fooblaster

A 5090 is likely 3x the cost of the steam machine. You are at the extreme high end of the gaming market here and not the target of the steam machine.

what's this for?

these are not remotely like anything Google uses in the datacenter, even a decade ago.

I have a lot of the details there. Suffice to say it's a nightmare:

https://www.google.com/url?sa=t&source=web&rct=j&opi=8997844...

AMD is likely to back away from this IP relatively soon.

The neural engine not having a native programming model makes it effectively a dead end for external model development. It seems like a legacy unit that was designed for cnns with limited receptive fields, and just isn't programmable enough to be useful for the total set of models and their operators available today.

I think I'd still rather have the hardware area put into tensor cores for the GPU instead of this unit that's only programmable with onnx.

MLX doesn't use the neural engine still right? I still wish they would abandon that unit and just center everything around metal and tensor units on the GPU.

I think the bigger deal for founders is that Nvidia can decide at a whim to deny you supply or, better yet give 100 billion dollars to your competitor.

POE is an quite expensive thing to implement on a board. Flyback transformers are essentially required to support the standard.

thanks

evidence?

The em4 does not cost 10000 dollars. It is going to sell for well under 1000$. Hesai's low end ATX is supposed going to be less than 300$, and the AT128 already sells for 400-500$.

Why ain't the first answer.. use a bazel provided tool chain instead of using system toolchains? This article is totally mad.

Why do all of you think prices haven't come down? I can buy an AT128 from hesai for a few hundred dollars in volume. It's higher performance than any spinning lidar I could buy in 2017.

Lidar is being manufactured in china in the volume of millions a year by robosense, Huawei, and hesai. Bom cost is on the order of a few hundred dollars - slightly more than automotive radar. The situation is a lot different in 2025 than in 2017.

The us provides has provided upkeep for pakistans f16s under strict supervision and according to some AI summary maintenance has been a huge issue for them that limited their effectiveness. So not sure the story you are painting is quite as rosy as the reality. Fighter jets are not easy to maintain, without a large domestic fighter jet industry.

Consider for example when one of the radar elements in the f35 burns out, among the thousands that are there. Where in Pakistan does on obtain custom GaN radar ASICs that integrate with the f35?

It actually requires substantially less maintenance time if you look at the equivalent force it replaces, for example the associated awacs capability needed for a sortie for previous gen fighters.

Honestly, it's just absurd to think that any jet fighter is somehow low maintenance. The issue here isn't the f35, it's the host country becoming a unreliable/hostile partner.

I think you are making the mistake of thinking that xilinx software can fix the programmability of their hardware. it cannot. If you have to solve a place and route problem or do timing closure in your software, you have made a design mistake in your hardware. You cannot design hardware such that a single FFT kernel takes 2 hours to compile and then fails, when nvcc takes 30 seconds and will always succeed. You have taken your software into the domain of RTL design. This is a result of the hardware design. Xilinx could have made their versal hardware a programmable parallel processor array that is cache coherent, where everyone has access to global memory. It fundamentally isn't that though. it's a bizarre data flow graph systemn that requires dma engines and a place and route, and a static fabric configuration. That's a fault of your hardware design!

the spending on gpus is completely unsustainable. It looks great for them, but not everyone is going to get a return on that investment.

Xilinx never had hardware that was even remotely capable of competing with Nvidia. So I don't think it's solely a software problem - they literally have never developed hardware that is programmable or general purpose enough. Even their versal hardware today is hideously difficult to program and has a very FPGA centric work flow.

My personal favorite is the Tesla apology bumper sticker. I want to signal I care, but I only care about 20$. Love the virtue signaling while driving round a 50k$ donation to Elon.

my experience with the mi300 does not mirror yours. If I have a complaint, it's that it's performance does not live up to expectations.

this hardware is ancient history. mi250 and mi300 are much better supported

They are. It's just not at the consumer hardware level.

Yes, the mi300x/mi250 are best supported as they directly compete with data center gpus from Nvidia which actually make money. Desktop is a rounding error by comparison.

My father died last week of cancer. I'm glad for the progress, but winning seems a bit much.

They reached orbit with new glenn in January. This just isn't true.

One of many: https://www.reuters.com/article/technology/researcher-in-fac...

see comments of hacker news threads: https://news.ycombinator.com/item?id=20800031

A clusterfuck generally implies a mishandled situation. The author is saying that there was intention behind the actions and that this outcome was the goal.

Where do you work that uses bluespec? Any open positions? Apologies for being forward.