We knew when we bought him that he had an overbite, clearly visible in the second image, but
I had been concerned about his lack of “canine” teeth. But they're there, and it seems that
it's normal for a dog in his age to have such small canines: they're still milk teeth, and
his adult teeth will look very different.
In the past we've made shepherd's pie out of
the remainder, but that's a little boring. How about something Mediterranean
like Moussaka? So Yvonne bought some
aubergines, and today I set to making it.
What a disaster! Preparations took 2 hours, mainly salting, washing, drying and grilling
the slices of aubergine. And of course the quantities in the recipe (from Australian Woman's Weekly) were all
wrong, and we didn't have nearly enough aubergines. Interestingly, the web site has 10
different moussaka recipes, none of which match the recipe in the cookbook. This one comes close, and like
so many asks for a 425 g can of tomatoes—in a country where tinned tomatoes come in tins of
400 g or 800 g. Were they once different in pre-“metric” times?
My best guess since then is that this is the content of an 800 g can, just to confuse
people.
The “Béchamel” sauce in particular
looked very dubious. For 1 litre of milk they wanted not only 100 g flour, but also 2 eggs.
My recipe uses half the proportion of flour, and
of course no eggs. What to do? Yvonne opined that we should just leave it out and coat
with potato slices instead. Did that, and it didn't taste bad, but what a lot of work!
Here's the recipe as we would do it next time, if
we ever do it again.
One of the nice things about eBay is that the
prices are generally lower, right? But somehow it doesn't always seem to be like that. A
couple of days I bought a second-hand Olympus Zuiko Digital ED 8
mm f/3.5 fisheye lens on eBay. After comparing the new price (US $799 from B&H Photo), the $455 I paid (including postage) seemed reasonable. I use B&H
as a reference because they're relatively cheap without being unreliable. Of course, the
price isn't directly comparable: there's at least $42 shipping, and so the whole thing
translates to AUD 890, almost double the price I paid.
I'm still waiting for it to arrive, but in the meantime eBay keeps popping up “This might
interest you” suggestions. One did: the same lens, brand new, for $1,860.08! That's well
over double the price that B&H charge. And then there's somebody selling a M.ZUIKO DIGITAL
ED 14-42mm F3.5-5.6 EZ, $349 at B&H, for AUD $6,500! The seller even has an
“explanation”:
Why are you offering this item for over $6000.00 when it doesn't worth that much?
qd_au: We have increased the price dramatically for items that are out of stock and
are currently under replenishment. Providing fast shipping is one of our top
priorities. We want to reduce the chances for buyers to wait for back orders as part of
our commitment. If you are really interested in this item, feel free to send us a message
:)
That's nonsense, of course. B&H have it in stock. And nobody's going to pay anything
like that price just to get it right now. In fact, it's one of the lenses that I was
considering for Yvonnea couple of months ago, and as I
noted at the time, there are plenty of alternatives.
So why are they doing it? The most likely reason seems to be to make it easier to increase
the prices people are prepared to pay for other auctions. The whole thing has an unpleasant
odour.
What am I missing here? You'd expect big international organizations to be correct in their
statements. Yes, maybe growth will drop, but the exchange rates tell a different story.
That doesn't look like the way a dog (or presumably a fox) would eat a rabbit. In
particular, it didn't eat very much of it. Today we returned and found less of it left, and
more fluff instead:
Australia doesn't have many carnivores. We guessed at
a Wedge-tailed eagle, made
easier by our recent
discovery of an eagle's nest not far away.
In the afternoon, visited our neighbour-to-be Graeme Swift, who suddenly became very
agitated because one of his (very small) dogs disappeared. The property is well fenced-in,
so there was no danger of him getting away. But it seems that Linda is terrified that an
eagle will get him.
Clearly no wedgie is going to carry off one of our dogs. But what about cats?
Theoretically it could, I suppose, but there are eagles round where we live now, and we
still have our cats. Probably Linda is just a little overly concerned.
My last attempt with sous-vide cooking was only
marginally successful: as I later discovered, ALDI's recommended 75° is far too hot to cook chicken breast. Jeff Potter's
Cooking for Geeks recommends a range between 60° and 65°:
OK, that made more sense. I've always had difficulties getting chicken thighs cooked
to the point where they come off the bone easily, and this looked like the solution. Given
the presence of a bone, it made more sense to cook longer, so I gave it 4 hours at 63°.
Here's before and after:
What went wrong there? Inaccurate temperatures? Indeed there was a discrepancy: although I
had set the cooker to 63°, spot checks showed that it thought that the temperature was 60°
or 61°. But the infrared thermometer showed pretty consistent temperatures within about
0.3° of 63°. Based on the method of measurement, if there were any error, I'd expect the
reading to be lower than the real temperature.
I think the real culprit is elsewhere: despite the extreme importance of exact temperatures
in sous-vide cooking, the documentation is full of errors. I've seen the first example in
the ALDI documentation. But that's only part of the story. SousVide SUPREME™ publish a table recommending 60-63° for white meat and 80° (!) for dark meat. Jeff doesn't make a
distinction, but it seems that he is thinking of breast.
The numbers are temperatures
in Fahrenheit
and Celsius. So soft-cooked eggs are
cooked either hotter or colder than hard-cooked eggs? What kind of sense does that make?
But the real problem was the rendering of the web page, which left out the times (minimum
and maximum):
But the “quick” solution isn't sous-vide cooking at all! The egg isn't supposed to reach
the ambient temperature. You can do the same thing even faster by raising the temperature
to 100° and cooking for 6 minutes.
In summary, then, it looks as if I'm going to have to spend a lot of time experimenting with
the temperatures. For the moment it looks as if I just chose the wrong kind of chicken for
the temperature in my first two experiments.
For the last 9 months, since his operation, Zhivago has been eating
special pellets to avoid a recurrence of
his uroliths. It's very expensive and
doesn't taste good. We talked about it with Pene on Tuesday, and she recommended giving him
normal food again: it took 8 years for the problems to occur, and it's unlikely he'll live
to 16. In the meantime we can keep an eye on what he eats.
Sent a message to Ron Frolley today. It didn't get delivered:
<ronfrolley@bigpond.com>: host extmail.bigpond.com[61.9.189.122] said: 552 5.2.0
Mexs1o00h1sUVRc01extSj Suspected spam message rejected. IB703 (in reply to
end of DATA command)
Next question: who are you sending the mail to? Gave him the email ID. Who is that? A
friend of mine. Are you near Ron's computer now?
Round about this time I asked to be transferred to his supervisor. Sorry, not until we
check your computer and ports. You should be connecting to mail.bigpond.com on ports
587 or 465, which proved to be the numerical equivalent of submission
and smtps. That doesn't work, of course:
=== grog@www (/dev/pts/0) /usr/local/etc/postfix 14 -> telnet mail.bigpond.com submission Trying 61.9.189.249...
Connected to mail.BIGPOND.com.
Escape character is '^]'.
ehlo lemis.com Connection closed by foreign host.
I tried to explain to him how SMTP works,
and where the problem was detected. I don't think he was listening. Once again I asked for
a supervisor, in retrospect a silly thing, since I've already established that, if there are
supervisors, they never get involved. Matteo told me that his supervisor was nowhere near
him, and he'd have to go halfway through the building to find him. Some supervision, eh?
But finally he offered to escalate the matter. That at least is an improvement over my
last interactions with them, where I
specifically noted the lack of escalation procedures.
Once again he asked for my Big Pond account number. After an explosion on my part he
finally gave me the reference number 140703-3820287 and told me they would monitor the
connection 24×7, whatever that is supposed to mean. Based on my experience, that'll
probably be the last I hear of the matter:
They have never, ever, solved a reported problem.
Maybe it's better for my blood pressure if they don't call back.
And the problem? It seems that in the last two days, since I last sent a message to Ron,
they've decided that digitally signed messages are spam. Removed the signature, and it went
through. Despite all my distaste for Telstra, I think that the real techies might be
interested in that.
Yes, the first photo is out of focus, but these are just experiments.
The panoramas were interesting. I tried the verandah
centre panorama, for which I need 20 images with the Zuiko Digital ED
9-18mm F4.0-5.6 lens: two rows exposed at 45° increments, and a single row at the top
at 90° increments. It's clear that both of the lower rows can be replaced by a single row
at 90° increments. And the third row should be replaced by a single image pointing straight
up.
Running Hugin was interesting: it
doesn't seem to know how to recognize a fisheye lens—not this one, anyway—and I had to tell
it manually. That meant I couldn't use the “assistant”, because it doesn't give you the
chance before trying to align the images. And for some reason it seemed to think that the
image taken straight up was taken with a different lens with a different focal length.
Finally corrected that and got a result that was less than optimal:
Yes, the coverage of the lens was enough to cover the roof of the verandah—almost. But the
field of view is rectangular, not circular, and I really needed a second one at right
angles. More to the point, though, the roof doesn't fit the rest of the image. Further
investigation showed that the control point detector had only found two control points
between the roof and the other images, both on the same image and very close together, at
bottom centre in the first image:
Made a second attempt with a total of 6 images, and this time the control point detector
didn't find any control points between the roof images and the others. Clearly
there's more work to be done here.
The next step in the house construction: we now have our septic tank permit. All we need
now is a new planning permit and two building permits. Still, it seems like progress.
There's a good reason for that: both are forest plantations. Still, it's a change from the
virgin forest that we're used to, and there's that too. Walked almost up to
the Ballarat-Colac
road, and then back to the car, about 3 km:
After a lot of trying to straighten it, I ended up wedged between the tree on the right and
a ridge on the left. No way to get out without outside help. Went to the houses along
Westons Road. At the first nobody was home, at the second they weren't able to help, and at
the third people were home, but there was no way to attract their attention. Why do so few
Australian houses have doorbells?
Gradually it was getting dark, so bit the bullet and walked home, a total distance of 6.8
km:
The tow bar was the obvious place to attach the rope, but pulling sideways might not be the
best thing for it: they're designed for longitudinal stresses. In the end decided that the
best would be to go into the paddock behind the car and pull it out from there. For that we
needed the permission of the owner, who today was fortunately there, and more than prepared
to let us in. Chris' tow rope proved too short, and Steve (the owner) produced a chain to
connect to it.
That was all we needed. A quick pull back by a metre or so, and the thing was done. Thank
God for that!
My last two attempts with sous-vide
cooking weren't unqualified successes. And today we had planned
a Girello beef roast for dinner. Again,
what temperature? What time? I was pretty sure that 53° would be right, since that's how I
cook the meat in the oven. With some difficulty found some more links and consolidated them
on a sous-vide page. In particular this page, though a little busy, suggested 50° for rare (“not recommended”) and 55°
for medium rare, which rather confirmed my expectations. But for that I needed to cook
either 2.5 to 3 hours for thinner cuts or 4.5 to 6.5 hours for thicker cuts. It seems that
there's a no-man's-land between 3 and 4.5 hours.
In the end I decided on the 6.5 hours; after all, what difference does it make? The result
showed that a considerable amount of reddish fluid had escaped from the meat in that time,
and the result didn't look overly rare:
Yes, it's not as rare as I wanted. Next time I'll do it less. But how much less? Are 2°
or 3° enough? We'll see.
In passing, it's interesting to note a few quirks of the cooker. It has a timer, but it's
not clear what good it is. Clearly it needs to come up to temperature before you can even
start the timer, and it keeps that temperature regardless. So you don't need the timer to
cook. What good is it? After it's finished, it beeps and turns off the cooker, but the
water temperature stays relatively constant, so it doesn't stop cooking. I suppose you
could consider it a marginal energy saving device.
The other oddness is the temperature display. When you turn it on you set the temperature,
to the nearest degree, but when you turn on the heating it switches to a time display
(static until you turn on the timer). To see the temperature, you need to press the
“Function” button, and then it shows the temperature for a couple of seconds before
switching back to the time display again.
But the temperature display is wrong! Or at least, that's the way I see it. Every time I
tried, it showed a temperature of 50° or 51°, though the infrared thermometer showed 53°,
and a meat thermometer in the meat after completion showed 52.6°:
As if that wasn't strange enough, they've been dismantling Chris' horse infrastructure to
half fence in their new animals, without so much as a by-your-leave. And today it seems
that borrowing the Landcruiser today wasn't allowed, and there was more bad blood. Enough
is enough. David is leaving for Singapore in a week or two, and Tuyết goes
into Ballarat all day Tuesday. So we're
planning to move out Chris, her dogs, her horses and her furniture on that day. It'll be
interesting to see how Tuyết copes with what's left.
The first question is: what does “focal length” mean with fisheye lenses? It's a
rectilinear concept. Maybe it works if you consider the immediate area round the optical
access, but the comparison with the 9 mm lens shows how little that means:
Focal length:
8.00 mm
9.00 mm
Horizontal FOV:
94.47°
87.73°
Diagonal FOV:
107.04°
100.49°
Vertical FOV:
78.19°
71.68°
The real diagonal field of view of the fisheye is 180°, but you can't even calculate that
based on the conventional formulas. And the reality looks very different:
The obvious thing to do there is to mount the camera horizontally, as shown here. That's
something I haven't done for a long time. And that showed another problem. When mounted
horizontally, the optical axis of the lens must intersect the horizontal rotation axis, like
here:
But since I last did that, I've changed my rails, on the one hand to give sufficient
clearance between lens and rail, and on the other hand to compensate for the offset tripod
mount of the Olympus OM-D
E-M1. And now I can't position the L bracket low enough. It fouls the bottom of the
panorama bracket:
Still, for the time being it'll work, though I see a more sophisticated panorama bracket on
the horizon.
Next question: how many images? I've already established that it's
considerably fewer. In practice, there's only a certain number I can choose: angles of 45°
or 60° or multiples thereof. 180° is clearly too few; but what about 120°. It almost
works, but the panorama preview showed missing areas:
They're not missing in the final image, but I didn't notice that until later, so dropped to
90° increments (4 images per row instead of 3). That certainly improves the height:
Other points of interest: the Olympus Viewer doesn't seem
to apply any distortion corrections. Either distortion isn't the issue in fisheyes that it
is in rectilinear lenses, or they don't know how to correct it. And Hugin doesn't calculate the focal length
correctly. It reads 8 mm from the EXIF
data, but when I change then lens type to “fisheye” it changes it to 10.487 mm. And if I
leave it like that, the control point detectors have problems. It seems that a better
“focal length” is 7.78 mm.
Finally, the minor differences make surprising differences in the output. I took a total of
7 shots of the “verandah east” view: 3 at 120° and 4 at 90°. They were all taken without
moving the tripod, so I was able to choose which images to use. Here's a view with all 7
and then only the four at 90°. Run the cursor over the images to
compare with the partner:
Somehow the area immediately to the right of centre hasn't been correctly joined in the
first image.
Apart from that, though, things worked “normally”. I was able to stitch the “verandah
centre” panorama by not pointing the lens directly upwards. Given the mapping of a fisheye
lens, it's difficult to see which of these two images was taken with the fisheye. Again, run the cursor over the images to compare with the partner:
Came into the lounge room this afternoon to find evidence of dogs having been misbehaving.
My back-scratcher and the mouse for teevee were both on the ground, superficially
undamaged. But the mouse didn't work. And then I found out why:
Chris and Yvonne were going to bring a couple of horses over
here this afternoon. It didn't quite work out the way they expected: David Yeardley had
decided that he needed
the Landcruiser. What for?
To steal firewood from Chris' property! I can't believe he's stoop so low. I was going to
go over and take photos, but Chris didn't want me to.
Over to the Stones Road property with CJ Ellis this morning to look at the fences more
seriously. In the process confirmed that the “rear fence” was in fact an incomplete low
fence about 20 m from the real end of the property, and that the real boundary fence had
many fallen trees on it, and that about a third of it was missing, presumably never built.
It's all in dense forest, and just getting machinery there will be a problem
Spent some time trying to decide on what to do, including the possibility of Yet Another
boundary fence at the end of the clear area, but in the end decided that we'd have to bite
yet another bullet and get Stewart Summersby to come and clear the area so that we could put
the fence where it belongs.
By contrast, the rest of the fencing was fairly straightforward, and we should be able to
start putting up respectable fences some time next week. Called up Stewart and Craig Mayor,
who will also be able to cut down the trees to the immediate west of the house.
Called up John Willowhite at Powercor. He
had previously offered us single phase at 40 A or dual phase at 25 A. 25 A isn't enough to
drive an air conditioner. Could he give us dual phase 40+10A? No, he couldn't do that, but
he noted the problem, and agreed to supply single phase 50 A. That's still almost nothing.
And can he get the meter connected? No, for that we need a retailer. Went out looking:
should I stay with my current retailer (Red
Energy)? Why not do some comparisons?
Off to Google to see what was there. Not surprisingly, there were many hits. Surprisingly,
none of them detailed what the tariffs were, not even the Government web site. All wanted to start with
my current electricity bill. What happens if I have just arrived in Australia (but not by
boat)? No bill, no idea.
Apart from that, each came up with its own strangenesses. The Government web site came up
with offers that represented kWh prices as low as 18.2¢, including all additional costs. No
explanation. Almost as soon as I filled in the form at Energy Watch I got a phone call from a girl with a
penetrating fake American accent, who promised to have a connection tomorrow—until she heard
that we didn't have a meter. Sorry, can't help you there. Come back when you have a meter
(and an electricity supplier). Why should I?
As if that weren't bad enough, ten minutes later I got a call from somebody else from Energy
Watch. No, thanks.
Then there was http://youcompare.com.au/, who came up with the least
number of suppliers, but at least mentioned the prices per kWh. But no sum! Only “$796
1st year saving*”. That reminds me of this stupid “60% fat free” on foodstuffs.
I understand that iSelect recommends plans from a range of providers on its Approved
Product List. []”
But at least they came up with most of the known suppliers, on a narrow display with space
on both side and a clever scrolling method to ensure that you can't compare easily, spread
over 14 subpages.
OK, so how about the retailers themselves? You'd think it would be a lot more work, but
given the reluctance of the “comparison” web sites to provide details, there's not much
difference.
In fact, as I found from Red Energy,
there's very little difference. I couldn't find their rates there either. I have come to
the conclusion that the entire electricity market is run by cowboys.
Into Ballarat this morning to meet Jo
Forbes and discuss colours for the new house. Based on prior experience, I had expected it
to take a long time, but in fact we got through relatively quickly. About the only thing
that we were still uncertain about was the flooring. We had chosen wood-finish vinyl, but
comparing it to the flooring in our current house, we were no longer so sure. Back to
Delta to reconsider, and
ended up with fake stone instead.
While at JG King, also spoke with Tom
Tyler. I had sent him a list of errors in the plans last week, but—despite our insistence
on quick processing—he hadn't done anything about it yet. Next time I'll come in and see
him personally. Spent longer going through that than we had with Jo, but hopefully there's
nothing else to worry about now.
The first two were taken with the fisheye and converted by Hugin to an
equirectangular and cylindrical projection respectively. Looking at the two, cylindrical
looks closer to what the 9 mm lens sees (last photo). But what's the field of
view? Hugin doesn't know either. It claims a horizontal field of view of 179° for
the first and 360° for the second, clearly both wrong. It's even difficult to get an
approximate crop of the image, but it would look something like this:
That crop is 2356×2033 out of an original 3403×2678, or 69%×76%, That's a diagonal crop
factor of √(2356² + 2033²) ÷ √(3403² + 2678²), or about 72%. Since we're talking
linear concepts with focal lengths, that corresponds to about 72% of 9 mm, or about 6.5 mm.
More experimentation with sous-vide
cooking today, and still more to come during the week. What would be more suited to
sous-vide than Hainan Chicken
Rice? It's one of the few dishes that is traditionally cooked below 100°. So I found a
recipe in the Codlo cookbook and followed
it. Once again chicken thighs, but this time deboned and cooked at 65°—not the conclusion I
came to last week.
But that's the part of the issue: I don't have a good feeling for how much difference 2°
makes. So I followed the recipe.
The results weren't bad. Juicy meat, though the liquid that came out of it was a little
reddish. The real issue was that the recipe didn't include the herbs and spices that my
traditional recipe uses. Time to write down my version.
I've already commented that the timer on a sous-vide cooker is pretty useless. Today I
confirmed: 15 minutes after removing the food from the cooker, the water temperature had
dropped from 65° to 59°, and after an hour it was still 35°.
But there was also a surprise at the beginning: I put the chicken in before the bath had
come up to temperature. Don't do that! When it finally came up to temperatures, I measured
temperatures of up to 70° in places. I took the chicken out immediately, and I don't think
it had any affect on the food, but it's clearly something that shouldn't be done.
Lately Leonid has been somewhat destructive.
He has been chewing on various things in the lounge room, such as the bouquet of dried
flowers in the fireplace on the right:
I'm continuing my experiments
with sous-vide cooking, and today we
tried a piece of salmon according to this
recipe, somewhat derived from one in the Codlo cookbook. Codlo wants the fish to be brined first to avoid albumen secretion,
but we didn't have time for that. On the other hand, the piece had skin on it, and this
proved to be sufficient: there was no leaching.
The taste was amazing. You can eat salmon raw or cooked normally, and it tastes good either
way. But here it was so soft it almost melted in your mouth. For the first time, I've seen
results from sous-vide cooking that I've never seen done any other way.
But wait! There's more! Yvonne bought
some ossi buchi while shopping
yesterday, and we're planning to eat them on Saturday. Several people have claimed that
they're ideal candidates for sous-vide
cooking. But how?
Went looking on the web and came up with a number of opinions. This recipe wants to cook them
at 137° F (corresponds to an unsettable 58⅓°) for 24 to 36 hours. This one, a suspect URL that may change, wants 61°
and 48 to 72 hours. This
one wants 62° for 72 hours, and notes that the result is flaky, dry, and apparently to
his taste. And this one describes
his experiment of cooking at 60°: he removed one bone after 12 hours and found it excellent,
while the second, cooked for 22 hours, was too long.
What can I make of all that? Sounds like more experiments. And none of them go into the
issue of how the flavour of the meat suffuses
the soffritto. One leaves it
out altogether, but most of the recipes put it in with the bones, which means that they,
too, would cook for up to 72 hours, well beyond what makes sense for vegetables. So for
this attempt I've decided to cook them without the soffritto and prepare that at the end
along with the juices from the bags.
Also had another issue with the vacuum pump: there was liquid in the bags, and it sucked it
out, too:
It's difficult to see, but the channel above the end of the bag is full of wine. It makes
sense when you think about it, but clearly it means that you should hold the bags vertically
when sealing.
Since temporarily giving
up on shepherd last month,
I've been using the electronic programme guides that are broadcast with the TV multiplex to
plan my TV recording. They've made me appreciate shepherd. What a pain! In some
programmes the description gets mixed in with the title, the categories are mainly empty,
both of which make it really difficult to present the programme in an easy-to-understand
way. And to add insult to injury, they keep changing the start and end times, so unless I
keep an eye on them all the time, I miss recordings:
My ossobuco cooked all day. But not quite as
planned. When I came into the kitchen this morning, the cooker had turned itself off.
That's possibly because I didn't start the silly timer in a reasonable period of time. So I
turned it on again and came up with further issues:
Although some recipes specify cooking times of 72 hours, possibly more, the designers of
the cooker have decided that the timer can be set for a maximum of 24 hours. That
doesn't make things any easier for the programmers: the counter runs in hours and
minutes, so a maximum of 99:59 would have been easier to achieve. One more example of
bad specs.
Once the cooker is running, you can't change anything. When the timer threatens to run
out, you have the choice of waiting until it does or just turning it off and setting a
new temperature and time.
As I've noted, when the cooker is turned on, it heats considerably beyond the target
temperature, by about 3°. That is completely uninteresting when the food is first put
in, because it'll cool the water back to temperature, but in the case I described, there
seems to be no way to avoid a temperature spike. Presumably it won't do that much harm,
but it's irritating.
Called up Momentum on 1300 662 778 and
selected “New Connections” from the phone menu. Was connected to Daniel, who told me he
would have to connect me to New Connections, where Sal told me effectively the same
information as Red Energy. Called up Stewart Summersby. Yes, that's the way they do
things. He'll start work on the connection on Monday. But until I get a supplier to
connect the meter for me, I can't even compare prices! What kind of consumer-friendly
situation is that?
Into Ballarat this morning with
Nikolai and Leonid for dog training. Ian and Sandra both recognized
Tanya, until we told them that it was really
Leonid.
This was my (and Niko's) first training session since late February, and Niko was very
agitated at the start. He calmed down after a while, and maybe that's a reason to continue
training for a while, so that he can socialize better. I can't think of any other reason.
Clearly it's high time to make more accurate measurements of my domestic electricity
consumption. I've already established that eureka, my main computer, along with
other components that are powered on all the time, use about 130 W, and this can increase to
150 W or more when eureka is busy. dischord, the Microsoft box, uses another
130 W by itself, while the monitors use about 120 W—a total of 400 W or so when everything's
running.
But what does that mean in terms of power consumption? I'll know in a week or so when I've
had them running that long. So far, though, the consumption is far less than I feared.
Finally the Ossobuco is cooked! How did it
taste? OK, but possibly not as good as cooked normally. The meat was firmer than normal,
but quite juicy, unlike the experience of this
person or this person. But
Yvonne thought it too firm, not disintegrating enough. That
can be changed, of course, with higher temperatures, but do we want to? It takes a lot
longer than conventional methods.
It's been a very mild winter so far, but today Yvonne told me
that there had been a mild frost early in the morning; it was gone by the time I got up.
And the minimum temperature that my weather stations measured so far this winter is still
+1°.
It's finally time to fix the fences on the new property. The west end has a fence
for about ⅔ of the length, but lots of tree branches have fallen on it. Today CJ Ellis
headed down to remove them so that Stewart can come in with his bobcat and tidy things up.
It's quite heavily wooded:
Looking at the first photo, and others like it, it only later occurred to me that Craig
was doing all the work and Stewart was just standing round looking stupid.
Later heard that Craig is moving back
to Bannockburn. That's a
double problem: firstly, it'll make him less available, and secondly it means that his wife
Leah will no longer be able to clean the house for us. Damn.
One of the minor irritations I have
with VoIP via my NetComm
V210P ATA is that I
still have to use the phone keypad to make calls. Wouldn't it be so much easier to click on
a number on the screen and have the ATA dial it? Of course it would. And as far as I know,
nobody has ever implemented such an obviously good idea.
But then, the ATA can store frequently called numbers and call them up via an abbreviation.
Can't I store the number in one of those registers via a web application? It turns out that
the answer is “yes, sort of”. The instructions tell you how to do it. You fill out a web form:
But what do the fields mean? The instructions are vague. After some experimentation, it
seems that the “Number” field is the same as the “Position” field, filled with leading
zeroes to make up 3 digits. And URL? The instructions say:
Enter the URL, VoIP Phone Number, Remote WAN IP Address of VoIP Gateway
It seems that this is normally a phone number, but it can be an email address or an IP
address with an optional port number. The instructions show a variety of formats, none of
them URLs. One example is:
Example 1: Position: 0, Name: "test@iptel", Number: 000, URL: test@iptel.org
When the user dials the Number 000, it will dial to the VoIP User test who registers to
the SIP Server iptel.org.
000 is the Australian emergency number! Thus there's no way to call it with this device.
Isn't that a clever thing to do in a device intended for the Australian market?
Still, this can make life easier. Reserve a specific number for a “quick dial”, put it in
via the web interface, and dial it via a dedicated “saved number” button on the phone.
In passing, it's interesting how the web form limits the utility of the feature. The ATA
can store 140 phone numbers, but the form will only display 10 at a time:
Over to Stones Road this morning to see how Stewart was getting on. Not badly, it seems.
He had cleared the whole front of the property, so CJ can get started there:
The southern half is less work. All he needs to do is to clear a few tree trunks, and he's
done. But then, just after midday, he left. No idea when he'll continue.
One wasn't registered, the second didn't do sheds, and the third gave me voicemail with the
encouraging information that he was a plumber.
Also called up a bloke who had called me a while back, sent by Widespan. At the time he said that he would come here,
though he's from the other side
of Melbourne. Today he said he
wouldn't. So I still don't have a builder.
Finally got on to Anthony Steinman about the shed. Yes, he can do it, will cost about
$4,000, and he won't be able to do it for at least 7 weeks. In that time I can get my
owner-builder permit. Started filling out the application, which requires all sorts of
things, even a statutory declaration. That won't be done before Monday. How these things
drag!
In fact, he lived in the same house before the O'Deas moved in, but he didn't update
his address. Given the experience that we had with him, that's not surprising.
So I did that a couple of days ago, and got a ballpark figure of $2,000 to $3,000, compared
to Stewart's $4,500. That's significant, and I told Stewart so. Now today Stewart
suggested I ask him about wiring the house too. Why? I thought Stewart was really keen on
doing it.
In any case, along came Jim and did his measurements, and was agreeable to wiring the house,
so now at the last minute it looks as if we'll have a change of electrician.
Finally got on to Peter O'Connell about bridging finance for the house. A little late, it
seems. I had understood that I should wait until I needed the money and then contact him.
Now I find it will take up to 3 weeks, by which time some big bills will be coming in.
Hopefully we won't run into cash flow problems.
Last week my panoramas with
my new Olympus
Zuiko Digital ED 8 mm f/3.5 fisheye lens were less than stellar. With a little
checking, found out why. My entrance
pupil page includes tables for the equipment I was using at the time. Since then I've
changed to another temporary setup and not updated the page; I knew that the 7 mm position
on the old rail (for the Zuiko Digital ED
9-18mm F4.0-5.6 at 9 mm focal length) corresponds to 70 mm on the new rail. So the 12
mm for the fisheye correspond to 75 mm, right?
Wrong! The scale goes in the other direction, so I needed 65 mm. Set that today, and the
results were much better. But there's still purple fringing; can that be fixed?
Ten years ago I bought my first colour laser printer, a
Brother
HL-2700CN. It didn't last very long: on 6 December 2005, while I
was flying back from Europe, a power surge killed it, though I didn't discover that for
nearly a week. I replaced it with the same model because
I had the consumables and also the duplex unit.
That printer is now 8½ years old and showing its age. It prints unevenly and makes
particularly worrying grinding noises while doing so. Time for a replacement.
After some research decided on another Brother, the HL-3170CDW. It has faster print, built-in duplex, and I got it for $249, less than
the price of replacement toner for the old printer. Today Yvonne picked it up.
How have things changed? The installation instructions were excellent, until they petered
out. Install a driver? What's that? The old one didn't need a driver. This is a network
printer (even does 802.11, for what use
that is to me), but they don't tell me how to set up the network. Probably that's on the
real manual on CD-ROM. But you'd expect at least a description of the front panel menus.
After some experimentation found that it had obtained
a DHCP address, and I was able to change it
to a static address. Accessed the web server, and all looked OK. Printed a document. No
duplex!
I've seen this before with the first printer. There the duplex unit was optional, so it
made sense that the default was off. But here it's integral, and it was still off. Finding
the knob and setting it to on was the only thing I needed to do, apart from setting the new
IP address.
And how does it compare? Much quieter, much smaller, much lighter. Maybe a little faster.
But the print quality for photos is no better, potentially even worse. Nobody expects good
photo print quality from a laser printer, but I had expected some improvement in that time.
I've had my Android tablet now for nearly a year. In that time I haven't exactly made friends with it, but there are
some things that aren't too painful.
It has a “16 GB” SD card, but it kept telling me that storage was full. When I look at the
output of df(1), I saw:
Why are there two partitions on the SD card? And how do I fix it? I've been pondering this
for weeks. Somehow Android has taken Linux and castrated it. Where are the normal
utilities? Where are the myriad file systems (12 of them, compared to 4 on eureka)
defined? There's no /etc/fstab. Even /etc is just a symlink
to /system/etc, but I can't find much recognizable there. I've found various
suggestions on the web, all requiring external programs, and I've been meaning to follow up
on them.
The most obvious recognition from these pages was that the SD card is formatted as a FAT 32
file system. So maybe a Microsoft box can shed some light on it. Shut down the tablet, put
the SD card in dischord, my Microsoft box. And there's only one file system on it!
And of course it has about 710 MB of used data.
Looking at the sizes, that's obvious. Adding the two together goes well over 16 GB, and
storage manufacturers have always ignored computer usage and defined their units as powers
of 10, so there are really only 14.9 GB on the card—exactly what /storage/sdcard1
reports. So what's this /storage/sdcard0? It must be somewhere in device memory.
Went searching for sdcard0 sdcard1, and sure enough I came up with a lot of hits. This one started off
with all sorts of ideas about swapped storage areas, but it's not until quite a way down
the thread that somebody mentioned that this is a storage setting. And sure enough,
After taking time to admire the grammar, did that, and now it works! But it took me nearly
a year. How many non-technical users out there have ever worked out what's going on? I
continue to be amazed how bad Android is.
How did that happen? Chris Bahlo has finally had enough living with the Yeardleys, and
though her new house won't be finished for some months yet, she's moving out and will live
in our guest cupboard for the intervening time. And now she and Yvonne can train horses in peace, without worrying whether David will dismantle the
infrastructure overnight.
So far we've paid cash for all expenses on the new house, but it's running out. This was
planned: we'll need a bridging credit from now until we sell our current house. And it
looks as if we may have left it a little late, since they will apparently take up to 3
weeks, and we have some big bills coming in on 8 August. Into town to take in all the
documentation they need; hopefully it will be fast enough.
I'm storing more quick dial numbers in my NetComm
V210P ATA, not something
that is normally of very much interest. I'm putting different categories on different
“pages” (really just web pages), and I've chosen page 3 for numbers related to house
construction. So: entry 20, code 020# was JG King. 21 was Tom Tyler. But I entered the
number as 021 just to see what would happen. Nothing appeared. OK, bug.
Then I went back to page 2. There was an entry for Tom, entry 17.
How did that get there? It took me about 15 seconds to realize how. And as Peter Jeremy
suggested, 0xe also created an entry 14. I couldn't use 2-digit hex numbers,
though, because the input field was restricted to 3 characters.
The big issue with the cooking is to keep the meat rare while cooking (melting) most of the
fat. Is this a case for sous-vide cooking? Clearly that doesn't address the fat, but there are other ways to deal with that.
Found a recipe in the Codlo cookbook and
bent it a little. After cooking the breast in the cooker, removed the entire fat layer,
which was surprisingly easy:
Yvonne came to me this morning to tell me that the photo
processing software had hung itself up again. That's normal enough for the Microsoft-based
products, but this was FreeBSD, and she was
just running make. ps(1) showed that the X server was no longer running. And /var/log/messages?
...
Jul 22 11:29:03 lagoon kernel: swap_pager_getswapspace(8): failed
Jul 22 11:29:03 lagoon kernel: swap_pager_getswapspace(9): failed
Jul 22 11:29:03 lagoon kernel: swap_pager_getswapspace(16): failed
Jul 22 11:29:04 lagoon kernel: pid 67064 (Xorg), uid 0, was killed: out of swap space
Decades ago I ran computers with 8 and 16 MB of memory. lagoon has 2 GB. In
addition it has 1 GB of swap space. That should be enough for anybody. But not for
firefox, which leaks memory like
a sieve. It looks like she's going to have to shoot down her firefox process at
regular intervals.
Discussing the matter with David Newall on IRC, he opined that it's inappropriate to shoot
down arbitrary processes. Yes, of course. But I recall some explanation why it's done this
way; time to UTSL, I suppose.
For some reason my MPEG recodings
on cvr2, my Ubuntu Linux computer video
recorder, ground to a halt today. No obvious reason, but mythtranscode ran at a
snail's pace. No error messages, just very slow. Anything in /var/log/messages?
No, there never is, just:
Jul 22 13:16:10 cvr2 -- MARK --
Jul 22 13:36:10 cvr2 -- MARK --
Jul 22 13:56:10 cvr2 -- MARK --
I don't know why it bothers.
cvr2 is Linux, not Microsoft, but since there was nothing obvious, I tried rebooting.
And mythbackend didn't come back for a long time. When it finally did, I got an
error message telling me that MySQL wasn't
running. How do you start MySQL? Let me count the ways... No, just ask on IRC. It seems
that for Ubuntu, or at least my version, the invocation is
service mysql start
Tried that and got an error message: can't start, disk full.
Why didn't anything else complain? In FreeBSD you'd have error messages all over the place. Does Ubuntu not want to scare
the user, maybe?
Stewart Summersby along this afternoon at 13:45 with a bill for $1,800. Where did that come
from? It all seems to have been for last week. We had agreed $600 (somewhat against my
better judgement) for removing the trees, and that was part of the bill. But he's not done
yet, and for some obscure reason he has thrown all the small branches into the pond:
And the other $1,200? It seems to be for the bobcat work, for which we had agreed $600
(clearly a favourite value of his) per day. But as near as I can see, he has only worked 4
hours. Yes, they were spread over 2 days, but he can't use a calendar day as a
measurement. When I asked him about it, he had a temper overload event, grabbed the slips
of paper that he uses instead of invoices, and left.
Later Craig Mayor came by and we discussed the matter. In principle it seems that my
estimate of the time was right. Stewart seems to think that he won't get paid at all, but
clearly that's not the way to do things. About the only thing is that we now need somebody
to complete what little remaining work there is. Finding a new electrician will be easier.
The connection hung. Why? I still don't know. What I've established so far is that it
only happens from lagoon, her machine, and only when accessing Coursera. It all goes
via an instance of squid running
on eureka, so you'd expect there to be no difference between individual machines.
It's also not the browser: I tried both firefox and Chrome, and the results were the same. The squid logs
aren't very helpful:
It's interesting that the second column (presumably transfer size) is different
between lagoon and teevee for the initial connection. But that doesn't help
me much. I should put a wireshark on it, but I don't have the time.
We're supposed to be moving out of here soon, when the Stones Road house is finished. But
first we're doing the opposite: Chris Bahlo is gradually moving in, and this evening she got
a gift horse, “Ruffian”, whom she had trained and sold some time ago. The new owners
weren't able to keep her under control (maybe that's how she got the name), and when they
moved house, she was in no condition to sell. So now Chris will have to train her again.
Then down to the police station to have my voluminous application for owner-builder status
for the shed to be witnessed by
a Justice of the peace—and I
discovered I had left my driver license behind in the scanner. I had the copy, but that
didn't help much. Damn!
Well, almost. We do move around a little, and it's good to have a phone with you. But now
it has to be a smart phone. Yvonne has one from Chris Bahlo,
and last week I got a call on it. Buzz, green and red phone symbols light up on the screen.
And it also volunteered the information that the call was from Tom Tyler, with whom I wanted
to speak. Intuitively I guessed that green meant “answer” and red meant “reject”. So I
pressed on “green”. It wouldn't be fair to say that nothing happened: the image lit up a
little, so clearly the touch screen was working. But it didn't answer. I had to give up
and call him back on a VoIP phone.
Today while—coincidentally—I was with Tom, he went out to make some copies, so I tried
calling the Android tablet I had with me from his phone. Sure enough, same thing, no response.
Then the penny dropped: I shouldn't just press on the phone symbol, I had to slide it
(hold my finger on the display and move it) to the right. Now isn't that obvious? Isn't it
intuitive? Of course not. On
old-fashioned GSM phones it was a physical
button that you had to press. Why do I now have to slide? Where's that documented?
Nowhere, of course: it's intuitive.
In February I drove
round Great Ocean Road with Jörg
Micheel, and on the way I received a phone call from Yvonne.
I wasn't able to answer it, and Jörg (mobile phone expert, but with an Apple phone) didn't know how to do it either. Was this the
problem?
I cooked them at 47° for about 45 minutes, and they came out pretty much the same as the
salmon we made two weeks ago, very, very
tender and not completely opaque:
On the whole I'm wondering whether it was worth it. One distinct issue is that 47° is tepid
at best, and the consistency takes a little getting used to. Yvonne thinks it might be good for fish in aspic, though. But then there are issues
of health: at that temperature many bacteria are not killed.
Another short power failure this morning at 1:02. It was long overdue; normally we have about one power failure a week,
but the previous was nearly 7 weeks ago.
Into Ballarat again this morning to have
the statutory
declaration for the Owner Builder application witnessed, and also a certificate for the German Pension Insurance to prove that I exist. Things went relatively smoothly, but
somehow I ended up frustrated beyond belief, and decided not to do anything else for the
rest of the day.
Later Yvonne came in and said “we have a problem”. It seems
that things went fine until he tried to back out of the driveway, and then he suddenly
subsided by about 25 cm, and might have gone deeper if the fuel tank hadn't hit the ground:
CJ tried to tow him out, but his tractor wasn't strong enough, and just made a mess of the
driveway, as the second photo above shows. Sean called up his people, and after a bit of
phoning around they sent up a prime mover
from Geelong. But as I discovered in the
afternoon, he hadn't been able to drag him out either. He had got him out of the original
rut only to have him sink in elsewhere, this time on the driveway proper:
But the real issue is that the driveway was completely incorrectly constructed. As several
of the people observed, notably Mike, the excavator driver, a driveway needs 30 cm of
aggregate under the surface. Stewart had put so little of anything that in places the
ground showed through. So it looks like we can start all over again. I hope Stewart has
good insurance; it looks like the matter will cost us at least the $1,500 we paid him and
the quarries for work and materials. Hopefully Metroll won't ask for compensation.
Still, every cloud has a silver lining. This was just a light truck. In a few weeks' time
we'll be having concrete trucks coming in to lay the slab. By then we need a real
driveway. In addition, it's good that it happened after Stewart gave me good reason not to
employ him any more.
It's been a month since I applied for an amendment to the planning permit, and Peter had said it would only
take a couple of weeks. But nothing has arrived. Called him up and discovered that the
delay was due to his attempts to have the biodiversity requirement removed altogether. He
was unsuccessful, but it's nice of him to try. He basically reflected my own opinion, that
it's ridiculous to require an offset for such a small amount of vegetation. But rules are
rules, and we have far too many of them, many badly thought out.
It looks like I'm going to have to work out a way to correct things manually.
The other thing I have to do is to start a new page for the new house. In the process I
should learn from my experience of the last 7 years, in particular regarding image names.
Steve and Zali O'Dea along for dinner this evening. Zali asked
for kosher food, which matches
the Star of David necklace that she
wears. OK, kosher isn't that difficult. The main issue is the question of milk products
and meat, so we cooked the scaloppine al
limone with olive oil instead of the traditional mixture of butter and oil. But
what to do with the entrée? Smoked salmon with bread—and no butter. We didn't have
margarine, so Yvonne suggested some strange “olive spread”
that she had. Finally found the stuff and discovered that it, too, had milk products in it.
And then Steve said “No problem, we're not Jewish. We just adhere to the code in the
Bible”. That just says (three times) “Thou shalt not cook a kid in its mother's milk”, but
Steve agrees with me that it's difficult to construe that this means “don't eat meat and
milk together in one meal”. Later I found this explanation, but I'm still not convinced. And the explanations in Wikipedia are different again.
Still, pleasant meal and pleasant company. The O'Deas have their
own Borzoi, Bindy, Zhivago's daughter, but they didn't bring her with
them. They did bring their 4-month-old daughter Roni, who elicited much interest from
Nikolai in particular:
We've been planning for some time that Chris Bahlo move in until her own house is
completed—potentially later than our new house, but at least 6 months. Today was the day:
she arrived with 3 stallions, 3 cats and 2 dogs (third dog to come tomorrow):
Last
month I tried my first sous-vide
recipe, chicken tanduri. It wasn't an
unqualified success, and I guessed—correctly—that the temperatures were too high. So today
we tried again with suggestions from the Codlo
cookbook, which suggested 62° for an hour or so.
Tried that, and it was better, but not much. The issue is maybe that the meat needs to be
grilled after cooking, and this could heat it up to pretty much the 75° we had last time.
As I see it, we have three alternatives next time:
Cook at 62° and allow to cool completely before grilling.
Don't grill at all.
Use chicken thighs, like everybody else I know.
Certainly the promise of sous-vide can only be attained with some careful experimentation.
I'm with MyNetFone, and they have other
tariffs: in fact, the tariff I'm on (no rental, pay only per call) is no longer being
offered. But the calls are more expensive, and they have a tariff for $9.95 a month (or $99
a year) that includes 200 landline calls, more than I'd ever use. And calls to mobile
phones are down to $0.20 per minute. In addition, for that I get two lines. And the price
corresponds to break-even.
How does the second line work? Called them up and asked if I could just add a
second ATA. I knew I
was going to have a hard time when Gavin, the first consultant, asked me what ATA meant.
Finally I got connected to technical support, where Marsha told me yes, I could use a dual
port ATA. I repeated my question and she sounded particularly unsure of herself, but in the
end said yes, I could use two ATAs.
Can I believe her? Hard to say. But as luck would have it, the cheapest new ATA I could
find on eBay was a Linksys PAP2T-NA, a
now-obsolete model for $24.38. And it has 2 ports, so there's no issue. Signed up for a
new connection and got two separate sets of authentication parameters. So Marsha was right:
I could use two separate ATAs.
So I now have three separate VoIP accounts. I had intended to use the old one for my
mobile phonetablet, where it could masquerade as a landline phone. It took a
while longer for me to realize that that doesn't really make much sense: not only are the
calls more expensive, I have to pay for all of them. So is there any use for the old
account? It's a separate connection, so I suppose it could be useful at times, but not very
often.
One of the advantages of my new MyNetFoneVoIP account is that I can call in on the
numbers as well. I can either transfer an existing number or be allocated a suitable one.
Just go to the web site and start searching...
And searching. And searching. All I found was my existing account. Finally called up
MyNetFone technical support and spoke to somebody who mumbled her name as “Nutella”. She
didn't seem to understand the issue, and I had to explain things to her several times. Then
she decided to reset my login password, occasioning further delay, and in the end decided
that she couldn't fix it, so she opened a ticket and told me I'd hear back from them.
The call came at 22:10, when I'm normally in bed. And they wanted me to go to the computer
and check! No way.
But that was premature. It failed again after 8 minutes, and stayed down. Called Aussie
again and got confirmation that others had had the same problem. But they still needed to
treat it as an isolated fault, because the National
Broadband Network hadn't reported any failure. Why not?
The NTD was showing
normal status, so it makes sense to guess that the problem was in the link
between Dereel
and Ballarat.
Called up the NBN on 1800 687 626 and found a voice menu option “Press 1 if your phone
service was disconnected in error”. Well, I suppose this is an error, so pressed that and
got connected to Sean after 2:15 minutes. He listened carefully and took down a complaint,
less carefully: I had to correct him several times. The way I see it, there are multiple
issues:
The NBN should perform enough network monitoring to notice when problems of this
magnitude occur. As I see it, they only become aware of faults when a customer reports
them.
Customer reporting is further complicated by the rigmarole that Every Single Customer
must complete, including crawling on the floor with a mirror to look at the LEDs next to
the RJ45 jacks, which point down at the floor. It takes at least 15 minutes to complete
it, including a reboot of the NTD. And if only one customer reports—perhaps because of
the time of day—it might still not get the attention it deserves.
Customers should get feedback on why the outage occurred.
I asked Sean what the availability goals for the NBN were. He didn't understand the
question, and I had to explain it to him. Finally he went searching and came up with a
figure: 99.9% availability averaged over a year.
That's terrible! It's nearly 9 hours a year. It means that what I've experienced so far (1
hour in June, 5 hours
in July, and at least an
hour today), is roughly all I can expect. And that's for a new, modern replacement for
Telstra's aging copper infrastructure, which here
in Dereel hasn't gone down at all in the 7
years I have been here. How can they claim to offer improved infrastructure with that level
of service?
In any case, it looks as if they'll look into it. They'll assign a “case manager” and a
“specialist” to it, and they gave me the reference number 511066-91183999 (hopefully the
size of the number isn't an indication of the number of complaints they're receiving).
And at the end: “Please hold on and answer a short three question survey”. Did that. Beep,
beep, beep. It's not the first time I've had that, either.
And how long was the outage? No idea. After 2 hours, 20 minutes I still had no connection,
and then I discovered that after shooting down
the DHCP client, I had forgotten to restart
it Started it again and got an immediate connection. Not all problems are NBN's fault.
For about a month now Nikolai has been
scratching and biting himself more than should be normal for a dog. We've checked for
insects and found nothing, we've washed him, all to no avail. It doesn't look like insects
anyway, because the other animals aren't scratching, though just recently Leonid started. So off
to Bannockburn today to
have the two of them looked at.
Fleas! How did we miss that? I can see two possibilities: either the fleas are new and
there's some other issue as well, or he had been so good at killing them that they've only
just become visible and transmittable. In any case, we'll treat them for fleas and see if
that makes all the problems go away. $179 for that!
The drive home was horrible. The road points roughly west-north-west, and today, like so
often, I was driving directly into the setting sun. To add to the pain, it had rained
heavily, and the trucks in front of me threw up a spray that glistened in the sunlight and
limited visibility to nothing from that point on. I was exhausted by the time I got home.
Signing the contracts was a surprise: there must have been 200 pages of mainly boilerplate,
and we had to initial Every Single Page, even some that we had already signed. But there
was little in the way of surprises, just the usual sloppy details—for example, I've told
them numerous times that the power supply to the house is only 50 A, but the contract states
that the main switch must be 80 A. It's silly to have that in the first place, but you'd
think that they'd get things right. Still, the fine tooth-comb treatment seems to have paid
off.
As a result of all the red tape, I was about half an hour late for Brendan. But that, too,
seems to be going well. Hopefully we'll have our money by next Friday, when the credit card
bill is due.
Over to Stones Road in the evening to meet Warrick Pitcher to discuss the works that Stewart
Summersby left unfinished. He has much larger equipment than Will Tattnell, and in
particular he thinks he can get the tree stumps out in less than an hour—Will was talking
about some time to remove them. Things look good, and he can do the work next week. Now we
just need to hear what it will cost.
Once again there's a discussion on the German Olympus Forum about image stabilization. Specifically: why is it recommended not to use image
stabilization when the camera is mounted on a tripod? I tried this four years ago and came to the conclusion
that yes, indeed, it doesn't work well.
But on that occasion I was using a 50 mm lens on my Olympus E-30, and the
stabilizer was set to a focal length of 800 mm. Clearly that shouldn't work. And as Markus
Burak noted, there have been significant improvements in image stabilization since the E-30,
notably with the OM-D
E-M1. So today I repeated the experiment. Here a detail comparison with the E-M1
(left) and the E-30 (right):
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.
