[dead]

GN used Bloomberg clips of US Gov officials speaking on AI chip matters, fully under fair use.

And Bloomberg did a DMCA takedown through youtube, copystrike in parlance which pulled the video down for a week. GN had no recourse other than to wait and counterclaim.

Week timed out, Bloomberg did nothing but be the bully.

Louis Rossmann's excellent explainer video here on the Bloomberg bit: https://www.youtube.com/watch?v=6RJvrTC6oTI

[flagged]

Pretty much the same experience (on a much smaller scale). And just open up one of their servers and compare the engineering to a Dell or HPE server. Anything that can be cheaped out is. Corrugated plastic for cooling air channels, FRU assemblies held in place with sheet metal screws, all very bargin basement.

I haven't worked with anything at that scale, but the little bit that I was SuperMicro adjacent I was always unimpressed by the "fit and finish" of the entire experience, as compared to Dell and HP. (Having said that, the entire x86 commodity server experience is shitty anyway. I had a brief time, early in my career, when I did work with DEC Alpha machines. Man, they had their shit together. Stuff was expensive as sin, but stuff worked together and worked well. Build quality was tank-like.)

Pretty much. But at one point you could buy 2 to 3 units to every equivalent Dell or HP unit unless you had enough scale to get volume discounts. At $30M I expect the price to be a lot closer though.

Then it’s a matter of how well your engineering/ops org is setup to deal with silly hardware issues and annoyances. Some orgs will burn dozens of hours on a random failure, some will burn an hour or treat the entire server as disposable due to aforementioned cost differences. If you are not built to run on cheaply engineered gear that has lots of “quality of life” sharp edges (including actual physical sharp edges!) then you are gonna have a bad time. Silly things like rack rails sucking will bite you and run up the costs far more than anyone would expect unless you have experience to predict and plan for such things beforehand.

Of course you do have the risk of a totally shit batch or model of server where all that goes out the window. I got particularly burned by some of their high density blade servers, where it was a similar story to yours. Total loss in the 7 figures on that one!

Totally agreed on their BMC/firmware department. Flashbacks to hours of calls with them trying to explain the basics. My favorite story from that group is arguing with them over what a UUID is - they thought it was just a randomly generated string. Worked until one didn’t pass parsing on some obscure deeply buried library and caused mysterious automation failures due to being keyed against chassis UUID… and that’s when they’d actually burn one into firmware in the first place.

It was also always a tradeoff of having to deal with cheaped out hardware engineering with supermicro or with some horrible enterprise quarterly numbers driven sales process with Dell.

what chaff? Just search, find what you want and buy. It's like ebay.

And ideal effective market must have a zero margins. That's normal, what the economy strives for, what customers want.

If some market has large margins, it means it has some inefficiencies.

> Remember the 2018 accusations of spy chips implanted in supermicro motherboards that everyone denied so strongly

It'd be easy to prove the existence of a pervasive "spy-chip" problem using a camera or a microscope. Unsurprisingly, neither Bloomberg nor it's quoted "experts" ever managed to do so, deapite loudly banging that drum.

This news doesn't magically make those 2018 accusations true.

For fun, I will sometimes buy trivial positions in solid companies whose stock price falls 8-10% or so due to some minor temporary bad press and then resell in a month or two when the news cycle forgets about them and price rebounds. I make a decent amount of play money this way.

SMCI has a pattern of missteps over the years, I would not qualify them as a solid future bet.

(And in case someone asks the question, no that is not a viable long-term strategy one's retirement savings because it's very much speculating and doesn't work AT ALL when the market is volatile or falling as a whole.)

External factors can be a quick recovery. Internal factors are often a long road. Accounting and corruption failures sounds internal to me.

You could be right. But reading the comments here it seems it's had 2-3 scandals in the last 4 years, which makes me suspect that more could be brought to light.

You have to be greed motivated to become wealthy at all. People don't get there by being satisfied with a few million, there's a selection effect.

I wonder which of these the intelligence services prefer. Every one of them has their own advantages and drawbacks in terms of predictability, reliability, long term stability and chances of double dipping/playing both sides.

interesting insight

They definitely are using Nvidia. Part of deepseek's special sauce was using an "undocumented" ptx instruction to get a cute microoptimization with the memory hierarchy.

https://youtube.com/watch?v=iEda8_Mvvo4

They don't work. Chinese are skilled enough to desolder and smuggle just the ships themselves. They make the rest of GPU in-house. With more VRAM than the nvidia offers, comically, in case of 4090.

Of course they're using the best chips they can get, but this is a bizarre theory. English is the lingua franca of scientific publishing.

I'm kind of surprised by this comment.

I wonder if someone made a comment citing "fundamental cultural differences" with how Israeli people did business, it would be as well received.

From my experience, dealing with Israeli companies and Chinese companies are pretty much the same.

If you're so brave, you should state what these fundamental cultural differences are.

Some of the big LLM labs have written about their training hardware.

DeepSeek v3 was trained on 2,048 NVIDIA H800s. https://arxiv.org/abs/2412.19437

MiniMax M1 used 512 H800s. https://arxiv.org/abs/2506.13585

The H800 wasn't banned in the first round of export controls - but was after October 2023: https://www.cnbc.com/2023/10/17/us-bans-export-of-more-ai-ch...

Z.ai say they used Huawei hardware: https://www.theregister.com/2026/01/15/zhipu_glm_image_huawe...

Qwen and Kimi haven't disclosed their hardware as far as I can tell.

They have done a bit of this. SMIC is basically operating off of a cloned TSMC N7 node that they have since iterated on to get to a 5nm class node.

Exactly, you can build 12nm but you can't quadruple the speed of light

You can you just have to use a tiled architecture. And microprocessors already have far shorter wiring distances than the simple speed of light calculation because it takes time for the gates to make the transition as well.

With processors it's customary to use the "Fan out of 4" metric as a measurement of the critical paths. It's the notional display for a gate with fan out of 4, which is the typical case for moving between latches/registers. Microprocessor critical paths are usually on the scale of ~10 FO4.

The largest chip at the moment is Cerebras's wafer scale accelerator. There the tile is basically at the reticule limit, and they worked with TSMC to develop a method to wire across the gaps between reticules.

If you can make much cheaper chips and use much cheaper energy it might not matter.

well, physics does work that way, depending on what you mean by performance. (in the sense that power is normally part of performance when we're talking about chips).

you could certainly use a larger process and clone chips at an area and power penalty. but area is the main factor in yield, and talking about power is really talking about "what's the highest clockrate can you can still cool".

so: a clone would work in physics, but it would be slow and hot and expensive (low yield). I think issues like propagation delay would be second- or third-order (the whole point of GPUs is to be latency-tolerant, after all).

Thank you stock astrologist

How did you learn algotrading?

Well, also had other pen testers come forward saying that they had found implants on supermicro servers and had talked to federal authorities who had said it was a known relatively large issue they were trying to get a handle on while keeping it under wraps.

And if it were posted to move the market, that would have been about the most cut and dry SEC violation possible, posted at a time when the federal government still enforced such things.

Whenever some soylent-drinking, impossible foods-eating dilettante says "debunked" I find myself not fully believing them. And Supermicro has always been sus. I can't believe people are only just now noticing.

Yeah my point was more,.. let's examine what led to this article being created. I.E. if the cofounder was up to something shady, maybe he also made that story up for some reason (buy more stock at a discount?). Just a baseless suspicion, I admit.

> Otherwise you could have situations like businessmen profiting by selling weapons to the enemy to kill his own countrymen

We do this already, though—we sell weapons to israel to kill americans living in palestine—Israel has certainly killed many more americans than Iran ever has. And yet, the sanctions are applied as if the situations were the opposite. Make it make sense!

This entire line of thinking just seems like delusion to comfort yourself for having to live under a shitty state.

It depends on what you consider confirmed. It was kind of corroborated, at least. There was a CEO of a hardware security firm that came forward after the original article. He claimed that his firm had actually found a hardware implant on a board during a security audit. It wasn't exactly as Bloomberg described, though.

His take was that it was very unlikely that it impacted exclusively Supermicro, though.

It was covered various places, including The Register https://www.theregister.com/2018/10/09/bloomberg_super_micro...

A supply chain attack similar to Supermicro's would be much more targeted and recalls with national security implications do get flagged via a separate chain.

What Bloomberg proposed - sniffing the TTL signal between BMC and boot ROM and flipping a few bits in transit - is far from science fiction. It would be easy to implement in the smallest of microcontrollers using just a few lines of code: a ring buffer to store the last N bits observed, and a trigger for output upon observing the desired bits. 256 bytes of ROM/SRAM would probably be plenty. Appropriately tiny microcontrollers can also power themselves parasitically from the signal voltage as https://en.wikipedia.org/wiki/1-Wire chips do. SMBus is clocked from 10khz to 1mhz, assuming that's what the ROM was hanging off of, which is comfortably within the nyquist limit on an 8 - 20mhz micro.

Something similar has been done in many video game console mod chips. IIRC, some of the mod chips manage it on an encrypted bus (which Bloomberg's claims do not require).

Here's one example of a mod chip for the PS1 which sniffs and modifies BIOS code in transit: https://github.com/kalymos/PsNee

"On PsNee, there are two separate mechanisms. One is the classic PS1 trick of watching the subchannel/Q data stream and injecting the SCEx symbols only when the drive is at the right place; the firmware literally tracks the read pattern with a hysteresis counter and then injects the authentication symbols on the fly. You can see the logic that watches the sector/subchannel pattern and then fires inject_SCEX(...) when the trigger condition is met.

PsNee also includes an optional PSone PAL BIOS patch mode which tells the installer to connect to the BIOS chip’s A18 and D2 pins, then waits for a specific A18 activity pattern and briefly drives D2 low for a few microseconds before releasing it back to high-impedance. That is not replacing the BIOS; it is timing a very short intervention onto the ROM data bus during fetch."

Nope. Bloomberg doubled down on it and even Bruce Schneider accepted it despite initially being a skeptic.

I'd like to think that modern centers are water cooled so it'd be more quiet these days unless you are implying that this application of theirs is running on legacy hardware? :P