M
More Than Moore
morethanmoore.substack.com · Personal blog · on HN since 2022
More Than Moore is Dr. Ian Cutress's newsletter on semiconductors, CPU and AI chip design, process nodes, and industry earnings.
What HN saysHN links it for deep dives such as IBM's 0.7nm process and HBM-on-logic, and for Nvidia financial analysis.
29HN people shared it
15linked it in comments
63stories submitted
1.1Kstory points
4 months agolast shared
2022people per year2026
Most-shared posts
- 1The Race to 2nm: RISC-V Chips in Japan – By Dr. Ian Cutress3 people · 90 points · first shared 2024 · 29 comments on HN
- 2Unboxing the Tenstorrent Grayskull AI Developer Kit3 people · 3 points · first shared 2024 · discussion on HN
- 3Efficient Computer's Electron E1 CPU – 100x more efficient than Arm?2 people · 267 points · first shared 2025 · 111 comments on HN
- 4AMD's AI Future Is Rack Scale 'Helios'2 people · 136 points · first shared 2025 · 82 comments on HN
- 5Major Changes at RISC-V Designer SiFive2 people · 58 points · first shared 2023 · 10 comments on HN
- 6Deep Dive into IBM's new NanoStack 0.7nm Process Node for Chips – 666 MTr/mm22 people · 7 points · first shared 2026 · discussion on HN
- 7Intel Optane, RIP2 people · 5 points · first shared 2022 · discussion on HN
- 8Intel 2025 Q1 Financials2 people · 2 points · first shared 2025 · discussion on HN
- 9Solving the Problems of HBM-on-Logic2 people · 2 points · first shared 2025 · discussion on HN
- 10Nvidia Enabling Computational Lithography on GPUs2 people · 1 points · first shared 2023 · discussion on HN
- 11Intel CEO Letter to Employees1 people · 292 points · first shared 2025 · 554 comments on HN
- 12The Future of Big Iron: An Interview with IBM’s Christian Jacobi1 people · 84 points · first shared 2024 · 49 comments on HN
- 13AMD Q3'2024 Financials1 people · 43 points · first shared 2024 · 40 comments on HN
- 14No Time Like MEMS Time1 people · 39 points · first shared 2024 · 21 comments on HN
- 15How to Solve LLM Hallucinations1 people · 18 points · first shared 2024 · 7 comments on HN
What HN people wrote when they linked it
IanCutress · 2026-06-25
For what it's worth, here's my 7000+ word deep dive into the technology.
https://morethanmoore.substack.com/p/ibms-announces-07nm-process-node
alright2565 · 2026-02-14
First one that comes to mind is https://morethanmoore.substack.com/
Scene_Cast2 · 2025-12-25
There was previous discussion a few days ago on "Solving The Problems of HBM-on-Logic" that I think is relevant.
https://news.ycombinator.com/item?id=46302002
https://morethanmoore.substack.com/p/solving-the-problems-of-hbm-on-lo…
hhh · 2025-09-19
No. It used to be more even between datacenter and gaming for NVIDIA, but that's not been the case for a few years. Gaming has brought in less money than networking (mellanox) since '24 Q4.
https://morethanmoore.substack.com/p/nvidia-2026-q2-financial-results
lproven · 2025-08-18
> For whatever reason, Acorn dropped the ball.
Acorn's CPU division is the most successful CPU design house in the world and sells around 10x more than all forms of Intel and Intel-compatible chips put together.
It was named after its first product, the Acorn RISC Machine: ARM. It is still called Arm Ltd. today.
Arm alone is one half of the entire CPU market.
https://morethanmoore.substack.com/p/arm-2025-q4-and-fy-financials
An Acorn-compatible CPU is inside half of the processor-powered devices in the world.
How is that "dropping the ball"? It is the most successful processor design of all time, bar none.
ac29 · 2025-07-24
> Intel headcount in 2022 was 131,900.
Not sure if you intentionally picked the all time high headcount, but you did.
Intel's headcount was relatively stable between 100-110k people between 2014 and 2021 [0]. So, getting down to 75k is definitely still a major reduction, but 2022 was also an outlier. A lot of companies overhired during Covid, and Intel particularly was the beneficiary of WFH pulling in a lot of corporate spending on laptops etc.
[0] https://morethanmoore.substack.com/p/intel-2025-q1-financials
LargoLasskhyfv · 2025-05-02
To read instead, from the same guy:
https://morethanmoore.substack.com/p/a-quiet-giant-steps-forward-arm-v…
uluyol · 2025-01-11
AMD is not ready to have a full announcement for their new consumer GPUs: https://morethanmoore.substack.com/p/where-was-rdna4-at-amds-keynote
PhilippGille · 2024-03-10
Dev kits:
- Grayskull e75 | drawing 75W | 96 Tensix cores | 1 GHz clock | 96 MB SRAM | 8GB LPDDR4 @ 102.4 GB/s | $599
- Grayskull e150 | drawing 200W | 120 Tensix cores | 1.2 GHz clock | 120 MB SRAM | 8GB LPDDR4 @ 118.4 GB/s | $799
It will be interesting to see their inference performance, compared to graphics cards. Will they be interesting for home labs?
I found one interview with unboxing a preview version (if I understood correctly), with some background info, but no performance numbers: https://morethanmoore.substack.com/p/unboxing-the-tenstorrent-grayskul…
Pet_Ant · 2024-03-07
I mean you could paste into ChatGPT:
> Please summarise https://morethanmoore.substack.com/p/the-race-to-2nm-risc-v-chips-in-j…
But this is my _personal_ takeaway: there are companies building (or trying to) Risc-V cores at advanced geometries. That is somewhat unheard of for a rather new and not-yet popular core. This suggests a lot of momentum for it. It's actually a new Japanese company called "Rapidus" that has some major backers and it's working with Tenstorrent (which has had a lot of buzz over the last couple of years). So it's exciting because a new company is entering the ring, from a new country, and they are benefitting Risc-V which is gaining traction.
There is another article on this here: https://www.anandtech.com/show/21281/tenstorrent-licenses-risc-vcpu-ip…
dkjaudyeqooe · 2023-10-25
So in reality they let go about 20% of their workforce. This seems to be roughly inline with what's happening elsewhere in the industry.
I suspect they've realised (like others) that funding is going to be much tighter going forward, or at least valuations less attractive, so they're adjusting their burn rate accordingly.
This seems a lot less apocolyptic than what < https://morethanmoore.substack.com/p/the-risk-of-risc-v-whats-going-on > would have you believe.
brucethemoose2 · 2023-03-23
He was just interviewed a month ago!
https://morethanmoore.substack.com/p/interview-with-jim-keller-tenstor…
If he quit Tenstorrent, that must have been very recent.
DeepYogurt · 2023-03-23
cuLitho isn't an AI process. Ian Cutress covered that in his article
https://morethanmoore.substack.com/p/nvidia-enabling-computational-lit…
> Of course when speaking about NV, it’s hard not to mention machine learning. Currently cuLitho is purely a mathematical solution tool, and not a machine learning accelerated tool. The engineering team we spoke to, when asked about adding ML tools to something like this, had a gleam in their eyes and essentially said ‘that would be the dream’. So not yet, but perhaps in the future.
pella · 2024-03-10
As I see this is : TT-Metalium [1] = a low level API [1]
"TT-Metalium is a platform for programming heterogeneous collection of CPUs (host processors) and Tenstorrent acceleration devices, composed of many RISC-V processors. Target users of TT-Metal are expert parallel programmers wanting to write parallel and efficient code with full access to the Tensix hardware via low-level kernel APIs."
https://tenstorrent-metal.github.io/tt-metal/latest/tt_metal/programmi…
[1]
"The software stacks come in two varieties – a high level and a low level. The high-level is called TT-Buda, using higher-level APIs to get things up and running, along with interfaces into modern machine learning frameworks. The lower level is TT-Metalium, which provides fine-grained control over the hardware for custom operators, custom control, and even non-machine learning code. Tenstorrent states that there are no black boxes, no encrypted APIs, and no hidden functions." https://morethanmoore.substack.com/p/unboxing-the-tenstorrent-grayskul…
CaliforniaKarl · 2024-05-13
More context: This is related to today's release of the Spring Top 500 list (https://news.ycombinator.com/item?id=40346788). Aurora rated 1,012.00 PetaFLOPS/second Rmax, and is in 2nd place, behind Frontier.
In the November 2023 list, Aurora was also in second place, with an Rmax of 585.34 PetaFLOPS/second.
See https://www.top500.org/system/180183/ for the specs on Aurora, and https://www.top500.org/system/180047/ for the specs on Frontier.
See https://www.top500.org/project/top500_description/ and https://www.top500.org/project/linpack/ for a description of Rmax and the LINPACK benchmark, by which supercomputers are generally ranked. The Top 500 list only includes supercomputers that are able to run the LINPACK benchmark, and where the owner is willing to publish the results.
The jump in Aurora's Rmax scope is explained by Aurora's difficult birth. https://morethanmoore.substack.com/p/5-years-late-only-2 (published when the November 2023 list came out) has a good explanation of what's been going on.