Yes, but bigger models are still more capable. Models shrinking (iso-performance) just means that people will train and use more capable models with a longer context.
Well, to be fair, the authors propose this thesis: "Although the vectorization of Verilog designs does not change the hardware they describe, it reduces their symbolic complexity, enabling faster and more scalable analysis and verification."
Maybe it doesn't help Design Compiler turn your shitty design into gold, but faster verification is an unalloyed good.
> Hierarchical physical design tends to be worse than flat PD because there are many variables to optimize (placement density, congestion, IR drop, thermal, parasitics, signal integrity, di/dt, ...) and even if you have some solution in mind that optimizes area for a highly regular block, that layout could be worse than a solution that intersperses lower-power cells throughout that regular logic to reduce hotspots.
This paragraph goes hard. And this is exactly why design space exploration is essential. I think you're right that basically, simplistic delay/area models are insufficient and the exploration must be driven by actual metrics of complete P&R flows.
> [...] it's a pretty challenging problem to design a formal specification language that is simultaneously high level enough and yet allows a compiler to do a good job of finding the optimal chip design
My experience in this domain is that actually the challenging problem isn't so much the design of a formal language. Instead the challenge lies primarily in expressing your design in such a way that both generalizes over and meaningfully exposes the freedom in the design space.
> [...] solving the problem of "what is the most efficient way to approximate this algorithm with N% precision"
> [...] in many domains it's hard to formulate an error metric that isn't either too conservative or too permissive.
I think the latter comment alludes to my objection about the former: It all depends on what "N% precision" means.
Does it mean that for every input/output pair, the output is always within N% of the correct value?
Or does it mean that for N% of the inputs, the output is correct? Is that weighted by the likelihood distribution of getting those inputs?
Or does it mean that the total mean squared error over all input/output pairs is within N%? Etc. etc.
In other words, I think it goes beyond even conservative vs. permissive; simply, the devil is in the details, and digital circuit design is a difficult multi-objective optimization problem.
I know this is an unfathomable concept, but to actually "leave fossil fuels [...] in the ground" you have to stop using fossil fuels. Burning fossil fuels someone else refused to leave in the ground means--surprisingly--that fossil fuels weren't left in the ground after all.
And it turns out that we actually live on a shared planet with a common atmosphere; sourcing your fuels from abroad does nothing to prevent climate change. But it does mean that you are unable to secure some of the most fundamental inputs to your economy.
Plus you have no control over the standards for extractions (e.g. methane leaks), and shipping it causes more pollution.
They're actually worse off, and they pay more for it instead of creating jobs and keeping the money in their own economy. Meaning less money for e.g. green programs to move away from fossil fuels.
It's not clear to me that the radar is necessarily a separate constellation. SpaceX hasn't said a whole lot, but for example, SpaceX sells "Starshield" comms to the military, which uses the Starlink satellites for communications, except apparently the military has separate ground stations.
There's also been some published work where people have been talking about using Starlink signals passively for aircraft tracking. I'm not sure you couldn't use Starlink as a "stealth radar", in that the waveforms that are coming out of the satellites look like normal communications, but the satellites can also look at the echoes. Having the entire Starlink constellation form the radar is also pretty attractive from a cost and resiliency perspective since there are so many satellites to shoot down.
Everything is a holocaust or genocide to pro-Islamist lefties. Y'all've so discredited those terms that if at some point Israel actually does start a genocide, people will just shrug.
"There's a genocide going on in Gaza? Yeah I know, you've been whining about it for years now."
60% is actually very close to 93%. To go from natural uranium (<1% U235) to 60% represents the vast majority of the effort. From 60% to 93% is actually quite quick; most of the material is already U235. And they already have enough to build maybe a dozen bombs.
They also have (had?) a very active ballistic missile program, and have conducted implosion experiments.
The constellation of evidence is quite clear: Iran is a threshold nuclear state with all the pieces necessary to credibly threaten the region (and soon the US homeland) with nuclear weapons.
We've gone from, "The amazing Islamic Republic of Iran isn't even capable of building deliverable nuclear weapons and they have lots of peaceful reasons to do enrichment to 60%!" to "Yeah OK, they are capable and they are indeed enriching Uranium for their weapons program--hey, look over here! USA and Israel!!!"
The clock was programmed to count down from 8,411 days, corresponding to a 2015
statement by Supreme Leader Ayatollah Ali Khamenei, who predicted that "Israel
won't exist in 25 years". He claimed in his statement that there will be nothing
left of the Jewish state by 2040. The statement was made in the aftermath of a
September 2015 nuclear deal that had a timeline of 25 years to complete. He
predicted that it would not take that long for Israel to cease existing.
Protesters annually chant "Death to Israel". The installation was part of a much
broader demonstration involving over a million participants, where anti-Israel
slogans and imagery were prominently featured.
> The US has absolutely no business attacking Iran.
Iran's theocratic regime just murdered tens of thousands of protestors, regularly organizes chants of "Death to America", calls the US "The Great Satan", sponsors terror organizations all around the region, has (through their Houthi proxies) cut off critical sea lanes in one of the most strategic areas, is very close to developing nuclear weapons (with enough HEU already to build maybe a dozen bombs), has extensive ballistic missile magazines and expertise, and is working on ICBMs explicitly to reach the US homeland.
But oh yeah, this is totally unprovoked and the US has no business attacking Iran. Right.
It's amazing to see the justification done by some people to attack other sovereign countries. Did not america learn from the fake WMD fiasco with Iraq?
The IAEA estimates that Iran had 440.9 kg of uranium enriched to up to
60% before last year's Israeli-U.S. attacks - enough, if enriched further,
for 10 nuclear weapons, according to an IAEA yardstick.
The agency and Western powers believe the bulk of that is still intact.
Washington wants Tehran to give it up.
I seem to have missed the IAEA report on Iraq's 400+ kilos of HEU.
Jm2c but I wonder how people can be surprised that Iran wanted to build a nuclear weapon, especially after the US under Trump's first presidency pulled out the nuclear deal struck under Obama and cornered Iran even more.
Like do people in US realize that countries around the world take notes about what happened to the Libyas and Iraqs and many others (like Colombia recently) and see that the US will attack other countries with impunity.
The US feels threatened by Iran's nuclear and ballistic missile programs and has the military force to stop them, probably. Diplomatic avenues bore no fruit. Military force is now being used to--hopefully--end the threat definitively.
Yes, of course we are aware of what happened to Ghaddafi. It's very en vogue to point out the game theoretical incentives to develop nuclear weapons.
But seemingly people never bring up South Africa's disarmament. And nobody ever mentions that game theory also incentivizes the US prevent their adversaries from developing nuclear weapons where possible.
Giving up or stopping development of nukes may invite attack. Refusing to stop developing them may also invite attack.
What I said is factual. We're already seeing pictures of murdered children coming out of Iran, just like we did with Gaza, the West Bank, Lebanon... Not a single shred of evidence has been produced to back the claim that Iran murdered anyone, let alone tens of thousands of people.
I'll save everyone the clicks: there's no evidence of Iran claiming they killed over 3000 people in any of these articles. There's a claim they said this, but as with all reporting about Iran, no proof. Also, 3000 is not "tens of thousands".
"The Human Rights Activists News Agency says it confirmed more than 7,000 deaths and that it is investigating thousands more. The government has acknowledged more than 3,000 killed, though it has undercounted or not reported fatalities from past unrest." - https://apnews.com/article/iran-campus-protests-crackdown-54...
'"I would put the minimum estimates to be 5,000 plus," Mai Soto, the United Nations’ special rapporteur on Iran, said in an interview with ABC Australia. Soto noted 5,000 dead is a "conservative" or "the minimum" estimate. Other credible estimates, she said, indicate as many as 20,000.' - https://www.usatoday.com/story/news/world/2026/02/20/how-man...
> as with all reporting about Iran, no proof.
In the same way there's no proof humans ever walked on the moon, I suppose.
It's constant deflection with you people. You can never actually address the facts head on; just deflect with "there's no evidence", pivot to "there's no proof", then cast aspersions on the most disagreeable messenger.
You have previously intimated that you are also in the United States. Should I dismiss your arguments because you're allegedly based in the US, too?
Yes. In my view, the negative payoff from getting locked out of a machine due to a key file mishap is more severe than the payoff of typing passwords all the time. I also use machines of various distributions and eras, and so the configurations would all differ and create hindrances.
I realize the security relevance of that, but I do not have daily images to restore from if something happens. I got locked of a key only box one time with an error after a reboot, and never want it to happen again. It felt like being robbed.
reply