In fact, Philips heads are quite common for machine screws. You don't usually find wood screws chrome plated, and it would be a very odd wood screw that had such a shallow head. Wood screws by necessity cut and tap a hole in the wood by themselves, to they have to take a much stronger force from a screwdriver. Thus, deeply cut heads. Machine screws are threaded into pre-tapped holes, usually in some other metal, so they don't take a significant force until the last bit of tightening. Thus, they're fine with much shallower heads.
I don't think these are countersunk at all, or intended to be. They look like pan head machine screws, not flat head. A final version of this might use flat heads that sit flush in a shallow countersink. A wood screw would have a much deeper, more gradual taper for the countersink anyway, which would have them sticking way the hell out from this unit, not neatly on top as they are.
Those original prototype Amigas (with the 5.25" drives and the expansion chimney) were in metal cases, not wooden. These generally had the older chips, Portia instead of Paula, Daphne instead of Denise, as I recall.
This is a natural part of product development... final, consumer-ready products don't spring to life fully borne, and in the case of something like the Amiga, the developer's units (which only went out to a handful of developers) were designed to get hardware into hands as quickly as possible, rather than waiting for final system details, final silicon, shipping OS, etc. Developers who get these things understand this well.
As for the nVidia thing... those were hardly wood screws, those are chromed machine screws most likely. I've been a carpenter and cabinetmaker for longer than a computer designer... and that, pretty long. Does this guy even know what a wood screw looks like?
And as well... demonstrating a "Fully Functional Fermi" is hardly the same as claiming to be demonstrating a production ready board. They're not necessarily claiming a final PCB design, final hardware around that PCB, completed drivers, or even finial silicon. Anyone in the actual hardware business would understand this. People who don't understand the development process should maybe stick to videogame reviews.
Actually, you want discrete graphics for anything really graphically intensive: 3D CAD, even video work can be GPU-accelerated these days (in fact, you can't really play full 1080/60p video on even very fast machines, and possibly 1080/60i depending on the PC, without multithreaded rendering and GPU-acceleration). Not that you need this for some kinds of development work, but "Developer" covers a broad area.
Obviously, this is up to the individual, but development machines often need more resources, of many kinds (I/O, HDD, CPU speed, screen resolution, and GPU speed) than the average consumer stuff. I wouldn't imagine doing video work with anything less than a nVidia 9600, at least looking forward. The video acceleration APIs were greatly improved in Vista (and might actually get supported come Win7), as well as nVidia and AMD/ATi's own private interfaces. At least one of these are supported by all modern video players, and should be coming to NLEs in this next generation. Video rendering, as well, is starting to use GPU processing as an accelerator (this is in TMPGenc already, and will very likely be moving in as a standard thing).
Well, back in December of 2007, I did the comparison. I looked at the 17" MacBook versus the 17" HP I eventually bought. The HP exceeded the MacBook in every aspect except two: the MacBook offered 800Mb/s Firewire, versus 400Mb/s on the HP. And it could be had with a 2.6GHz processor -- only Apple was getting those back then, at least that month... and it was something like $300 more than the 2.4GHz processor. The HP ran me $1280, the Apple was $2999. And I don't believe the Apple was of greater build quality in any significant way.... I'm not gentle on my laptops, most only last me about two years, but this one is coming on on the two-year anniversary still going strong.
My HP dv9500 (older model... but same applies to newer ones) offers your choice of VGA or HDMI, GbE, three USBs, Bluetooth and 802.11n, and some kind of docking connector. Given the internal stuff, a docking module was never that interesting, this may have been traded for eSATA on some of the newer models, which I'd find more useful, personally. This has a 1680x1050 display... I'd go for 1920x1200 today, and if I could afford it, quad instead of dual processor. This one ran me $1280 back in December of 2007.
And, like most 17" models, dual internal drives (I cranked it up to 640GB... that includes both 64-bit Vista and Ubuntu Studio, with extra partitions for various sorts of video capture, though that's a bit low compared to the 3TB internal on my desktop, clearly one can go higher nowadays) and a flash card reader (unless you're married to Compact Flash). And Firewire. This is the unit I use very often for both HW and SW development.
Hydrogen holds some promise, but it's questionable right now. It's green to burn (or otherwise use) Hydrogen, just as it's green to use electricity. Both have the same original problem, though... you can't mine or otherwise locate sources of hydrogen anymore than you can do so with electricity. H2 is just a chemical answer for the battery.
Now, what you left out.. the big piece... is how that H2 is converted to electricity. Are you buring it, or feeding a fuel cell? The Fuel Cell is great idea... over 65% efficient, no "burning", thus, few if any pollutants (you would still have NO2 and other pollutants burning H2). We've been making these for a long time to power spacecraft... but they have the budget for it. Traditional fuel cells use lots of Platinum... same problems as large BEVs... no one wants to spend $150,000 on an economy car. Newer designs with engineered materials are promising, but there's more work to do. H2 storage is another issue... compressed gas is a hazard and also limited in capacity, while chemical storage (very similar to a NiMh battery) is higher density, but the cells wear out.
And you still want this to be a hybrid... a fuel cell likes to deliver a steady power output, it's not surgey at all.
Then there's the production of the H2... where does it come from? Like electricity, you can make it many ways... like, from electricity mixed with water to release H2 and O2. But that's not terribly efficient. You can make it from petroleum products, or from alchols, but there are also efficiency issues. In fact, very similar to those of the battery EV world.
And there's also the infrastructure problem. H2 refueling might be faster than electric recharging (it is now... it won't necessarily always be). Power distribution would ultimately have to be beefed up to support a BEV infrastructure, but it does exist today. H2 is non-existant... no one's building fueling stations unless they're in on the experiment.
While that's true, you have to figure "well to wheel" for both technologies. For ICE cars, you're lucky to get 15%.
For electric, the car itself rocks... the typical electric motors (they're all three phase AC motors, Toyota's use permanent magnets, other EVs like the Tesla use linear induction motors) are over 95% efficient. You drop a bit charging then discharging the battery, but that's at least 85% efficient. And a bit is lost to heat in the inverter. Still, very good.
Operating the vehicle, the EV gets more efficiencies back.. electric motors have peak torque at zero RPM.. they have a nearly ideal power curve for driving, so the practical benefit exceeds what you see from the motor and battery efficiencies, particularly when you factor in regenerative braking (which is also a wonderful thing for the back-up friction brakes.. my 2003 Prius has 115,000 miles on it, and the original brakes). On the downside, EVs are heavier than ICE cars, due to the batteries (some of which is offset by the weight of the motor... the Tesla Roadster delivers supercar performance off the line with a 125lbs motor about the size of a big watermelon).
You also lose an average of 7.2% in the process of distributing electricity.
I did the back-of-the-napkin math on this once, based on good electric car efficiencies and the typical use of cars. If just the passenger cars all went electric (not other forms of transportation), the US power output would have to at least double.
And it could be worse... moving peak demands into overnight, when normally peak-friendly power sources like solar and wind don't work so well.
It's virtually impossible BEV use would grow fast enough to cause problems, but if we all woke up on Christmas morning with BEVs in our garage, it would destroy power distribution as we know it.
Most likely it's the pollution rationale.. that's all that cities care about. They really don't give a hoot about gas use, just pollution release. If they get two people in a car, that's half of the pollution, even if it's two in a Hummer versus two Hummers (obviously, not a perfect system if the compact owner carpools in the Hummer-guy's Hummer all the time).
Prius entered the market at SULEV level, and now resides as PZEV. That basically means that, particularly within a city, it moves sometimes without making any pollution, and it's really clean otherwise. The cutoff for SULEV means the car produces 90% fewer emissions (in detail:.01 grams/mile hydrocarbon emission, 1.0 grams/mile of carbon monoxide emissions,.02 grams/mile of nitrous oxide and.01 grams/mile of particulate matter emissions) than a "regular" car. SULEV was originally California-only, now recognized in New York, New Jersey, Vermont, and Massachusetts as well.
I do agree that this should be entirely based on the pollution certs of the car, not the type of car. The Yaris only meets ULEV last I checked, but if a non-hybrid meets the same specs as the Prius, it ought to get the carpool rights... and if a hybrid doesn't, it ought to be banned. Stupidity about the implementation rather than the desired result is what caused that whole electric car fiasco in California. There's one Ford Focus build that meets this spec, non-hybrid.
Anyway, the magical juice is really low emissions. That was the original goal of the Prius, not strictly high mileage (the two are related, but you can optimize for either one... the original 67mpg Honda Insight with manual transmission only met ULEV standards). The SULEV is based on a level of pollution equal to or less than the level an electric car would produce, based on the average pollution (ish) due to power generation in California back to sometimes in the mid-1990s, when these standards were all worked out.
NiMh cells don't have an aging issue, at least not anything significant. There have been plenty of Prius out there at over 200,000 miles, still going strong (and it's not an option.. if the traction battery dies, you can't even start the Prius.. all the 12V battery does it boot the computers).
Li-ion, in its typical forms, unfortunately have had aging issues.. a Li-ion cell will usually lose some important bit of its capacity just sitting there. It'll also self-destruct if discharged below a threshold, and most of the Li-ion chemistrys don't handle as many charge/discharge cycles as NiMh. But the energy density of Li-ion blows away NiMh.
Both of these cells last dramatically longer when only partially cycled. The 2001-2003 Prius cycles from 40% to 80% of battery capacity. Toyota increased the range to 60% of capacity for the 2004 model (and cut out a bunch of the cells, so the effective power is the same). Cycling that way, the "number-of-charge/discharge cycles" figure goes away. This is apparently true of Li-ion cells. Naturally, more practical in a hybrid than a full BEV.
The holy grail is certainly improved battery technology. But for a BEV in the long run, while 500 miles sounds great, that's not the big problem. First big problem is battery life.. if I have to spend $10,000 or even $3,000 every 3-5 years on a new battery pack, that's going to be a significant problem for most potential buyers. There has already been lots of promising work on engineered materials for long-life, if not yet immortal, anode and cathode materials. Many of those also improve the charge rate and discharge peaks, both good things.
Charge rate isn't a big deal if you're charging in your garage. Certainly NiMh supports 4*C charging (though not necessarily without reduced life), which gives you a charge in 15 minutes... still too long. But you don't have remotely enough power coming to the typical home anyway, so that's a moot point, and you generally have all night for a charge once that car is at home.
But ultimately, for BEV to succeed as a realy alternative to ICE, it has to, well, be a real alternative to ICE. I need to be able to stop somehere and "fuel" up, and in some reasonable amount of time. If not, BEVs may be a very nice business, but they're always be a niche.
Prius efficiency has nothing whatsoever to do with moving the efficiency point to 20mph.
Yeah, you get more mileage in city-type situations... at least for the duration of an EPA test. That's the effect of running on electric-only for a fair percentage of the time, and particularly, off the line, where electric is so much more efficient than gasoline.
The big issue with the Prius that people have is the same issue they have with every other car -- they just don't know it with most cars, because they don't have mileage computers. The EPA numbers are based on slower travel. If I drive my 2003 Prius (less efficient than the newer models) at the posted speed limit, with 40psi in the tires and other basic good sense, using a little feedback from the computer, I get over 50mpg, unless it's the dead of winter. If I drive like a bandit (which, alas, is more typical).. 75-80mph on the highway, then I'll see those 42-46mpg results. Duh! Even with the Cds of the Prius, you still have a significant component of turbulent airflow... that component is based on the square of your speed.
What you're missing is the whole basic understanding of how the Prius works. There's a small win on efficiency due to using the Atkinson cycle on the ICE, rather than the Otto cycle.. more efficiency, less peak power -- which you don't need, because the MG2 motor provides that for peak power demands. So the ICE doesn't need the same power peaks. Then there's the transmission... which doesn't exist. A set of fixed gears, two motors, and a computer that can dial in the peak possible RPMs out of the engine for any given load requirement. And power normally wasted by the ICE, and by braking, is put in the battery.
Also, keep in mind that the original (G0/G1) Prius was designed primarily for low emissions, with fuel economy secondary. Compare that to the Honda Insight, the original model designed for maximal mileage (1L 3-cylinder engine, electric boost, well over 60MPG in regular use, beyond that if you drive hypermiler style).
Why the Prius mileage isn't better is simple -- ICE engines just suck. A typical car gets about 15% efficiency on a good day. A Prius is likely more in the 25% range... still sucks, compared to a 95+% efficient electric motor coupled to an 85+% efficient battery.
"Good writers borrow from other writers. Great writers steal from them outright." -Aaron Sorkin
"Immature poets imitate; mature poets steal." -T. S. Eliot
"Plagiarize
Let no one else's work evade your eyes
Remember why the good Lord made your eyes
So don't shade your eyes
But plagiarize, plagiarize, plagiarize
Only be sure always to call it please "research" -Tom Lehrer
Before guns, you basically needed one guy with a sword to fight another guy with a sword. Give one of them a gun, and now he can kill dozens of guys with swords, without even necessarily taking on much personal risk himself. So the killing is more likely, just because it's possible where once it wasn't.
Before computers, you basically needed one guys with a phone tap to spy on one guy on the phone. But add in speech recognition software, and you might have dozens of taps being monitored at once, with a human only standing around to listen with the tapping program thinks it has something of interest. So the tapping in more likely, just because it's possible where once it wasn't.
Same story with the advent of face recognition software, surveillance drones, robots, all kinds of things out there that allow one person to do the work that formerly required many. This isn't anything all that new... the only real advantage is that GPS surveillance (and keep in mind, there has to be a second radio too to do the reporting back to the cops... a GPS itself is just a receiver) is ridiculously easy to defeat with jammers. And a planted GPS transponder is going to be relatively easy to find, if you're looking for it.
This is pretty scary sounding in the Big Brother way, but extremely easy to defeat.
A GPS only works with a clear view of the sky... that means, your GPS bug needs an antenna somewhere, you can't simply hide the whole thing under a car.
You also need a backchannel -- cellular modem or whatever you have access too... police and Feds probably have their own private frequencies to use, but then they'd have the problem of range... you're not going very far on high frequencies at all, or low frequencies without a pretty obvious antenna. So you can probably bet most would use cellular modems, unless they also have you under active surveillance.
Both GPS and cellular are crazy-easy to jam. GPS in particular.. GPS chips are built with a -140dB or so input sensitivity. There are no naturally occurring noise sources at GPS frequencies, but you could build a tiny, very low power noise source and keep any GPS signals from working near your car. In fact, you can probably just buy one of these easily enough: http://boingboing.net/2009/07/30/gps-jammer-plugs-int.html http://www.gpsjammers.net/gmc07.html
Yeah, guess you can. These are all blocking the L1 band for consumer GPS, not the military band, but chances are, that's good enough to block the police, and the military frequencies aren't any more difficult to block, if perhaps less off-the-shelf. Same goes with cellular jammers... you can buy these pretty easily.
Now, obviously, the Good Guys hope that the Bad Guys are not as smart about these things... and they won't be, initially. But soon enough.
All RAID is software RAID. The only difference between "hardware" RAID and software RAID is where the software runs.. does it get its own CPU or not? And there are plenty of applications for "hardware" RAID... external "box of drives" when your PC is full, large NAS (I have one of these), SANs (not yet), etc. From the prespective of the RAID hardware, you may well be running Linux with ZFS or BTRFS or something entirely different, and that's software... but from the external clients prespective, it's hardware RAID. And this is the way its always been done... there were "hardware RAID" cards for the PC's ISA bus... I recall one had an embedded SPARC processor running the RAID software, back in the mid-80s or so.
Yes, it's true... I do run a RAID on my personal PC (well, in a little box next to it), and it's also true, I can't afford a 6TB SSD. Sad, but true.
However, I also know a secret... you can re-write a sector on an HDD many, many times more than the same physical sector on an SSD. The SSD gets you around the mechanical issue of seek times, but it doesn't get around the idea that things fail. And, given that SSDs usually do wear levelling to ensure no one sector wears out prematurely (depending on the specifics of the NAND flash used in the SSD, whether it's MLC or SLC, etc... some MLCs are only good for a few tens of thousands of writes), by the time you start to see a failure or two, chances are, the whole thing's ready to go. They're also typically slower at dealing with writes, particularly many small writes, since you have very large block sizes at the flash-memory level (up to about a megabyte) on large SSDs.
Obviously, there are plenty of usage patterns that would yield a mechanical failure from an HDD long before there's a flash cell failure from an SSD. In both cases, the realistic life expectancy may be fine... I don't expect an HDD to last forever... I'm happy when the last to the point at which a replacement's cost is essentially a "no brainer" (eg, when 2TB drives start showing up below $100 or so, my 6TB RAID is likely to become an 8TB RAID).
The FOSS world has payment, too, it's just a different form of currency... sometimes. You have company-supported open source projects, paid in cash just like any other software engineering work. And you have the volunteers, who work on projects for fun, renown in the FOSS community, "building one's own toys"... eg, it's a hobby, or a technical challenge they may not find in the day job. But they all have day jobs. A sterling reputation in the FOSS community doesn't put food on the table... even Linus has a day job (he was at Transmeta, at least in their better years).
You don't understand the beauty of the Zune marketplace!
Of course practically no one bought the original Zune at the original price... but when I bought my brown Zune for $75, that was a damn good buy for a 30GB player at the time. And say what you will about Microsoft in general (and I usually do), but they've keep the upgrades getting better, and kept them free. That doesn't suck.
As for the dock, when I got my Altec-Lansing dock for the orginal Zune for $30, I got to laugh at all the iPodden out there buying the very same dock with iPod connector for $120 or so.
Hopefully the same things happen for the Zune HD... MS already released it at a somewhat decent price. So no one buys it, the $220 model drops to $120 or so, the $90 dock (which is MSRP, anyway) drops to $30, and then I upgrade.
Anyway, if you really want a PDA+MP3 player, you're probably in line for an iPod anyway... as long as you're happy running applications only blessed by The Steve. Otherwise, go with an Android device. But if you just want a decent PMP, the choice right now has to be between the Zune HD (assuming they keep the quite-a-bit-better-than-iPod sound quality of the originals) and the Cowon S9. Both have OLED screens, which pwn the iPod ten ways to Sunday. When you can have OLED, you don't want LCD... period.
As for free applications... does anyone here really think there's a free lunch, particularly from companies like Apple or Microsoft. When they promised free games, my initial assumption was that a few freebies would come out, either trivial ones or those that function as ads for paid games... just like you have on the iTunes store (eg, it's either relative crap, or you pay for it). There is absolutely no way they would have a continued supply of new games and applications being developed with no chance at revenues of some kind... same reason you get ads on Google and YouTube and pretty much anything else related to Web 2.0 hysteria. Oh yeah, and slashdot!
That they chose an advertising model, rather than a cash-up-front-model, is fine with me. I'm adult enough to not expect a free lunch from Microsoft.
So, you're suggesting Apple get into the business, and produce a few dozen lenses to go with their DSLR? If not, there's zero point in building a DSLR... not that Apple would be taken seriously.
That's also not a market that Apple would be or should be interested. Keep in mind that over 80% of the DSLR market is owned by Canon (41%) and Nikon (40%).. they're the guys who make those dozens of lenses and accessories that make the DSLR worth the price in the first place. The rest is being sliced up between Pentax, Olympus (6%), Fujifilm, Sony (6%.. they're one of the big four in high-end video cameras... the bought up what was Minolta-Konica's SLR works), just recently Panasonic (they're one of the other major powers in high-end video cameras, along with Canon and JVC), and a few others in specialty markets (Leica, Ricoh, Hasselblad). Serious users aren't likely to buy a DSLR from an electronics company without years of proof in the business.. that's why some of video camera companies compete, but few of the other film or CE companies who make digital P&S models (Casio, Epson, HP, Kodak, etc).
The global market for DSLRs in 2010 is expected to be a bit under 12 million units (that's an estimated 9% growth in DSLR shipments)... so that's 4.92 million Canons, 4.8 million Nikons, and 2.28 million left for everyone else to fight over. Do you really think that's an iPod-like market? Apple sells nearly twice that number of iPods... every MONTH.
No, Apple's doing the right thing here... "embrace and extend". They made a decent MP3 player, but really made the sale by delivering the iTunes store. Lots of people make better MP3 players, so Apple pushed in photos, then video. Lots of people came out with better PMPs, so Apple added PDA function, and a phone. Now they toss in a "free" webcam (some prespective here.. this is a crappy, crappy camera for any serious use, but it's fine for the kids buying crappy Flip cameras and delivering video only for Facebook and Youtube) and push the iPod/iPhone as a gaming platform, taking on the Sony PSP and the Nintendo DS. That's the smart thing... that's a $1.00 camera added to the basic stuff the iPod already has. Not a couple of billion to develop a line of lenses no one will ever buy.
Certainly, such "free" cameras will get better over the year, but the iPod is competing directly with the "Flip" and other web-cam + flash units, and to an extent with cell phone cameras. They aren't even close to being as good as a modern point-and-shoot model from Canon, Panasonic, Sony, Nikon, or many others. And they never will be.
... tried the first one a few years back, it just didn't "pop". Can't answer for the whole series, of course.
On the other hand, Brandon Sanderson is a fine writer/storyteller IMHO... I loved the "Mistborn" trilogy. So if you are a WOT fan, the series might well go out with a good story. Finally.
Said it once, and I'll say it again. CDMA is a dead end, the world is moving to LTE. Why would anyone waste their resources on a technology with such a limited lifespan.
3G is now (HSPA or EvDO).. and of course, they're rolling it out still. LTE is 4G, it's a sea change, and such rollouts take time. Verizon is starting next year, on 700MHz, but they won't be finished for several years at best. AT&T is planning trials next year, with the real rollout starting in 2011.
Now, get some perspective... the iPhone is a product with a 2-year expected life. Apple introduces a new model once a year.. the product doesn't even last two years in the channel, just ni your pocket. And Apple's never been worried about keeping up with technology on it, anyway. So, even if they launch a CMDA phone in 2010, most places won't have LTE coverage even when those phones are retired. You'll be trading in your second generation CDMA iPhone for an LTE iPhone, if you're lucky.
Why do it? Money. CDMA isn't new technology... it's a relatively minor revision to an existing technology. You don't think a company like Apple can get QualComm engineers to provide a guaranteed works-first-rev CDMA solution, and get Verizon to pay for it, assuming the agree on a deal? If you doubt this for a femtosecond, you haven't worked in hardware engineering as long as I have (26 years). Apple isn't going to make a CDMA phone just hoping someone will buy into it, as they did with the first iPhone.. but they've established a market and a brand. If Verizon want the iPhone, and can deal with whatever hoops Apple wants them to jump through, they'll get the iPhone next year.
There's not even a tiny question about this.. the development costs are insignificant compared to the money involved. Apple themselves spends several times more money just on ads for any of their products then they spend in product development. As do most hardware companies.
Keep in mind, cellphone vendors aren't designing the cellphone chips themselves, in most cases (and certainly Apple's case), any more than Dell or Acer are designing their CPUs or Wifi chips -- they get a ready-made, working solution from the chip vendor. At worst, they're doing some of the baseband processing on their own host CPU, and they'll port code from the chip supplier to that host CPU. The iPhone, as you might expect, has a separate baseband processor... baseband doesn't run on the application processor.
The world will move to LTE. That hasn't really started yet. There are plenty of locales that don't even have decent 3G coverage yet (either kind). Bringing the iPhone to a CDMA network is a next-year thing, and it'll happen if there's a business case for it. This is not a 5-year investment, it's a small change to an existing product.
Verizon is moving to LTE for 4G, just like everyone else... well, except Sprint, who's already on WiMax (they're virtually the same downstream, anyway). Even QualComm, the originator of CDMA2000 and other atrocities, gave up their 4G plans, and will be integrating LTE into their chips for CDMA networks, just as all the GSM chips will. It's not much of a problem for anyone.
There are no iPhones on LTE, and in fact, they're kind of dodgy on 3G today. When Apple's no longer bound by their exclusive deal with AT&T (it runs out next year, unless there's some kind of renewal agreement... which, given the iPhone's success, would cost AT&T dearly), don't even be slightly surprised if there's a CDMA/EvDO iPhone ready to go. It's not that big of an engineering deal -- one chip in the system and some new baseband software. And Apple can, at this point, get Verizon to pay for that development if they're interesting in carrying the iPhone.
Apple, on the other hand, has much to gain from such a deal.. Verizon is slightly larger than AT&T, so they could, in theory, double their user base going to Verizon. It's easy to claim they'll offer the iPhone through T-Mobile, and they might well, but that wouldn't help them much, other than to maybe force from contract differentiation into the market. And it wouldn't even change much there, if Apple gets the same kinds of kick-back deals from other telcos that it got from AT&T (which is the factor that pushed the exclusivity deal into 2010).
Slashdot Mirror
In fact, Philips heads are quite common for machine screws. You don't usually find wood screws chrome plated, and it would be a very odd wood screw that had such a shallow head. Wood screws by necessity cut and tap a hole in the wood by themselves, to they have to take a much stronger force from a screwdriver. Thus, deeply cut heads. Machine screws are threaded into pre-tapped holes, usually in some other metal, so they don't take a significant force until the last bit of tightening. Thus, they're fine with much shallower heads.
I don't think these are countersunk at all, or intended to be. They look like pan head machine screws, not flat head. A final version of this might use flat heads that sit flush in a shallow countersink. A wood screw would have a much deeper, more gradual taper for the countersink anyway, which would have them sticking way the hell out from this unit, not neatly on top as they are.
Those original prototype Amigas (with the 5.25" drives and the expansion chimney) were in metal cases, not wooden. These generally had the older chips, Portia instead of Paula, Daphne instead of Denise, as I recall.
This is a natural part of product development... final, consumer-ready products don't spring to life fully borne, and in the case of something like the Amiga, the developer's units (which only went out to a handful of developers) were designed to get hardware into hands as quickly as possible, rather than waiting for final system details, final silicon, shipping OS, etc. Developers who get these things understand this well.
As for the nVidia thing... those were hardly wood screws, those are chromed machine screws most likely. I've been a carpenter and cabinetmaker for longer than a computer designer... and that, pretty long. Does this guy even know what a wood screw looks like?
And as well... demonstrating a "Fully Functional Fermi" is hardly the same as claiming to be demonstrating a production ready board. They're not necessarily claiming a final PCB design, final hardware around that PCB, completed drivers, or even finial silicon. Anyone in the actual hardware business would understand this. People who don't understand the development process should maybe stick to videogame reviews.
Actually, you want discrete graphics for anything really graphically intensive: 3D CAD, even video work can be GPU-accelerated these days (in fact, you can't really play full 1080/60p video on even very fast machines, and possibly 1080/60i depending on the PC, without multithreaded rendering and GPU-acceleration). Not that you need this for some kinds of development work, but "Developer" covers a broad area.
Obviously, this is up to the individual, but development machines often need more resources, of many kinds (I/O, HDD, CPU speed, screen resolution, and GPU speed) than the average consumer stuff. I wouldn't imagine doing video work with anything less than a nVidia 9600, at least looking forward. The video acceleration APIs were greatly improved in Vista (and might actually get supported come Win7), as well as nVidia and AMD/ATi's own private interfaces. At least one of these are supported by all modern video players, and should be coming to NLEs in this next generation. Video rendering, as well, is starting to use GPU processing as an accelerator (this is in TMPGenc already, and will very likely be moving in as a standard thing).
Well, back in December of 2007, I did the comparison. I looked at the 17" MacBook versus the 17" HP I eventually bought. The HP exceeded the MacBook in every aspect except two: the MacBook offered 800Mb/s Firewire, versus 400Mb/s on the HP. And it could be had with a 2.6GHz processor -- only Apple was getting those back then, at least that month... and it was something like $300 more than the 2.4GHz processor. The HP ran me $1280, the Apple was $2999. And I don't believe the Apple was of greater build quality in any significant way.... I'm not gentle on my laptops, most only last me about two years, but this one is coming on on the two-year anniversary still going strong.
My HP dv9500 (older model... but same applies to newer ones) offers your choice of VGA or HDMI, GbE, three USBs, Bluetooth and 802.11n, and some kind of docking connector. Given the internal stuff, a docking module was never that interesting, this may have been traded for eSATA on some of the newer models, which I'd find more useful, personally. This has a 1680x1050 display... I'd go for 1920x1200 today, and if I could afford it, quad instead of dual processor. This one ran me $1280 back in December of 2007.
And, like most 17" models, dual internal drives (I cranked it up to 640GB... that includes both 64-bit Vista and Ubuntu Studio, with extra partitions for various sorts of video capture, though that's a bit low compared to the 3TB internal on my desktop, clearly one can go higher nowadays) and a flash card reader (unless you're married to Compact Flash). And Firewire. This is the unit I use very often for both HW and SW development.
Hydrogen holds some promise, but it's questionable right now. It's green to burn (or otherwise use) Hydrogen, just as it's green to use electricity. Both have the same original problem, though... you can't mine or otherwise locate sources of hydrogen anymore than you can do so with electricity. H2 is just a chemical answer for the battery.
Now, what you left out.. the big piece... is how that H2 is converted to electricity. Are you buring it, or feeding a fuel cell? The Fuel Cell is great idea... over 65% efficient, no "burning", thus, few if any pollutants (you would still have NO2 and other pollutants burning H2). We've been making these for a long time to power spacecraft... but they have the budget for it. Traditional fuel cells use lots of Platinum... same problems as large BEVs... no one wants to spend $150,000 on an economy car. Newer designs with engineered materials are promising, but there's more work to do. H2 storage is another issue... compressed gas is a hazard and also limited in capacity, while chemical storage (very similar to a NiMh battery) is higher density, but the cells wear out.
And you still want this to be a hybrid... a fuel cell likes to deliver a steady power output, it's not surgey at all.
Then there's the production of the H2... where does it come from? Like electricity, you can make it many ways... like, from electricity mixed with water to release H2 and O2. But that's not terribly efficient. You can make it from petroleum products, or from alchols, but there are also efficiency issues. In fact, very similar to those of the battery EV world.
And there's also the infrastructure problem. H2 refueling might be faster than electric recharging (it is now... it won't necessarily always be). Power distribution would ultimately have to be beefed up to support a BEV infrastructure, but it does exist today. H2 is non-existant... no one's building fueling stations unless they're in on the experiment.
While that's true, you have to figure "well to wheel" for both technologies. For ICE cars, you're lucky to get 15%.
For electric, the car itself rocks... the typical electric motors (they're all three phase AC motors, Toyota's use permanent magnets, other EVs like the Tesla use linear induction motors) are over 95% efficient. You drop a bit charging then discharging the battery, but that's at least 85% efficient. And a bit is lost to heat in the inverter. Still, very good.
Operating the vehicle, the EV gets more efficiencies back.. electric motors have peak torque at zero RPM.. they have a nearly ideal power curve for driving, so the practical benefit exceeds what you see from the motor and battery efficiencies, particularly when you factor in regenerative braking (which is also a wonderful thing for the back-up friction brakes.. my 2003 Prius has 115,000 miles on it, and the original brakes). On the downside, EVs are heavier than ICE cars, due to the batteries (some of which is offset by the weight of the motor... the Tesla Roadster delivers supercar performance off the line with a 125lbs motor about the size of a big watermelon).
You also lose an average of 7.2% in the process of distributing electricity.
I did the back-of-the-napkin math on this once, based on good electric car efficiencies and the typical use of cars. If just the passenger cars all went electric (not other forms of transportation), the US power output would have to at least double.
And it could be worse... moving peak demands into overnight, when normally peak-friendly power sources like solar and wind don't work so well.
It's virtually impossible BEV use would grow fast enough to cause problems, but if we all woke up on Christmas morning with BEVs in our garage, it would destroy power distribution as we know it.
Most likely it's the pollution rationale.. that's all that cities care about. They really don't give a hoot about gas use, just pollution release. If they get two people in a car, that's half of the pollution, even if it's two in a Hummer versus two Hummers (obviously, not a perfect system if the compact owner carpools in the Hummer-guy's Hummer all the time).
Prius entered the market at SULEV level, and now resides as PZEV. That basically means that, particularly within a city, it moves sometimes without making any pollution, and it's really clean otherwise. The cutoff for SULEV means the car produces 90% fewer emissions (in detail: .01 grams/mile hydrocarbon emission, 1.0 grams/mile of carbon monoxide emissions, .02 grams/mile of nitrous oxide and .01 grams/mile of particulate matter emissions) than a "regular" car. SULEV was originally California-only, now recognized in New York, New Jersey, Vermont, and Massachusetts as well.
I do agree that this should be entirely based on the pollution certs of the car, not the type of car. The Yaris only meets ULEV last I checked, but if a non-hybrid meets the same specs as the Prius, it ought to get the carpool rights... and if a hybrid doesn't, it ought to be banned. Stupidity about the implementation rather than the desired result is what caused that whole electric car fiasco in California. There's one Ford Focus build that meets this spec, non-hybrid.
Anyway, the magical juice is really low emissions. That was the original goal of the Prius, not strictly high mileage (the two are related, but you can optimize for either one... the original 67mpg Honda Insight with manual transmission only met ULEV standards). The SULEV is based on a level of pollution equal to or less than the level an electric car would produce, based on the average pollution (ish) due to power generation in California back to sometimes in the mid-1990s, when these standards were all worked out.
NiMh cells don't have an aging issue, at least not anything significant. There have been plenty of Prius out there at over 200,000 miles, still going strong (and it's not an option.. if the traction battery dies, you can't even start the Prius.. all the 12V battery does it boot the computers).
Li-ion, in its typical forms, unfortunately have had aging issues.. a Li-ion cell will usually lose some important bit of its capacity just sitting there. It'll also self-destruct if discharged below a threshold, and most of the Li-ion chemistrys don't handle as many charge/discharge cycles as NiMh. But the energy density of Li-ion blows away NiMh.
Both of these cells last dramatically longer when only partially cycled. The 2001-2003 Prius cycles from 40% to 80% of battery capacity. Toyota increased the range to 60% of capacity for the 2004 model (and cut out a bunch of the cells, so the effective power is the same). Cycling that way, the "number-of-charge/discharge cycles" figure goes away. This is apparently true of Li-ion cells. Naturally, more practical in a hybrid than a full BEV.
The holy grail is certainly improved battery technology. But for a BEV in the long run, while 500 miles sounds great, that's not the big problem. First big problem is battery life.. if I have to spend $10,000 or even $3,000 every 3-5 years on a new battery pack, that's going to be a significant problem for most potential buyers. There has already been lots of promising work on engineered materials for long-life, if not yet immortal, anode and cathode materials. Many of those also improve the charge rate and discharge peaks, both good things.
Charge rate isn't a big deal if you're charging in your garage. Certainly NiMh supports 4*C charging (though not necessarily without reduced life), which gives you a charge in 15 minutes... still too long. But you don't have remotely enough power coming to the typical home anyway, so that's a moot point, and you generally have all night for a charge once that car is at home.
But ultimately, for BEV to succeed as a realy alternative to ICE, it has to, well, be a real alternative to ICE. I need to be able to stop somehere and "fuel" up, and in some reasonable amount of time. If not, BEVs may be a very nice business, but they're always be a niche.
Prius efficiency has nothing whatsoever to do with moving the efficiency point to 20mph.
Yeah, you get more mileage in city-type situations... at least for the duration of an EPA test. That's the effect of running on electric-only for a fair percentage of the time, and particularly, off the line, where electric is so much more efficient than gasoline.
The big issue with the Prius that people have is the same issue they have with every other car -- they just don't know it with most cars, because they don't have mileage computers. The EPA numbers are based on slower travel. If I drive my 2003 Prius (less efficient than the newer models) at the posted speed limit, with 40psi in the tires and other basic good sense, using a little feedback from the computer, I get over 50mpg, unless it's the dead of winter. If I drive like a bandit (which, alas, is more typical).. 75-80mph on the highway, then I'll see those 42-46mpg results. Duh! Even with the Cds of the Prius, you still have a significant component of turbulent airflow... that component is based on the square of your speed.
What you're missing is the whole basic understanding of how the Prius works. There's a small win on efficiency due to using the Atkinson cycle on the ICE, rather than the Otto cycle.. more efficiency, less peak power -- which you don't need, because the MG2 motor provides that for peak power demands. So the ICE doesn't need the same power peaks. Then there's the transmission... which doesn't exist. A set of fixed gears, two motors, and a computer that can dial in the peak possible RPMs out of the engine for any given load requirement. And power normally wasted by the ICE, and by braking, is put in the battery.
Also, keep in mind that the original (G0/G1) Prius was designed primarily for low emissions, with fuel economy secondary. Compare that to the Honda Insight, the original model designed for maximal mileage (1L 3-cylinder engine, electric boost, well over 60MPG in regular use, beyond that if you drive hypermiler style).
Why the Prius mileage isn't better is simple -- ICE engines just suck. A typical car gets about 15% efficiency on a good day. A Prius is likely more in the 25% range... still sucks, compared to a 95+% efficient electric motor coupled to an 85+% efficient battery.
"Good writers borrow from other writers. Great writers steal from them outright." -Aaron Sorkin
"Immature poets imitate; mature poets steal." -T. S. Eliot
"Plagiarize
Let no one else's work evade your eyes
Remember why the good Lord made your eyes
So don't shade your eyes
But plagiarize, plagiarize, plagiarize
Only be sure always to call it please "research" -Tom Lehrer
That's technology for you.
Before guns, you basically needed one guy with a sword to fight another guy with a sword. Give one of them a gun, and now he can kill dozens of guys with swords, without even necessarily taking on much personal risk himself. So the killing is more likely, just because it's possible where once it wasn't.
Before computers, you basically needed one guys with a phone tap to spy on one guy on the phone. But add in speech recognition software, and you might have dozens of taps being monitored at once, with a human only standing around to listen with the tapping program thinks it has something of interest. So the tapping in more likely, just because it's possible where once it wasn't.
Same story with the advent of face recognition software, surveillance drones, robots, all kinds of things out there that allow one person to do the work that formerly required many. This isn't anything all that new... the only real advantage is that GPS surveillance (and keep in mind, there has to be a second radio too to do the reporting back to the cops... a GPS itself is just a receiver) is ridiculously easy to defeat with jammers. And a planted GPS transponder is going to be relatively easy to find, if you're looking for it.
This is pretty scary sounding in the Big Brother way, but extremely easy to defeat.
A GPS only works with a clear view of the sky... that means, your GPS bug needs an antenna somewhere, you can't simply hide the whole thing under a car.
You also need a backchannel -- cellular modem or whatever you have access too... police and Feds probably have their own private frequencies to use, but then they'd have the problem of range... you're not going very far on high frequencies at all, or low frequencies without a pretty obvious antenna. So you can probably bet most would use cellular modems, unless they also have you under active surveillance.
Both GPS and cellular are crazy-easy to jam. GPS in particular.. GPS chips are built with a -140dB or so input sensitivity. There are no naturally occurring noise sources at GPS frequencies, but you could build a tiny, very low power noise source and keep any GPS signals from working near your car. In fact, you can probably just buy one of these easily enough:
http://boingboing.net/2009/07/30/gps-jammer-plugs-int.html
http://www.gpsjammers.net/gmc07.html
Yeah, guess you can. These are all blocking the L1 band for consumer GPS, not the military band, but chances are, that's good enough to block the police, and the military frequencies aren't any more difficult to block, if perhaps less off-the-shelf. Same goes with cellular jammers... you can buy these pretty easily.
Now, obviously, the Good Guys hope that the Bad Guys are not as smart about these things... and they won't be, initially. But soon enough.
All RAID is software RAID. The only difference between "hardware" RAID and software RAID is where the software runs.. does it get its own CPU or not? And there are plenty of applications for "hardware" RAID... external "box of drives" when your PC is full, large NAS (I have one of these), SANs (not yet), etc. From the prespective of the RAID hardware, you may well be running Linux with ZFS or BTRFS or something entirely different, and that's software... but from the external clients prespective, it's hardware RAID. And this is the way its always been done... there were "hardware RAID" cards for the PC's ISA bus... I recall one had an embedded SPARC processor running the RAID software, back in the mid-80s or so.
Yes, it's true... I do run a RAID on my personal PC (well, in a little box next to it), and it's also true, I can't afford a 6TB SSD. Sad, but true.
However, I also know a secret... you can re-write a sector on an HDD many, many times more than the same physical sector on an SSD. The SSD gets you around the mechanical issue of seek times, but it doesn't get around the idea that things fail. And, given that SSDs usually do wear levelling to ensure no one sector wears out prematurely (depending on the specifics of the NAND flash used in the SSD, whether it's MLC or SLC, etc... some MLCs are only good for a few tens of thousands of writes), by the time you start to see a failure or two, chances are, the whole thing's ready to go. They're also typically slower at dealing with writes, particularly many small writes, since you have very large block sizes at the flash-memory level (up to about a megabyte) on large SSDs.
Obviously, there are plenty of usage patterns that would yield a mechanical failure from an HDD long before there's a flash cell failure from an SSD. In both cases, the realistic life expectancy may be fine... I don't expect an HDD to last forever... I'm happy when the last to the point at which a replacement's cost is essentially a "no brainer" (eg, when 2TB drives start showing up below $100 or so, my 6TB RAID is likely to become an 8TB RAID).
The FOSS world has payment, too, it's just a different form of currency... sometimes. You have company-supported open source projects, paid in cash just like any other software engineering work. And you have the volunteers, who work on projects for fun, renown in the FOSS community, "building one's own toys"... eg, it's a hobby, or a technical challenge they may not find in the day job. But they all have day jobs. A sterling reputation in the FOSS community doesn't put food on the table... even Linus has a day job (he was at Transmeta, at least in their better years).
You don't understand the beauty of the Zune marketplace!
Of course practically no one bought the original Zune at the original price... but when I bought my brown Zune for $75, that was a damn good buy for a 30GB player at the time. And say what you will about Microsoft in general (and I usually do), but they've keep the upgrades getting better, and kept them free. That doesn't suck.
As for the dock, when I got my Altec-Lansing dock for the orginal Zune for $30, I got to laugh at all the iPodden out there buying the very same dock with iPod connector for $120 or so.
Hopefully the same things happen for the Zune HD... MS already released it at a somewhat decent price. So no one buys it, the $220 model drops to $120 or so, the $90 dock (which is MSRP, anyway) drops to $30, and then I upgrade.
Anyway, if you really want a PDA+MP3 player, you're probably in line for an iPod anyway... as long as you're happy running applications only blessed by The Steve. Otherwise, go with an Android device. But if you just want a decent PMP, the choice right now has to be between the Zune HD (assuming they keep the quite-a-bit-better-than-iPod sound quality of the originals) and the Cowon S9. Both have OLED screens, which pwn the iPod ten ways to Sunday. When you can have OLED, you don't want LCD... period.
As for free applications... does anyone here really think there's a free lunch, particularly from companies like Apple or Microsoft. When they promised free games, my initial assumption was that a few freebies would come out, either trivial ones or those that function as ads for paid games... just like you have on the iTunes store (eg, it's either relative crap, or you pay for it). There is absolutely no way they would have a continued supply of new games and applications being developed with no chance at revenues of some kind... same reason you get ads on Google and YouTube and pretty much anything else related to Web 2.0 hysteria. Oh yeah, and slashdot!
That they chose an advertising model, rather than a cash-up-front-model, is fine with me. I'm adult enough to not expect a free lunch from Microsoft.
So, you're suggesting Apple get into the business, and produce a few dozen lenses to go with their DSLR? If not, there's zero point in building a DSLR... not that Apple would be taken seriously.
That's also not a market that Apple would be or should be interested. Keep in mind that over 80% of the DSLR market is owned by Canon (41%) and Nikon (40%).. they're the guys who make those dozens of lenses and accessories that make the DSLR worth the price in the first place. The rest is being sliced up between Pentax, Olympus (6%), Fujifilm, Sony (6%.. they're one of the big four in high-end video cameras... the bought up what was Minolta-Konica's SLR works), just recently Panasonic (they're one of the other major powers in high-end video cameras, along with Canon and JVC), and a few others in specialty markets (Leica, Ricoh, Hasselblad). Serious users aren't likely to buy a DSLR from an electronics company without years of proof in the business.. that's why some of video camera companies compete, but few of the other film or CE companies who make digital P&S models (Casio, Epson, HP, Kodak, etc).
The global market for DSLRs in 2010 is expected to be a bit under 12 million units (that's an estimated 9% growth in DSLR shipments)... so that's 4.92 million Canons, 4.8 million Nikons, and 2.28 million left for everyone else to fight over. Do you really think that's an iPod-like market? Apple sells nearly twice that number of iPods... every MONTH.
No, Apple's doing the right thing here... "embrace and extend". They made a decent MP3 player, but really made the sale by delivering the iTunes store. Lots of people make better MP3 players, so Apple pushed in photos, then video. Lots of people came out with better PMPs, so Apple added PDA function, and a phone. Now they toss in a "free" webcam (some prespective here.. this is a crappy, crappy camera for any serious use, but it's fine for the kids buying crappy Flip cameras and delivering video only for Facebook and Youtube) and push the iPod/iPhone as a gaming platform, taking on the Sony PSP and the Nintendo DS. That's the smart thing... that's a $1.00 camera added to the basic stuff the iPod already has. Not a couple of billion to develop a line of lenses no one will ever buy.
Certainly, such "free" cameras will get better over the year, but the iPod is competing directly with the "Flip" and other web-cam + flash units, and to an extent with cell phone cameras. They aren't even close to being as good as a modern point-and-shoot model from Canon, Panasonic, Sony, Nikon, or many others. And they never will be.
... tried the first one a few years back, it just didn't "pop". Can't answer for the whole series, of course.
On the other hand, Brandon Sanderson is a fine writer/storyteller IMHO... I loved the "Mistborn" trilogy. So if you are a WOT fan, the series might well go out with a good story. Finally.
4) You're calling on an iPhone.
That's because you're not using an iPhone. My sister calls me all the time on hers... it drops calls like crazy.
Said it once, and I'll say it again. CDMA is a dead end, the world is moving to LTE. Why would anyone waste their resources on a technology with such a limited lifespan.
3G is now (HSPA or EvDO).. and of course, they're rolling it out still. LTE is 4G, it's a sea change, and such rollouts take time. Verizon is starting next year, on 700MHz, but they won't be finished for several years at best. AT&T is planning trials next year, with the real rollout starting in 2011.
Now, get some perspective... the iPhone is a product with a 2-year expected life. Apple introduces a new model once a year.. the product doesn't even last two years in the channel, just ni your pocket. And Apple's never been worried about keeping up with technology on it, anyway. So, even if they launch a CMDA phone in 2010, most places won't have LTE coverage even when those phones are retired. You'll be trading in your second generation CDMA iPhone for an LTE iPhone, if you're lucky.
Why do it? Money. CDMA isn't new technology... it's a relatively minor revision to an existing technology. You don't think a company like Apple can get QualComm engineers to provide a guaranteed works-first-rev CDMA solution, and get Verizon to pay for it, assuming the agree on a deal? If you doubt this for a femtosecond, you haven't worked in hardware engineering as long as I have (26 years). Apple isn't going to make a CDMA phone just hoping someone will buy into it, as they did with the first iPhone.. but they've established a market and a brand. If Verizon want the iPhone, and can deal with whatever hoops Apple wants them to jump through, they'll get the iPhone next year.
There's not even a tiny question about this.. the development costs are insignificant compared to the money involved. Apple themselves spends several times more money just on ads for any of their products then they spend in product development. As do most hardware companies.
Keep in mind, cellphone vendors aren't designing the cellphone chips themselves, in most cases (and certainly Apple's case), any more than Dell or Acer are designing their CPUs or Wifi chips -- they get a ready-made, working solution from the chip vendor. At worst, they're doing some of the baseband processing on their own host CPU, and they'll port code from the chip supplier to that host CPU. The iPhone, as you might expect, has a separate baseband processor... baseband doesn't run on the application processor.
The world will move to LTE. That hasn't really started yet. There are plenty of locales that don't even have decent 3G coverage yet (either kind). Bringing the iPhone to a CDMA network is a next-year thing, and it'll happen if there's a business case for it. This is not a 5-year investment, it's a small change to an existing product.
Verizon is moving to LTE for 4G, just like everyone else... well, except Sprint, who's already on WiMax (they're virtually the same downstream, anyway). Even QualComm, the originator of CDMA2000 and other atrocities, gave up their 4G plans, and will be integrating LTE into their chips for CDMA networks, just as all the GSM chips will. It's not much of a problem for anyone.
There are no iPhones on LTE, and in fact, they're kind of dodgy on 3G today. When Apple's no longer bound by their exclusive deal with AT&T (it runs out next year, unless there's some kind of renewal agreement... which, given the iPhone's success, would cost AT&T dearly), don't even be slightly surprised if there's a CDMA/EvDO iPhone ready to go. It's not that big of an engineering deal -- one chip in the system and some new baseband software. And Apple can, at this point, get Verizon to pay for that development if they're interesting in carrying the iPhone.
Apple, on the other hand, has much to gain from such a deal.. Verizon is slightly larger than AT&T, so they could, in theory, double their user base going to Verizon. It's easy to claim they'll offer the iPhone through T-Mobile, and they might well, but that wouldn't help them much, other than to maybe force from contract differentiation into the market. And it wouldn't even change much there, if Apple gets the same kinds of kick-back deals from other telcos that it got from AT&T (which is the factor that pushed the exclusivity deal into 2010).