Jensen Huang – Nvidia GTC 2025 Keynote • Hacker News Board

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#51

Post by KeplerBoy » Tue, Mar 18, 2025, 9:30 PM UTC

adrian_b wrote: post #48

KeplerBoy wrote: post #47

Earlier quoted context omitted.

It will be better than strix halo. Even the Orin AGX GPU is better than strix halo and that is Ampere.

Perhaps you are confusing Strix Halo with Strix Point. The GPU of Strix Halo is many times faster than the GPU of the biggest Orin AGX. The Strix Halo has 25% more "CUDA cores" (2560 vs. 2048), which work at a clock frequency that is more than double (2.9 GHz vs. 1.3 GHz) and which can have for some operations a double throughput even at the same clock frequency. The memory throughput is also higher by 25%. The GPU o…

You're right, my bad.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#52

Post by canucker2016 » Tue, Mar 18, 2025, 9:42 PM UTC

Gracana wrote: post #37

nialv7 wrote: post #34

DGX Spark only has 273 GB/s memory bandwidth?! Massive disappointment, this is just on par with Framework's desktop PC.

That's what everybody was predicting based on the memory technology and the pictures. I'm just glad to hear about DGX Station, though I doubt I'll be able to afford one.

theVerge article says the DGX Spark will cost US$3K, about the same ballpark range as the Framework desktop with the maxed out 128GB RAM option.

https://www.theverge.com/news/631957/nvidia-dgx-spark-statio...

10Gb vs 5Gb Ethernet

1 HDMI 2.1, 4 x USB TypeC vs 1 HDMI 2.1, 2 Displayport, 2 USB TypeC

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#53

Post by KeplerBoy » Tue, Mar 18, 2025, 9:44 PM UTC

adrian_b wrote: post #44

baobabKoodaa wrote: post #33

The video glitches out at 2:14:30 where Jensen would be introducing DGX Spark and Station. Does anyone have a link to a working video for that segment?

The specifications are at: https://www.nvidia.com/en-us/products/workstations/dgx-spark... DGX Spark has the same memory as AMD Strix Halo, a weaker CPU, but perhaps a stronger GPU, except that for now there is no data about the GPU, besides that it might be stronger for AI inference (only FP4 speed is given). For now it is not known whether for graphics it will be better than Strix Halo. While DGX Spark might be wea…

I'm somewhat surprised they are not explicitly mentioning the networking capabilities. There should be two QSFP(?) ports but they don't mention if their speeds and supported protocols (possibly not just Ethernet but also infiniband, but probably not).

Could be 2x200 Gbit/s, could also be much less.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#54

Post by InTheArena » Tue, Mar 18, 2025, 10:09 PM UTC

canucker2016 wrote: post #52

Gracana wrote: post #37

Earlier quoted context omitted.

That's what everybody was predicting based on the memory technology and the pictures. I'm just glad to hear about DGX Station, though I doubt I'll be able to afford one.

theVerge article says the DGX Spark will cost US$3K, about the same ballpark range as the Framework desktop with the maxed out 128GB RAM option. https://www.theverge.com/news/631957/nvidia-dgx-spark-statio... 10Gb vs 5Gb Ethernet 1 HDMI 2.1, 4 x USB TypeC vs 1 HDMI 2.1, 2 Displayport, 2 USB TypeC

The unified memory model w/ CUDA support should be the huge differentiator for this hardware.

I think a better comparison might be the Mac Studio ultra.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#55

Post by mepian » Tue, Mar 18, 2025, 10:10 PM UTC

comeonbro wrote: post #45

lousken wrote: post #41

I feel like I am seeing same stuff every year for like half a decade Self driving cars? Simulations? Robots? What's changed?

You seem to have somehow missed one of the most significant advancements in human history suddenly exploding out of the pages of science fiction and fundamentally altering our existence forever. Minor oversight.

Ah yes, a chat bot that is consistently failing to do such advanced tasks as counting letters in words is definitely life-changing, right next to the metaverse and NFTs.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#56

Post by davely » Tue, Mar 18, 2025, 10:22 PM UTC

incognito124 wrote: post #42

PeterStuer wrote: post #18

Earlier quoted context omitted.

Yep, all 10 they shipped sold instantly at 2x MSRP. Prices on 3090's (not a typo) are going up ffs, because there is nothing out there on the NVIDIA front.

My 3090 Ti appreciated in price for about 20% in 2 years (1300->1600) nVidia GPUs have basically been an inflation hedge for the past few years lol

This made me curious, so I just took a look on eBay.

My 4070 Ti that I bought (new, oops) this past December has appreciated 20 - 25%! At least according to what people have them listed for... no idea if anyone is actually buying them at those prices.

Of course even if I managed to sell it, everything else has gone up in that time, so it's not like I'd get to make money on the deal. Pretty wild, nonetheless!

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#57

Post by nialv7 » Tue, Mar 18, 2025, 10:24 PM UTC

Gracana wrote: post #37

nialv7 wrote: post #34

DGX Spark only has 273 GB/s memory bandwidth?! Massive disappointment, this is just on par with Framework's desktop PC.

That's what everybody was predicting based on the memory technology and the pictures. I'm just glad to hear about DGX Station, though I doubt I'll be able to afford one.

I think most people (including me) were expecting 384-bit and ~400G/s. Can't believe they went with 256.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#58

Post by davely » Tue, Mar 18, 2025, 10:25 PM UTC

mepian wrote: post #55

comeonbro wrote: post #45

Earlier quoted context omitted.

You seem to have somehow missed one of the most significant advancements in human history suddenly exploding out of the pages of science fiction and fundamentally altering our existence forever. Minor oversight.

Ah yes, a chat bot that is consistently failing to do such advanced tasks as counting letters in words is definitely life-changing, right next to the metaverse and NFTs.

Pointing out this one error is reductive and greatly simplifies the profound effect that LLMs (and research around them) are having across a wide range of industries, unlike the metaverse and NFTs.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#59

Post by SV_BubbleTime » Tue, Mar 18, 2025, 10:39 PM UTC

codedokode wrote: post #43

SV_BubbleTime wrote: post #12

Earlier quoted context omitted.

AMD’s is coming up in support for things like torch and xformers. But it isn’t there yet. If you want for AI, your best bet is a used 3090 for $700-800. The ram is more important and that card is still faster than a 4070.

AMD has documentation for its assembly language as I remember. Or one needs more documentation to run AI?

Isn’t a documentation issue.

The entire ecosystem is built around CUDA currently. There was someone implementing Cuda on AMD hardware but as I understand it, AMD shut that down.

There are two alternatives to CUDA for AMD right now, but I forget their names.

AMD has STRIX coming, which is a single board embedded ram machine, and is being targeted at AI loads with lots of low latency ram. So we’ll see what happens there.

Re: Jensen Huang – Nvidia GTC 2025 Keynote

#60

Post by numba888 » Tue, Mar 18, 2025, 10:54 PM UTC

Gracana wrote: post #37

nialv7 wrote: post #34

DGX Spark only has 273 GB/s memory bandwidth?! Massive disappointment, this is just on par with Framework's desktop PC.

That's what everybody was predicting based on the memory technology and the pictures. I'm just glad to hear about DGX Station, though I doubt I'll be able to afford one.

> glad to hear about DGX Station

how much is it expected, my guess barely fits in 5 digits. Would be nice to have something in between Spark and Station. I.e. some desktop withing $20K.