shaklee3

5 ms·

1.
by shaklee3 28d ago
RCS is not terrible. Apple's implementation of RCS is terrible.
2.
by shaklee3 3mo ago
I've seen lots of successful large k8s installs. This same argument from the nomad crowd comes up constantly. There's a reason why nomad lost, and it's not because it's better.
3.
by shaklee3 3mo ago
They will be using a Samsung screen, so just no. Stop repeating this false trope. Didn't happen with the vision pro, and won't happen here.
4.
by shaklee3 5mo ago
It happened 2 years ago: https://developer.nvidia.com/blog/nvidia-transitions-fully-t...
5.
by shaklee3 7mo ago
I built the same thing and it gets calendar updates from the Google API and weather from wunderground or AccuWeather.
6.
by shaklee3 7mo ago
Which is also bad compared to 5.3 codex. People don't seem to realize that this is not codex 5.3 quality. It's a large step down on the benchmarks to get lower latency.
7.
by shaklee3 7mo ago
They don't even have GPU systems, so...no?
8.
by shaklee3 8mo ago
They have someone in the driver's seat
9.
by shaklee3 1y ago
Or green contexts
10.
by shaklee3 1y ago
800Gbps
11.
by shaklee3 1y ago
you can do this with home assistant already
12.
by shaklee3 1y ago
dp4a, ldg. just Google it. there's a whole page of them
13.
by shaklee3 1y ago
GBps
14.
by shaklee3 1y ago
not true; there are a lot of simd instructions on GPUs
15.
by shaklee3 1y ago
This is not true. GPUs are SIMT, but any given thread in those 32 in a warp can also issue SIMD instructions. see vector loads
16.
by shaklee3 1y ago
That's quite an interesting way to explain why Apple does poorly in blind tests. The real reason though is that Apple's cameras are just not as good, but I suppose it's easier to explain away by making up biases.
17.
by shaklee3 1y ago
there's a documentary about this, and the people who own the rights screwed over Bob and are purely there to exploit him. they have nothing to do with his family.
18.
by shaklee3 1y ago
C++ and its libraries can do everything erlang does with concurrency, and faster. erlang remains niche mainly for that reason.
19.
by shaklee3 2y ago
scott grey figured out this exact thing and more back in 2015 for maxwell, and it's been written about many times since by other people.
20.
by shaklee3 2y ago
no. they just released the Fitbit ace
21.
by shaklee3 2y ago
this is required for writing generic code when the function may return different types
22.
by shaklee3 2y ago
As someone not that familiar with haas, can someone explain why there's not a clear path to replace Alexa or Google home? I considered using haas recently to get a gpt like response after being frustrated with Google home, but it seems
23.
by shaklee3 2y ago
the new version is censored. they removed things like blowing up the rhino with dynamite
24.
by shaklee3 2y ago
This is totally inaccurate. There is nothing 20x slower about a GPU core unless you're nitpicking, in which case you can show those same GPU cores are 20x faster than CPU for other things.
25.
by shaklee3 2y ago
No. D2D is arroubd 3TB/s
26.
by shaklee3 2y ago
this is bidirectional bw and it's gh200, not gh100
27.
by shaklee3 2y ago
what does a laptop GPU have to do with a server grade xeon that the article is comparing against?
28.
by shaklee3 2y ago
To save some time for people, the title is absolutely incorrect. The GPU is significantly faster for this test, but the author is measuring PCIe bandwidth from a PCIe generation of 10 years ago. If instead they had used pcie5 the bandwidth
29.
by shaklee3 2y ago
The public sector is typically air-gapped, so not really.
30.
by shaklee3 2y ago
This is exactly what the Fitbit Ace LTE does. So far it's been great since my daughter has no desire to ask anything else on it besides increase physical activity to get prizes

More ›