News for hackers. Stuff that matters.

deviantbit

Re:I want a good parallel computer on Saturday March 22, 2025 @11:57PM (#43450806)
Not in those terms, but an autobiography is coming, and bits and pieces are being explained. I expect about 10 people to buy the book, as all of the socialists will want it for free. I am negotiating with a publisher as we speak on the terms.
Re:I want a good parallel computer on Saturday March 22, 2025 @11:43PM (#43450753)
"financial incentives"

This is the most ignorant response I've seen yet. We don't expect monetary gain from publishing a book. We expect our costs to be covered.

This is about the consumer, not the publisher. If we lived in a socialist system, they would still pirate our publications and we will still be in debt over it.

Re:I want a good parallel computer on Friday March 21, 2025 @09:11PM (#43442396)
So, there are books out there. I use Computer Architecture: A Quantitative Approach by Hennessy and Patterson. Recent revisions have removed historical information. I understand why they did remove it. I wanted to use Stallings book, but the department had already made arrangements with the publisher.

The biggest problem on why we don't write books is that people don't buy them. They take the PDF and stick it on github. Publishers don't respond to the authors on take down requests, github doesn't care about authors, so why spend the time on publishing a book? We can chase grant money. I'm fortunate enough to not have to chase grant money.

Re:I want a good parallel computer on Friday March 21, 2025 @08:30PM (#43442176)
Sure you did. What were they? The only successful transputer was the T414 and it never made it outside academia.
Re:I want a good parallel computer on Friday March 21, 2025 @08:24PM (#43442135)
I tell students today that the best way to learn is by studying the mistakes others have already made. Dismissing the solutions they found isn’t being independent or smart; it’s arrogance that sets you up to repeat the same failures.

Sounds like you had a good mentor. Buy them lunch one day.

Re:I want a good parallel computer on Friday March 21, 2025 @08:16PM (#43442090)
... and maybe ChatGPT can design this dream system of yours.

You live in a fantasy—equal parts hand-wavey, ahistorical, and dangerously naive.

Let’s unpack your utopia: suggesting that we flatten address spaces, eliminate syscalls, remove hardware memory protection, and run everything inside a universal software sandbox (like a JVM/CLR/WASM)—because it’ll simplify the CPU, save power, and increase portability.

No. This is a complete misunderstanding of why modern OS and hardware architectures exist the way they do, and an even bigger misunderstanding of the real costs of doing all this “in software.”

Do you like not having random apps reading your banking info from memory? Good. That’s thanks to hardware-enforced memory isolation using virtual-memory.

Do you know what syscalls and context switches really cost? A few hundred nanoseconds. On a modern processor, that’s background noise unless you're writing ultra-low-latency code (and if you are, you're already using things like io_uring or DPDK to bypass the kernel).

Just go away.

Re:I want a good parallel computer on Friday March 21, 2025 @08:05PM (#43442030)
I see what you did there.
Re:I want a good parallel computer on Friday March 21, 2025 @08:00PM (#43441999)
Obviously you missed the sarcasm.
Re:I want a good parallel computer on Friday March 21, 2025 @07:56PM (#43441973)
Inmos was a disaster. No application ever shipped on one. EVER. It used a serial bus to resolve the problems that should have never been problems. Clearly you never wrote code for one. Each oslink couldn't reach more than 3 feet. What a disaster that entire architecture was.
Re:I want a good parallel computer on Friday March 21, 2025 @07:47PM (#43441930)
"I believe there are two main things holding it back."

He really science’d the heck out of that one. I’m getting tired of seeing opinions dressed up as insight—especially when they’re this detached from how real systems actually work.

I worked on the Cell processor and I can tell you it was a nightmare. It demanded an unrealistic amount of micromanagement and gave developers rope to hang themselves with. There’s a reason it didn’t survive.

What amazes me more is the comment section—full of people waxing nostalgic for architectures they clearly never had to ship stable software on. They forget why we moved on. Modern systems are built with constraints like memory protection, isolation, and stability in mind. You can’t just “flatten address spaces” and ignore the consequences. That’s how you end up with security holes, random crashes, and broken multi-tasking. There's a whole generation of engineers that don't seem to realize why we architected things this way in the first place.

I will take how things are today over how things used to be in a heart beat. I really believe I need to spend 2-weeks requiring students write code on an Amiga, and the programs have to run at the same time. If anyone of them crashes, they all will fail my course. A new found appreciation may flourish.

Re:I want a good parallel computer on Friday March 21, 2025 @07:29PM (#43441816)
Yep, transputers failed miserably. I wrote a ton a code for them. Everything had to be solved in a serial bus, which defeated the purpose of the transputer.
Re:I want a good parallel computer on Friday March 21, 2025 @07:27PM (#43441799)
Let’s be honest, saying “just fix the page tables” is like telling someone they can fly if they “just rewrite gravity.”

Yes, on Apple Silicon, the hardware supports shared physical memory, and with enough “convincing”, you can rig up a contiguous virtual address space for both the CPU and GPU. Apple’s unified memory architecture makes that possible, but Apple’s APIs and memory managers don’t expose this easily or safely for a reason. You’re messing with MMU-level mappings on a tightly integrated system that treats memory as a first-class citizen of the security model.

I can tell you never programmed on an Amiga.

Re:I want a good parallel computer on Friday March 21, 2025 @07:24PM (#43441779)
Please explain how these "worker cores" should operate.
Re:I want a good parallel computer on Friday March 21, 2025 @07:18PM (#43441741)
Unified memory doesn't mean unified address space. It frustrates me when no one understands unified memory.
Wait until they get to code generated by AI. All of this Rust code that people are using and sticking in various Linux services. It won't be covered under the GPL. It will flat out be public domain.
Re:Why Layoffs Don't Work on Thursday March 13, 2025 @01:46PM (#43355642)
Don't you love people make claims and the can't substantiate the claim? They just down vote your comment.
Re:Why Layoffs Don't Work on Monday March 10, 2025 @12:33PM (#43322294)
Which company?
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @10:37PM (#43316457)
Balance sheets state otherwise.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @05:58PM (#43314331)
"Studies"

I love studies. Which study, who funded it, and why?

Widespread layoffs are not necessarily a sign management screwed up, in most cases this is not the case. For example, who would have been able to predict COVID? I did, but that was in the first week of January 2020. I was able to profit from it. The largest total number of layoffs in history occurred in 2020. Market conditions change.

Is it managements faults that productivity is increasing at exponential rates because of machine learning? Did they screw up If they fire everyone because of automation, and lower costs? That is not a screw up. Automation creates opportunity. It also creates opportunity to fire people to increase the value of the share holders equity.

Buckle up butter cup, your emotions and feelings about to become even less relevant. There is going to chasm in labor like never before. We have an entire generation that has been sold they are entitled to a job. Shareholders, management and politicians are sick of them and we're doing everything possible to avoid them.

Re:Why Layoffs Don't Work on Sunday March 09, 2025 @05:47PM (#43314237)
Your opinion and feelings. Balance sheets say otherwise.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @05:43PM (#43314209)
Which dot com was that?
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @05:33PM (#43314109)
Activist investors are a problem. Most recent activist investors have been centered around climate and DEI. Exxon had activist investors try to get on the board. There was a lawsuit over it, I believe it was Arjuna Capital.

Whoever is funding these activist investors, probably a nation/state, is doing it on purpose. Natasha Lamb has to be the dumbest investors to ever live, next to Cathie Wood, IMO. You could invest in an S&P index fund and perform better at 264% for the past 11 years, compared to her 132% realized, and taking far less risk.

I get everyone has their idealistic views of how the world should work, but capitalism dominates every other society, providing better living standards, and security. It is unfortunate we have had leftists in the Democratic party take control of it, and pushing some very strange agendas, that doesn't reflect reality. These strange agendas have been exploited by other nations, like China.

Some activist investors have brought better function management and boards. Carl Icahn is a great example. But he has also brought his fair share of problems.

Re:Why Layoffs Don't Work on Sunday March 09, 2025 @04:15PM (#43313270)
No, it really doesn't.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @04:10PM (#43313215)
He lied to you then.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @03:09PM (#43312530)
That is what happened in recent layoffs. You can look at the company as a whole and say, well they were making money. Specific divisions of the company were in the red. Many companies rotate the bottom N percent to dump the bad performers. If you're not doing your job to the best of your ability, there is someone else out there that is willing to do the job. You're not special.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @02:48PM (#43312328)
No, it's not a choice we made as a society. It was a choice the shareholders made. You didn't make that choice, you were not part of it.
Re:Why Layoffs Don't Work on Sunday March 09, 2025 @02:30PM (#43312152)
This is a terrible opinion piece. Layoffs do work. Cutting hours does not work. I am always amazed how few people understand how a business operates. They think its a community organized event where there is unlimited revenue.

There are a few basic accounting principles employees need to understand. It all revolves around Assets = Liabilities + Owner’s Equity. If you think otherwise, take a "Cost Accounting" course. This is a pure numbers game. Everyone wants to wrap a psycho analysis into something that has ZERO relevancy.

If a company is bleeding revenue, it cannot sustain the overhead of employees. They have to go. Cutting hours does nothing for those on a salary, and those that are hourly, benefit costs are far more expensive than their wage. If a company has stagnant growth, that means leadership has made bad decisions, and things have to change.

Employees feelings don't matter on a Balance Sheet, Income Statement and Cash Flow Statement. There is not a "Employee's Feelings" column on the ledger. Everyone can be replaced. No one is special, unless you're a majority shareholder.

Re:Partnering with the Shawnee Tribe for Civilization VII on Sunday February 23, 2025 @07:51PM (#43154766)
How? If you enter into a contract with them, it has to be litigated in their tribal court. Who do you think is going to win? The chief sits on the court as the judge and jury.
Re:Partnering with the Shawnee Tribe for Civilization VII on Sunday February 23, 2025 @07:44PM (#43154725)
I've worked with the Shawnee Tribe, they don't do anything for free. In fact, when they owe you money, they will send you a letter notifying you of a generous contribution you made to their hospital. They are flat out thieves and you can't do anything about it. They are a sovereign nation.
Re:The Fastest Mutexes on Thursday October 03, 2024 @11:18AM (#41731584)
This codes benchmarks mutex contention, not mutex lock performance. If you're locking like this, you should reevaluate your code. Each thread locks and unlocks the mutex for every increment of g_chores. This creates an overhead of acquiring and releasing the mutex frequently (100,000 times per thread). This overhead masks the real performance differences between locking mechanisms because the benchmark is dominated by lock contention rather than actual work. Benchmarks such as this one are useless.

Older »