of thing, because adjusting the height
of the camera also moves it off the subject, but in practice that was the least of my
worries. How do you focus? Manually, of course, but the real issue there is that the LCD
display is just not bright enough. In the studio it's easy: the magnification function is
better than anything else I've ever seen. But it's almost unrecognizable in the sun. In
addition, I had thought that the yellow part of the flower was a stamen, but in fact it's
just a coloration of that part of the petals. Despite my best attempts, I just couldn't get
the thing focused. In the end tried a shot anyway, and in fact it wasn't too bad:
The problem was that I just couldn't see it in the viewfinder, not even with the magnifier.
I suppose I should come with a magnifying glass.
In addition, there was noticeable camera shake. The exposure time was 1/25 s, which isn't
that bad, and I had set a 2 s delay on the shutter, but it's not the firmest ground. Next
time I should try flash as well.
Over to the site to see what hadn't happened. Duncan had said that they would start
erecting the frame today, but nobody was there. My guess is that he had forgotten that
today was supposed to be Ballarat Cup Day, though the Ballarat Turf Club, who organize it, seems to
think that it will be on 22 November 2014. But maybe that's why nobody
was working.
That didn't stop Jim Lannen, though, who came over with paperwork for connecting
electricity, which hopefully will be connected by the end of next week. Over again to
discuss some details with him.
After my photos in the nature strip, got into the car and started to drive home. Something
was on my neck, so I brushed it off and watched a relatively large, bulbous spider going
down the outside of the leg of my pants. Stopped immediately, of course, and looked for it.
Gone.
Where did it go? About the only thing I could think of was up the inside of my pants. No
time to be prudish: the thing could be
a redback. Pants off, and drove
home in my underwear. I must get a can of insect spray for the car.
The new version of DxO
Optics “Pro” has a number of improvements over the previous version: it can now
display properly on medium resolution (2560×1440) screens, and hopefully also on 3840×2160
screens. It has reinstated the old image comparison function (original/modified), which
they had previously put on the Ctrl-D key. But that repeats, so I ended up with a
quick alternation of the two. Now there's a button you can hold down with the mouse.
Best, though, they have finally found a (cumbersome) way of telling the crop function
not to enforce aspect ratio, at least for the “elite” version: you can save the information
with a “preset”. And of course that's about the first thing I did, both for myself and for
Yvonne.
But today Yvonne told me that it didn't work for her. Investigated this horrible “preset
editor”, which has completely unexpected behaviour, and discovered that every time I tried
to save the edits, it reverted to “preserve aspect ratio”.
Why? I know that the configuration is stored in text files. I have Cygwin installed, so the obvious thing to do was to take a
look at it with Emacs.
Tried to start it.
C:\Users\grog\AppData\Local\DxO_Labs\DxO OpticsPro 10\Presets>emacs 'emacs' is not recognized as an interneal or external command,
operable program or batch file.
Huh? Has my cygwin flown away? Tried cygwin. Nothing. bash. Nothing.
How did I start the thing? Spent 20 minutes looking through the Cygwin
documentation, The Quick
Start Guide for those more experienced with UNIX seemed to be a good start. But
although it was offered as a “User's Guide”, it was all about installation, not starting.
Finally, in despair, gave up and asked on IRC. Simple, says Daniel O'Connor: click on the
Cygwin shortcut (whatever that may mean) on the desktop (whatever that may mean). In good
Unix, that's find the Cygwin icon on the root window and click on it. But why should I have
to do that? That's not the Unix Way.
So finally I got the thing started. How do I access Yvonne's config file? It seems that,
although I have root administrator privileges, I can't access
Yvonne's AppData directory. Finally gave up, printed out my file, and over to her
machine to compare.
There was no configuration file! The Mysettings shown on her DxO window was in fact
an empty directory. And DxO was far too considerate to worry her with error messages. So
basically I just needed to save the configuration again with the modifications I wanted.
Aren't computers wonderful? But out of this I've gained some information about how the
configuration files work. The amateur version of DxO Optics “Pro” doesn't have the editing
feature, but I'm sure you can still edit the configuration files, if you can only find them.
Clearly the curly braces are for grouping things, and almost everything is inside
the Base group. Looking at my config file, the important entry seems to be:
CropRatio = -1,
It's at the top level inside Base.
Comparing with DxO's own preset files, it seems that there are a lot of defaults. I can't
find a CropRatio entry in any of them, for example.
It's frustrating that it only appears on the console. I had hoped that the numbers at the
beginning of the line would give some indication, but they don't make much sense. The first
could conceivably be the PID;
the second is the same as last
time.
On IRC, the opinion was that it was a hardware issue, maybe overheating. But the fact point
elsewhere: I have an xearth process complaining about a missing marker file every 5
minutes. And the hang message is always at the bottom, so it must appear round the time I
power on the monitors. It has been warm lately, but I haven't had any problem at any time
except when turning on the monitors. Could there be an issue there? It only happens on
X server 0, so I'll try leaving server 1 running
when I turn off the monitors.
It's a supplementary handle to hold the camera, and the screw is to screw into the central
flash contact—not exactly something I'd want to do, for a number of reasons. I'm still
trying to decide whether it would be of any use.
Another GPU hang this morning! This time I had left X server 1 active, but once again it was server 0 that was affected. Something to do with
the clients? Ran ps and got hundreds of processes, including 300 zombie ssh-agents, to be inspected later.
But that's not the way you solve a problem nowadays. Google is your friend. And, as it proved, also the enemy. This article describes a
(perhaps only marginally) different issue, but it makes it clear that the problem is
associated with Google chrome. That may
mean that it's sufficient to shoot down chrome, and not the X server, if it happens again.
They were supposed to start on the frame of the house on Friday, but there's no evidence of
that. Only Jim Lannen, who has now completed the preparations for installing the meter:
Yvonne went out riding with Chris today, and for some reason
used Chris' camera rather than her own. Back home it was up to me to read in the card, for
which I didn't have a cable. And at least 3 images didn't get read:
Nov 9 13:48:31 eureka kernel: (da4:umass-sim2:2:0:2): READ(10). CDB: 28 40 00 00 22 00 00 00 10 00
Nov 9 13:48:31 eureka kernel: (da4:umass-sim2:2:0:2): CAM status: SCSI Status Error
Nov 9 13:48:31 eureka kernel: (da4:umass-sim2:2:0:2): SCSI status: Check Condition
Nov 9 13:48:31 eureka kernel: (da4:umass-sim2:2:0:2): SCSI sense: HARDWARE FAILURE asc:0,0 (No additional sense information)
Nov 9 13:48:31 eureka kernel: (da4:umass-sim2:2:0:2): Retrying command (per sense data)
Nov 9 13:48:32 eureka kernel: (da4:umass-sim2:2:0:2): READ(10). CDB: 28 40 00 00 22 00 00 00 10 00
Nov 9 13:48:32 eureka kernel: (da4:umass-sim2:2:0:2): CAM status: CCB request completed with an error
Nov 9 13:48:32 eureka kernel: (da4:umass-sim2:2:0:2): Retrying command
But that was only one of the images. The others were created, but they were stored with
length 0. Tried it on lagoon, Yvonne's computer. Same thing.
Could this be a FreeBSD problem? What does
Microsoft think? Tried it on dischord, and it had no trouble reading the images.
Except that some of them were junk: it couldn't display them. Clearly a problem, possibly
file system metadata, which Microsoft is too considerate to complain about.
So Chris tried, both with Microsoft and Apple. And they could even display the
broken JPEGs. She copied them to my system,
and I still couldn't display them. Is this good or bad? I need to investigate what's
really wrong with the card.
Over to the site this morning to see how they were getting on with erecting the frame. They
weren't. They were scheduled to start last Friday, but by this evening nothing had
happened. Duncan tells me that they're still on track to have the frame up by Friday this
week.
One of the reasons why things are happening so fast might be that by the time the frame is
up, we'll be due to pay half the total cost of the house. Received an invoice for the first
progress payment (“Base stage”) from JG King, with the usual accuracy:
Chris Bahlo wasn't as successful today. She received a planning permit a couple of weeks
ago, and last week I dropped in her application for a septic tank permit. And then the
planning department decided that she needed to submit an application for amendment to the
planning permit. Why? Because the house plans she had submitted were mirror images. But
she told them that, and made it obvious on another sheet of the application. So the
application as approved contains a contradiction. Why did they approve it? And why is
it her fault that they didn't do their job right? Chris is understandably pissed
off.
The exposure meter that I bought from eGlobaL last weekfinally arrived. Only
took a week. Yes, they said “three to five working days”, but that's what most people say
for parcels inside Australia, and usually they're there in two. The problem is that eGlobal
are in Hong Kong, something that you
have to dig to find out. Is it worth it? I still don't know.
Finally got round to reconsidering solar electricity. Spoke to David Grigg of BREAZE, phone number 0401-763-553. He certainly knows
his stuff, and we spent 50 minutes on the phone (roughly doubling my normal monthly phone
bill). The big question for me was pricing, and he helped there:
For battery operation, we also need a DC/DC regulator ($1,500) and a (Selectronic)
inverter charger, $6,000.
The batteries themselves are
currently lead-acid gel
batteries. To maintain a battery life of 12 to 15 years, they should not be
discharged by more than 30%. He quoted a price of $9,500 for a 48 V, 660 Ah set. That
corresponds to 31.6 kWh, but at 30% it's only 9.5 kWh.
So: doing the sums, it would seem that the battery setup would cost $17,000 on top of the
standard solar panel system. For that I get a maximum of 9.5 kWh per day, or about 42 MWh
over 12 years. At $0.30 per kWh, that corresponds to a savings of $12,500 odd.
Is that worth it? On the one hand, electricity prices will go up, so things may not be
quite as bad. On the other hand, LiIon
technology is just round the corner, and it could be a lot cheaper. And
maybe finally the governments of this country will come up with better incentives for
renewable energy. For the moment, though, it looks as if it's a bad time to install a
battery backup system. I'll put in the separate power circuits, but for the moment it looks
like a better idea to have a big UPS there rather than batteries.
So what do we do in the meantime? I can't use 5 kW sustained power for any length of time.
It would make more sense to put in, say, 2 kW and add later if we get batteries. Even then
much of the power would go to waste or to Powercor, which, at $0.08 per kWh, is pretty much the same thing. 2 kW is too little
for heavy consumers like dishwashers and ovens, but they're used so little that it doesn't
make much sense to install a higher capacity just for that. 1 kW of panels cost $1,000.
Running a dishwasher might use 2 kWh, so we'd have to run it 1,500 times to make up for
that—and we can't rely on the sun shining when we want to wash the dishes.
So: what do we get for 2 kW? Price is round $3,500 + installation, say another $1,000. For
that I get, say, 1 kW for about 8 hours a day, and another 1 kW for another 3 hours not of
my choosing. Over 10 years that's a savings of about $8,800—if I can use that
power. Somehow I get the feeling that the whole thing isn't worth the effort.
A builder must not demand or recover or retain under a major domestic building contract of
a type listed in column 1 of the Table more than the percentage of the contract price
listed in column 2 at the completion of a stage referred to in column 3.
Those sums are 10% for the base stage—I paid 20%, and 15% for the frame stage—they're going
to ask for 25%. How can they get away with that?
Spent some time looking round. There are no less than three empty spaces in the walls. I
had noticed them long ago in the display home, and they're clear in the plans, here in the
kitchen at top and right, and at bottom left:
It's been nearly 50 years since I last
bought an exposure meter. Things have changed since then, and today I spent a bit of time
getting used to my new Sekonic L-308S. No great surprises, since I had read reviews and instruction
manuals. But I'm still puzzled by some of the design decisions.
It seems that it's standard practice on modern exposure meters to append decimals to
aperture values. Thus there are 10 steps between f/2.8 and f/4. They have the approximate
values in the left-hand column, but the exposure meters display the base stop and a decimal,
as in the right-hand column:
2.80
2.8
2.89
2.8.1
3.00
2.8.2
3.10
2.8.3
3.21
2.8.4
3.32
2.8.5
3.44
2.8.6
3.56
2.8.7
3.69
2.8.8
3.82
2.8.9
3.95
4.0
The list doesn't end exactly at f/4 because f/2.8 isn't exactly one stop from f/4. Clearly
these numbers are irritating, and they don't match the ⅓ stop increments that most cameras
use, which in this case are f/3.2 and f/3.5 on my camera, though more exact thirds would be
3.15 and 3.55.
Does this make sense? At some level, I suppose it does. I might find out when I get a
flash unit with digital power control. In the meantime, though, the meter has an irritating
“feature”: you can select between tenth stops, half stops and third stops. But the tenth
stop setting also displays (only) whole stop increments in shutter speed, and the others are
clearly lower resolution. Fortunately the EV display is always in tenth stops, so I can
work around it like that.
Apart from that, it's pretty much as expected. I now also have the first opportunity to
measure flash guide numbers exactly: hold the meter 1 m from the flash source and measure.
The aperture display corresponds directly to the guide number. I need to measure the
distance more accurately, but the first attempt confirms that yes, my ring flash probably
does have guide number 14.
Processing today's photos with DxO Optics “Pro” showed an
interesting issue: I couldn't load presets. Why? Here I have tried to set a black and
white preset (as the most obvious kind), and the image remains in colour. But the two
smaller previews are black and white:
Yvonne is very taken by him, and there's a good chance that,
despite my protests about the cost, she'll buy him.
Amy also brought an amazing amount of buffalo cheese and yoghurt, must have been over 2 kg:
they run Shaw River Buffalo, making
dairy products. We'll have
real Mozzarella for our pizzas for a
long time to come.
It's been a few days since I
last had the dreaded “GP process hung” message. Maybe a web page that I have since removed
from the browser state? But today I got another message, not once but dozens of times:
Google confirms that this, too, comes from chrome. Unlike the other message, though, it wasn't
associated with an X hang. It seemed related to
some issues I had mounting an SDHC card; the
message output came in bursts related to inserting and removing the card. Somehow there's
strangeness in the FreeBSDUSB stack.
The Panasonic microwave oven that I bought seven years ago has died, presumably magnetron failure. Hardly worth replacing,
given the rapid progress in domestic appliances. So I went looking for a new one,
preferably one that isn't designed for illiterates (the Panasonic doesn't have a numeric
keypad; instead you have to push buttons up to 10 times to count).
And it's amazing how few ovens there are. When I bought the last oven, I expressed my hope
that inverter technology would give more even heating. That hope was fulfilled, so I went
looking for a new one with inverter technology, expecting it to be universal. Far from it,
at least if I can believe the descriptions. Finally settled on a 1200 W Sharp model, and in to town to look for the best price,
first at The Good Guys.
But the Sharp was far smaller than the old oven. Looked around and found a suitable looking
unit from LG. But it didn't say “inverter”. On
the other hand, it wasn't in the specs of the high end Panasonics either, just written on
the door. Why can't people write specs any more? Maybe the others also have inverter
technology.
Finally found a salesman who seemed to know what he was talking about, and he confirmed that
only those marked “inverter” really had inverter technology. Why? It can't be the price.
I ended up buying a Panasonic NN-ST641W, pretty much the same as the old oven, for $162.
The LG I was looking at came $30 more expensive.
And how does it compare to the old one (NN-ST666W)? The controls are very similar, but
there are now only 6 power levels instead of 10, and instead of being numbered (something
that is clearly alien to Panasonic's target clientele), there are names: High, Med-High,
Med, Med-Low, Def and Low. It almost reminds me of the CFA's bushfire danger levels. On the positive side,
it now displays the power level when heating; previously if you forgot the level, the only
way to be sure was to stop the oven and set it up (presspresspresspresspresspress) again.
But the naming suggests that they had run out of LED indicators on the panel.
And then there's the (moisture) sensor. I had already established that it's useless, but
the new oven doesn't have it at all. On the other hand, it does offer the ability to set
three different times and levels in succession, though I can't imagine what good that would
be. And after heating, it carries on running the fan for a couple of minutes, along with
the display “Cool”. Is this progress?
Sometimes I think that Yvonne and Chris don't want me to come
riding with them. My riding helmet disappeared years ago, and they lent me a substitute—for
a while. But it belongs to David Yeardley, and we no longer have access to his things. So
while in town, bought a new helmet. I'm amazed how little choice I had, but I found pretty
much what I was looking for.
The Colac-Ballarat
road, particularly
round Enfield between here and
Ballarat, is not in the best of conditions. People have been complaining about it for
years, though objectively it's not that bad. Earlier this year it was voted the worst road in the state, apparently because a truck overturned on a “particularly nasty bend of the road”.
Now the state elections are coming up. And what do we get? They're completely resurfacing
the “particularly nasty stretch”.
Of course it's just a coincidence that this is happening just before the state elections.
But more to the point, this “particularly nasty stretch” wasn't all that bad; my guess is
that the truck was going too fast for the bend. The worst part of the road is to the north
of the area that's being resurfaced, with large subsidences of up to 20 cm. And they're not
doing anything about that (yet). After all, nobody rolled a truck there.
The frame of the house is due to be finished by the weekend, but so far they've only done
one day's work. Today a bloke showed up with
a Bobcat and evened out the
soil round the house, then covered it with stones:
Finally got round to measuring the guide numbers of my flash units today. I have a total of
9, but the mecablitz 58 AF-1 still
needs to be repaired, I haven't worked out how to manually fire the two tiny Olympus flashes
that came with the OM-D
E-M1 and E-PM2 units, and I didn't think about the 40
CT 4. That left the two 110 Joule
studio flash units, the mecablitz 58-AF-2, the Meike MK-300 and the Viltrox JY-670.
The test setup was simple enough: place the flash on a stand (here the mecablitz with
Yvonne's E-PM2) one metre away from the flash unit and
measure the intensity of a manual flash at 1/250 s and (in case it makes a difference) at
1/125s.
There are quite a few measurements to be done, since each unit has variable power output.
The results were amazing. None of
the flash units came close to the rated guide number. Here's a summary of the full power
output from the details page:
How can that be? It wasn't the measurements: the photo above was taken with the studio
flashes bounced off the ceiling, and the exposure suggested by the exposure meter was
exactly correct. On the other hand, I've been complaining for years about flash
(under)exposure problems, and these measurements support my observations.
The Meike MK-300 was particularly disappointing, but even the mecablitz only managed 40%
of the rated light. One possibility is in
the wikipedia article:
guide numbers of units with a focused reflector have a virtual origin behind the unit.
Well, it's what they call a guide number. A real guide number is the aperture times
the distance from the flash, and that's what I measured. But it's clear that I'm going to
have to try again with, say, 5 m distance, at least for those units that can flash that far.
The raw output from the exposure meter tests wasn't very helpful. Here the readings for the
MK-300:
1
11:3
1/2
11:2
1/4
8:9
1/8
8:1
1/16
5.6:5
1/32
4:5
1/64
2.8:0
1/128
1:9
The first column is the power level, and the second column is the exposure meter reading, in
aperture and tenths. That's really difficult to interpret. After a bit of playing
around with a calculator, decided that the issue was complicated enough to warrant a
program, which produces output like this:
Power ratio
Raw GN
Real GN
Expected
Deviation %
1
1
11:3
12.21
1/2
0.50000
11.0:2
11.79
8.63
87
1/4
0.25000
8.0:9
10.93
6.10
221
1/8
0.12500
8.0:1
8.28
4.32
268
1/16
0.06250
5.6:5
6.66
3.05
376
1/32
0.03125
4.0:5
4.76
2.16
386
1/64
0.01562
2.8:0
2.80
1.53
237
1/128
0.00781
1.0:9
1.37
1.08
60
The first and third columns are the input data. The “real GN” is converted into real
numbers, and the last two columns compare the measured data to the data expected based on
the full power output. As this one shows, the Meike MK-300 is wildly inaccurate. When set on 1/32 output, it produces roughly 1/6
of the full power. But the others are only marginally more accurate; the mecablitz 58-AF-2, a high-end flash unit, deviated by up to 50% from the calculated
values.
I've already mentioned that the Sekonic L-308S calculated
the correct flash exposure for the photo above. That's what I bought it for, so it's a
success. But it has some really strange limitations. I've already mentioned this strange
tenths-of-a-stop readout, but it gets worse: there's this coupling between EV steps and
display. I have three choices: 1/10 EV aperture steps and whole EV shutter speed steps, or
½ or ⅓ EV steps both for shutter speed and aperture. Neither is really satisfactory.
Probably ⅓ EV steps is best for natural light, and I can get by with the 1/10 EV steps for
flash because the shutter speed isn't so important, though of course it makes it impossible
to set the normal flash shutter speed of 1/320 s.
But what about the shutter speed value with flash measurements? Some of my flashes have flash
durations of more than 4 ms (1/250 s) on full power, but none of my measurements showed
results for 1/125 s that differed significantly from the 1/250 s results. Does it really
just measure for that period of time?
A phone call this afternoon from a woman with a south-east Asian accent, telling me she came
from some debt collection agency, and wanting my personal details. She seemed to be
surprised when I refused.
Gradually it became apparent that she was following up with the repair to the damage to a Telstra telephone cable last August. As I noted at the time, they only wanted to know where the fault occurred, and not a
billing address. So it seems that they sent the invoice to Stones Road, and of course it
couldn't be delivered. Instead of calling me on the phone—they had those details—they
turned over the matter to a debt collection agency. And they wanted money today.
Sorry, I don't know who you are, and I have no business with you. If Telstra wants money
from me, they should send me an invoice. But no, the agency wanted to send me email and get
me to call them back. No. So she put me on hold and finally came back, agreeing to my
request.
Would you expect any less from Telstra?
A month later I had still not heard from Telstra. I've heard from others, too, that
they never received a bill for legitimate damage. Is this yet another example of their
incompetence, or were the people who called me just scammers? I may never know.
That's a horse fly net, a cushion, a sheepskin and the remains of some polystyrene
packaging. Fortunately the only thing that he destroyed was the polystyrene, but it does
mean that it's difficult to leave him alone.
Into town early this morning to pick up an ALDI special offer: car radio, reversing camera and GPS navigator. At $299 it was on the
expensive side, but the good thing about ALDI is that you can return things within 60 days
if you don't like them. In this case the problems started (don't they always?) with the
documentation, in this case the advertising. No description of the mounting, just a montage
of the radio and the navigator display:
There were at least four of them at the same checkout as I was. At $30 it wasn't even
cheap. What's the attraction? Those things went out of fashion 50 years ago.
Back home and took a look at my new toy. That wasn't a montage! The display slides out of
the top of the radio, so it really sits directly above it.
Is that a good idea? The obvious thing is to try it out. Where's the documentation? 33
pages, including 6 pages of safety information, 4 about the remote control and 6 on
installation. And half a page about the GPS navigator, the only part of which was any use
was the information that the manual could be downloaded from http://www.ndrive.com/sites/default/files/UG_V11_PND.pdf. But the safety
information included the following message, no less than four times:
But the worst thing of all was the connections. Nowadays you'd expect every aftermarket
radio to have a DIN connector, like my last ALDI radio did 5
years ago. But no, it has another custom connector:
Can I be bothered? It makes it easier to first decide whether it will fit in the position
it would operate in. And that would obstruct the clock, a couple of controls and the
central heating vents:
The position is also probably too low, though it would be worth checking. And it would
almost certainly point straight back, so at an angle to the driver. Is it worth trying? It
would certainly be less messy than the current arrangement. But before I can try, I need to
either find a real adapter for ISO plugs, or make one myself. I can't be bothered, and I'm
sure that others won't either. A pity: it could have been a good unit.
Spring is drawing to a close, and we've seen dozens of baby rabbits run around the area.
Today Leonid saw one too, close to the corner
we call the “eagle's nest”.
And he caught it almost immediately.
But the details were very different. Niko's rabbit was fully grown, while Leo had caught a
baby. Niko let me put him on a leash and take him home; Leo didn't let anybody near him,
and he played around with it like a cat does with its prey. Niko took over an hour to eat
his; Leo had completely eaten it before we got home. And of course, for once in a while I
didn't have a camera with me, so we don't have any photos.
Last Friday I
measured the light output of many of my flash units, and was surprised by the results: the
guide numbers I measured represented between 15% and 50% of the claimed light output for the
units. One possible explanation was that I measured all units at a distance of 1 m, which
may not be typical. So today I gathered my remaining flash units and measured the guide
numbers of all of them at 2 m and 4 m.
The results are interesting. To a
certain extent they bear out the hypothesis that the guide number is lower at close
distances (because of the virtual origin of the beam), but even at 4 m they're roughly 1.5
times less than the advertised guide number, or a little under half the power output. With
the exception of the Meike MK-300, that's relatively consistent. But it explains why I have to set my
camera to +1 EV flash compensation.
Call from Zali O'Dea this afternoon: they were on the Stones Road property to pick up some
firewood, and they wanted us to come over and bring the dogs.
Problem: I was flat out doing my flash tests, and Yvonne and
Chris were off recovering from the Yeardley's latest trick (turning off the water to Chris'
property). But I didn't want to be rude, so over with Nikolai for 10 minutes or so: he hasn't had much running
around lately.
Earlier this year I
got involved with the Dereel History
Group, run by Julie Donaghy. It clearly needed some structure, and Scott Weston kindly
donated space on his server for a Dereel
History Wiki. We put in a couple of pages, and I tried to interest people to add to
it, but somehow the whole thing went to sleep.
Then I saw Julie last month, and
she was talking about reviving the group, and bemoaning the fact that nobody had signed up.
So I waited, and last week she posted—on facebook—a request to get together on Tuesday to discuss the book she wants to
write. A number of people replied, but today she decided to cancel the meeting:
look its ok Lora Ardran Sharon Hegarty Breen Greg Lehey, Caroline Everett we might leave
it til early nexdt year for now, as i really need to do loads of photocopies, and get the
website up to date, this i can do on my own. then i shall need a lot of help after that
i have a photographer Lisa Hunter to do the photos of any interesting places around
dereel, i have a scribe who offered to do any typing, once i work out what other tasks are
required i can ask for help then so thank you all, so tuesday is cancelled
Photocopies? That's a sure-fire way of reducing the quality of documents which are already
in not very good condition. Web site? Which web site? I think she's talking about
her blog, which is limited
to her own use. There's no evidence that she has even tried to use the real web site.
A scribe? Is that something like a secretary who types in what other people have to say?
That's not going to work, and I said so. And why have only one photographer in a
collaborative project?
From there things went to hell in a handbasket. How dare I criticize Julie? I don't
know Lora Ardran, but she needs a good dose of disinfectant in her mouth:
who the fuck do you think you are Greg Lehey. ... talking to Julie Donaghy like that you
fucking wana be know it all .....with you head stuck so far up your iwn arse I can't
believe you are still alive ?????? WTF
It's amazing that people will express themselves like that in public, but I suppose Facebook
has a clientele that hasn't been through the flame wars of the 1980s and 1990s. One day it
might come back and bite Lora. Julie, who also runs the Facebook group, deleted the thread,
but the NSA have it, and so do I.
Still, it looks like nothing will come of the history group. Julie hasn't updated her blog
in any way that I can see in the eight months since I put up the wiki, and if she wants to
do everything herself, that's not a group. But I should learn from my experience with the
Friends of Ballarat Botanical Gardens and be
glad that I'm not involved. It's just a pity for the history of Dereel.
So today I tried again, aided by this
recipe and this one. Neither are ideal, but the first one has some interesting ideas. And of
course the google search returned many others, including my own at position 6. There was
also this
recipe, with the startling statement:
The first thing we did was to brine the chicken in 9% brine for 6 hours (actually close to
7). This is to make the chicken more juicy because the osmotic pressure in the cells of the
chicken will change, and holds more water.
I wonder where they got that idea from. The brine removes water from the meat. In any
case, I've never seen a recipe like that.
In the end I cooked the chicken whole in the cooker for 2 hours at 65°, and then one
drumstick for another hour or so at 80°. That proved to be the right thing to do: the
breast and that drumstick were done the way I wanted them, and the other drumstick and the
wings were underdone. At least I'm getting closer, and I've updated recipe accordingly.
Yvonne's monitor died today. Not the first monitor that has
died on us, but since moving to LCD screens, there seems to be only one failure mode: the
power supply dies.
Why? We've had switching power supplies for decades, and while they have certainly evolved,
you'd think it was well understood technology. LCD digital displays, on the other hand, are
only now receding from the bleeding edge. You'd really expect them to be more likely to die
than the power supply.
The other obvious thing is that it's hardly worth getting somebody to repair it. A
replacement costs $129 at Officeworks,
and repair technicians can easily take that even for a simple repair.
Over to the site this morning with the generator, to find only a single person there
attaching the fascias. He didn't want the generator, so I took it with me.
In the afternoon over again to meet Karl Waldron, the air conditioner man. The Actron units aren't without their down sides: they
have a minimum air flow of 140 l/min, which is a lot when cooling a single bedroom. So
we're going to have to have a “spill”, an extra outlet, somewhere near the reverse air. I'm
not overly happy with that, but I don't see much choice.
While there, noticed that the CFA had been
there to burn off our excess timber:
Into town this morning to pick up a new
monitor for Yvonne, and while I was at it picked up a second
cheaper one—$129 for a 21" 1920×1080 monitor—to be prepared for the next
failure. My Tandem background
must be showing.
Later did some searching about the causes of monitor failures. Things aren't quite as
clear-cut as I thought, but articles like this one suggest
that power supply failure is one of the leading causes.
I've been trying since Friday to get an National
Broadband Network connection for Stones Road. Called up Aussie Broadband and spoke to Fabien, who told
me that the trouble is the lack of a location ID. It seems that NBN has assigned location
IDs to all houses, but since we don't have one (yet), they don't have a location ID. Why is
it taking so long to assign one? NBN can be slow.
Called up the NBN and spoke to Jane, who asked me for the geographical coordinates—good
thing I had them—and told me that it could take up to a week, but would probably be here by
the end of the week. Hopefully; I hadn't expected this to be a hold-up, and we're going to
be wiring the house soon.
I've already commented on
the abundance of roses we have this year, and there's a particularly pleasant view from the
bathroom window. But how do I get a photo? So far I have failed. There are two problems:
firstly, I can't open the bathroom window, and both dirt and reflections are a problem. And
secondly the photos I did take just don't show the abundance:
To add insult to injury, I tripped while removing the camera, and gave the camera a knock
which seems to have disrupted something in the viewfinder. There's a slight green flare to
the right of the centre. It seems to be related to the illumination, and it doesn't show up
in the images, so it's not that big a deal, but it's still annoying.
This was unrelated to the knock. The camera in question was an Olympus OM-D E-M1 Mark I, a
model that has an issue with the viewfinder burning if exposed to the sun under the right
circumstances for even a second. More here
Some years ago we saw
a Western ringneck in the
garden. They don't come from round here, and we wondered whether it was an escaped cage
bird. We've seen it—always alone—several times since, and today while taking some
photos of Yvonne and Keldan, saw it again:
Not much happening on site at the moment. According to the schedule that Duncan gave me
last month, the
roof should have gone on today. All that was to be seen was the handrail (presumably put in
yesterday, as scheduled):
What kind of Xanthorrhoea? I had
thought Xanthorrhoea
australis, but that is clearly quite a different species. Finding details isn't easy,
but possibly it's Xanthorrhoea minor.
I wonder what other surprises the nature strip has in store.
So what kind are the Xanthorrhoeas
in Stones Road? It seems
that Xanthorrhoea
australis has prominent trunks below the grass, and these ones don't. And the only
photos I've found show different flowers. It seems more likely that
they're Xanthorrhoea minor.
But the obvious thing to do would be to take some photos of the flower spikes, which look
like this:
They're not the only Xanthorrhoea species we have. In the Kleins Road garden we have
a Xanthorrhoea
quadrangulata, which unfortunately has already finished flowering. The good news is
that there are plenty of seeds, so we can take some with us when we move:
So now I have my shiny new Sekonic L-308S exposure
meter—just what I need to take the photos of
the Xanthorrhoeas. It showed f/64.9
at full power, so I turned it down to ⅛ power and took a photo at f/22. That should have
been 1/16 power, except that the lens extension reduced the effective aperture by about one
stop. But I wasn't prepared for what I got:
What went wrong there? Shutter speed (1/250 s) was OK, but just to be sure I reduced to
1/80 s, with no change. More power? Finally I increased the power to full power, which
should have been enough for an aperture of f/64, but it was only barely more than enough at
f/11:
That's a difference of 5 stops! What caused that? The flash went off, so it couldn't be
that. The exposure was set to manual, so it couldn't be the camera. The aperture was the
smallest, so it couldn't be a problem like the stuck diaphragm that I had a couple of months ago. What
was it? Could it be that the flash synchronization was completely incorrect, and the flash
either went off before the shutter opened, or long after it closed?
Still puzzling, went to Stones Road to take photos of the other Xanthorrhoea. Same
distances, same settings, and complete overexposure:
Yvonne decided to go
to Melbourne tomorrow to visit
Equitana. We had really wanted to plan it
with an overnight visit, but somehow that didn't happen, so Yvonne decided to go just for
the day.
Under those circumstances it probably doesn't make sense to go by car. 270 km at 10 l/100
km is 27 litres of petrol, or about $40. Then there's the pain of navigating through
Melbourne and the probably horrendous parking fees. Wouldn't it be simpler to go by train?
What does it cost?
After fighting http://www.vline.com.au/ for 15 minutes, I still
didn't know. I was able to establish the timetable, sort of, but nothing I could do
divulged the fares. In desperation called the phone service on 1800 800 007, where Mark
told me the fee immediately: $12.74 return. That's amazingly cheap. But why is the web
site so completely broken? Mark confirmed that he couldn't find it there either, and that
the site to use is http://ptv.vic.gov.au/. Now why didn't I think of
that? Put in a formal complaint; I wonder if it will help.
Later took a look at the PTV web site. I couldn't find it there either, just the reference
to a myki.
It seems that you can't just buy tickets any more. But it still doesn't explain why you
can't find the fare. Oh brave new world, that has such people in 't!
The roofers were there again today, and spent a surprising amount of time adding battens to
the roof structure—it seems to have taken as long as erecting the entire frame and roof.
Yvonne went by in the evening and found the roof half
mounted; it seems that that's the easy part.
We still don't have power. Called Red
Energy, and Carl told they couldn't find my application (put in by the electrician)
unless I gave them the date. What kind of organization is that? Called up Jim Lannen, who
told me that the inspector had been there, required him to make some changes, and then taken
the application to forward to Red Energy. Somehow this isn't going as smoothly as I had
expected.
Another call from Duncan with the next milestones: mid to late next week they'll wrap the
house in foil (this I have to see), and they'll also do the windows in the course of that
week. Plumbing rough-in will be on 28 November, and they need the electrician to do his
rough-in at the beginning of the following week. They'll also need 20 l of water for the
plumber to test his joints, and if the power still hasn't been connected, he'll need
a generator on Friday. Called up Jim again and arranged for him to do the rough-in,
starting round Monday midday. Why do I get the feeling that it's going to be Tuesday?
The week after next is also the best time to install the antenna for the National Broadband Network, so called up Aussie Broadband, who had promised to call me
back when the NBN allocated a location ID. That was yesterday, but by this evening they
hadn't called, so called up, spoke to Hayley, and arranged for the installation, hopefully
in the week of 1 December. Application number is 141 830.
Over to Stones Road mainly to take my weekly house photos, and to my surprise found the
roofers stil hard at work. By the time I left, round midday, there wasn't much more to be
done:
It's fairly clear now that I'll postpone solar electricity until I can store it more
efficiently, or until there are more government incentives for solar electricity—something
that could happen as a result of the upcoming state election. In the meantime, it seems to
be a good idea to use
a UPS to supply the
circuits that will later be battery fed, so that we can weather the interminable power failures. Spent some time today
thinking about electrics in general, without coming to any firm conclusions. I need to get
my act together.
A few days ago I bought a
spare monitor, in case one failed. But presumably it would be of better quality than my
older monitors, so swapped out my oldest monitor for the new one. I can use the old one for
machines that have previously been headless.
That wasn't as simple as it sounded: yes, they have the same resolution and pretty much the
same dimensions, but the old one was connected
by HDMI, and the new one doesn't have an
HDMI connection. So I ended up having to move monitors around, and since the other monitor
with HDMI had a different resolution, I had to restart (but not reconfigure) X.
But that was all. Was it worth it? Yes, I think. It has more contrast and brightness than
the old one, so it's more suited for photo processing. And finally people are marking the
locations of the controls visibly on the front, so at least I can find them when I need
them.
This progress means that the “frame stage” is complete, and that's worth another 25% of the
house price. So of course a bill arrived. We've now paid half the total price; wouldn't it
be nice if the other 50% took the same amount of time?
Today the infrared filter that I bought on eBay arrived. What do I do with it? Tried some sample photos from the verandah. Unprocessed
they're particularly uninteresting:
In the process, noted that the image exhibits the dreaded “hot spot” syndrome. That's
dependent on the lens, but how do I find out which lenses exhibit it and which don't? I
need a different filter for nearly every lens I have.
Time to read up and try a few different processing tricks.
The bad weather was accompanied by lots of thunder and lightning. At one point I heard a
spark right in front of me—it sounded as if a monitor had arced. But a second later there
was a clap of thunder, so it must have been a lightning strike. Surprisingly, nothing was
obviously damaged.
And then an hour later two of my monitors died. It took me a while to realize that the rest
of the system was running
on UPS, and that
something had tripped the circuit breaker in the switchboard.
In the afternoon, wanted to make a phone call. Phone was dead. After a lot of checking,
discovered that yes, the phone was dead. So was my new Linksys PAP2T that I got only 4 months ago. Damn.
That's not the first CiscoATA to die on me. In
fact, it's the fifth. Nine years ago two died in similar
circumstances, and the two replacements also died later for no apparent reason. In every
case there's just no power indication, although the (external) power supply is functional.
But why did the devices die? Since both of them were phones, you'd expect that the problem
was with the phone line. But the connections aren't direct: the phone line goes into the
other ATA (undamaged), from there to the phone (damaged), and via the dual line circuitry to
the other ATA (damaged). Is that really what happened?
So: Once Again I need to buy a phone and an ATA. Neither is simple. I paid $25 for the
Linksys 4 months ago, but the supply seems to have dried up, and I'm not sure I want to buy
one anyway. The cheapest I could find was an auction for a NetComm V210P, the same as the
other ATA. But that won't be over until tomorrow.
The other issue was the phone. It has two lines and a headphone jack, both of which are
important for me. But they don't make them like that any more. While tidying up the
garage, I had noticed an old dual line phone, but it has one obvious disadvantage:
It also has a less obvious disadvantage, the real reason why I stopped using it: it doesn't
have a headphone jack.
So why can't I get anything like that any more? There
are VoIP phones on the market, of course,
but that's a whole different level of either complexity (how do I connect a set of portable
phones to it?) or restrictions (you don't). Found one dual line phone
offering Bluetooth connectivity. That
sounds like an excellent alternative to a corded headset, but is it compatible with my
headset? If not, how do I find a compatible one? In addition, it's used and from the USA,
where they have different mains voltages. More complexities.
It's been nearly 50 years since I last dabbled in
infrared photography. I took one film, developed it, and, I fear, never printed it. Now
things are completely different, of course, especially as I don't have the equivalent of
infrared film: the cameras all have filters to block infrared light, significantly
increasing exposure times. Yes, they can be removed, but first I need to know if it's worth
it.
The results with my OM-D
E-M1 were not spectacular. There were a number of possible reasons: the presence of
the infrared block filter in the camera, the high speed (45°/25,000 ISO) with which I took
the test photos, and the poor lighting, which may have preferentially blocked infrared
light.
But what about the other cameras? We also have Yvonne's
E-PM2 and my old
E-30. How do they
compare?
Out to the paddocks in reasonable light, but was pursued by Estrella, the “orphan” foal whom
we are hand-feeding, and she wouldn't leave me alone. To another paddock, where I had
marginally less difficulty keeping horses away, and took three pairs of photos, at 12 mm and
60 mm focal length.
The results weren't at all what I expected. Each camera produced very different results,
and it seems that the E-30 is the most sensitive. Here comparisons:
What is not immediately apparent is the exposure: the E-M1 and E-PM2 needed about 8 seconds
at f/8 (EV 3) for the close-up, while the E-30 needed only 1 second (EV 6). On the other
hand, the E-M1 and E-PM2 both calculated the exposure correctly, while the E-30 underexposed
by about 3 EV (giving an exposure of 1/10 s for the first images, not shown here). The
exposures on the left are less consistent because of the differences in the sky.
None of these look good, of course. I need to follow up on how to process these images.
Call from Tom of Bushmans today—for about
5 seconds. Then the connection was dropped. That wouldn't be that unusual, except that it
happened twice yesterday to other callers. Was there something wrong with my
other ATA? Tried using
a straight telephone without ATA, and it worked fine. Checked later: Daniel O'Connor called
me on the Telstra line (passthrough through the Netcomm V210P) and
on VoIP via the same ATA. Yes, it's
repeatable: the VoIP circuits work well, but
the PSTN connection repeatedly gets dropped
after 5 seconds.
That clarifies a number of things: first, it supports the hypothesis that the damage came
in through the phone line. Secondly, it makes the V210P less attractive, especially since
the auction price exceeded my generous bid of $41.50, while I can get the Linksys ATA (2
lines) for $25. So I went for that.
What does that mean for PSTN? I still can't make up my mind whether I should even have a
landline in the new house. Telstra have
put up their prices again, so I'd end up paying $25 a month for a line that I would only use
in emergencies. In addition, I'd need to pay $299 for connection, and get somebody
to dig a trench to the cable. For that price I could buy an antenna and use mobile
telephones for the few times when I needed it.
And Tom? They'll deliver the water tanks on Monday.
Also finishing the specs for the electrical wiring of the house. At one point I thought
that it was trivial to find cheap glass fibre. The more I look, the more it seems to be a
can of worms. The only reason to put in fibre now is to save on installation expenses
later, if I ever need it. But it's not clear that I ever will,
and Cat 6A will be
enough for 10 Gb/s, so I think I'll just take the easy way out and forget it for the time
being.
We have a number of geriatric domestic animals. Our cat Lilac will be 18 years old in January, and when Chris
Bahlo moved in she brought two cats (TC and Shadow) and a dog (Vito), all in advanced years,
but all in reasonable health.
Vito was clearly going deaf, and he was nowhere near as mobile as he was, but yesterday he
still found the energy to bark at me as I went by. But this morning when Chris went out,
she found that he had just crawled into a corner and died:
And although we've agreed for double glazing, the panel beside the door is single-glazed.
There's nothing in the contract about that.
The shed hasn't been started. Spoke to Norm Baker and discovered that the rain on Monday
had softened up the soil again. He's hopeful it will be dry enough by the end of the week.
While looking round the house, found a strange marking in the pantry:
The GPU hang messages that I had earlier this month are back. They had caused me to restart X, and were thus particularly irritating. But the seemed to
come from Google Chrome, so tried
shooting it down. Success! Another X session saved. And probably another indication that
it's high time to upgrade my system software.
Walking the dogs this morning (without a camera), we were followed by another dog: Nina,
Chris' other Maremma. Now that
Vito is dead, she's on her own and looking for company. Leonid didn't take kindly, and barked at and threatened
her. We'll have to work on that.
Chris' house is almost ready to start construction. Over with her to her building site this
morning to meet her site supervisor, Chris Stoel, who came along with a hired measuring
wheel—$36 per day, when you can buy digital versions for $49. Spent a surprising amount of
time trying to locate the house site, not helped by the terrain, electric fences and
incorrect site plans:
Finally all was done. It seems that Chris S comes
from Oss
in Noord-Brabant, and he's a cousin
by marriage with Mari Hendriks. Sounds like a reasonable bloke, but very different in
character from Duncan Jackson, our site supervisor.
Chris doesn't have any sheep. Whose are they? David Yeardley's. That's his property on
the other side of the fence in the third photo. He doesn't have proper sheep-proof fencing,
and he just lets them wander onto her property with no regard for property rights or the
welfare of the animals. People like that shouldn't be allowed to keep sheep.
It seems that the “wrapping” (green) is simply part of the wall structure, mainly for
insulation purposes, and it's there to stay. They've also put some wooden structures in the
bathrooms:
We still don't have any electricity! Called Jim Lannen again, who confirmed that he had
spoken to the inspector, and that the paperwork had been submitted a couple of weeks ago.
Called Red Energy again and spoke to
Seima, who told me that the application had been received on 24 November 2014, and that it would be connected within 20 working days. How I hate
bureaucrats!
Also heard from Aussie Broadband that my National Broadband Network antenna
would be installed next week. Do the installers know what will await them? On the one hand
it's the correct time to install the cabling, but we won't have any walls to attach
the NTD to. Asked Aussie to confirm
with the installers, but no, they have no contact with the installers, who work for NBN.
But then I got a call from a Walter Bonilla from the NBN, who told me what they needed and explained that they had exact instructions on
what to do, and they weren't allowed to diverge from them at all. I explained that the NTD
just needed to be screwed to a wall, but no, he said, there's a metal frame that goes with
it. So I checked: there are two screws that go into the wall, no frame to be seen. Then
Walter admitted that he didn't know the details, and he would get the techies to call me a
day or two before they came. Finally broke out of the bureaucratic loop!
As a result of construction activity it's been a few weeks since we've let the dogs run free
in Stones Road. But now it looks as if they won't be in the way, so back over there again.
The grass is really high, and one of the advantages of the soil is that it's still green,
where just about everywhere else it's going brown, like Chris' property (second photo):
One of the nice things about the Olympus OM-D E-M1 is that it
can take up to 10 full-size photos per second, and I frequently take several in a second,
for example when taking horses or dogs. The trouble is that the standard settings display
the image just taken for 0.5 s before reverting to the normal viewfinder. In that time you
can lose your subject.
Finally fought my way through the excuse for an instruction manual to find out how to change it. What's the function
called? I couldn't think of anything useful, but clearly it should be in the “Display”
menu. But I didn't find anything there. I recalled that the E-30 also had a function like
that. Looked through that manual too. Finally I found it in the setup (spanner) menu with
the completely non-intuitive name Rec View. While I was at it, also found the
“Built-In EVF” menu (J), which sets various info settings and also displays more information
in the viewfinder. You have to find it out yourself. This appears to be the sum total of
the documentation:
It seems that Style 3 is the default, and that Style 1 also shows further information at the
bottom of the viewfinder. Style 2 is the same as Style 1 except that the information is in
monochrome.
It seems that they had been following the thread, but didn't make any comment. They gave
Reinhard a PDF document („Blitztipp 17“, which I haven't found on their web site). He
quoted (slightly reformatted by me):
Die meisten der im Handel erhältlichen Blitzbelichtungsmesser sind für den professionellen
Einsatz mit Studioblitzanlagen konzipiert. Dadurch kann es in vielen Fällen bei der
Leitzahlmessung von Elektronenblitzgeräten zu Fehlern kommen. Die Gründe hierfür sind
folgende:
Zur Bestimmung der Leitzahl eines Blitzgerätes gibt es ein genormtes Messverfahren,
welches für ein Blitzgerät mit einer bestimmten Lichtleistung einen einzigen
Normleitzahlwert ergibt. Für die normgemäße Kalibrierung von Blitzbelichtungsmessern ist
dagegen ein Eichungsbereich zugelassen, dessen Grenzwerte einen Unterschied von ca. einer
dreiviertel Blendenstufe ergeben. Gleichzeitig ist gemäß der Norm für
Blitzbelichtungsmesser zusätzlich eine Messungenaugigkeit von ca. + ½ Blendenstufe
zugelassen. Das heißt, im Extremfall kann die Anzeige von zwei verschiedenen
Blitzbelichtungsmessern um ca. 1 ¾ Blendenstufen unterschiedlich sein und doch
entsprechen beide der Norm. Daraus geht klar hervor, dass mit normalen
Blitzbelichtungsmessern keine genaue Messung der Leitzahl von Blitzgeräten möglich ist.
1.1 Praktische Erfahrungen:
Messungen der Leitzahlen von Blitzgeräten mit verschiedenen Blitzbelichtungsmessern haben
in der Regel zu geringe Leitzahlwerte ergeben. Das heißt, die Hersteller von
Blitzbelichtungsmessern kalibrieren ihre Geräte nach anderen Kriterien, als zur Messung
von Blitzgeräte-Leitzahlen, in erster Linie für den Einsatz im Studiobetrieb. Bei
Leitzahlmessungen mit Blitzbelichtungsmessern entstehen oftmals zusätzliche Fehler
dadurch, dass bei den Messungen ein zu geringer Abstand zwischen Blitzgerät und
Blitzbelichtungsmesser verwendet wird. Durch die dann kurzzeitig sehr hohe Lichtstärke
des Blitzgerätes wird der Fotosensor des Belichtungsmessers übersteuert. Dies führt zu
einem zusätzlichen Fehler in der Anzeige. In der Regel wird dadurch ein zu geringer Wert
am Blitzbelichtungsmesser angezeigt. Hiervon sind Blitzgeräte mit kurzzeitig extrem hohen
Lichtstärken besonders betroffen.
In summary: most flash exposure meters are intended for studio use, and they can be
inaccurate in measuring guide numbers. The reasons are: the standard defines a calibration
range of ⅔ EV, and in addition meters are allowed an inaccuracy of another ½ EV. It's not
completely clear what they mean here, but it implies that in extreme cases two exposure
meters that meet the standards can show readings that differ by up to 1¾ EV.
In general, guide number measurements with flash exposure meters show too low a value,
because the manufacturers calibrate the exposure meters for different purposes. If a
powerful flash is too close to the exposure meter, the sensor can be overloaded.
Does that make sense? It certainly explains my results. But I have difficulty in believing
that exposure meters with a resolution of 0.1 EV can have an error in either direction of
0.88 EV. The standard may allow it, but the customers wouldn't. So far the exposure meter
has shown itself to be much more accurate than the TTL exposure produced by my TTL flash
units and my cameras. But the people on the Olympus forum are surprisingly hostile to the
idea that anything could be wrong, so I'll have to do some more testing.
Also got the confirmation that the drain in the laundry went through the frame. Here
yesterday and today (in fact, two sequential images in the camera):
Spoke with Martin, who had been given plans with additional markup that I hadn't
approved, like gas-fired central heating. Who did that? And why? No difficulty getting it
changed, but it was a good thing that I was there.
Also issues with the placing of the solar panels for the hot water. Somehow we had
forgotten that they have to face north, and there was concern whether they would fit
(they're three panels 1.20 × 2.4 m, with 10 cm in between, for a total area of 3.8 × 2.4
m). After some measurement, it seems that they will fit.
Back in the evening, where we discovered a number of things that need attention. The gas
connection in the kitchen (on the wooden cross-member) is at the extreme right, which is not
where the stove will go:
And that probably means that now is the time to put in a similar cross-member to hold the TV
holder on the wall. The outlets for the showers are also surprisingly low, about my nose
height. We need to check what the shower equipment will look like, but there's a good
chance that we'll have to lift them. And, of course, we've found more places where we need
power points.
State election day today. Nothing that unusual, except that it was much more in evidence
than before, and, it seems, campaigning wasn't as dirty as usual.
Who to vote for? A year and a
half agoDenis Napthine, the
premier, promised Dereel a mobile phone
tower. Where is it? Even talk about it seems to have subsided. I'm surprised that nobody
mentioned it in the course of the election campaign.
In fact, it shouldn't have been that unexpected. It's Max Mitchell, our new neighbour from
across the road, whom we've known from Olivaylle
for decades.
Lately while walking the dogs, we've me a new bloke who seems to have a similar schedule.
He has a Dachshund cross, which makes
for some amusing contrasts:
The other dog to whom we're trying to get our dogs accustomed is Chris
Bahlo's Maremma Nina.
Leonid had shown some aggression, so we
decided to let them run free in Stones Road. No aggression. In fact, very little interest
on either side; Maremmas obviously have a completely different personality
from Borzois:
Summer doesn't start until tomorrow, but it seems to have got its calendar mixed up. Today
we had a top temperature of 36.2°, and Yvonne and Chris spent
all day carting hay and moving horses:
Do you have a comment about something I have written? This is a diary, not a
“blog”, and there is
deliberately no provision for directly adding comments. It's also not a vehicle
for third-party content. But I welcome feedback and try to
reply to all messages I receive. See the diary overview for more details. If you do
send me a message relating to something I have written, please indicate
whether you'd prefer me not to mention your name. Otherwise
I'll assume that it's OK to do so.
