Tobias Mann
@tobiasmann.bsky.social
Systems Editor at TheRegister / SitPub
— hiker, animal lover, photographer, blogger, and tech journo.
— hiker, animal lover, photographer, blogger, and tech journo.
I'll be at SC25 next week repping @theregister.com for the fourth year running. Give me a shout if you're going to be in St Louis for the event.
November 10, 2025 at 4:33 PM
I'll be at SC25 next week repping @theregister.com for the fourth year running. Give me a shout if you're going to be in St Louis for the event.
Nvidia's biggest scale up domain is 72 GPUs. Google's is 9,216 TPUs.
Historically TPUs have trailed on FLOPS, memory, & bandwidth. That's no longer the case with Ironwood.
Google has a Blackwell-class TPU with absurd scale. More on @theregister.com ⬇️
www.theregister.com/2025/11/06/g...
Historically TPUs have trailed on FLOPS, memory, & bandwidth. That's no longer the case with Ironwood.
Google has a Blackwell-class TPU with absurd scale. More on @theregister.com ⬇️
www.theregister.com/2025/11/06/g...
TPU v7, Google's answer to Nvidia's Blackwell is nearly here
: Chocolate Factory's homegrown silicon boasts Blackwell-level perf at massive scale
www.theregister.com
November 7, 2025 at 4:16 PM
Nvidia's biggest scale up domain is 72 GPUs. Google's is 9,216 TPUs.
Historically TPUs have trailed on FLOPS, memory, & bandwidth. That's no longer the case with Ironwood.
Google has a Blackwell-class TPU with absurd scale. More on @theregister.com ⬇️
www.theregister.com/2025/11/06/g...
Historically TPUs have trailed on FLOPS, memory, & bandwidth. That's no longer the case with Ironwood.
Google has a Blackwell-class TPU with absurd scale. More on @theregister.com ⬇️
www.theregister.com/2025/11/06/g...
I might be crucified for saying this, but OEM GPU servers are boring now.
Everything is just a rebadged HGX box or NVL rack now.
The only point of differentiation becomes whose lights out management interface does it have.
Everything is just a rebadged HGX box or NVL rack now.
The only point of differentiation becomes whose lights out management interface does it have.
October 21, 2025 at 6:29 PM
I might be crucified for saying this, but OEM GPU servers are boring now.
Everything is just a rebadged HGX box or NVL rack now.
The only point of differentiation becomes whose lights out management interface does it have.
Everything is just a rebadged HGX box or NVL rack now.
The only point of differentiation becomes whose lights out management interface does it have.
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
Full review is already in the works #AI #Workstation #watercooling #Tech
October 21, 2025 at 4:08 PM
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
Full review is already in the works #AI #Workstation #watercooling #Tech
This was a fun review. I don't think folks realize how big a deal the DGX Spark is. A year ago an #Nvidia #workstation with 128GB+ of VRAM cost tens of thousands of dollars. Is it better than AMD's Strix Halo? Now that's the real question.
#DGXSpark #AI
www.theregister.com/2025/10/14/d...
#DGXSpark #AI
www.theregister.com/2025/10/14/d...
DGX Spark Nvidia's desktop supercomputer: first look
hands on: This relatively-affordable AI workstation isn’t about going fast; it’s about doing everything well enough
www.theregister.com
October 14, 2025 at 3:11 PM
This was a fun review. I don't think folks realize how big a deal the DGX Spark is. A year ago an #Nvidia #workstation with 128GB+ of VRAM cost tens of thousands of dollars. Is it better than AMD's Strix Halo? Now that's the real question.
#DGXSpark #AI
www.theregister.com/2025/10/14/d...
#DGXSpark #AI
www.theregister.com/2025/10/14/d...
On the road again.
September 28, 2025 at 4:49 PM
On the road again.
Reposted by Tobias Mann
If you can't use AI then it's bye bye, Accenture tells staff.
If you can't use AI then it's bye bye, Accenture tells staff
ai-pocalypse: Consultancy says machine learning advice is making bank
www.theregister.com
September 26, 2025 at 8:05 PM
If you can't use AI then it's bye bye, Accenture tells staff.
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
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.
I love my dog. He's usually a very good boy. But having adopted him as a 10 week-old puppy less than a year before COVID hit, his anxiety can be overwhelming at times.
He had a vet visit today. He got his shots, but wouldn't let the Dr. do a proper exam. We get to come back in 4 weeks and try again
He had a vet visit today. He got his shots, but wouldn't let the Dr. do a proper exam. We get to come back in 4 weeks and try again
September 12, 2025 at 5:27 PM
I love my dog. He's usually a very good boy. But having adopted him as a 10 week-old puppy less than a year before COVID hit, his anxiety can be overwhelming at times.
He had a vet visit today. He got his shots, but wouldn't let the Dr. do a proper exam. We get to come back in 4 weeks and try again
He had a vet visit today. He got his shots, but wouldn't let the Dr. do a proper exam. We get to come back in 4 weeks and try again
Look what just landed in the lab
September 12, 2025 at 2:36 PM
Look what just landed in the lab
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
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
Ever since Nvidia started talking about disaggregated inference architectures at GTC this spring, I had a feeling a HBM-less prefill accelerator was only a matter of time.
My latest for @theregister.com
www.theregister.com/2025/09/10/n...
#Nvidia #AI #Datacenter #Servers #HPC
My latest for @theregister.com
www.theregister.com/2025/09/10/n...
#Nvidia #AI #Datacenter #Servers #HPC
Nvidia's context-optimized Rubin CPX GPUs were inevitable
Analysis: Why strap pricey, power-hungry HBM to a job that doesn't benefit from the bandwidth?
www.theregister.com
September 10, 2025 at 3:05 PM
Ever since Nvidia started talking about disaggregated inference architectures at GTC this spring, I had a feeling a HBM-less prefill accelerator was only a matter of time.
My latest for @theregister.com
www.theregister.com/2025/09/10/n...
#Nvidia #AI #Datacenter #Servers #HPC
My latest for @theregister.com
www.theregister.com/2025/09/10/n...
#Nvidia #AI #Datacenter #Servers #HPC
Oof. I can relate. Last night my email was filled with TrueNAS warnings. A drive reported Smart Errors.
Logs: extended test failed.
short test fails. Yep dead drive. 😟
Backup my core files and drop in the cold spare. 6 hours of resilvering left to go.
www.theregister.com/2025/09/03/m...
Logs: extended test failed.
short test fails. Yep dead drive. 😟
Backup my core files and drop in the cold spare. 6 hours of resilvering left to go.
www.theregister.com/2025/09/03/m...
Matrix.org homeserver grinds to a halt after RAID meltdown
: Engineers wrangle 55 TB restore and traffic replay as millions of messages queue up
www.theregister.com
September 3, 2025 at 2:54 PM
Oof. I can relate. Last night my email was filled with TrueNAS warnings. A drive reported Smart Errors.
Logs: extended test failed.
short test fails. Yep dead drive. 😟
Backup my core files and drop in the cold spare. 6 hours of resilvering left to go.
www.theregister.com/2025/09/03/m...
Logs: extended test failed.
short test fails. Yep dead drive. 😟
Backup my core files and drop in the cold spare. 6 hours of resilvering left to go.
www.theregister.com/2025/09/03/m...
New bench who this?
August 30, 2025 at 12:56 AM
New bench who this?
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
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
In my latest hands on for @theregister.com I break down everything you need to know to run large language models in the privacy of our home using Llama.cpp.
www.theregister.com/2025/08/24/l...
#AI #PC #LLM #HomeLab
www.theregister.com/2025/08/24/l...
#AI #PC #LLM #HomeLab
How to run LLMs on PC at home using Llama.cpp
Hands on: Everything you need to know to build, run, serve, optimize and quantize models on your PC
www.theregister.com
August 25, 2025 at 1:52 PM
In my latest hands on for @theregister.com I break down everything you need to know to run large language models in the privacy of our home using Llama.cpp.
www.theregister.com/2025/08/24/l...
#AI #PC #LLM #HomeLab
www.theregister.com/2025/08/24/l...
#AI #PC #LLM #HomeLab
Reposted by Tobias Mann
Some stories are just made for puns, not to mention highlighting F-grade security.
Yes, I did have fun with this.
Yes, I did have fun with this.
McDonald's not lovin' it when hacker exposes rotten security
: Burger slinger gets a McRibbing, reacts by firing staffer who helped
www.theregister.com
August 20, 2025 at 2:21 PM
Some stories are just made for puns, not to mention highlighting F-grade security.
Yes, I did have fun with this.
Yes, I did have fun with this.
So wait. Is Arm really getting into silicon? Seems like a sure fire way to piss off your customers.
Maybe chiplets would be okay? Most Arm customers take off-the-shelf cores anyway.
My latest for @theregister.com
www.theregister.com/2025/08/19/a...
#Arm #CPU #Chips #tech
Maybe chiplets would be okay? Most Arm customers take off-the-shelf cores anyway.
My latest for @theregister.com
www.theregister.com/2025/08/19/a...
#Arm #CPU #Chips #tech
Top AWS chip engineer reportedly defects to Arm
: Rami Sinno led Trainium and Inferentia development at Amazon
www.theregister.com
August 20, 2025 at 3:17 PM
So wait. Is Arm really getting into silicon? Seems like a sure fire way to piss off your customers.
Maybe chiplets would be okay? Most Arm customers take off-the-shelf cores anyway.
My latest for @theregister.com
www.theregister.com/2025/08/19/a...
#Arm #CPU #Chips #tech
Maybe chiplets would be okay? Most Arm customers take off-the-shelf cores anyway.
My latest for @theregister.com
www.theregister.com/2025/08/19/a...
#Arm #CPU #Chips #tech
The way the report describes the B30A it sounds a lot more like a B300 NVL than a H20 replacement...
www.theregister.com/2025/08/19/n...
www.theregister.com/2025/08/19/n...
Nvidia reportedly plotting cut-down B300 for Chinese market
: It's that or a replacement for its aging H200 NVL PCIe cards
www.theregister.com
August 19, 2025 at 7:40 PM
The way the report describes the B30A it sounds a lot more like a B300 NVL than a H20 replacement...
www.theregister.com/2025/08/19/n...
www.theregister.com/2025/08/19/n...
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?
Which do you opt for?
August 19, 2025 at 3:48 AM
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?
Which do you opt for?
Reposted by Tobias Mann
Big Dyson Sphere can be sung to Pink Pony Club
anyone who has ever spent time with a 5 year old has heard stories like this, except 5 year olds are cute and their tall tales are charming
August 19, 2025 at 3:14 AM
Big Dyson Sphere can be sung to Pink Pony Club
The funniest bit to me was the comparison of US chip tracking to the machines in The Matrix.
Last I checked, Uncle Sam wasn't using human batteries to power major industry or AI development.
My latest for @theregister.com
www.theregister.com/2025/08/18/c...
Last I checked, Uncle Sam wasn't using human batteries to power major industry or AI development.
My latest for @theregister.com
www.theregister.com/2025/08/18/c...
China labels US as 'surveillance empire' over chip tracking
Comment: Spy vs spy in the chips
www.theregister.com
August 18, 2025 at 10:27 PM
The funniest bit to me was the comparison of US chip tracking to the machines in The Matrix.
Last I checked, Uncle Sam wasn't using human batteries to power major industry or AI development.
My latest for @theregister.com
www.theregister.com/2025/08/18/c...
Last I checked, Uncle Sam wasn't using human batteries to power major industry or AI development.
My latest for @theregister.com
www.theregister.com/2025/08/18/c...
Cisco, Arista, Broadcom and every other Ethernet equipment vendor is jazzed about AI and its not hard to see why when when ever GPU is an excuse to sell 3-5 of the fastest switch ports money can buy:
My latest for @theregister.com
www.theregister.com/2025/08/15/e...
My latest for @theregister.com
www.theregister.com/2025/08/15/e...
Cisco, other Ethernet switch vendors, high on AI networks
: When one GPU translates into three to five of the fastest switch ports money can buy, can you blame them?
www.theregister.com
August 18, 2025 at 7:25 PM
Cisco, Arista, Broadcom and every other Ethernet equipment vendor is jazzed about AI and its not hard to see why when when ever GPU is an excuse to sell 3-5 of the fastest switch ports money can buy:
My latest for @theregister.com
www.theregister.com/2025/08/15/e...
My latest for @theregister.com
www.theregister.com/2025/08/15/e...
Called it! GPT-5 was a cost cutting measure.
"We have better models, and we just can't offer them because we don't have the capacity. We have other kinds of new products and services we'd love to offer," Altman said.
ICYMI: www.theregister.com/2025/08/13/g...
"We have better models, and we just can't offer them because we don't have the capacity. We have other kinds of new products and services we'd love to offer," Altman said.
ICYMI: www.theregister.com/2025/08/13/g...
OpenAI's GPT-5 is a cost cutting exercise
Analysis: Gotta pay for all those GPUs somehow
www.theregister.com
August 18, 2025 at 1:56 PM
Called it! GPT-5 was a cost cutting measure.
"We have better models, and we just can't offer them because we don't have the capacity. We have other kinds of new products and services we'd love to offer," Altman said.
ICYMI: www.theregister.com/2025/08/13/g...
"We have better models, and we just can't offer them because we don't have the capacity. We have other kinds of new products and services we'd love to offer," Altman said.
ICYMI: www.theregister.com/2025/08/13/g...
Sneak peek at what I've got cooking:
August 18, 2025 at 5:28 AM
Sneak peek at what I've got cooking: