* Posts by user555

40 publicly visible posts • joined 29 Apr 2023

Day after nuclear power vow, Meta announces largest-ever datacenter powered by fossil fuels

user555

Thre is a fix for this, and for everything else too

Ban user tracking! Without the huge data gathering of user behaviour these companies will stop spending on AI.

Boeing busted by employee over plans to surveil workers, quickly reverses course

user555

"... a system for managing energy and space usage in selected office areas ..."

The bullshit alerts are firing today. There is super cheap motion sensors plentifully used already for, I don't know, every automatic door opener ever made. Don't need a zillion cameras and hulking "AI" processors for that!

Fission impossible? Meta wants up to 4GW of American atomic power for AI

user555

LMRs maybe?

Maybe those being built in Canada should be called Large Modular Reactors then.

Ten years under Dr Su: How AMD went from budget Intel alternative to x86 contender

user555

It isn't performance or efficiency that Nvidia has, it is the ecosystem built around CUDA. Just like Apple does with self-cultivated OS for iPhone and Mac. It isn't a performance game, it's the software platform lock-in. M$ does the same with Windoze/MS Office where it can.

BASIC co-creator Thomas Kurtz hits END at 96

user555

Forth is just cryptic

If you don't have the workings of the dictionary memorised, you're lost. Assembly is a breeze in comparison.

Mozilla's Firefox browser turns 20. Does it still matter?

user555

Ad industry and privacy won't improve without regulations

The idea that Mozilla, or anyone other than law makers, could move the needle on privacy with ads is a complete dream. The very reason why traditional formats (newspapers/network TV) are all dying is simply because there is a competitive advantage when privacy is violated. It's not a level playing field without regulations to make it level.

Europe's largest local authority slammed for 'poorest' ERP rollout ever

user555

Re: An orderly transition would seem sensible

Given the actual outcome, using such excuses would all sound like upfront laziness with ongoing incompetence ... or outright lying to cover corruption.

user555

An orderly transition would seem sensible

It always surprises me in these type of operations that the old systems are just flat abandoned without any attempt to prove the new system first. I guess these are ample proof of the old adage, “Those who do not learn history are doomed to repeat it.”

The Register takes AMD's Ryzen 9800X3D for a spin

user555

Re: Sold out

It's called a paper launch. Stock is still to arrive at the shops. Intel is doing the same. I'm seeing stock arriving in a week.

Cast a hex on ChatGPT to trick the AI into writing exploit code

user555

Re: Squirrel?

Or the wasteful and useless LLM crap gets the plug pulled instead.

Energy exec punts datacenter power options out to long term

user555

Re: I'm all in on datacenter nukes. . .

The only people seeming to be giving a time estimate for fusion is all the "perpetually 20 years" doubters. I don't see news pieces setting any dates.

user555
Thumb Up

Flawless!

CISA boss: Makers of insecure software must stop enabling today's cyber villains

user555

The comparisons being made

are against engineering for safety. Where other engineering disciplines have had heavy regulation for many decades, if not centuries, those were for safety rather than security. All the bleating that somehow software engineering has had is easy by not being punished for insecure code is missing the fact that security is not safety, and safety is not security.

user555

Re: None Of It True

And none of those are for safety reasons.

user555

Safety is not security

They are different terms for different purposes. Safety is the protection against unintentional harm to humans. Security is the protection against intentional harm to humans. Engineering for safety is the norm in many practices. Engineering for security is not the norm at all. If someone wants to throw another person off a bridge then there is no substantial protection against it. Just some minor safety barriers for accident reduction.

Lawyers need reminded of this distinction sometimes too.

Intel has officially entered the grin and bear it phase of its recovery

user555

Buying every High-NA machine

without any of them earning an income probably hasn't helped the finances.

AGI is on clients' radar but far from reality, says Gartner

user555

IBM Watson anyone?

IBM went through this on a smaller scale well over a decade ago. Goggle has been quietly pouring money into the tech for a similar amount of time. Why anyone thought AI was magically so much better now, surprised me a lot.

What's up with Mozilla buying ad firm Anonym? It's all about 'privacy-centric advertising'

user555

Re: If it looks like bullshit and smells like bullshit

The networks are doing it for one simple reason - It's a competitive advantage over traditional ad methods. This in turn is wiping out the competition.

What is really needed is the top to bottom banning of tracking entirely. This then puts them back on a level playing field.

Defiant Microsoft pushes ahead with controversial Recall – tho as an opt-in

user555

Re: Camel and straw

Corporates will be using this to monitor employees. There'll be more "managers" than ever!

OpenAI co-founder to depart ChatGPT

user555

Re: Tulip mania

Not yet. It's a big-ass bubble for sure but the money is very much still flowing.

user555

Speaks nice, presuambly gets his golden handskake

Are real safeties actually a thing? Easy enough to fool from reports I've heard. Ad-hoc would be the term.

I'm thinking the real story more likely a classified military one.

Aurora breaks the exaFLOPS barrier but falls short of the final Frontier once again

user555

Kudos to Tobias Mann

Cheers for the informative news piece.

ASML profits plunge 40% amid dip in chipmaking tool orders

user555

Re: Kind of predictable

It's not the bureaucrats making all the fuss, nor the decisions. All they do is carry out the orders/laws handed down from on high.

ASML ships another high NA EUV lithography machine to mystery client

user555

Three was an IBM linked announcement from last year, of a private/public partnership at a NY State research facility, posted on Anand indicating a likely candidate - https://research.ibm.com/blog/high-na-euv-lithography-albany

96% of US hospital websites share visitor info with Meta, Google, data brokers

user555

I avoid the worst by not owning a cellphone

If work assigns me one, I then use it for work alone. A prepaid burner is an option for travelling though.

Standardization could open door to third-party chiplets in AMD designs

user555

That was quite the segue.

ZenHammer comes down on AMD Zen 2 and 3 systems

user555

Re: Stay on topic

Just to be clear, this is a bigger problem than exploits. Corrupting of neighbouring rows is entirely possible in regular use. This is a general reliability issue.

user555

Stay on topic

You guys have wandered off into talking about the micro architecture vulnerabilities which the rowhammer exploits are nothing to do with. This is a DRAM problem that has possibly existed since the original invention of DRAM. Although I suspect it's only a recently exploitable problem due to tight cell density of modern DRAMs.

It will be solvable at the DRAM level. Just it's not as quick a fix as simply tweaking the timings ... and the industry probably still has to focus on the flaw seriously. The fact that one row can mess with its neighbouring row even while fully refreshed says there is need for design improvements in the DRAMs themselves.

It'll be a form of crosstalk. And that sucks because the way to deal with crosstalk in both cabling and board layout is with shielding. Shielding in a memory array is going to cost space, and that means reduced cell density. :(

Voyager 1 starts making sense again after months of babble

user555

Re: (Sighs)

Well, Boeing for sure.

Are you ready to back up your AI chatbot's promises? You'd better be

user555

Re: I'm not a Luddite, but

Fear-of-missing-out (FOMO)

Drowning in code: The ever-growing problem of ever-growing codebases

user555

Re: Optane is basically ROM

On the ROM side, "Mask ROM" would be the equivalent exception. Both SRAM and Mask ROM are integral to modern CPU fabrication.

user555

Re: Optane is basically ROM

Um, I made the point about SRAM because it's the sole exception by being made of the same transistors as the logic structures around it. That's why CPU caches are built with SRAM. All other RAM types, as you imply, are specially crafted cell structures.

user555

Re: Optane is basically ROM

Lol. I mean since a flip-flop is integral to compute designs, as soon as the IC came along it's natural to pack them into arrays for blocks of RAM.

My point there is SRAM isn't a special construction, it's built the same way as the logic circuits around it.

user555

Re: Optane is basically ROM

Yes, Core memory is a RAM. So that's now four RAMs in total - in chronological order: Core, SRAM, DRAM and MRAM.

PS: SRAM is a space-optimised array of flop-flops that naturally formed out of integrated circuits.

user555

Re: Optane is basically ROM

Everything is random access to varying degrees. Fixating on that one word is missing the point of what RAM really does.

What I was getting at is files are a placeholder for ROM too. It's a means to hold a non-volatile copy for indefinite time. Number of writes (endurance) is a bonus. And from there Optane fits the same definition.

MRAM is the only candidate that can bridge both worlds because it is also unlimited writes.

user555

Optane is basically ROM

Because Optane (Flash/ReRAM/whatever) wasn't unlimited writes it couldn't be classed as RAM. That means it's a form of ROM. Execute-in-place (XIP) like games cartridges.

Similar for file systems. They aren't XIP but rather loads into the temporary working RAM.

MRAM is probably the only non-volatile memory tech than can be classed as a RAM. But it can't compete with Flash on density. At least not yet.

Persistent memory to replace DRAM, but it could take a decade

user555

Re: Isn't MRAM wear resistant?

Yes, for this very reason MRAM is the only candidate for a DRAM replacement. None of the others can't last the distance. In fact MRAM is the only one in that group that can be classed as an actual RAM. The others are technically all ROMs.

The real significance of Apple's Macintosh

user555

Um, the relevant part is " ... they had a working GUI then so it wasn’t something written in the time since the Mac came out."

Millions of Gigabyte PC motherboards backdoored? What's the actual score?

user555

Re: You missed a question.

But M$ loves auto-run features.

Shocks from a hairy jumper crashed a PC, but the boss wouldn't believe it

user555

USB is easily locked up

USB is well known for its poor handling of static discharges, it can lock the whole chain of controllers and hubs totally. Not even a forced hard computer reset will recover it at times. Only way is to remove the computer's power. Some hubs are much better than others I've noticed.

I've had to add extra discharge protections (extra earthing of equipment and metalwork around it, shielding of cables) to equipment using USB devices to get some reliability.