DUMP YOUR RASPI NODES. Get a used mini PC or used laptop for cheap!
More powerful, only $45-75 on eBay (plus ~$65 2TB ssd, but you'd need that anyway for your Raspi node).
This HP EliteDesk 705 G4 is **overkill** for a node. 5 yrs old. Costs $75*.
Fan noise? Meh. Power draw? CPU's max draw is 35W.
Once it's done syncing, the video can't even pick up the fan running at idling speeds:
Cost to run:
15W * 24hrs / 1000W/kW * 12¢/kWh * 30 days = $1.30/mo
This cpu is 8.5x more powerful than a Raspi4.
A more pleb-friendly option is a used laptop. Has everything you need built-in (esp the BATTERY!!! Huge advantage vs Raspi). This HP ProBook 450 G3 is running an 8yr old i5-6200U (3.5x more powerful than a Raspi4). Also super-quiet and low power (15W max cpu draw).
Motion blur would just create some ugly synthetic frames. You just inherently can't stuff 24 non-divisible things in 25 boxes, for example.
I'd love to see a test where all the source framerates are the same AND the final output framerate matches. Just curious to see if there are any playback differences in terms of smoothness and consistency when there are no inherent framerate conflicts.
Okay, so the day's productivity turned around!
I had a messy branch of experimental nostr:npub17tyke9lkgxd98ruyeul6wt3pj3s9uxzgp9hxu5tsenjmweue6sqq4y3mgl changes that needed to be separated into 3-4 different PRs.
* Managed to isolate each major chunk of changes and create two new PRs + one still in-progress branch.
* Created a bonus new PR that was just some quick cleanup in a totally unrelated area.
* Pulled off my first `git rebase -i` with a bunch of `drop`s to cut out some earlier changes that went into the other PRs while preserving the new stuff in the most recent commit (I am NOT a git ninja so this was a bit terrifying; really shocked that it worked so well).
I'm going to chalk up today mostly to residual 'rona. About 4hrs of my afternoon was either napping or only being half awake while scrolling on my phone. Just kind of worn out.
Ha, actually working on isolating some messy code changes I made into an acceptable nostr:npub17tyke9lkgxd98ruyeul6wt3pj3s9uxzgp9hxu5tsenjmweue6sqq4y3mgl PR so there's hope for today yet!
A week in Atlanta for TABConf straight to a week in Nashville made number go up.
🫂 I appreciate the perspective!
At least nostr:npub1kumadawgefxjt2c293whsan5sfmdg5tylm0uk39enc6uxxfy5m4scrz2v7 got her usual two walks and napped with me a lot so her day was pretty solid!
Can't tell if it's residual covid hangover or just being a lazy af pos, but I accomplished absolutely nothing today.
Last year my rusty Spanish that I learned in high school was almost barely conversational. Way better than I was expecting! That was really encouraging. But haven't directly spoken with anyone in Spanish since so I have no clue what to expect this time.
Most of the last year on Duolingo has been review, but it has reminded me of a lot of things I'd forgotten. How much that helps...?
I'm actually giving a talk in Spanish in Nov so potential harsh reality coming my way!
Mostly, yes, but we'll see how I fare in El Salvador in Nov compared to last year.
Duolingo obv aims for practice and repitition instead of direct instruction (e.g. conjugate -ir verbs in the past tense by doing blah).
My engineer's brain would prefer to know those rules (to the extent that they exist and aren't a mess of exceptions), so it feels like a bit of a learning style mismatch. But I believe them when they say they've leveraged the experience of teaching millions of people. The way I think I want to learn might not be the best way for me to learn.
The "you" = "ustedes" or "tú" ambiguity feels unfair to get this marked wrong. Though I guess my "a" is wrong, too?
Whoa, just discovered the add-your-own reaction option in Amethyst! Nice job!
More nostr:npub12vkcxr0luzwp8e673v29eqjhrr7p9vqq8asav85swaepclllj09sylpugg feedback: My heart/like clicks sometimes take a few moments to fill in the outline to red. I assume there's some broadcast/relay comms delay going on in the background.
Maybe show a "pending" colored state (or circling/loading animation around the heart?) while awaiting a response from relays? I just want some kind of feedback to know that it's in progress and not just my browser tab going totally unresponsive.
Animation on success would be fun, too!
A quick video transcoding test. Let me know which you can/can’t see and which looks better. 🐶🐾🫡 (for geeks, it’s h264, h265, av1 in order)
https://v.nostr.build/OqDG.mp4
H265 for the way it smooths its blocks and banding (esp the clocktower shot w/the moon moving behind it). Fast areas of the waterfall shots also highlight H264's sharper blockiness making for an uglier presentation whereas H265's smoothening calls less attention to itself in high motion, hard to encode areas.
But all three have some really bad judder(?), esp the very last clip. Watching the water is painful. 25fps mismatch w/source to blame? Are all the sources 24fps? Mix of 24 and 29.97fps?
My M1 Pro should be able to do smooth decoding, though the browser embed might mess with that.
tfw "Stephan Livera" likes your tweet (yeah, right. Fuck off, bots!!!)...
...but it turns out to actually be nostr:npub1r8l06leee9kjlam0slmky7h8j9zme9ca32erypgqtyu6t2gnhshs3jx5dk!
Eh, badge of honor at this point.
Email is a failed / dead-end communication medium for me for the most part, anyway.
Just like voicemail. I switched cell carriers and ported my phone number over, but somehow the new carrier could never get my voicemail set up. Meh.
Probably been a decade with no working voicemail. It's a feature now, not a bug.
I'm not up to date on the various note kinds, but I'm mostly:
* NACK: nostr notes don't need editing, period. And it does screw up the context for, potentially, the entire reply / quote ref interactions.
* or meh, even with editing, the original versions should always be available (when possible) in the UI. And / Or always have to assume that some relay somewhere still has the original version around, anyway. Or some clients refuse to show the updated version.
So the worst case scenario can't be resolved (no such thing as truly erasing a mistake) while the more innocuous cases (just a little cleanup) aren't enough of a win to justify the effort / hassle / UI muddle of a somewhat clunky solution that maybe / probably won't get broad adoption anyway.
