Veedrac

95 ms·

811.
by Veedrac 7y ago
You're arguing against a position that nobody sensible holds. It's like you're yelling at the invention of the train because it can't fly to the moon. Well, sure, but it's a necessary step on the way. Researchers ar
812.
by Veedrac 7y ago
I have no expertise at all in the matter, but my understanding is that the evidence for dark matter is overwhelming. In particular, this Reddit comment (which is mostly just a TL;DR of [2]) seems pretty convincing: https://www.re
813.
by Veedrac 7y ago
Honestly, on HN I feel there is much more disdain for my position. Given the world we're in, Facebook being able to assist law enforcement around child abuse, sexual slavery and human trafficking provides a lot of value. I feel people
814.
by Veedrac 7y ago
The 10¹⁷ includes that factor. 10¹⁰ neurons by 10⁵ synapses by 10² Hz. It seems unlikely to me that the meaningful temporal resolution is going to be 1000x that of the firing rate, but if you want to add a factor of 10 or so I wouldn't
815.
by Veedrac 7y ago
The higher numbers there (eg. 10²⁸) are irrelevant. It claims a single neuron is operating at 10¹⁶ operations per second—that is, as much computation is happening within a single neuron as is the sum of all computation between neurons i
816.
by Veedrac 7y ago
This is actually pretty interesting. The statistic most people noticed is frequency, which climbed exponentially between 1995-2003 and came to a dead stop a year later. https://i.imgur.com/QAEqcie.png However, almost immedi
817.
by Veedrac 7y ago
The A12 has better single-threaded SPECint2006 scores than EPYC7742 almost across the board. It's about 10% worse than the 3900X, and 20% worse than the i9-9900K, so if the 20% uplift holds true for SPECint scores then the A13 should b
818.
by Veedrac 7y ago
KOReader for the Kobo has a skim feature which works OK for this purpose, though you're always going to run against framerate issues on these devices until CLEARink or such becomes widely available.
819.
by Veedrac 7y ago
1TB 970 EVO is $170 on Amazon.com, not $300.
820.
by Veedrac 7y ago
The highest speed consumer M.2 drives go around $180/TB. Apple doesn't pay retail prices. $40 could well be a significant overestimate, especially given it's displacing an existing chip.
821.
by Veedrac 7y ago
Eww, they made this opt-out now? That might actually convince me to migrate...
822.
by Veedrac 7y ago
This is a rather different claim to saying there wouldn't be a benefit. I'm not saying filesystem accelerator SSDs would necessarily be in sum a good thing, just that in principle it seems like the kind of thing that should perfor
823.
by Veedrac 7y ago
You are rather missing the argument. If KV SSDs are much faster than doing the same with a software DB on top of a block abstraction, is that not proof that these are inequivalent? GPUs run software too, but that doesn't mean it's
824.
by Veedrac 7y ago
I confess I'm not sure why people are talking about bugs now. I was talking about how hardware can access higher levels of performance and efficiently guarantee more things (eg. early persistence guarantees) than software can.
825.
by Veedrac 7y ago
We're nowhere near ‘post-Moore’ in persistent storage, though. Although Dennard scaling for CPUs has faltered (if not Moore proper), storage is still getting faster and denser at the same time... it just does the speed upgrades in disc
826.
by Veedrac 7y ago
The point is there, though: hardware has more room to improve than software does, when it comes to IO-heavy tasks.
827.
by Veedrac 7y ago
Which is often just meaningless pedanticism. There are a lot of questions about C APIs where it's irrelevant whether the answer uses C or C++. The latter might not be an exact superset, but in most cases it's easily close enou
828.
by Veedrac 7y ago
You don't even have to have an account to post an answer, nor any points to post a question. You need only 15 points to upvote (2 upvotes on an answer, or three on a question).
829.
by Veedrac 7y ago
> it's still a pretty big deal since I haven't heard of anything I'd truly call next-gen happening to transistors There are several, for example nanoscale vacuum-channel transistors (think nanoscopic vacuum tubes), and the
830.
by Veedrac 7y ago
Again, I suggest you give direct sources for your arguments.
831.
by Veedrac 7y ago
“Not all that difficult” is in the context of the brain, where things tend to vary between ‘pretty difficult’ and ‘seemingly impossible’. I say ML shows pattern matching of this sort isn't all that difficult because progress has been s
832.
by Veedrac 7y ago
Elon Musk isn't part of OpenAI any more. I don't think his stance should hold that much weight any more. > Sikka argued that "openness" was the fundamental reason he supported the project. I believe the relevant quote
833.
by Veedrac 7y ago
> When they refused to release GPT-2 the OpenAI researchers took an extreme on that spectrum, asserting that essentially they had built a mind control device. Having read everything I could find from OpenAI on this topic, I have no recol
834.
by Veedrac 7y ago
> I'm convinced that the people behind OpenAI understand this nuance, even if most non geeks struggle with the difference between possible and feasible. This means that OpenAI mixed the two up on purpose. You should probably actuall
835.
by Veedrac 7y ago
Please try to argue in good faith. I've already said ML models have caveats, obviously I don't think they're perfect or par-human.
836.
by Veedrac 7y ago
Yes, you are capable of pattern matching. The point is that this is a distinct thing from intellectual ability, since we know having orders of magnitude less sensory input doesn't much seem to limit the ability to do high-level reasoni
837.
by Veedrac 7y ago
> So when a human thinks about a cat, she can probably picture the way it moves, hear the sound of purring, feel the impending scratch from an unsheathed claw. She has a rich store of sensory information at her disposal to understand the
838.
by Veedrac 7y ago
They say they support efficient execution of smaller batches. They cover this somewhat in their HotChips talk, eg. “One instance of NN, don't have to increase batch size to get cluster scale perf” from the AnandTech coverage. If this d
839.
by Veedrac 7y ago
Maybe it's best to wait until after we've hammered away at the problem on warehouses of human-scale hardware for 50 years before calling it inscrutable? The paper that put neural networks on GPUs is ten years old . I get that the
840.
by Veedrac 7y ago
A core receives data over the interconnect. It uses its fast memory and local compute to do its part of the matrix multiplication. It streams the results back out when it's done. The interconnect doesn't give you single-cycle acce

More ›