wtracy

35 ms·

331.
by wtracy 8y ago
Care to elaborate? Nvidia used to be the only discrete GPU manufacturer that had Linux support worth mentioning. I know that AMD/ATi has made a lot of progress, but is Nvidia really the worst at this point?
332.
by wtracy 8y ago
Why do you say that OSS developers are against that? Personally, I am all in favor of any monetization scheme that doesn't infringe on the four freedoms described by the FSF: https://www.gnu.org/philosophy/free-sw.
333.
by wtracy 8y ago
I would love to see a similar "jumping off point" for audio programming. Anybody know of one?
334.
by wtracy 8y ago
Sorry, I misunderstood the question to mean, "What is the advantage to having the microcontroller on the CPU die, as opposed to having separate chips on the same board."
335.
by wtracy 8y ago
From a user/developer perspective, the advantage would be low latency. I imagine that the more interesting advantages are from the perspective of the board designer.
336.
by wtracy 8y ago
The value proposition of a $12k conference is not in the seminar content. It's in getting into a room of sales prospects and potential investors who have already demonstrated that they can afford to spend $12k on a conference . It&#x2
337.
by wtracy 8y ago
Relying on your customers to use password managers to protect themselves is not very useful in the Eternal September age of the internet. (I do see your point that a Google redirect is only marginally more likely to be effective for phishin
338.
by wtracy 8y ago
That might not be an issue for Google, but I could see it being a big problem for a company that relies heavily on projecting a "family friendly" image (think Disney).
339.
by wtracy 8y ago
A decade ago people were using Google redirects to lure people into visiting shock sites. Back then, it seemed reasonable to not consider this a real security flaw. Now that everyone has a Google account, the possibility of credential theft
340.
by wtracy 8y ago
Yeah, the rebasing complaint isn't fair if you're using a modern VCS. I used to work on a large team at Cisco Systems that used Clearcase. Clearcase does not do merges. If anything has changed in master, you have to check out ag
341.
by wtracy 8y ago
Well then, Mr. Taleb needs to learn how to use satire constructively. We started with what could have been an interesting discussion about the validity of IQ tests, and suddenly it turned into a passive-aggressive discussion about fashion.
342.
by wtracy 8y ago
I'm not familiar with how monorepos work in practice, but it seems obvious to me that it's going to complicate everyday tasks. Ready to commit? Whoops, another team made a bunch of commits to their project, and you need to rebase
343.
by wtracy 8y ago
Again, "shock humor" and satire are antithetical to "cutting through the bullshit".
344.
by wtracy 8y ago
To expand, it makes sense in situations where the "expert" is unable to scale up any farther (lack of capital or labor come to mind) or cannot expand farther without assuming an uncomfortable amount of risk. The second possibility
345.
by wtracy 8y ago
> Yes, you can try to sue for anything. Point taken. > the lawyer will try to show it was not "good faith" but a failure of software AFAIK, the phone calls are being placed by Facebook employees, not software. Presumably, th
346.
by wtracy 8y ago
> His whole approach designed to cut through bullshit and seek the truth. The ad hominem attacks (particularly the "losers in Birkenstocks" remark) in this essay are the exact opposite of that.
347.
by wtracy 8y ago
Your post addresses my biggest problem with the article: It reads more like a wishlist than a guide. I would like to these issues addressed going forward: (And please don't take this as a list of reasons why this won't work; these
348.
by wtracy 8y ago
Regardless of the reasons for the call, can you really pursue damages from someone who made a good faith call to emergency services? The only legal issue I could see is violation of privacy, but I bet their EULA covers them there. Now, if t
349.
by wtracy 8y ago
I would say yes, because the developers would find something equally inane to spend money on. Perhaps a more relevant example: Look at the bottled water industry. Even if all our creature comforts are provided for free, someone will pay for
350.
by wtracy 8y ago
There are people who spend thousands of dollars on Candy Crush. There will always be something more. Also, I have to point out, if you had made that list a hundred years ago, electricity wouldn't have been on it. Luxuries have a way of
351.
by wtracy 8y ago
> Most players just restart each battle until deathless runs are achieved- either due to emotional or practical concerns. So, when someone dies, it’s not sad, it’s annoying. How about balancing a handful of levels so that it's impo
352.
by wtracy 8y ago
>Same number of misses, but each miss is effectively half as costly Wow, I'm still trying to wrap my head around that. I was going to argue that in cases where there's no way to split an algorithm across multiple threads withou
353.
by wtracy 8y ago
I'm also curious if it affects all USB ports on the computer, or only some. (Some PC's have "always on" outlets that supply power even when the PC is turned off, and I imagine that those take a slightly different route t
354.
by wtracy 8y ago
Call it a "capacitor array" and you're set.
355.
by wtracy 8y ago
It seems to me that the speed increase in your example comes from having twice the cache and therefore half the misses. You can do that without adding cores. (Barring edge cases where two threads use different pages that happen to share the
356.
by wtracy 8y ago
I suspect the answer is, Rust ignores the general case, and simply doesn't let you do anything that's not provably correct.
357.
by wtracy 8y ago
SIMT architecture is why GPUs are actually not that great for ray tracing. In a nontrivial scene, some rays will bounce more times than others, and that means that the whole parallel execution unit is tied up until every single ray terminat
358.
by wtracy 8y ago
I'm having trouble imagining a performance benefit to two (otherwise identical) 1ghz cores over one 2ghz core, unless: 1. The two processes cause frequent evictions of the CPU cache in some way that two cores can mitigate by having sep
359.
by wtracy 8y ago
Nothing that processes nodes in parallel can guarantee either correct depth first or breadth first ordering. That said, it's easy enough to get effectively correct ordering by reassembling the results in the correct order as the work
360.
by wtracy 8y ago
While the article talks about not being able to run x86 binaries on Arm, this is the real drawback to the platform. I really wish we would see a standard Arm kernel instruction set come along and get adopted by every manufacturer who isn&#x

More ›