As I understand it, the US capacity factor has been improving over the years:

around 88% from 2006-2012, but only 70% averaged from 1970-2009 - http://www.nei.org/Knowledge-C... and http://www.nei.org/Knowledge-C... from https://en.wikipedia.org/wiki/...

I think he's talking about nameplate capacity vs. capacity adjusted for capacity factor.

Nameplate capacity - The power the system generates at full rated capability.

Capacity factor - Actual production divided by nameplate capacity averaged over time. https://en.wikipedia.org/wiki/...

Nuclear stations usually have 80-90% capacity factor as do most other "baseload" plants including coal

Natural gas plants often run intentionally at lower capacity factor since they're usually built specifically for peaking. In the US, that's around 42%

PV Solar is usually only 13-20% (13-15% in MA, 19% in Arizona)

Concentrated solar power often has a lot of "inertia" in the plant along with built-in storage, so apparently CSP in California achieves a 33% capacity factor

Wind is 20-40%

Hydro varies widely since many countries intentionally overbuild nameplate capacity in order to use a hydro dam for energy storage. (I believe Norway's hydro stations operate at a pretty low capacity factor, but this is partly because Norway acts as Denmark's "battery" and is the sole reason Denmark can achieve around 20% grid penetration of wind/solar.)

So if the installations of solar nameplate capacity matched new coal nameplate capacity installations, in terms of actual contribution to the grid, solar is only contributing 20-30% of what the new coal/nuke/whatever plants are contributing. Another way of thinking about it is that you need MUCH more solar nameplate capacity along with a vast improvement in energy storage in order to match a baseload plant such as a nuclear station.

Also note that this is new installations - most gas/coal plants have already been built, and when renovated/modernized they don't count as "new".

The limitation of the consumer nVidia cards is double precision floating point. He may not need that. There are plenty of problems that need only single precision math, the extra precision is wasted. In that case, you don't see much benefit going to the pro cards, certainly not enough to justify the price.

They created new rules very recently about reddit being a "safe space". This is something that is, of course, extremely vague. What the hell is a "safe space"?

So suddenly some long time subreddits are getting banned for violating that. They are all shitty splaces, but then other shitty places seem to get left alone. As such people are rightly saying "What the fuck?"

Basically the rule is an arbitrary one. They are saying "We can ban you if you say things we don't like." Now its their site, they can do that if they wish, of course, but that is why users are reacting so negatively. It isn't a clear rule that is being consistently applied, rather it is deliberately vague and being targeted in a scattershot fashion.

"Oh look at me, I'm so awesome, I don't need that high end technology! I'm just so great and productive that this old stuff is excellent!"

The reason I say that is because I've always seen it on Slashdot. Many people here seem to take pride in using old systems. Even back in the P2 days when a brand new system was still "slow" for a lot of things you'd have people humblebraging on how they were using a 486 and it was fine.

While I'll certainly agree that machines have WAY more life these days (a 5 year old machine is perfectly serviceable at work for most things) it has always been something I've observed on Slashdot. Rather than a bunch of people bragging on the high end hardware they have, as you tend to see on gaming forums, you have a bunch of people bragging on the low end hardware they have.

I've not tried GTA 5 yet but the GTA world is generally very limited to do what it does. A great example would be GTA 3 and Vice City. Open world games that ran on PS2 hardware. Amazing... However they did it by tracking very little. Only things in your FOV and relevant to what was happening (quest NPCs, police chasing you) were handled. Everything else was not there. Turn around and then around again, and traffic would be totally different because it was not tracked off screen. Drops/pickups disappear when you go slightly out of range. Most objects couldn't be interacted with past them being damaged, which would fix offscreen.. Stuff like that.

Fallout/Skyrim track quite a bit, some of it in a very permanent fashion. Granted not all of it is in memory or simulated at one time (there are a certain number of grids simulated at once) but it is still pretty complex. You can go in to an area, interact with things, pick them up drop them off move them around, travel far away, come back and they'll be in the same state you left them.

Not saying clever optimization can't fix some thing, but there's limits. Also there are limits to how much time it is worth spending. Say you can engineer a clever system that uses all kinds of hacks and tricks to reduce what is tracked and how it is tracked, and then you optimize the shit out of it to reduce the space it takes... great but how many man-hours did that take? Is it worth it? Time is money in games, and you don't want to spend it on things unless it is needed.

So if projected sales from the older consoles aren't enough to justify the development costs and/or offset the cuts that have to be made, you don't do it.

They were having real, real problems getting the kind of game they wanted in to the very limited memory of the last gen consoles. Cutting down graphics only goes so far, there are just limits to how large a world you can easily have, and how many things you can keep track of at once. They did a lot of creative things to manage that, but it was causing issues and they were reaching their limit.

Some games scale more easily but the big open world types that Bethesda likes do not do as well. Hence it makes sense to target only the current gen stuff.

If you want to university back in the 80s and you STILL haven't repaid your loans, well you are the one who is failing. Not only was university cheaper, and loan terms better, then but you've had like THREE DECADES to pay it back. Student loans are a bitch, and take a long time to pay back for sure. However unless you really fucked up, you can manage it in 30 years.

A good way to tell is with nickle metal hydride rechargables. They have a 1.2v cell voltage. So does a device work with them? Then it'll work on less than 1.3v. Of the things in my house that take batteries, all but one have worked with them. That device, a swifer wetjet, specifically says no rechargables so I haven't tried.

NiMH batters work well in devices including, but not limited to, remotes, wireless microphones, EOTech sights, laser pointers, cordless phones, headphone amplifiers, and wireless mice.

Every time this happens, the carriers blame the OEM and the OEM blames the carriers.

But when the device is 99% identical to updated devices from the same OEM on another carrier, it becomes very clear who is at fault.

No, they're not.

THEY HAVE NOT RELEASED LOLLIPOP BINARIES FOR THE DEVICE. Which means they are under NO obligation to provided Lollipop kernel source.

They have provided KitKat kernel source for their devices. So they're fully in compliance with the GPL here.

Outdated software - which they have released source for.

Their smartphones are usually pretty well built.

The problem is that software and hardware are so tightly integrated these days that if a company can't produce reliable embedded software, all of their hardware looks like shit.

That's why I never had issues with my Nexus 4 or Nexus 5 but, as stated before, will NEVER touch any non-Nexus LG phone.

Yeah, the Nexus devices often have their own software flaws (oops, considering they're supposed to be reference devices), but nothing nearly as bad as the shitfest that is LG's own software.

Yeah. So far, fracking has been primarily used in areas that are sparsely populated and/or don't have much surface water (e.g. dry with not much rainfall).

There are lots of examples of local contamination... Drinking water supplies in Dimock, PA have basically been destroyed by fracking.

The big controversy is the Marcellus Shale in NYS - It's a massive resource pretty much located over either the Susquehanna watershed, or worse, the NYC water supply - http://en.wikipedia.org/wiki/N...

NYC's water supply is the largest untreated or one of the least-treated drinking water supplies in the world, partly due to much stricter conservation easements of the water supply than in many other areas. Allowing fracking in that region could completely change the results of this report in terms of "widespread" contamination.

See, the problem is, no one has had issues with widespread systemic impacts.

They have issues with an ALARMINGLY high number of local impacts. Also, I wonder if this is just evaulating the actual fracturing process itself, or if it is including things such as companies dumping produced water 100 feet from a stream (It's happened multiple times - they're not allowed to do it, but underpaid truck drivers take shortcuts.)

Also, part of the reason we haven't had widespread impacts is because people who live in areas with large surface drinking water supplies (as opposed to primary drinking water being underground aquifers) have been fighting hard - New York City has one of the largest untreated water supplies in the world, and it is fed by a network of reservoirs and streams upstate. NYS has been good about keeping fracking AWAY from this infrastructure.

It's just a matter of time before those local impacts become systemic if fracking is allowed in more areas.