I was wondering why I saw them for cheap on aliexpress…

And if you order XXXXL it might fit. A little tight, but bearable.

were*

As the article says, they most likely changed their mind, probably following quite a bit of background discussions and industry influence.

Interesting that t8i is both for post-training and inference.

Next: replaceable storage? Since flash-based storage is widely known as a consumable that tends to fail first.

You mean the kind of "first" that comes right after shattered screens, worn batteries, and loose charging ports? :P

I agree in spirit though: storage chips wearing out seems to be common from my limited experience and it would be good if you could solder on, or slide in, a new chip with some standard procedure

Screen, battery and charging port replacements are already routine at any repair shop, at least for the devices I buy.

Sure, but you said storage chips wearing out is the first thing that fails in a device

And battery replacements usually leave devices in a degraded state, not sure about screen or charging port, so it's still good that this becomes more accessible

Why wouldn't it move the needle? Less time spent, less frustration, more performance, more resources focused on the business?

What's the plan to migrate off of it ? https://www.atlassian.com/licensing/data-center-end-of-life#...

It might be a bit CPU and RAM starved… Which in theory should be OK, but in practice you'll find production workloads that struggle because of this. Just make sure whatever you want to run on this is indeed extremely GPU-bound, or you might have bad surprises later.

Thanks for the context. Since I'm not interested in betting, I had not clicked on the grey on white About link at the bottom, which says:

> All the trains, delays, and data on this app are real.But the money isn't – because for that I'd need to move to Malta. Or Cyprus. Or Schleswig-Holstein.

It would be fun if Google lost its months of edge in the LLM value race because it alienated early adopters paying $250/month by using a 0-strike system with no customer support.

Honestly, I think it was probably a few users abusing the system like crazy. I've been building with Gemini CLI the past few days and had an increasing amount of issues getting a request through.

The GH issue trackers were full of people bitching and moaning about it. I think it might be a worse thing to alienate your users who use your product in the intended way - through Google's tooling.

But I agree the 0 strike rule seems really excessive.

It is also a possible scenario that a single individual sets up 10+ AI Pro subscriptions to blast through tokens like crazy - not sure how the economics of the daily allowances compare to the API pricing here.

> Honestly, I think it was probably a few users abusing the system like crazy.

It's not unusual that 10% of users use up 80% of the capacity; first saw this when home internet started getting ubiquitous.

By dropping the problematic 10%, the remaining 90% are:

a. Much happier, and

b. Much more profitable for the provider.

Resulting in a sustainable service for those who don't abuse the hell out o the ToS/FUP.

> 880mm^2 die

That's a lot of surface, isn't it? As big an M1 Ultra (2x M1 Max at 432mm² on TSMC N5P), a bit bigger than an A100 (820mm² on TSMC N7) or H100 (814mm² on TSMC N5).

> The larger the die size, the lower the yield.

I wonder if that applies? What's the big deal if a few parameter have a few bit flips?

> I wonder if that applies? What's the big deal if a few parameter have a few bit flips?

We get into the sci-fi territory where a machine achieves sentience because it has all the right manufacturing defects.

Reminds me of this https://en.wikipedia.org/wiki/A_Logic_Named_Joe

Also see Adrian Thompson's Xilinx 6200 FPGA, programmed by a genetic algorithm that worked but exploited nuances unique to that specific physical chip, meaning the software couldn't be copied to another chip. https://news.ycombinator.com/item?id=43152877

I love that story.

2000s movie line territory:

> There have always been ghosts in the machine. Random segments of code, that have grouped together to form unexpected protocols.