Forgot your password?

typodupeerror

Comment: Re:It's... OK. (Score 1) 144

by dinfinity (#44043955) Attached to: Google Enables VP9 Video Codec In Chromium

Most of the early results show that, while VP9 isn't better than h265, it's within a percentage point or two. That's not its problem.

Bull. Percentage points mean diddly squat when comparing different video codecs. Visually, VP9 doesn't even come close to h.265 (HEVC).

Convince yourself: http://forum.doom9.org/showthread.php?p=1620230#post1620230

Comment: Re:Copies are not you! (Score 1) 382

by dinfinity (#43999789) Attached to: Dmitry Itskov Wants To Help You Live Forever Via an Android Avatar

You define whether 'you' die in copy/teleportation thought experiments.

No, Reality does, aka cause & effect. Believe what you will, but that won't change reality.

You misunderstand. The discussion revolves around identity and my claim that it only exists by definition. The reality is the physical existence of a collection of elementary particles and their configuration. Nobody is doubting whether the collection of particles would appear in a recognizable configuration in a teleportation thought experiment.

By your logic some random guy on the street IS you if you are killed and that new guy is brainwashed into thinking he is you.. BS.

I never gave you a single definition, yet you base your argument on one.

Just because you don't truly understand the physical phenomenon of Consciousness, that which experiences reality, doesn't mean you can just wave your hands and say that there is no local consciousness.

I never claimed or implied that.

If you are defining "You" as the abstraction you are experiencing at any particular moment then no wonder you are confused.

I'm not confused.

You are not the fleeting patterns your mind feeds your consciousness, you are that which experiences those patterns. You are a physical, pure droplet of Reality itself interfaced in an amazing way with a system able to modulate you with complex patterns.

Poetic writing does not an argument make.

Comment: Re:Copies are not you! (Score 1) 382

by dinfinity (#43999651) Attached to: Dmitry Itskov Wants To Help You Live Forever Via an Android Avatar

It's a great defense in court, too

There actually are instances in which you are not punished for actions of a previous 'you', at least in some countries. But the crux here is that as a society, we have a definition of the 'things' that can be held accountable and punished, and that that definition is pretty much completely independent from whether or not that thing has consciousness. The latter is, of course, what people care about in copy/teleportation thought experiments.

The legal aspects of how to deal with sufficiently advanced AI / cyborgs / uploaded brains is of course a very interesting (and ultimately unavoidable) debate.

Comment: Re:Copies are not you! (Score 4, Insightful) 382

by dinfinity (#43993277) Attached to: Dmitry Itskov Wants To Help You Live Forever Via an Android Avatar

This is the only truly insightful comment in this thread.

Everybody is so hung up on the pervasive illusion of a spatiotemporally continuous consciousness that they forget that nothing on any reasonable macro level even exists without a definition.

For some definitions of 'you', you didn't exist a minute ago. For others, it is perfectly reasonable to assume that there are multiple instances of 'you'. It just happens that those definitions are not as useful to work with in daily life. It is more effective for an organism to have any instance of consciousness feel responsible for the next one that may arise in it and the ones that previously arose in it. We can't prove that our current consciousness is 'the same' as it was yesterday. We can only define that it is.

Which leads to the only reasonable conclusion: You define whether 'you' die in copy/teleportation thought experiments.

Comment: Re:What else can provide enough clean power? (Score 1) 293

by dinfinity (#43991079) Attached to: <em>Pandora's Promise</em> and the Problem of "Solutionism"

I was going for a kind of worst case scenario (apart from using the average household), so yes: going completely off-grid using only solar is very doable (in Western countries) today.
- Most European households consume at most half the power of US households.
- The figures I quoted for Li-Ion batteries were on the low end (low power / USD and low power / liter).
- Nighttime+evening power consumption is only part of the quoted daily power consumption.
- Wind power and other renewable sources of power were ignored (although this means being on-grid).
- Efficiency of power generation during the daytime also increases by using batteries as a local buffer (vs transferring excess power to the grid or worse, discarding it)

These oversights with positive influence are of course partly mitigated by these:
- To retain current convenience levels, the batteries and power production must support the worst case. Being left without heating in cloudy winters is not an option.
- The cost of the system regulating the interaction between appliances, solar panels and batteries was omitted.

All in all, though, it seems pretty clear that technically and economically, a society almost completely powered by renewable energy (and specifically solar) is very realistic.

Comment: Re:What else can provide enough clean power? (Score 1) 293

by dinfinity (#43985647) Attached to: <em>Pandora's Promise</em> and the Problem of "Solutionism"

Wind, solar and geothermal can't ramp up fast enough to meet power demand, AFAIK.

When considering total power demand, that is probably wrong (2027 is probably a bit optimistic):
http://en.wikipedia.org/wiki/Growth_of_photovoltaics

Of course, demand response is an issue.
Solving peak demand during the day is easy: have installed capacity equal or surpass peak demand. Turning solar panel power production on and off should be pretty much instant.

Nights are the obvious real (technical) issue for solar. Probably the most viable solution is introducing buffers of batteries, hydrogen or carbohydrates.

Let's look at the matter from a household perspective. According to what I can find, the average US household uses about 15,000 kWh annually. Let's go naive and assume those are evenly spread across the year: 15000/365 =~ 41kWh/day.
Current Lithium-Ion cells are about 250Wh/liter and about 2.5Wh/USD. That means that the volume of cells required to store a day of electricity usage in the average US household is about 41000/250 = 164 liters (dm^3). Stacking these blocks one high, we end up with a volume of roughly 1.3m x 1.3m x 0.1m (which is obviously negligible). The naively calculated cost of such a storage unit would be 16.400 USD.

My calculation seems to be fairly accurate: http://www.wholesalesolar.com/battery-banks.html

Now the above is based on current day technology, supply and demand. I'm pretty sure technological progress will make the above economically even more viable.

Comment: Re:About to change (Score 1) 316

by dinfinity (#43842469) Attached to: Console Manufacturers Want the Impossible?

The problem is that different graphics cards give different output for the same input. So a pixel that would be white when calculated by an Nvidia shader, is black when calculated by the Intel integrated graphics shader, resulting in an infinite death cycle at an early stage of the game.

You'll probably agree with me that that doesn't sound like a very robust game engine.

The Skyrim glitches are an odd case. In a sense, they're unforgivable. But in another sense, the scope of that game is so huge, if they polished it all, they'd never finish.

I'm pretty sure Skyrim isn't the only game released on consoles that required patching: http://www.xbuc.net/ (Xbox game patches site)
Of course the point stands that developing for a single target hardware platform is easier than for the multitude of PC configurations out there.

Comment: Re:About to change (Score 1) 316

by dinfinity (#43841499) Attached to: Console Manufacturers Want the Impossible?

Although your point is kind of valid, Closure looks like it could run on a fist-generation Pentium with a 2MB ATi VLB card. Apparently their programmers are just terrible.

Terrible programming generally fails on consoles too:
http://answers.yahoo.com/question/index?qid=20120408105239AA6tkkR

What would help for PC gaming is if all game programmers could rely on some set of metrics provided by a unified benchmarking application. Creating settings profiles that would 'just work' and even predicting whether a game would run (well) on your system would become fairly trivial.

System going down in 5 minutes.

Working...