A DC motor is using a split commutator to convert the input AC into rectified AC.
If you spin it, it will generate rectified AC, not DC. Actually, due to the leading advance built into the motor, it'll likely not switch at the right point. But you could make it switch at the right point.
Rectified AC useful too, but it's not DC and you can't treat it as such.
Anyway, having a split commutator is problematic when you want to produce a lot of power. It's easier to diode rectify.
To get DC you, um, spin a coil in a magnetic field, then rectify it, then put a huge capacitor on there to flatten out the humps.
There's just no good method for generating DC. And even if there were, electric companies aren't going to run two new phases (DC+ and DC-) to get it to you from the source.
Instead, the power is going to come to near you as 3-phase, then be rectified. There is a loss in that rectification, but sadly, you can't eliminate it, just change where it happens. Moving it to the other side of your power meter will have an advantage since you theoretically wouldn't have to pay for the losses, although the electric companies would surely change their rates to recoup this lost money. But note that even if they don't change their rates, you haven't saved any energy, just not paid for as much.
So my guess is this experiment bought into this fallacy, that they measured their power usage at DC levels, found it was lower and reported that as a win, when without a source of DC power that doesn't involve rectification it really isn't.
I'm sure they save some electricity due to the increased voltage. That reduces current, which decreases power lost. This is the same reason electric companies use high voltages for power transmission.
The article seems to imply that power supplies convert 120VAC to 381VDC internally. This just isn't true. They never raise the voltage, and 120VAC peaks at 175V or something like that. Even 240V input would peak at 350V. So I don't get this. I think they just messed up a few numbers and really in the experiment connected rectified 240V (UK 240V, which is one phase double high, not the US one 120V phase over another) directly into the power supplies after the point where the rectifier would normally be.
From what I can tell, going to DC just would save you the cost of lots of little rectifiers in favor of the cost of one big one. To be honest, since the small rectifiers come in commodity ATX power supplies, you're paying almost nothing for them anyway. So I don't see that it's all that valueable to consolidate them.
I would recommend that if we wanted to save the most power on servers, we should just go to 3-phase 440V AC power supplies. A new connector would have to be designed, as the current 440V 3-phase connector would barely fit on the back of a tower, and wouldn't fit on a 1U server. This would save the most possible in losses without having to buy external rectifiers or force the electric companies to install one on site (and charge you back in increased rates).
The idea that someone's idea of foreign bacon could be colored (coloured?) by them getting hold of a bad example of it. Quite a novel idea. Did you think of this idea before making your post about your views of American bacon?
As an additional note, I think it's quite possible that your equivalence between streaky bacon and American bacon might be flawed. I have had what I though was American bacon outside the US before and disliked it. I wonder now if I really had streaky bacon instead.
The reason I didn't like the bacon I was served is because it was smoke cured instead of sugar cured. American bacon gets much of its flavor from the way the sugar water it is cured in interacts with the fat that comes off as you cook it. When you smoke cure it, none of this happens and it tastes a lot different. It tastes inferior in my opinion, but that is a matter of taste.
Perhaps the streaky bacon is smoke cured instead of sugar bacon and so you can't really judge American bacon from streaky bacon flavor?
One more note is that what you get in the US as Canadian bacon isn't the same as what you get in Canada as back bacon. What you get in Canada as back bacon is what you talk about as English bacon. What you get in the US is even more like just a piece of ham than back bacon is. The US version of Canadian bacon is basically just a thick circle cut of ham. It's usually perfectly circular (like on an Egg McMuffin) and thus doesn't have the fat on the borders. To be honest, it's a very boring cut of meat.
I just thought I'd mention that, because it seems like there's more than just two types of bacon (streaky and back) and I think I found another.
Having had some of your bacon (and being an American).
Your bacon (and Canadian/back bacon) is just a slice of ham.
Honestly, it can't hold a candle to American bacon. Back bacon is barely as tasty as a regular slice of ham, and American bacon is far more tasty.
As to American bacon being horrendeously fattening, well, fattening is a function of calories. If your food has calories in it, it's fattening too. And if american bacon has more calories than your bacon, perhaps you could eat a smaller portion.
Perhaps Beirut has people who just show up at bombsites hoping to get their picture in the paper. It wouldn't surprise me. We have the same kind of media whoring in the US.
This doesn't necessarily mean the photos were faked or taken on the same day.
He apparently knows how to get the most impactful photos. It might involve having friends who will post for him.
I'm sure this is very common, not just from the Hezbullah point of view.
It's one of the fallouts of a freelance system. He wants to earn the most money, which means selling the most photos, which means getting the most striking photos possible.
As to the one guy in a bucket truck and 30 minutes, yeah, you're right. But the type of employees you will get aren't used to working at full speed for 8 hours a day.
The relationship with the county will help with access to poles, presumably for no fee. But I didn't count right of way costs in there. This is pure labor, and the government knows nothing about reducing the cost of installation labor.
Mesh isn't going to work in rural Washtenaw. You're talking 1/2 mile between houses. And using unlicensed bandwidth (802.11a) for your backhaul is stupid anyway. Anyone can legally interfere with your communications and there is nothing you can do about it.
802.16/WiMax will be pretty cheap by the time this network gets rolled out. There's nothing difficult about doing it, current 802.11 chipsets only need a few changes to do WiMax. There's currently no competition in that market, but a few large-scale rollouts and there will be. Plus it seems Intel is going to roll it into their chips soon anyway. Either way, out in the countryside of Washtenaw, you'll have a choice between giving someone a WiMax card or installing one AP per household. The installation costs of the AP will be much higher than the couple hundred for the WiMax card.
6,000APs. Lets say those only cost $200 each (AP + DSL modem). That's $1,000,000 right there.
But honestly, that's the cheap part. People don't just want Wi-Fi, they want Wi-Fi that connects them to the internet. So every AP needs a backhaul connection and it needs to be mounted on a post.
It'll easily cost you $2000 each to get these up on posts. That's if a crew puts up 3 or more a day. And then for every unit that is more than a trivial distance from a phone line, you need to hook up the backhaul wire, that'll cost at least $2000 in the city, and far more out in the countryside (which Washtenaw has a lot of). If you have to run a wire a mile to hook up these access points, that's a couple thousand dollars.
And you still haven't accounted for the costs of the equipment to hook this stuff up to the internet. DSLAMs, whatever.
Out in the countryside you'll be paying at least $5K per household, because the houses are too far apart to share access points.
And in the end, you end up with patchy coverage anyway.
I just don't get why to do this when WiMax is already here. Even using WiMax as the backhaul would save hundreds of thousands of dollars.
I'd also note that Washtenaw County used to be the center of the NSFNET.
From my Mac to my phone, I'm lucky to get 35KBps. That's 250kbps.
Bluetooth has some concept of whether each slot should be allocated for upload or download, if you rig them all the right direction, you could in theory get 700kbps. But you won't ever see that.
I guess from your numbers it might be just fast enough, but it's pretty close. Many phones restrict the data rates between the Bluetooth modem and the cell radio to 115kbps or 230kbps (because they're using RS-232 in there!), which would bone you pretty bad. I think the Treo 650 is limited to 230kbps, with the 700 being a lot faster (in theory).
I have the first Bluetooth (integrated) for Sprint, the Sony-Ericsson T608. It's not crippled, but it's the buggiest phone I've ever used. I switched to Sprint to get it because Verizon was dragging their feet on Bluetooth and I didn't want to leave CDMA. I was right to leave Verizon (with their crippling policy), but I was wrong to go to Sprint. They may not cripple their phones, but their phone selection is truly awful. When I switched away to Cingular, I was using DUN on my T637 long before the Treo 650 even came out (even on GSM), and when I got mine, I was already a year behind those who did it on the T616.
I just can't be held captive to what a provider wants to offer anymore. I need a choice. It's too bad too, because GSM stinks.
But, as we all know, we don't do all that much science here in the USA anymore. If anyone does this, my hope is Haier in China will do it, as that several million dollars should be used here in the USA to buy another executive jet or yacht so the guy at the top whose time is worth several million dollars a year isn't wasted in line at an airport or is inadequately entertained.
What a stupid comment. US companies do a lot more research than Haier does.
Chinese companies do a good job of producing existing technologies more cheaply. But don't look to Haier for new technologies. The key to making commodity products cheaply is not to not spend a lot of money on R&D. Instead they work on minimizing expenditures.
And if you think the execs at Haier don't make a lot of money and live well on it, you're not paying attention to the nouveau riche in coastal China. They're not afraid to spend money on themselves.
"This is effectively a form of lossy audio data compression."
A-law and mu-law are both non-linear 8-bit representations of audio samples. They are generally considered to compress because they provide the same dynamic range as a 16-bit representation but use save space. But the intention of a-law and u-law is to represent the original signal as accurately as possible.
And no, u-law and a-law aren't used to improve the dynamic range of the signal, they are used to preserve it (and as much detail as possible) while using as little digital bandwidth as possible.
A system could be devised that used linear 8-bit samples instead of a-law or u-law, and it would also preserve the dynamic range of the sound, it just wouldn't preserve as much detail in low sounds. A logarithmic representation like a-law or u-law preserves dynamic range and still can represent signals with small dynamic range fairly well.
In short, an a-law or u-law encoding is used to get as much of the fidelity of a 16-bit linear system as possible in the bandwidth of an 8-bit system. It's thus a compression scheme.
It is not a system for improving dynamic range nor a system that is dependent on the frequency ranges of a human voice.
If you're gonna be pedantic about what others say, try to get your own stuff impeachably correct first.
Apple's strategy is not to do what the competition does.
Apple shipped two Yonahs in desktop systems. No other major player (including Dell) did that.
Apple pretty much does what they feel like. For example, they already make an iMac with integrated graphics. It was released a couple weeks ago.
Merom is the same price as Yonah for any MHz rate, at least in the markets we see (who knows what Apple pays). I can't see how Conroe would be cheaper.
Your idea of a 60W thermal envelope is odd. First of all the iMac G5 started with the 90+W G5. Second of all, Apple completely revamped the cooling zones inside the iMac when it went Intel. So any thermal envelope that existed before is invalid now. It could be higher, it could be lower.
Whether Conroe will work in there or not, I think Apple will use Merom.
I think Apple will make an iMac Merom in the study with an X3000 graphics solution. Heck, they'll probably make one with the 3000 graphics too, to replace the current low-end education iMac. I think Apple will at least extend the iMac to a lower-end market, but perhaps not shift it down there completely. Even if customers prefer a tower+display solution, Apple doesn't have a tower that is affordable right now, nor one that fits under most desks.
We're both just guessing though, as the GP pointed out.
I don't think there'd be a heat problem. Yes, Conroe uses more power. But Merom does too.
I think the iMac won't get Conroe for cost reasons. Even if the chips cost the same, there is a quantifiable cost to adding more traces to the motherboard, and since the iMac isn't a top-performer, they don't need to spend that extra money.
Although when Intel's G965 and Q965 express chipsets come out, I think they'll revise the iMac to use those, because the integrated graphics is a lot better than the current 850 stuff. Yes, I think that means they'll dump the ATI in there to save money. C'est la vie.
And 16KHz just isn't that high. You should be able to find many cops who can hear it. At least half of cops on patrol are under 40.
Also, note to the designers of the Mosquito:
Something isn't ultrasonic if humans can hear it. Early TV remotes were ultrasonic. They were 38KHz or 44KHz. That's well above the range of human hearing and humans couldn't hear them.
To those who say this wouldn't show up on a sound meter, A-weighting is only down 10dB at 20KHz and 7 at 16KHz. 7 is a lot, but the difference between audible and annoyingly loud is far more than 10dB, so it would show up.
Like I said, they have to make the checks in any case. Because they need to reduce the detail in order to maintain frame rates on older hardware.
Yes, there are very few or perhaps even no games that absolutely cannot be played on a DX8 card (really driver). Developers don't want to restrict their potential customer base to only people with the newest cards.
I don't understand your response...
on
Halving Half Lives
·
· Score: 2, Informative
The three posts above yours talk about breeder reactors. That is, reactors that can can turn some isotopes into useful fuel as they create energy.
Your post talks about prefab reactors, like the French have been using for years and are improved further (it seems) with pebble-bed designs. These are not breeder reactors.
Also, the US has used breeder reactors. Fermi 1 even operated for a short time as a commercial breeder reactor.
Why do you turn one thing into another?
Toshiba's design uses liquid sodium as a coolant. These designs have been problematic in the past, for example Fermi 1 or Soviet nuclear subs.
I do agree with you that nuclear power is very misunderstood.
For example, your argument about the difference of piecemeal acceleartion in OpenGL versus presence or absence of capabilities in Direct X. This is known as the caps bits (or caps flags) argument. You present the factual part, then you skip over some of the intermediate steps and go straight to the (incorrect) conclusion that you can't run Direct X 9 games if you don't have a Direct X 9 card.
First of all, you can run Direct X 9 games on Direct X 8 cards as long as the games check to see if the card has the new Direct X 9 capabilities (using caps bits) before using them. If the card (or more accurately the driver) doesn't have them, it had better not try to use them or it will fail. The short version is that you can use Direct X 9 games on Direct X cards, they just won't look like Direct X 9 games because they won't have true displacement mapping or some stuff that the card doesn't offer.
This seems vastly different then OpenGL where OpenGL can do the new operations in software and thus you can use the new capabilities even if your card doesn't have them. The problem however is that the moment you use a capability that is emulated in software, the chances are it will be very slow. Draw any frame that uses the capability in over half the frame (like water using displacement mapping) and the frame rate will drop 10X. To a developer, this is death. Gamers will tolerate variable frame rates, but they won't tolerate their frame rate dropping from 60 to 6. The developer will have to avoid using the technique when it isn't present in hardware.
And that, in a nutshell, is the reason that it doesn't make any difference whether you use the OpenGL piecemeal acceleration model or the DirectX caps bits model. The developer still cannot use new techniques that aren't implemented in software. In either case they'll instead use some other technique that doesn't look as good but maintains their frame rate.
OpenGL's system is more academic. That means if you are making less frame-intensive programs (non games), you can render a still life with all the newest techniques regardless of how old your video card is, it'll just take a while. In that way, OpenGL is probably a little better for learners.
You also mix in a lot of opinion. Like that no developer would use Direct3D by choice. That's just ridiculous. Many developers would prefer OpenGL, but by no means all. Like all kinds of other things in life, some people like one option and some like another.
Other things like: >OpenGL 3d drivers can provide 2D AND 3D acceleration.
Yes, that's true from an academic point of view. Many 2D operations can be represented in 3D. Not quite all though. Some operations require access to the current contents of the framebuffer during rasterization that are not possible with 3D acceleration. These things are done in 3D by compositing portions of the image into offscreen buffers and then recompositing them back to the screen. This isn't as fast as the equivalent 2D operations. Other things require information about the pixel registration of the patch compared to the display so they can do things like ClearType does. Finally, I don't know (I could be wrong on this) that it is currently possible to do prepress-type color matching (like Apple's ColorSync) in any current 3D system.
>OpenGL 3d drivers can provide faster 2D acceleration then what the legacy 2D drivers can do. (due to the nature of the hardware GPU, not so much the drivers)
That's patently untrue. If a 2D operation could be done in a faster way using the installed video hardware, the driver for the 2D card would use it. This includes by making use of 3D hardware.
>3D acceleration is a hard requirement for a modern desktop.
Not that I can tell. Anyone who only runs Excel may do no 3D at all, and any 3D they could do could even be done on the main processor instead of a GPU. Many Linux machines have no 3D acceleration at all and they seem to be considered modern desktops.
I don't see why OpenGL should take over the screen completely like DirectX does. Sure, make a standardized API to do it, but why add it to OpenGL?
GSM uses a TDMA infrastructure which is greatly inferior to CDMA.
How inferior? The new GSM 3G standard uses CDMA. That's a pretty ringing endorsement by those who actually understand and design GSM. Perhaps you could take some clues from them?
Also, in HDTV, it's COFDM, not OFDM.
I'd also point out x86 took off in the US when it was up against Apple's 6502-based machines and 68000-based machines back in the early 80s. PPC didn't come around until 1993/4. To say the US chose x86 over PPC is really odd.
And furthermore, it's pretty difficult to argue HD-DVD is technically superior when BluRay and HD-DVD support the same DRM and content formats, and at the physical later, BluRay stores twice as much.
Slashdot Mirror
A DC motor is using a split commutator to convert the input AC into rectified AC.
If you spin it, it will generate rectified AC, not DC. Actually, due to the leading advance built into the motor, it'll likely not switch at the right point. But you could make it switch at the right point.
Rectified AC useful too, but it's not DC and you can't treat it as such.
Anyway, having a split commutator is problematic when you want to produce a lot of power. It's easier to diode rectify.
To get AC, you spin a coil in a magnetic field.
To get DC you, um, spin a coil in a magnetic field, then rectify it, then put a huge capacitor on there to flatten out the humps.
There's just no good method for generating DC. And even if there were, electric companies aren't going to run two new phases (DC+ and DC-) to get it to you from the source.
Instead, the power is going to come to near you as 3-phase, then be rectified. There is a loss in that rectification, but sadly, you can't eliminate it, just change where it happens. Moving it to the other side of your power meter will have an advantage since you theoretically wouldn't have to pay for the losses, although the electric companies would surely change their rates to recoup this lost money. But note that even if they don't change their rates, you haven't saved any energy, just not paid for as much.
So my guess is this experiment bought into this fallacy, that they measured their power usage at DC levels, found it was lower and reported that as a win, when without a source of DC power that doesn't involve rectification it really isn't.
I'm sure they save some electricity due to the increased voltage. That reduces current, which decreases power lost. This is the same reason electric companies use high voltages for power transmission.
The article seems to imply that power supplies convert 120VAC to 381VDC internally. This just isn't true. They never raise the voltage, and 120VAC peaks at 175V or something like that. Even 240V input would peak at 350V. So I don't get this. I think they just messed up a few numbers and really in the experiment connected rectified 240V (UK 240V, which is one phase double high, not the US one 120V phase over another) directly into the power supplies after the point where the rectifier would normally be.
From what I can tell, going to DC just would save you the cost of lots of little rectifiers in favor of the cost of one big one. To be honest, since the small rectifiers come in commodity ATX power supplies, you're paying almost nothing for them anyway. So I don't see that it's all that valueable to consolidate them.
I would recommend that if we wanted to save the most power on servers, we should just go to 3-phase 440V AC power supplies. A new connector would have to be designed, as the current 440V 3-phase connector would barely fit on the back of a tower, and wouldn't fit on a 1U server. This would save the most possible in losses without having to buy external rectifiers or force the electric companies to install one on site (and charge you back in increased rates).
The idea that someone's idea of foreign bacon could be colored (coloured?) by them getting hold of a bad example of it. Quite a novel idea. Did you think of this idea before making your post about your views of American bacon?
As an additional note, I think it's quite possible that your equivalence between streaky bacon and American bacon might be flawed. I have had what I though was American bacon outside the US before and disliked it. I wonder now if I really had streaky bacon instead.
The reason I didn't like the bacon I was served is because it was smoke cured instead of sugar cured. American bacon gets much of its flavor from the way the sugar water it is cured in interacts with the fat that comes off as you cook it. When you smoke cure it, none of this happens and it tastes a lot different. It tastes inferior in my opinion, but that is a matter of taste.
Perhaps the streaky bacon is smoke cured instead of sugar bacon and so you can't really judge American bacon from streaky bacon flavor?
One more note is that what you get in the US as Canadian bacon isn't the same as what you get in Canada as back bacon. What you get in Canada as back bacon is what you talk about as English bacon. What you get in the US is even more like just a piece of ham than back bacon is. The US version of Canadian bacon is basically just a thick circle cut of ham. It's usually perfectly circular (like on an Egg McMuffin) and thus doesn't have the fat on the borders. To be honest, it's a very boring cut of meat.
I just thought I'd mention that, because it seems like there's more than just two types of bacon (streaky and back) and I think I found another.
Having had some of your bacon (and being an American).
Your bacon (and Canadian/back bacon) is just a slice of ham.
Honestly, it can't hold a candle to American bacon. Back bacon is barely as tasty as a regular slice of ham, and American bacon is far more tasty.
As to American bacon being horrendeously fattening, well, fattening is a function of calories. If your food has calories in it, it's fattening too. And if american bacon has more calories than your bacon, perhaps you could eat a smaller portion.
Steve stumbled a bit, that's true.
And the lineup of stuff to announce wasn't mindblowing.
So yeah, I would say yesterday was a bit disappointing.
But I don't think it means Steve's lost his touch or something. Give him a chance to wow you next time.
It does look like the same woman though.
Perhaps Beirut has people who just show up at bombsites hoping to get their picture in the paper. It wouldn't surprise me. We have the same kind of media whoring in the US.
This doesn't necessarily mean the photos were faked or taken on the same day.
He apparently knows how to get the most impactful photos. It might involve having friends who will post for him.
I'm sure this is very common, not just from the Hezbullah point of view.
It's one of the fallouts of a freelance system. He wants to earn the most money, which means selling the most photos, which means getting the most striking photos possible.
Wimax will decrease backhaul cost, like I said.
As to the one guy in a bucket truck and 30 minutes, yeah, you're right. But the type of employees you will get aren't used to working at full speed for 8 hours a day.
The relationship with the county will help with access to poles, presumably for no fee. But I didn't count right of way costs in there. This is pure labor, and the government knows nothing about reducing the cost of installation labor.
Mesh isn't going to work in rural Washtenaw. You're talking 1/2 mile between houses. And using unlicensed bandwidth (802.11a) for your backhaul is stupid anyway. Anyone can legally interfere with your communications and there is nothing you can do about it.
802.16/WiMax will be pretty cheap by the time this network gets rolled out. There's nothing difficult about doing it, current 802.11 chipsets only need a few changes to do WiMax. There's currently no competition in that market, but a few large-scale rollouts and there will be. Plus it seems Intel is going to roll it into their chips soon anyway. Either way, out in the countryside of Washtenaw, you'll have a choice between giving someone a WiMax card or installing one AP per household. The installation costs of the AP will be much higher than the couple hundred for the WiMax card.
6,000APs. Lets say those only cost $200 each (AP + DSL modem). That's $1,000,000 right there.
But honestly, that's the cheap part. People don't just want Wi-Fi, they want Wi-Fi that connects them to the internet. So every AP needs a backhaul connection and it needs to be mounted on a post.
It'll easily cost you $2000 each to get these up on posts. That's if a crew puts up 3 or more a day. And then for every unit that is more than a trivial distance from a phone line, you need to hook up the backhaul wire, that'll cost at least $2000 in the city, and far more out in the countryside (which Washtenaw has a lot of). If you have to run a wire a mile to hook up these access points, that's a couple thousand dollars.
And you still haven't accounted for the costs of the equipment to hook this stuff up to the internet. DSLAMs, whatever.
Out in the countryside you'll be paying at least $5K per household, because the houses are too far apart to share access points.
And in the end, you end up with patchy coverage anyway.
I just don't get why to do this when WiMax is already here. Even using WiMax as the backhaul would save hundreds of thousands of dollars.
I'd also note that Washtenaw County used to be the center of the NSFNET.
From my Mac to my phone, I'm lucky to get 35KBps. That's 250kbps.
Bluetooth has some concept of whether each slot should be allocated for upload or download, if you rig them all the right direction, you could in theory get 700kbps. But you won't ever see that.
I guess from your numbers it might be just fast enough, but it's pretty close. Many phones restrict the data rates between the Bluetooth modem and the cell radio to 115kbps or 230kbps (because they're using RS-232 in there!), which would bone you pretty bad. I think the Treo 650 is limited to 230kbps, with the 700 being a lot faster (in theory).
I have the first Bluetooth (integrated) for Sprint, the Sony-Ericsson T608. It's not crippled, but it's the buggiest phone I've ever used. I switched to Sprint to get it because Verizon was dragging their feet on Bluetooth and I didn't want to leave CDMA. I was right to leave Verizon (with their crippling policy), but I was wrong to go to Sprint. They may not cripple their phones, but their phone selection is truly awful. When I switched away to Cingular, I was using DUN on my T637 long before the Treo 650 even came out (even on GSM), and when I got mine, I was already a year behind those who did it on the T616.
I just can't be held captive to what a provider wants to offer anymore. I need a choice. It's too bad too, because GSM stinks.
On Cingular.
I have to agree that it's just greed.
But anyway, Bluetooth (1.1) isn't fast enough for EVDO, it's barely fast enough for EDGE.
Bluetooth 2.0 support should rectify that.
I buy my GSM phones unlocked so I can use data over Bluetooth. Verizon/Sprint customers don't generally have this option.
But, as we all know, we don't do all that much science here in the USA anymore. If anyone does this, my hope is Haier in China will do it, as that several million dollars should be used here in the USA to buy another executive jet or yacht so the guy at the top whose time is worth several million dollars a year isn't wasted in line at an airport or is inadequately entertained.
What a stupid comment. US companies do a lot more research than Haier does.
Chinese companies do a good job of producing existing technologies more cheaply. But don't look to Haier for new technologies. The key to making commodity products cheaply is not to not spend a lot of money on R&D. Instead they work on minimizing expenditures.
And if you think the execs at Haier don't make a lot of money and live well on it, you're not paying attention to the nouveau riche in coastal China. They're not afraid to spend money on themselves.
unimpeachably correct I mean.
silly me.
http://en.wikipedia.org/wiki/Companding
"This is effectively a form of lossy audio data compression."
A-law and mu-law are both non-linear 8-bit representations of audio samples. They are generally considered to compress because they provide the same dynamic range as a 16-bit representation but use save space. But the intention of a-law and u-law is to represent the original signal as accurately as possible.
And no, u-law and a-law aren't used to improve the dynamic range of the signal, they are used to preserve it (and as much detail as possible) while using as little digital bandwidth as possible.
A system could be devised that used linear 8-bit samples instead of a-law or u-law, and it would also preserve the dynamic range of the sound, it just wouldn't preserve as much detail in low sounds. A logarithmic representation like a-law or u-law preserves dynamic range and still can represent signals with small dynamic range fairly well.
In short, an a-law or u-law encoding is used to get as much of the fidelity of a 16-bit linear system as possible in the bandwidth of an 8-bit system. It's thus a compression scheme.
It is not a system for improving dynamic range nor a system that is dependent on the frequency ranges of a human voice.
If you're gonna be pedantic about what others say, try to get your own stuff impeachably correct first.
Apple's strategy is not to do what the competition does.
Apple shipped two Yonahs in desktop systems. No other major player (including Dell) did that.
Apple pretty much does what they feel like. For example, they already make an iMac with integrated graphics. It was released a couple weeks ago.
Merom is the same price as Yonah for any MHz rate, at least in the markets we see (who knows what Apple pays). I can't see how Conroe would be cheaper.
Your idea of a 60W thermal envelope is odd. First of all the iMac G5 started with the 90+W G5. Second of all, Apple completely revamped the cooling zones inside the iMac when it went Intel. So any thermal envelope that existed before is invalid now. It could be higher, it could be lower.
Whether Conroe will work in there or not, I think Apple will use Merom.
I think Apple will make an iMac Merom in the study with an X3000 graphics solution. Heck, they'll probably make one with the 3000 graphics too, to replace the current low-end education iMac. I think Apple will at least extend the iMac to a lower-end market, but perhaps not shift it down there completely. Even if customers prefer a tower+display solution, Apple doesn't have a tower that is affordable right now, nor one that fits under most desks.
We're both just guessing though, as the GP pointed out.
I don't think there'd be a heat problem. Yes, Conroe uses more power. But Merom does too.
I think the iMac won't get Conroe for cost reasons. Even if the chips cost the same, there is a quantifiable cost to adding more traces to the motherboard, and since the iMac isn't a top-performer, they don't need to spend that extra money.
Although when Intel's G965 and Q965 express chipsets come out, I think they'll revise the iMac to use those, because the integrated graphics is a lot better than the current 850 stuff. Yes, I think that means they'll dump the ATI in there to save money. C'est la vie.
http://paulstamatiou.com/2006/03/02/intel-mac-mini -is-upgradable/
I don't actually keep track, but I think the only things that aren't socketed are the laptops.
There's nothing magic about age 40.
And 16KHz just isn't that high. You should be able to find many cops who can hear it. At least half of cops on patrol are under 40.
Also, note to the designers of the Mosquito:
Something isn't ultrasonic if humans can hear it. Early TV remotes were ultrasonic. They were 38KHz or 44KHz. That's well above the range of human hearing and humans couldn't hear them.
To those who say this wouldn't show up on a sound meter, A-weighting is only down 10dB at 20KHz and 7 at 16KHz. 7 is a lot, but the difference between audible and annoyingly loud is far more than 10dB, so it would show up.
A-law is an audio compression scheme.
GM committed to putting line-in jacks on all their affordable cars standard. Some high end ones too.
n -aux-in-jack-to-their-cars/
http://www.engadget.com/2005/03/18/gm-adds-line-i
They started this a couple years ago, so the Cobalt, Sky, Vue, Ion, Solstice, HHR, G6, Impala, Monte Carlo and many others already have it.
It baffles me why it took so long and why other companies like VW who claim they are in touch with America's youth still don't have them.
Intel has 70% of the market for servers built on personal computer chips.
What a stupid article.
Like I said, they have to make the checks in any case. Because they need to reduce the detail in order to maintain frame rates on older hardware.
Yes, there are very few or perhaps even no games that absolutely cannot be played on a DX8 card (really driver). Developers don't want to restrict their potential customer base to only people with the newest cards.
The three posts above yours talk about breeder reactors. That is, reactors that can can turn some isotopes into useful fuel as they create energy.
Your post talks about prefab reactors, like the French have been using for years and are improved further (it seems) with pebble-bed designs. These are not breeder reactors.
Also, the US has used breeder reactors. Fermi 1 even operated for a short time as a commercial breeder reactor.
Why do you turn one thing into another?
Toshiba's design uses liquid sodium as a coolant. These designs have been problematic in the past, for example Fermi 1 or Soviet nuclear subs.
I do agree with you that nuclear power is very misunderstood.
You mix facts with opinion.
For example, your argument about the difference of piecemeal acceleartion in OpenGL versus presence or absence of capabilities in Direct X. This is known as the caps bits (or caps flags) argument. You present the factual part, then you skip over some of the intermediate steps and go straight to the (incorrect) conclusion that you can't run Direct X 9 games if you don't have a Direct X 9 card.
First of all, you can run Direct X 9 games on Direct X 8 cards as long as the games check to see if the card has the new Direct X 9 capabilities (using caps bits) before using them. If the card (or more accurately the driver) doesn't have them, it had better not try to use them or it will fail. The short version is that you can use Direct X 9 games on Direct X cards, they just won't look like Direct X 9 games because they won't have true displacement mapping or some stuff that the card doesn't offer.
This seems vastly different then OpenGL where OpenGL can do the new operations in software and thus you can use the new capabilities even if your card doesn't have them. The problem however is that the moment you use a capability that is emulated in software, the chances are it will be very slow. Draw any frame that uses the capability in over half the frame (like water using displacement mapping) and the frame rate will drop 10X. To a developer, this is death. Gamers will tolerate variable frame rates, but they won't tolerate their frame rate dropping from 60 to 6. The developer will have to avoid using the technique when it isn't present in hardware.
And that, in a nutshell, is the reason that it doesn't make any difference whether you use the OpenGL piecemeal acceleration model or the DirectX caps bits model. The developer still cannot use new techniques that aren't implemented in software. In either case they'll instead use some other technique that doesn't look as good but maintains their frame rate.
OpenGL's system is more academic. That means if you are making less frame-intensive programs (non games), you can render a still life with all the newest techniques regardless of how old your video card is, it'll just take a while. In that way, OpenGL is probably a little better for learners.
You also mix in a lot of opinion. Like that no developer would use Direct3D by choice. That's just ridiculous. Many developers would prefer OpenGL, but by no means all. Like all kinds of other things in life, some people like one option and some like another.
Other things like:
>OpenGL 3d drivers can provide 2D AND 3D acceleration.
Yes, that's true from an academic point of view. Many 2D operations can be represented in 3D. Not quite all though. Some operations require access to the current contents of the framebuffer during rasterization that are not possible with 3D acceleration. These things are done in 3D by compositing portions of the image into offscreen buffers and then recompositing them back to the screen. This isn't as fast as the equivalent 2D operations. Other things require information about the pixel registration of the patch compared to the display so they can do things like ClearType does. Finally, I don't know (I could be wrong on this) that it is currently possible to do prepress-type color matching (like Apple's ColorSync) in any current 3D system.
>OpenGL 3d drivers can provide faster 2D acceleration then what the legacy 2D drivers can do. (due to the nature of the hardware GPU, not so much the drivers)
That's patently untrue. If a 2D operation could be done in a faster way using the installed video hardware, the driver for the 2D card would use it. This includes by making use of 3D hardware.
>3D acceleration is a hard requirement for a modern desktop.
Not that I can tell. Anyone who only runs Excel may do no 3D at all, and any 3D they could do could even be done on the main processor instead of a GPU. Many Linux machines have no 3D acceleration at all and they seem to be considered modern desktops.
I don't see why OpenGL should take over the screen completely like DirectX does. Sure, make a standardized API to do it, but why add it to OpenGL?
CDMA is technically superior to GSM.
GSM uses a TDMA infrastructure which is greatly inferior to CDMA.
How inferior? The new GSM 3G standard uses CDMA. That's a pretty ringing endorsement by those who actually understand and design GSM. Perhaps you could take some clues from them?
Also, in HDTV, it's COFDM, not OFDM.
I'd also point out x86 took off in the US when it was up against Apple's 6502-based machines and 68000-based machines back in the early 80s. PPC didn't come around until 1993/4. To say the US chose x86 over PPC is really odd.
And furthermore, it's pretty difficult to argue HD-DVD is technically superior when BluRay and HD-DVD support the same DRM and content formats, and at the physical later, BluRay stores twice as much.