Somewhat dubious claim if it was right or not, the only benefit was that asm.js was backwards compatible and set the stage for Mozilla to lose out by simply having the slower JS engine whereas NaCL/PNaCL proposal was "performance neutral" between browsers.

They used a lot of NASA resources to bootstrap, makes sense they later lobbied to destroy them once they could do the expensive testing on their own

O2 GPU was slower than other SGI options at the time, however it could use hilariously larger pool of memory without copying, which meant that O2 could use approaches that were punishingly hard (very tight transfer loops) or impossible (huge textures that couldn't be virtualized due to needing whole texture).

That was because unlike other GPUs at the time, O2's didn't have dedicated memory but shared the memory with CPU - way slower, but zero copies and bigger.

Arguably early home computers and workstations also used "unified memory" :D

Every AMD APU since introduction of HSA did it, which is how AMD ended up doing SoCs for PS4, PS5, and Xbox

Ok, so which one of these contemporary or previous chipsets could compete with the M1 for inference? Perhaps I'm missing some major detail.

The M1 isnt particularly good at inference, so pretty much every major current competitor with a 256+ bit unified memory system is better: AMD Strix Halo, NVIDIA DGX Spark, possibly Intel Panther Lake

Sure, but none of these shipped before the M1. That was the first chip I encountered that managed to do something useful without a discrete GPU.

Even in Apple land M1 isn't the first with unified memory - pretty much all intel on-chip GPU (Sandy Bridge and newer) count - it was even a reason for driver issues early in intel's new dedicated GPU lineup, as the drivers expected unified memory - but M1 is essentially modification of an iPad chip, and you can see "unified memory" there going all the way back to first chip Apple bought from Samsung to power iPhone 2G

You were likely only using Intel systems then. AMD systems have had iGPUs capable of light gaming for a very long time. It took Intel a long time to get to that level, after M1 (2020).

You are missing a major detail: integrated GPUs are crap. They win on efficiency but not on raw compute. Before AI (and crypto too, I guess) people bought GPUs to render graphics and that was their main consumer. People built more and more demanding games that required increasingly powerful GPUs to render well. Gaming systems always had a discrete GPU so there was no reason to scale up integrated GPUs because they wouldn't sell, or they would be a waste of die space.

I don't think the M1 specifically focused on inference. Their goal was to replace Intel/AMD/Nvidia with their own chips, and since the previous Macs shipped discrete GPUs, they had to match or beat those so they don't ship something slower.

SGI O2 was the famous "unified memory architecture" graphics system, two years before i740 that didn't really do UMA.

O2 was popular in systems where large textures or textures generated dynamically (like mapping external video input to texture) was important

And don't forget it's not really about chipset, it's about market segmentation and up selling you for setting a boolean flag in firmware.

Surprising amount of gear required extra licenses or even parts to have fully functional remote console that does all the stuff PiKVM 4 brings to the table, and not always do you have the benefit of properly operational netboot infrastructure (especially in corporate settings with their usually broken by design networks) - and remote boot media might be gated behind license.

So you end up with ipkvm because there might be windows server that requires remote display, or because for various reasons IPMI SOL or equivalent does not work properly (BIOS mode enabled instead of UEFI, SOL requiring extra license, etc). Or even not being able to setup autoconfiguration for BMC but for various reasons it's easier to setup ipkvm.

And so easy to make into bottleneck.

Yes, zygote pattern makes it easy to make fork() into bottleneck - it requires a lot more discipline and low level tricks (linker scripts, compiler-specific extensions, custom sections, low level dependencies on pagesize that get "fun" on ARM servers).

If you don't, you might wake up with fork() causing latency issues.

You mean Chrome? (Looking at the 5GiB usage here...)

Google Chrome is using 5 GB of VRAM?! Are you running something with WebGPU or something?

It's called prerendering tabs or something :V

At 4k it adds up, especially since if you look into DevTools Chrome actually renders more at least vertically if not across major scroll regions.

Unfortunately, the manager pushing the outsourcing pockets the money, outsourcing company pockets the money, and the local and indian teams get to be abused and latter gets extra blame for things outside their control.

I would say the problem is heavily structural and linked to few companies that very much pursue lowest possible effort, harming both employee and customer. We had a company that infamously pursued a somewhat similar (just with not as much leverage over employees) strategy in Poland - a common refrain was how other companies would get people jumping jobs from them who gave variants of "I needed some spending money while finishing my degree, escaped as soon as I could" story.

Unfortunately it seems a bit harder to do when some places apparently hire with only category appearing to be "according to census they speak english", then put them on project with minimal training and zero time or space to acquire more training.

Yes - I remember one time hearing someone from a big consulting company make the argument that even the top Indian graduates would work for peanuts (even by local standards) and asking them how likely it was that they’d be that smart but unable to determine their market value.

This was unsurprising after my entire life hearing people complain about declining quality and imported junk while always picking the cheapest item on the shelf at the store, even when they had plenty of money for a better quality option. There’s just a certain mindset which can’t look past the lowest possible price.

Also sometimes you have few big companies dominate local market to detriment of choice, even if your plan is on getting further away.

The polish company I mentioned was infamous for using still in education students especially from an university where its founder was a professor, and for his motto of "you can replace any specialist with finite number of students, and the number is usually one". I met some of them later who worked there for first few years only to setup conditions to jump ship.

But along the line I heard how few big outsourcing/consulting companies dominated IT hiring in bunch of cities in Poland, with Warsaw being the odd one out because suddenly you had to fight for candidates with 100-150 person "minnows". I imagine without such minnows finding sensible work would be way harder