Tobias Mann
banner
tobiasmann.bsky.social
Tobias Mann
@tobiasmann.bsky.social
Systems Editor at TheRegister / SitPub
— hiker, animal lover, photographer, blogger, and tech journo.
Say what you will about the hardware or the software, @tenstorrent.bsky.social's Blackhole QuietBox is a gorgeous bit of kit.

Full review is already in the works #AI #Workstation #watercooling #Tech
October 21, 2025 at 4:08 PM
On the road again.
September 28, 2025 at 4:49 PM
I’m embarrassed to admit that I’ve never held a QSFP DD cable before. It’s enormous compared to the SFP+ DACs I’m used to.
September 12, 2025 at 7:55 PM
Look what just landed in the lab
September 12, 2025 at 2:36 PM
Something curious I’ve noticed is I’m using virtualization less in my homelab preferring instead to run bare metal with containers.

I still keep a PVE box for VMs when they’re warranted but a lot of the stuff I’m doing can be achieved using containers.

#Linux #VM #Homelab #tech
September 11, 2025 at 4:36 PM
R9 9900X on a MSI X870E-P MB.
September 7, 2025 at 5:50 PM
Yes indeed. Stout little card. Though I’m rather eager to put the Radeon Pro AI r9700 to the test.

Here’s batch 1 perf figs if you’re curious.
August 30, 2025 at 2:49 AM
New bench who this?
August 30, 2025 at 12:56 AM
After three years of 24/7 service in my homelab my R9 3900X met its end on Sunday.

During routine thermal paste change the cooler ripped it from the socket bending several pins.

Alas even when bent back into position it refused to post. RIP my friend.

#AMD #CPU #PC #Homelab #Tech
August 26, 2025 at 2:37 PM
AMD Radeon Inference perf in Llama.cpp is an interesting conundrum. The Vulkan backend offers higher token gen but ROCm is vastly superior in prompt processing.

Which do you opt for?
August 19, 2025 at 3:48 AM
Sneak peek at what I've got cooking:
August 18, 2025 at 5:28 AM
Compiling code on a Raspberry Pi 5 is painfully slow.
August 16, 2025 at 12:38 AM
Still need to do some cable management and it remains to be seen whether that PSU will be adequate, but it posted.

24 Zen4 cores
128GB DDR5 5600 RDIMM
2x RTX 3090 TI 24GB w/ NVL

And don’t worry those GPUs are power limited to 250W. The PSU shouldn’t explode.
August 14, 2025 at 12:49 AM
Challenge: go a week with gpt-oss as your only AI assistant. Could you do it?

I have a feeling it's gonna be painful, but I'm going to try.
August 13, 2025 at 4:34 AM
I did a thing…
August 12, 2025 at 11:31 PM
After having a few days to mess with OpenAI's gpt-oss — specifically the 21B parameter version — I gotta admit, these are the most impressive open models to come out of the US so far this year. A 🧵 [1]
August 7, 2025 at 5:16 PM
What again are we advancing? I can’t remember.

#AMD #AI
June 10, 2025 at 3:08 AM
Supermicro is putting Grace Grace in a storage server? 🤯

I wonder if we’ll get Vera Vera superchips too.
March 21, 2025 at 12:22 AM
Why did Nvidia announce four generations of GPUs at #GTC this week? Because everyone else needs to catch up.
March 19, 2025 at 8:39 PM
If it weren’t for the NVLink attach points you’d think you were looking at an NVL4 not a GB300 Superchip #GTC
March 19, 2025 at 7:57 PM
New DGX Station boards on the way with a GB300 desktop superchip on board. #GTC
March 18, 2025 at 6:59 PM
Nvidia reveals its smallest DGX system with a petaOP of AI compute. Called DGX Spark #GTC
March 18, 2025 at 6:58 PM
Vera Ruben Ultra will be a 600kW rack 🥵 #GTC
March 18, 2025 at 6:39 PM
Jensen teases Vera Ruben NVL144 (same # GPUs as NVL72, just a different naming convention)

#GTC
March 18, 2025 at 6:38 PM
Jensen unveils all-new #Blackwell Ultra platform boasting HBM3e memory by 50 percent to 288GB and dense FP4 compute to 15 PFLOPS. #GTC
March 18, 2025 at 6:35 PM