dagmx

This is incredibly sad for a lot of my friends who are finding themselves out of work despite delivering well received products.

But at the same time I appreciate the candor of Asha saying that the corporate management are to blame and letting studios go back to being independent where possible.

Phil Spencer really messed up. Everyone in the industry knew Microsoft were making bad calls trying to dig themselves a hole with gamepass and simultaneously digging a hole with their acquisition spree. I’m glad that Asha is laying this bare even though it sucks to be brought in as the hatchet person.

This is an example of the glass cliff and I’m hoping she can help right the ship. I think they need to split to a wholly owned subsidiary rather than be in Microsoft proper, and I expect that to be announced at the Q1 investor meetings.

Phil really dug their hole deep. Microsoft themselves encouraged it. It’s been a decade of sheer incompetence at the highest level so I’m hoping they can right this without taking out half the industry in their wake.

I found this paper quite disappointing.

It requires a fairly expensive precomputation pass and can only work for static scenes.

Meanwhile interactive path tracing is fast enough that the scenes they showed would only be minorly slower to be truly interactive with dynamic scenes.

I wish they’d showed this with scenes that don’t fit in GPU memory so it could show the benefits for CPU only renderers, otherwise GPU based renderers would be fairly fast with these scenes.

The only big thing for me was the multi view lighting. The painted light to light parameters is a neat trick but been done quite a few times in the past with traditional techniques too.

Anything AI focused in silicon is also valuable for a ton of other use cases. If LLMs and GenAI don’t pan out, that silicon just gets used for other processing. Then they scale back on the dedicated die space in subsequent generations.

It’s all fairly easy bets to make and correct.

Where would they make that which isn’t constrained already?

They’re just as ugly and huge as you spec them out to be. There are tons of ITX / SFF builds that look just as good or better.

Simply because nobody has successfully pitched the Unicode group to add one.

Anyone can pitch new emoji, they just have to go through the fairly easy but strict formal process.

I’m really shocked you can’t see the sheer hypocrisy on your part here again.

You have literally added nothing to the discussion, other than levying criticism and denigrating any points made just because you either don’t agree or don’t actually have anything material to add.

Anyway, it’s the internet. No point getting annoyed by drive by responses from a stranger. Enjoy your weekend.

There’s an irony in you claiming my comment is uncharitable and claiming the article is a discussion point when you haven’t even taken up the question I posited and framed them as a strawman.

Can it be possible to make everything during a user interaction perfect looking? What does that mean?

That was part of my original comment that you skimmed past rather than engaging with. So if you truly believe this should be a starting point of a discussion, then why not even bother taking that point to build a discussion point off of?

I’d also argue that this is not presented as just an idea. That is disingenuous. The author clearly writes “I call these situations “The technology has outsmarted the programmer” “

That is more than just a discussion point but a levied criticism at the skill of the people making things. In turn therefore they’ve set the framing of skill as part of the discussion and I don’t think it’s unfair to point out their own lack of ability to execute to their high standard.

But let’s go by your standard of saying that we shouldn’t take the authors own execution or their own words into consideration but only focus on the core idea they present.

Then that still leaves defining what they consider every frame being perfect. What are the bounds of that statement?

I have a lot of thoughts about this:

I think this is a weakly presented argument. The article doesn’t actually present stronger alternatives or even why anything shown is negative to the user. It might be negative but otherwise this is the same vacant critique that is levied by pointing at smear frame or transitional points in media to critique it.

The user also has an untenable maxim. Every frame must make sense? I would posit this is impossible, or I’d ask the author how they’d handle window resizing while keeping every frame perfect.

I also think the author themself finds it easier to point out flawed frames (again without actually explaining why they’re issues) than doing as they say. Tap the header links on their blog and see the animations play after the click is complete. Or go see their own UI projects and see how text and objects don’t stay within their containers. Surely someone saying that this is a tenet that should be followed could demonstrate it themselves.

I think this is just a very hollow critique on their end.

A more competently written article would have focused on why anything shown is bad for the end user, and how they might handle it instead. A good critique should actually include some substance and point to more than just the what, but the why and how.

Ironically the two studios you mentioned don’t actively render on GPUs and it’s an area which shows that even these small SoCs can punch way above their weight if you look at their pure compute power.

Disney’s Hyperion is CPU based and RenderMan XPU is just exiting beta after over a decade.

But while they do stack their workstations with higher end GPUs for artist throughput in viewports it’s mostly just for the higher memory to fit unoptimized scenes in. None of the studios or major films I’ve worked on have had their on desk artists be raster rate gated but just memory gated.

But again, besides the point, because it’s still valuable as a metric to compare with when comparing perf between similar chipsets.

There are already more creatives using their consumer grade hardware to make stuff. And even the studios you mentioned do actually use laptops on the go for parts of their creation pipelines for various things like virtual production scouting etc.

As a professional in the space, a ton of people DO care about mobile performance. If you’d go to SIGGRAPH in the last few years you’d see how the landscape has really changed.

It’s not going to be the primary place of creation but there’s a lot of usefulness in having a portable workstation or that entire segment of the laptop workspace wouldn’t exist.

In either case, it’s besides the point because the point is talking about the compute levels of a GPU in the same form factor.

What x86 chips have the same or higher number of cores in the form factors that these chips are available in and are also more performant?

Strix Halo is 16 cores. Intel Core Ultra 9 285HX is 24. Apple is 18. Qualcomm is something similar too but I can’t recall. NVIDIA is 20.

Until you get to threadripper/epyc or Xeon territories (completely different form factors and TDPs) the arm chips are ahead on both power and perf than the x86. And even when you get to those areas, arm is equivalent or out performs them as can be seen by the recent neoverse x3 and Vera benchmarks.

That’s only if you consider FP32 specifically. On average the M5 Max will pull ahead for tasks like GPU raytracing (it’s currently the fastest mobile GPU for Blender rendering) and token generation and other things that benefit from the higher memory bandwidth.

I’d also mention that you’re comparing peaks which the RTX Spark won’t be hitting. The top TDP is less than that of the DGX Spark.

I just think anyone calling this a beast and a game changer are conflating/extrapolating from different form factors and constraints

Ah I see. But the only C2C equivalent on the Apple side is the UltraFusion which is 2.5TB/s if I recall correctly.

This feels fluff to me on the part of the author (whose work I don’t want to trivialize) but I don’t think they’ve actually looked deeper than a paper spec sheet on this.

1. Yes it has the same number of cores as a 5070 mobile. It’s also running at a shared peak of 2/3 the bandwidth and a shared peak of 2/3 the TDP. The GPU by itself will likely perform at half the dedicated units performance

2. Apple may not have SVE2 but they do have the AMX (private) and SME. I don’t see why he thinks the SVE2 will give him more performance than the SME.

3. He mentions a single core type but doesn’t mention the total makeup. We already have known for a year how the DGX Spark compares to Apple chips. For CPU it’s roughly equivalent to an M3 Pro and for GPU compute (not rasterization) it’s between an M4 Pro and M4 Max without considering bandwidth.

The real advantage to these is that they run CUDA. That’s it. Otherwise when they launch they’ll be 2-3 generations behind where Apple is and 1 gen behind AMD.

The other super power of the DGX Spark was the NIC for pairing them together. But that’s been removed here too.

Where did you get those numbers from?

DGX Spark has a maximum of 273 GB/s bandwidth in ideal scenarios (hard to reach)

That puts it between an M5 (153) and M5 Pro (307)

Splats would work and it’s what a lot of automation folks use. But the risk is that your splats aren’t tight to the surface and that can cause false positives.

For training you can do a hybrid geometry plus splats workflow. Have geometry that you can constantly raycast against and have as an input to your vision training or to get accurate depth buffers.

The workflow for splats and photogrammetry are very similar.

I know this comes up a lot on HN because its not primarily a graphics community but:

1. Gaussian Splats are very expensive to render. They capture a lot of detail which makes them seem cheaper than an equivalent raster render of that quality, but they wouldn't meet real time AAA game performance requirements

2. Gaussian Splats don't have a concrete surface. Want to cast shadows or do physics? It's doable but very tricky. Want to relight them? Also tricky. What is the exact surface point that you want to affect or sample for any particular operation? Deformations also become very difficult to do well.

3. Gaussian Splats are not sharp. You can get sharper with different kernel types or higher density of points, but your costs go up as well.

4. Gaussian splats are awful for any kind of path tracing. You can do it but you go back to the issues above. So mixing and matching traditional content with splats becomes a performance bottleneck.

I don't think you'll see a AAA game use splats for more than something like cinematics in the near term.

Why would you limit one to your local hardware and one to a cloud infrastructure?

Both can be done locally or on cloud? the comparison point becomes moot if you change the parameters that drastically

I'm curious how the RTX Spark will perform and what it'll cost.

The big things it has going for it are CUDA and DX support...but performance wise this is pretty weak.

It's a 5070 (mobile) equivalent core set but in a <100W envelope for combined components, whereas the full 5070 mobile is more than that by itself.

It also has a very constrained bandwidth in comparison, and a lot of older ARM cores. NVIDIA and Microsoft better be able to be competitive on price because unless you specifically need CUDA or Windows, this just doesn't feel like a very competitive product to a Mac Studio?

Not traditional Gaussian splats since they have their lighting baked in but there are alternate forms that have the ability to do so but largely in academia or very specific use cases

Without some extra solving most full body IK systems won’t give you walking motion. It’ll just be like pulling a doll by its limb or QWOP

You need a proper motion controller on top to actually have realistic walking motion driven by steps.

It means that you can lock the end point to a position and figure out the rest of the pose.

Imagine a character walking with forward kinematics. Every time you move the characters hips , you’d have to rotate the leg joints and make sure the foot doesn’t slide. Remember virtual characters don’t have friction.

IK lets you lock the foot in a spot so you can animate the body above it without having to spend time matching the foot position.

In a real world, this is like if you tried to put your hand on a door handle while jumping up and down. It’s easier to keep position when you hold the handle than if you were to just touch it.

Brain Control Interface support was already announced last year and afaik is part of iOS already.

https://developer.apple.com/documentation/accessibility/brai...

It’s not trying to be a gotcha, it’s pointing out the sheer arrogance to criticize someone for your own decision to take a commonly used metaphor literally, and then to go on and act high and mighty because you’ve decided to be the outlier in society who’s decided that you get to pick and choose language norms at your whim.

But you do you. I’m sure it gives some self satisfaction to pick pseudo intellectual arguments. There’s no point continuing this discussion because it’s clearly been in bad faith from the get go.

You must be a hoot at parties. Getting to pick and choose what you and everyone else around you should care about and how they define terms to fit exactly what’s important to just you.

Orthography may be independent of language, but shockingly in society, it’s part of written language.

You know many words can be used colloquially and not literally? Language is contextual and ever evolving.

I’d also be hesitant to call people out for sloppy language when you can’t be bothered to even capitalize stuff.

Nobody is moving the goalposts. You just poorly choose words.

I don’t think you actually understand the problem space of what you’re describing or you wouldn’t call it DRM.

And as such I don’t think you actually have a grasp of what is affected. But you’ve deles that everyone who doesn’t think it’s the same level of issue as you is somehow beneath your intellect based on your other comments.

What do you think that does with regards to DRM? And do you really think most people are running that to use their Mac?

Could you explain what CLI you think needs to be used to install software on a Mac?