|
| Greg's diary |
| December 2024 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Sunday, 1 December 2024 | Dereel | Images for 1 December 2024 |
| Top of page | ||
| next day | ||
| last day |
|
Pork and dofu revisited
|
Topic: food and drink, opinion | Link here |
It's been over a month since I tried my new pork and dofu recipe. Time for another attempt.
Once again a question of quantities. More hoisin and bean sauce. Less dofu. And somehow there are too many steps. But in principle the same thing. I thought that it had looked considerably darker this time, but in fact I could hardly see any difference. Here last time (first image) and today:
|
|||||||||||||||||||||||
|
Raclette quantities
|
Topic: food and drink, opinion | Link here |
Raclette for dinner today. And the eternal question: how much of each ingredient should we use? Started writing down things, but so far it seems, for us 2½ non-representative guests:
| Ingredient | Weight | |
| Cheese | 350 g | |
| Potatoes | 600 g | |
| Ham | 150 g |
I'm not sure that we got the ham right. It seems far too little. But I do recall having plenty left over in previous years.
|
First Sunday in Advent
|
Topic: general | Link here |
Today was the beginning of summer and coincidentally the first Sunday in Advent. Jane had brought us an advent calendar:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's not a custom known in Australia, and the origin was clear in the name of the supplier: Gewürzhaus, a company I had never heard of, though it seems that they have shop as close as Daylesford. This is a calendar with a difference: instead of sweets and things, each door contains a packet of some kind of spice:
|
|||||||||||||||||||||||
Not to be outdone, we dragged out our old Christmas pyramid:
|
|||||||||||||||||||||||
And of course Yvonne had to have her fun:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
We had thought of lighting up the Hanukkiah as well, but although Hanukkah almost always overlaps with Advent, this year it will be different, not starting until Christmastide.
| Monday, 2 December 2024 | Dereel | Images for 2 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Tidying the freezers
|
Topic: general, food and drink, technology | Link here |
It's been at least 14 years since I started keeping computer records of what's in our deep freezes. It's not easy to keep things in sync. I can't go to a computer every time I take something out of the freezer, and for some baskets with frequent change, such as the breakfast ingredients, I no longer even keep track: it's easier just to remember, including where the things are in the basket.
But the thing's full, and we have this cheese that I bought over a month ago. It keeps reasonably in the fridge, but it's time to freeze it.
But where? I've started tidying out the thing, basket by basket. Yesterday it was basket 2, today basket 1, where the cheese should go. It's hard work correlating the lists with the contents, and at the end I didn't have time to cut up the cheese. But I did end up with basket 1 only half full.
|
Mail bloat
|
Topic: technology, history, opinion | Link here |
A couple of days ago I received:
Date: Sat, 30 Nov 2024 03:14:15 +1100 (AEDT)
From: Mail Delivery Subsystem <MAILER-DAEMON@hydra.lemis.com>
>>> MAIL From:<root@hydra.lemis.com> SIZE=755923075
<<< 552 5.3.4 Message size exceeds fixed limit
554 5.0.0 Service unavailable
755 MB in a single mail message! It would have been some nightly maintenance message, but why so big? I'll never know: it's gone to the bit bucket in the sky. Yes, of course I was able to increase the limit (to 1 GB, both for message and mailbox size), but how times have changed! When I first set up my mail server, a message of this size would have tied up the phone line for a little over 9 days and cost me 340,165 DM in data charges (about US $464,000 today).
| Tuesday, 3 December 2024 | Dereel | Images for 3 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Cheese: done!
|
Topic: food and drink | Link here |
Still more work in the kitchen today, cutting up and freezing cheese. Why does it take such a long time? But at least I still have space in the freezer.
|
Mona and Jane
|
Topic: photography, animals, opinion | Link here |
Seen this evening:
|
|||||||||||||||||||||||
That was on the spur of the moment: minimal lighting, f/4 lens, so the camera chose 6400/39° ISO, not the best image quality. But what I didn't expect was this crop from another image:
|
|||||||||||||||||||||||
The camera (OM System OM-1 Mark II) told me that it was in focus. What went wrong there? It wasn't the only out-of-focus image. And to add insult to injury, exiftool didn't want to copy the Exif data.
| Wednesday, 4 December 2024 | Dereel → Ballarat → Dereel | Images for 4 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More atrial fibrillation
|
Topic: health, opinion | Link here |
Today was the end of Jane's stay with us, and Yvonne was to take her to the railway station.
But it didn't work out like that. She woke up with a bad case of atrial fibrillation, the complaint of which she was ”cured“ only a month ago. Is that due to the stress or activity of the past week? And what should we do? More Flecainide? Tried to contact both Rodney Reddy and Professor Peter Kistler, but only spoke to the receptionists. Reddy: take half a flecainide now and one in the evening, go to hospital if you feel bad. Kistler no answer, but Reddy's receptionist told us that he had sent a letter asking for continuation of the medication unchanged. More emails, no response, pulse rate at a constant 135 or so all day long. How serious is that? What worries me is that the flecainide doesn't seem to have helped. This is really worse than it has ever been.
|
More GPS strangenesses
|
Topic: technology, general, opinion | Link here |
Heading off to Ballarat, my phone claimed “GPS signal lost”, and with a brief interruption stayed like that until it was restarted. Is that an issue with Google Maps or with the phone? I have a log from the Mendhak GPS logger which should tell me something—once I do a lot of mouse and keyboard manipulation.
|
Shopping in the times of strike
|
Topic: general, food and drink, opinion | Link here |
Another consequence of Yvonne's atrial fibrillation was that I had to do the shopping. That wasn't as painful as I had feared, but there was another unexpected factor:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
That was at Woolworths, who are suffering from an extended strike action. About the only bottled water I found had a sting in the tail:
|
||||||||||||||||||||||||
All with barely stated added flavouring. That should be in big letters! Finally I found some San Pellegrino at a significantly higher price. Thanks, Woolworths.
They have also changed their do-it-yourself checkout. No longer this stupid bagging stuff, though remnants of it exist in the messages displayed on the registers. But I didn't get a receipt! Pressed the “call assistant” button, then turned to the assistant standing right next to me, who hadn't heard the call. She was able to print out a receipt for me after dismantling the printer, though it was a software issue, not a hardware one. She agreed to me that the system was stupid, though I suspect that the register would have given me the opportunity if I had asked for one during its limited attention span. Instead I committed the sin of bagging my purchases first, something that they no longer believe in, so the offer disappeared from the screen about half a second before I wanted to select it.
|
Who wants a greyhound?
|
Topic: animals, general | Link here |
Yesterday Yvonne and Jane went to Petstock in La Trobe Street to wash the dogs, and also to weigh them: Larissa 38.3 kg, Elena 40.1 kg, both at the top of the scale for females. While they were there, they saw a rescue cat who looked “just like” Mona, so while I was in town I went to take a look.
And there I bumped into Peggy Naumov, who has just got a new black greyhound, Cody, and was having him washed. Oh. Yvonne had asked me just yesterday if I could consider adopting a rescue greyhound, one whom she knew (answer: no). Peggy would have been just the right person to take her, but it's too late now.
And the cat? Yes, she looks quite like Mona. But we can't take her. The same logic as with the greyhound applies, but more to the point, her name is Luna. How could you distinguish between Luna, Mona, Lara, Lena and Bruno? More seriously, apart from the fact that we don't need another cat, she doesn't get on with other animals.
|
Now that's a bone!
|
Topic: animals, food and drink, general | Link here |
Larissa and Elena have had a rough time of the last week. Yvonne has been busy with horses, and we only took them for a walk once or twice. Today I went off with them to the „Große Linde“, where Lara found the biggest bone (kangaroo?) that I have seen:
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||
It weighed 363 g, while I suspect that they normally weigh under 100 g.
|
Camouflaging cats
|
Topic: animals, general | Link here |
Lately Bruno has been sitting on the chair by the door of Yvonne's bedroom.
|
|||||||||||||||||||||||
In principle that should make him easy to see from the hallway, but in practice the backlight makes him blend into the cushion. With Mona it's easier:
|
|||||||||||||||||||||||
For once the photo makes it easier to see than in reality.
| Thursday, 5 December 2024 | Dereel → Ballarat → Dereel | Images for 5 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More atrial fibrillation
|
Topic: health, opinion | Link here |
Yvonne was feeling no better this morning, and her ECGs confirmed it. Consistent tachycardia with a pulse round 133. No response from Professor Peter Kistler. Called up Rodney Reddy, who recommended going to the emergency department; under the circumstances that sounded like the only option, so packed her up and carted her off to the Ballarat Base Hospital.
She had the usual interminable wait there. She didn't get seen to until 4 hours after arrival, but then they decided that the (lack of) medication that Kistler prescribed was the problem, and that maybe a cardioversion would solve the issue. They did that round 19:45, and I was told that I could pick her up at midnight.
Cardiversion went as planned, and was successful—something that, I suspect, was not a foregone conclusion. And by 22:00 she was cleared to go home! In the end she caught a taxi. What a day!
|
The Botanical Gardens again
|
Topic: gardening, opinion | Link here |
After dropping Yvonne, on to the Ballarat Botanical Gardens for the first time in four years and walked around a bit. Things don't seem to have changed much, though the decrepit fern house has been partially replaced. Here 14 years ago (first two photos) and today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Walking round the west, found this plant:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||
I know that plant! Just by chance, I took a photo of another one this morning, in our garden:
|
|||||||||||||||||||||||
So it's an Agapanthus inapertus! Strange. It doesn't look much like the Agapanthus that I know.
Oh. I check my links, and it seems that Agapanthus inapertus looks like this:
That looks more like what I would have expected. So the sign is wrong, probably left over from a previous planting. And I still don't know what my grass is called.
|
Collecting the washing machine
|
Topic: general | Link here |
Then along to Cambridge Street to pick up the old washing machine from Glen Castleman. As he had suspected, there was little damage, just a water sensor.
|
Animals come closer?
|
Topic: animals, general | Link here |
Yvonne wasn't at home, but the animals congregated in her bedroom:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
That's normal enough for Bruno and the dogs, but it's the first time I've seen Mona join in.
They were clearly bored. Earlier I had found this in the hallway:
|
||||||||||||||||||||||||
|
||||||||||||||||||||||||
| Friday, 6 December 2024 | Dereel | Images for 6 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Paying bills: unWise
|
Topic: technology, general, opinion | Link here |
My domain registration is coming up, and it seems to be getting more expensive all the time.
Today it would have been over $59. But Gandi is in France. What
is it in Euros? 31,98 €, which
corresponds to about $52.80. OK, that's what I have TransferWise for. Signed
in, which for once worked without problems, and discovered that I should top up my credit.
And somehow that, too, went without too many problems.
Then pay. “Your card is frozen. Unfreeze?”. Now why did I do that? Still, unfreeze. “Your card is now unfrozen”. Gandi says “Our online banking partner refused your payment”. And sure enough, “Wise” has changed its mind. Card still frozen.
After several tries I gave up and tried another number, which worked. But why is “Wise” always such a pain?
|
More OM-1 Mark II fun
|
Topic: photography, general, opinion | Link here |
Spent some time updating camera firmware today. Not as successful as I might have thought. I was sure that my OM System OM-1 Mark II was down-rev by one update, but OM Workspace, the only program allowed to update my cameras, told me that it was up to date. And the firmware update page told me that a number of my lenses have updated firmware, some of which it refused to supply, along with Yvonne's Olympus OM-D E-M5 Mark III. Much messing around, during which I was reminded that Workspace doesn't save the configuration for the E-M5 Mark III. All the firmware updates relate to lenses that we don't have, so I left it at that.
Then there are a few Panasonic/Leica lenses. The Vario-Elmar 100-400 mm updated nicely, but Workspace claimed that I was up to date with the 15 mm f/1.7 Summilux and the 25 mm f/1.4 Summilux. It wasn't until afterwards that it occurred to me: it didn't give me the name of the lens, and the claimed revision was 1.0. By contrast, with Olympus lenses it gave the name of the lens:
|
|
|||||||||
It should look something like this:
|
|
|||||||||
Something to do with Leica? It's not that simple, since the update for the 100-400 mm worked. As it is, I don't even know if the lenses are up to date or not.
In passing, came across an alternative: update without “Workspace”. It looks interesting, but it has two disadvantages. Firstly, from my point of view, it only relates to camera firmware, and secondly it's a delicate procedure. If it goes wrong, it could potentially brick the camera.
| Saturday, 7 December 2024 | Dereel | Images for 7 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More KardiaMobile pain
|
Topic: technology, general, health, opinion | Link here |
Last month we had untold pain with Yvonne's KardiaMobile 6L, which repeatedly refused to work. Without the aid of any diagnostics, I guessed that it was a pairing issue. But the indicator said “battery discharged”, and so I ordered some replacement batteries (CR2016). But then it ran without that kind of problem for over a month.
Until today. OK, change the battery. Yes, the old battery had a voltage of 3.00 V, while the new one had 3.20 V. But I still couldn't pair!
Much investigation. The problem occurred on fossil, Yvonne's Xiaomi Redmi Note 9T. What about a different phone? Tried it on albo, my Redmi Note 13, and it worked. Well, for a while. In the middle of a reading I got:
|
|||||||||||||||||||||||
That makes sense at some point. But here it suddenly occurred after 22 seconds of a 30 s recording, displaying the “battery empty” symbol:
|
|||||||||||||||||||||||
My best bet is that the battery symbol is very misleading. It displays it when it isn't paired, for example, where it can't know anything about the state of the battery. It took me three attempts to get a good reading.
Back to fossil. Still no success. Dammit, I just want to record an ECG, not debug broken firmware. Dragged out hirse, my old Redmi 9T and installed the app there. Works. But that includes my fingerprint authentication, and I don't want to reset that. So dragged out enzian, the glacially slow old Redmi 9A and installed there.
|
|||||||||||||||||||||||
What's that? I forgot to record, but I got it going in the end. But what's wrong with the device? Is it maybe really defective?
|
Anti-semitism at work
|
Topic: politics, opinion | Link here |
Yesterday unknown attackers set fire to the Adass Israel Synagogue in Melbourne. An abhorrent act. Everybody has condemned it.
But Bibi got up on his hind legs, and, after condemning the action (of course), tweeted (or is that now “Xed”?):
Unfortunately, it is impossible to separate this reprehensible act from the extreme anti-Israeli position of the Labor government in Australia, including the scandalous decision to support the UN resolution calling on Israel "to bring an end to its unlawful presence in the Occupied Palestinian Territory, as rapidly as possible", and preventing a former Israeli minister from entering the country.Anti-Israel sentiment is antisemitism.
What's that nonsense? Bibi, you are the problem. Yes, you're Jewish, a thing many Jews are ashamed of. But criticizing people who object to genocide is not an honourable thing to do. Have you considered that your actions are contributing to anti-semitism amongst people who don't know any better? Go away!
|
Crash!
|
Topic: technology, opinion | Link here |
All day long the /Photos disk on eureka was making a rhythmic churning sound. SMR catching up with a backup? Firstly it took far too long, but much more importantly, it's not an SMR disk.
In the evening I went feeling through the disks on the machine. Yes, definitely /Photos. Left it at that and went back to the lounge room and did some web surfing. And then I couldn't connect to wwww.lemis.com, my local web server. I couldn't start an ssh on eureka either.
Damn. How do I even access the console on eureka any more? It was the right-hand monitor, and it was spewing swap pager errors. I couldn't start any programs locally either. Nothing for it: Big Red Button.
It came up relatively easily, modulo a syslog.conf misconfiguration that ended up in a loop reminiscent of what I had seen in September. Oh:
*.* @eureka
I'm logging to myself! Is that the problem I had in September? Why did I do that? Remove the line and all was well.
But why did I run into swap problems? Fought my way through 4 GB of /var/log/messages and found:
Dec 7 21:50:13 eureka kernel: ada2 at ahcich4 bus 0 scbus2 target 0 lun 0
Dec 7 21:50:13 eureka kernel: ada2: <PNY CS900 250GB SSD CS900613> s/n PNY21152104160104730 detached
Dec 7 21:50:13 eureka kernel: (ada2:ahcich4:0:0:0): Periph destroyed
Dec 7 21:50:14 eureka kernel: swap_pager: I/O error - pagein failed; blkno 172127,size 4096, error 6
Dec 7 21:50:14 eureka kernel: vm_fault: pager read error, pid 2694 (Xorg)
The time matched when I was checking for the thumping. Did I accidentally disconnect a cable? That seems more likely than an error on the disk itself.
| Sunday, 8 December 2024 | Dereel | Images for 8 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Still more KardiaMobile pain
|
Topic: health, technology, general, opinion | Link here |
Why did fossil, Yvonne's phone, suddenly stop communicating with her KardiaMobile 6L? I suspected firmware issues. Bluetooth is flaky at the best of times, but it seems that KardiaMobile has made things worse. It doesn't adhere to normal pairing procedures: it requires fingers on the contacts to even pair. But the status this morning was that enzian, hirse and albo all worked, and fossil didn't.
OK. First, disable Bluetooth on fossil, then reenable it. Also restart the app. No improvement. Microsoft solution? Uninstall and reinstall the app (why is it 89 MB in size? Only web browsers are that size). Restart. No go.
Back to phone settings. Remove the saved device, pair—not from KardiaMobile, but from the phone. That still requires fingers on the device, but it connected. So there's no basic communication problem.
Back to yet another instance of KardiaMobile. It works! So clearly there's something very wrong in their implementation of the communications protocols. The only question is: how long will it work for? In any case, we have three other phones with which it is currently prepared to communicate.
I've already noted that the device documentation is terrible. How much help was the web? This page tells me
Do not pair the KardiaMobile 6L outside of the Kardia app in your phone's bluetooth settings
If your 6L was paired outside of the app, go to your phone's bluetooth settings, select forget device, then open the Kardia app and take steps to re-pair the KardiaMobile 6L within the app
“Take steps”. Thanks for the clear explanation. Apart from the fact that that was the only way I got it to work at all, why not? Another indirect admission of serious bugs. People, this device is for people with cardiac problems. They should avoid stress, and this adds to it.
It also contained the statement
Make sure you are not near electronics like computers, large TV's, and that your phone isn't plugged into a charger
I'm not sure what to make of that. I'd classify it in the same category as warnings not to use mobile phones at petrol pumps or near large Lithium batteries. More a documentation superstition than anything else.
I also found a compatibility list of phones that work with the device, dated “11/07/2024”. What's that? I read July, but it could equally well be November. Fortunately the otherwise superfluous statement “Updated · 1 month ago” clarified their ambiguity: it was indeed 7 November, one of the days where we had these communication problems. But why should there be a compatibility list? This is just a question of Bluetooth communication, so any well-behaved Bluetooth implementation should work. Given that there was only one Xiaomi phone there, it's clear that this is another admission of their failure.
|
What do I want from new photo software?
|
Topic: photography, technology, opinion | Link here |
I currently have test installations and offers for a number of photographic software packets: ACDSee, for which I already have a license that I don't use, Affinity, for which I have had a 6 month trial since August, and which I also found too hard, Radiant, and DxO PhotoLab 8. I've spent time reading about them, even tried a couple. And there are online videos that I've tried watching and found completely irrelevant to my requirements. Somehow they don't do it for me. So what do I really want?
Moiré reduction. I can't get better than this in yesterday's photo of my phone:
|
|||||||||||||||||||||||
Easy, accurate background removal. Luminar didn't quite make the grade. I bought it mainly to manipulate this image:
|
||||||||||||||||||||||||
It did greatly improve things, until I looked at the details:
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
Good, easy to use noise reduction. The latest version of DxO PhotoLab claims to do this better, but I have more things I want of it:
DxO: I've mentioned noise reduction, which may work well. But there are others. In particular, I can't even properly test the current version because it doesn't find my “palettes” and “presets”. Well, it doesn't display the “presets”, though it finds them for existing photos (via the “sidecar” file, presumably). But just testing it requires significant pain, and there's no evidence that it has fixed any of my complaints, like inability to set default crop behaviour or reinstating the rectangle recognition code.
Recovering old photos. I have a whole lot of old photos that I would like to make less “old”, like this one (Big Jim McGibbon, April 1970):
|
|||||||||||||||||||||||
A related issue: correct second order colour casts. A surprising number of photos have blacks that go into deep blue or purple. I can't just change the colour, since otherwise the highlights would go into yellow or similar. The Big Jim photo is one example, but it happens with modern photos too.
No subscription. By coincidence, all the packages I have been offered come without a subscription. Can it be that this is generally unpopular?
There's also an ongoing issue with the Exif data from the OM System OM-1 Mark II, but I think there's little overlap there.
|
Flowers in garden
|
Topic: gardening | Link here |
Summer is on its way, and for once we have some flowers in the garden:
|
|||||||||||||||||||||||
Two stand out, though. First, the Hibiscus rosa-sinensis “Uncle Max”, which has suffered considerably over the last few years. But now it's flowering more than I have seen at this time of year:
|
|||||||||||||||||||||||
And the Clematis “Edo Murasaki” that was mutilated by Nick Macdonald and mate in October is now coming back:
|
|||||||||||||||||||||||
There's even a barely visible bud.
|
Why did my swap fail?
|
Topic: technology | Link here |
Yesterday's eureka crash was clearly related to the disappearance of the swap device. There's also a smoking gun pointing to me messing around in the cabinet round that time. But is that the only reason? Can I read the entire device?
=== root@eureka (/dev/pts/2) /usr/local/etc/postfix 40 -> time dd if=/dev/ada2 of=/dev/null bs=1m
^C^C43313+0 records in
43313+0 records out
45416972288 bytes transferred in 212.532209 secs (213694538 bytes/sec)
No! The ^Cs came after a lot of these messages:
Dec 8 12:54:03 eureka kernel: ahcich4: Timeout on slot 19 port 0
Dec 8 12:54:03 eureka kernel: ahcich4: is 00000000 cs 00000000 ss 00080000 rs 00080000 tfd 40 serr 00000000 cmd 0000d317
Dec 8 12:54:03 eureka kernel: (ada2:ahcich4:0:0:0): READ_FPDMA_QUEUED. ACB: 60 00 00 81 49 40 05 00 00 01 00 00
Dec 8 12:54:03 eureka kernel: (ada2:ahcich4:0:0:0): CAM status: Command timeout
Dec 8 12:54:34 eureka kernel: ahcich4: Timeout on slot 30 port 0
Dec 8 12:54:34 eureka kernel: ahcich4: is 00000000 cs 00000000 ss 40000000 rs 40000000 tfd 40 serr 00000000 cmd 0000de17
Dec 8 12:54:34 eureka kernel: (ada2:ahcich4:0:0:0): READ_FPDMA_QUEUED. ACB: 60 00 00 85 49 40 05 00 00 01 00 00
Dec 8 12:54:34 eureka kernel: (ada2:ahcich4:0:0:0): CAM status: Command timeout
Dec 8 12:55:05 eureka kernel: ahcich4: Timeout on slot 9 port 0
Dec 8 12:55:05 eureka kernel: ahcich4: is 00000000 cs 00000000 ss 00000200 rs 00000200 tfd 40 serr 00000000 cmd 0000c917
Dec 8 12:55:05 eureka kernel: (ada2:ahcich4:0:0:0): READ_FPDMA_QUEUED. ACB: 60 00 00 89 49 40 05 00 00 01 00 00
Dec 8 12:55:05 eureka kernel: (ada2:ahcich4:0:0:0): CAM status: Command timeout
Dec 8 12:55:35 eureka kernel: ahcich4: Timeout on slot 20 port 0
Dec 8 12:55:35 eureka kernel: ahcich4: is 00000000 cs 00000000 ss 00100000 rs 00100000 tfd 40 serr 00000000 cmd 0000d417
Dec 8 12:55:35 eureka kernel: (ada2:ahcich4:0:0:0): READ_FPDMA_QUEUED. ACB: 60 00 00 8d 49 40 05 00 00 01 00 00
Dec 8 12:55:35 eureka kernel: (ada2:ahcich4:0:0:0): CAM status: Command timeout
Dec 8 12:55:35 eureka kernel: (ada2:ahcich4:0:0:0): Retrying command
But what do they mean? Clearly it's the disk ada2, but there are four different slots, and the status differs each time. It does suggest, though, that the drive is on its way out. The addresses in question are out of the range of the swap partition, however. Now that small SSDs are so cheap, there's no point worrying about it: next time I get this kind of failure, I'll swap swap for a spare new drive that I have waiting for “just in case”.
In passing, the messages from one system after eureka came back up again:
nfs snerfvser serveureekar:/ : enoutr reeskpondinag:
/: not responding
nfs server nfesur eka:s/:e rivs earli ve againe
ureka:/: is alive again
We could fix that, of course: it's a locking issue, but it's so amusing.
|
The battle of the rangefinders
|
Topic: photography, history | Link here |
Seen today: The battle of the rangefinders, a comparison of the development of the Leica camera and Zeiss Contax.
| Monday, 9 December 2024 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Medication: life or death?
|
Topic: health, opinion | Link here |
Yvonne is recovering well from her cardioversion last week, but a question remains: what medication should she be taking? Various statements suggested that the attack was due to inappropriate medication. But despite our repeated requests, she has had nothing in writing about what the correct medication should be. Yvonne tried to get an appointment with Rodney Reddy, but he didn't want to preempt Professor Peter Kistler. And Peter hadn't answered our email from last Wednesday. People, this is almost a matter of life or death!
Sent off a rather more strongly worded message to both of them, asking for clear medication instructions in writing today. And how about that, we really did get a response from Peter:
We received your email Friday afternoon.I would recommend flecainide 50mg bd regularly doubling the dose if the arrhythmia recurs. Once you are back in sinus rhythm then the dose can return to 50mg bd. Resume digoxin at its previous dose.
Well, it's not clear where the email was received on Friday afternoon. It was delivered to their designated mail server at 15:12 on Wednesday, 4 December:
Dec 4 04:12:11 lax postfix/qmgr[35201]: 11F6C280C7: from=<yvonne@lemis.com>, size=14538096, nrcpt=2 (queue active)
Dec 4 04:12:16 lax postfix/smtp[84548]: 11F6C280C7: to=<reception@melbourneheart.com.au>, relay=melbourneheart-com-au.mail.protection.outlook.com[52.101.149.9]:25, delay=49, delays=44/0.02/0.61/4.3, dsn=2.6.0, status=sent (250 2.6.0 <20241204041120.GI33477@lagoon.lemis.com> [InternalId=5583457491310, Hostname=SY0PR01MB8858.ausprd01.prod.outlook.com] 14547204 bytes in 2.238, 6346.342 KB/sec Queued mail for delivery)
The times are UTC, and we're 11 hours ahead. But somehow this outlook.com domain looks suspicious. Could they really have held on to it for over 24 hours? This time he got it quickly enough:
Dec 9 02:24:34 lax postfix/qmgr[35201]: EB47928093: from=<yvonne@lemis.com>, size=2533, nrcpt=2 (queue active)
Dec 9 02:24:36 lax postfix/smtp[77395]: EB47928093: to=<reception@melbourneheart.com.au>, relay=melbourneheart-com-au.mail.protection.outlook.com[52.101.149.1]:25, delay=3, delays=0.2/0.02/0.71/2, dsn=2.6.0, status=sent (250 2.6.0 <20241209022432.GA40794@lagoon.lemis.com> [InternalId=142348101104641, Hostname=ME3PR01MB6195.ausprd01.prod.outlook.com] 11494 bytes in 0.223, 50.237 KB/sec Queued mail for delivery)
And the response came at Mon, 9 Dec 2024 03:38:03 +0000, only a little over an hour later.
And “bd”? I guessed “bi-daily”, but it proves that it's short for “bis die”, not a mixture of German and English, but Latin for “twice per day”. It's interesting to read the comment in the link:
AMA style avoids use of this abbreviation (spell out "twice a day")
Ultimately we also got confirmation from Rod's nurse on the phone, and after a request she sent a confirmation that agreed with Peter. So at least we now have agreement. But wouldn't it have been so much clearer to say this?
| Trade name | Medication | Dose | Frequency/day | |||
| Digoxin | 62.5 μg | 1 | ||||
| Eliquis | Apixaban | 5 mg | 2 | |||
| Tambocor | Flecainide | 50 mg | 2 |
|
Another technology failure
|
Topic: general, technology, opinion | Link here |
Yvonne has had trouble with fossil, her phone, Yet Again. It failed in the most basic of functions, making phone calls. It went through the motions, but didn't connect, and incoming calls went straight to voice mail. Much messing around. The settings screen showed that it had forgotten its phone number—maybe. At least it wasn't showing, while it does show on mine. Power down, up, no change. Power down, reseat SIM card, power on. Now it works.
Somehow we have had so many issues with modern technology recently. Will it continue?
|
More photo software thoughts
|
Topic: photography, technology, opinion | Link here |
After yesterday's consideration, I tried DxO PhotoLab 8 again. There's really very little difference, and for that the price of US $229 is just too high. So I'll just forget all of them.
But then I heard of a new product with a hard-to-spell name, Aiarty. Free! Installed that, and later went back to the web site, where everything seems to have changed. Ah, it was Aiarty Image Matting, just part of their product, specializing in background removal. Can I even use it on its own? I can find out later.
| Tuesday, 10 December 2024 | Dereel | Images for 10 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Larissa problems?
|
Topic: animals, opinion | Link here |
We've known that Larissa has conformation problems almost since we got her, and she's been receiving shots of catrophen every few months.
Lately Yvonne had the impression that things were getting worse. She set off to see Pene Kirk, braving multiple road works in the process, and Pene agreed with her. She came back with a note from Pene that read:
Bilat — Rt looser than left
Cruciate laxity with anterior drawer motion.
Hyperextended hocks → abnormal weight bearing.
I'm pretty sure that I understand that, but it's better to record the original than an interpretation. Pene can't guess how the problem might progress. She might get to the point where she needs daily painkillers within months, or it could go on for years. Lara is nearly 4 now, about half way through a normal Borzoi life span. For the moment it's “wait and see”.
|
I'm not making this up
|
Topic: technology, language, fiction, opinion | Link here |
Seen in my email today:
8 N + 10-12-2024 To groggyhimself@lemis.c ( 292) HR Department N + Annual Leave Nonfiction
Spam, of course. We shut down our HR Department last year (no, come to think of it, that's fiction). But what kind of illiterate scammer can confuse a presumed “notification” with “nonfiction”?
|
Israeli non-aggression
|
Topic: politics, opinion | Link here |
The fall of the Assad regime in Syria was one of the biggest surprises we've seen in a surprising year. And one of the questions, of course, was “what will Israel do now?”. But we got reassurances: Israel would not invade Syria, just form a defensive barrier round the Jolan Heights.
What did they do? In contravention of UN resolutions, they attacked Syria. Just to destroy weapons that could fall into enemy hands. Israel, you are the enemy in Syria, and war criminals to boot. And with Donald Trump behind them, they're safe—for the moment.
Of situations like this world wars are made.
|
Letting Bruno outside
|
Topic: animals | Link here |
Bruno really, really wants to go outside. Why aren't we letting him? He's the first cat we have ever kept only inside.
Our concern is simply that he will get out onto the road and get run over. But will he even go in that direction? Piccola never did. Today we let him out under supervision. He really wanted to go behind the house, but the gate was shut, so he ran into the hay shed:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
And after a while he was happy to be picked up and taken inside. Can we repeat this? I'm still uncomfortable.
| Wednesday, 11 December 2024 | Dereel → Sebastopol → Dereel | Images for 11 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Car servicing
|
Topic: general | Link here |
Yvonne's car was due for a service, so she had it done today after shopping. And she had had my car booked in for service after that. Certainly it was time,
|
Chasing up photo processing issues
|
Topic: photography, technology, opinion | Link here |
I still haven't really identified the cause of the strange problems that I have had with Exif data processing on my OM System OM-1 Mark II. I have established that the problems occur with DxO PhotoLab (release 5), but not with OM Workspace. For some reason, this particularly boring photo resisted correct processing:
|
||||||||||||||||||||||||
The steps and issues were:
So, what about DxO PhotoLab 8? Does the problem still occur? Put all of the photos for 5 December through PhotoLab 8, using the new DeepPrime XD/XD2s processing, whatever that means (excess?).
DxO has always been glacially slow, but now there's a difference:
|
|||||||||||||||||||||||
7 minutes, 22 seconds per image! That's unbelievable. A normal image takes about 6 seconds.
Yes, it's unbelievable. Microsoft joined in the fun and put distress to sleep while I was in town swapping cars with Yvonne. A more accurate measure of progress was 11 images after 42:9 minutes, or about 3:45 minutes per image, still impressive.
And what did they look like? The good news: the Exif data for the problem image was correct. So maybe it is time to upgrade my DxO.
The DeepPrime images were more of an issue. I didn't get the same parameters, because I couldn't set the “MySet” profile with version 8. So the new results were generally darker. They did look better, though. Here the result from version 5 with MySet and version 8 without (run the cursor over an image to compare it with its neighbour):
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Yes, there's considerably less noise, here to the left of Bruno's head:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
On the other hand, the images have been processed differently, so I'll need to repeat things after I find a way (probably somewhere in this diary) to move my settings to the new version.
But what's this below?
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It's on a number of photos in exactly the same place:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
This place in the other images is much brighter, so it's difficult to identify it, but clearly it's a dead pixel (or even group of pixels). Damn! Hopefully I can fix that.
And the other thing that I hadn't expected: the detail images, created by ImageMagick's convert program, also have broken Exif:
=== grog@hydra (/dev/pts/32) ~/Photos/20241205 1575 -> exifcopy orig/XC050208.ORF Mona-Bruno-9-X-detail-1.jpeg
exiftool -overwrite_original_in_place -TagsFromFile orig/XC050208.ORF -all>all -title=XC050208 Mona-Bruno-9-X-detail-1.jpeg
Error: Format error in file - Mona-Bruno-9-X-detail-1.jpeg
0 image files updated
1 files weren't updated due to errors=== grog@hydra (/dev/pts/32) ~/Photos/20241205 1576 -> exifx Mona-Bruno-9-X-detail-1.jpeg
File Mona-Bruno-9-X-detail-1.jpeg
Date taken: Thursday, 5 December 2024, 20:13:20
Size: 300 x 225 pixels (0.07 megapixels)
Author: Greg Lehey
So I'm not out of the woods yet.
| Thursday, 12 December 2024 | Dereel | Images for 12 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Where did that file come from?
|
Topic: technology | Link here |
While syncing my diary to www.lemis.com, found this:
rsync: write failed on "/home/grog/www.lemis.com/grog/diary-dec1964.php": No space left on device (28)
rsync error: error in file IO (code 11) at receiver.c(381) [receiver=3.1.3]
Huh? Overflowing logs again? But no, after tidying up there I still had no space on disk. Finally I traced it to a file /var/tmp/webfoo36468, 1.6 GB in size, starting on 12 October. Where did that come from? It contained Apache logs, but they're normally stored in /var/log/www, and that's what I had just rotated. Where's lsof when you need it? Not installed, and not installable without kernel sources. While I was pondering what to do, it went away. Why? It didn't occur to me until later that apachectl graceful may leave processes behind that keep files open until they finish.
Things didn't stop there. A couple of hours later I had:
=== root@lax (/dev/pts/1) /var/tmp 39 -> df .
Filesystem 1048576-blocks Used Avail Capacity Mounted on
/dev/ufs/rootfs 54,550 31,561 18,624 63% /
What was all that? Has my Apache been storing in strange places?
|
Still more pixel peeping
|
Topic: photography, technology, opinion | Link here |
So why do I have this light spot in the images I took yesterday? It looks for all the world like a stuck pixel. OK, we have a pixel remap function in the menus. Try that and take a photo at 100,000/51° ISO with the lens cap on:
|
|||||||||||||||||||||||
|
|
|||||||||
Oh. That's not at all better. After a lot of comparison processing, came up with an “unprocessed” conversion (“no correction” in DxO-speak), which came up with this detail:
|
|
|||||||||
DxO problem? I should at least try with OM Workspace. Not surprisingly, there was no difference:
|
|
|
|||||||||
|
|
|||||||||
It's not nearly as bad at lower sensitivities. Here the same detail view at 6400/39° and 200/24°:
|
|
|||||||||
|
|
|||||||||
But clearly there's something seriously wrong there. I fear I'm going to have to brave the grey market warranty service. And in that connection it seems that there's another issue that had puzzled me for some time, a white spot in the viewfinder round the same position, to bottom left of the detail image:
|
|||||||||||||||||||||||
|
|
|||||||||
Could it be that this is something other than a sensor issue? I can't see how this display could be influenced by a minor sensor defect.
In the process, it has become clear that the Exif data fails even under normal circumstances, processing with the OM System software. I couldn't find a way to copy the Exif data to any of the detail images. At least I now have enough info to put in an error report.
|
DeepPrime XD/XD2s in more detail
|
Topic: technology, photography, opinion | Link here |
So how much better are the results from DxO PhotoLab's new XD/XD2s noise reduction? Yesterday it looked as if there was a significant improvement, but the processing parameters were different. Today I tried a different motive:
|
|||||||||||||||||||||||
I took three images at 200/24°, 6400/39° and 100,000/51° and processed them five ways: DxO 5 normal, DxO 5 DeepPrime, DxO 8 normal, DxO 8 DeepPrime and DxO 8 DeepPrimeXD. I couldn't see any difference between DxO 5 and DxO 8. For normal sized images there was almost no difference to be seen between the processing options under DxO 8. The image above was the one taken at 100,000/51° and normal processing. But looking more carefully shows noise in the background, and DeepPrimeXD removes it almost completely (in all these examples, run the cursor over an image to compare it with its neighbour):
|
|
|||||||||
|
|
|||||||||
But looking more carefully, things aren't all that good. What's the text on the second line of the Praktica camera? Here another example from the same view:
|
|
|||||||||
|
|
|||||||||
That's extreme, of course, and it's not clear if the DeepPrimeXD image isn't better anyway. At more sane sensitivities, here 6400/39°, and with normal DeepPrime in the middle, we have:
|
|
|||||||||
|
|
|||||||||
|
|
|||||||||
Arguably DeepPrimeXD is still better, but is it better enough that I should pay US $229 for it? That seems to be the only advantage now that Exif data is broken here too.
In passing, it's interesting that DeepPrime usually reduces the size of the output image:
-rw-r--r-- 3 grog lemis 3,008,451 12 Dec 11:26 Cameras-DxO-8-24.jpeg
-rw-r--r-- 3 grog lemis 2,693,444 12 Dec 11:26 Cameras-DxO-8-24-DeepPrime.jpeg
-rw-r--r-- 3 grog lemis 2,786,031 12 Dec 11:26 Cameras-DxO-8-24-DeepPrimeXD.jpeg
-rw-r--r-- 3 grog lemis 5,025,613 12 Dec 11:26 Cameras-DxO-8-39.jpeg
-rw-r--r-- 3 grog lemis 3,347,113 12 Dec 11:26 Cameras-DxO-8-39-DeepPrime.jpeg
-rw-r--r-- 3 grog lemis 2,934,782 12 Dec 11:26 Cameras-DxO-8-39-DeepPrimeXD.jpeg
-rw-r--r-- 3 grog lemis 8,346,955 12 Dec 11:26 Cameras-DxO-8-51.jpeg
-rw-r--r-- 3 grog lemis 4,916,517 12 Dec 11:26 Cameras-DxO-8-51-DeepPrime.jpeg
-rw-r--r-- 3 grog lemis 3,384,887 12 Dec 11:26 Cameras-DxO-8-51-DeepPrimeXD.jpeg
|
Bruno's next adventure
|
Topic: animals | Link here |
Let Bruno out again today, on the verandah. He went round the bushes on the north side of the house and ended up in front of my office with a big, bushy tail for which I could find no reason. I left it at that and brought him back outside.
| Friday, 13 December 2024 | Dereel | Images for 13 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
ALDI phở
|
Topic: food and drink, opinion | Link here |
While in town last week, I bought some instant soup packs at ALDI. Here their beef Phở:
|
|||||||||||||||||||||||
Why? I already have my own recipe for phở, and I'm relatively happy with it. But how authentic is it? And the real reason was the other package, ramen soup. I don't really have a good recipe for that, so I bought a pack to compare. And the phở was as a check for the ALDI brand.
Surprise, surprise. The recipe takes longer than my own!
|
|||||||||||||||||||||||
Part of it involved cooking these spices in advance:
|
|||||||||||||||||||||||
And the cooking time for the noodles ranks with the worst I have seen: “Cook Rice Noodles in 4 cups of boiling water for 2-3 minutes or until noodles are cooked”. Which? I cooked them until they were cooked, not in cups, but in a saucepan, and it took 9 minutes.
The result? Acceptable, but no more. The broth was boring. I have another serving, and I think I'll add a considerable quantity of my existing phở concentrate.
|
More E-M1 Mark II strangenesses
|
Topic: photography | Link here |
Yesterday I got evidence of a spot to the left of the monitor display on my new OM System OM-1 Mark II, slightly above the middle:
|
|||||||||||||||||||||||
But what are those symbols to the left of the imperfection? Clearly the left-hand icon and percentage is information about the battery charge. But the one to the right of it? At normal magnification (above) it's unrecognizable. Only with the detail image can I recognize anything:
|
|
|
|||||||||
It looks like a crossed-out battery symbol. But why? And what does PBH mean? After considerable searching, once again with the aid of Google Gemini, discovered that it means “Power Battery Holder”.
But I don't have no steenking Power Battery Holder. Why do I get the display in the first place? At Gemini's suggestion, went through the menus and found that the PBH was set as preferred power source. That's easy enough to change, but how did that end up that way? And it didn't make any difference to the display. Is that maybe normal, or is there something (else) wrong with the camera?
| Saturday, 14 December 2024 | Dereel | Images for 14 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
ALDI ramen soup
|
Topic: food and drink, opinion | Link here |
Had the other ALDI soup pack today, the Ramen. It's not very interesting, and I didn't take any photos. There was a problem: the noodles had to be cooked in the soup, and they were a single mass stuck together, so I had to prepare both portions at once. And once again the noodle cooking times were hopelessly optimistic. After the specified 2 minutes, I wasn't even able to separate them! But it “only” took 4 minutes, as opposed to 9 minutes for the rice noodles yesterday.
And the flavour? Again, somewhat boring. About the only surprise was that it was somewhat sweet, not an improvement. Somehow I was hoping for more insights. But of course there's no reason to believe that either of these soup packs was in any way authentic.
|
Letting Bruno out
|
Topic: animals | Link here |
Once again I let Bruno outside today. I lost him for a minute, but he had a guard:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
As expected, he didn't try to go very far from the house. My main concern was that he was so close to the neighbours' fence.
|
Understanding NTFS SD card
|
Topic: technology, photography, opinion | Link here |
What's on the new SD card that I bought last month? It claims to be an NTFS card. Can that be right? Certainly my FreeBSD machines couldn't recognize it, though it worked fine on distress, the Microsoft box. In principle that points to a deficiency in FreeBSD, and I know that our NTFS implementation is not the best, but can cameras understand NTFS? Today I put to test the hypothesis that it might really be a FAT32 file system with incorrect partition information.
=== root@hydra (/dev/pts/4) /usr/ports/print/texlive-full 327 -> gpart show da3
=> 63 124211137 da3 MBR (59G)
63 32705 - free - (16M)
32768 124178432 1 ntfs (59G)=== root@hydra (/dev/pts/4) /usr/ports/print/texlive-full 328 -> gpart delete -i 1 da3
da3s1 deleted=== root@hydra (/dev/pts/4) /usr/ports/print/texlive-full 333 -> gpart add -b 32768 -t fat32 da3
da3s1 added=== root@hydra (/dev/pts/4) /usr/ports/print/texlive-full 335 -> gpart show da3
=> 63 124211137 da3 MBR (59G)
63 32705 - free - (16M)
32768 124178432 1 fat32 (59G)=== root@hydra (/dev/pts/4) ~ 339 -> mount -t msdosfs /dev/da3p1 /mnt
mount_msdosfs: /dev/da3p1: Invalid fstype: Invalid argument
Not exactly a confirmation. I'll have to look more deeply. But in passing, it's interesting that this layout wastes 16 MB at the beginning of the card! Once upon a time that was a large proportion of the space on the card. Twenty years ago I bought a new camera with a 128 MB SD card; this layout would have wasted ⅛ of the total space.
|
Merging file systems: the pain
|
Topic: technology | Link here |
It's been well over a year since I got my new machine, hydra.lemis.com, and I still haven't migrated everything from eureka. I know why. Today I ran into problems with my Emacs initialization file ~/.emacs. I had an older version on hydra than on eureka. OK, simple: move it and its RCS file ~/RCS/.emacs,v from eureka to hydra and replace the versions on eureka with a symlink to hydra.
But it didn't work! I already had symlinks in place, and I managed to blow away my control file. Much searching through backups, taking up much of the afternoon. And that for one file! No wonder I'm slow on moving entire directory hierarchies.
|
Love cakes
|
Topic: food and drink, language, opinion | Link here |
The Advent calendar that Jane Ashhurst gave us proves to be a source of considerable interest. It has a French flavour also evident in the intricate decoration, but today's spice was Gingerbread spice, something that we would consider German rather than French, and which was somewhat confirmed by the detailed description.
But that description omits the recipe that was on the packaging:
|
|||||||||||||||||||||||
Love cake? Why? Does it relate to the German name for gingerbread, Lebkuchen? Spent some time considering: „Kuchen“ means cake. And „Leb“? “Live” maybe, but it's also possible that they (mis?)interpreted it as “love”. In German that would be something like „Liebkuchen“.
OK, what does Duden know? Checked out the blue Duden, which says that the name is inconclusive, but could be related to „Laib“ (“loaf”).
|
More X strangeness
|
Topic: technology, opinion | Link here |
Turned on the TV in the afternoon and... no signal. What, has teevee gone down? No, it was still up, but somehow X had stopped. And this stupid breakage with the display meant that I had to restart X from another system.
That worked, but lots of settings were wrong. In particular, the display of special characters was wrong. Starting new xterms worked, but what was wrong with the old ones? Here the parameters for the old and new:
grog 8663 0,0 0,1 39940 25724 9 I Sa.17 0:00,30 xterm -name xterm-lx -bg BlanchedAlmond -s -sl 2048 -sb -ls -j -rw -display :0.0 -geometry 90x50+15+0 -fn 10x20 -e /usr/local/bin/bash
grog 9288 0,0 0,0 20400 8184 9 I Sa.19 0:00,02 ssh -A tiwi LC_ALL=en_AU.UTF-8 /usr/local/bin/uxterm -name xterm-lx -s -sl 2048 -sb -ls -j -rw -display teevee:0.0 -geometry 90x50+53+0 -e /usr/local
It looks as if it started with old configuration files. What a mess this is!
|
Ham in the hair dryer again
|
Topic: food and drink, opinion | Link here |
At Easter we cooked a netted roast ham in the “bathtub” “air fryer”. Not a complete failure, but getting the net off proved to be more trouble than I had expected. Today I took it off first, still not the easiest thing, and cooked it at 160°. The ham weighed only 320 g, but it took 40 minutes, at least 10 minutes longer than I expected. The results were acceptable, though the meat was still quite browned, and normal carving knives couldn't cut through the crust:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
| Sunday, 15 December 2024 | Dereel | Images for 15 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
“Gardening” again
|
Topic: gardening, food and drink | Link here |
Somehow it seems ridiculous to buy Thai basil and then freeze it—from which it suffers significantly—when I can grow my own. So I ordered some from eBay, because it was not available in the shops in Ballarat. $2.50 including postage of $1.50, as compared to $4.50 or $5.00 for similar seeds in Ballarat.
So for the first time in I don't know how long I planted something! Judging by the potting mix, it must have been lying there for a couple of years.
|
More Microsoft surprises
|
Topic: technology | Link here |
One of the strangest things about my remote desktop connection to distress (a Microsoft machine) is that setting up the connection keeps changing. It works much worse for Yvonne than for me, but recently I have been getting this message:
|
|
|||||||||
I can select either one, but I can't find a way to close them. There's nothing to select in the message beyond the choice of session. Leaving the session with “Disconnect” works, but the session is still there when I return. I suppose it will go away when I get the next enforced reboot.
|
More Emacs problems
|
Topic: technology, opinion | Link here |
On teevee, tried to edit something. I have C-x B bound to buffer-menu, but it didn't work, and of course I couldn't remember the name of the function. And other key bindings, including the ones that are stored in my cerebellum, didn't work, notably M-p, which is bound to repeat-complex-command. I couldn't even copy text!
It wasn't due to the version of Emacs: on teevee it's 29.4, the latest and greatest, and on tiwi it's 28.1. But in particular It Used To Work. And, started from teevee, it misbehaved on tiwi and hydra as well. Only on eureka did it work correctly.
And it complained about the function set-default-font. That's in ~/.emacs, which I messed around with yesterday, and Emacs stops processing the file if it finds an error. But this was at the very end of the file, so it shouldn't have been a problem.
More looking around on eureka showed:
set-default-font is an alias for `set-frame-font' in `frame.el'.
(set-default-font FONT &optional KEEP-SIZE FRAMES)
This function is obsolete since 23.1;
use `set-frame-font' instead.
Oh. So why has it worked so far? Changed the name of the function, read in ~/.emacs again, and that worked, but other things still didn't work as I expected. The header menus on teevee and tiwi looked different, as did the highlight colours.
And then I discovered that I had a new ~/.emacs on eureka! I had carefully replaced it with a symlink yesterday. How did it get back?
This has been a particularly strange issue, but it helps me justify the lack of progress in moving the files around.
| Monday, 16 December 2024 | Dereel | Images for 16 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More SD card investigations
|
Topic: technology, photography, opinion | Link here |
So why can't hydra recognize the file system on my new SD card? I suspected that the partition type (NTFS) was wrong, and I changed it to FAT32. But that didn't work either.
OK, is the partition correct? Does Microsoft even look at the partition type? Put the card into distress, which happily read the files. So no, Microsoft doesn't seem to look at the partition type.
So: is it even FAT32? What does a hex dump show?
=== grog@hydra (/dev/pts/32) ~/Photos/20191215 1755 -> hd /dev/da3s1 | less
00000000 eb 76 90 45 58 46 41 54 20 20 20 00 00 00 00 00 |.v.EXFAT .....|
00000010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
Is there a file system called EXFAT? Yes. Well, exFAT. And how about that, it shares a partition type with NTFS:
ntfs A partition that contains a NTFS or exFAT
filesystem. The scheme-specific type is "!7"
for MBR.
So it seems that the card was correctly formatted. But do normal cameras understand it? It seems that it has been a Microsoft proprietary file system until 2019. Put it into the oldest camera I had that supports SD cards, my Nikon “Coolpix” L1, introduced in September 2005. It didn't want to know. Considering that exFAT was introduced in 2006, that's not overly surprising. OK, the next oldest is the Olympus E-PM1, introduced in June 2011. And yes, it has no trouble reading and formatting the card.
So somehow both FreeBSD and Mtools are at least 13 years out of date.
More searching, and found this web discussion, which required some interpretation to come to a useful understanding. Why do people set up system configuration and reboot when they're just testing something? That comes (much) later. But with
=== root@hydra (/dev/pts/4) ~ 439 -> pkg install fusefs-exfat
...=== root@hydra (/dev/pts/4) ~ 442 -> mount.exfat /dev/da3s1 /mnt
FUSE exfat 1.4.0 (libfuse2)=== root@hydra (/dev/pts/4) ~ 443 -> ls -lR /mnt
total 1
drwxrwxrwx 1 root wheel 131072 16 Dec 12:14 ALBM
drwxrwxrwx 1 root wheel 131072 16 Dec 12:14 DCIM
/mnt/ALBM:
total 1
-rwxrwxrwx 1 root wheel 326940 16 Dec 12:14 ALB27ZY1.BIN
/mnt/DCIM:
total 1
drwxrwxrwx 1 root wheel 131072 16 Dec 12:14 108OMSYS
/mnt/DCIM/108OMSYS:
total 35
-rwxrwxrwx 1 root wheel 17867479 16 Dec 12:14 AC160264.ORF
-rwxrwxrwx 1 root wheel 17973579 16 Dec 12:15 AC160265.ORF
And then I can add this to /boot/loader.conf:
fusefs_load="YES"
Others want to load fuse as well, but until proof of the contrary fusefs is enough.
In passing, while messing around I saw this:
WARN: volume was not unmounted cleanly.
You don't see that with FAT. And how do you recover? In my case, I can just format the card in the camera. But do I want to? Although the mount variety looks promising, it breaks my syp script, which uses mtools to establish which files to copy. Now I need to mount the file system—can I do that as a normal user?—and umount it when I'm done. Can't I just format the card as FAT32? It seems that FAT32 can go way beyond the artificial 32 GB limit that Microsoft has set. But that's for another day.
|
Another X crash?
|
Topic: technology, multimedia | Link here |
Turned the TV on in the afternoon. No display! I had this only two days ago. Back to hydra to restart X, but it was already running:
=== grog@teevee (/dev/pts/7) ~ 1 -> ps aux | grep Xorg
root 8655 0.0 0.4 25300588 84728 9 S Sat17 0:57.91 /usr/local/libexec/Xorg :0 -config xorg.conf -listen tcp -auth /home/grog/.serverauth.8636
What was it? Cable problem? Keyboard? Nothing obvious. Wobbling the cables had no effect, and when I restarted it (without difficulty) the keyboard worked normally.
| Tuesday, 17 December 2024 | Dereel | Images for 17 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
DxO PhotoLab 8: last chance
|
Topic: technology, photography, opinion | Link here |
My trial of DxO PhotoLab 8 is running out. Do I want to buy it? The only feature of any use is the improved DeepPrimeXD/XD2s. By chance, while testing the exFAT file system, I ended up with a relatively noisy test photo, taken with the Olympus E-PM1 at 3200/36° ISO. How does it compare? Here first normal conversion, then the “conventional” DeepPrime that I have with DxO PhotoLab 5, then the DeepPrimeXD/XD2s that comes only with PhotoLab 8 (run the cursor over an image to compare it with its neighbour):
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Is there a difference? Yes, definitely. In particular, the DeepPrimeXD/XD2s image has slightly different colour gradation. But even with this relatively noisy image, I really need to look to see it, for example under the desk. And it took 2 minutes, 10 seconds, compared to 51 seconds for the normal DeepPrime and 4 seconds for normal conversion. I think I'll wait with any DxO upgrade until I really need the difference.
|
More file system experiments
|
Topic: technology, photography, opinion | Link here |
Yesterday I established that the file system on my new 64 GB SD card was exFAT, and that that's correct for this size card. But it's inconvenient: my processing scripts are based on FAT, and mtools can't handle exFAT. On the other hand, there's no restriction that prevents me from creating a FAT file system on the card.
Well, that's what I thought:
=== root@hydra (/dev/pts/4) ~ 396 -> gpart show da3
=> 63 124211137 da3 MBR (59G)
63 32705 - free - (16M)
32768 124178432 1 ntfs [active] (59G)=== root@hydra (/dev/pts/4) ~ 397 -> gpart delete -i 1 da3
gpart: Device busy
OK, remove and replace card.
=== root@hydra (/dev/pts/4) ~ 402 -> gpart delete -i 1 da3
da3s1 deleted=== root@hydra (/dev/pts/4) ~ 403 -> gpart add -t fat32 da3
da3s1 added=== root@hydra (/dev/pts/4) ~ 404 -> gpart show da3
=> 63 124211137 da3 MBR (59G)
63 124211137 1 fat32 (59G)=== root@hydra (/dev/pts/4) ~ 405 -> mformat d:
Disk size not known
Oh.
=== root@hydra (/dev/pts/4) ~ 408 -> fdisk da3s1
...
fdisk: invalid fdisk partition table found
The data for partition 1 is:
sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
start 63, size 124198452 (60643 Meg), flag 80 (active)
beg: cyl 0/ head 1/ sector 1;
end: cyl 562/ head 254/ sector 63
That looks like guesswork on the part of fdisk. There's nothing else in the FreeBSD space that will help, so I tried Microsoft. But:
|
|
|||||||||
It doesn't even offer the option of FAT32! How about the good old FORMAT program from MS-DOS days? No:
|
|
|||||||||
Off to visit my bard, which I'm using more and more with mainly positive results.
Due to limitations in Windows' built-in formatting tools, you'll need to use a third-party tool to format a 64GB SD card to FAT32.
It went on to recommend MiniTool Partition Wizard, though it was too polite to supply a URL, and since then it has changed its mind anyway. But I found it and was able to format the card and access it from FreeBSD.
But not from my cameras. “Card error”. I reformatted it, and it came up as exFAT. It seems that they insist on exFAT for larger SD cards. I wish I had known that before I bought the card; I only bought it because it was so cheap by comparison.
OK, back to see what I can do with exFAT. Gemini had given me a pointer to a package called exfat-utils, which has amazingly poor documentation. The only information I could find was in /usr/ports/sysutils/pkg-descr, which reads:
Utilities to manage extended file allocation table filesystem. This package
provides tools to create, check and label the filesystem.
It contains dumpexfat to dump properties of the filesystem, exfatfsck to report
errors found on a exFAT filesystem, exfatlabel to label a exFAT filesystem and
mkexfatfs to create a exFAT filesystem.
On the positive side, it does have man pages for the programs.
OK, try that.
=== root@hydra (/dev/pts/4) /usr/ports/filesystems/exfat-utils 483 -> mount.exfat /dev/da3s1 /mnt
FUSE exfat 1.4.0 (libfuse2)
WARN: volume was not unmounted cleanly.
Oh. I saw that yesterday. But now I have an exfatfsck. First umount and remount:
=== root@hydra (/dev/pts/4) /usr/ports/filesystems/exfat-utils 484 -> umount /mnt=== root@hydra (/dev/pts/4) /usr/ports/filesystems/exfat-utils 485 -> mount.exfat /dev/da3s1 /mnt
FUSE exfat 1.4.0 (libfuse2)
And the message is gone! Still,
=== root@hydra (/dev/pts/4) /usr/ports/filesystems/exfat-utils 487 -> exfatfsck -a -n /dev/da3s1
exfatfsck 1.4.0
Checking file system on /dev/da3s1.
File system version 1.0
Sector size 512 bytes
Cluster size 128 KB
Volume size 59 GB
Used space 52 MB
Available space 59 GB
Totally 3 directories and 3 files.
File system checking finished. No errors found.
OK, that seems reasonable enough. Now I just need to wonder how to get the files off the file system. I can use cp, of course, but that doesn't copy to a directory related to the date taken. I suppose I can modify my syp script, but that's irritating.
In passing, discovered a new FreeBSD program, fstyp, which identifies a file system on a disk. It doesn't appear to be overly clever, but it does recognize my SD card:
=== root@hydra (/dev/pts/4) ~ 438 -> fstyp /dev/da3s1
exfat
|
Australia Post excels itself
|
Topic: general, photography, opinion | Link here |
On 1 December I bought a camera on eBay, a particularly primitive Voigtländer Bessa that must date to the mid-1930s:
It only has two shutter speeds, 1/25 s and 1/75 s, and the lens has a maximum aperture of f/7.7. I bought it mainly because it was cheap ($26.50). I'm in no hurry to get it here, but of course I've been following its progress here. Today, according to eBay, Australia Post tried to deliver it, but I wasn't home.
|
|
|||||||||
Nonsense, of course. I need to pick it up in Napoleons. But why only today? That's 16 days since I paid for it! Here an overview:
| Date | Location | |
| 1 December | Purchased | |
| 5 December | Posted, Adelaide | |
| 6 December | Adelaide airport | |
| 7 December | Arrived in Melbourne | |
| 13 December | Arrived in Ballarat | |
| 17 December | Sent to Napoleons |
That's amazing! A week from Adelaide airport to Ballarat, and another 4 days from one side of Ballarat to the other. Even Australia Post must think so: despite their own proof and claims that it's still in Winter Valley, a new suburb in the west of Ballarat, they claim that it will be delivered between last Friday and yesterday:
|
|||||||||||||||||||||||
|
|
|||||||||
I don't know which is worst, their amazingly slow delivery, their amazingly stupid tracking information or eBay's claims. To be fair to AusPost, though, it seems that something in their system has flagged a problem:
|
|
|||||||||
|
Still more problems on teevee
|
Topic: technology | Link here |
Turned on teevee again today. Was X working? Yes, with firefox displaying ancient photos that I was looking at yesterday. Iconified that, looked at something else, deiconified firefox and—it hung. Why? I restarted it from the fvwm menu and it hung again. I had to start it from an xterm.
Clearly this has something to do with the way I started X, but what?
| Wednesday, 18 December 2024 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Voigtländer's bottom of the line
|
Topic: photography, opinion | Link here |
My “new” camera is finally here, after only 17 days: a Voigtländer Bessa „Singlo“, if I interpret the Wikipedia page correctly:
When was it made? Not before 1935, but nobody can tell me how old it is. Still, it seems likely that it was made before the Second World War, and is thus the oldest camera I have, and with a 6×9 cm image size also the largest I currently have, 6.25 times the size of a “full frame”.
It's also the most primitive. That's not because of the age: both the Leica II and the Contax were on the market when it entered. The Contax had shutter speeds between 1/2 s and 1/1000 s, and an f/1.5 Sonnar lens was available for it. But the Bessa is clearly a cheap version, “bottom of the line” as I read somewhere. 10,5 cm f/7.7 „Voigtar“ lens, a shutter with only two speeds, 1/25 s and 1/75 s. I spent some time looking for a way to cock the shutter, confirming that there is no connection between the film advance lever and the shutter. Even more than with the Diaxette, which cocks the shutter when winding the film, the entire mechanism is in the lens assembly. On the Bessa, the shutter doesn't need cocking. Just press it, making double exposures child's play.
|
Bruno outside again
|
Topic: animals | Link here |
Bruno out again today, for no less than 4 times. Having the dogs there keeps him at bay: they chase him, and he's not overly happy about that. They also keep track of him.
Unfortunately he's still as hyperactive inside. I suppose we should give him even more exercise.
|
Google Gemini draws cats
|
Topic: technology, animals, opinion | Link here |
A couple of days ago Google Gemini offered to create photos for me. I know they've done this for some time, but this time I tried it out: “Make a photo of a Borzoi chasing a chocolate Burmese cat”. And how about that, not too bad:
Clearly we have a running Borzoi and a running cat. But the dog isn't chasing the cat, he's just running. And the size relationships are wrong: the dog should be about twice the size.
Time to investigate the details. More pictures. Twice the same thing: “draw me a photo of a chocolate burmese cat and a brown burmese cat lying in a basket together”. I was expecting the same photo twice, but no:
My bad. I really wanted a soft basket. “draw me a photo of a chocolate burmese cat and a brown burmese cat lying in a soft basket together”:
Not quite a soft basket, but OK. “draw me a photo of a chocolate burmese cat and a lack and white Borzoi lying in a soft basket together”. And yes, I didn't notice the typo until later, and Gemini didn't complain:
Blue eyes! And once again the dog was too small.
So far in the two cat photos, the brown has been on the left. What happens if I change the sequence in my question? “draw me a photo of a chocolate burmese cat and a brown burmese cat lying in a soft basket together”
Hmm. Is the one on the left supposed to be brown? But it's still marginally darker than the one on the right. Try again: “draw me a photo of a brown burmese cat and a chocolate burmese cat lying in a soft basket together”:
Now the choc is definitely on the right. And the basket has hardened. “draw me a photo of a brown burmese cat and a chocolate burmese cat lying in a soft basket together and licking each other”:
Somehow it must be getting tired. The cats both seem to be chocolate, the basket is still hard, and the left-hand cat is sticking out his tongue. That's not licking.
So: the photos look good with some reservations. The language parser still seems to need attention. But it's quite impressive. I tried to get some photos of Yvonne on Samba, expecting confusion on the part of Samba, but it refuses to draw photos of humans.
|
Wise?
|
Topic: technology, general, opinion | Link here |
A few days ago I received an email claiming to be from Wise, the banking service that used to be called Transferwise. It was quite well done, and it passed some of my usual checks, but it was clearly phishing, in particular because it was sent to the address that I have only given to eBay. Worth reporting to Wise so that they can warn others. But how did I do it? They only accept with topics that they have thought of. Finally fought my way through their mess and sent the message. At least I was able to attach the original, unchanged message. No response.
Two days later I received an HTML-only email from support@wise.com showing that they had not even read what I had written:
In all likelihood, the message or call you received was not sent by Wise.
Emails from @wise.com, @info.wise.com, and @feedback.wise.com are legitimate. Wise doesn’t send emails from @gmail.com, @yandex.com, @aol.com, or similar free email providers.
Our security team can investigate fraudulent messages. Please provide the full original copy of the fraudulent email with headers. How to get it depends on the mail provider, for more information see: Trace an email with its full headers.
After getting the email headers, forward them to phishing@wise.com. You have the option to send them as an attachment (.eml file) or by copying and pasting the email headers in plain text. Alternatively, you can also forward the original email that you received.
Grrr! According to that message, the message itself was invalid, since it didn't come from any of the named email addresses. And in fact, the phishing message did. You need to look at the headers to see the rather silly clues that something was wrong:
From info@wise.org Tue Dec 10 06:11:56 2024
Return-Path: <info@wise.org>
...
Received: from smtp.voyager.co.nz (mail1.voyager.co.nz [114.23.5.28])
by lax.lemis.com (Postfix) with ESMTP id 2263B280CE;
...
To: Recipients <info@wise.org>
From: "wise.com" <info@wise.org>
So I sent an annoyed message suggesting that they should read before replying, and got another message that actually had a good suggestion:
I understand that you would like us to forward this to our relevant team.
We have a separate team dealing with this, unfortunately we don't have any visibility on this matter.
If you think you’ve received a phishing email from somebody pretending to be Wise, report it here so we can look into it.
With this link, you can report the phishing incident and it will be directed to our relevant team.
In the mean time, to protect yourself against phishing - here's what you can do:
You can set up a Secure communication code. It will help you to tell if the emails you receive are from Wise or are phishing attempts from fraudsters. Once you’ve set your communication code, it will be included in all genuine marketing and automated Wise emails.
Most of that is abject stupidity, of course: “If you think you’ve received a phishing email from somebody pretending to be Wise, report it here so we can look into it”. And how do I report it so that they actually understand? But the last paragraph is good: something like a reverse password, and something that I have been advocating for some time. So I did that, and discovered that I could only use digits and letters, not even something like “finally a good solution”. But it worked, although I needed to look carefully at the response to find the code.
So today I got what appears to be a legitimate message: “You have a reply from Wise.com”. No “secure code”.
*sigh*
| Thursday, 19 December 2024 | Dereel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
More system strangenesses
|
Topic: technology | Link here |
Checking the overnight logs today brought a couple of surprises. First, from hydra only,
Limiting closed port RST response from 29 to -4 packets/sec
Limiting closed port RST response from 11 to -5 packets/sec
Limiting closed port RST response from 2 to 26 packets/sec
Limiting icmp unreach response from 13 to 2 packets/sec
Limiting icmp unreach response from 28 to -3 packets/sec
Limiting icmp unreach response from 28 to -6 packets/sec
That relates to a change I made last year to stop ridiculous quantities of PING ICMP packets I receive. At the time I reduced the number from 200 per second to 5, but I've since increased it again to 10. But that was on eureka, where it works as expected:
Limiting icmp ping response from 11 to 10 packets/sec
Limiting icmp ping response from 12 to 10 packets/sec
Limiting icmp ping response from 13 to 10 packets/sec
In passing it's interesting to note how few messages are dropped. It seems that the intruders are adapting. But why so many from hydra? They need to go through eureka to get there. And why such obviously incorrect values?
And then there are more issues with the swap SSD on eureka:
Dec 18 22:56:32 eureka kernel: swap_pager: indefinite wait buffer: bufobj: 0, blkno: 755058, size: 4096
Dec 18 22:56:35 eureka kernel: ahcich4: Timeout on slot 10 port 0
Dec 18 22:56:35 eureka kernel: ahcich4: is 00000000 cs 00000000 ss 00000c00 rs 00000c00 tfd 40 serr 00000000 cmd 0000cb17
Dec 18 22:56:35 eureka kernel: (ada2:ahcich4:0:0:0): READ_FPDMA_QUEUED. ACB: 60 08 f8 83 c7 40 08 00 00 00 00 00
Dec 18 22:56:35 eureka kernel: (ada2:ahcich4:0:0:0): CAM status: Command timeout
Dec 18 22:56:35 eureka kernel: (ada2:ahcich4:0:0:0): Retrying command
That repeated a few times, but stopped after about 3 minutes, apparently with no failures as a result. But clearly that SSD needs replacing.
|
Larissa approaches Mona
|
Topic: animals, opinion | Link here |
Mona still hasn't made friends with the dogs, but things are looking up. Today Larissa approached her and wanted to sniff. Mona arched herself, puffed up her tail, growled and hissed, but she didn't go away. And Lara didn't go away either, at least not at first.
Not exactly friendship, but at least they're coming closer. I should leave a camera nearby.
| Friday, 20 December 2024 | Dereel | Images for 20 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Relearning basic mathematics
|
Topic: photography, general, opinion | Link here |
Years ago I answered a question on Quora: In what two positions will a converging lens of focal length 7.5cm form images of an object on a screen 40cm from the object? Textbook answers are 10cm and 30cm from object. And today I got my first upvote.
Oh. Looking at the answer, it was Just Plain Wrong. I don't know why, but it clearly needs fixing. But how? How do you calculate this thing? Simple algebra, of course, but I'm so rusty that it took me some time. And I was surprised to find that the correct answer is the solution to a quadratic equation, though I suppose the fact that there are two solutions should have pointed me to it.
|
Bushfire?
|
Topic: general, photography, opinion | Link here |
In the late morning I received a message on my phone from Vic Emergency: small bushfire west of Smythesdale, I think. It must have been small: they removed it from their web site after only an hour or two, while normally they clutter it up with reports of expired bushfires that are weeks or even months old.
But then the sun went yellow. I've seen that before, from bushfires. But this bushfire was so small that VicEmergency had already forgotten about it.
Ah, not that one, says Yvonne. It's the one in the Grampians, 80 km away. And yes, that's a big fire. Well, according to VicEmergency, a pair, one covering 28,157.00 ha (281.5700 km²), one small, 30 m apart:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
VicEmergency is too polite to give coordinates, but that's what you get clicking on the bottom left of this group of five (or six?) fires, all the same:
|
|||||||||||||||||||||||
That's not the only time I've seen such pairs, almost invariably 30 m apart from each other. There's another one near Creswick, north of Ballarat and close to where Chris Bahlo and family intend to build a house:
|
|||||||||||||||||||||||
How can I get the colour of the smoke on a photo? Out with my Olympus M.Zuiko Digital ED 8 mm f/1.8 Fisheye PRO to see what I could get, and in fact it wasn't that bad:
|
|||||||||||||||||||||||
The lighter area on the left is normal sky, so it seems that the colour has been rendered better than I expected. I just wasn't expecting the oval sun.
For the fun of it, I also tried the full spectrum Olympus E-PM1 to see what difference that might make, but it looks as if I left it too long. Before processing, it looked about as boring as most such photos:
|
|||||||||||||||||||||||
| Saturday, 21 December 2024 | Dereel | Images for 21 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Morning parrots
|
Topic: animals, general | Link here |
Up this morning and off to look for my phone, and saw this just outside the lounge room window:
|
|||||||||||||||||||||||
They even stayed long enough for me to get my camera.
They (crimson rosellas) are not the only birds in the area. Apart from the New Holland honeyeaters and wattle birds, we have a couple of superb fairy wren pairs, and they're particularly evident. Today I had one about 1.5 m away from me on the windowsill, though they don't stay long enough for photos.
|
Breakfast from leftovers
|
Topic: food and drink, opinion | Link here |
We've accumulated a lot of leftover ingredients recently. What should I do with them? My standard dish is nasi goreng, but you should be able to do that with noodles too, so that's what I did today.
One of the ingredients wasn't exactly leftovers: it was the outer leaves of Brussels sprouts, which Yvonne normally discards. To my surprise they went a particularly bright green:
|
|||||||||||||||||||||||
Also surprisingly, they were tough. Most brassica leaves don't need much cooking, and that's how I cooked these ones, but they're are an exception. Next time they'll need significantly longer cooking.
| Sunday, 22 December 2024 | Dereel | Images for 22 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Letting Bruno free
|
Topic: animals, general | Link here |
Let Bruno out again today, a number of times. The first time round was a bit of a surprise for him:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
He didn't seem to worry too much, and in the afternoon he stayed out for the longest time so far, well over an hour. It seems that he stays round the garden, so there's not too much danger of him going onto the road—yet. I'm still concerned, but it's clear that he's much happier now, and he spent much of the evening on my lap, something that he hasn't done for a long time.
|
Garden flowers in early summer
|
Topic: gardening | Link here |
It's the summer equinox, time for the monthly garden photos.
It's been a dry spring, and I have my concerns about the irrigation system. Even the Carpobrotus looks yellowish:
|
|||||||||||||||||||||||
After Nick Macdonald of macGARDENS came with his mate to tidy up the garden two months ago, a number of plants looked worse than they did. The damage to Clematis “Edo Murasaki” was definitely caused my Nick or his mate. Here before and after their visit, then one month ago and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Last month I wrote “maybe it will flower by autumn”, but it could be sooner than that.
Then there's this Leucadendron:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It doesn't look that bad. I'll keep an eye on it.
And the wild grass with the seed spikes that got destroyed last year has recovered. It seems that this could be Phormium tenax. Here September 2023, last month and today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
The Marriotts may be gone from next door, and Diane's Callistemon has stopped flowering, but now we have another plant:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
It seems that this could be a Veronica speciosa, which once we would have called a Hebe. It seems to be doing much better than our own.
The Strelitzia nicolai is flowering again:
|
|||||||||||||||||||||||
And to my surprise, a Strelitzia reginae has recovered from being mown down completely:
|
|||||||||||||||||||||||
It still doesn't look happy, but maybe it will flower after all.
The Camellia japonica that we transplanted last year still doesn't look good, but it's a lot better than it was. Here last year and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
In particular, the remaining leaves are green, not yellow, and I think there are more of them. Hopefully it will flower better next spring.
As expected, the Buddleja x weyeriana are growing well:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
Maybe because of that, the wild bush that Yvonne planted last year now seems to be recovering. Here in April and now:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
My guess is that it really doesn't like the sun, and now that the Buddleja x weyeriana
Apart from that, our trees mainly seem to grow at a snail's pace, if at all. It's hard to believe that this Ginkgo biloba is coming on 10 years old:
|
|||||||||||||||||||||||
But at least it isn't dead. About the only surprise was our Grevillea robusta, only 5 years old, which has now decided to grow after all.
|
|||||||||||||||||||||||
And other trees nearby, to the west of the house, are also looking better than they have been:
|
|||||||||||||||||||||||
|
Still more X problems on teevee
|
Topic: technology | Link here |
To the lounge room in late afternoon, turned on the TV and started looking at something. And the display on teevee froze. What caused that? Mouse or keyboard issue? The machine was still running, since it was playing music from Radio Swiss Classic. But then that stopped too. To the office, where I found that teevee was still running normally, but the xterm with the X display was not reacting. Shot it down and restarted it, and all worked normally.
That's the fourth time now. This time I caught it in the act. What did the log file say? Nothing of use:
[2281147.654] (--) NVIDIA(GPU-0): HISENSE (DFP-1): connected
[2281147.654] (--) NVIDIA(GPU-0): HISENSE (DFP-1): Internal TMDS
[2281147.654] (--) NVIDIA(GPU-0): HISENSE (DFP-1): 600.0 MHz maximum pixel clock
[2281147.654] (--) NVIDIA(GPU-0):
That appears to happen when I turn on the TV. There are multiple messages like that, all repeated several times.
[2281865.117] (EE) event5 - Telink 2.4G Mouse, class 0/0, rev 1.10/1.00, addr 1: client bug: event processing lagging behind by 23ms, your system is too slow
[2281865.117] (II) event5 - Telink 2.4G Mouse, class 0/0, rev 1.10/1.00, addr 1: SYN_DROPPED event - some input events have been lost.
[2281865.117] (II) event4 - Logitech USB Receiver, class 0/0, rev 2.00/29.00, addr 2: SYN_DROPPED event - some input events have been lost.
That makes sense. It would explain the lack of response. But continuing,
[2281865.161] (II) UnloadModule: "libinput"
[2281866.191] (II) NVIDIA(GPU-0): Deleting GPU-0
[2281866.528] (II) Server terminated successfully (0). Closing log file.
That last message is clearly when I shot down the server. But it's only 0.411 seconds after the mouse and keyboard messages, and 618 seconds after the TV turn-on messages. So the mouse and keyboard messages are clearly part of the shutdown sequence. Between turning on the TV and that there are no messages at all.
So what is it? It's most likely to be a hardware issue. Next time it happens (and I fear there's no doubt that there will be a next time) I'll try to find out what the server is doing, in particular what it's waiting on.
| Monday, 23 December 2024 | Dereel | Images for 23 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Unexpected domestic animals
|
Topic: animals, photography, general, opinion | Link here |
What will we do for a Christmas photo this year? Over the last 12 years we have taken varied photos of ourselves and as many as we could muster, and we're running out of motives.
OK, ten years ago the Stones Road house was still under construction, but we managed this photo:
|
|||||||||||||||||||||||
That's (now) the north wall to Yvonne's bedroom. How about a repeat, 10 years later? The chairs are still there, piled up. And when I took them apart, I found this:
|
|||||||||||||||||||||||
It was still alive, but it stayed around long enough to allow me to take a photo of it. What is it? Is it a whistling tree frog? The description suggests that it's darker on the back, and that's not the case here. Maybe a Litoria ewingii?
|
Protecting the Grevillea bedggoodiana
|
Topic: gardening, general, animals | Link here |
While walking the dogs, a car came past and asked whether the dogs were Salukis. No, of course, but we got talking: Les and Paula, who live in Enfield, just round the corner from Incolls Road, where Zali and Steve O'Dea, whom we visited ten years ago. And somehow we ended up talking about Grevillea bedggoodiana: they're arranging for protection status, since the recent burnoffs have ignored the presence of the plants, which have protected status. He also tells me that there are some rare orchids currently flowering down Incolls Road; I should go and take a look.
|
Google Maps fail
|
Topic: technology, general, opinion | Link here |
Yvonne out to harrow the paddocks today. Out of interest I checked her location with Google Maps. But before I could zoom in close enough, she was finished and came back in the driveway.
But she had disappeared from the map! And she didn't return. In to check and discovered that her phone had decided to turn off accurate reporting, or some such status. In any case, the compass wheel on the display, normally blue, had gone red. Click on it. Do you want accurate reporting? Of course, fool. OK, done. And the position on the map returned immediately.
How do these things happen? Yvonne was driving at the time, and she barely uses the app at the best time. Couldn't have been finger trouble.
| Tuesday, 24 December 2024 | Dereel | Images for 24 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
New Year's photo, finally?
|
Topic: general, photography, history, opinion | Link here |
Preparing for this year's New Year's letter brings a problem: what photo motive should we use? Yesterday we decided to do a repeat, and today we managed it with relatively little difficulty. Here ten years ago, and then today:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
I was expecting for us to look much older and more decrepit, but to my surprise, today's photo looks better. We should consider repeating the photos of 10 years ago as time goes on.
|
Christmas: no ducks!
|
Topic: food and drink, general, opinion | Link here |
Yvonne off to town a day early for the weekly shopping. Somehow Christmas is meaning less and less every year, but at least we can eat something special. Goose? Out of the question in Australia, but surely we can find a duck, which would be a better choice from the size point of view. But no, not even a duck! How boring Australian food has become! How meaningless Christmas has become!
| Wednesday, 25 December 2024 | Dereel | Images for 25 December 2024 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
|
Damaged gate
|
Topic: general | Link here |
How did this gate get damaged like this?
|
|||||||||||||||||||||||
That's quite a solid gate, and it's in the middle of the paddocks, where nothing much can get at it. Yvonne thinks kangaroos, but I don't think that even a horse could have done that damage. She was harrowing two days ago, and she could have done the damage with the car or the harrow, but she would have noticed that, and it would be unlikely that she would have hit it exactly in the middle.
What do we do? Reposition the chain that keeps it shut; currently it's too tight.
|
More VicEmergency nonsense
|
Topic: rant, general, opinion | Link here |
The next couple of days will pose an extreme fire danger, so we're keeping our eyes on emergency.vic. Oh, no, that's just the URL. The real name is Vic Emergency. And because of the danger, they're even more obfuscated than usual. Today I saw:
|
|||||||||||||||||||||||
That's us in the middle. Why is the map orange? I haven't found anything on this appalling app that tells me what the colours mean, but in the Grampians it represents the most dangerous parts of the 5 or 6 bushfires, totalling about 440 km², that are still not under control. Even the normal 2 bushfires 30 m apart are only on a greenish-yellow background:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
But look at the bottom of my display. Eight incidents!
|
|||||||||||||||||||||||
OK, list them.
Would you settle for 4?
|
|||||||||||||||||||||||
They're not incidents at all, just warnings and prohibitions. I'm in the Central Fire Weather District, so only two of them even apply to me!
OK, try again. Now we're down to one incident!
|
|||||||||||||||||||||||
I tried selecting it, but ended up in a loop. People, this is a matter of life and death. The reporting is broken, and has been for at least 14 years. In that time, the biggest changes have been gratuitous changes to the way information has been reported, including moving the web site I still can't get a report on things that endanger me (or just about anybody else in my area). I still get house fires and six-month-old chicken flu warnings.
Only one possible thing might explain some of the problems. Why am I getting “incidents” for the South-Western fire district? My 20 km radius “watch zone” barely cuts into the South-Western district:
|
|
|||||||||
Could that be the reason? Tried to reduce it to 19 km, but no, the app is too clever for that and only goes in 5 km decrements. And when I tried it, all such warnings had gone away, though in the meantime I had also received some for the North Central district, suggesting that it was only a coincidence.
|
Bah! Humbug!
|
Topic: food and drink, general, history, opinion | Link here |
Yet another Christmas Day with almost nothing to show for itself. We didn't even cook anything special, just leftover Wiener Schnitzel from Sunday. But then, it seems that there's a certain tradition there too: we also ate leftovers in 1965 and 2019.
Somehow it seems that festivals like Christmas belong to a time when basic items like food cost more of a family budget than they do today.
| Thursday, 26 December 2024 | Dereel | Images for 26 December 2024 |
| Top of page | ||
| previous day |
|
Bushfires: good news, bad news
|
Topic: general | Link here |
Today was slated to be one of the worst fire danger days in years: high temperatures, strong winds. Yes, the temperatures were high—we reached 36.5°—but they weren't as high as forecast. Normally we measure temperatures round 3° higher than the forecast. Yesterday they forecast 34°, and we got 37.4°. But today's 36.5° were 0.5° lower than forecast. And the wind wasn't that bad.
The bad news: the fire in the Grampians got worse, as forecast. Now there are 740 km² under fire, more than double what it was a couple of days ago. As the media put it, that's more than the size of Singapore (726 km²). And it's still nothing like under control.
|
Roast pork again
|
Topic: food and drink, opinion | Link here |
Chris Bahlo, now a rare guest, along for Christmas dinner this evening. Roast pork, something that I have never had completely mastered. Looking back, it seems that Australian pork hasn't either: the recipe there now diverges significantly from what I found there three years ago, coincidentally the last time Chris ate here. And in the meantime I had used the “hair dryer” “air fryer”, but reading my diary it seemed that that wasn't the optimum.
And to my immense surprise I found a useful recipe on the (ALDI) packaging:
|
|||||||||||||||||||||||
Yes, whoever wrote the heading couldn't count, but the instructions looked detailed enough.
OK, salt, oil and put in the oven at 220°.
|
|||||||||||||||||||||||
Cook for 26 minutes until it starts crackling, then turn the temperature down to 180°. My roast weighed 1.576 kg, so from this point it should have taken between 47 and 63 minutes to get up to “between 70 and 75°” meat temperature. In fact it took 73 minutes to reach 72°:
|
|||||||||||||||||||||||
And how was it? Parts of it were excellent. The issue is that the heat was mainly from above, while the crackling was wrapped round 270°.
|
Silly dinner photos: out of practice
|
Topic: photography, opinion | Link here |
In the past we have always taken silly photos at dinners, like this one taken at our last roast pork dinner:
|
|||||||||||||||||||||||
But it's been a while, and I'm out of practice. In the past there were more people and I did a panorama. This time I used the M.Zuiko Digital ED 12-100 mm f/4.0 IS PRO at 12 mm, and it wasn't really wide enough:
|
|||||||||||||||||||||||
I don't think it would have helped much to be wider: the angle of view is too wide for a rectilinear projection. I think that cylindrical would have been better.
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.
| Top of page | Previous month | Greg's home page | Today's diary entry | Greg's photos | Copyright information |
