beastlykings

I've been toying with Linux on and off for almost 20 years now.

Started with damnsmalllinux on some ancient 600mhz Thinkpads. Dual booted Ubuntu for a long time, back when 3d desktop cubes were all the rage, so I'm used to gnome, synaptic and apt.

Tried to stick with it, but never could get away from Windows entirely. Especially for gaming, and a few critical apps. Eventually I kind of drifted away, and went full Windows for years. I always keep an Ubuntu LTS thumb drive around, and would use it occasionally for various reasons, testing etc etc.

Recently I installed Ubuntu 24.04, and had tons of stability issues. Mostly involving video output and the GUI. Screen would jitter left and right a few pixels. And sometimes maximized windows would be transparent to clicks, so you'd be clicking random stuff below the window. This was especially bad with Firefox and VLC, separately. I also had issues with removable drives not mounting properly. Standard stuff, I wasn't doing anything weird. Practically a fresh install.

So I tried Mint, cinnamon. And so far I really like it! I've not been running it daily, but just the same tinkering. And so far no issues at all. But that got me thinking, what else am I missing?

I'm comfortable in the command line, but not proficient, I appreciate a good GUI for most things.

I plan to do some gaming, so steam proton compatibility is important. I don't think that's hard to achieve, but I wanted to make sure, it's important to me.

Last time I played with KDE was a decade ago, I hear there's lots of new developments going on there? In plasma? Unless plasma is different now, IDK I haven't looked extremely hard.

I don't care much about customization, I don't want arch. I want something that is a pretty solid base, with decent features, and good support for when this go sideways. I feel like that's not Ubuntu anymore. Especially with them pushing into Wayland and flat packs.

I guess my question is, does Mint seem like a good distro to start with? Or am I not looking hard enough?

Thanks!

My CPU is an AMD 5600X. My test video is 1080p 30fps, I'm trying to bring it down to 480p using AV1.

This is my first time playing with AV1. I bought an A310 to play around with because I read that the hardware encoder was faster than software, albeit lower quality and slightly larger.

Quality isn't important to me, I have 700gb of 480i video that was saved at 1080p and inefficiently encoded, I want to reduce its size dramatically.

I'm using handbrake on Windows, and I chose AV1 SVT at first, and I average about 150fps, sometimes 120 sometimes 180. My CPU sits at 100%.

But if I choose AV1 Intel QSV, I average only about 40fps. And the GPU sits at 68%.

What am I missing? Thanks in advance

Edit: I found a thread from a year ago saying that encoding performance dropped after driver 4887, so I rolled way way back to 4885 from October 2023, and my performance almost doubled to 70-80 fps. But this is still far worse than SVT on CPU alone.

I read about a bug where the whole card can only be utilized if running two jobs simultaneously, so I tried this. The second job runs much slower, about 10-20fps. But that does bring me closer to 90-100fps combined, sometimes 🤷‍♂️

Something has got to be wrong, or maybe I'm expecting too much performance for the job I'm doing? I don't have any special filters set up. You'd think encoding 1080 to 480 would be lightning fast.

Some background: I love flashlights. Been in the hobby for years. I have an original group buy version of "Giggles" the GigaThrower, and upgraded it with the new emitter and driver when the GT70 came out.

I've got a group buy LT1 and C01 as well.

I've EDCed a D4 for as long as I can remember, added Lexel custom aux emitters before they came standard. Wore out my first D4, had to replace the switch. Upgraded to a D4v2, used it for years. I forgot it on a job site a few weeks ago and it got stolen. So I have a D4v2 dual channel in the mail.

I'm pretty sure I was the first to mod a D18 with aux emitters, there's proof on Reddit somewhere. Definitely modded my fair share of headlamps with custom Lexel drivers and emitters, upgraded springs, etc.

All this to say: I've been around. But I definitely haven't been active in a long time, and even when I was I definitely focused on stuff that runs anduril. So I don't know what's out there these days.

I love the idea of a big fat heavy flashlight. I have a maglite, and I keep it by my bed. I know it's silly, but I like the idea of having a short club in hand if I hear a bump in the night..

I put an LED conversion bulb in it. But what I want is something like an xhp50.3, and anduril. I could modify it myself, 3d printed trays for 18650s and such, but I worry it won't have the weight I want.

Giggles is too big and expensive for the task. I want something maglite shaped and weighted, running anduril, but not ridiculously expensive. Doesn't even have to be that crazy bright either. Just big lol.

Any suggestions?

I love the steam deck, and recently, so does my wife! Don't get me wrong, I couldn't be happier. And she would find something to play on her switch if I asked for the deck back. But I don't want to stifle this new found love for the deck.

So I decided to get creative. What you're seeing is horizon forbidden West on maximum settings, streaming from my desktop to an over 8 year old Chromebook, with an Xbox controller.

It's honestly pretty good, considering. Every minute or two I get some hitching and stuttering, something to do with the Wi-Fi. Also there's a weird green bar at the bottom I can't seem to remove 🤔 But otherwise it's smooth and clear. Good enough to get me through until we can justify a second Deck 😬

I'm surprised it works at all! But it's steams built in streaming, when the Chromebook steam app 🤷‍♂️

Edit: in case anyone comes by later and wants to know or could find this useful: the Chromebook is an Acer R13 from 2016, mediatek CPU. My wifi is a 1st gen Google WiFi puck. And the stuttering started turning into zero audio at times.

I bought a random $15 gigabit Ethernet USB C adapter with pass through charging, ran cat 6 tm under my couch to where I sit, and now things run 99.9% smooth. Very pleased. This will hold me over until I can justify another steam deck.

Edit 2: I fixed the green bar by realizing I was an idiot. I never looked at my PC when it was streaming, and it was using my crappy secondary display at 1400x900. This Chromebook is 1920x1080, so it was doing weird things. Moved it to my main display at 1080, and now it looks even better, and no more green bar.

Cat tax: That's not a small pineapple, he's just a big cat.

I'm cheap, and I also have barely any time for breakfast in the morning, and my wife likes it when I make her breakfast but she leaves for work an hour after me.

So this is what I do, and have done for almost three years straight now.

You get yourself some Some And a big bag of shredded cheese, your choice

First thing I do when I walk into the kitchen is start the toaster oven, getting it hot. Then I take a chicken patty out of the freezer and break it in half on the edge of the counter while it's still in the bag. Then I take a half sheet paper towel, and fold it in half, because I hate doing dishes. I put both halves of the chicken patty on it, pop it in the microwave for one minute, 30 seconds per half if I'm only doing one. Then while that's going, I slap two tortillas on the counter, sprinkle a healthy dose of cheese on them and spread it out evenly. By the time I'm done, so is the chicken, so I put each half on one side of each tortilla. Next comes the flavor. You can sprinkle a little garlic salt and pepper, or a dash of worcestershire, or my favorite was a dab of Chick-fil-A sauce under the patty. Then, slide it onto the rack in the piping hot toaster oven. Then I walk away to go start getting ready for work, just a simple task like finding socks or something, then I come back a minute or two later and the cheese is nice and bubbly, the tortilla is browning on the edges, it's just about ready to pull out. Then I pull them out, fold them in half, put mine on my water bottle to cool, and hers goes back into the toaster oven, but it's a fancy oven so I set the temp to 160f so it's nice and hot when she gets up, but doesn't keep cooking too much.

The whole process takes less than 10 minutes, maybe even 5 minutes if I'm really on my game in the morning.

The whole thing costs like 50 cents, and is plenty filling for me. It's probably not the healthiest option, but.. 🤷‍♂️

Why don't I use something more breakfasty, like sausage? Because I can't find it as cheap as the chicken. Funny enough, I actually started this whole process during COVID, with frozen precooked sausage patties. We got a bag of them with one of our low income commodity boxes, and couldn't figure out what to do with them. So I started doing this. Then when the bag ran dry, I transitioned to chicken. Not as good, but still good, and like I said, I'm cheap lol.

I'm going to do some a/b testing to compare to the rubber duck antennas I've been using. I don't expect much more, but perhaps a little bit more range?

Going from the little stubby coil of wire included with the heltec v3s, this should be amazing.

Edit: I tested it and it works just fine, or at least within the margin of error of my test. It wasn't significantly better or worse than the Amazon bought high gain rubber duck antennas, a little disappointing but 🤷‍♂️

As has been said before, height is might. Line of sight counts for more than increased power.

I'll still probably use them on some permanent nodes 🤷‍♂️

Just putting this here in case it's useful to someone else.

I'm still working on setting up my local nodes and meshing to my buddies houses. Part of my testing involved going to the top of the tallest hill in my town, (1300 feet above ground level, the rest of the ground is relatively flat around here).

The problem was that I couldn't connect to my home node, around 2.5 miles away, good line of sight. Testing showed that my home node received all my messages, but the return confirmations never made it back.

The problem? The tallest hill in town is also home to two fully loaded cell towers, blasting away, presumably deafening my node, overloading the front end with strong out of band signals. 900mhz GSM perhaps?

At any rate, I looked into band pass filters. I'm a ham and I've dealt with poorly filtered front ends by the likes of baofeng, so I know filtering can do a lot, and I wasn't sure how much a standard node came with.

The answer? Some, not much. Enough for most use cases. Oddly, most information I found on the subject, specific to Lora, advocated against using filters, saying they are usually unnecessary, etc etc.

While that's probably true most of the time, it's definitely not true all of the time. I'd be interested to see some a/b comparisons of 20-40 foot high nodes in urban environments, both with and without filters.

From experience I know that a baofeng 144mhz radio (known to have poor filtering) with a 1/4 wave vertical antenna up 40 feet, was mostly deaf to any distant signals, and actually performed better in some instances by just using the stock antenna and standing on the ground. Likewise when using the 40 foot tall antenna and adding a filter, the reception was massively improved.

Add to that experience my most recent test. I added a 915mhz band pass filter to my node and brought it on top of the hill, next to the cell towers, and was able to make full duplex communication with my home node.

I'll be doing more testing, some a/b testing with the filter on my home node to see if it improves my range tests.

Filters probably aren't for everyone. And they aren't free performance gains, you can't forget insertion loss. But don't be afraid to buy one and try it if you think you might be having desense issues.

It was supposed to rain, but we pressed on, and it ended up being beautiful. Chilly but pleasant. Just cold enough to make you appreciate the warm fire. Plus the sound of the river to lull you to sleep.

Hard to beat!

Hear me out. I've been thinking about the best way to put up solar nodes in my area.

I live in Michigan so the winters are long and usually cold, and can get quite cold on occasion. Less so these days, but still. We all know that lithium doesn't like to be charged below 32f, so that's a problem.

I have one idea for a remote node on my property, to just use a 6v sealed lead acid with a 6v solar trickle charger, maybe a diode in series to the node if the voltage from the panel threatens to go over 7 or 8 volts. Or some kind of shunt, idk that idea isn't fully baked.

I'm also looking into a thermostatically controlled resistive heater. Bump up the capacity of the battery and make sure the panel is big enough to run the heater non-stop if need be. But that idea is also still baking, parts are in the mail and prototyping is yet to be done.

But then I had another idea. What about super capacitors? You can get a 5.5v 10F super capacitor on Amazon for $6. Some chatgpt math (and a proper understanding of the difference between power and energy, or rather ma vs mah) tells me that the storage 10F at 5 ish volts is equal to a nominal lipo at 3.7v and 10mah. Ignoring the fact that the capacitors voltage would drop sooner, and thus you'd lose some energy on the bottom end...

The T114 v2 has a built in solar charger circuit, a standby current of 9ma, and a TX draw of 150ma. So with a super capacitor you could get roughly an hour of idle time, or 4 solid ish minutes of nonstop transmitting. That's more than enough to account for a big cloud passing by, especially if you beefed up the solar panel, or easier yet just doubled or tripled the capacitors, they are pretty small.

The charging circuit might not like the low internal resistance of an empty capacitor every sunrise, but a couple ohm resistor in series would probably solve that.

Yes it's not ideal to have your nodes turning off every night, especially in the winter when days are short to begin with. But could it technically work? I feel like it could technically work. It'd be great in the summer,

My only concern would be the node getting stuck in a weird state if the sky is cloudy and the CPU browns out. It'd take a whole day to power cycle. Probably put a megohm resistor across the capacitor to ensure that it drains fully overnight in that scenario.

Are there any concerns with constantly hard power cycling a node like that? Data corruption?

The lead acid is probably the safest solution, though heavy. And the heater is probably the most compact solution, though more complicated and prone to catastrophic failure. But maybe there is room for super capacitors?

Sorry for the wall of text. Just spit balling.

Edit: HOLD THE PHONE! You can get 500F caps for $7! The reviews say they're actually more like 300F, but even that could idle a node for well over 24 hours, and TX nonstop for 2 hours, which is unlikely to happen. This is a game changer. This could solve the winter lithium problem.. I'm going to try this and I'll report back.

Edit 2: Various super caps and LiCs still on order. But I just came across this in the discord:

"Dendritic degradation happens when charging below freezing. However it’s a slow process. And the slower the charge, the longer the lifetime. With IoT devices the charge rate is quite low and not as critical compared to higher draw devices. Up here winters reach -40 sometimes, and in a lot of cases it sits below freezing for weeks, even months at a time. We’re now in our second winter on the same batteries with no failures (yet). Regular lithium Ion and LiPo. I’ve only done LTO for high traffic nodes that are very hard to access. Otherwise just plan to replace them when they die.

Slower the charge rate / the higher the capacity / the newer the battery / the better the quality the battery, all compound into longer cold weather performance.

There’s other factors too, like the higher the charge rate usually means the more sun/solar activity, which also means if you have a properly setup enclosure, the sun hitting it can be enough to increase the temp by over +20c or even more. This also doesn’t account for the charge/discharge heat being released by the batteries themselves in the enclosure." -Cully@KBOXLabs

So it seems to me I'm overthinking this whole thing.

Maybe a super cap or LiC would be good for extreme longevity. But a decent pair of 18650s might be plenty for something you won't have to touch for a couple years 🤷‍♂️

I'll still do some testing and report back, but I might not spend the extra money making each node supercapped.

I'm still fighting with Amazon to get my first pair of devices delivered, I went with heltec v3s to get started.

But looking at the meshmap.net, it's very very sparse around me. How likely is it that the are nodes that aren't on the map?

Hopefully I can check it out myself soon!

The civet processed beans? I'm wondering if it's all hype? Or worth buying a bag to try?

Edit: Thanks for the responses everyone! I'm gonna pass on it 👍