You can't disprove the existence of something which is not sure to exist.
(Of course, you can't turn off a light that's already off, but that's too obvious)
So saying the LHC cannot disprove the HB is like saying no one can disprove Unicorns
nice try
Erm... no
Actually, the LHC can't prove the existence of the Higgs Boson
They can show overwhelming evidence for it, up to the point the Higgs Boson is as accepted as the electron for instance
But until then, you can't "disprove" a thing that has the best evidence for existing @ 3 sigma...
Yeah, I mean, you cannot pay for a photo with Richard Stallman for $3.41
And the worse thing is, there isn't a shortage of new candidates for this list
"much harder to refute" how about no
- Neutrinos through Interstellar medium vs. neutrinos through "planet earth" (almost the same thing to neutrinos, sure, still, almost)
- Neutrino interactions with interstellar medium
- Neutrino oscillations
- Neutrino generation process in a supernova event How do we know the level of neutrino generation didn't begin to raise 4 years before it went supernova? Far fetched yes, impossible, no.
Yes, the stock is clearly increasing over time...
No, my CPU doesn't stress out while playing an MP3 using any program while PulseAudio is disabled
Resampling is a rather more complex operation than you might think
Well, I actually studied signal processing at university, so I have an idea of what it takes to resample a sound. (I actually did it with images once, but it wasn't very fast - processing a 2D signal is very complicated).
For sounds it's easier since you can have very good and still very fast methods. PA may be picking between perfect-slow and very fast-bad (And no, if I can guess what the 'trivial' method does, I wouldn't recommend it)
But I guess some programs already resample to 48000 internally, like mplayer (and maybe they can do it inside the MP3 decoding process, simplifying it considerably)
You can't rely on SIMD instructions always being available
Nowadays? Yes, you can. The machine I bought in 2002 (Athlon XP) has SIMD instructions. You can pretty much count on any machine today having SSE
Unless you're using a machine older than that in which case PA is the smallest of your problems.
Exactly
Also, JACK helps with latency, instead of totally screwing it up like PA
Or you can make the audio card run in the sampling rate of the audio file...
And resampling still does not justify the CPU usage (unless who did the resampling algorithm used a naive algorithm, which may be very well possible), especially with SIMD instructions.
Remember what the % CPU usage of winmodems was (with significantly worse processors)? And that's way more difficult than resampling a sound file
No VDPAU here, for an SD video mplayer uses ~10%CPU
I'll not reinstall PA here, but I've seen it use more than 10% cpu on a number of occasions (besides all the other problems)
I confess I tried to use PA. When using Ubuntu 9.04 PA lasted 1h on my machine, for 11.04 it clearly improved, it lasted almost a week before I uninstalled it.
No one in their right mind would ship Pulseaudio (amongst other things)
Right, nobody in their right mind would do that.
Except that everybody is doing it, and it works just fine. But the dinousars who got an opinion about pulsesaudio several years ago and don't know that these days pulseaudio works fine and provides features that nobody else does will never change their mind.
No, It doesn't work "just fine"
No, my opinion is not based on an older version of PA. PA is bad NOW, before it was even worse
Except that everybody is doing it, and it works just fine.
Maybe that's why there are several discussions about removing/disabling PA from Ubuntu (and others)
provides features that nobody else does
Like what? Using more CPU for audio than for playing a video? (yes, I did this test)
Increased delay from 'start playing' to actually playing (try it with mplayer)
So yeah, I guess I'm a dinossaur for wanting my SW to work without BS, without hogging my CPU and failing to work properly half of the time (again, newest version). Yeah, sound mixing, per app volume, makes me wonder when was the last time I needed that (especially with most apps having their own volume control)
This
And especially true regarding Ubuntu lately
No one in their right mind would ship Pulseaudio (amongst other things)
But of course, to hell with the user
Well, since the 'neutrino' name is (kind of) Italian...
Then on the next couple of years we get congestion at the 5GHz band
It's not about band, it's about intelligently using the resources. Unfortunately, 802.11g is very bad at this, 802.11n is better, still, I think there's a long way to go still...
Great points!
I think you nailed it, "rapid prototyping is useful when you only need to make one of something, and you need it right now"
One thing I've heard about 3D printed things is that things made with it are very brittle, still, I know it's used for implants, I think they do a prototype then make it into a mold and cast something with other materials.
If you have a procedure with 10 parameters, you probably missed some.
