My friend and colleague @leomironov went on Bloomberg’s Asia Centric pod to discuss the challenges of investing in China against the backdrop of the unfolding balance sheet recession of recent years. A great listen… https://t.co/CwLM8hPgBS
$NVDA claimed Vera superiority in recent benchmarks
But there are caveats
They compared Vera, a 2026 CPU, against $AMD’s EPYC 9575F, a 2024 CPU
Vera was 10% faster than AMD’s EPYC 9575F on the geometric mean of the tested workloads
However, the real competitor to Vera will be AMD’s Venice / Zen 6
Also, the benchmarks were limited. NVIDIA controlled the initial access, and the workload set was focused on areas closer to Vera’s intended use cases
These were NVIDIA-permitted benchmarks, power consumption could not be measured, and Phoronix explicitly said this was only a subset of its normal full testing process
Bullshit. The hardware was always excellent, and the software has improved to the point that the moat is irrelevant. It was inevitable that the crunch would be the base catalyst, we already knew that from the 2024 h100 crunch.
AMD stepped up to the plate. That’s the real story here.
This is so fake news. Again. They had poor initial metrics because their CI wasn’t working and a bunch of PRs to improve things. Now they are using that to make up that there was some giant improvement. Wtf.
@benitoz@deepseek_ai This has nothing to do with AMD and everything to do with InferenceX and their chart generation. But sure Ben, create as much uneducated narrative as you want.
https://t.co/uc0wiHUGOt
https://t.co/dlXa8taHPO
Artifical Garbage Item
ARM internal uarch team has been repeatedly humiliated by Apple/AMD/Intel for 15+ years.
Every time you hear "Nvidia Grace is so bad", think of ARM Neoverse V2. Nvidia restarted internal custom CPU uarch team for good reason.
@CarsonTalkMoney I am extremely bullish on AMD in the mid to long term but i dont expect AMD's share prices to rip in 1H 2026; broad mkt still expecting AMD to show them the money before they pull the trigger
@jukan05 I don’t know where you get the idea that this is an “industry first” when AMD has been doing this for years.
I guess sometimes you would like to think AMD doesn’t exist, but it does and it is coming in hot.
https://t.co/StLqAW1qi1
@canyoudugit8@jukan05 to be fair while AMD is the first to rollout 3.5D packaging and hybrid bonding with MI300 , Broadcom's claim is to be first in 3.5D Face-to-Face hybrid bonding (AMD uses Face - to - Back stacking.
Still i agree that AMD's achievemetns are ignored here!
Ever wonder why junior miners keep diluting shareholders? High-risk exploration means no revenue guarantees - just drilling into the unknown. Commodity hype is the lifeline for raises. Spot the winners before the next boom! ⛏️
Read on 🧾
https://t.co/3DgDXeAOc1
AI has been built on one vendor’s stack for too long.
AMD’s GPUs now offer state-of-the-art peak compute and memory bandwidth — but the lack of mature software / the “CUDA moat” keeps that power locked away. Time to break it and ride into our multi-silicon future. 🌊
It's been a blast working with the amazing @_williamhu, @Drewwad and team; we present HipKittens!
You’re just cherry-picking data to fit your confirmation bias. If you go to InferenceMAX dot ai and look at the @OpenAI GPT-OSS 120B model and change the Y-axis selector to “TCO per million tokens,” you’ll see that the MI355X performs competitively with the B200 across certain interactivity levels. There are examples in the data that show B200 is way better than MI355X on the current software stack. As we mentioned in the article, there’s a lot of nuance in the data, the world isn’t black and white.
"The problem, which isn’t getting enough attention on Wall Street ... " I think any semi-competent trader would take into account the interest rates, when looking at the curve shape. Calculating rates-adjusted backwardation can me done with a small f-la in Bloomberg or Excel -
https://t.co/KyYnwn4xtA
GS had some article with backwardation regressions on interest rates ...
Second - "Using the 10-year US Treasury yield as a proxy" - why? You are financing inventory for a few months max - why would a 10y rate be relevant? Short rate should be taken. Sure, it is higher now, than pre-covid but not critically higher and going down.
Third- it will all boil down to the size of the surplus. OECD inventories are not high so a moderate surplus could be partially looked through. If indeed a "cartoonish" surplus materialises, prices would react more.
PS Fwiw I was saying oil will be in a fairly large surplus in Q4 25-q4 26 since April, so I am not disputing the fact that the curve would be in contango and would not be surprised to see oil lower in Q4-Q2. But I think the scale of the effects above is overstated