Actually, some cameras do use CMY, or more likely, CMYG. My old Canon Pro90IS had such a sensor. Maybe they're trying to minimize the color error rather than maximize sensitivity? Hard to say.
More interestingly, the very first HDV camcorder, the JVC HD-GR1, used both of these techniques back in 2003... see http://www.jvc.com/promotions/grhd1/unprecedent/s_ right.html. Their sensor is White (clear), Yellow, Cyan, and Green in a Bayer-like pattern. They made similar claims: the effect is 50% luma, rather than the 33% you'd presumably get with all filtering. This didn't really help -- that camera had really bad low-light performance, less than HD actual resolution, but they did make an interesting argument.
They also had the advantage that, knowing your output is going to be 4:2:0 encoded, it's an easy argument to trade luminance for color accuracy that's getting tossed out anyway. Kodak's approach is likely to result in even worse edge effects (eg, interpolations that "guess wrong"), all things being equal. Of course, as you increase pixel density, this is less of an issue anyway for a unit-sized still photo. And, I suspect, their main goal here is to deliver higher resolution sensors without increasing the die size or making them crappy at low-light performance. This is more important in the highly competitive consumer camera market, but not a non-issue in pro-level stuff, as they continue to squeeze more pixels within something between APS and 35mm sized sensors, and simultaneously have a demand for better low-light performance.
Planned obsolescence, sure... the manufacturers like the idea that, at some point, you'll run out an buy a new camera, and new cards for that camera. But the notion that "it didn't make sense to design a standard for 4G+ cards in the days of 8M and 16M cards" is just silly. Particularly when you consider the history of things like SmartMedia, a card format that ran out of steam at 128MB, and alienated those buyers not prescient enough to know this was a stupid format.
And particularly, consider MMC and SD media... these are LOGICALLY addressed, for christsakes. One or four bits of data, to a logical model that have been anything they wanted. It would have been trivial to include address space for these from the get-go, and it wouldn't have cost anything more. This isn't reasonable shortsightedness, it's planned.
And take a look at the oldest memory card format on the market: Compact Flash. Sure, it has had a tiny bit of evolution (direct address vs. packet addressing, at least an issue on PCMCIA, from which CF derives), and there's an upper limit in there somewhere (132GB?), but it was a perfectly reasonable one for the day.
They're kind of fooling you on the 2003 to 2004 issues here. The 2003 Prius (I own one) is composed of 38 separate 7.2V Panasonic "Prismatic" NiMH modules, the system runs directly from that battery (more like 300V with a reasonable charge), and runs it over 40% of its useful range, which is what keeps it alive for a very long time.
Beginning with the 2004 model, they didn't substantially change that 7.2V module, but the battery itself is now composed of 28 separate 7.2V NiMH modules, for a nominal voltage of 201V. However, they've extended the range of charge/discharge, and the inverter is also now a step-up regulator, delivering a regulated 500V to the motors.
Thus, yes, the new full battery is smaller, cheaper, lighter, and has a higher specific power, but that's due to the new system, not any dramatic change in the battery technology. The smaller, cheaper, and lighter come from using fewer cells, the higher specific power comes from using more of the available battery capacity.
And the 2004 is actually the third generation Prius, the model I have (2001-2003) is the second. The first generation unit was on sales in 1997 in Japan only, and used what were essentially individual D cells. There were other differences... it ran out of battery power much easier (you see a Tortoise icon when this happens, but that's almost unheard of in the recent models), and the battery pack was larger and heavier, but less efficient. There were more significant changes in the battery formulation before the 2001 model was introduced.
And they've gone on record claiming that the 2009 model will run from Li-ion cells rather than NiMH. That's been kind of a holy grail, due to both the higher energy density and lack of entangling patents on Li-ion vs. NiMH. On the other hand, NiMH have traditionally had much higher current peak capacities, the well established long life as used in Hybrids (no so much in a full BEV that runs the full cycle), and they don't easily explode. Toyota may be planning to use some of the emerging nanotech anode technology for these cells, which promises faster charging, higher currents, and dramatically longer battery life... some day.
As many have pointed out, the patent was applied for in 1987, but only granted in 1991... thus, its in effect for 17 years from the date it was granted... that's the system prior to 1995.
However, it should be painfully obvious this is the illegal sort of submarine patent. Usually, companies used to attempt the legal sort -- they'd apply, they'd try to drag the process along as slowly as possible, counting on the 17 years they'd get once the patent was granted, the extra time before the patent is granted would hopefully get that patent into common use... particularly when the patent, such as this one, was so frickin' obvious that people re-created it continuously. The 1995 changes were designed to stop this as much as possible.
But tabbed interfaces have been around for quite some time. There were tabbed interfaces in AmigaOS programs... I can't swear they showed up before 1987 (AmigaOS 1.x), but this was the latest thing in the early 1990s. I know I've seen computer Rolodex-type programs using a tabbed interface (since that's basically the metaphore), very likely before 1987, though I can't think of the specific example. But obviously, this has been used in Windows and many other programs since then. They ought to lose enforcement rights on this patent due to the fact it hasn't been enforced for the last 15+ years.
I spoke with one of the "fixer" guys Verizon sends out when their first repair guy can't solve your problem; he's done all kinds of DSL installations. The DSL board set can live in any local node, it doesn't have to be at the central office... he knew the precise board set and software revision for the local node that's less than 1000ft from my driveway, and promised that Verizon would NEVER support DSL there. Period. Thus, the 0.97m 2-way satellite dish on my roof, and a hefty internet bill.
See, the telcos view DSL as a necessary evil today, but want to invest in it as little as possible. They're spending the crazy money on FIOS, which is the only way they stay competitive with the cablecos in delivering voice, internet, and TV. Naturally, of course, they're rolling out FIOS in areas already well served by DSL and Cable... I have a few buddies in the West Chester/Exton PA area who are getting FIOS-TV, 15Mb/s downlinks, etc.... must be nice. But if you're in a non-coverage area, don't hold you breath for DSL expansion.
Oh, phleeze! Is it actually a requirement to be an idiot to work for the FCC.
I'm one of those guys who can't get wired broadband, and I'll betcha they have me listed as "served" (it's ok, I have satellite, for about $100/month, that's delivers reasonable performance, unless latency is your big thing). I live in rural South Jersey, in a town covering about 45 square miles that's apparently too small to have its own zipcode. So we get to use a bunch of zipcodes from other towns... that's how zipcodes work, after all... they just link your house to the post office serving that house.
So you can get DSL in some areas of this zipcode, and cable in others... but the zipcode is for Monroeville, NJ. Monroeville is about 15 houses and a post office... most of the people in the Monroeville zipcode don't actually live in Monroeville. In fact, that zipcode covers part of two different counties. MOST of the people in the Monroeville zipcode probably can't get either form of wired broadband offered in this area, but we're no-doubt counted, because the zipcode covers a huge area.
Consider also what Commodore brought the Amiga. First of all, the Amiga, Inc. people really, really didn't want to be acquired by Atari, and would have made out very poorly, financially. Commodore, as it turns out, paid them very nicely. And while I agree on the marketing side of things (C= didn't market the Amiga any better than anything else C= marketed), the technical issues were day and night, Commodore vs. Atari. Commodore did actually have a fab that could make the chips for the Amiga. They brought additional engineers into the projects, expanded the projects, set up a truly functional Developer program, evolved the OS and the hardware (not as much as any of us wanted, but way more than anything Atari ever did).
And in fact, the Amiga did ok under Commodore for awhile. It wasn't a rapidly growing business, but it was steady until the early 1990s, which a whole new level of mismanagers were hired. They managed to royally screw up what could have been some really interesting years for the Amiga.
In the end, the notion of doing in-house graphics hardware had to die, just as doing in-house CPUs had once worked for Commodore, but didn't by the time the Amiga came along. We in Engineering knew this; in fact, there was a project in 1993 indended to deliver a PCI-based successor to the Amiga graphics chips... it even had a real GPU (actually based on the PA-RISC processor ISA) and 3D engine. This was at least potentially something that would have sold in the PC market as well as being sold in Amiga computers.
Toshiba announced research on a technology for fast charging li-ions over two years ago. This was using nanotech materials for an improved anode (maybe cathode too), enabling fast charging (80% charge in one minute) and long life (99% capacity after 1,000 charges). A similar approach was also annouced, about the same time, by Altair Technologys in Reno. It's all about increasing the effective surface area of the anode, and perhaps making it from stronger stuff.
In traditional Li-ion cells, a big wear factor is that the anode can form a parasitic battery with the electrical contact, causing the terminal to eventually wear out, faster as you approach full cycling the battery. Heat is also a factor, in both terminals and the full cell... the higher internal resistance of the Li-ion vs. NiMH (or better still, NiCAD) limits peak power, and also increases the risk of damage or, particularly in quesitonably made cells, explosions.
Dramatic improvements in both of these are necessary to enable practical (in a commerical sense) pure electric vehicles (BEV). There's no conspiracy necessary... traditional NiMH cells are a problem for full electrics.. which the actual reason none of these cars have been successful. Not to mention the expense... the Toyota EV-RAV4, for example, cost $42,000 and gave you about 100 miles on a charge.. and that with Toyota still selling them at a loss (as they did in the early days of the Prius, too).
In a hybrid, the batteries are only partially cycled (my 2003 Prius runs the NiMH cells over 40% of their capacity range; Toyota extended this to about 60% on the models starting in 2004), and that keeps them very long lived. Natrually, better batteries make a better hybrid, but the fact my Toyota can only go about 2-3 miles on a full charge doesn't impact its general use; the issues around battery technology today make the BEV a small niche product.
But the energy density is just too low even full cycling NiMH to make a BEV with mass appeal... most people would demand at least 200-300 miles of range, charging times on-the-road similar to that of petrol fueling (not the minimum of 15-30 minutes you'll have with today's cells), and long life (full cycling NiMH, they're good for about 500-1000 charges).
Once you have a higher density cell that doesn't wear out and can be charged in under 5 minutes, full EVs will be practical enough for a mainstream automaker to POSSIBLY launch a full production car, not just an experiment. This is critical technology for improving hybrids as well, and keep in mind that all practical FCEVs will also be hybrids (fuel cells suck at peak power demands, they like to be slow and steady, so you need a battery or supercapacitor to enable the peaks).
Well, I did read the article, and I can't tell what they're claiming is "new"... my first design with "upgradeable hardware" was in the Metabox 500 settop box (1998); this device included a modem for a datacasting technology called BOT (Broadcast Online Television, developed in conjunction with the University of Dresden), and the FPGA/LCA design was uploaded to the device as part of the device driver initialization for the modem. Anytime we released a new device driver, we could change the hardware.
And we could make some radical changes, since nearly the whole design was implemented in a low-cost FPGA. At $5.00, maybe a full custom chip would have been $1.00 (at least without amortizing NREs), but it didn't really matter, it was cheap enough. Anyway, even with a full FPGA design, there are always issues of routability... you can't make any arbitrary change and guarantee you meet timing requirements, or even if it'll fit. So while useful, this isn't as magic as it might sound to laypeople.
There would be serious issues with embedding this into CPUs or other complex devices. Other than the aforementioned speed issues (LCA/FPGA is always slower than gate array, which is slower than full custom, based on the same technology), it's only so good as where the FPGA is embedded. You can't fix any problem unless the whole design is implemented in an FPGA. There's also the startup issue... if you embed the FPGA design in Flash, you have the additional complexity of putting flash down in a complex CPU core (there are plenty of CPUs with on-chip flash, but they're usually things like ARM7/9, or simpler still), and the reality that Flash only lasts so long -- do the chips start to fail completely in 10-something years? If it's loaded at boot time, then obviously you can't FPGA any of the hardware needed at boot time.
I'm in the no-man's land of South Jersey, where broadband is dicey at best. Before I decided to upgrade, I had a POTS line pinned (with another bonded channel optional), truely unlimited use (well, other than the limits imposed by ~50kb-100kb/s) for $15 a month.
So here, there's no DSL, no cable... dicey. Last year, I hooked up to Verizon EvDO. In theory, I was paying $80 a month (ouch) for unlimited service. Well, except for the "unlimited" and "service" parts. At best, my EvDO Rev 0 connection would hit 500-800kb/s down, 60-80kb/s up. That happend twice, for about a week each. More usually, I'd get 100-150kb/s down, 10-30kb/s up (yeah, slower than dialup)... and this even with the addition of a 3W amp/LNA and a roof antenna. They were playing games here.
Even with unlimited EvDO, you get into weird and broken definitions of "unlimited". Essentially, Verizon knows it's a limited resource, but wants to compete with other hookups, so they call it unlimited. And it is -- no extra charges for usage... only, once you pass a magic double-secret download level, they toss you off.
Now I'm on HughesNet satellite.. for $100 a month I get 1.5Mb/s down and 250-500kb/s up. And while all networks are subject to aggregate load, at least in my area it's pretty dependable. This is definitely no unlimited, but they have very specific rules about "fairness", and if you exceed them, you're throttled back until you drop below a threshold (based on your service agreement recovery time). But it's totally OK to hit that limit, it's just enforcing fairness (and I gather, it's only implemented when there's actual contention for bandwidth), not a warning or anything.
While I'd much prefer FIOS (friends in West Chester, PA, now have Comcast cable and Verizon FIOS competing... competition does exist), I'm also happy to know the actual service agreement, not some double-secret (and thus, floating on arbitary whims) level you're not going to be told about up-front.
I loved the story when I read it back in High School. It actually does properly lend itself as an ET-style tale, since it's about young kids being taught very different ways of thinking. It's easy to believe Hollywood would mess this up like they have so many before it, but don't immediately discount it based on the fact that kids are prominent in the story.
There were a number of these systems over the years, ranging from glorified teletext systems (teletext itself is digitally encoded on one of the vertical blanking interval lines, so are close captioning texts) to something more complex. While working at Metabox AG in Germany, I helped deliver a system called BOT (Broadcast Online Television, originated at the University of Dresden) which allowed datacasting from 80kb/s (using just some spare space in horizontal intervals) up to a theoretical 4Mb/s (taking over a whole channel). While obviously just one-way, we used this to broadcast web content to our set-top boxes. The STBs ran web servers which could avoid expensive connect charges (usually POTS or ISDN in those days, and in Germany, there was no unlimited local calling, all dialup was pay by the minute) for content found locally. This was started in 1998 or so.
Intel had a somewhat similar technology for analog TV datacasting, though it didn't exactly set the world on fire, either. The advent of digital television and real broadband pretty much rendered these technologies pointless.
And these are not what's being discussed here. Reliable last mile broadband wireless, whether based on Wifi, WiMax, or something new, would be a great thing for folks like me, who are too far out in the boonies to get wired broadband. My current choices are cellular broadband (I've had Verizon EvDO for the last year, and it's flakey here, running from near DSL speeds to slower-than-POTS to nothing, even with the 10dBi roof antenna and 3W PA/LNA) or satellite (HugesNet guys are coming tomorrow). Both are dramatically more expensive and slower than wired broadband, and have other issues (satellite's latency, cellular's flakiness and double-secret usage limits).
If you're upconverting VHS to DVD, you can pretty easily get 6 hours of VHS-quality video on today's DVD. The 720x480 format (NTSC and NTSCfilm resolutions) is fine for commercial films (and on a dual layer, you can store well beyond two hours worth, even at that resolution), but for VHS, it's overkill. DVD also supports 352x480 MPEG-2 and 352x240 MPEG-1 resolutions, either of which will do justice to plain old VHS tape (you'll want full resolution for well recorded SVHS).
VHS vs. DVD was a different story -- analog linear tape vs. digital random access disc. VHS sucked (in the sense of the engineering term "it sucks"), DVD isn't bad, even on a good HDTV.
The post-DVD war, unfortunately, is looking more like DVD-Audio vs. SACD than VHS vs. Beta. Yeah, they'll be better. But worth the extra cashish to most people? That's debatable, for exactly the same reasons as the extended rez audio discs. While predictions claim as much as 40% HDTV penetration by 2007, that's still missing the point: that's at least one set per house. You might watch HD-DVD/Blu-Ray on your HD set, but it doesn't play in the kids' room, in the car, on the portable, on the older PCs, etc.
Meanwhile, you can buy a DVD player that'll play DiVX-HD, Microsoft WMV9/VC-1, and HDV-compatible MPEG-2.... today. The hardware guys already grok the fact that the 5" round shiny thing is just a disc, and the player can host any number of applications. And must, to remain competitive in a hard market.
Whoops, hit SUBMIT rather than PREVIEW. The actual number would be, er, $30 per charge, best case, plus the power costs. But that's a subsudized battery pack, maximum expected life, and NiMH. You're probably looking at $60-$120 additional cost per charge, for the batteries, using state of the art Lithium and commercially profitable predictions for cell life. Then there's the fact that the RAV4-EV only got 80-120 miles on a charge; you're going to need to about double that for consumer acceptance. That could double it, or more, so you're probably getting close to a worst-case charge of $300 per "fillup". Obviously that won't fly.
The problem isn't so much EVs in general, or even battery life. The problem is the time it takes to charge (the best technology runs about 4C, 15 minutes for a full charge, IF you can supply that much current, which you might at a service station, not at most homes) and the cost of the batteries. There are some promising ideas using nanomaterials that might achieve a very fast charge, but the cost issue is still a big problem.
There are two big problems with battery exchange. First is capacity -- as they age, the batteries hold less charge. People aren't going to like old cells, and they certainly aren't going to want to pay per replacement pack.
Second is cost -- you have to amortize the cost of the battery pack across the life of the battery, and I'm guessing service stations will pass that on to the customer. Consider the cost of a NiMH pack, such as that in the Toyota RAV4-EV, was listed at about $30,000, and that with Toyota selling the $42,000 car at a loss. Even assuming a Li-ion pack cost no more, you're only getting about 1,000 charges out of a Li-ion or Li-poly pack these days, at best. So figure you have to tack on something in excess of $300 for each fillup, just to cover the batteries. Consumers would not accept that, even if the power in the cells came for free.
Actually, no, that's not true, the "55 by itself" thing.
The way the Prius works, you actually have two motor/generators. MG1, the small motor, which starts the engine and, at low speeds, acts as a generator, and MG2, the large motor, which is on the final drivetrain and gives you that big torque when you start up.
MG1 serves another function: it balances the power between the ICE and MG2/wheels, providing the effect of a CVT, even though the Prius has fixed gearing. So when you're going slow, MG1 acts as a generator, dragging on the "power-split device" and providing the effect of a low gear (it's running against the movement of the ICE).
But above about 60mph, MG1 has to, instead, push in the direction of the ICE motion, acting as a motor rather than a generator, to provide the effect of a high gear. During this, MG2 is engaged as a generator to provide MG1 the power needed to do this. So there's no charging, period. But also, no dependency on the batteries: you can and do get to the top speed of 99mph (for the 2001-2003 car) regardless of the battery charge.
Hey, I'd go for the Jetsons thing, but hack in a manual control, so I could still "go stealth" if need be in my Prius.
Thing is, the Prius being as small as it is, at least WE can both see and hear you, on your bike. You're taller than we are, and louder. And since it's a green-tech vehicle, you can also bet that a significant percentage of Prius drivers, such as myself, are also cyclists. So we're watching your back, even when you don't know we're there.
Compare that to the Hummer driver. Between the blare of bad country music, the loud roar of the disel or that curiously-even-less-efficient V8 in the H2, and the perpetual state of shock of the average Hummer driver, given the $100-or-so cost of a fillup, they don't know you're there. And even the satisfying crunch the typical SUV driver gets destroying cyclists is lost on the Hummer driver, as the combination of active suspension and >6000lbs curb weight conspire to deprive said pilots of the usual SUV pleasures.
Panasonic/Matsushita, like every other NiMH battery licensee, pays all their royalties to Ovonics. If they had not, Ovonics could have actually blocked the import of the Prius to the USA. Unlike most of the licensees, they've done their own share of improvements to the technology, and have their own patent portfolio.
The thing with Panasonic and Ovonics is not the royalties per se, but the patent license itself. Ovonics has had no interest in making D-Cells, they were, from the start, after the Hybrid car market... only it didn't quite exist. At the time Toyota started work on the Prius, and even until recently, Ovonics was only R&D, no production facilities at all. Their NiMH license was for small cells only: AAs, Cs, Ds, cellphone and laptop batteries, but not the giant cells you'd need for Hybrid cars.
So Panasonic and Toyota got together on this, and built a hybrid anyway. The original (1997-2000) Prius actually used plain old everyday high-output D-Cells. For the 2001-2003 model, Panasonic supplied prismatic cells in packs of six, to Toyota. They, in turn, not being subject to any legal agreement between Panasonic and Ovonics, built up 38 of these (28 in the 2004 model) and voila -- hybrid power pack.
So basically, it sounds like Panasonic had smarter lawyers than Ovonics. The main focus of the lawsuit by Ovonics was to prevent Panasonic from doing similar things in deals with Ford or GM, in any hybrid projects they had in the works. As well, it's likely they were after some of the Panasonic technology -- Panasonic has the highest output cells, by far, of any NiMH vendor.
And it's also a limited time thing... Li-Ions or Li-Polys are likely to take over in the hybrid market, once they figure out how to keep them alive for 150,000 miles. And, I suppose, ensure they won't spontaneouslyh explode.
Honda, an investor in Ovonics, apparently doesn't have any issue. But they make very weak hybrids anyway. The Toyota system is much cooler... and I'm not just saying that as a 2003 Prius owner.
Hmmm.... that might explain a few things. I went to the Salem County fair (NJ) last night, to watch the pig races and pig scramble (really, me and my kids). I might be wrong, but I'm fairly sure I saw two prize sows with stupid grins on their faces, wearing SCO logo T-shirts.
Some years back, when I was VP of Technology at Metabox AG, I wanted to sell off a fairly large number of share to apply to the US branch of the company. The German company, on the other hand, didn't want it to seem that some high ranking corporate dude was getting ready to bail. So I cashed out some (ok, in hindsite, it should have been all of it) in dozens of smaller transactions. Sure, you'll find these, as in that Yahoo link, if you go looking, but it apparently prevents any big alarms from sounding.
Looks to me the SCO guys are "gettin' while the gettin's good", using the same techniques to hide it as best as they can. The obvious flaw in their logic is that 10's of thousands of Linux fans are watching their every move. You'll even see that some guys, like Broughton, are making multiple same-day transactions under different corporate titles.
What HAVE I been doing lately? Well, in The Work Thing, I'm the principle HW guy at Sizig, Inc; a small startup company, based in Philadephia. Our primary project is a consumer robot, but probably not what you're thinking. If successful, maybe a sort of "C64 of home robots", in most of the ways you could apply that idea. This is the best thing I've been involved with since Commodore. It's a good feeling!
We're well aware of the bad investment environment these days -- you're crazy to do a start-up now, right? But we have, it's been going over a year-and-a-half (I was involved since April 2002, and fulltime since October), so we've kept the development smart, as opposed to BIG, and will not rush this project. I'm still working most days out of my home office/lab (about 1000 sq. ft. in my cellar, more space than I personally had a Commodore).
In the shorter term, we have a spin-off of some elements of this technology into another market, which I won't discuss right now. This is one of those curious markets that's fairly large, not understood to be that large by most people, and totally back in the 40's or so, technology-wise. We think that dragging them into the 21rst century will be well met.
I'm also doing more video than ever... which I really didn't do back in the Amiga days, there just wasn't time for hobbies or any sort of side business. I'm shooting weddings and other events, making DVDs, etc. Not really THAT strange; before I was a computer nerd I was a photography nerd. The DVD is the ideal place to unify all my skills: computers, video, audio, photography, artwork, music, etc. I did a "Deathbed Vigil" re-release last year, as a justification to buy the software needed, and I have a PAL version on the way REAL SOON NOW.
Thanks for the interest, and a chance to write about some of the good stuff [other bad stuff intentionally left out]...
Jay was certainly THE father, but the whole team in Los Gatos was responsible for the Amiga. I was certainly part of the second-wave effort, in West Chester, and sure, as a group WE took on the role of the godparents -- bringing her up right in the absence on the original parents. I don't like to be singled out that way, this was a whole team effort. Some journalist made up that "godfather" thing. I really found it embarrassing.
Of course I worked my ass off on the Amiga, but most of us did. That's way a team of around 50 people, total, could do what most companies didn't when they had hundreds or thousands involved. My name got out, I suppose, because I wrote a lot -- on usenet (I still have those "Boing!" awards on the wall here), in magazine articles, etc.
I love to hear good said of what we did, but keep it as WE if you can. Thanks.
There actually was a Mac "emulator" for the Atari ST (which everyone called the "Jackintosh" when it came out) first. I didn't agree with the term "emulator" everyone used, since it really wasn't an emulator, but a port of MacOS to the Atari hardware, without Apple's permission. I dubbed this a "Hostile Port".
The early versions for the Amiga worked as well, but eventuall you got versions that ran as a more-or-less well behaved task under AmigaOS. That was pretty cool, if you needed Mac software... you could have Mac and Amiga at the same time. In those days, the Amiga had one of the fastest Mac hard drives, thanks to DMA, available -- dramatically faster than any "real" Mac.
I was a founder of Metabox, along with Andy Finkel (ex-Director of Software at Amiga) and two German businessmen, Stefan Domeyer and Geerd Ebeling. We were originally called PIOS Computer, back in the Mac Clone days. PIOS/Metabox had the first 300MHz Mac Clone shipping -- that should set the coordinated for your way-back machine. We bought the motherboards from UMAX, which also carried the license, and made our own CPU cards (actually designed by Thomas Rudloff).
I was working on a CHRP system, which wasn't terribly easy in the day. It had a separate CPU module, along the lines of what they had planned for the second generation BeBox (not precisely the same, but had they gone forward, it probably would have become so), and we had single and quad processor modules in development, G2 stuff in Apple terms. Future modules could have done G3, G4, or PPC970 for that matter. But Apple did pull the plug before this was finished, and Metabox [rightly] didn't see a viable market in a PPC machine that couldn't run MacOS. Of course, the Mac had over twice the market it has today.
The CPU modules kept selling, and Metabox acquired a US branch, based in Austin Texas, to bring some of this to the US market, but it wasn't expecially good timing, since Apple finally got aggressive with G3 machines.
We had three STBs -- the Metabox 500, based on the PC architecture and OS/2, the Metabox 100, which was an OEM from Teknema/Ravisent, and the never-completed Metabox 1000. That was my design, Thomas joined in later, and we had more people building add-ins for it, like a DVD/DVB decoder. This was roughly DVD-player-shaped. It ran a proprietary, AmigaOS-like OS developed under Andy and one of the Germans, Carsten Scholte(sp), called CaOS. The Amiga coonection was pretty key -- we tapped into numerous, well developed technologies like MUI (OO-graphics), Voyager (a browser), etc. This all ran on a ColdFire 5307/5407, not my top choice for a CPU, but a decent enough CPU if you had hardware for MPEG.
Metabox failed when the management got totally nuts, due to the stock prices rising (my shares, which I couldn't sell then, peaked at about US$5.8 million, but I got out of Metabox in terrible financial shape, with them owing me about $75,000 in salary alone). Basically, they spent money on nutty sponsorships: they tried to create a German basketball league, they sponored Forumla 1 racing, Soccer teams, etc. They bought a small film studio.
Meanwhile, the engineering team wasn't getting paid regularly, as the shares started falling in the fall of 2000. They pulled some maneuvers, probably illegal, that effectively stole all of my and Andy's shares in the company, replacing them with then-worthless, unregistered shares, all without our permission. A year of in-and-out of bankruptcy killed off the positive happenings at the US branch (I was CTO there in late 2000/early 2001, we were getting serious interest in the STB from Blockbuster, Enron, and others... ok, so maybe it was fated, anyway, to fail:-).
They went into another bankruptcy late last year, more of the Chapter 7 than Chapter 11 sort from what I heard, but I don't know the German rules that well. Basically, the management proved, in less successful times, to be a bunch of criminals, stabbing their own partners in the back this way. I'd love to report they're all in jail now, but German law doesn't seem to have much to say unless you're German (they actually have excellent protections for employees - thankfully, most of our crew didn't get hosed).
No, it won't die like Be died. And that's because it won't live like Be lived. BeOS was actually something new and better, not something "just different" running on slow hardware. So yes, it'll die, but not like BeOS.
AmigaOS development over the last nearly-eight years did happen, sorta-kinda. But it was very much gypsies in the palace. None of what made the AmigaOS great, none of the innovation continued. It was largely a mix of rehashes, ugly kludges, and territorial pissings.
If you want a useful non-MS-OS, run Linux. If you want a very interesting non-MS-OS that may yet have a real future, run OpenBeOS. Don't worry about going non-x86; despite some ugly legacy, x86 is the only architecture today with true chops. PPC is falling further behind every day. And even IBM's PPC970 will be little more that foot-dragging.
Slashdot Mirror
Actually, some cameras do use CMY, or more likely, CMYG. My old Canon Pro90IS had such a sensor. Maybe they're trying to minimize the color error rather than maximize sensitivity? Hard to say.
_ right.html. Their sensor is White (clear), Yellow, Cyan, and Green in a Bayer-like pattern. They made similar claims: the effect is 50% luma, rather than the 33% you'd presumably get with all filtering. This didn't really help -- that camera had really bad low-light performance, less than HD actual resolution, but they did make an interesting argument.
More interestingly, the very first HDV camcorder, the JVC HD-GR1, used both of these techniques back in 2003... see http://www.jvc.com/promotions/grhd1/unprecedent/s
They also had the advantage that, knowing your output is going to be 4:2:0 encoded, it's an easy argument to trade luminance for color accuracy that's getting tossed out anyway. Kodak's approach is likely to result in even worse edge effects (eg, interpolations that "guess wrong"), all things being equal. Of course, as you increase pixel density, this is less of an issue anyway for a unit-sized still photo. And, I suspect, their main goal here is to deliver higher resolution sensors without increasing the die size or making them crappy at low-light performance. This is more important in the highly competitive consumer camera market, but not a non-issue in pro-level stuff, as they continue to squeeze more pixels within something between APS and 35mm sized sensors, and simultaneously have a demand for better low-light performance.
Planned obsolescence, sure... the manufacturers like the idea that, at some point, you'll run out an buy a new camera, and new cards for that camera. But the notion that "it didn't make sense to design a standard for 4G+ cards in the days of 8M and 16M cards" is just silly. Particularly when you consider the history of things like SmartMedia, a card format that ran out of steam at 128MB, and alienated those buyers not prescient enough to know this was a stupid format.
And particularly, consider MMC and SD media... these are LOGICALLY addressed, for christsakes. One or four bits of data, to a logical model that have been anything they wanted. It would have been trivial to include address space for these from the get-go, and it wouldn't have cost anything more. This isn't reasonable shortsightedness, it's planned.
And take a look at the oldest memory card format on the market: Compact Flash. Sure, it has had a tiny bit of evolution (direct address vs. packet addressing, at least an issue on PCMCIA, from which CF derives), and there's an upper limit in there somewhere (132GB?), but it was a perfectly reasonable one for the day.
They're kind of fooling you on the 2003 to 2004 issues here. The 2003 Prius (I own one) is composed of 38 separate 7.2V Panasonic "Prismatic" NiMH modules, the system runs directly from that battery (more like 300V with a reasonable charge), and runs it over 40% of its useful range, which is what keeps it alive for a very long time.
Beginning with the 2004 model, they didn't substantially change that 7.2V module, but the battery itself is now composed of 28 separate 7.2V NiMH modules, for a nominal voltage of 201V. However, they've extended the range of charge/discharge, and the inverter is also now a step-up regulator, delivering a regulated 500V to the motors.
Thus, yes, the new full battery is smaller, cheaper, lighter, and has a higher specific power, but that's due to the new system, not any dramatic change in the battery technology. The smaller, cheaper, and lighter come from using fewer cells, the higher specific power comes from using more of the available battery capacity.
And the 2004 is actually the third generation Prius, the model I have (2001-2003) is the second. The first generation unit was on sales in 1997 in Japan only, and used what were essentially individual D cells. There were other differences... it ran out of battery power much easier (you see a Tortoise icon when this happens, but that's almost unheard of in the recent models), and the battery pack was larger and heavier, but less efficient. There were more significant changes in the battery formulation before the 2001 model was introduced.
And they've gone on record claiming that the 2009 model will run from Li-ion cells rather than NiMH. That's been kind of a holy grail, due to both the higher energy density and lack of entangling patents on Li-ion vs. NiMH. On the other hand, NiMH have traditionally had much higher current peak capacities, the well established long life as used in Hybrids (no so much in a full BEV that runs the full cycle), and they don't easily explode. Toyota may be planning to use some of the emerging nanotech anode technology for these cells, which promises faster charging, higher currents, and dramatically longer battery life... some day.
As many have pointed out, the patent was applied for in 1987, but only granted in 1991... thus, its in effect for 17 years from the date it was granted... that's the system prior to 1995.
However, it should be painfully obvious this is the illegal sort of submarine patent. Usually, companies used to attempt the legal sort -- they'd apply, they'd try to drag the process along as slowly as possible, counting on the 17 years they'd get once the patent was granted, the extra time before the patent is granted would hopefully get that patent into common use... particularly when the patent, such as this one, was so frickin' obvious that people re-created it continuously. The 1995 changes were designed to stop this as much as possible.
But tabbed interfaces have been around for quite some time. There were tabbed interfaces in AmigaOS programs... I can't swear they showed up before 1987 (AmigaOS 1.x), but this was the latest thing in the early 1990s. I know I've seen computer Rolodex-type programs using a tabbed interface (since that's basically the metaphore), very likely before 1987, though I can't think of the specific example. But obviously, this has been used in Windows and many other programs since then. They ought to lose enforcement rights on this patent due to the fact it hasn't been enforced for the last 15+ years.
I spoke with one of the "fixer" guys Verizon sends out when their first repair guy can't solve your problem; he's done all kinds of DSL installations. The DSL board set can live in any local node, it doesn't have to be at the central office... he knew the precise board set and software revision for the local node that's less than 1000ft from my driveway, and promised that Verizon would NEVER support DSL there. Period. Thus, the 0.97m 2-way satellite dish on my roof, and a hefty internet bill.
See, the telcos view DSL as a necessary evil today, but want to invest in it as little as possible. They're spending the crazy money on FIOS, which is the only way they stay competitive with the cablecos in delivering voice, internet, and TV. Naturally, of course, they're rolling out FIOS in areas already well served by DSL and Cable... I have a few buddies in the West Chester/Exton PA area who are getting FIOS-TV, 15Mb/s downlinks, etc.... must be nice. But if you're in a non-coverage area, don't hold you breath for DSL expansion.
Oh, phleeze! Is it actually a requirement to be an idiot to work for the FCC.
I'm one of those guys who can't get wired broadband, and I'll betcha they have me listed as "served" (it's ok, I have satellite, for about $100/month, that's delivers reasonable performance, unless latency is your big thing). I live in rural South Jersey, in a town covering about 45 square miles that's apparently too small to have its own zipcode. So we get to use a bunch of zipcodes from other towns... that's how zipcodes work, after all... they just link your house to the post office serving that house.
So you can get DSL in some areas of this zipcode, and cable in others... but the zipcode is for Monroeville, NJ. Monroeville is about 15 houses and a post office... most of the people in the Monroeville zipcode don't actually live in Monroeville. In fact, that zipcode covers part of two different counties. MOST of the people in the Monroeville zipcode probably can't get either form of wired broadband offered in this area, but we're no-doubt counted, because the zipcode covers a huge area.
Someone, I shoulda seen this one coming...
Consider also what Commodore brought the Amiga. First of all, the Amiga, Inc. people really, really didn't want to be acquired by Atari, and would have made out very poorly, financially. Commodore, as it turns out, paid them very nicely. And while I agree on the marketing side of things (C= didn't market the Amiga any better than anything else C= marketed), the technical issues were day and night, Commodore vs. Atari. Commodore did actually have a fab that could make the chips for the Amiga. They brought additional engineers into the projects, expanded the projects, set up a truly functional Developer program, evolved the OS and the hardware (not as much as any of us wanted, but way more than anything Atari ever did).
And in fact, the Amiga did ok under Commodore for awhile. It wasn't a rapidly growing business, but it was steady until the early 1990s, which a whole new level of mismanagers were hired. They managed to royally screw up what could have been some really interesting years for the Amiga.
In the end, the notion of doing in-house graphics hardware had to die, just as doing in-house CPUs had once worked for Commodore, but didn't by the time the Amiga came along. We in Engineering knew this; in fact, there was a project in 1993 indended to deliver a PCI-based successor to the Amiga graphics chips... it even had a real GPU (actually based on the PA-RISC processor ISA) and 3D engine. This was at least potentially something that would have sold in the PC market as well as being sold in Amiga computers.
-Dave
Toshiba announced research on a technology for fast charging li-ions over two years ago. This was using nanotech materials for an improved anode (maybe cathode too), enabling fast charging (80% charge in one minute) and long life (99% capacity after 1,000 charges). A similar approach was also annouced, about the same time, by Altair Technologys in Reno. It's all about increasing the effective surface area of the anode, and perhaps making it from stronger stuff.
In traditional Li-ion cells, a big wear factor is that the anode can form a parasitic battery with the electrical contact, causing the terminal to eventually wear out, faster as you approach full cycling the battery. Heat is also a factor, in both terminals and the full cell... the higher internal resistance of the Li-ion vs. NiMH (or better still, NiCAD) limits peak power, and also increases the risk of damage or, particularly in quesitonably made cells, explosions.
Dramatic improvements in both of these are necessary to enable practical (in a commerical sense) pure electric vehicles (BEV). There's no conspiracy necessary... traditional NiMH cells are a problem for full electrics.. which the actual reason none of these cars have been successful. Not to mention the expense... the Toyota EV-RAV4, for example, cost $42,000 and gave you about 100 miles on a charge.. and that with Toyota still selling them at a loss (as they did in the early days of the Prius, too).
In a hybrid, the batteries are only partially cycled (my 2003 Prius runs the NiMH cells over 40% of their capacity range; Toyota extended this to about 60% on the models starting in 2004), and that keeps them very long lived. Natrually, better batteries make a better hybrid, but the fact my Toyota can only go about 2-3 miles on a full charge doesn't impact its general use; the issues around battery technology today make the BEV a small niche product.
But the energy density is just too low even full cycling NiMH to make a BEV with mass appeal... most people would demand at least 200-300 miles of range, charging times on-the-road similar to that of petrol fueling (not the minimum of 15-30 minutes you'll have with today's cells), and long life (full cycling NiMH, they're good for about 500-1000 charges).
Once you have a higher density cell that doesn't wear out and can be charged in under 5 minutes, full EVs will be practical enough for a mainstream automaker to POSSIBLY launch a full production car, not just an experiment. This is critical technology for improving hybrids as well, and keep in mind that all practical FCEVs will also be hybrids (fuel cells suck at peak power demands, they like to be slow and steady, so you need a battery or supercapacitor to enable the peaks).
Well, I did read the article, and I can't tell what they're claiming is "new"... my first design with "upgradeable hardware" was in the Metabox 500 settop box (1998); this device included a modem for a datacasting technology called BOT (Broadcast Online Television, developed in conjunction with the University of Dresden), and the FPGA/LCA design was uploaded to the device as part of the device driver initialization for the modem. Anytime we released a new device driver, we could change the hardware.
And we could make some radical changes, since nearly the whole design was implemented in a low-cost FPGA. At $5.00, maybe a full custom chip would have been $1.00 (at least without amortizing NREs), but it didn't really matter, it was cheap enough. Anyway, even with a full FPGA design, there are always issues of routability... you can't make any arbitrary change and guarantee you meet timing requirements, or even if it'll fit. So while useful, this isn't as magic as it might sound to laypeople.
There would be serious issues with embedding this into CPUs or other complex devices. Other than the aforementioned speed issues (LCA/FPGA is always slower than gate array, which is slower than full custom, based on the same technology), it's only so good as where the FPGA is embedded. You can't fix any problem unless the whole design is implemented in an FPGA. There's also the startup issue... if you embed the FPGA design in Flash, you have the additional complexity of putting flash down in a complex CPU core (there are plenty of CPUs with on-chip flash, but they're usually things like ARM7/9, or simpler still), and the reality that Flash only lasts so long -- do the chips start to fail completely in 10-something years? If it's loaded at boot time, then obviously you can't FPGA any of the hardware needed at boot time.
I'm in the no-man's land of South Jersey, where broadband is dicey at best. Before I decided to upgrade, I had a POTS line pinned (with another bonded channel optional), truely unlimited use (well, other than the limits imposed by ~50kb-100kb/s) for $15 a month.
So here, there's no DSL, no cable... dicey. Last year, I hooked up to Verizon EvDO. In theory, I was paying $80 a month (ouch) for unlimited service. Well, except for the "unlimited" and "service" parts. At best, my EvDO Rev 0 connection would hit 500-800kb/s down, 60-80kb/s up. That happend twice, for about a week each. More usually, I'd get 100-150kb/s down, 10-30kb/s up (yeah, slower than dialup)... and this even with the addition of a 3W amp/LNA and a roof antenna. They were playing games here.
Even with unlimited EvDO, you get into weird and broken definitions of "unlimited". Essentially, Verizon knows it's a limited resource, but wants to compete with other hookups, so they call it unlimited. And it is -- no extra charges for usage... only, once you pass a magic double-secret download level, they toss you off.
Now I'm on HughesNet satellite.. for $100 a month I get 1.5Mb/s down and 250-500kb/s up. And while all networks are subject to aggregate load, at least in my area it's pretty dependable. This is definitely no unlimited, but they have very specific rules about "fairness", and if you exceed them, you're throttled back until you drop below a threshold (based on your service agreement recovery time). But it's totally OK to hit that limit, it's just enforcing fairness (and I gather, it's only implemented when there's actual contention for bandwidth), not a warning or anything.
While I'd much prefer FIOS (friends in West Chester, PA, now have Comcast cable and Verizon FIOS competing... competition does exist), I'm also happy to know the actual service agreement, not some double-secret (and thus, floating on arbitary whims) level you're not going to be told about up-front.
-Dave
It's actually based on the classic "Mimsy Were the Borogoves", a 1943 short story written by Henry Kuttner and Catherine Moore (under the pseudonym Lewis Padgett). See http://math.cofc.edu/kasman/MATHFICT/mfview.php?ca llnumber=mf300.
I loved the story when I read it back in High School. It actually does properly lend itself as an ET-style tale, since it's about young kids being taught very different ways of thinking. It's easy to believe Hollywood would mess this up like they have so many before it, but don't immediately discount it based on the fact that kids are prominent in the story.
There were a number of these systems over the years, ranging from glorified teletext systems (teletext itself is digitally encoded on one of the vertical blanking interval lines, so are close captioning texts) to something more complex. While working at Metabox AG in Germany, I helped deliver a system called BOT (Broadcast Online Television, originated at the University of Dresden) which allowed datacasting from 80kb/s (using just some spare space in horizontal intervals) up to a theoretical 4Mb/s (taking over a whole channel). While obviously just one-way, we used this to broadcast web content to our set-top boxes. The STBs ran web servers which could avoid expensive connect charges (usually POTS or ISDN in those days, and in Germany, there was no unlimited local calling, all dialup was pay by the minute) for content found locally. This was started in 1998 or so.
Intel had a somewhat similar technology for analog TV datacasting, though it didn't exactly set the world on fire, either. The advent of digital television and real broadband pretty much rendered these technologies pointless.
And these are not what's being discussed here. Reliable last mile broadband wireless, whether based on Wifi, WiMax, or something new, would be a great thing for folks like me, who are too far out in the boonies to get wired broadband. My current choices are cellular broadband (I've had Verizon EvDO for the last year, and it's flakey here, running from near DSL speeds to slower-than-POTS to nothing, even with the 10dBi roof antenna and 3W PA/LNA) or satellite (HugesNet guys are coming tomorrow). Both are dramatically more expensive and slower than wired broadband, and have other issues (satellite's latency, cellular's flakiness and double-secret usage limits).
If you're upconverting VHS to DVD, you can pretty easily get 6 hours of VHS-quality video on today's DVD. The 720x480 format (NTSC and NTSCfilm resolutions) is fine for commercial films (and on a dual layer, you can store well beyond two hours worth, even at that resolution), but for VHS, it's overkill. DVD also supports 352x480 MPEG-2 and 352x240 MPEG-1 resolutions, either of which will do justice to plain old VHS tape (you'll want full resolution for well recorded SVHS).
VHS vs. DVD was a different story -- analog linear tape vs. digital random access disc. VHS sucked (in the sense of the engineering term "it sucks"), DVD isn't bad, even on a good HDTV.
The post-DVD war, unfortunately, is looking more like DVD-Audio vs. SACD than VHS vs. Beta. Yeah, they'll be better. But worth the extra cashish to most people? That's debatable, for exactly the same reasons as the extended rez audio discs. While predictions claim as much as 40% HDTV penetration by 2007, that's still missing the point: that's at least one set per house. You might watch HD-DVD/Blu-Ray on your HD set, but it doesn't play in the kids' room, in the car, on the portable, on the older PCs, etc.
Meanwhile, you can buy a DVD player that'll play DiVX-HD, Microsoft WMV9/VC-1, and HDV-compatible MPEG-2.... today. The hardware guys already grok the fact that the 5" round shiny thing is just a disc, and the player can host any number of applications. And must, to remain competitive in a hard market.
Whoops, hit SUBMIT rather than PREVIEW. The actual number would be, er, $30 per charge, best case, plus the power costs. But that's a subsudized battery pack, maximum expected life, and NiMH. You're probably looking at $60-$120 additional cost per charge, for the batteries, using state of the art Lithium and commercially profitable predictions for cell life. Then there's the fact that the RAV4-EV only got 80-120 miles on a charge; you're going to need to about double that for consumer acceptance. That could double it, or more, so you're probably getting close to a worst-case charge of $300 per "fillup". Obviously that won't fly.
The problem isn't so much EVs in general, or even battery life. The problem is the time it takes to charge (the best technology runs about 4C, 15 minutes for a full charge, IF you can supply that much current, which you might at a service station, not at most homes) and the cost of the batteries. There are some promising ideas using nanomaterials that might achieve a very fast charge, but the cost issue is still a big problem.
There are two big problems with battery exchange. First is capacity -- as they age, the batteries hold less charge. People aren't going to like old cells, and they certainly aren't going to want to pay per replacement pack.
Second is cost -- you have to amortize the cost of the battery pack across the life of the battery, and I'm guessing service stations will pass that on to the customer. Consider the cost of a NiMH pack, such as that in the Toyota RAV4-EV, was listed at about $30,000, and that with Toyota selling the $42,000 car at a loss. Even assuming a Li-ion pack cost no more, you're only getting about 1,000 charges out of a Li-ion or Li-poly pack these days, at best. So figure you have to tack on something in excess of $300 for each fillup, just to cover the batteries. Consumers would not accept that, even if the power in the cells came for free.
Actually, no, that's not true, the "55 by itself" thing.
The way the Prius works, you actually have two motor/generators. MG1, the small motor, which starts the engine and, at low speeds, acts as a generator, and MG2, the large motor, which is on the final drivetrain and gives you that big torque when you start up.
MG1 serves another function: it balances the power between the ICE and MG2/wheels, providing the effect of a CVT, even though the Prius has fixed gearing. So when you're going slow, MG1 acts as a generator, dragging on the "power-split device" and providing the effect of a low gear (it's running against the movement of the ICE).
But above about 60mph, MG1 has to, instead, push in the direction of the ICE motion, acting as a motor rather than a generator, to provide the effect of a high gear. During this, MG2 is engaged as a generator to provide MG1 the power needed to do this. So there's no charging, period. But also, no dependency on the batteries: you can and do get to the top speed of 99mph (for the 2001-2003 car) regardless of the battery charge.
Hey, I'd go for the Jetsons thing, but hack in a manual control, so I could still "go stealth" if need be in my Prius.
Thing is, the Prius being as small as it is, at least WE can both see and hear you, on your bike.
You're taller than we are, and louder. And since it's a green-tech vehicle, you can also bet that a significant percentage of Prius drivers, such as myself, are also cyclists. So we're watching your back, even when you don't know we're there.
Compare that to the Hummer driver. Between the blare of bad country music, the loud roar of the disel or that curiously-even-less-efficient V8 in the H2, and the perpetual state of shock of the average Hummer driver, given the $100-or-so cost of a fillup, they don't know you're there. And even the satisfying crunch the typical SUV driver gets destroying cyclists is lost on the Hummer driver, as the combination of active suspension and >6000lbs curb weight conspire to deprive said pilots of the usual SUV pleasures.
That's not quite the story.
Panasonic/Matsushita, like every other NiMH battery licensee, pays all their royalties to Ovonics. If they had not, Ovonics could have actually blocked the import of the Prius to the USA. Unlike most of the licensees, they've done their own share of improvements to the technology, and have their own patent portfolio.
The thing with Panasonic and Ovonics is not the royalties per se, but the patent license itself. Ovonics has had no interest in making D-Cells, they were, from the start, after the Hybrid car market... only it didn't quite exist. At the time Toyota started work on the Prius, and even until recently, Ovonics was only R&D, no production facilities at all. Their NiMH license was for small cells only: AAs, Cs, Ds, cellphone and laptop batteries, but not the giant cells you'd need for Hybrid cars.
So Panasonic and Toyota got together on this, and built a hybrid anyway. The original (1997-2000) Prius actually used plain old everyday high-output D-Cells. For the 2001-2003 model, Panasonic supplied prismatic cells in packs of six, to Toyota. They, in turn, not being subject to any legal agreement between Panasonic and Ovonics, built up 38 of these (28 in the 2004 model) and voila -- hybrid power pack.
So basically, it sounds like Panasonic had smarter lawyers than Ovonics. The main focus of the lawsuit by Ovonics was to prevent Panasonic from doing similar things in deals with Ford or GM, in any hybrid projects they had in the works. As well, it's likely they were after some of the Panasonic technology -- Panasonic has the highest output cells, by far, of any NiMH vendor.
And it's also a limited time thing... Li-Ions or Li-Polys are likely to take over in the hybrid market, once they figure out how to keep them alive for 150,000 miles. And, I suppose, ensure they won't spontaneouslyh explode.
Honda, an investor in Ovonics, apparently doesn't have any issue. But they make very weak hybrids anyway. The Toyota system is much cooler... and I'm not just saying that as a 2003 Prius owner.
Hmmm.... that might explain a few things. I went to the Salem County fair (NJ) last night, to watch the pig races and pig scramble (really, me and my kids). I might be wrong, but I'm fairly sure I saw two prize sows with stupid grins on their faces, wearing SCO logo T-shirts.
Yeah, it's significant.
Some years back, when I was VP of Technology at Metabox AG, I wanted to sell off a fairly large number of share to apply to the US branch of the company. The German company, on the other hand, didn't want it to seem that some high ranking corporate dude was getting ready to bail. So I cashed out some (ok, in hindsite, it should have been all of it) in dozens of smaller transactions. Sure, you'll find these, as in that Yahoo link, if you go looking, but it apparently prevents any big alarms from sounding.
Looks to me the SCO guys are "gettin' while the gettin's good", using the same techniques to hide it as best as they can. The obvious flaw in their logic is that 10's of thousands of Linux fans are watching their every move. You'll even see that some guys, like Broughton, are making multiple same-day transactions under different corporate titles.
What HAVE I been doing lately? Well, in The Work Thing, I'm the principle HW guy at Sizig, Inc; a small startup company, based in Philadephia. Our primary project is a consumer robot, but probably not what you're thinking. If successful, maybe a sort of "C64 of home robots", in most of the ways you could apply that idea. This is the best thing I've been involved with since Commodore. It's a good feeling!
We're well aware of the bad investment environment these days -- you're crazy to do a start-up now, right? But we have, it's been going over a year-and-a-half (I was involved since April 2002, and fulltime since October), so we've kept the development smart, as opposed to BIG, and will not rush this project. I'm still working most days out of my home office/lab (about 1000 sq. ft. in my cellar, more space than I personally had a Commodore).
In the shorter term, we have a spin-off of some elements of this technology into another market, which I won't discuss right now. This is one of those curious markets that's fairly large, not understood to be that large by most people, and totally back in the 40's or so, technology-wise. We think that dragging them into the 21rst century will be well met.
I'm also doing more video than ever... which I really didn't do back in the Amiga days, there just wasn't time for hobbies or any sort of side business. I'm shooting weddings and other events, making DVDs, etc. Not really THAT strange; before I was a computer nerd I was a photography nerd. The DVD is the ideal place to unify all my skills: computers, video, audio, photography, artwork, music, etc. I did a "Deathbed Vigil" re-release last year, as a justification to buy the software needed, and I have a PAL version on the way REAL SOON NOW.
Thanks for the interest, and a chance to write about some of the good stuff [other bad stuff intentionally left out]...
Jay was certainly THE father, but the whole team in Los Gatos was responsible for the Amiga. I was certainly part of the second-wave effort, in West Chester, and sure, as a group WE took on the role of the godparents -- bringing her up right in the absence on the original parents. I don't like to be singled out that way, this was a whole team effort. Some journalist made up that "godfather" thing. I really found it embarrassing.
Of course I worked my ass off on the Amiga, but most of us did. That's way a team of around 50 people, total, could do what most companies didn't when they had hundreds or thousands involved. My name got out, I suppose, because I wrote a lot -- on usenet (I still have those "Boing!" awards on the wall here), in magazine articles, etc.
I love to hear good said of what we did, but keep it as WE if you can. Thanks.
Whoa! My ears are burning!
:-).
There actually was a Mac "emulator" for the Atari ST (which everyone called the "Jackintosh" when it came out) first. I didn't agree with the term "emulator" everyone used, since it really wasn't an emulator, but a port of MacOS to the Atari hardware, without Apple's permission. I dubbed this a "Hostile Port".
The early versions for the Amiga worked as well, but eventuall you got versions that ran as a more-or-less well behaved task under AmigaOS. That was pretty cool, if you needed Mac software... you could have Mac and Amiga at the same time. In those days, the Amiga had one of the fastest Mac hard drives, thanks to DMA, available -- dramatically faster than any "real" Mac.
I was a founder of Metabox, along with Andy Finkel (ex-Director of Software at Amiga) and two German businessmen, Stefan Domeyer and Geerd Ebeling. We were originally called PIOS Computer, back in the Mac Clone days. PIOS/Metabox had the first 300MHz Mac Clone shipping -- that should set the coordinated for your way-back machine. We bought the motherboards from UMAX, which also carried the license, and made our own CPU cards (actually designed by Thomas Rudloff).
I was working on a CHRP system, which wasn't terribly easy in the day. It had a separate CPU module, along the lines of what they had planned for the second generation BeBox (not precisely the same, but had they gone forward, it probably would have become so), and we had single and quad processor modules in development, G2 stuff in Apple terms. Future modules could have done G3, G4, or PPC970 for that matter. But Apple did pull the plug before this was finished, and Metabox [rightly] didn't see a viable market in a PPC machine that couldn't run MacOS. Of course, the Mac had over twice the market it has today.
The CPU modules kept selling, and Metabox acquired a US branch, based in Austin Texas, to bring some of this to the US market, but it wasn't expecially good timing, since Apple finally got aggressive with G3 machines.
We had three STBs -- the Metabox 500, based on the PC architecture and OS/2, the Metabox 100, which was an OEM from Teknema/Ravisent, and the never-completed Metabox 1000. That was my design, Thomas joined in later, and we had more people building add-ins for it, like a DVD/DVB decoder. This was roughly DVD-player-shaped. It ran a proprietary, AmigaOS-like OS developed under Andy and one of the Germans, Carsten Scholte(sp), called CaOS. The Amiga coonection was pretty key -- we tapped into numerous, well developed technologies like MUI (OO-graphics), Voyager (a browser), etc. This all ran on a ColdFire 5307/5407, not my top choice for a CPU, but a decent enough CPU if you had hardware for MPEG.
Metabox failed when the management got totally nuts, due to the stock prices rising (my shares, which I couldn't sell then, peaked at about US$5.8 million, but I got out of Metabox in terrible financial shape, with them owing me about $75,000 in salary alone). Basically, they spent money on nutty sponsorships: they tried to create a German basketball league, they sponored Forumla 1 racing, Soccer teams, etc. They bought a small film studio.
Meanwhile, the engineering team wasn't getting paid regularly, as the shares started falling in the fall of 2000. They pulled some maneuvers, probably illegal, that effectively stole all of my and Andy's shares in the company, replacing them with then-worthless, unregistered shares, all without our permission. A year of in-and-out of bankruptcy killed off the positive happenings at the US branch (I was CTO there in late 2000/early 2001, we were getting serious interest in the STB from Blockbuster, Enron, and others... ok, so maybe it was fated, anyway, to fail
They went into another bankruptcy late last year, more of the Chapter 7 than Chapter 11 sort from what I heard, but I don't know the German rules that well. Basically, the management proved, in less successful times, to be a bunch of criminals, stabbing their own partners in the back this way. I'd love to report they're all in jail now, but German law doesn't seem to have much to say unless you're German (they actually have excellent protections for employees - thankfully, most of our crew didn't get hosed).
No, it won't die like Be died. And that's because it won't live like Be lived. BeOS was actually something new and better, not something "just different" running on slow hardware. So yes, it'll die, but not like BeOS.
AmigaOS development over the last nearly-eight years did happen, sorta-kinda. But it was very much gypsies in the palace. None of what made the AmigaOS great, none of the innovation continued. It was largely a mix of rehashes, ugly kludges, and territorial pissings.
If you want a useful non-MS-OS, run Linux. If you want a very interesting non-MS-OS that may yet have a real future, run OpenBeOS. Don't worry about going non-x86; despite some ugly legacy, x86 is the only architecture today with true chops. PPC is falling further behind every day. And even IBM's PPC970 will be little more that foot-dragging.