AI Face Anonymizer Masks Human Identity in Images
We’re all pretty familiar with AI’s ability to create realistic-looking images of people that don’t exist, but here’s an unusual implementation of using that technology for a different purpose: masking …read more
#hacking #projects
https://hackaday.com/2024/11/13/ai-face-anonymizer-masks-human-identity-in-images/
Within 2 Weeks: We squashed our #GitHubActions spending from $4,900 (weekly) down to $890 ... Here's what we fixed in the (semi-chaotic) 2 Weeks for Apache #NuttX RTOS
We take the TinyEMU #RISCV Emulator for Ox64 BL808 SBC ... Tweak it for #Sophgo SG2000 SoC
Article: https://lupyuen.codeberg.page/articles/sg2000b.html#update-the-memory-map
Optimising the Continuous Integration for Apache #NuttX RTOS 🎉
@freemo They break ties and count electoral votes and other such minor duties. Other than that, not sure why people think VPs have any power to do anything. After a few decades it's hard to remember their names unless they end up as president one way or another.
I have to admit I'm so used to Asian women white women kind of look odd, but they're supposed to look that way.
How we upload the #NuttX Test Logs ... As GitHub Release Notes
Article: https://lupyuen.codeberg.page/articles/sg2000a.html#upload-the-test-log
Well, partial success - I did actually get heard through the space station and relayed back down to Earth.
The down side here is I didn't hear anything on this pass, possibly due to atmospheric conditions. I did hear the ISS beacon one pass back, so I can confirm send/receive through the kernel AX.25 stack does work, I just don't see to be able to it on a single orbital pass of the station.
The reason I don't want ECC is the cost. I have already blown up my budget with ASUS TUF Gaming motherboard and brand new intel 13th gen i5 because I don't want to gamble on cheap stuff.
I do plan on using those cheap stuff until I saw the comment saying it's frustrating when you have to going back and forth with the customer support team.
I'm planning to build my own NAS and now decide to use Unraid instead of TrueNAS because ECC memory will be important to ZFS compared to other filesystems. (I saw this from a lot of discussions)
My question is why? I know ZFS rely on RAM for caching, but why ZFS is more vulnerable to memory corruptions? And why other systems like unraid and btrfs are not as emphasize ECC memory?
It's quite interesting to see two major voices. People from TrueNAS side suggest you must have ECC memory if you care about your data. But people from unraid side said ECC it's not worth the money.
----
I also think if I keep a list of hashes when I back up my data, then ECC might not be nessary at all.
Well I threw my QSL cards up for sale. This is the first time I've ever sold a physical protocol, so hopefully everything goes well: https://ncommander-shop.fourthwall.com/products/kd2jrt-qsl-card-oregon-trail-design?
Thoughts welcome, but really I'm just excited that I got this all setup and working.
Boo👻 The #Pinecil V2, #PineTime, #PinePhone Pro as well as other devices, accessories and bundles are back in stock
-Devices: https://pine64eu.com/product-category/devices/
-Bundles: https://pine64eu.com/product-category/bundles/
-Accessories: https://pine64eu.com/product-category/device-accessories/
Well I'm pretty sure I just found a race condition in the Linux kernel involving the fact that AX.25 probably has only ever been tested at 1200 and 9600 baud.
This post to linux-hams (https://lore.kernel.org/linux-hams/CANnsUMHwRsUEhwwvTn1pi6Ct3U=Xw77ESz5xB+fhtyfDc_J8Ww@mail.gmail.com/T/#t) is the same exact problem I ran into.
I haven't tinkered in the kernel in a very long time, and as far as locking and queue interfaces, I'm probably going to need someone to point my in the right direction. If you're interested, let me know.