Gabriele Svelto

Old school nerd
Knows things about computers that would drive you insane
Hacks on Firefox at Mozilla

web: https://setphaserstostun.org
github: https://github.com/gabrielesvelto
gpg: https://keys.openpgp.org/vks/v1/by-fingerprint/D71D73D63144D1AD45F71705763A5AD66BC5A398
pronouns: he/him

Public Key

npub1vh5wrsqqcwcmyujgpfdjh6qymt5wghvrge4tnjlaytwqw38mlpysj2r3th

NIP-05 Address

[email protected]

Profile Code

nprofile1qqsxt68pcqqv8vdjwfyq5ketaqzd468ytkp5v64ee07j9hq8gnalsjgpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dszdg03y

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub1vh5wrsqqcwcmyujgpfdjh6qymt5wghvrge4tnjlaytwqw38mlpysj2r3th

Show more details

Last Notes

2026-05-05 19:52:01 UTC

The fact that Google decided to dump a 4 GiB language model file on every Chrome installation is yet another sign of how the generative AI craze is unsustainable. Don't look at it from the side of users, look at it from Google side. Having every user download a 4 GiB monster which will need to be routinely updated is a significant cost. It takes a ton of bandwidth to do that, far more than Chrome updates consume. And yet they're doing it because they're desperate to externalize the cost of "AI".

2026-04-24 17:12:18 UTC

https://media.mas.to/media_attachments/files/116/460/794/799/509/668/original/c7a384434dba1279.jpg

2026-04-13 20:29:21 UTC

- reply

yeah, but we're talking resources here. How much fuzzing and analysis would a few billion $ buy? A few 10s of billions? Remember that the total capex behind these technologies over the past three years is now in the 13-digits range. Spend that money on anything and it will fly.

2026-04-13 19:26:31 UTC

- reply

isn't fuzzing a number game though? LLMs are fuzzers backed by billions, they'll absolutely find something, but so would everything else given the same resources and no restrain on how to spend them, no matter how wasteful.

2026-04-10 13:45:26 UTC

You should ask yourself, why are the vibe bros obsessed with coding now? Weren't they busy trying to put artists out of work? Writers? Translators? Well, it turns out that in practically every other sector gen AI has basically become a liability. You use it hoping to save time and money, it turns into a PR disaster, you get to redo the work again, you saved nothing, you won't try again. So that's why. They're running out of options to peddle their snake oil.
https://www.animenewsnetwork.com/news/2026-04-10/wit-studio-apologizes-for-using-generative-ai-in-opening-sequence-of-ascendance-of-a-bookworm-part-/.236271

2026-04-09 09:30:26 UTC

Surprising no-one providing not only legal access but also legal production of drugs makes crime go away.
https://www.dutchnews.nl/2026/04/one-year-of-legal-dutch-cannabis-with-no-signs-of-crime/

2026-04-05 05:14:34 UTC

This reminded me of something my father once told me about space missions. He was on the ground staff for two Space Shuttle missions and he mentioned that any realistic depiction of mission control would involve people eating snacks all the time because - when things went well - there was really not much else to do but sit around and wait.
https://arstechnica.com/space/2026/04/artemis-ii-is-going-so-well-that-were-left-to-talk-about-frozen-urine/

2026-03-28 13:07:01 UTC

We're all wondering when this bubble will burst, but what if it already has? The flash of a distant explosion might not look like much, but only until the shockwave reaches you
https://media.mas.to/media_attachments/files/116/303/349/202/130/128/original/54a4bdd139f6a9bc.png

2026-03-25 16:29:51 UTC

Given what it's gonna do to the world economy when the bubble will burst I think it's better if we drop all references to intelligence and just start calling it subprime computing

2026-03-24 15:38:03 UTC

Check the benchmarks that can't scale to tons of cores or can't leverage dedicated hardware and you'll see that they show only a ~10x speedup. 18 years, two order of magnitudes in transistor counts, hundreds of billions in hardware development and the best we can do for user-facing workloads is an order of magnitude speedup. The tech sector needs to get its act together on software quality and optimization because frankly we suck.
https://www.phoronix.com/review/intel-penryn-to-panther-lake

2026-03-17 19:56:13 UTC

Technological progress is amazing these days
#DLSS5
https://media.mas.to/media_attachments/files/116/246/263/856/279/647/original/3e0de70b0831000d.png

2026-03-09 20:23:11 UTC

- reply

https://media.mas.to/media_attachments/files/116/201/074/402/149/781/original/e7e7f58ea3fc1eca.jpg

2026-03-09 16:33:38 UTC

- reply

yes, but why? Which packages where taking so long? Firefox has almost 4 millions of lines of Rust and it takes only a few minutes to build them.

2026-03-09 16:20:55 UTC

- reply

why did the project take so long to build?

2026-03-09 11:38:14 UTC

I've seen people claiming - with a straight face - that mechanical refactoring is a good use-case for LLM-based tools. Well, sed was developed in 1974 and - according to Wikipedia - first shipped in UNIX version 7 in 1979. On modern machines it can process files at speeds of several GB/s and will not randomly introduce errors while processing them. It doesn't cost billions, a subscription or internet access. It's there on your machine, fully documented. What are we even talking about?

2026-03-05 10:52:11 UTC

Pedro Sánchez' refusal to bow to Trump's demands is proof that European politicians don't have to be spineless cowards, most of them simply chose to be

2026-03-04 16:06:30 UTC

- reply

And for the record I'm looking at this mostly on computers and phones, but this affects *every* device. Routers, printers, etc... you name it. That fancy ARM-based MacBook with RAM soldered on the CPU package? We've got plenty of crashes from those, good luck replacing that RAM without super-specialized equipment and an extraordinarily talented technician doing the job. 5/5

2026-03-04 16:05:18 UTC

- reply

And to reinforce this estimate I've looked at the numbers we got from the users who run the memory tester after having experienced a crash: for every two crashes we think are caused by a bit-flip the memory tester found one genuine hardware issue. Keep in mind that this is not doing an extensive test of all the machine's RAM, it only checks up to 1 GiB of memory and runs for no longer than 3 seconds... and it has found lots of real issues! 4/5

2026-03-04 16:04:38 UTC

- reply

In other words up to 10% of all the crashes Firefox users see are not software bugs, they're caused by hardware defects! If I subtract crashes that are caused by resource exhaustion (such as out-of-memory crashes) this number goes up to around 15%. This is a bit skewed because users with flaky hardware will crash more often than users with functioning machines, but even then this dwarfs all the previous estimates I saw regarding this problem. 3/5

2026-03-04 16:04:17 UTC

- reply

In the last week we received ~470000 crash reports, these do not represent all crashes because it's an opt-in system, the real number of crashes will be several times larger. Still, out of these ~25000 crashes have been detected as having a potential bit-flip. That's one crash every twenty potentially caused by bad/flaky memory, it's huge! And because it's a conservative heuristic we're underestimating the real number, it's probably going to be at least twice as much. 2/5

2026-03-04 16:03:58 UTC

A few years ago I designed a way to detect bit-flips in Firefox crash reports and last year we deployed an actual memory tester that runs on user machines after the browser crashes. Today I was looking at the data that comes out of these tests and now I'm 100% positive that the heuristic is sound and a lot of the crashes we see are from users with bad memory or similarly flaky hardware. Here's a few numbers to give you an idea of how large the problem is. 🧵 1/5

2026-03-04 10:41:19 UTC

- reply

the reason why the botlickers have become so loud lately is that it's their toy that's about to taken away from them. The AI industry is crumbling under a mountain of debt and gigantic losses. Hardware and now energy too just got a lot more expensive, making those gigantic losses even larger. The collapse will be absolutely spectacular and will hurt us too, but it is coming.

2026-02-13 10:18:07 UTC

When talking about the energy consumed by LLMs don't be fooled by arguments focusing solely on the direct power consumption of these models, because they are externalizing a lot of it. Browsers are now doing proof-of-work calculations to access many websites, because websites need to protect themselves from AI scrapers. That takes power! Let that sink in: every computer, tablet or phone on earth is now consuming more power *every time it accesses a webpage* because of "AI".

2026-02-12 21:02:24 UTC

- reply

absolutely, and it gives people the impression that they have failure modes, which they don't. Their output is text which they cannot verify, so whether the text is factually right or wrong is irrelevant. Both are valid and completely expected outputs.

2026-02-12 19:10:44 UTC

Don't anthropomorphize LLMs, language is important. Say "the bot generated some text" not "the AI replied". Use "this document contains machine-generated text" not "this work is AI-assisted". See how people squirm when you call their slop out this way.

2026-02-11 19:04:03 UTC

Toots don't go viral, they escape containment

2026-02-09 12:13:02 UTC

A new argument that is cropping up among LLM apologists is that when things go wrong it's not because of the tools but it's because of the people.
That's literally the core argument of pro-gun rights advocates in the US, the country with the highest number of gun-related deaths in the world.
If the tool's primary use is to cause other people harm then it's 100% the tool, not the people.

2026-02-01 14:35:08 UTC

RE: https://mstdn.social/@990000/115993630141888644
How did we end up in the timeline where instead of writing a note on a post-it and sticking it to the door one needs A HUNDRED THOUSAND GPUS WHICH COST A TRILLION DOLLARS AND BURN A TERAWATT OF POWER WHILE GIVING ZERO FUCKS ABOUT COPYRIGHT OR CONSENT?
#note1nfy…l3nf

2026-01-29 18:15:47 UTC

It's meme time
#meme #StarTrek #StarWars #Spaceballs
https://media.mas.to/media_attachments/files/115/979/741/626/703/815/original/bb2ece5adead7ea4.jpg

2026-01-26 11:47:41 UTC

RE: https://ec.social-network.europa.eu/@HennaVirkkunen/115961104718375855
Now remove all the accounts of EU institutions from X. They have no business being there.
#note1zvx…z5vk

2026-01-22 16:07:12 UTC

- reply

Bonus end-of-thread post: when you encounter these bugs try to cut the hardware designers some slack. They work on increasingly complex stuff, with increasingly pressing deadlines and under upper management who rarely understands what they're doing. Put the blame for these bugs where it's due: on executives that haven't allocated enough time, people and resources to make a quality product.

2026-01-22 16:05:08 UTC

- reply

All in all modern CPUs are beasts of tremendous complexity and bugs have become inevitable. I wish the industry would be spending more resources addressing them, improving design and testing before CPUs ship to users, but alas most of the tech sector seems more keen on playing with unreliable statistical toys rather than ensuring that the hardware users pay good money for works correctly. 31/31

2026-01-22 16:04:56 UTC

- reply

And remember there's a lot of variables involved: timing broadly depends on transistor sizing and wire resistance. Higher voltages improve transistor performance but increase power dissipation and thus temperature. Temperature increases resistance which decreases propagation speed in wires. It's a delicate dance to keep a dynamic equilibrium of optimal power consumption, adequate performance and reliability. 30/31

2026-01-22 16:04:50 UTC

- reply

When CPUs ship their most performance critical circuits are supposed to come with a certain timing slack that will compensate for this effect. Over time this timing slack gets smaller. If a CPU is already operating near the edge, aging might cut this slack all the way down to zero, causing the core to fail consistently. 29/31

2026-01-22 16:04:43 UTC

- reply

You might also remember that Raptor Lake CPU problems get worse over time. That's because circuits degrade, and applying the wrong voltage can cause them to degrade faster. Circuit degradation is a research field of its own, but its effects are broadly the same: resistance in wires go up, capacity of trench capacitors go down, etc… and the combined effect of these changes is that circuits get slower and need more voltage to operate at the same frequency. 28/31

2026-01-22 16:04:33 UTC

- reply

I can't be sure that this is exactly what's happening on Raptor Lake CPUs, it's just a theory. But a modern CPU core has millions upon millions of these types of circuits, and a timing issue in any of them can lead to these kinds of problems. And that's without saying that voltage delivery across a core is an exquisitely analog problem, with voltage fluctuations that might be caused by all sorts of events: instructions being executed, temperature, etc... 27/31

2026-01-22 16:04:20 UTC

- reply

This is a circuit with two sets of 8 wires that go into it, plus one wire to select which inputs will go to the output, and a single set of 8 wires going out. Depending on the value of the select signal you'll get one or the other set of inputs. Guess what happens if the select signal arrives too late, for example right after the end of the clock cycle? You get the wrong set of bits in the output. 26/31

2026-01-22 16:04:12 UTC

- reply

In Raptor Lake's case a very common pattern that me and others have noticed is that sometimes the wrong 8-bit value is delivered. This happens when reading 8-bit registers such as AH or AL, which are just slices of larger integer registers, and don't have dedicated physical storage. The operation that pulls out the higher or lower 8 bits of the last 16 bits of a regular register is usually done via a multiplexer or MUX. 25/31

2026-01-22 16:04:05 UTC

- reply

The speed at which signals propagate in circuits is proportional to how much voltage is being applied. In older CPUs this voltage was fixed, but in modern ones it changes thousands of times per second to save power. Providing just as little voltage needed for a certain clock frequency can dramatically reduce power consumption, but providing too little voltage may cause a signal to arrive late, or the wrong signal to reach the pipeline register, causing in turn a cascade of failures. 24/31

2026-01-22 16:03:56 UTC

- reply

When a clock cycle ends, all the signals resulting from a pipeline stage are stored in a pipeline register. A storage element - invisible to the user - that separates pipeline stages. So if a stage adds two numbers for example, the pipeline register will hold the result of this addition. The next cycle this result will be fed to the circuits that make up the next pipeline stage. If the result of the addition I mentioned is an address for example, then it might be used to access the cache. 23/31

2026-01-22 16:03:49 UTC

- reply

To understand how this works keep this in mind: the maximum frequency at which a CPU can operate is dictated by the longest path through the circuits that make up a pipeline stage. Signals propagating via wires and turning transistors on and off take time, and because modern circuit design is strictly synchronous, all the signals must reach the end of the stage before the end of a clock cycle. 22/31

2026-01-22 16:03:41 UTC

- reply

But this is just logic bugs and unfortunately there's a lot more than that nowadays. If you've followed the controversy around Intel's first-generation Raptor Lake CPUs you'll know that they had issues that would cause seemingly random failures to happen. These bugs were caused by the wrong voltage being provided to the core under certain conditions which in turn would often cause a race condition within certain circuits leading to the wrong results being delivered. 21/31

2026-01-22 16:03:34 UTC

- reply

However not all bugs can be fixed this way. Bugs within logic that sits on a critical path can rarely be fixed. Additionally some microcode fixes can only be made to work if the microcode is loaded at boot time, right when the CPU is initialized. If the updated microcode is loaded by the operating system it might be too late to reconfigure the core's operation, you'll need an updated UEFI firmware for some fixes to work. 20/31

2026-01-22 16:03:28 UTC

- reply

Microcode can also be used to work around conditions caused by data races, by injecting bubbles in the pipeline under certain conditions. If the execution of two back-to-back operations is known to cause a problem it might be possible to avoid it by delaying the execution of the second operation by one cycle, again trading performance for stability. 19/31

2026-01-22 16:03:21 UTC

- reply

When implementing a new core it is commonplace to implement new structures, and especially more aggressive performance features, in a way that makes it possible to disable them via microcode. This gives the design team the flexibility to ship a feature only if it's been proven to be reliable, or delay it for the next iteration. 18/31

2026-01-22 16:03:14 UTC

- reply

As an example microcode can be used to disable certain circuits. Imagine something like a loop buffer, a structure that captures decoded instructions and re-executes them in a loop bypassing instruction fetches. If it turns out to be buggy a microcode update might disable it entirely, effectively sacrificing an optimization for stability. 17/31

2026-01-22 16:03:07 UTC

- reply

Now you might wonder if these kinds of bugs can be fixed after the fact. Well, sometimes they can, sometimes they can't. CPUs are not purely hard-coded beasts, they rely on microcode for part of their operation. Traditionally microcode is a set of internal instructions that the CPU ran to execute external instructions. That's mostly not the case anymore, and modern microcode ships not only with implementations of complex instructions but also a significant amount of configuration. 16/31

2026-01-22 16:03:00 UTC

- reply

Reading the errata of any relatively recent CPU you will find the same wording applied to every known issue: "Under complex microarchitectural conditions...". That's hardwarese for "a state we had not anticipated we could end up in". Try looking it up yourself on an errata document such as this one: https://edc.intel.com/content/www/us/en/secure/design/confidential/products-and-solutions/processors-and-chipsets/tiger-lake/11th-generation-intel-core-processor-family-specification-update/errata-details/ 15/31

2026-01-22 16:02:53 UTC

- reply

In all these cases the likely culprit is a bug in the machinery that tracks the internal CPU state when an unlikely sequence of events happens: a rapid series of interrupt or context switches, execution timing of certain instructions while the processor leaves or enters a particular mode of execution. These are not unlikely software bugs where you missed checking a particular condition at a specific time, and most of the time it doesn't matter except for that one time when it does. 14/31

2026-01-22 16:02:45 UTC

- reply

Other bugs could have even worse effects, such as AMD's infamous Barcelona TLB bug which would put the core in a state from which recovery wasn't possible, effectively halting execution: https://arstechnica.com/gadgets/2007/12/linux-patch-sheds-light-on-amds-tlb-errata/ 13/31

2026-01-22 16:02:38 UTC

- reply

The instruction pointer associated with an instruction is corrupted, and what appears to be the currently executing instruction is not. You can tell because a load causes a store exception, or a jump causes an access exception: https://bugzilla.mozilla.org/show_bug.cgi?id=1820832 12/31

2026-01-22 16:02:31 UTC

- reply

The code expected a piece of data to be loaded from memory, but the load/store unit returned stale data from a previous fetch: https://bugzilla.mozilla.org/show_bug.cgi?id=1687914 11/31

2026-01-22 16:02:23 UTC

- reply

Here's a few examples I've encountered: the instruction pointer is fetched from the stack while returning from a function call but it appears wrong, possibly because the wrong instruction pointer was sent to the instruction fetch pipeline: https://bugzilla.mozilla.org/show_bug.cgi?id=1746270 10/31

2026-01-22 16:02:16 UTC

- reply

The execution of every single instruction can alter a significant chunk of this enormous amount of state and must do so reliably. But as I mentioned before it's impossible to test all possible combinations and some sequences might lead to inconsistent or corrupted state, which in turn will manifest itself as a software bug. 9/31

2026-01-22 16:02:08 UTC

- reply

And that's without mentioning the large amount of hidden state carried by a CPU purely for performance reasons: physical-to-memory address translations are done via tables stored in memory, but this data needs to be cached in a translation lookaside buffer inside the core. Cache lines can be shared by different cores and must track their state, are they owned by a core? Shared? Is the data dirty locally and needs to be fetched remotely? 8/31

2026-01-22 16:02:00 UTC

- reply

Instruction faults cannot be predicted either: if a memory access fails because it access a protected memory area the flow of instructions must be stopped, undoing everything that should have happened before the faulting instruction and steering the core towards executing code provided by the operating system for such case, giving the impression that execution stopped right there in a perfectly sequential way. 7/31

2026-01-22 16:01:48 UTC

- reply

Instruction addresses, operands, dependencies are tracked as the CPU fetches and executes a stream of instructions which might or might not have to be executed, depending on branch prediction. In the case of a misprediction the state of the "wrong" instructions needs to be discarded. Similarly instruction timing is not predictable anymore. Memory accesses can take anything from a few cycles to hundreds, and fetch their data through different structures both inside and outside of the core. 6/31

2026-01-22 16:01:35 UTC

- reply

Modern CPUs contain a tremendous amount of state that they need to track. Hundreds of instructions can be in-flight at any given moment, each of them operates on physical registers which are assigned just-in-time via a mechanism that maps the registers in the ISA to spare physical slots in very large banks. Each instruction is associated with a set of data that is entirely speculative for a very long time, including the instruction itself. 5/31

2026-01-22 16:01:27 UTC

- reply

In the early days of integrated CPUs this state was not only visible to the user, but corresponded physically to what was in the core. The registers corresponded to entries in an actual bank of SRAMs inside the core, the instruction pointer was a physical register that would be read every cycle to fetch instructions from memory. In today's CPUs all these things are merely abstractions and the underlying physical reality is dramatically more complex. 4/31

2026-01-22 16:01:17 UTC

- reply

Let's start with logic bugs. As you probably know a CPU has a certain amount of visible state: a set of registers that contain data and which is manipulated by the instructions, an instruction pointer pointing holding the address of the executing instructions, a set of special registers to alter the core's behavior, for example by changing how floating-point operations round their operands. 3/31

2026-01-22 16:01:05 UTC

- reply

The root of all these issues is fundamentally the same: complexity. Modern cores have become so complex that it's impossible to demonstrate at design time that they will work reliably under all possible conditions, and thoroughly testing them is also infeasible. In addition to ever increasing logic complexity the the conditions in which they operate have also changed: fixed voltages and frequencies are a thing of the past, complicating physical design. 2/31

2026-01-22 16:00:44 UTC

In the early days of personal computing CPU bugs were so rare as to be newsworthy. The infamous Pentium FDIV bug is remembered by many, and even earlier CPUs had their own issues (the 6502 comes to mind). Nowadays they've become so common that I encounter them routinely while triaging crash reports sent from Firefox users. Given the nature of CPUs you might wonder how these bugs arise, how they manifest and what can and can't be done about them. 🧵 1/31

2026-01-18 13:28:06 UTC

When talking about DRAM please don't use the term "shortage", call it what it is: price manipulation.
Shortages are when things you want to buy cannot be bought anywhere, but DDR5 memory kits are widely available. There's hundreds of different types in stock at online retailers. I've yet to see a kit that is actually out of stock. This is the tech industry attempt at keeping high margins at a time when computer sales were already predicted to be declining. Don't fall for it. It's a scam.

2026-01-05 10:13:13 UTC

«Please, pretty please, don't call it slop. Do you have any idea how much money we sunk on this thing? Can you imagine what kind of about-face I'll have to make if this thing tanks for good? Our investors are going to cry so hard when they figure out that we lit their money on fire. So please be kind to this faceless megacorp and don't call it slop.»
https://futurism.com/artificial-intelligence/microsoft-satya-nadella-ai-slop

2025-12-29 09:12:01 UTC

My mother-in-law deleted her Facebook account because she was pissed of all the spam, bots and scam accounts pestering her. The importance of non-tech users dropping commercial social networks cannot be underestimated.

2025-12-26 22:56:48 UTC

RE: https://fosstodon.org/@atoponce/115787010150256693
I swear I read "shitting velocity" instead of "shipping velocity" the first time around
#note1ezk…jxgv

2025-12-22 21:48:56 UTC

If your big ass AAA videogame studio is now scrambling to contain the gigantic PR damage caused by your potential use of genAI, please consider solving the problem at the root by not using genAI in the first place.

2025-12-21 20:55:33 UTC

just one more USB port bro. i promise bro just one more port and we'll plug all the dongles bro. just one more port. please just one more bro. one more port and we can fit all the peripherals bro. bro c'mon just one more port. bro bro please just one more
https://media.mas.to/media_attachments/files/115/759/531/693/120/725/original/109664fc90dccb28.webp

2025-12-15 21:39:50 UTC

https://media.mas.to/media_attachments/files/115/725/740/573/920/462/original/e57e63c0cf80d877.png

2025-12-01 13:17:13 UTC

The price increases on RAM and SSDs are out of this world. Possibly above the 2020 profiteering. Just look at the graph below showing an almost 800% jump that went down to only 400% with the "generous discounts" of this time of year.
There is no way "demand" justifies this. This is vendors agreeing on jacking up the prices because they hoped that AI or Windows 10 planned obsolescence would drive up demand and it didn't. So now they have to find another way to please investors and shareholders.
https://media.mas.to/media_attachments/files/115/644/484/602/114/579/original/67dfe2670a3feb27.png

2025-11-19 09:22:13 UTC

I've seen a lot of posts focusing on the specific issues that caused the recent Cloudflare incident, but no discourse about how these companies are being run.
We've known how to build completely reliable computer systems for decades. When things fall apart, it's not because we don't know how to do it, it's because cutting corners has become the norm.
https://media.mas.to/media_attachments/files/115/575/608/173/704/599/original/70fe846b6a920997.jpg

2025-11-05 22:03:47 UTC

- reply

A few pictures of this machine and a few more ways in which this is better than a consumer one:
* the keyboard is solid and comfortable
* you get a touchpad with 3 separate buttons and a trackpad
* the hinges are sturdy
* it's very easy to open up for servicing
* the cooling system is well done and doesn't blow hot air on the screen
* the bottom case is solid magnesium
* you get USB A & C, HDMI and Ethernet ports, an SD card reader and a headphone jack
* the camera has a lid to block it
https://media.mas.to/media_attachments/files/115/499/292/598/730/320/original/f4ed2e757305bd74.jpg
https://media.mas.to/media_attachments/files/115/499/292/934/601/494/original/df3d28bfbc992a87.jpg
https://media.mas.to/media_attachments/files/115/499/293/239/706/952/original/a5e1e4b824d22a8b.jpg
https://media.mas.to/media_attachments/files/115/499/293/658/861/534/original/2773099520739f2d.jpg

2025-11-04 12:51:40 UTC

A friend needed a new laptop and was about to buy some consumer-grade garbage for ~500-600€. Fortunately he asked me before committing to it, and we found a second-hand AMD-based Thinkpad T14 for 300€. It's stupidly powerful for what he does, has much better connectivity and a much sturdier case. It was in really good condition too.
Long story short, stay away from consumer laptops. There's second-hand business machines that are both better and cheaper.

2025-11-01 09:11:14 UTC

> To be clear, everybody is losing money on AI. Every single startup, every single hyperscaler, everybody who isn’t selling GPUs or servers with GPUs inside them is losing money on AI. No matter how many headlines or analyst emissions you consume, the reality is that big tech has sunk over half a trillion dollars into this bullshit for two or three years, and they are only losing money.
https://www.wheresyoured.at/big-tech-2tr/

2025-10-21 07:36:53 UTC

About that AWS outage, this is certainly a coincidence and you shouldn't be reading anything into it: https://blog.stackademic.com/aws-just-fired-40-of-its-devops-team-then-let-ai-take-their-jobs-d9db9d298bfa

2025-10-18 15:34:18 UTC

Dale Luck, one of the engineers who designed the Amiga, let me touch the Lorraine, the first working prototype.
I am a happy man.
When I come down from nerd heaven I'll make an extensive thread about it, with plenty of pictures and loads of alt text for those who cannot see them. I want to share this joy with everybody.
#Amiga #Amiga40
https://media.mas.to/media_attachments/files/115/395/870/738/969/638/original/4faed68e2a1f36b5.jpg

2025-10-16 19:43:57 UTC

Dominion Media Television dropped this absolute #StarTrek banger the other day and I couldn't not share it here.
https://media.mas.to/media_attachments/files/115/385/531/546/103/277/original/17d2b8908ee082d2.jpg

2025-10-14 08:26:37 UTC

This isn't a pyramid scheme, it's not even a pump-and-dump one. God only knows what it is. Describing the scam that is the AI bubble needs brand new words because nobody ever did something so outrageous in the past.
https://finance.yahoo.com/news/very-troubling-ais-self-investment-spree-sets-off-bubble-alarms-on-wall-street-160524518.html

2025-10-10 07:53:01 UTC

Important reminder: the end of a war does not make war crimes disappear nor absolves war criminals.

2025-10-09 20:27:13 UTC

This is a great explanation of what the front-lines look like in Ukraine, which is nothing like any other war. Worth a read because this is the type of stuff that the mainstream press usually presents in simplistic and misleading ways.
https://texty.org.ua/projects/116021/20-kilometers-of-the-gray-zone-the-front-line-has-become-blurred/

2025-10-05 13:55:04 UTC

If new people are coming go check out those #introduction posts and do some boosting. We need to build an ever wider, more inclusive network.
If there's something that sets the fediverse apart is that we can build authentic social relationships here. This isn't just a surrogate of old media with the illusion of social interaction, like on the commercial platforms. Here the social aspect is at the very center of our experience.

2025-10-02 15:57:39 UTC

Counterpoint: there are millions upon millions of old consoles readily available in the second hand market. They come with gigantic catalogs of old games, some of which all-time classics which you can sink hundreds of hours in. The games are also cheap even without considering the modern not-exactly legal clones, and they won't disappear from a "digital" catalog one day.
There is no need to buy a new console. Buy an old one and enjoy your time with it.
https://www.theverge.com/report/782532/game-console-price-hikes-xbox-ps5-switch

2025-09-18 15:05:32 UTC

- reply

I played it as a kid and I'd love to see it on a CRT again. It's hard to describe just how impressive that dithering looks like unless you see the final effect on a CRT. It gets blurred in ways that makes the image appear more detailed and sharper than it actually is, it's amazing.

2025-09-15 17:48:48 UTC

I'm convinced that in the history books we'll credit BitTorrent for having preserved a significant chunk of humanity's cultural production

2025-09-05 10:27:22 UTC

Is there a guide somewhere on how to properly upload a music album to the Internet Archive? I've read about the basic stuff but I'd like to also add more stuff besides the music and cover: a scan of the booklet/back/spine as well as stuff like CUE/TOC files. Basically everything that's needed to preserve it entirely.
#cd #preservation #DigitalPreservation

2025-09-02 16:47:40 UTC

One of the side-effects of the AI bubble is that we stopped talking about alternative solutions to computing problems. Everything needs to be solved with "AI" - meaning an LLM - and nobody stops and asks if there is an alternative. This drives me mad because LLMs are not only unreliable, but also shockingly inefficient. And I'm positive that for most tasks there *is* an alternative algorithmic solution that is more reliable and less expensive.

2025-08-21 16:23:41 UTC

I have not been waiting for a game release in a long, long time, because I usually don't have the time to play games the minute they drop. But I'll make an exception for Silksong. I'm genuinely excited about it. Hollow Knight has been one of the best platformers of this century and I can't wait to get more of it.
https://www.gog.com/news/hollow_knight_silksong_will_be_available_on_september_4th

2025-08-06 09:56:35 UTC

Dear EU commission, as you may have noticed bowing to Trump didn't work and instead emboldened him to double-down on his bullying. We thought that you would have learned this particular lesson after years of trying to appease Putin, but no, you still haven't. Dear EU commission, it's about time your members grew a fucking spine. Sincerely, a European citizen.

2025-08-01 14:06:17 UTC

#Firefox users with older computers (at least 7-10 years old), if you have a minute could you check something for me? Go into about:support and scroll down to the "Codec Support Information" table. I'd like to know if it shows hardware decoding as being supported for the HEVC codec (it shouldn't). Also let me know what OS are you on.

2025-07-31 16:16:40 UTC

We are modernizing the user interface
Pray we do not modernize it further

2025-07-24 13:37:53 UTC

- reply

entire data centers are being built using borrowed money. The loans for those data centers are being sold as part of retail financial products with the value of the data center (and the GPUs inside of it) as a collateral. When the bubble bursts those data centers and their contents will be worth very little. That's subprime mortgages but with data centers instead of houses. Given 10s if not 100s of billions are involved it's a massive financial time bomb.

2025-07-24 12:15:44 UTC

Slowly reading through Ed Zitron's last article on AI and there's one thing that stands out: it's clear that the AI bubble might send us straight into another 2007-style financial meltdown. When it happens it's crucial that we point the finger at those responsible for it, because - with most of the industry tied up into this madness - the scapegoating will be one for the ages.

2025-07-16 22:16:30 UTC

The situation in Syria is very complex, far more than the simplistic reporting you will find in the mainstream press. However if you're trying to understand why the Israel Defense Forces got involved then this is the crucial bit of information you need to be aware of.
https://media.mas.to/media_attachments/files/114/865/221/684/901/621/original/7151cba2e146f329.png

2025-07-07 17:34:33 UTC

If you have an Intel Raptor Lake system and you're in the northern hemisphere, chances are that your machine is crashing more often because of the summer heat. I know because I can literally see which EU countries have been affected by heat waves by looking at the locales of Firefox crash reports coming from Raptor Lake systems.

2025-06-25 22:14:36 UTC

As you probably heard animated PNG images (APNG) have been standardized! So here's a story on how I used an APNG file to cut a phone's power consumption by 10 times in a common scenario during the weeks leading up to the launch of Firefox OS 1.0. We'll talk about webpage layout, rendering, mobile hardware intricacies, driver bugs and of course APNG files. 🧵 1/16
https://social.linux.pizza/@knurd42/114743763415059021
#FirefoxOS #PNG

2025-06-25 07:05:09 UTC

- reply

one wonders how hollow are the lives of these people that they feel the need to fill them with so much vanity, at the cost of setting the world on fire in the process