Uncategorized

ASUS ROG Laptops for AI Development

Leave a comment

Say you want a laptop with decent power for local AI development. If you don’t want to spend a lot of money, and you want a portable, sleek laptop (not a clunky “portable desktop”), consider the ASUS ROG Zephyrus G14. You can get one of these from BestBuy, on a deal, for under $2000. It comes with 1TB of NVME – but for ~$200 you can upgrade this yourself to 4TB using a Samsung Pro NVME. I did the same and ended up with a 5070ti, 32GB of ram, all in a fairly sleek laptop.

Hardware out of the way here are some tips:

  • You can kick rocky 9.6
  • You should add in elrepo and upgrade the kernel using the kernel-ml repo. At the time of this article this pulled me to kernel version 6.16.4-1
  • Unfortunately X11 doesn’t work – use Wayland.
  • I’ve found a lot of python libraries are not yet built with cuda 13 support – so stick with cuda 12.9. Also it seems to matter which version your driver targets, so i found 575.64.05 work well – still supports the 5070 ti, but targets cuda 12.9
  • I was able to control the keyboard LEDs, screen brightness, and power profiles via KDE. I also tried the asusctl and supergfxctl – but ended up not needing these.
  • As for screen control, for some reason the screen got VERY dark on bootup – so i ended up writing a screenbrightness systemd script that would set screenbrightness to max on bootup. Once in KDE i could control this via KDE’s controls
  • I also found sddm works well, but could not get gdm to work – so stick with sddm/wayland
  • One downside to this platform – when not on wall power, the GPU halves the amount of power it is willing to source. In practice for AI inference tasks this only seems to make a small dent in performance.
  • Also on power modes: the Quiet mode is so pleasent – little to no fan noise, and the performance is stilll quite good. If you flip to Performance mode, the fans are so loud – you have to really need the extra performance.
  • Lastly, sddm seemed to crash on boot – so i wrote a timer based systemd script to restart it 7 seconds after boot, and this fixed that

Here’s a kickstart for rocky 96: https://github.com/carsonfenimore/kickstart

  • Be sure to disable secure boot
Uncategorized

tts Text to Speech with Conversion

Leave a comment 
import torch
from TTS.api import TTS

# Get device
device = "cuda" if torch.cuda.is_available() else "cpu"

# List available 🐸TTS models
print(TTS().list_models())

# Init TTS
#tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)

tts = TTS("tts_models/de/thorsten/tacotron2-DDC")
tts.tts_with_vc_to_file(
    "We are seeing multiple sensors with bad values, please pay attention!",
    speaker_wav="/data/apu_america.wav",
    file_path="/data/myapu3.wav"
)
rust

Cargo build freezing…

Leave a comment

Occasionally cargo build freezes near the last step (26/27 for me) and i ctrl-c. Then running cargo build again I get this error:

“Blocking waiting for file lock on build directory”

Searching the internets people claim this can be fixable by clearing out the package-cache (“rm -rf ~/.cargo/.package-cache”), however this doesn’t work for me. What DOES work is cargo clean then rebuild. This takes WAY too long. A better way is to just pkill the running process (ps aux | grep cargo, or pkill cargo).

rust

rust cxx crate: linking to external c++ shared lib

Leave a comment

I have a custom c++ shared library. I also have a rust application that needs to use that library. I had no problem making a build.rs using bindgen, however I wanted a little better c++ support so I turned to the cxx crate. All went well until link time as I was getting “unresolved symbols”… but the linker output showed the library WAS being linked in! What gives?!

First, here was my source BEFORE the fix:

    // linking to c++ "libcamlite"
    println!("cargo:rustc-link-lib=dylib=camlite");
    // .... code removed for brevity
    cxx_build::bridge("src/main.rs")
        .file("cxx/wrap.cxx")
        .std("c++20")
        .debug(true)
        .compile("libcamlite-rs");

The solution? Behold!


    cxx_build::bridge("src/main.rs")
        .file("cxx/wrap.cxx")
        .std("c++20")
        .debug(true)
        .compile("libcamlite-rs");
    // .... code removed for brevity
    // linking to c++ "libcamlite"
    println!("cargo:rustc-link-lib=dylib=camlite");

YES… I simply moved the cxx::build command above the link directive…! Seems like a bug, eh? And I don’t have time to care much more about it – but hope this helps someone.

Uncategorized

Tesla Model Y After 4 Years

Leave a comment

About 4 years ago I bought an early edition of the Tesla Model Y. I immediately loved it. I still think it is the best car available.

When I purchased the car I also paid for Full Self Driving (FSD). It was understood that FDF wasn’t actually ready – but that it would mature over the coming months… or years…

Steady progress in FSD was made – but it seemed to stall for about a year and a half. Every new release would be proclaimed as “mind blowing” – but was only a small step change. This all got old. By summer 2024 I was pretty well resigned that there was no reasonable timeframe in which one would expect real progress.

This all seems to have changed in FSD 12.5.4 – for the first time, it seems smooth, doesn’t have the nag that just annoys. Doesn’t beep at me. It just drives. It’s really nice.

I hope the Elon haters just try to let the car drive you. Congrats to the Tesla team for sticking to it!

Uncategorized

Apple Watch Solo

Leave a comment

Apple makes some good products… I’m not so sure the Apple watch can be counted as one of them. Reasons: the battery life isn’t stellar, and, most importantly, it is an utter pain to set one up for family member.

The setup problem is because apple requires an iphone to set up the apple watch and, to travel the royal path, said iphone should be on a plan that supports the apple watch. Since I am cheap, I am not on one of those plans.

After much pain and wasted time I found the one true way to connect the watch to cellular while keeping a cheap plan on your phone. The process is superficially easy, but unnecessarily painful. Roughly it goes like this:

  1. Call verizon. Yes. Call. In the year 2023 – you MUST call them.
    • You must connect to their “inside sales” departemnt
  2. Tell the sales associate you want to connect an apple watch to cellular in standalone mode
  3. Given them the watch IMEI
  4. Ask them to activate the watch. Then activate the watch using your iphone. Do this with them on the phone
    • If they don’t activate now, you will try endlessly to get it to work in vain. You will call them the next day and have to pick right up where you left off.

Linux

Note to self: Even rocky 9.2 uses bridge-utils

Leave a comment

Sometimes I decrypt LUKS volumes via dracut/dropbear (ssh). To enable this I follow the instructions from the dracut-ssh project. However since I usually use my linux boxes as hypervisors, and vms need a bridge, I end up with bridged networking.

Now for the problem: if your dracut networking config does not align with your normal post-dracut networking, you end up with a weird blending of the two. So to avoid this I just keep dracut the same, building by bond/bridges in dracut to match the normal configuration.

Here’s a sample portion of my /etc/default/grub:

rd.neednet=1 ip=10.10.2.2::10.10.0.1:255.255.240.0:server-2:bridge0:none:8.8.8.8 bridge=bridge0:enp47s0

This all works well, so long as you are sure to install bridge-utils – otherwise dracut silently fails.

yum install -y bridge-utils

Be sure to rebuild your dracut

dracut --force --regenerate-all

Viola! Fin.

Uncategorized

Disneyworld sucks

Leave a comment

It’s not magical – no matter how many times they say it is. It is as magical as having a pickpocket empty your wallet everyday and giving you back only lost days of your life and a sunburn.

I hope someone starts a new theme park that is actually based on some new ideas and real entertainment, not hackneyed recycled flufff from yesteryear. Walt would be so ashamed.

Starlink

Review of Starlink Internet

Leave a comment

I have been using starlink for several months and have been thoroughly impressed! I switched to Starlink even though I live in an area with multiple high-speed internet options: Verizon FIOS, Comcast xfinity, etc… The “big boys” of internet. They all tout multi-hundred Mbit/sec internet downlink; in the case of fios symmetric uplink. Starlink, in the other hand, barely breaks 100mbit/sec down. So why bother with it?

Normally you don’t care about your ISP – you just use your internet and it’s great… but what about when it isn’t so great?

Enter exhibit A – fios performance against starlink. Fios basically went to pot around january. No manner of rebooting of routers would resolve it. Starlink didnt tank – just verizon.

Now enter Verizon’s consistently crappy service: i call verizon and get bounced between 4 or 5 different “agents.” Nobody could help – but they acknowledged the problem. They assured me they valued me as a customer. Then they started blaming my router… but in my case, they didn’t realize i have redundant routers running pfsense. I’m not a black belt network engineer but I am fairly capable of handling my home internet router. I tried in vain to explain to them that it wasn’t my router. This proved a waste of time so I asked them to cancel. Even getting them to cancel took forever! And to top it off they informed me that since I was cancelling at the beginning of a billing cycle i would still have to pay the full amount – no proration – “per the contract.” Well, so much for valuing me as a customer – they don’t care and never did, even up to the very bitter end.

Now enter starlink. Normally one has no bargaining chip if they have a single internet uplink. I have pfsense load balance my internet across starlink and fios. This meant I could cancel Verizon without any interruption of service. So satisfying.

Granted Starlink is NOT as fast as FIOS. Not even close. But the fact that “Starlink != Verizon” is good enough for me. I’m even willing to pay more, obviously; starlink is now $120/mo for me, and verizon was only $80/mo. That’s what sheer disdain does.

Oh and user experience with starlink is way better. Everything is managed out of the starlink app. Want to cancel? Upgrade? Just hit the button, for goodness sakes! And never are you required to interact with anyone to set it up, change, or shut it down. Take a hint Verizon – nobody wants your service reps.

Also starlink is improving: Around march the already-low latency dropped another 10ms. See in the image below around March.

Now for the bad… obviously uplink is weak – single digit Mbit/sec. Also during torrential rains I lose internet for a minute or two at a time.

Elon: I’m hoping starlink can add a cheaper tier – perhaps $60/mo? I’d even be good with $60/mo for 60 Mbit/sec downlink? But please don’t drop uplink speed. Its already too low.

Linux

Probably time to dump Redhat

Leave a comment

Late in 2020 Redhat made its first attempt to kill CentOS. Then came the heroic rescue of Alma and Rocky distros. Now just last month, June 2023, Redhat is attempting to kill these “downstream distros.” They – wrongly, I might add – assert “recently, we have determined that there isn’t value in having a downstream rebuilder.”

We could argue with redhat or just move on. There are other distros. But the tragedy here is that redhat is missing the entire point of open source. Everyone contributing anything that worked on CentOS was bolstering Redhat’s offering. It is arrogant for anyone, Redhat included, to view the community as a bunch of freeloaders. Opensource isn’t narrowly defined as sharing of code – it is a community of sharing. In that light Redhat is saying they are done sharing. So they are done with open source. So I am done with them. Time to move on.

The only question now is which distro will be used next. In the past I’ve avoided other distros because there was simply no compelling reason to switch. There is now a reason to switch. Perhaps Ubuntu or Debian? Arch?

Goodbye Redhat – have a good time sliding further into irrelevance.