CIRCLE WITH A DOT

@kunev This depends entirely on the efficiency of the pizza oven. When measuring the sustainability of such systems (and you must consider the entire system rather than components in isolation), it it’s important to remember that pizzas are fungible.

@BenAveling Easy is relative. When you double the number of transistors available for a design like the 8086, there are a load of things you can do with them that will have an immediate impact on performance for most workloads. The same doubling for a modern CPU will need to be mostly spent on clever structures for trying to keep execution units busy. Doubling the number of execution units would have almost no impact on performance.

@david_chisnall the obvious answer is "labor i like should not be replaceable but labor i dislike should be" which doesn't yield itself for philosophical fedi posts, just actions

@david_chisnall I'm seeing a lot of talk about reputation systems at the moment, applying to open source contributing and social media.Every time, I'm reminded of how awful it was getting started on Stack Overflow.I had an account for years before I ground through the painful process of building a reputation.I'm not surprised that they're dying, it's not just AI; if you build walls in front of new users they'll give up and go somewhere else.Much of my angst was that I'd put in the work elsewhere, but there was seemingly no means of transferring that reputation.But there will always be new people trying to start from scratch, and somehow we need to welcome them whilst keeping out the abusers.

@jmorris @bsdphk @cheri_alliance Yup. Microsoft also maintains builds for the Arty A7

@radhitya How are you monitoring memory usage?

@forestfoxx Fuzzing is great, but it needs to be usefully tied to coverage and that's tricky. A simple case of fuzzing a CPU, you can fairly trivially generate every 32-bit instruction and feed them through an RTL simulator. But that will mostly test the same things. You really want to test things like different pipelines with different timing with dependent instructions. Defining a coverage model that you can feed into a fuzzing tool and get useful output is tricky.That said, being able to just throw compute at the problem is a great way of increasing confidence.

@david_chisnall You're right, but there are good principled reasons for tackling both the supply and demand side of things. Allowing the government to skirt the intent of the 4th amendment by buying the stuff is also bad, including where it uses our own tax dollars against us.

@david_chisnall I think the key is to keep the timescale for your predictions marginally longer than each wave of hype. By the time you've been proved wrong the world has already moved on to the next "big thing" and nobody cares about your previous prediction.

@david_chisnall Ouch! I understood that reference

@NicelyManifest That entire series is great. The level design is much more linear (except 5, which actually has some big open spaces). Fun in split-screen co-op mode, which seems to be dying out.Also totally unrealistic in every way,

@datarama @paul_ipv6 @david_chisnall Another mom here. I think you have the right kind of bias. (I learned to code 50 years ago.)

Welcome to the Sovereignty Queue, another benefit brought to you by Brexit!

@dexter @9vlc @emaste would know, I think?

@laxorr3 @Cheeseness @cmccullough There are also a lot of reasons to avoid the project. Their anti-D&I stance was enough for me to avoid them.

@david_chisnall well done!

@david_chisnall I use more green and blue.

@armb No idea. I've also no idea if 747s ever came in a configuration with little cabins. From the planes I've been on, there isn't enough physical space for the things that were depicted.

@david_chisnall my first computer had all of 1kb of ram. The web browser I'm typing this on is currently using up 869376 ZX81s of memory!

@david_chisnall you'd think that none of these people know what the tradeoffs of an abstraction layer are.What you said... if this were easy formal verification would already be ubiquitous. The whole point of constantly advancing our programming languages is to have more refined specifications that allows everything to be reused more easily and more broadly.Simply using an LLM as an abstraction layer doesn't eliminate the chaos underneath it all. It just pushes it farther away. That can't see how that works out in the long run.If LLMs rule the world and nobody is guiding the development of underlying specifications, wouldn't we eventually end up with one of two extreme undesirable outcomes?1) A programming landscape that's entirely standards free.2) A programming landscape that's homogenized as diversity and critical edge cases get factored out by LLMs as undesirable.