SoylentNews is people
SoylentNews
from the extraordinary-claims-require-extraordinary-evidence dept.
Shock result in particle experiment could spark physics revolution
Scientists just outside Chicago have found that the mass of a sub-atomic particle is not what it should be.
The measurement is the first conclusive experimental result that is at odds with one of the most important and successful theories of modern physics.
The team has found that the particle, known as a W boson, is more massive than the theories predicted.
[...] The scientists at the Fermilab Collider Detector (CDF) in Illinois have found only a tiny difference in the mass of the W Boson compared with what the theory says it should be - just 0.1%. But if confirmed by other experiments, the implications are enormous. The so-called Standard Model of particle physics has predicted the behaviour and properties of sub-atomic particles with no discrepancies whatsoever for fifty years. Until now.
CDF's other co-spokesperson, Prof Georgio Chiarelli, from INFN Sezione di Pisa, told BBC News that the research team could scarcely believe their eyes when they saw the results.
"No-one was expecting this. We thought maybe we got something wrong." But the researchers have painstakingly gone through their results and tried to look for errors. They found none.
The result, published in the journal Science, could be related to hints from other experiments at Fermilab and the Large Hadron Collider at the Swiss-French border. These, as yet unconfirmed results, also suggest deviations from the Standard Model, possibly as a result of an as yet undiscovered fifth force of nature at play.
Also at Nature and Ars Technica.
Journal Reference:
T. Aaltonen. S. Amerio. D. Amedei, et. al.,High-precision measurement of the W boson mass with the CDF II detector, Science, (DOI: https://www.science.org/doi/10.1126/science.abk1781)
(Score: 4, Interesting) by Snotnose on Saturday April 09 2022, @04:39PM (22 children)
3 different teams did this measurement 3 different ways. 2 agree with the theory. This one doesn't.
Odds are they made a mistake somewhere. It's gonna be subtle because you know that team went over everything with a fine tooth comb. But there's a mistake in there somewhere.
Relationship status: Available for curbside pickup.
(Score: 2) by captain normal on Saturday April 09 2022, @04:44PM
Maybe a programming error...?
"It is easier to fool someone than it is to convince them that they have been fooled" Mark Twain
(Score: 5, Interesting) by Mojibake Tengu on Saturday April 09 2022, @05:18PM (15 children)
I think the software package used for analysis mentioned in the original paper (CERN ROOT version 5.34/12) needs some serious audit.
The edge of 太玄 cannot be defined, for it is beyond every aspect of design
(Score: 3, Insightful) by istartedi on Saturday April 09 2022, @05:25PM (14 children)
It sounds like their real discovery might be a subtle floating-point error in somebody's CPU. I have no idea how much crunching they do to analyze the data, but if it's not too much then maybe they could re-run it on a different architecture to rule that out. A different compiler might help too.
Appended to the end of comments you post. Max: 120 chars.
(Score: 2, Troll) by Mojibake Tengu on Saturday April 09 2022, @07:45PM (4 children)
I suspect deliberate.
Creating a plausible controversial dramatic situation in opaque kind of research (opaque as in 'no true commoner could verify this' fallacy) is a model for asking more of huge amount of money pile. Those people are very smart to begin with, reusing an existing software glitch for the purpose is a no-brainer for them.
New paradigm: Hacked Science.
The edge of 太玄 cannot be defined, for it is beyond every aspect of design
(Score: 3, Funny) by Anonymous Coward on Saturday April 09 2022, @07:55PM
Hacked science? No, MBA science. Stacks of management, protocol, regulatory compliance, ethical compliance, budgetary projection... and 1 Chinese postdoc at the bottom doing the work (badly).
(Score: 4, Funny) by istartedi on Saturday April 09 2022, @08:24PM (1 child)
That's an interesting take, and from your original comment about the software package needing an audit, we can obviously infer it needs a lot of auditors who will be well paid.
Appended to the end of comments you post. Max: 120 chars.
(Score: 2, Touché) by Anonymous Coward on Saturday April 09 2022, @08:35PM
It's an interesting take in the same way that throwing spaghetti bolognese on a canvas is interesting.
(Score: 2) by PiMuNu on Sunday April 10 2022, @04:33PM
I don't think you know what you are talking about.
(Score: 0) by Anonymous Coward on Saturday April 09 2022, @07:52PM (5 children)
> subtle floating-point error
Please. You think nobody knows about floating point? I would guess members of their group designed floating point representation in digital computations. Subtle floating-point error, what a clown.
(Score: 1) by istartedi on Saturday April 09 2022, @08:21PM
You could say the same thing about Intel engineers, and yet that's where one of the most infamous floating-point bugs came from.
Appended to the end of comments you post. Max: 120 chars.
(Score: 2) by turgid on Sunday April 10 2022, @08:39AM (3 children)
They're human like the rest of us and make mistakes. Also, they're scientists and probably don't have very much training in software development so the quality of their code might not be that great. It's undoubtedly very clever code, but probably badly structured and full of subtle bugs that they haven't noticed particularly in corner cases.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 0) by Anonymous Coward on Sunday April 10 2022, @11:23PM (2 children)
Y'all software guys bitch about shitty software development, like, all the time. Don't be projecting your bullshit onto actually smart people. Physicists created The Bomb while you bitches only created Windows.
(Score: 0) by Anonymous Coward on Monday April 11 2022, @09:44PM
Funny how that turns out.
In my experience, the worst customers are doctors. "Rawr! I am doctor! I studied for more than a decade to be a doctor! You are a meat-thing, crawling on your lips through a sewer of electrons!"
"OK, great, then fix it with your scalpel. Or, you could listen to someone who spent years taking courses that weren't in med school, to resolve the problems that your med school training left you."
(Score: 2) by turgid on Friday April 15 2022, @10:28AM
I never said they weren't smart. I started out there myself and I know just how smart they are. What I was saying is that they haven't had the same amount of training in the ways of software, so their software tends to be of relatively poor quality. Mind you, there's a lot of software written by software professionals that's of poor quality, like Windows.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by RS3 on Saturday April 09 2022, @08:41PM
I don't know if there are still floating-point errors in CPUs, or if the researchers were using any such chips, but maybe some kind of math / rounding / data type conversion error?
(Score: 3, Insightful) by PiMuNu on Sunday April 10 2022, @04:32PM
> somebody's CPU
> I have no idea how much crunching they do to analyze the data
Just to be clear, analyses like this are typically not done on "somebody's CPU". The particle physics community operates a distributed set of clusters, with nodes in many universities worldwide and major nodes containing >> thousands of cores in the major laboratories.
As I mentioned elsewhere, the paper actually contains six separate analyses, each of which is statistically consistent with the others. The analyses are validated against known particle masses. The analysis was done in a fully blinded manner.
It is challenging to imagine that a "floating-point error" can cause such a discrepancy.
(Score: 0) by Anonymous Coward on Monday April 11 2022, @12:28AM
time to stop using the 586 for anything except Quake
(Score: 2) by HiThere on Saturday April 09 2022, @06:38PM (3 children)
You're probably right, but did they measure the mass the same way as the other two experiments?
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: -1, Troll) by Anonymous Coward on Saturday April 09 2022, @08:16PM
Great question, I'll get right on that. Handjob while you wait?
(Score: 3, Funny) by inertnet on Saturday April 09 2022, @08:24PM (1 child)
Yes, but right after Thanksgiving, hence the mass increase.
(Score: 0) by Anonymous Coward on Sunday April 10 2022, @02:17PM
Very lean Thanksgiving, I 'ave to say. Just 0.1% increase in mass. I gain a lot more on a Friday night, but then piss all the ingested beer and everything's dandy.
(Score: 4, Insightful) by PiMuNu on Sunday April 10 2022, @04:13PM
Just to be clear, the paper is very well thought through as one might expect. The team present 6 different measurements for the W boson mass, each of which is consistent with the others. The team is using a very mature analysis toolset, which has been used in countless other papers for similar analyses and not shown any problems. It is in tension with the standard model at ~ 7 sigma and in tension with the other experiments at about 3 sigma.
The analysis studies the mass of a W boson, a particle which decays radioactively after a vanishingly short lifetime. The experiment studies the kinematics of the decay products (momentum, energy) and uses this to infer the mass of the decaying primary particle.
One might question whether there are any systematic errors that may systematically effect all of the separate analyses. For example, such an analysis may be systematically biased by decays of particles that are not W bosons but do look like W bosons. The experiment studies two different decay modes, W -> electron + neutrino and W -> muon + neutrino. It would be surprising if there was an impurity that effected both decays in the same way, and of course the paper authors have thought of this and checked. The experiment uses the kinematic properties of the decay products. Electrons and muons behave differently and it would be surprising if both were biased in the same way. The paper authors checked by calibrating their instrumentation against a known decay having a known mass (J/psi).
Of course, mistakes do occur, and we will see what the experts say. But it is beyond "guy on the internet" level dismissals. It really needs a cross check from the other experiments.