piadodjanho

21 ms·

61.
by piadodjanho 7y ago
This is not the first time an airline rebrands an airplane due fearmongering after an accident. The Fokker-100 was rebranded as MK-28 in Brazil after an accident (trust reverse deployed after take-off). The news reported every incident with
62.
by piadodjanho 7y ago
Nice catch. > They make it clear in their notes that "are numbers" includes infinities but not NaNs. This is definitely not very clear on the form, thought.
63.
by piadodjanho 7y ago
Sorry for tricking you into giving real data for the researchers LOL. Another comment pointed out to the paper with the answers. A great resource to learn about the FP corner cases is the great Random ASCII blog: https://randomas
64.
by piadodjanho 7y ago
First, I would like to apologize for my attitude on the original post, my choice of words was very inappropriate for this forum. Second. I would like to thank you for bringing another point of view. I'm talking about the average joe e-
65.
by piadodjanho 7y ago
Hmm. Their results seems to good to be true without using quire. Look: The two largest/smallest exponent posit16 can represent are: min exp: 2^{ (+14<<es) or 0x01 } * (1+0) = 2^(-28) max exp: 2^{ (-14<<es) or 0x00 } * (1+0
66.
by piadodjanho 7y ago
There are a few FPGA implementations. The posit hub website provides a table with all of them. Each has different trade-offs. The implementation I working on is focused on hardware reuse and high frequency operation.
67.
by piadodjanho 7y ago
I will get in touch once I finish the section on Posit. I would appreciate some feedback! I'm writing in my language, but hopefully the google translator will work well. Gustation is being too ambitious trying to reinvent the whole flo
68.
by piadodjanho 7y ago
> They tell you that the quire is even faster than a traditional sum while being more precise but reading the associated reference reveals that it only hold with specific hardware which could also be used to use the quire with floats. Pr
69.
by piadodjanho 7y ago
They are not the same thing, but they are close enough most of the time. The issue with floating point arrises when comparing very close numbers.
70.
by piadodjanho 7y ago
Sorry, but you actually sounds like one those people who dosen't really know how FP work. > I shudder to think how many e-commerce sites use `float` for financial transactions! The float IEEE-754 represent up to 9 decimal digits (
71.
by piadodjanho 7y ago
He argues that underflow/overflow are usually caused by bugs in the code. Having so many special numbers implies reducing the numeric representation (there are a lot of NaN in IEEE-754) and a hardware overhead to deal with all the spec
72.
by piadodjanho 7y ago
The innovative idea from posits are they use Golomb-Rice prefix to encode exponent numbers. The Golomb Rice prefix let you encode exponent closer to zero using less space. For example, the posit16 with nbits=16 and es=1 encodes the exponen
73.
by piadodjanho 7y ago
I am implementing a high performance posit arithmetic unit for uni thesis. The idea of using Golomb-Rice prefix tree to encode the exponent is genius. Having said that. I totally agree his papers make a lot of bogus claim. For instance, the
74.
by piadodjanho 7y ago
In the paper "Do Developers Understand IEEE Floating Point" the authors surveys phd student and faculty members from computer science and found out that your observation is true: most people don't know how fp works. They have
75.
by piadodjanho 7y ago
Nice. But 17 digits of of precision is on the rendered drawing bit much.
76.
by piadodjanho 7y ago
For the western emacs user probably not much. Unfortunately, the ability to use non-latin typefaces in the apps are often (resasonably) neglect by most devs. Using HarfBuzz (that is a font shapping library), the dev don't have to worry
77.
by piadodjanho 7y ago
Such the yummy lead white paint that kids peel from the walls
78.
by piadodjanho 7y ago
XeLaTex uses the same text shaping engine Firefox uses, that's why it looks so much better than the default latex.
79.
by piadodjanho 7y ago
Pandoc creates PDF using LaTeX. You still constrained to LaTeX features, and therefore the lack of trivial accessibility.
80.
by piadodjanho 7y ago
Are you sure about your Photoshop claim? I'd assume they would have rewritten most of their filters code in Halide by now.
81.
by piadodjanho 7y ago
The article says it had 122 failure due bird strikes plus 85 unnamed problems in about 30 years of data. Considering the number of flights, that does sounds reliable to me. I still think two failures in the same sensor, in the same airplane
82.
by piadodjanho 7y ago
I have the impression that people are overlooking the sensors. They are suppose to be very, very, reliable. Two different planes got wrong reading from sensor in the same side, this seems to be a red flag for me. I wonder in what side of th
83.
by piadodjanho 7y ago
In Brazil the deforestation in mostly for agricultural active and live stock. I'm curious to know how much the land would have to cost so the country made as much tax with the planted land as with other activities.
84.
by piadodjanho 8y ago
In the academia people prefer PyTorch over TensorFlow.
85.
by piadodjanho 8y ago
I never thought about that. Now it makes sense why they still use film even though the prices to shot some shows, even though it a lot of extra costs (media is more expensive, it is time consuming, better lightning, etc)
86.
by piadodjanho 8y ago
For VR/AR the most important feature is be pixel density
87.
by piadodjanho 8y ago
There are other acting acting forces on the car other than gravity