nightski

31 ms·

511.
by nightski 3y ago
I'm following lab grown meat with optimism but I really hope they are required to label it as such.
512.
by nightski 3y ago
For a second there I thought I was reading boomer news instead of hacker news. I kid, but honestly there are plenty of us that still upgrade our systems piecemeal. I'm always buying new parts. Partly because I like having the latest
513.
by nightski 3y ago
I prefer an explicit function call over compiler voodoo magic, especially when it comes to reactivity (of which can have performance implications). Any solution where a compiler is changing the semantics of my code on the fly is just a har
514.
by nightski 3y ago
If it's only a 0.005% speed improvement the manual hacks probably aren't required :P
515.
by nightski 3y ago
I work on several large react apps that are very data heavy and what I would consider complex. Honestly we just use useState and prop drilling. Most state is stored at optimal position within the hierarchy, so there really aren't tha
516.
by nightski 3y ago
Interesting and disappointing that someone from Microsoft would convey that .NET Core is too slow and not portable enough :) VS Code is Microsoft's tool after all.
517.
by nightski 3y ago
Almonds really aren't expensive on a per calorie basis. They are far cheaper than tomatoes. For example at Walmart as of today I can buy 2,240 calories of blue diamond almonds for $6.98. The cheapest Roma tomato is $0.44 per item, w
518.
by nightski 3y ago
Where is your source? Last time I looked an almond took roughly the same amount of water to grow as a tomato on a per calorie basis.
519.
by nightski 3y ago
You are probably right, but at the same time there are probably plenty of people willing to step in and take their place.
520.
by nightski 3y ago
It's not a full dataset per say, but you might be interested in Sebastian Starke's DeepPhase work [1]. [1] https://youtu.be/wAbLsRymXe4 [2] https://github.com/sebastianstarke/AI4Animation
521.
by nightski 3y ago
I fully understand that maybe I lack the vision Apple has, but if I had to call it this is going to be a total flop. There's no vision here, it's just iPad glasses. They did not embrace spatial computing at all or build a new ex
522.
by nightski 3y ago
That demand will only increase, which further emphasizes my point. It can't be a luxury.
523.
by nightski 3y ago
I'm all for efficiency but if we are to progress as a civilization we need to find a way to not have power hold us back. We can get to a point where we have nearly unlimited power with minimal impact on the environment around us. Tha
524.
by nightski 3y ago
You might want to check out what Nvidia is doing with RTX Remix - https://www.nvidia.com/en-us/geforce/rtx-remix/
525.
by nightski 3y ago
I like the idea of decentralized but I'd suggest you don't need to go fully decentralized where every peer has a full copy. Actually I kind of like the Bitcoin approach in that you have the ability to create a full peer, but most
526.
by nightski 3y ago
This gives me Mixer vibes.
527.
by nightski 3y ago
I've been consulting for well over a decade. I used to fly to visit my clients. During the pandemic they all learned to work remotely and I haven't made a trip since. We actually communicate more than we did before, and it feel
528.
by nightski 3y ago
Yeah and those people aren't going to pay significantly more money for a Mac M2 Max.
529.
by nightski 3y ago
Game developers are not going to start mass adopting a JavaScript API.
530.
by nightski 3y ago
The GPU is $4500. That gives you $2000 to build a machine (which you can build a top of the line consumer PC for cheaper than that, I just did recently w/128GB DDR5 and Zen 4 7950X3D).
531.
by nightski 3y ago
I don't see why it is unreasonable to desire Apple to provide documentation and open up their hardware. Honestly I'd ask the same as any other hardware vendor. Intel for example provides very detailed technical documents on thei
532.
by nightski 3y ago
Specifically GPU drivers, which can dramatically impact performance. Especially if I am attempting to run any kind of ML workload from Linux. I'm assuming it's basically a non-starter at this point and one is forced to use Mac O
533.
by nightski 3y ago
This is great work and I commend it. But in other threads people are acting like Asahi Linux hardware support is 100% complete. My fear is that if I were to go this route and purchase the hardware I'd be seeing fraction of the perfor
534.
by nightski 3y ago
You can get an A6000 for cheaper than an M2 Ultra. That said it's still only 48GB.
535.
by nightski 3y ago
Its not that it is immoral, just that it indicates management lacks the imagination of what to do with it which can be problematic if it is a competitive industry.
536.
by nightski 3y ago
That's an interesting option. Can you get direct GPU passthrough similar to what is offered with CUDA & WSL(g) on Windows?
537.
by nightski 3y ago
AI is a pretty well defined field I feel like. It combines symbolists, connectionists, evolutionaries, bayesians, and analogizers. Just because those not in the field misuse the term does not make it less useful.
538.
by nightski 3y ago
You don't have that level of performance though. Not even close really. RAM amount is only one piece of the puzzle.
539.
by nightski 3y ago
High end consumer GPUs would be your 3090/4090 cards which both have 24GB.
540.
by nightski 3y ago
Where do you think company profits come from if not from customers?

More ›