Projects
The Website
With work, I have not had much time to write these blog posts outside of the weekends. As a result, I decided to consolidate Week 33 to Week 38, 2025 into this one post. I do not remember what kept me from working on this on the weekends which is the perfect time to do it. After this, I do plan to start releasing blog posts on time again.
For the longest time, this website still needs a lot of work done to it with missing pages and broken downloads. I am aware of this and would try get around to fixing it.
Starting with this blog post, blog icons would now be 256x256 instead of 512x512 to use less resources on the client (browser) and improve loading speed on slower connections.
On September 21, I was able to resize all of the icons in each page directory at once using this single ImageMagick command: find . -type f -iname 'wk*_icon.webp' -exec mogrify -resize 256x256 {} +
Shift from Git and VSCodium
During Week 35, 2025, I started to work on moving from GitHub for multiple reasons. I have been thinking about doing this for quite a long time now. I feel like I should not have to use stuff that I do not want to use and has me suffer using it.
I switched back to Geany from VSCodium for the time being as it is much lighter on resources and generally seems cleaner.
Git and GitHub
I started to use rsync to sync my website data from the PC to whatever laptop I am using. This is actually faster than using Git or vscode-remote.
On September 14, 2025, I was able to successfully download all of my repositories and upload them to the storage server. Now this storage server, or NAS, is to store repositories.
However, this means that changes to project files would no longer appear on GitHub. I may set up a mirror on this website but given my schedule, it may be a while before this happens.
At some point, when I build up the courage to want to touch GitHub again, I would start removing repositories.
MWDL/WBPC
I recently started to work on switching more systems to Fedora with Linux Mint being banned from my hardware in December. I plan to have Mint completely phased out by January 1, 2026.
SVCS
After working on the two SnapServer systems at 2nd Life, which both had Supermicro X10SL7-F motherboards with DDR3 memory which came to me as a surprise. I realized that the dual-X5670 X8DAi system that I have currently is much too powerful and inefficient for my workload that uses less than an eighth of system memory
On September 18, 2025, I ordered a pre-assembled X10SLH-F board; Intel Xeon E3-1275L, heatsink and some RAM included for about US$69.
"Polystyrene v2"
On August 13, I brought home the Dell Precision 690 I covered in the previous post.
Not until August 16, I powered up the system for the first time since I brought it home. I used an SSD from another system that already had Fedora installed. Immediately I tried to run Hiveswap Friendsim which was successful, which was not surprising given Ren'Py's requirements.
On August 23, I attempted to run the Minecraft 1.16.5 modpack Seaopolis (2021). Given Minecraft's resource usage, the game ran surprisingly well.
Desktop Monitor Switch
On September 3, when I was digging around in the monitor scrap bin at 2nd Life, I came across a 21.5" monitor from 2021 that had an Innolux panel. These units were ViewSonic VG2249, which are advertised to specifically have MVA type panels which Innolux is currently the main manufacturer of.
Soon after, I found two of these VG2249 on eBay for less than US$200 total (including coupon and sale discounts). While I would prefer Innolux's AAS (Azimuthal Anchoring Switch), I consider this the next best thing.
By September 10, 2025, this setup was finalized. It may take a while for me to used to these monitors. It probably does not help that my office chair is an inflatable armchair that is probably older than me that seems to develop another leak every time I sit on it.
Later on, on September 20, one of the VG2249 monitors just went to a black screen and no longer powered on. I had my old Acer X213H, albeit having an AUO TN panel, take place of it for now.
On September 21, 2025, wondering why these units I have here look so different from what I have at work. I checked inside both monitors and noticed that these have M215HJJ-L30 Rev.C5 while what I use at work has, presumably, the M215HJJ-L30 Rev.C6. Innolux panels often can be drastically different between revisions which was made apparent here. The Rev.C6 seems to have better viewing angles and overall is a better experience than the Rev.C5. This makes sense as both of what I have were made in 2019 while the one at work is from early 2021.
The iPhone XR
While riding a Lime scooter back to 2nd Life Inc., the iPhone XR I was using flew out of my hand when I hit a curb, having it slam into the road and crack the front glass. Fortunately, the touchscreen is still functional and the LCD itself sustained minimal damage.
It is about time I stop using this iPhone for the social-media-and-untrusted-apps phone. Updating to iOS 18.x has slowed down the phone significantly. I may find a Samsung Galaxy S21, specifically the SM-G991B/DS variant that uses the Exynos SoC.
It may be a while before I can do this.
MediaCow Touch 2
I am aware I have done absolutely nothing with the project in over several months. With the changes to my workflow as described above in "Shift from Git and VSCodium", I may start working on the project again.
2nd Life Inc.
As I am consolidating four weeks of activity into a single post, this section will be long.
Week 33, 2025
Day 16 - August 13, 2025
As mentioned in the above section, I brought home the Dell Precision 690.
I continued work on Cisco networking equipment.
Day 17 - August 14, 2025
While digging through a box of DDR3 ECC SODIMM RAM from what appeared to be telecommunication system boards, I found a set of Memphis-branded SODIMMs with Intelligent Memory-branded ICs. This is the first time I have seen Intelligent Memory-labeled parts in person after hearing about Intelligent Memory since 2020, even contacting in 2021 in an attempt to source ECC DRAM parts. The ICs themselves are presumably relabeled PSC (Powerchip) as seen by the die code marking style.
Later on, a couple of Dell OptiPlex systems from 1998 were brought up to the inspection area. I worked on one while a coworker inspected the other.
As it was Thursday, we worked on scrap starting noon.
Day 18 - August 15, 2025
In an attempt to find a way to speed up the clearing of hard drives, I tried to make use of an LSI 9211-8i but it needed to be flashed into "IT mode" which would allow access to every drive individually. However, the firmware needed to do this was removed by Broadcom likely due to age of the card, effectively making this useless for the task unless I were to dig around for it, which I do not have time for.
I eventually dropped the idea of using the Supermicro C2SBC-Q desktop for this and went for an already-configured setup, being a Dell PowerEdge T550.
Week 34, 2025
Starting this week, I work five days a week, making this a full-time job.
Day 19 - August 18, 2025
For clearing drives, I set up a Dell PowerEdge T550, allowing up to eight drives to be written to at once.
Day 20 - August 19, 2025
I switched over to a PowerEdge T430 to clear drives, appearing to be more effective in some ways. The T550 is also a much more valuable unit so it was a better idea to get it prepared for resale.
Day 21 - August 20, 2025
I continued HDD/SSD work. I found some RAMAXEL-branded SODIMMs that had non-relabeled Nanya DDR4 which I found interesting as that is a rare sight.
Day 22 - August 21, 2025
Again, this day surrounded the clearing of SSDs and HDDs. This would be going on for quite a while so many of the following sections are going to start with this.
I found some ASIC-based bitcoin miners in the power supply scrap bin. I noticed that the controller boards used a Xilinx ZYNQ XC consistently paired with US-diffused (Manassas, Virginia; 1X location code) flash, some boards having also US-diffused Micron DDR3 SDRAM while others had ESMT-labeled RAM.
Day 23 - August 22, 2025
Again, this day surrounded the clearing of SSDs and HDDs. Noting how quite powerful the fans are on these bitcoin miner systems, some of them being Nidec UltraFlo W12E12BS11B5-57 (12v 1.65A) possibly having 210-265 CFM. I started to salvage the fans off from them thinking they could be potentially useful for the "Polyvinyl Chloride" AI workstation build which is going to need massive amounts of airflow.
On these Antminer ASIC boards, I noticed that the power is actually wired in series from one ASIC to another. I have heard of this before but I would have never expected to see it for myself.
I came home to a wonderful surprise; moments after I got home, the power went out. The UPS cut out immediately as the battery pack is in need of replacement. This lasted quite a while and power was not restored until around 8-9 PM. While I waited, I went down to a playground I like to go sometimes when no one else is around. I needed to touch grass but it sucks when it is at the cost of my servers.
Week 35, 2025
Entering another week of staring at dd prompts
Day 24 - August 25, 2025
On this day, we started to pack up hundreds and hundreds of hard drives into boxes as part of one big order. How we packed them is that the first layer would be 3.5" drives, for any empty spaces, I would cut a piece of polyethylene packing foam to size and put it in that space. The second layer was all 2.5" drives, I was able to get an average of 97 2.5" drives in each box. This seemed to be a majority of the hard drives we had
Day 25 - August 26, 2025
Another day of clearing drives, I do not have much of note to put here.
Day 26 - August 27, 2025
Another day of clearing drives for the most part.
By the end of the day, I was able to come up with a way to print labels listing the capacity, type and interface to be put on SSDs/HDDs using a Brother QL-820NWB label printer. From experience, Brother tends to offer solid support for Linux-based operating systems so I was able to do this from Fedora 41 on the ThinkPad T430s.
This was significantly faster than writing on these labels by hand, with my dysgraphia, it would have become painful quick at this scale.
Later on, I found quite a bit of 4GB Kingston SODIMMs that were all relabeled Nanya, presumably A-die.
Day 27 - August 28, 2025
Another day of clearing drives for the most part.
I was given a Microsoft Sidewinder Precision Pro joystick to figure out how to get working. I found out it requires an interface called a Game Port, which I had nothing immediately on hand that could be used to interface it. Relatively low cost adapters can be found online for connecting this device over USB.
Looking for anything to work on, I found a couple of GE-branded Interlogix MVC-21S CRT monitors that at first look like cube-shaped televisions. I did not have anything that I could think of that had composite video out so I just tested the screen by using the OSD.
Day 28 - August 29, 2025
Apart from continuing to clear hard drives, this was a fairly interesting day. I got to crush hard drives that either failed before or during wiping/testing and those marked to be crushed due to security requirements. I unfortunately did not check the PCB of a WD Velociraptor from 2008 which had a Qimonda SDRAM IC on-board before putting it in the crusher. These DRAM ICs are becoming more and more rare so I must be more careful.
Later on, around noon, we had a company cookout that more or less was my lunch break for the day.
Week 36, 2025
Monday, September 1, 2025 was Labor Day so I had no work on this day.
Day 29 - September 2, 2025
When I started clearing drives again, I was tasked with putting about 850 32GB DDR4 RDIMMs into larger racks for shipping.
After finishing up the packing of RAM, I went back to hard drives where soon after, the power went out. When the warehouse went dark, I was like "welp" and started to do something else that did not require electricity. I even insisted on continuing to work since it is definitely more fun and interesting than waiting in the breakroom.
Day 30 - September 3, 2025
While waiting on hard drives, I went to the scrap area to check what is in these monitors. After digging through some monitors with AUO and LG panels, I finally found an Innolux panel which was in a ViewSonic VG2249 from January 2021. I eventually found more of these monitors online. I have already gone into detail about this in the above section "Desktop Monitor Switch"
Day 31 - September 4, 2025
I do not have much of note from this day. It is presumed I continued drive work.
Day 32 - September 5, 2025
Near the end of the day, I was tasked to put in 256GB SSDs and at least 8GB of RAM in ten Dell laptops that were sold to a client. Given that a vast amount of the RAM I had on hand were those 8GB Samsung 1Rx16 M471A1G44BB0-CWE modules, I was able to give all of them 16GB.
Week 37, 2025
I have started to see an end to the wiping of drives. I may finish the work by the end of the next week.
Day 33 - September 8, 2025
As I finished most of the SATA HDDs and SSDs at this point, I have started to look for a system that can interface more than one NVMe SSD at once. I eventually came across a Dell micro desktop but it seemed that just one of the two M.2 Key M slots had PCIe, required for NVMe.
Later I came across an AXIOM DDR4 module with chips with markings that I have not seen before. Me along with others online determined it to be Samsung given the 11x7.5mm package size and that the markings did not look like SK hynix or Micron.
Day 34 - September 9, 2025
I continued to use the Dell desktop until it was needed for a resale order. I went back to try using one of the iBUYPOWER prebuilts that had the ASUS PRIME Z490-P motherboards. Unlike before, the Ubuntu live USB booted successfully.
Using the two onboard M.2 connectors, M2_1 that is connected to the chipset and M2_2 that is presumably connected to the CPU, I was able to reliably write to two drives at once.
Day 35 - September 10, 2025
While continuing the "wiping" of NVMe SSDs, I started to sort RAM and putting the most common model in a labeled bin. For 8GB DDR4, this was Samsung M471A1G44BB0-CWE and SK hynix HMA81GS6AFR8N-UH. I am not really sure why I started doing this; it is likely to do something at least considered productive while I wait on the SSDs.
Day 36 - September 11, 2025
This day, I was able to get two PCIe NVMe adapter cards which were delivered before I got up to go to work. This doubled the amount of NVMe SSDs that could be wiped at once using the iBUYPOWER ASUS PRIME Z490-P desktop.
During scrap work, I found a Samsung SM-T560NU tablet in a scrap box in nearly perfect condition. However, the Android version is quite old at version 7.1.1, severely limiting what can be done with it. Versions of Samsung Internet and Google Chrome that are preinstalled are also quite old, having this website one of the few websites that load properly. I could go on about how this is forced obsolecense as I quite literally found this device in an e-waste pile.
Day 37 - September 12, 2025
While working on SSDs as usual, a coworker asked me to work on two SnapServer units which were 2U rack-mounted servers each having a Supermicro X10SL7-F, Intel Xeon E3-1220 v3 and 8GB of RAM. This did not take long as there were no drives installed apart from an 1GB USB drive connected to the on-board USB connector. This was definitely more interesting than watching dd.
By the end of the day, I was running out of M.2 NVMe SSDs to wipe, I found another bin of SSDs mixed with WLAN (Wi-Fi) cards in a cabinet that were not wiped yet. I added those to what I was working on. Soon I would move onto the M.2 SATA SSDs which cannot be used with the PCIe NVMe cards.
Week 38, 2025
I finally saw the end of wiping drives this week, at least for now.
Day 34 - September 15, 2025
This day, RAM sorting and SSD clearing continued. Near the end of the shift I found an HP thin client in the scrap pallet. Looking at it more I found out it had a VIA Eden 1GHz CPU. I decided to hold onto it as it could be useful for testing software on such kind of hardware.
Day 35 - September 16, 2025
I finished the last of the SSDs on this day. I got the desktop I was using put back together and back onto the pallet
Day 36 - September 17, 2025
I do not have much of note from this day apart from after work where I almost rode a Lime scooter all the way to Maymont and back from 2nd Life. About halfway I noped out as the rain continued. I rode 6.2 miles in the single session.
Day 37 - September 18, 2025
I started on networking equipment. First were some small managed switches that did not take much time at all to inspect. I moved onto routers which were more like a desktop motherboard put into a server chassis. Both of them had Pentiums without ECC memory support.
I came across thirteen USB floppy disk readers from 2003-2004. They worked out of box on Fedora 41 and was able to test reading and writing with each drive. This may have been my first time that I have actually done anything with floppy disks.
Scrap work commenced at noon, I took apart some fairly large Avaya G650 Media Gateway systems in order to have them light enough for me to pick up. Also in scrap, there was a new-in-box HP printer that could be worth US$400 on eBay so we set it aside.
Day 38 - September 19, 2025
Near the end of the day, I saw the sight I was waiting for for a week now: a pallet of just normal computers to work on which were some Dell laptops and micro desktops.
All of the BIOS were locked but they all seemed to use a common password that was likely provided to us by the former owner. The method of clearing BIOS to clear the password actually worked on one unit but doing this for all of them would be very tedious so I used the known password for the rest of the units.
For the Lime scooter ride, I was determined to reach Maymont from 2nd Life but ran into some challenges on the way. I just had enough of a charge to get from 2nd Life to somewhere near Forest Hill Park, where I had the choice of pushing myself on the scooter or walking. I ended up walking about eight minutes to reach the next ride. The bridge on Westover Hills Blvd I planned to cross was actually a "no go" zone and not just a "no parking" zone. Unaware of this, going about 30 miles per hour down a hill until someone yelled out as I was riding into what was likely a highway. It should be much more clear in the app what is a "No Parking" and a "No Go" zone. After this, I just wanted to get home alive. As other scooters around had little to no charge, I had to ride multiple in order to get back to 2nd Life. I may try doing this again but crossing the James on a bridge I actually know is safe to ride on, which consistently has been Mayo Bridge. Going through downtown would be much more convenient anyway.
Personal
Some major personal changes and events occured during this period of time.
Homestuck "ban"
Starting September 19, 2025, Homestuck would no longer be present in my projects and how I express myself. I would no longer continue to read it, use related games for hardware testing.
Riding in Richmond
Starting September 8, I have been riding rideshare scooters within Richmond. Most of the time I do it for fun though it is quite useful from getting to one building to another at VCU.
John Baldwin
On September 4, 2025, I went to a VCU LUG meeting where John Baldwin gave us a presentation about his work experience and work with the FreeBSD project. John Baldwin recently became a board member of the FreeBSD Foundation in June.
I did not really say much during the event other than asking about the status on the ARMv7/armhf port as I did once consider trying to get FreeBSD to run on the Sunplus SP7021.
The 2016 Subaru Outback
Following successful setup of a car loan in the previous days, on August 23, 2025, me and my father went to Virginia Beach to pick up the 2016 Subaru Outback that was picked out for me to have. I drove myself back from Virginia Beach after. The vehicle itself I actually found to be quite aesthetically interesting.
This allows me true independence on where and when I want to go. As a result of this, I have been hanging out within Richmond more.
September 12, 2025 outing
With a couple of friends, we went out to go to various places. Continuing my search from yesterday, we went to Green Top first to actually see a KRISS Vector in person which is something I wanted to do for several years now with my fascination in its design, origin and mechanics. I have became more serious about familiarizing and educating myself about firearms due to the chaotic events of the previous days. This is not new for me; I have always had an interest in firearms and their technology since 2020, especially 2021.
After that, we went out to a store that had mostly food from Asia. While I was there, I found a Chinese can of Red Bull that unlike the Red Bull in the US, it lacked carbonation, had less caffeine and considerbly had more flavor. A couple days later, I came back to pick up more cans.
I went by myself to a nearby store named Miniso. There was the most Minecraft merchandise I have seen in one place, 2014 me would be proud.
After that we went to Wawa to get "dinner" and went over to a neighborhood of one of the two where we walked around for a little while. It was at the point where I was starting to fall asleep so we called it for the night.