Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel Arc Pro B70 Review

pugetsystems.com

Post reply on HN
Print view View original

121–130 of 131 posts

Re: Intel Arc Pro B70 Review

#121

Post by spookie »

storus wrote: post #57
2001zhaozhao wrote: post #43

There's a tradeoff between dense models and MoEs on memory usage vs. compute for the same quality. For example, Qwen3.5 27B and Qwen3.5 122B A10B have similar average performance across benchmarks. The 122B is much faster to run than the 27B (generates more tokens at the same compute). The 27B, on the other hand, uses ~4x less VRAM at low context lengths (less difference at high context lengths). Right now, different…

DGX Spark is at the compute level of 5070. Its main issue is low memory bandwidth, i.e. it has quite fast token prefill but awful token generation. Strix Halo is just slow on every metric and used to be a cheap way to get 128GB unified RAM (now its prices are comparable to DGX Spark).

Strix Halo TDP is significantly lower. Comparing apples to oranges, really.

Re: Intel Arc Pro B70 Review

#122

Post by storus »

tehologist wrote: post #115
storus wrote: post #57

Earlier quoted context omitted.

DGX Spark is at the compute level of 5070. Its main issue is low memory bandwidth, i.e. it has quite fast token prefill but awful token generation. Strix Halo is just slow on every metric and used to be a cheap way to get 128GB unified RAM (now its prices are comparable to DGX Spark).

I have one, this isn't true. The wattage of a 5070 is about 300. The spark entire unit runs at 200 watts max. In reality it runs like a rtx 5060 with lots of vram. Very good for training, okay for inferencing if you are running batch jobs and don't mind waiting.

DGX Spark has actually the same compute as 5070Ti but its slower RAM and TDP brings it down to the 5070 territory.

Re: Intel Arc Pro B70 Review

#123

Post by marshray »

lukan wrote: post #53
marshray wrote: post #22

Earlier quoted context omitted.

I don't know about LLMs, but I tried an Intel card when Ubuntu Wayland couldn't initialize a 2 year old Nvidia. It just works.

That is just Linux and politics. Linux wants to force vendors to open source theirs, Intel plays along, Nvidia as the market lead does not, so you have to use their proprietary one, which most distros do not ship by default.

[dead]

Re: Intel Arc Pro B70 Review

#124

Post by nirav72 »

rekoil wrote: post #116
greybcg wrote: post #108

Something that is also cool with these cards is proper SR-IOV without hassle. Arc pro cards make for nice graphical acceleration devices for vms. I know ai gets all the hype but I also appreciate being able to accelerate multiple workstations with a single gpu and still get decent frametimes.

I bought an Arc A770 expecting that this feature would materialise based on it being available on other contemporary Intel GPUs, but it never did. Does the B70 definitely support SR-IOV from day one?

The B50 certainly does. So Likely same for B70.

https://youtu.be/xii8bqmE6jk?si=EqpzZYA-z-46F697

Re: Intel Arc Pro B70 Review

#125

Post by MisterTea »

AlotOfReading wrote: post #79
varispeed wrote: post #59

Earlier quoted context omitted.

I still not see the point running these models. I say they produce plausible garbage, nowhere near quality of frontier models (when they work). Why can't Intel look beyond this nonsense state of affair and build something with 1TB of RAM or more? What I am trying to say, I am yet to see anything competitive in the market. Cards very much stalled in sub 100GB region and best corporations can do is throw something to r…

What's wrong with Grace Hopper if you want to throw buckets of local memory at a problem?

What does Admiral Hopper have to do with this?

Re: Intel Arc Pro B70 Review

#126

Post by kinow »

embedding-shape wrote: post #33
kinow wrote: post #25

For those that use Blender, in their section about Blender: > We hope that, in the future, there will be real options other than NVIDIA for GPU-based rendering, as it is an area where competition is nearly non-existent. And Checking opendata.blender.org, a NVIDIA GeForce RTX 4080 Laptop GPU scores 5301.8, while Intel Arc Pro B70 is still at 3824.64. So there is still a bit more to go before Intel GPUs perform close t…

Also the first section I jumped to :) To Intel's credit, seems they're slowly improving, the section starts with: > Over the last year or two, Intel has worked to deliver serious optimizations for and compatibility with Blender GPU rendering on its Arc GPUs. Although NVIDIA has long held an advantage in the application, our last time looking at Intel’s cards indicated ongoing improvements. This round of testing is no…

Yeah, I checked an Intel GPU some years ago and I think it was scoring near 1000 or below in Blender's open data. Glad it's slowly improving, although I have to check if the price is also increasing or not, although I suspect it must still be cheaper than the other options.

Re: Intel Arc Pro B70 Review

#127

Post by dwoldrich »

als0 wrote: post #96
dwoldrich wrote: post #65

Hi Intel, I'm itching to buy an Xe3P! Or, Nova Lake? Crescent Island? Celestial? Jaguar Shores? Whatever the hell you name it doesn't matter to me, I just want a workstation with one of them bad boys attached to 160GB of RAM for legit inference power! I've been saving my money not paying for Claude Code so I can run my own agentic coding setup at home on yours. Please don't charge too much for the workstation class c…

> Please don't charge too much for it Intel wouldn’t decide to do this even to save their own life

The whole rest of the industry seems blind to the possibility of excellent personal/private agentic coding. There is a chance Intel could capitalize on that and steal a ton of mindshare in a flash.

Maybe a slim chance based on past performance, but it's there.

Re: Intel Arc Pro B70 Review

#128

Post by als0 »

dwoldrich wrote: post #127
als0 wrote: post #96

Earlier quoted context omitted.

> Please don't charge too much for it Intel wouldn’t decide to do this even to save their own life

The whole rest of the industry seems blind to the possibility of excellent personal/private agentic coding. There is a chance Intel could capitalize on that and steal a ton of mindshare in a flash. Maybe a slim chance based on past performance, but it's there.

Intel's strategy has consistently been that they do not consider doing any kind of business unless they earn at least 25% margin on each sale.

Re: Intel Arc Pro B70 Review

#129

Post by to11mtm »

girvo wrote: post #39
arjie wrote: post #27

I was looking into this for LLMs but it's clearly a graphics-processing focused card. The memory bandwidth is too low for that much RAM to be useful in an LLM context. The 5090 I have has the same amount of RAM but far more bandwidth and that makes it much more useful.

Oh wow, I really would've expected higher memory bandwidth. That's only ~2-3x the little DGX Spark-alike I have to play with. Would've expected more.

It would have been nice but I'm not mad about it; this is what I've been asking for since the 16GB Arc cards from earlier gens were so cheap; i.e. okay now do 'cheap but enough memory to run fancy models'.

IMO if this would have come out a year or so earlier, I feel like it would have been more like 750$ due to DRAM pricing and thus more compelling... [0]

[0] Yes It's likely that companies had secured DRAM contracts before the launch but I am also willing to bet that they also priced the card's launch price with at least some padding for increases in mind...

Re: Intel Arc Pro B70 Review

#130

Post by luckydata »

mrweasel wrote: post #114
luckydata wrote: post #58

this review was essentially pointless, they reviewed the card for a ton of workloads nobody in their right mind would pick it for, and left out the only use case where it makes sense. great job?

How you would you know not to pick this card for these types of workloads, without benchmarking it?

let's benchmark a truck for doing track days. pointless.
Post reply on HN