Me, a sysop in Ohio, trying to dial in to a friend's BBS in Florida, and finding it busy. Since I had S11 cranked down to 11 or 12ms (or whatever the fastest speed the phone company was capable of processing that week was), I was redialing like mad (God bless Supra for mailing me, -free-, firmware which supported such insane speeds), and with the busy detection also cranked down to just-barely-reliable periods, I was redialling like mad.
After a minute or two, instead of the usual touble-tap busy signal, half-second silence, and a rapid-fire redial, I heard prolonged silence and then a slight click.
Strange, I thought. I let the machine continue to redial.
After a few more minutes, the same thing happened again: no busy signal, no ringback, just dead silence.
I quickly exited to a terminal, picked up a phone, typed ATX3D, and hung the phone back up. I heard the modems negotiate and connect.
The guy in Florida, who thought he was calling a local BBS, was instead connected (at my expense) to my Telemate session..
We talked for a bit in Telemate's split-screen chat. He was obviously quite surprised to find me typing to him from Ohio when he thought he was calling a BBS across town. I told him how to fire up his WWIV BBS using the existing connection, without dropping carrier, and we talked for a bit more using WWIV's superior (and still un-matched) split-screen chat before I checked for new files and logged off.
But at the time, the situation dictated that I either install a locally-available, department store 7200 RPM drive, or wait at least a year to upgrade the box. Given those choices, it was pretty clear to everyone that getting a 7200RPM drive right now was the obvious way to go.
I spent half a day in the car and on the phone looking for a reasonably large 5400RPM drive, to no avail.
Ambient temperatures of 100F aren't hazardous to most consumer electronics. In terms of durability: Given exclusion from severe catastrophic failure syndrome catalysts such as lightning strikes and 2-year-old children, anything in your cabinet which doesn't rely on moving parts is very likely to outlive you and a number of your descendants.
That said, neither the bearings in your DVD player nor the VHS tapes you play are likely to be happy, long-term, with such elevated temperature.
Additionally, the properties of the individual components (caps, resistors, transistors) change with temperature, so well-designed analog electronics are engineered with a specific temperature range in mind. They'll certainly sound best when operated at whatever ambient temperature they were designed for, which is likely to be at or slightly above room temperature (72F).
The thermal switch in your amp is not likely to trip until the heatsink is justabout hot enough to boil water. It exists as a safety feature, like a fuse, to turn things off under abusive situations or in catastrophic modes of failure. It is not, in any way, a device intended to ensure proper fidelity.
My parents have a similar situation at their house. They've got a 36" Sony CRT, on top of a glass-doored Sony stand. Inside this stand resides all of the extra components associated with the TV - a DVD player, VCR, and DirecTV TiVo.
After adding an 80-gig, 7200RPM Maxtor to the TiVo, things would get hot enough inside of the cabinet that the TiVo would lock hard every couple of days.
They simply removed the glass doors, and everything has been rock-solid stable since.
I recommend you do the same.
Not only will your components be more accessible, you won't need to worry about things being too hot. It's also free.
In my own living room, I solve the heat problem differently. I've got the line-level (minimal BTW/hr) stereo components stacked neatly on a shelf, the TV on its own seperate stand along with the PSX and DVD player, and a fan-cooled power amp in its own rack back in the far corner of the room, hidden behind a plush chair.
By spreading things out and avoiding confining furniture, heat becomes a non-issue. And I also get to keep the more dangerous components (the ones with volume controls, capable of producing dangerously-loud, eviction-level radio static) up out of reach of my 2-year-old daughter.
If none of these solutions are appealing, simply install a largish, slow-moving fan near the top of whatever cavity houses your AV components, exhausting air out the back. Maybe something like this would do the trick. If such an arrangment turns out to be too loud, wire a rheostat in series with it to slow it down even more.
You could also use a low-voltage DC fan, but it'd take all of the fun out of it and require the use of a seperate power supply.
Whichever the case, the purpose here is not to actively cool the components, but to simply provide a mechanism for exchanging the stale, warm air inside of a cabinet with cooler air from outside, be it by convection (avoidance of enclosed cabinets and glass doors) or force (a fan to push things around).
That sounds very similar to an idea I was considering...
My variation is to attach a number of large IR LEDs around the front license plate and inside of the headlight reflectors, modulated so as to confuse laser speed measuring devices.
On a clear day this could provide some minimal added protection against speeding tickets.
Clearly. And in the context of the article (did you even RTFA?), there is no "smart charger" to be had. This makes the point rather moot, don't you think?
And now for something completely off-topic:
Have you ever, even once in your life, agreed with someone in an upfront manner, or admitted that you were wrong?
Or are you one of those people who are always right, and will always state fault with what someone else is saying, even when there's a chance that you're not right and that the other person's statement is without flaw?
I never said NiMH batteries suffered from memory effect, just that they have issues with being cooked too long in the charger.
Try it sometime.
NiMH and NiCads each have very similar charging voltage and current, so it's quite possible to use NiMH batteries in a NiCad charger.
Here's what you'll need:
A set of NiCads, a set of similar NiMH, and two examples of the same type of (dumb, timerless, non-smart, plain-old-voltage-source) charger, such as that found on a cordless phone base.
Load one charger with NiCads, and the other with NiMH, and make the chargers charge for 12 or 14 hours - whatever seems typical for that battery capacity in that charger.
After this, measure the capacity of each set of batteries. A small flashlight and a stopwatch will work fine for this (but don't let things discharge completely - that's another type of abuse, altogether). Record your findings. The NiMH battery will probably last somewhat longer than its NiCad counterpart.
After that, repeat the experiment using the same batteries, except increase charging time to one week instead of half of a day. The NiMH batteries will lose substantial capacity during this time, which you'll see with your stopwatch. The NiCads will still have similar capacity to what they did during the first experiment. Record your findings, and realize this:
No technology is perfect. Repeat it over and over until it becomes clear. No technology is perfect. No technology is perfect. No technology is perfect. Once you get that to sink in, start with the following: Nothing is what it seems. Nothing is what it seems. Nothing is what it seems.
After a few months of this, perhaps you'll be able to emerge from Marketroid Consumerland and join the rest of us thinkers.
While it's true that NiMH batteries do not suffer the same "memory effect," per se, of NiCads, they are not without their own particular problems and quirks. Chief among these is how susceptible they are to damage from being charged too long. They also leak (self-deplete/drain/reach maximum entropy) faster than NiCads, discharging themselves relatively quickly while not connected to anything at all.
From the tone of your article, the phone itself sounds like it is relatively old.
Nicad batteries wear out. This is normal. They're also the cheapest and safest rechargable batteries you're likely to find, bar none.
So, just replace them. Find a well-stocked Panasonic phone dealer (amazon?) and get the genuine article, or just pick up the appropriate off-brand replacement at the department store of your choice.
While it certainly wouldn't be impossible to design a cordless phone that used rechargable lithium cells, noone seems to market such a beast to the consumer. Probably because lithium batteries are very good, very light-weight, and also very expensive. They also tend to react explosively to mistreatment during the charging cycle. Stay away from lithium rechargables in devices not expressly intended to use them. [FWIW, Motorola Startacs have a microprocessor in the battery pack itself, expressly for the purpose of lawsuit prevention - I imagine most/all other companies do the same thing, but I've never taken apart any other Li-Ion packs.]
You might be able to change the nicad cells with NiMH, but you're not likely to get any net benefit from the exercise. Besides, NiMH suffer severe stamina problems after being charged for too long at one time, which is something that chargers designed for this type account for, and your cordless phone does not.
So. With NiMH, you'll have to remove the cordless phone from the base after no more than 12 or so hours of charging, followed by a substantial drain before the next re-charge. Unless you're able to do this every time you charge it, and are able to train your friends and family to do the same, I'd avoid a NiMH retrofit, unless you anticipate and accept that permenant battery damage will result. [At least they don't tend to explode, like lithium batteries do...]
And at any rate, if you had the discipline to follow such strict procedures, you'd have plain old, cheap NiCads that last forever, just like me.
I've got a 4-year-old, vanilla, Uniden 900MHz analog phone. It works well, having great audio quality at long distances through any number of obstructions.
I replaced the stock battery after a year or so with whatever compatible type Sears had at the time, and things have been golden since.
Tricks:
I use the phone until it's either inoperably dead, or starts in with its low-battery song and dance (blinking light, funky stutter ring when someone calls). When this happens, I place it on the charger, and leave it there until at least the next day. When I feel like using it again, anywhere from six hours to six days later (depending on alcohol intake), I keep it off the charger until it is once again dead. The whole cycle takes from 5 days to just over a week, depending on use.
Those who are keeping score will note that I've got three years of outstanding service from a cheap Sears battery. It takes a slight effort to keep things in-check, but the system (as a whole) is reliable and predictable, which is really all that matters.
That all said, I echo your sentiment: I've got a Nokia cell phone with their smallest available lithium battery. I only ever use it with the cheap, third-party charger I have in the car, which cooks the battery completely in just over an hour. Routine fast charging like that is supposed to always be bad for batteries. But Lithium batteries seem to Just Work(tm) no matter what I do to them, and I still get great talk time and days worth of standby. Must be that microprocessor...
I'd love for a cordless phone manufacturer to start using modern battery systems. I'm even willing to pay extra for it the convenience of not having to bother with the proper care and feeding fo the battery.
Even my UPS, a Best Ferrups 850, maintains its sealed lead-acid battery (a Panasonic) automagically. Once every 1 or 2 weeks, it'll switch over to battery power for a few minutes at a time. It then tops off the battery again and resumes keeping it on float charge. This is all part of the normal care and feeding of a lead-acid battery (your car does it with its battery, too, by virtue of you running the starter motor), but most UPSs don't do it at all. OTOH, I suppose this is no ordinary UPS, but why can't phone manufacturers adopt at least semi-modern battery technology? NiMH would be a good, safe, and inexpensive start. Even my RioVolt came with a pair of AA NiMH batteries, and enough smarts to avoid abusing them.
I'll be the first to say that I don't know much about drilling, except that when I turn the faucet on at my parents' house, water comes out. Every now and then, the lights dim briefly when the pump turns on... And that's nowhere near the scale being discussed.
I don't have any paraffin or diesel here, but if you do, try this quick-and-dirty experiment to measure 2.4GHz RF absorbtion:
Put a container of it in the microwave. Does it get hot? If so, how does the temperature compare to an equal amount of water in a similar container, in the same spot of the oven, after being nuked for the same amount of time?
Therein lies a rough answer to your question, though it does a ignore number of possibly important factors, such as reflection and refraction of the signals and probably a slew of others that few people outside the realm of microwave engineering really know about.
As long as you realize that your findings are based on absorbtion alone, you'll be doing fine. There's probably a proper method to quantify it with, with a proper, capitalized Unit to go along with it, if you feel like being really anal about it.
Now that I'm thinking about it more:
If the results turn out to be negative (as in, no substantial heating occurs), it might be interesting to ponder using the metal pipe as a waveguide. Such a transmission system would be incredibly efficient, but would place constraints on the type of joint used, and the length of the pipe segments would have to be precise, corresponding to some factor of the wavelength of the signal in whatever medium ends up filling the pipe.
High-frequency RF does not penetrate earth very well.
2.4GHz signals, in particular, are very trouble-prone in this application, as water converts it to heat more efficiently than any other frequency. Drilling is a very wet operation.
Hint: this is why your microwave operates at 2.4GHz, and why the band is unlicensed. Because it is so readily absorbed by masonry, trees, and other relatively wet objects, it was deemed (at least a few years ago) relatively unsuitable for serious communications and kept from being sold commercially since the beginning of time.
Have you never driven through a tunnel with the radio on, or while using a cell phone? FM radio is down near 100MHz, well into the range of relatively slow data transfer.
You need VLF radio to get through that much solid crap, and once you do that, you're back into the slothly realm of measuring things in bits per minute.
'sides, aiming a 1-megaWatt microwave oven down a drill pipe would not make their already-existing heat problems any better...
Uhhhh, some people get better cooling with the case OFF their computer. A lot of older cases (and many not-so-old designs) do not have ducts going to/from the hot spots. The case is seldom part of the nonexistant ducting system. Remove the case, and you HELP ambient air exchange, thus helping your ability to cool. So removing the case is a recommended MOD.
This isn't a shiney new aluminum Lian-Li, Coolermaster, or stealthly-quiet, duct-filled Dell. There's no Radeon here, no PC2700DDR, no dual-proc Athlon MP. It's an 8-year-old Infotel full-tower AT, with a 350MHz K6-2. With the top off, its trio of 7200RPM IBM Ultrastars, fan-cooled Plextor burner, and CPU all run cooler. The Voodoo3 overclocks better. The DDS-2 streamer acts healthier.
While I understand that you were just trying to be helpful, I'd like to assure you that your concern is unwarranted.
I'd also like to ask if you've got anything constructive to add to the topic at hand, which has everything to do with radiation and casemods, and nothing to do with cooling.
CRT monitors are affected more by low-frequency EMI than the RFI computers emit.
A simple test: Take the lid off of your case, and place it beside the monitor. Nothing strange happens.
Next, place an AC-operated fan, transformer-based soldering iron, or similar magnetic device next to the monitor, and watchen das blinkenrainbow.
That said, flat peices of steel (such as that which comprises your case) do very little to counteract low-frequency magnetism, while aluminum does absolutely nothing. Common steel can have some low-freuquency shielding effect if it's curved just so, but that's usually impractical. (there's other stuff, such as Mu-Metal, which is formulated with the specific goal of blocking EMI, and does work quite well. But it's expensive, and hard to find.)
I have to be careful where I put my Best FerrUPS because the large ferroresonant transformer in it will cause monitors to shake from several feet away.
Problems with computer-generated RFI generally show up with radio and television. I can't listen to an AM radio anywhere near my apartment with the PCs on, and there's a few FM stations that I can only recieve outside or in the back bedroom, away from the machines.
My neighbors must hate me for it, as I'm sure it's not much better anywhere in the building. But the 300-pound, heavy-footed woman upstairs has four kids who wake up at 5:30 AM daily, and the people directly beside me have a bad habit of listening to one-note bass lines with their lousy, one-note subwoofer, directly on the other side of the wall behind my desk.
So, I guess I care a lot less about RFI than I do about proper cooling. Thus, the top of the case is completely absent, allowing all kinds of natural, quiet convection cooling to take place.
I not only have a high-density 5.25" drive, but I've also got a sample of every tape drive I've ever owned.
I've still got my "important" data from 10 years ago living on CD, having recently been re-discovered (I didn't create much, back then, but the old MS-DOS utils directory has such oddities as pkarc and zoo which just aren't easy to find these days, along with a few hundred other fun things still applicable to my 386SL laptop).
I admit that I'm not very cautious with my choice of CD-R media. I generally choose the cheapest I can find locally (most recently: ~$17 for a spindle of 100 unbranded blanks), and my trustworthy 8x SCSI Plextor burner generally does a fine job of keeping the bits intact no matter what kind of trash I feed it. Time will tell as to whether or not this is a reliable way to archive things, but I'm betting that as soon as DVD-R media comes down to $1, I'll be jumping on that bandwagon instead.
Important stuff isn't very difficult to keep around, in my experience. In the early 90's, my father contracted an audiophile friend of mine to record some 1950-vintage wire recordings of his family to cassette. FWIW, the wire recorder itself was working JustFine, and required no repair or reconditioning to play these glorified spools of bailing wire with good fideliy.
A year or two ago, during the same family's Christmas gathering, I (on a whim) made audio CDs of this 50-year-old dialogue and distributed them. I used a 5-year-old Onkyo tape deck (which I'll probably never get rid of) and a well-designed Yamaha XG-based sound card. The transfer was, to my ears, perfect.
Does anyone make wire recorders anymore? Nope. Does anyone make good cassette decks anymore? Perhaps, but you'll have a hard time persuading me that it's not just old stock. 5.25" drives? I'll be honest: I never owned one, until a couple of years ago when I picked one up at a shop specializing in used PC parts. I cut my teeth on PCs in the late 80's, when 3.5" media was sure to be a success. (And what do you know - 3.5" drives are still standard on justabout every new non-Apple machine you can buy.)
So, no. I'm not worried about the shelf life of my CDs in the slightest.
Even color film isn't so good, either. I currently work in a department store photo lab, for lack of a better job. A couple of weeks ago, someone brought in a roll of 200-speed Kodak from 1984 (judging by the t-shirts worn, and the C-64 connected to a TV with knobs on it). The colors were abyssimal. Absolutely horrid. An artist's rendition of a technicolor nightmare. I was able to get a few good prints from the roll, but things were -really bad-, overall, with hideously-colored people ranging from magenta to olive drab on a frame-by-frame basis.
I expect my bargain-bin CD-Rs to fare at least as well as that, and hereby submit that CD-ROMs are the 35mm film of the digital age.
If it were me and I had the cash to drop, I'd go for a mini-ITX system, or one of Shuttle's prefab mini-boxen.
They really shouldn't require much time to set up - there won't be anything involved other than selection, ordering, and assembling of parts.
The laptop may actually represent a larger time-sink, unless it's got a vanilla CD-ROM drive built-in and fairly standard PCMCIA hardware, which would make it easy.
If you've got no removable mass storage for the CTX box, invest a couple of dollars in a 2.5" -> 3.5" IDE kit. This will let you use a desktop machine to do the base install of software.
Since storage is cheap and demands are few, you'll probably want to leave a copy of your Linux distro on the hard drive in its own partition, so you can get back to a bootable state without pulling the drive, in case Bad Things happen. Slackware still fits on one CD, FWIW - you won't give up too much space for it.
I don't know about ogg, but MP3 is not at all tied to having things sampled at a specific bit-depth - it just doesn't matter to it whether it is 16, 20, or 24 bit - the format has sufficient dynamic range. I know of at least one free 24-bit MP3 decoder, but I've not run across anything for the encoding side just yet (and I haven't been looking, either).
'sides, HDCD essentially only adds 4 bits worth of dynamic range, for a total of 20, which isn't such a stretch for current consumer audio gear. The human ear is likely to be a greater limiting factor than the equipment.
And finally, there will never be any "exclusive content" on the HDCD layer, because there is no HDCD layer. The extra four bits are gleaned from unused space on a CD and interspersed with everything readable by a normal CD player. HDCD is an adjunct to a standard red book audio CD, not a replacement for it. Take away red book compatibility, and you not only remove 16 of HDCD's 20 bits, but eliminate compatibility with existing HDCD players (which are numerous indeed).
You're thinking of Sony's SACD, which can include multiple layers, and currently lacks ripping tools and PC-compatible hardware. I'm sure that Sony would love to convert the world to SACD and then stop adding a red book layer to their music as a cost-saving measure.
I doubt it will ever happen, however. Remember, this is the same company that brought you Beta, Minidisc, and Memory Stick. All of them are good, well-designed products, but lack of interoperability has always held them down, just as it will SACD.
HDCD, while developed by a high-end audio company whose name currently escapes me, is now 100% owned by (you guessed it!) Microsoft along with the aforementioned original developer.
You can find an overview of how things work during the encoding process here, but don't expect anyone to tell you how to do any of those things without you first handing them vast fistfulls of cash.
You might be able to glean some useful information from the patent text, but probably the only sane way to go about this effort would be to read the bitstream coming from the CD. And while you should able to discern what the bitstream looks like without too much effort, it would probably be a fairly involved task to learn what it means.
So. My only suggestion would be to give up now before you've wasted any effort on trying. But if you insist on putting real time into this project, here's a couple of nice encapsulated postscript HDCD logos you can illegitimately use to adorn any illegitimate HDCD products you produce.
Since nowhere in your post did you mention that it must be a fast machine, and your desired goals are very light-weight, just buy a cheap, slow Pentium machine from Ebay or a place like this.
Save energy how you can, if it's important to you. Toss the CPU fan, and keep the heatsink. GlobalWin makes some huge Socket7 heatsinks which are suitable for this, all of which come with easily-removed fans screwed to them.
I've got a P133 which has been running various incarnations of Windows (now 2k) for years, with only a quiet PSU fan and a modest 6.5gig Seagate drive which spends most of its life spun down. It's nearly silent, doesn't make much heat, and I don't even think about its power consumption.
Configuring hdparm/apmd/kernel/BIOS to put the system to sleep would be good. As long as you don't let it drop into suspend mode, it'll come right back to life as needed.
Avoid hardware that you don't need. Don't use a sound card, find a slow/old/efficient video card. Keep things simple.
If you're worried about the reliability of a used machine, don't be. Remember, only the moving parts are subject to wearing out in normal use. Of these moving parts, you'll be completely eliminating the CPU fan. You can buy a nice new Sunon or Panaflow fan to replace whatever comes with the PSU, either of which should last a long, long time (the last dead fan I replaced was a Sunon that I've had spinning for 8 years).
All that's left is the hard drive, and you'll probably want to buy a couple of new ones no matter what you do, anyway, so that you've got two of them that match for your RAID.
That all said, I'm not exactly sure how this is Ask Slashdot material -- even if it's not something Google easily spits out answers for. Since specifications are so decisively absent, and cost is a factor, there's no way in hell you're going to listen to any of our suggestions, as none of them will be nifty enough or cheap enough for whatever purposes you actually end up using the thing for.
I strongly suspecct that you're either lost trying to figure out what kind of horsepower you need for the software you haven't picked yet, or that you already have a good idea of what you want and are looking for some sort of devine Slashdot Affirmation of your unspoken decisions. But you didn't ask us for software advice, or moral support - you asked us about hardware.
Running a glorified bulletin board for a small household is not a difficult task for a computer -- I had hundreds of users dialing into my 10MHz XT a decade ago, and things were plenty fast. WWIV under DOS, FWIW. In other words, the slowest computer capable of running Linux is more than fast enough for your purposes.
Find yourself a nice 386SL notebook, and be happy. The one I have here draws less than 12 Watts at full tilt, and the hard drive spins up in less than 2 seconds. Powersaving features are built-in, and the box supports killing the power to unneeded accessories. I just put Slackware 8.1 on it tonight, and things are looking good with a $2 PCMCIA network card. I bet an old Tandy/Northgate/AST/Blue Dolphin/Honeywell/AT&T/whatever 386 would work just as well, with a slight power-efficiency disadvantage.
And if you think you need anything faster than a 7-year-old Pentium desktop with RAID or a 10-year-old notebook for your family to write notes to eachother not more than several times a day, call Dell and buy yourself a new Optiplex or Dimension or whatever it is that they're hawking these days.
Or, stop complicating life by making things so simple, and invest in a corkboard and some scrap paper, plus a few moments to consider a proper location in which to put them.
Listen up, kids: If anyone here is currently going through the recruitment paperwork dance, and your recruiter is not buying you food every time you see them, try to find a different recruiter, or go to a completely different post. Not only is he lying to you (it's -all- lies, after all), he's just being plain unfriendly. They get reimbursed for every expense they incur before you leave for training.
And don't let him make you think, even as he makes a point of letting you see him hand his last twenty dollar bill to the waiter, that it was ever his money to begin with.
Use them for all they're worth, because that's exactly what they're actively doing to you.
A good recruiter will offer to drive you to work every day if your car breaks down, and will be Johnny-on-the-spot when you, blind-drunk at a party, call his cell phone at 3:00AM and tell him you need a ride home.
A good recruiter is like a puppy that you can kick all day long, scold constantly, and never give any water, that never grows tired of being subservient. Until you leave for training, it's their job to keep you happy and out of jail.
Don't think for a second that you're hurting their feelings by accepting, or even asking for, these favors. They've all been in the military long enough that they're quite used to being fucked.
At some point, all signals meant for human consumption are analog. Your user number is low enough that you've almost certainly seen discussion of ultimately bypassing any music copy protection ever created by clipping wires to the speaker, or the headphones, or whatever.
It would not be a terrible burden for an experienced TV tech with an accurate schematic to add an unencrypted analog output.
It may be in some strange form of RGB, or HSV, or something else entirely, but it's do-able. The most non-trivial aspect of it would probably be the act of opening up the set.
And as long as it remains do-able (ie, forever), and there is sufficient demand, off-the-shelf devices will exist which are capable of converting this signal into whatever it is that is people find useful for recording HD.
Utopian? Perhaps. But region-hacked DVD players are commonplace now, as are Macrovision scrubbers, and SCMS strippers. Not even the DRM king could keep SP1 away from illicit XP users.
What makes you think that hollywood would have any better luck with any of this than they did with the oddly annoying joke known as CSS?
I'm not too worried about it. 'sides, why would you ever want to record uncompressed analog HD in the home? The bandwidth demands would be atrocious.
Re:why does it hurt when i do this...
on
Unionfs for Linux?
·
· Score: 3, Insightful
Witty. Constructive. +4.
I like it. Not.
If "there are lots of solutions" to the problem, showing at least one of them would perhaps turn outright discouragement into something useful.
The problems are obvious. And they have been solved for a long time - witness TVFS under OS/2. Does something similar exist for a free OS? LVM might be close.
Mounting filesystems should be no more nor less complicated than is needed to achieve the desired goal. If this means that it's "as complicated as firewall rules," so be it.
Seems that this is already the case, anyway. I've got three lines of firewall rules. The fstab on the same machine trounces that handily, without doing anything creative or silly - just mounting various partitions to various points.
(and before anyone asks how it can possibly be secure, I'll say this: It's -STABLE, and that's good enough for me.)
Your statement makes as much sense as the following:
"He didn't build the computer, he built the case for it."
A loudspeaker includes a number of components. A driver, such as the Shiva contraption, is one of them. A box is another.
He assembled these components together. He built a speaker.
Your next statement is similarly non-sensical. I'll convert it to an analog which you might understand:
"...he used the BP-6 motherboard, which I believe is only available seperately, although there are some places building and selling computers with BP-6 motherboards already in them."
Does this statement not disprove itself? Or is there some magic additive which must be included in a loudspeaker/white-box computer before it may be considered as a whole instead the sum of its parts?
Is it a pre-requisite that such finished products carry an endorsement from Apple or Sony before they are recognizably complete systems? Or perhaps some other subtle nuance I've managed to miss in my many years of building complete systems myself?
Please explain further. I fear that if you're correct, my speakers and my computers may suddenly disintegrate themselves into seperate parts.
An interesting a do-it-yourself project, but can you talk with elephants using it, let alone load it into an Isuzu Trooper?
That seems to have been the going standard for servo-driven loudspeakers for a decade or two.
I had the unique experience of sitting rather directly beside a (somewhat lesser, but similar) dual-cone 15" unit for an evening as I played DJ. I'll spare the details, but do allow me to say that visiting The Throne after returning home that night after having everything homogenized (shaken, not stirred) by the subwoofers was uniquely euphoric experience.
[I'd include more prominent links to Intersonics/Servo-Drive, who still sells these monsters. But their web page is "currently down for reconstruction." Dumb shits. Their old page was at least informative, not to mention existant. groups.google.com for more information, I guess...]
A driver (typically a coil, cone, suspension, and magnet) is not a subwoofer. An enclosure (typically a pile of lumber, some screws or expensive clamps, a bunch of glue) is also not a subwoofer.
Putting both of them together may make a "subwoofer", if things end up wired appropriately, but neither a box nor a driver is very useful by itself. It's like having a CPU without a motherboard or any other requisite component, and calling it a "computer".
Next time, try to posess even a modest understanding of the topic before flaming the submitter/editor/whoever.
Your mention that making a proper enclosure isn't easy is quite possibly the understatement of the century. The moving parts (drivers) of a completed loudspeaker assembly generally only account for ~10% of the final retail cost - and you're lucky if they spend even that much on it. The rest goes toward crossover parts (if needed), hardware like fancy gold-plated terminals, paying someone to solder it all together, and, mostly, a quality cabinet to put it all in.
The cabinet is primarily responsible for making the speakers at your local high-end shop sound better than those sold by your friendly neighborhood appliance whore. It is big, expensive to build, and expensive to ship.
The selection of design and materials is a very non-trivial exercise. Not surprisingly, the more difficult, time-consuming, and heavy designs tend to work better. More expensive materials (think Corian, granite, or good MDF, versus OSB) tend to sound better. Good engineering early on in the project can yield a shift toward zero for all of these variables, but everyone here knows that good engineering is similarly non-free.
And speaking of engineering, it takes a lot of it to build a good driver from scratch. Common practise is geared more toward modifying an existing sample to do more of what you want, instead of building new. The variables are numerous, the parts non-obvious, and the mathematical predictions for what-does-what lacking. But if you really want to know about it, a good place to start might be back-issues of Voice Coil Magazine.
Slashdot Mirror
scratch that. ATX3D should read ATA. Indeed, how AT commands fade from memory...
:-/
Now I feel like a senile old man, instead of just clever grown-up kid.
Naah, these were the days:
Me, a sysop in Ohio, trying to dial in to a friend's BBS in Florida, and finding it busy. Since I had S11 cranked down to 11 or 12ms (or whatever the fastest speed the phone company was capable of processing that week was), I was redialing like mad (God bless Supra for mailing me, -free-, firmware which supported such insane speeds), and with the busy detection also cranked down to just-barely-reliable periods, I was redialling like mad.
After a minute or two, instead of the usual touble-tap busy signal, half-second silence, and a rapid-fire redial, I heard prolonged silence and then a slight click.
Strange, I thought. I let the machine continue to redial.
After a few more minutes, the same thing happened again: no busy signal, no ringback, just dead silence.
I quickly exited to a terminal, picked up a phone, typed ATX3D, and hung the phone back up. I heard the modems negotiate and connect.
The guy in Florida, who thought he was calling a local BBS, was instead connected (at my expense) to my Telemate session..
We talked for a bit in Telemate's split-screen chat. He was obviously quite surprised to find me typing to him from Ohio when he thought he was calling a BBS across town. I told him how to fire up his WWIV BBS using the existing connection, without dropping carrier, and we talked for a bit more using WWIV's superior (and still un-matched) split-screen chat before I checked for new files and logged off.
Those were the days.
Probably, yes.
But at the time, the situation dictated that I either install a locally-available, department store 7200 RPM drive, or wait at least a year to upgrade the box. Given those choices, it was pretty clear to everyone that getting a 7200RPM drive right now was the obvious way to go.
I spent half a day in the car and on the phone looking for a reasonably large 5400RPM drive, to no avail.
Ambient temperatures of 100F aren't hazardous to most consumer electronics. In terms of durability: Given exclusion from severe catastrophic failure syndrome catalysts such as lightning strikes and 2-year-old children, anything in your cabinet which doesn't rely on moving parts is very likely to outlive you and a number of your descendants.
That said, neither the bearings in your DVD player nor the VHS tapes you play are likely to be happy, long-term, with such elevated temperature.
Additionally, the properties of the individual components (caps, resistors, transistors) change with temperature, so well-designed analog electronics are engineered with a specific temperature range in mind. They'll certainly sound best when operated at whatever ambient temperature they were designed for, which is likely to be at or slightly above room temperature (72F).
The thermal switch in your amp is not likely to trip until the heatsink is justabout hot enough to boil water. It exists as a safety feature, like a fuse, to turn things off under abusive situations or in catastrophic modes of failure. It is not, in any way, a device intended to ensure proper fidelity.
My parents have a similar situation at their house. They've got a 36" Sony CRT, on top of a glass-doored Sony stand. Inside this stand resides all of the extra components associated with the TV - a DVD player, VCR, and DirecTV TiVo.
After adding an 80-gig, 7200RPM Maxtor to the TiVo, things would get hot enough inside of the cabinet that the TiVo would lock hard every couple of days.
They simply removed the glass doors, and everything has been rock-solid stable since.
I recommend you do the same.
Not only will your components be more accessible, you won't need to worry about things being too hot. It's also free.
In my own living room, I solve the heat problem differently. I've got the line-level (minimal BTW/hr) stereo components stacked neatly on a shelf, the TV on its own seperate stand along with the PSX and DVD player, and a fan-cooled power amp in its own rack back in the far corner of the room, hidden behind a plush chair.
By spreading things out and avoiding confining furniture, heat becomes a non-issue. And I also get to keep the more dangerous components (the ones with volume controls, capable of producing dangerously-loud, eviction-level radio static) up out of reach of my 2-year-old daughter.
If none of these solutions are appealing, simply install a largish, slow-moving fan near the top of whatever cavity houses your AV components, exhausting air out the back. Maybe something like this would do the trick. If such an arrangment turns out to be too loud, wire a rheostat in series with it to slow it down even more.
You could also use a low-voltage DC fan, but it'd take all of the fun out of it and require the use of a seperate power supply.
Whichever the case, the purpose here is not to actively cool the components, but to simply provide a mechanism for exchanging the stale, warm air inside of a cabinet with cooler air from outside, be it by convection (avoidance of enclosed cabinets and glass doors) or force (a fan to push things around).
It won't take much.
That sounds very similar to an idea I was considering...
My variation is to attach a number of large IR LEDs around the front license plate and inside of the headlight reflectors, modulated so as to confuse laser speed measuring devices.
On a clear day this could provide some minimal added protection against speeding tickets.
> This is what smart chargers are for.
Clearly. And in the context of the article (did you even RTFA?), there is no "smart charger" to be had. This makes the point rather moot, don't you think?
And now for something completely off-topic:
Have you ever, even once in your life, agreed with someone in an upfront manner, or admitted that you were wrong?
Or are you one of those people who are always right, and will always state fault with what someone else is saying, even when there's a chance that you're not right and that the other person's statement is without flaw?
Just curious.
I never said NiMH batteries suffered from memory effect, just that they have issues with being cooked too long in the charger.
Try it sometime.
NiMH and NiCads each have very similar charging voltage and current, so it's quite possible to use NiMH batteries in a NiCad charger.
Here's what you'll need:
A set of NiCads, a set of similar NiMH, and two examples of the same type of (dumb, timerless, non-smart, plain-old-voltage-source) charger, such as that found on a cordless phone base.
Load one charger with NiCads, and the other with NiMH, and make the chargers charge for 12 or 14 hours - whatever seems typical for that battery capacity in that charger.
After this, measure the capacity of each set of batteries. A small flashlight and a stopwatch will work fine for this (but don't let things discharge completely - that's another type of abuse, altogether). Record your findings. The NiMH battery will probably last somewhat longer than its NiCad counterpart.
After that, repeat the experiment using the same batteries, except increase charging time to one week instead of half of a day. The NiMH batteries will lose substantial capacity during this time, which you'll see with your stopwatch. The NiCads will still have similar capacity to what they did during the first experiment. Record your findings, and realize this:
No technology is perfect. Repeat it over and over until it becomes clear. No technology is perfect. No technology is perfect. No technology is perfect. Once you get that to sink in, start with the following: Nothing is what it seems. Nothing is what it seems. Nothing is what it seems.
After a few months of this, perhaps you'll be able to emerge from Marketroid Consumerland and join the rest of us thinkers.
While it's true that NiMH batteries do not suffer the same "memory effect," per se, of NiCads, they are not without their own particular problems and quirks. Chief among these is how susceptible they are to damage from being charged too long. They also leak (self-deplete/drain/reach maximum entropy) faster than NiCads, discharging themselves relatively quickly while not connected to anything at all.
From the tone of your article, the phone itself sounds like it is relatively old.
Nicad batteries wear out. This is normal. They're also the cheapest and safest rechargable batteries you're likely to find, bar none.
So, just replace them. Find a well-stocked Panasonic phone dealer (amazon?) and get the genuine article, or just pick up the appropriate off-brand replacement at the department store of your choice.
While it certainly wouldn't be impossible to design a cordless phone that used rechargable lithium cells, noone seems to market such a beast to the consumer. Probably because lithium batteries are very good, very light-weight, and also very expensive. They also tend to react explosively to mistreatment during the charging cycle. Stay away from lithium rechargables in devices not expressly intended to use them. [FWIW, Motorola Startacs have a microprocessor in the battery pack itself, expressly for the purpose of lawsuit prevention - I imagine most/all other companies do the same thing, but I've never taken apart any other Li-Ion packs.]
You might be able to change the nicad cells with NiMH, but you're not likely to get any net benefit from the exercise. Besides, NiMH suffer severe stamina problems after being charged for too long at one time, which is something that chargers designed for this type account for, and your cordless phone does not.
So. With NiMH, you'll have to remove the cordless phone from the base after no more than 12 or so hours of charging, followed by a substantial drain before the next re-charge. Unless you're able to do this every time you charge it, and are able to train your friends and family to do the same, I'd avoid a NiMH retrofit, unless you anticipate and accept that permenant battery damage will result. [At least they don't tend to explode, like lithium batteries do...]
And at any rate, if you had the discipline to follow such strict procedures, you'd have plain old, cheap NiCads that last forever, just like me.
I've got a 4-year-old, vanilla, Uniden 900MHz analog phone. It works well, having great audio quality at long distances through any number of obstructions.
I replaced the stock battery after a year or so with whatever compatible type Sears had at the time, and things have been golden since.
Tricks:
I use the phone until it's either inoperably dead, or starts in with its low-battery song and dance (blinking light, funky stutter ring when someone calls). When this happens, I place it on the charger, and leave it there until at least the next day. When I feel like using it again, anywhere from six hours to six days later (depending on alcohol intake), I keep it off the charger until it is once again dead. The whole cycle takes from 5 days to just over a week, depending on use.
Those who are keeping score will note that I've got three years of outstanding service from a cheap Sears battery. It takes a slight effort to keep things in-check, but the system (as a whole) is reliable and predictable, which is really all that matters.
That all said, I echo your sentiment: I've got a Nokia cell phone with their smallest available lithium battery. I only ever use it with the cheap, third-party charger I have in the car, which cooks the battery completely in just over an hour. Routine fast charging like that is supposed to always be bad for batteries. But Lithium batteries seem to Just Work(tm) no matter what I do to them, and I still get great talk time and days worth of standby. Must be that microprocessor...
I'd love for a cordless phone manufacturer to start using modern battery systems. I'm even willing to pay extra for it the convenience of not having to bother with the proper care and feeding fo the battery.
Even my UPS, a Best Ferrups 850, maintains its sealed lead-acid battery (a Panasonic) automagically. Once every 1 or 2 weeks, it'll switch over to battery power for a few minutes at a time. It then tops off the battery again and resumes keeping it on float charge. This is all part of the normal care and feeding of a lead-acid battery (your car does it with its battery, too, by virtue of you running the starter motor), but most UPSs don't do it at all. OTOH, I suppose this is no ordinary UPS, but why can't phone manufacturers adopt at least semi-modern battery technology? NiMH would be a good, safe, and inexpensive start. Even my RioVolt came with a pair of AA NiMH batteries, and enough smarts to avoid abusing them.
Good luck!
I'll be the first to say that I don't know much about drilling, except that when I turn the faucet on at my parents' house, water comes out. Every now and then, the lights dim briefly when the pump turns on... And that's nowhere near the scale being discussed.
I don't have any paraffin or diesel here, but if you do, try this quick-and-dirty experiment to measure 2.4GHz RF absorbtion:
Put a container of it in the microwave. Does it get hot? If so, how does the temperature compare to an equal amount of water in a similar container, in the same spot of the oven, after being nuked for the same amount of time?
Therein lies a rough answer to your question, though it does a ignore number of possibly important factors, such as reflection and refraction of the signals and probably a slew of others that few people outside the realm of microwave engineering really know about.
As long as you realize that your findings are based on absorbtion alone, you'll be doing fine. There's probably a proper method to quantify it with, with a proper, capitalized Unit to go along with it, if you feel like being really anal about it.
Now that I'm thinking about it more:
If the results turn out to be negative (as in, no substantial heating occurs), it might be interesting to ponder using the metal pipe as a waveguide. Such a transmission system would be incredibly efficient, but would place constraints on the type of joint used, and the length of the pipe segments would have to be precise, corresponding to some factor of the wavelength of the signal in whatever medium ends up filling the pipe.
Bzzt.
High-frequency RF does not penetrate earth very well.
2.4GHz signals, in particular, are very trouble-prone in this application, as water converts it to heat more efficiently than any other frequency. Drilling is a very wet operation.
Hint: this is why your microwave operates at 2.4GHz, and why the band is unlicensed. Because it is so readily absorbed by masonry, trees, and other relatively wet objects, it was deemed (at least a few years ago) relatively unsuitable for serious communications and kept from being sold commercially since the beginning of time.
Have you never driven through a tunnel with the radio on, or while using a cell phone? FM radio is down near 100MHz, well into the range of relatively slow data transfer.
You need VLF radio to get through that much solid crap, and once you do that, you're back into the slothly realm of measuring things in bits per minute.
'sides, aiming a 1-megaWatt microwave oven down a drill pipe would not make their already-existing heat problems any better...
Uhhhh, some people get better cooling with the case OFF their computer. A lot of older cases (and many not-so-old designs) do not have ducts going to/from the hot spots. The case is seldom part of the nonexistant ducting system. Remove the case, and you HELP ambient air exchange, thus helping your ability to cool. So removing the case is a recommended MOD.
This isn't a shiney new aluminum Lian-Li, Coolermaster, or stealthly-quiet, duct-filled Dell. There's no Radeon here, no PC2700DDR, no dual-proc Athlon MP. It's an 8-year-old Infotel full-tower AT, with a 350MHz K6-2. With the top off, its trio of 7200RPM IBM Ultrastars, fan-cooled Plextor burner, and CPU all run cooler. The Voodoo3 overclocks better. The DDS-2 streamer acts healthier.
While I understand that you were just trying to be helpful, I'd like to assure you that your concern is unwarranted.
I'd also like to ask if you've got anything constructive to add to the topic at hand, which has everything to do with radiation and casemods, and nothing to do with cooling.
CRT monitors are affected more by low-frequency EMI than the RFI computers emit.
A simple test: Take the lid off of your case, and place it beside the monitor. Nothing strange happens.
Next, place an AC-operated fan, transformer-based soldering iron, or similar magnetic device next to the monitor, and watchen das blinkenrainbow.
That said, flat peices of steel (such as that which comprises your case) do very little to counteract low-frequency magnetism, while aluminum does absolutely nothing. Common steel can have some low-freuquency shielding effect if it's curved just so, but that's usually impractical. (there's other stuff, such as Mu-Metal, which is formulated with the specific goal of blocking EMI, and does work quite well. But it's expensive, and hard to find.)
I have to be careful where I put my Best FerrUPS because the large ferroresonant transformer in it will cause monitors to shake from several feet away.
Problems with computer-generated RFI generally show up with radio and television. I can't listen to an AM radio anywhere near my apartment with the PCs on, and there's a few FM stations that I can only recieve outside or in the back bedroom, away from the machines.
My neighbors must hate me for it, as I'm sure it's not much better anywhere in the building. But the 300-pound, heavy-footed woman upstairs has four kids who wake up at 5:30 AM daily, and the people directly beside me have a bad habit of listening to one-note bass lines with their lousy, one-note subwoofer, directly on the other side of the wall behind my desk.
So, I guess I care a lot less about RFI than I do about proper cooling. Thus, the top of the case is completely absent, allowing all kinds of natural, quiet convection cooling to take place.
I not only have a high-density 5.25" drive, but I've also got a sample of every tape drive I've ever owned.
I've still got my "important" data from 10 years ago living on CD, having recently been re-discovered (I didn't create much, back then, but the old MS-DOS utils directory has such oddities as pkarc and zoo which just aren't easy to find these days, along with a few hundred other fun things still applicable to my 386SL laptop).
I admit that I'm not very cautious with my choice of CD-R media. I generally choose the cheapest I can find locally (most recently: ~$17 for a spindle of 100 unbranded blanks), and my trustworthy 8x SCSI Plextor burner generally does a fine job of keeping the bits intact no matter what kind of trash I feed it. Time will tell as to whether or not this is a reliable way to archive things, but I'm betting that as soon as DVD-R media comes down to $1, I'll be jumping on that bandwagon instead.
Important stuff isn't very difficult to keep around, in my experience. In the early 90's, my father contracted an audiophile friend of mine to record some 1950-vintage wire recordings of his family to cassette. FWIW, the wire recorder itself was working JustFine, and required no repair or reconditioning to play these glorified spools of bailing wire with good fideliy.
A year or two ago, during the same family's Christmas gathering, I (on a whim) made audio CDs of this 50-year-old dialogue and distributed them. I used a 5-year-old Onkyo tape deck (which I'll probably never get rid of) and a well-designed Yamaha XG-based sound card. The transfer was, to my ears, perfect.
Does anyone make wire recorders anymore? Nope. Does anyone make good cassette decks anymore? Perhaps, but you'll have a hard time persuading me that it's not just old stock. 5.25" drives? I'll be honest: I never owned one, until a couple of years ago when I picked one up at a shop specializing in used PC parts. I cut my teeth on PCs in the late 80's, when 3.5" media was sure to be a success. (And what do you know - 3.5" drives are still standard on justabout every new non-Apple machine you can buy.)
So, no. I'm not worried about the shelf life of my CDs in the slightest.
Even color film isn't so good, either. I currently work in a department store photo lab, for lack of a better job. A couple of weeks ago, someone brought in a roll of 200-speed Kodak from 1984 (judging by the t-shirts worn, and the C-64 connected to a TV with knobs on it). The colors were abyssimal. Absolutely horrid. An artist's rendition of a technicolor nightmare. I was able to get a few good prints from the roll, but things were -really bad-, overall, with hideously-colored people ranging from magenta to olive drab on a frame-by-frame basis.
I expect my bargain-bin CD-Rs to fare at least as well as that, and hereby submit that CD-ROMs are the 35mm film of the digital age.
If it were me and I had the cash to drop, I'd go for a mini-ITX system, or one of Shuttle's prefab mini-boxen.
They really shouldn't require much time to set up - there won't be anything involved other than selection, ordering, and assembling of parts.
The laptop may actually represent a larger time-sink, unless it's got a vanilla CD-ROM drive built-in and fairly standard PCMCIA hardware, which would make it easy.
If you've got no removable mass storage for the CTX box, invest a couple of dollars in a 2.5" -> 3.5" IDE kit. This will let you use a desktop machine to do the base install of software.
Since storage is cheap and demands are few, you'll probably want to leave a copy of your Linux distro on the hard drive in its own partition, so you can get back to a bootable state without pulling the drive, in case Bad Things happen. Slackware still fits on one CD, FWIW - you won't give up too much space for it.
Good luck.
Do you really think so?
I've been downloading high-quality rips of entire albums lately, without difficulty, zipped or tar'd for convenience and completeness.
I've even donwloaded monolithic files of several hundred megabytes, containing entire artist discographies, with ease.
I've found these techniques to be far superior to trying to find good rips of individual songs.
I don't know about ogg, but MP3 is not at all tied to having things sampled at a specific bit-depth - it just doesn't matter to it whether it is 16, 20, or 24 bit - the format has sufficient dynamic range. I know of at least one free 24-bit MP3 decoder, but I've not run across anything for the encoding side just yet (and I haven't been looking, either).
'sides, HDCD essentially only adds 4 bits worth of dynamic range, for a total of 20, which isn't such a stretch for current consumer audio gear. The human ear is likely to be a greater limiting factor than the equipment.
And finally, there will never be any "exclusive content" on the HDCD layer, because there is no HDCD layer. The extra four bits are gleaned from unused space on a CD and interspersed with everything readable by a normal CD player. HDCD is an adjunct to a standard red book audio CD, not a replacement for it. Take away red book compatibility, and you not only remove 16 of HDCD's 20 bits, but eliminate compatibility with existing HDCD players (which are numerous indeed).
You're thinking of Sony's SACD, which can include multiple layers, and currently lacks ripping tools and PC-compatible hardware. I'm sure that Sony would love to convert the world to SACD and then stop adding a red book layer to their music as a cost-saving measure.
I doubt it will ever happen, however. Remember, this is the same company that brought you Beta, Minidisc, and Memory Stick. All of them are good, well-designed products, but lack of interoperability has always held them down, just as it will SACD.
HDCD, while developed by a high-end audio company whose name currently escapes me, is now 100% owned by (you guessed it!) Microsoft along with the aforementioned original developer.
It is also patented.
You can find an overview of how things work during the encoding process here, but don't expect anyone to tell you how to do any of those things without you first handing them vast fistfulls of cash.
You might be able to glean some useful information from the patent text, but probably the only sane way to go about this effort would be to read the bitstream coming from the CD. And while you should able to discern what the bitstream looks like without too much effort, it would probably be a fairly involved task to learn what it means.
So. My only suggestion would be to give up now before you've wasted any effort on trying. But if you insist on putting real time into this project, here's a couple of nice encapsulated postscript HDCD logos you can illegitimately use to adorn any illegitimate HDCD products you produce.
Why wait until "days later" for a webcast?
You can watch NASA TV live every day, for free.
Though you'll get -much- better quality from a nice C-band feed, at least you can get Realvideo of the event live.
Assuming their servers aren't full.
Which, especially after this posting, they most certainly will be...
Since nowhere in your post did you mention that it must be a fast machine, and your desired goals are very light-weight, just buy a cheap, slow Pentium machine from Ebay or a place like this.
Save energy how you can, if it's important to you. Toss the CPU fan, and keep the heatsink. GlobalWin makes some huge Socket7 heatsinks which are suitable for this, all of which come with easily-removed fans screwed to them.
I've got a P133 which has been running various incarnations of Windows (now 2k) for years, with only a quiet PSU fan and a modest 6.5gig Seagate drive which spends most of its life spun down. It's nearly silent, doesn't make much heat, and I don't even think about its power consumption.
Configuring hdparm/apmd/kernel/BIOS to put the system to sleep would be good. As long as you don't let it drop into suspend mode, it'll come right back to life as needed.
Avoid hardware that you don't need. Don't use a sound card, find a slow/old/efficient video card. Keep things simple.
If you're worried about the reliability of a used machine, don't be. Remember, only the moving parts are subject to wearing out in normal use. Of these moving parts, you'll be completely eliminating the CPU fan. You can buy a nice new Sunon or Panaflow fan to replace whatever comes with the PSU, either of which should last a long, long time (the last dead fan I replaced was a Sunon that I've had spinning for 8 years).
All that's left is the hard drive, and you'll probably want to buy a couple of new ones no matter what you do, anyway, so that you've got two of them that match for your RAID.
That all said, I'm not exactly sure how this is Ask Slashdot material -- even if it's not something Google easily spits out answers for. Since specifications are so decisively absent, and cost is a factor, there's no way in hell you're going to listen to any of our suggestions, as none of them will be nifty enough or cheap enough for whatever purposes you actually end up using the thing for.
I strongly suspecct that you're either lost trying to figure out what kind of horsepower you need for the software you haven't picked yet, or that you already have a good idea of what you want and are looking for some sort of devine Slashdot Affirmation of your unspoken decisions. But you didn't ask us for software advice, or moral support - you asked us about hardware.
Running a glorified bulletin board for a small household is not a difficult task for a computer -- I had hundreds of users dialing into my 10MHz XT a decade ago, and things were plenty fast. WWIV under DOS, FWIW. In other words, the slowest computer capable of running Linux is more than fast enough for your purposes.
Find yourself a nice 386SL notebook, and be happy. The one I have here draws less than 12 Watts at full tilt, and the hard drive spins up in less than 2 seconds. Powersaving features are built-in, and the box supports killing the power to unneeded accessories. I just put Slackware 8.1 on it tonight, and things are looking good with a $2 PCMCIA network card. I bet an old Tandy/Northgate/AST/Blue Dolphin/Honeywell/AT&T/whatever 386 would work just as well, with a slight power-efficiency disadvantage.
And if you think you need anything faster than a 7-year-old Pentium desktop with RAID or a 10-year-old notebook for your family to write notes to eachother not more than several times a day, call Dell and buy yourself a new Optiplex or Dimension or whatever it is that they're hawking these days.
Or, stop complicating life by making things so simple, and invest in a corkboard and some scrap paper, plus a few moments to consider a proper location in which to put them.
Listen up, kids: If anyone here is currently going through the recruitment paperwork dance, and your recruiter is not buying you food every time you see them, try to find a different recruiter, or go to a completely different post. Not only is he lying to you (it's -all- lies, after all), he's just being plain unfriendly. They get reimbursed for every expense they incur before you leave for training.
And don't let him make you think, even as he makes a point of letting you see him hand his last twenty dollar bill to the waiter, that it was ever his money to begin with.
Use them for all they're worth, because that's exactly what they're actively doing to you.
A good recruiter will offer to drive you to work every day if your car breaks down, and will be Johnny-on-the-spot when you, blind-drunk at a party, call his cell phone at 3:00AM and tell him you need a ride home.
A good recruiter is like a puppy that you can kick all day long, scold constantly, and never give any water, that never grows tired of being subservient. Until you leave for training, it's their job to keep you happy and out of jail.
Don't think for a second that you're hurting their feelings by accepting, or even asking for, these favors. They've all been in the military long enough that they're quite used to being fucked.
Harumph.
At some point, all signals meant for human consumption are analog. Your user number is low enough that you've almost certainly seen discussion of ultimately bypassing any music copy protection ever created by clipping wires to the speaker, or the headphones, or whatever.
It would not be a terrible burden for an experienced TV tech with an accurate schematic to add an unencrypted analog output.
It may be in some strange form of RGB, or HSV, or something else entirely, but it's do-able. The most non-trivial aspect of it would probably be the act of opening up the set.
And as long as it remains do-able (ie, forever), and there is sufficient demand, off-the-shelf devices will exist which are capable of converting this signal into whatever it is that is people find useful for recording HD.
Utopian? Perhaps. But region-hacked DVD players are commonplace now, as are Macrovision scrubbers, and SCMS strippers. Not even the DRM king could keep SP1 away from illicit XP users.
What makes you think that hollywood would have any better luck with any of this than they did with the oddly annoying joke known as CSS?
I'm not too worried about it. 'sides, why would you ever want to record uncompressed analog HD in the home? The bandwidth demands would be atrocious.
Witty. Constructive. +4.
I like it. Not.
If "there are lots of solutions" to the problem, showing at least one of them would perhaps turn outright discouragement into something useful.
The problems are obvious. And they have been solved for a long time - witness TVFS under OS/2. Does something similar exist for a free OS? LVM might be close.
Mounting filesystems should be no more nor less complicated than is needed to achieve the desired goal. If this means that it's "as complicated as firewall rules," so be it.
Seems that this is already the case, anyway. I've got three lines of firewall rules. The fstab on the same machine trounces that handily, without doing anything creative or silly - just mounting various partitions to various points.
(and before anyone asks how it can possibly be secure, I'll say this: It's -STABLE, and that's good enough for me.)
Your statement makes as much sense as the following:
"He didn't build the computer, he built the case for it."
A loudspeaker includes a number of components. A driver, such as the Shiva contraption, is one of them. A box is another.
He assembled these components together. He built a speaker.
Your next statement is similarly non-sensical. I'll convert it to an analog which you might understand:
"...he used the BP-6 motherboard, which I believe is only available seperately, although there are some places building and selling computers with BP-6 motherboards already in them."
Does this statement not disprove itself? Or is there some magic additive which must be included in a loudspeaker/white-box computer before it may be considered as a whole instead the sum of its parts?
Is it a pre-requisite that such finished products carry an endorsement from Apple or Sony before they are recognizably complete systems? Or perhaps some other subtle nuance I've managed to miss in my many years of building complete systems myself?
Please explain further. I fear that if you're correct, my speakers and my computers may suddenly disintegrate themselves into seperate parts.
Feh.
An interesting a do-it-yourself project, but can you talk with elephants using it, let alone load it into an Isuzu Trooper?
That seems to have been the going standard for servo-driven loudspeakers for a decade or two.
I had the unique experience of sitting rather directly beside a (somewhat lesser, but similar) dual-cone 15" unit for an evening as I played DJ. I'll spare the details, but do allow me to say that visiting The Throne after returning home that night after having everything homogenized (shaken, not stirred) by the subwoofers was uniquely euphoric experience.
[I'd include more prominent links to Intersonics/Servo-Drive, who still sells these monsters. But their web page is "currently down for reconstruction." Dumb shits. Their old page was at least informative, not to mention existant. groups.google.com for more information, I guess...]
A driver (typically a coil, cone, suspension, and magnet) is not a subwoofer. An enclosure (typically a pile of lumber, some screws or expensive clamps, a bunch of glue) is also not a subwoofer.
Putting both of them together may make a "subwoofer", if things end up wired appropriately, but neither a box nor a driver is very useful by itself. It's like having a CPU without a motherboard or any other requisite component, and calling it a "computer".
Next time, try to posess even a modest understanding of the topic before flaming the submitter/editor/whoever.
Your mention that making a proper enclosure isn't easy is quite possibly the understatement of the century. The moving parts (drivers) of a completed loudspeaker assembly generally only account for ~10% of the final retail cost - and you're lucky if they spend even that much on it. The rest goes toward crossover parts (if needed), hardware like fancy gold-plated terminals, paying someone to solder it all together, and, mostly, a quality cabinet to put it all in.
The cabinet is primarily responsible for making the speakers at your local high-end shop sound better than those sold by your friendly neighborhood appliance whore. It is big, expensive to build, and expensive to ship.
The selection of design and materials is a very non-trivial exercise. Not surprisingly, the more difficult, time-consuming, and heavy designs tend to work better. More expensive materials (think Corian, granite, or good MDF, versus OSB) tend to sound better. Good engineering early on in the project can yield a shift toward zero for all of these variables, but everyone here knows that good engineering is similarly non-free.
And speaking of engineering, it takes a lot of it to build a good driver from scratch. Common practise is geared more toward modifying an existing sample to do more of what you want, instead of building new. The variables are numerous, the parts non-obvious, and the mathematical predictions for what-does-what lacking. But if you really want to know about it, a good place to start might be back-issues of Voice Coil Magazine.