Tsiklon

That memory speed drop going from single DIMM per channel DDR5 to dual DIMM per channel is very much notable - 3600MT/s is barely any faster than what DDR4 could do in it's final days.

The CPU market seems to be missing the "HEDT" platform we used to enjoy.

This looks like a very cool high end x86 workstation. CPU alone is $12000 retail.

Theres a few potential “problems” I see with it - main system RAM is about 30-40x slower than that of the GPU’s and you will have PCI-e as a bottleneck getting working data to or between the GPUs.

In practice is the 400-500GB/s main memory limiting some use cases, or is the PCI-E bus the main issue with these types of systems?

I am a big fan of the original Altar. I like that you’ve been trying new things with that design like the novel keycap profile.

The switches look like cherry ultra low profile switches. I’ve not tried these out, how do they feel compared to the linear kailh choc v1’s you used before.

This cycle is different. Prices have increased for both Sony and Microsoft’s consoles and no higher efficiency versions have been released (ala the PS3, X360).

Partly, windows defender can open and scan the contents of many archives, but extraction of several popular formats isn’t possible

Poland I imagine is a tremendous exponent of South Korea’s arms industry.

I always thought of arms procurement as a tedious and slow process but the Polish/Korean agreements have turned that on its head in my eyes.

Simultaneously supplying new tanks, artillery, and munitions, while quickly standing up local factories to build them in Poland too is impressive, undoubtedly giving the Polish government some well needed breathing room.

Given that the technical package is included here, there must be some form of non-competition clause

I just grabbed a Tissot Gentleman 38mm as my first Swiss machine. Same movement but a little smaller, and in my opinion a more handsome case

Thank you, I’ll keep this in my hat for the next time.

XFS as implemented in RHEL8+ (the only places i've used it in anger) tends to handle being full very badly, leading to system lockups and blocked tasks necessitating a hard reboot. Worse yet is when it's in this state the journal fills and nothing can be done with the volume.

To recover from this on a volume mounted at boot mandates going to either a live disk, or stopping boot in initramfs and running xfs_repair there, I've fruitlessly attempted to play back the journal on many separate occasions by attempting to mount the filesystem (again causing a lock up due to no space) in that state you have to drop the journal, run xfs_repair and then clean up the detritus from /lost+found (and then the location that caused the disk to fill altogether).

EXT4 has other issues certainly, but at least it reserves blocks for the root user explicitly so the system doesn't stop.

Service dependency resolution and parallel startup is super in systemd. Big fan of what I can do with it.

XML - I see you’ve used macOS’ LaunchD, the system that inspired Systemd

Apple’s AMD GPU driver had issues gracefully handling hot plugging monitors via thunderbolt. It reliably panic’d every other time I plugged in my ‘19 16” machine to a thunderbolt 3 dock.

I argued with GPT-OSS 120B about cascade lake Xeon workstation CPU parts not having a GPU when it vehemently said otherwise

Wonderful. I wonder is there a plan for a 30th anniversary edition given that this December marks that point

As a blow in to New York, I find the bodega cats a very charming tradition. Reminding me of “tough as old boots” farm cats, working animals with a purpose and a style all their own.

The extension cable they used to include in box with the computers, it has the third pin for the charger brick connector which is wired to ground

The panda were wonderful, and based off a headphone I loved to bits in the Oppo PM-3.

If Corsair still own the design, they could simply reimplement the original earcups/hinge mechanism from the Oppo as it was super solid.

That won’t work for the home hobbyist 2.4KW of GPU alone plus a 350W threadripper pro with enough PCIe lanes to feed them. You’re looking at close to twice the average US household electricity circuit’s capacity just to run the machine under load.

A cluster of 4 Apple’s M3 ultra Mac studios by comparisons will consume near 1100W under load.

It has a HDMI port and its USB-C ports also support display out. But I believe most who buy it intend to use it headless. The machine runs Ubuntu 24.04 and has a slightly customised Gnome (green accents and an nvidia logo in GDM) as its desktop.

Fortnite is 9 years old this year. Epic brought in biblical amounts of money from just this one property over this time. Where and how did they spend this money?

Surprisingly the original developers Milestone are doing a reboot/reimagining of the series which is out this week.

Tangentially related, that screenshot of Screamer 2 caught me off guard completely, I loved that game to death, and I feel I was the only one of my friends to have played it. Tremendous handling model and superb music.

I’ve had no issues running GPT-OSS 120b with decent performance on the machine (HP Zbook Ultra G1a). Running on Bluefin/Universal Blue and Windows.

Strix Halo’s unified setup is pretty cool. In systems with 128GB of memory, in BIOS set the dedicated GPU memory to the smallest permitted and the Drivers will use the whole main memory pool appropriately in Linux and Windows

The difference between DDR4 and 5 is quite substantial. I have a fully loaded Cascade Lake Mac Pro - 6 channels of DDR4-2933 gets me to about 120GB/s or 960Gb/s. PCIe 3.0 is a major Achilles heel of what would be a capable workstation system with modern nvidia GPUs precisely for the reason you document.

When did nvidia drop PNY as ODM for their reference cards? I recall my A5000 (now 2 Gen old) was made by PNY.

To my recollection on the machines that shipped with Lion (circa 2011) you’ll want to set up a protective MBR with the appropriate drive dimensions on the GPT, to get it to install windows like with boot camp.

To my recollection on machines with discrete GPUs this is what triggers the appropriate hardware configuration (BIOS boot, disabling the internal GPU and switching the MUX to only route via the AMD card)

But I do recall getting the internal GPU working with this trick https://github.com/0xbb/apple_set_os.efi

Beautiful, it's typeset in classic cló Gaelach.

Intel also sees the value in decent onboard GPUs now, their newly announced laptop processors have solid onboard GPUs too

From looking at the WiFi ssid’s broadcast at the New York subway stations, I believe Boldyn also does the phone coverage here too