Most (all?) modern American cars have such governers. They're in place to prevent you from exceeding the speed rating of the OEM tires. The idea is to keep centrifugal force and heat from ripping the tires apart, which can be rather devestating at triple-digit speeds.
My 1995 Beretta kills the fuel supply at 112MPH, due to the S-rated Generals that were fitted at the factory.
Back on topic: the system is not smart enough to recognize that it now has H-rated (140MPH) rubber at all corners, thus requiring a payment of several hundreds of dollars to a company like Neuspeed for a custom ROM to eliminate the "feature". Which sucks, and is expensive, and complicated.
OTOH, I've never found a good reason to go any faster than that, although I do find myself creeping up on 110MPH with some frequency.
Is it illegal, then, to listen to other peoples' shouted conversations in public places? What if it's an interesting conversation, and you crane your neck in to hear more of it? What if it proves sufficiently enticing to you that you choose to join in with your own anecdotes?
Does it make a difference if the shouting consists of pressure waves in air (sound) or electromagnetic radiation (RF)?
There is not a more public slice of spectrum than the 2.4GHz band used by 802.11b.
As a user of a medium-range (several miles) 802.11b link, by which I'm accessing Slashdot right now, I'm acutely aware of this. Anything I don't want to say in public is treated, at least, with SSH before it hits the air.
Nobody has written a proper interactive HTTP download client that doesn't look and act like a web browser.
Try lftp. It's a featureful FTP client with a remarkably simple textual interface. Most, if not all, of its commands also work transparently with HTTP URLs. Things such as reget, bandwidth throttling, mirroring, and a few other fairly nice toys seem to work justfine for me with HTTP.
Perhaps I failed to emphasize the part about only one type of 74-minute Verbatim disc ever having compatibility trouble.
I use these discs everywhere. I hand them out to customers. I sell them. People like my the products I make on CD-R, no matter what equipment they're using.
Additionally, the link provided (http://www.digit-life.com/articles/cdrwtestidep12/index.html) is a comparison of an AOpen (read: Acer) and Plextor burner. The AOpen produced CDs with consistantly higher block error rates, and (somehow!) managed to write discs slower at 20x, than the Plextor at 16x.
I, henceforth, would like to iterate again my point about the quality of hardware.
While the instructions say to use a shortwave radio tuned to 10MHz, I found that a regular broadcast-band AM radio worked fine. Just chop a zero off of the frequency, and tune in somewhere around 1000. (1030 was what my tuner said, at the point where the "music" was most plainly heard).
The issue of CD-R compatibility has a great deal to do with the quality of the burner.
Three years ago, I picked up an 8x Plextor burner, and decided to run some tests in order to determine what kind of (then expensive) blanks I should be using.
I ordered 2 each of ~12 different makes of CD-R blank -- as many as I could find at one place at one time -- and tried them all. As burning at 8x was the latest-and-greatest thing at the time, none of the discs were "rated" for the speed except for a special, higher-dollar one from Kodak.
After burning identical audio tracks at 8x to half of the discs, I listened to them with a well-calibrated ear on a good mid-fi stereo.
They all worked fine. No ticking, no artificial harshness, nadda. The twelve of them sounded exactly the same.
This was frustrating, because I still had no idea what brand to buy - so I downgraded the CD player from a Carver TL-3300 to a friend's borrowed (and quite abused and tired) Playstation.
Some variations finally showed. Surprisingly, the 8x-rated Kodak disc worked least well - slow seek times, and strange behavior. I don't remember the rest of it that battery, unfortunately.
The next test involved ripping the CDs with cdparanoia, and timing how long it took for a good rip, on a 32x Plextor reader. All discs ripped without reported error, which was good - the Kodak was slowest of the bunch, and TDK the fastest.
Interestingly, the burner came packaged with a TDK blank as a gimme.
But what makes this interesting is that some of the discs were not rated for speeds anywhere near 8x - IIRC, the Maxell blanks said 2x on the package. They all passed. Even the Kodak, though strange in some way, was acceptable (and guaranteed to be readable for 100 years or somesuch).
Since then, I've not done anything as exhaustive. But I have had 74-minute Verbatim discs fail to read -at all- in any Playstation, while the box of 50 80-minute Verbatims (with, appearently, the same dark blue dye formulation) work well with everything.
At the studio, I use the cheapest, no-name unbranded shiney CDs I can get, with an identical Plextor burner. I've never had any customers say anything to me about an unreadable burn, and I never burn at 4x, 2x, or 1x - why waste the time?
Meanwhile, my sister has a 4x Phillips CD-RW, which always produces skip-prone CDs, that occasioally tick and pop - no matter what media is used.
To conclude the sermon, I'd like to point out that Plextor makes the most expensive burners available to consumers, and that by most (if not all) accounts, they're worth every cent. For a more lengthy and detailed analysis, on more modern hardware, see: http://www.digit-life.com/articles/cdrwtestidep12/ index.html
Further, so long as I'm ranting, the color of the
dye means absolutely nothing, except to visually compare two or more CDs. Even if CD players used some binary reflective/nonreflective technique to read a disc, which they do not, they use infrared light to do their business. Which is to say that unless you're wearing IR goggles, you have no idea what the dye looks like to a CD player.
All that happens (or, rather, needs to happen) when a CD is burned is that it makes a longer path for the light of the reading laser wherever a pit should be, by putting a hole in the dye layer below the reflective metal (which, FYI, was 24k gold on the Kodak disc, and silver (not aluminum) on the TDK).
You might've noticed that mass-produced, stamped/injection-molded CDs have no visible contrast (or IR contrast, for that matter), and all of -those- work justfine.
At a bit before 4AM in Ohio, I saw several meteors within about a minute and a half - with unadjusted eyes, and a well-lit garage behind me.
More interesting to me than the fireballs in the sky, however, is their (apparent) effect on FM broadcast radio.
Since around 8PM, I haven't been able to listen to local radio stations with any clarity - they're all being stepped on by something else sharing the frequency. I can, however, get good reception of a few stations that typically require driving at least 50 miles away to recieve at all.
Reflections from outer space, or strange atmospheric stuff? It -is- a lot foggy tonight...
After 9/11 I realised that (a) it is possible that this war will see no end for a long time (b) we'll need help.
So. In a few months, I'll be a network admin for the US Army. It's a nice ride.
If you think your boys might eventually be drafted (which isn't terribly likely to happen), the last thing you want to do is keep them from the skills they'll need to stay alive. And, meanwhile, have fun with them.
An out of sight, out of mind mentality doesn't work very well against something as blind as the draft.
But is just adding boxes not cheaper, more flexible, and less prone to catastrophic failure?
By my reckoning, half-width 1U rackmount PCs stack more densely than mid-tower cases each with a half-dozen cards in it. Overall reliability increases, as well: you don't need to rely on a single motherboard and power supply to keep things going.
Putting a bunch of rackmount PCs into a single (portable, or not) box is also a very trivial exercise.
Is realestate at the desk really at such a premium that anyone need care about these things?
I've been using LAME 3.89 for awhile, with the following command line:
lame -r -o -k -v -V0 -x -b128 -mj -h -q1
After much figuring and careful listening, I've found this to work well enough that I never hear the difference between the encoded file and its original wav. Average bitrate works out to 190-220kbps, depending on source material, which seems quite acceptable in terms of disk use.
Notice the lack of an upper limit on VBR bitrate on the command line; if you're going through all the trouble of doing VBR, why bother with arbitrary limits? LAME's VBR makes very minimal use of the bit resevoir - the same feature which allows it to work as well as it does on CBR encodings. Don't set limits. It's mathematically shown that you'll hear the difference, and you won't care notice the slightly increased filesize of having a few 320kbps frames - especially on a 100GB drive.
Equipment is as follows:
Recent ALSA
CMI-based sound card from Zoltrix
Audio Alchemy DDE 1.1, fed with TOSLINK from the CMI card
Rotel RTC-940AX preamp
Ashly FTX-1001 power amp
Midrange Sony headphones (which I don't have in front of me)
Speakers of my own build (parts from Vifa and Madisound, simple crossovers, -good- cabinets)
The magic here on the hardware end is the CMI card, which doesn't resample its digital output to 48KHz, like most other consumer devices (SBLive, Yamaha XG come to mind) - thus, it is bit-perfect. Being samplerate-locked causes a bit of trouble on non-44.1 files, as ALSA doesn't have good software resampling, but it's a CD player replacement - not a gaming machine - and for that, it works perfectly.
You seem pleased with the battery life of your Sony telephone - a 7x improvement in use over the originals that were new a couple of decades ago..
I can't imagine being pleased with a watch that would last for 3.5 days between charges. I like to strap them onto my arm, and forget about them until I need to do something with it.
Dealing with the battery every few years is bad enough for something as easily depended upon as a wristwatch, let alone having to do it twice weekly .
...Nevermind that we'd need to wait 20 years to get such impressive stamina out of the thing.
A large portion of eyestrain at a computer is probably caused by poorly configured monitors.
It's easy to set them up in a more-or-less correct fashion, without such gizmos as color analyzers (though they certainly make things easy).
Some of this is might seem trivial and obvious. For some reason, however, most people still don't get it right. The following is a quick-and-dirty guide to setting up all manner of CRT displays.
First, get the monitor's geometry right. You'll need a solid image for this - almost anything will do, as long as it stretches to all edges of the display. A solid-color desktop is fine. Tweak the size and pincushion and slant and rotation and whatever else you've got, until the displayed image stretches as far to the edges as possible without going over. After that, work on making it square, from -your- perspective, as you normally use it. Some, most, or all of these adjustments will interact with eachother, so you may have to revisit each several times to get things right.
Now that your display is as big as possible (always a good thing), it's time to start on the finer points of eyestrain reduction.
Find a screen that is mostly (or preferably all) black, and use this to adjust black level using the brightness control. Start by cranking it up until everything turns grey, and then back it down until that grey is invisible under ambient light, and then reduce it another step or so. The idea is that black should -always- be as black as possible under your typical lighting, but not so dark that shades of grey also become invisible.
After that, adjust the white level, otherwise known as "Contrast" or "Picture." Find or make a large (full-screen is nice) greyscale gradient, with consistant steps from white to black. This might be easier with a palletized 8-bit image, so that there's fewer levels of grey for your eye to discern. Turn up the contrast until the white end of this image loses its definition, and then turn it down until you see even steps of grey. No two strips should looks any closer to eachother in shade than any other two, ideally.
Double-check that your black level adjustment is sane, as changing the contrast will often interact with that. *sigh*
At this point, your monitor will probably look much darker than usual. It also is not overdriven anymore. This is not a coincidence.:) Your monitor will last markedly longer, and your eyes will be appreciative.
The darkness is due to a CRT and the associated electronics are only capable of producing a certain amount of light. Ask it to do more of this, and it won't. This has the effect of pushing midtones up in luminosity, while flattening out the top end. This is not good for the electronics, or your eyes, as an overdriven white will not appear as sharply focused as one which is within the display's range of linearity.
Your display is now vastly more accurate than most others.
So. It is now time to adjust gamma, which will bring up midtones without overdriving things. There is, unlike NTSC, no standard gamma for PC displays, and thus there is no right or wrong way to do it.:-/ I like to set mine up so that it has a final gamma of 1, which is to say that if the video card produces 50% of maximum voltage, the monitor shows 50% of maximum brightness. As a bonus, this matches the characteristics of my printer, so things really are WYSIWYG.
It's easy to adjust gamma, provided that your video card and OS support it (xgamma under xfree86). Do a quick Google search for a gamma correction image - what you want is something with a checkerboard of small white and black squares, and a 50% grey box for comparison. Turn the gamma up until the checkerboard pattern matches the brightness of the box, and leave it there. You might find it helpful, depending on the image you use, to squint slightly and blur the checkerboard into something solid while adjusting this. Fortunately, since this happens digitally, there's no impact on any other adjustments, and you're done with it.
Your display should now appear glorious.
Next, color temperature. This is really the first step, but since most people -never- care about it, I put it last. Which doesn't make it unimportant, just neglected and largely unknown. An explanation is thus in order.
Color temperature represents the color of white. You've noticed that a white shirt appears differently under incandescent lights, or on a cloudy day, or in stark sunlight, or under harsh flourescents, or at a football game under metal halides -- even though, in all instances, the white shirt is still obviously white. This is color temperature. Metal halides, as used in stadiums, often have color temperature in excess of 15,000 Kelvin, which is largely blue. Incandescents are often as low as 4300k, which is yellow or brown (depending on your interpretation). The higher the number, the more blue it is. Lower, is more yellow/brown.
Like gamma, and also unlike NTSC (standard of 6500k) there's no standard color temperature for computer displays. I like my monitor at ~7000k, which to my eyes, looks like the sunlight that comes through my NW Ohio window as it appears on a sheet of paper.
Most monitors (and TVs alike, dispite the known, functional standard) have ghastly high color temperatures from the factory. This, as far as anyone can tell, is because of some human tendancy to think that a bluer picture is brighter, and thus better. Sony displays are notoriously blue out of the box.
I haven't seen a monitor sold in the past couple of years which did not allow adjustment of color temperature, and I also haven't seen one at all which has temperature presets which match reality. Ignore the presets.
So, first, you'll need to find a reference. A Kodak grey card and a lamp of known temperature would be nice. A color analyzer would be better, if one can be borrowed. Or, just use a sheet of plain white paper (not coated stock for inkjets) and ambient sunlight.
Use a full-screen image that's just plain white, or grey. You'll then tweak the individual gain of the red, green, and blue guns on your monitor until the displayed image matches the color (-not- brightness) of your reference. Alternate between white and grey images until you get as close a match as you can on each. Then (as this is the first step), go through and do all the other things mentioned above.
Done.
It will look darker again, but in reality, the light output hasn't been changed much by the temperature adjustment, but perception brightness has. Let it sit that way for a few days, and you'll soon look at other monitors with disdain.
Load up some pr0n, or fire up your favorite black-on-white application, and notice how much easier things are to read and discern. You'll see details in photographs that you positively couldn't before, because your monitor couldn't show them to you as it was stuck in "buy-me" mode, like a stereo turned up to 11.
If, after this rough calibration, you find that your monitor is overcome with glare, the obvious solution is to reduce ambient light. Don't try to turn up the monitor, because it can't do it. If you can't reduce the amount of light in the room, buy a new monitor. The one you have presently is aging, as vacuum tubes (particularly CRTs) tend to do, and it can't produce as much light as it used to be able to.
I was reminded of the old PCs just last night, when trying to configure LILO on a new PII-based Linux box.
On the first reboot, it came up, POSTed, and acted normal. Instead of LILO, what instantaneously appeared on the CRT was a message saying "CANNOT FIND ROM BASIC - SYSTEM HALTED".
It was beautiful, in a 40-column white-on-black font that I hadn't seen used in years.
I stared at it in awe for several minutes before rebooting to see it again. =)
FWIW, xmp claims to support hardware reverb and chorus on AWE cards under *nix, using OSS's/dev/sequencer.
I haven't yet tried that end of it. A couple of weeks ago, I decided to start collecting mods (and relatives) again, and I'm just too thrilled at the way they sound when played with xmp through a good external DAC to bother making an AWE32 work. Last time I heard most of this music, I was using an SB 1.5 on a slow 386SX. The difference is striking.
OTOH, the AWE32 cards do have an SP/DIF output straight from the Ensoniq DSP... =)
Perhaps my view is somewhat skewed. I've had limited experience with DSL, and I haven't read the RFCs. If anyone with more of a clue cares to correct my notions, I invite them to do so.
In my area, Ameritech offers two kinds of end-user equipment. One is a DSL "router" which uses DHCP, performs NAT and includes a 4-port 10mbps ethernet hub, from Efficient. The other is a DSL "modem" which talks PPPOE to the customer's PC, and is made with the Westell mark.
The router works just as one might expect. It has a reasonably complete command line interface, allowing a fair amount of creativity with forwarded ports and other simple router tricks. I do not know what is involved with its initial setup, as Ameritech does this themselves. I also do not know if it is possible to use it without any NAT at all. And I've got no idea what the transport layer back to the DSLAM consists of.
And I just don't care. It works well.
The DSL "modem" is just like any other modem. POTS into one jack, serial into another. Except in this case, that serial data is encapsulated into ethernet frames. So what?
It talks PPP, quite obviously. I'm using the Roaring Penguin software, with superb reliability. On 160/768 ADSL, I see a few percent of CPU usage on an absolutely horrible Cyrix MII firewall/NAT box under one of the 2.2 linux kernels. If such a meager amount of CPU time can't be spared (and, according to some of you, it's impossible), I guess you've already got your priorities in line and need a router of some form to offload the task. Nevermind that the cute perl script you whacked together to paste together fragments of porn from usenet, the cause of all the bandwidth usage, is already eating more CPU than that.
This latter arrangement also sports a dynamic IP address. Who cares? It was trivial to have the address updated automagically on one of the dynamic DNS services. It's also at the whim of the ISP - I'm sure that I could get a static IP, if I wanted one. It's certainly not an issue of the capabilities of PPP, but more of a social thing. I've had a static IP PPP dialup nailed up at home for years, without an ounce of trouble. If the connection drops, it reconnects. Things then continue where they left off, like it never happened.
PPP also supports routing entire networks, or subnetworks, or whatever. If you want a/29 or somesuch, have your ISP provision you one. If they won't, you again have a social problem, not a technological problem.
Which is to say that PPPOE is just as capable as anything else DSL, given some sort of router. Is the need for a router at home something new to this world of supposed geeks? Stop whining and fire up ipchains, natd, ipfwadm, or a pretty box that says Linksys, D-Link, Netgear, ZyXEL, or some other such name.
People complain that PPPOE links aren't always-on. Of course they aren't. NOTHING in this world is.
Deal with it. The PPPOE-connected linux firewall does justfine handling such things as bringing the connection back in the event of it dropping. It takes seconds. It happens infrequently enough that I don't notice unless I'm looking at logs.
This stuff works over a pair of amazingly thin copper wires, strung crazily around the streets at the whim of city leaders, and operated by the telephone company. Nevermind that the design specification for these wires is 300-3,000Hz, and that DSL in any incarnation is an ugly fucking hack to begin with. It's amazing it works at all. And you want perfection?
This change is so fucking insubstantial to the way things work in practise that it's absolutely laughable to see so many people upset about it. When's the last time you had a network problem, and found PPP (over -any- medium) to be at fault?
Get a life. Stop whining. Even if PPPOE -does- incur a real performance hit, you'll never notice. And if you do notice, and still care, and still feel like whining about insignificant things, find something else. Vote with your wallet. Show those money-grubbing assholes as SBC just how you want to be treated.
Resampling digital audio always has detrimental effects. Doing it twice (44.1 to 48 and back) makes these issues worse. Three times (44.1 to 48 to 44.1 and 48 again on playback through an SBLive!) is obviously quite a bit of comb-filtering bit-munging.
Better to just buy a card with a non-resampling assortment of SP/DIF I/O, such as the plethora of "pro" cards from Lexicon, M-Audio and the like.
Or, a $30 Zoltrix Nightingale (or about any other card [including some motherboards] based on the CMI8738 chip) will do the trick nicely with coax or toslink. Also works well as a hardware format converter, and an SCMS stripper.
While I'm on the subject, the error correction of a CD player takes place well before the bits reach the digital output.
While I'm on the subject, it occurs to me that such things as SafeAudio lend a hand toward legitimizing filesharing services. "Well, your Honour, I didn't have any way to utilize Fair Use and use this CD in the MP3 player that came with my new Mazda, so I downloaded the files from someone else who was able to figure it out."
Check local surplus houses. Mendelson Electronics (www.meci.com) is near enough to me that I can swing down in an afternoon, and pick up rack of almost any size for less than $100-150. Some of these are nice, half-sized racks with glass doors. They say things like "Compaq" on them.
Buy a road case from someone like Anvil, Starcase, or SKB. They're durable. You can take them with you to a LAN party. And your equipment will be safe in transit, or as luggage on a 747, or whatever.
Alternatively, build a rack. Parts for serious road cases and other racks (aluminum extrusions, hasps, heavy steel corners, plywood with colorful vacuum-laminated fiberglass, pre-tapped rails, etc) can be found at TCH (the URL escapes me). They seem to be the same to road cases as Black Box is to networking gear.
Rails are also available at Parts Express (www.partsexpress.com), for cheap.
Whatever you do, be sure to compare the depth of the equipment to the depth of the rack. It might not be a big deal if the back of your server hangs out of the cabinet a few inches, unless it bothers you to look at. But it would be somewhat troublesome if the back cover for your new road case doesn't fit once it's loaded with equipment.
Companies like Starcase and Anvil are completely willing to build custom projects, so if you -really- want something special...
Slashdot Mirror
Most (all?) modern American cars have such governers. They're in place to prevent you from exceeding the speed rating of the OEM tires. The idea is to keep centrifugal force and heat from ripping the tires apart, which can be rather devestating at triple-digit speeds.
My 1995 Beretta kills the fuel supply at 112MPH, due to the S-rated Generals that were fitted at the factory.
Back on topic: the system is not smart enough to recognize that it now has H-rated (140MPH) rubber at all corners, thus requiring a payment of several hundreds of dollars to a company like Neuspeed for a custom ROM to eliminate the "feature". Which sucks, and is expensive, and complicated.
OTOH, I've never found a good reason to go any faster than that, although I do find myself creeping up on 110MPH with some frequency.
So.
Is it illegal, then, to listen to other peoples' shouted conversations in public places? What if it's an interesting conversation, and you crane your neck in to hear more of it? What if it proves sufficiently enticing to you that you choose to join in with your own anecdotes?
Does it make a difference if the shouting consists of pressure waves in air (sound) or electromagnetic radiation (RF)?
There is not a more public slice of spectrum than the 2.4GHz band used by 802.11b.
As a user of a medium-range (several miles) 802.11b link, by which I'm accessing Slashdot right now, I'm acutely aware of this. Anything I don't want to say in public is treated, at least, with SSH before it hits the air.
I think I like this picture better.
-
Try lftp. It's a featureful FTP client with a remarkably simple textual interface. Most, if not all, of its commands also work transparently with HTTP URLs. Things such as reget, bandwidth throttling, mirroring, and a few other fairly nice toys seem to work justfine for me with HTTP.
-
Perhaps I failed to emphasize the part about only one type of 74-minute Verbatim disc ever having compatibility trouble.
2 /index.html) is a comparison of an AOpen (read: Acer) and Plextor burner. The AOpen produced CDs with consistantly higher block error rates, and (somehow!) managed to write discs slower at 20x, than the Plextor at 16x.
I use these discs everywhere. I hand them out to customers. I sell them. People like my the products I make on CD-R, no matter what equipment they're using.
Additionally, the link provided (http://www.digit-life.com/articles/cdrwtestidep1
I, henceforth, would like to iterate again my point about the quality of hardware.
-
While the instructions say to use a shortwave radio tuned to 10MHz, I found that a regular broadcast-band AM radio worked fine. Just chop a zero off of the frequency, and tune in somewhere around 1000. (1030 was what my tuner said, at the point where the "music" was most plainly heard).
Spooky stuff, this.
The issue of CD-R compatibility has a great deal to do with the quality of the burner.
/ index.html
/injection-molded CDs have no visible contrast (or IR contrast, for that matter), and all of -those- work justfine.
Three years ago, I picked up an 8x Plextor burner, and decided to run some tests in order to determine what kind of (then expensive) blanks I should be using.
I ordered 2 each of ~12 different makes of CD-R blank -- as many as I could find at one place at one time -- and tried them all. As burning at 8x was the latest-and-greatest thing at the time, none of the discs were "rated" for the speed except for a special, higher-dollar one from Kodak.
After burning identical audio tracks at 8x to half of the discs, I listened to them with a well-calibrated ear on a good mid-fi stereo.
They all worked fine. No ticking, no artificial harshness, nadda. The twelve of them sounded exactly the same.
This was frustrating, because I still had no idea what brand to buy - so I downgraded the CD player from a Carver TL-3300 to a friend's borrowed (and quite abused and tired) Playstation.
Some variations finally showed. Surprisingly, the 8x-rated Kodak disc worked least well - slow seek times, and strange behavior. I don't remember the rest of it that battery, unfortunately.
The next test involved ripping the CDs with cdparanoia, and timing how long it took for a good rip, on a 32x Plextor reader. All discs ripped without reported error, which was good - the Kodak was slowest of the bunch, and TDK the fastest.
Interestingly, the burner came packaged with a TDK blank as a gimme.
But what makes this interesting is that some of the discs were not rated for speeds anywhere near 8x - IIRC, the Maxell blanks said 2x on the package. They all passed. Even the Kodak, though strange in some way, was acceptable (and guaranteed to be readable for 100 years or somesuch).
Since then, I've not done anything as exhaustive. But I have had 74-minute Verbatim discs fail to read -at all- in any Playstation, while the box of 50 80-minute Verbatims (with, appearently, the same dark blue dye formulation) work well with everything.
At the studio, I use the cheapest, no-name unbranded shiney CDs I can get, with an identical Plextor burner. I've never had any customers say anything to me about an unreadable burn, and I never burn at 4x, 2x, or 1x - why waste the time?
Meanwhile, my sister has a 4x Phillips CD-RW, which always produces skip-prone CDs, that occasioally tick and pop - no matter what media is used.
To conclude the sermon, I'd like to point out that Plextor makes the most expensive burners available to consumers, and that by most (if not all) accounts, they're worth every cent. For a more lengthy and detailed analysis, on more modern hardware, see: http://www.digit-life.com/articles/cdrwtestidep12
Further, so long as I'm ranting, the color of the
dye means absolutely nothing, except to visually compare two or more CDs. Even if CD players used some binary reflective/nonreflective technique to read a disc, which they do not, they use infrared light to do their business. Which is to say that unless you're wearing IR goggles, you have no idea what the dye looks like to a CD player.
All that happens (or, rather, needs to happen) when a CD is burned is that it makes a longer path for the light of the reading laser wherever a pit should be, by putting a hole in the dye layer below the reflective metal (which, FYI, was 24k gold on the Kodak disc, and silver (not aluminum) on the TDK).
You might've noticed that mass-produced, stamped
-
That explains why cows in India are worshipped, instead of eaten. They don't have enough gravity to burn off the calories of beef.
-
Are there any resellers other than Tux Games who are going to carry the linux version of the title?
I do like a bit of choice of who (and how much) I pay.
Can I expect to see it on the shelf at Software Etc., or will it be strictly a special-order item?
-
At a bit before 4AM in Ohio, I saw several meteors within about a minute and a half - with unadjusted eyes, and a well-lit garage behind me.
More interesting to me than the fireballs in the sky, however, is their (apparent) effect on FM broadcast radio.
Since around 8PM, I haven't been able to listen to local radio stations with any clarity - they're all being stepped on by something else sharing the frequency. I can, however, get good reception of a few stations that typically require driving at least 50 miles away to recieve at all.
Reflections from outer space, or strange atmospheric stuff? It -is- a lot foggy tonight...
After 9/11 I realised that (a) it is possible that this war will see no end for a long time (b) we'll need help.
So. In a few months, I'll be a network admin for the US Army. It's a nice ride.
If you think your boys might eventually be drafted (which isn't terribly likely to happen), the last thing you want to do is keep them from the skills they'll need to stay alive. And, meanwhile, have fun with them.
An out of sight, out of mind mentality doesn't work very well against something as blind as the draft.
But is just adding boxes not cheaper, more flexible, and less prone to catastrophic failure?
By my reckoning, half-width 1U rackmount PCs stack more densely than mid-tower cases each with a half-dozen cards in it. Overall reliability increases, as well: you don't need to rely on a single motherboard and power supply to keep things going.
Putting a bunch of rackmount PCs into a single (portable, or not) box is also a very trivial exercise.
Is realestate at the desk really at such a premium that anyone need care about these things?
Stuff like this started in the PC world, IIRC, with 386s on 16-bit ISA cards.
Nobody cared then.
Why would anyone care now?
Please explain your point using no more than 100 words.
-
Anyone feeling brave enough to actually try it, now that they're dumping them as surplus?
I've been using LAME 3.89 for awhile, with the following command line:
lame -r -o -k -v -V0 -x -b128 -mj -h -q1
After much figuring and careful listening, I've found this to work well enough that I never hear the difference between the encoded file and its original wav. Average bitrate works out to 190-220kbps, depending on source material, which seems quite acceptable in terms of disk use.
Notice the lack of an upper limit on VBR bitrate on the command line; if you're going through all the trouble of doing VBR, why bother with arbitrary limits? LAME's VBR makes very minimal use of the bit resevoir - the same feature which allows it to work as well as it does on CBR encodings. Don't set limits. It's mathematically shown that you'll hear the difference, and you won't care notice the slightly increased filesize of having a few 320kbps frames - especially on a 100GB drive.
Equipment is as follows:
Recent ALSA
CMI-based sound card from Zoltrix
Audio Alchemy DDE 1.1, fed with TOSLINK from the CMI card
Rotel RTC-940AX preamp
Ashly FTX-1001 power amp
Midrange Sony headphones (which I don't have in front of me)
Speakers of my own build (parts from Vifa and Madisound, simple crossovers, -good- cabinets)
The magic here on the hardware end is the CMI card, which doesn't resample its digital output to 48KHz, like most other consumer devices (SBLive, Yamaha XG come to mind) - thus, it is bit-perfect. Being samplerate-locked causes a bit of trouble on non-44.1 files, as ALSA doesn't have good software resampling, but it's a CD player replacement - not a gaming machine - and for that, it works perfectly.
You seem pleased with the battery life of your Sony telephone - a 7x improvement in use over the originals that were new a couple of decades ago..
I can't imagine being pleased with a watch that would last for 3.5 days between charges. I like to strap them onto my arm, and forget about them until I need to do something with it.
Dealing with the battery every few years is bad enough for something as easily depended upon as a wristwatch, let alone having to do it twice weekly .
...Nevermind that we'd need to wait 20 years to get such impressive stamina out of the thing.
You mean like this switch in ALSA, on my $12 Zoltrix Nightingale sound card with SP/DIF IO:
switch("SPDIF Copyright", false)
Just a thought...
I believe they call these access points .
A large portion of eyestrain at a computer is probably caused by poorly configured monitors.
:) Your monitor will last markedly longer, and your eyes will be appreciative.
:-/ I like to set mine up so that it has a final gamma of 1, which is to say that if the video card produces 50% of maximum voltage, the monitor shows 50% of maximum brightness. As a bonus, this matches the characteristics of my printer, so things really are WYSIWYG.
It's easy to set them up in a more-or-less correct fashion, without such gizmos as color analyzers (though they certainly make things easy).
Some of this is might seem trivial and obvious. For some reason, however, most people still don't get it right. The following is a quick-and-dirty guide to setting up all manner of CRT displays.
First, get the monitor's geometry right. You'll need a solid image for this - almost anything will do, as long as it stretches to all edges of the display. A solid-color desktop is fine. Tweak the size and pincushion and slant and rotation and whatever else you've got, until the displayed image stretches as far to the edges as possible without going over. After that, work on making it square, from -your- perspective, as you normally use it. Some, most, or all of these adjustments will interact with eachother, so you may have to revisit each several times to get things right.
Now that your display is as big as possible (always a good thing), it's time to start on the finer points of eyestrain reduction.
Find a screen that is mostly (or preferably all) black, and use this to adjust black level using the brightness control. Start by cranking it up until everything turns grey, and then back it down until that grey is invisible under ambient light, and then reduce it another step or so. The idea is that black should -always- be as black as possible under your typical lighting, but not so dark that shades of grey also become invisible.
After that, adjust the white level, otherwise known as "Contrast" or "Picture." Find or make a large (full-screen is nice) greyscale gradient, with consistant steps from white to black. This might be easier with a palletized 8-bit image, so that there's fewer levels of grey for your eye to discern. Turn up the contrast until the white end of this image loses its definition, and then turn it down until you see even steps of grey. No two strips should looks any closer to eachother in shade than any other two, ideally.
Double-check that your black level adjustment is sane, as changing the contrast will often interact with that. *sigh*
At this point, your monitor will probably look much darker than usual. It also is not overdriven anymore. This is not a coincidence.
The darkness is due to a CRT and the associated electronics are only capable of producing a certain amount of light. Ask it to do more of this, and it won't. This has the effect of pushing midtones up in luminosity, while flattening out the top end. This is not good for the electronics, or your eyes, as an overdriven white will not appear as sharply focused as one which is within the display's range of linearity.
Your display is now vastly more accurate than most others.
So. It is now time to adjust gamma, which will bring up midtones without overdriving things. There is, unlike NTSC, no standard gamma for PC displays, and thus there is no right or wrong way to do it.
It's easy to adjust gamma, provided that your video card and OS support it (xgamma under xfree86). Do a quick Google search for a gamma correction image - what you want is something with a checkerboard of small white and black squares, and a 50% grey box for comparison. Turn the gamma up until the checkerboard pattern matches the brightness of the box, and leave it there. You might find it helpful, depending on the image you use, to squint slightly and blur the checkerboard into something solid while adjusting this. Fortunately, since this happens digitally, there's no impact on any other adjustments, and you're done with it.
Your display should now appear glorious.
Next, color temperature. This is really the first step, but since most people -never- care about it, I put it last. Which doesn't make it unimportant, just neglected and largely unknown. An explanation is thus in order.
Color temperature represents the color of white. You've noticed that a white shirt appears differently under incandescent lights, or on a cloudy day, or in stark sunlight, or under harsh flourescents, or at a football game under metal halides -- even though, in all instances, the white shirt is still obviously white. This is color temperature. Metal halides, as used in stadiums, often have color temperature in excess of 15,000 Kelvin, which is largely blue. Incandescents are often as low as 4300k, which is yellow or brown (depending on your interpretation). The higher the number, the more blue it is. Lower, is more yellow/brown.
Like gamma, and also unlike NTSC (standard of 6500k) there's no standard color temperature for computer displays. I like my monitor at ~7000k, which to my eyes, looks like the sunlight that comes through my NW Ohio window as it appears on a sheet of paper.
Most monitors (and TVs alike, dispite the known, functional standard) have ghastly high color temperatures from the factory. This, as far as anyone can tell, is because of some human tendancy to think that a bluer picture is brighter, and thus better. Sony displays are notoriously blue out of the box.
I haven't seen a monitor sold in the past couple of years which did not allow adjustment of color temperature, and I also haven't seen one at all which has temperature presets which match reality. Ignore the presets.
So, first, you'll need to find a reference. A Kodak grey card and a lamp of known temperature would be nice. A color analyzer would be better, if one can be borrowed. Or, just use a sheet of plain white paper (not coated stock for inkjets) and ambient sunlight.
Use a full-screen image that's just plain white, or grey. You'll then tweak the individual gain of the red, green, and blue guns on your monitor until the displayed image matches the color (-not- brightness) of your reference. Alternate between white and grey images until you get as close a match as you can on each. Then (as this is the first step), go through and do all the other things mentioned above.
Done.
It will look darker again, but in reality, the light output hasn't been changed much by the temperature adjustment, but perception brightness has. Let it sit that way for a few days, and you'll soon look at other monitors with disdain.
Load up some pr0n, or fire up your favorite black-on-white application, and notice how much easier things are to read and discern. You'll see details in photographs that you positively couldn't before, because your monitor couldn't show them to you as it was stuck in "buy-me" mode, like a stereo turned up to 11.
If, after this rough calibration, you find that your monitor is overcome with glare, the obvious solution is to reduce ambient light. Don't try to turn up the monitor, because it can't do it. If you can't reduce the amount of light in the room, buy a new monitor. The one you have presently is aging, as vacuum tubes (particularly CRTs) tend to do, and it can't produce as much light as it used to be able to.
I was reminded of the old PCs just last night, when trying to configure LILO on a new PII-based Linux box.
On the first reboot, it came up, POSTed, and acted normal. Instead of LILO, what instantaneously appeared on the CRT was a message saying "CANNOT FIND ROM BASIC - SYSTEM HALTED".
It was beautiful, in a 40-column white-on-black font that I hadn't seen used in years.
I stared at it in awe for several minutes before rebooting to see it again. =)
FWIW, xmp claims to support hardware reverb and chorus on AWE cards under *nix, using OSS's /dev/sequencer.
I haven't yet tried that end of it. A couple of weeks ago, I decided to start collecting mods (and relatives) again, and I'm just too thrilled at the way they sound when played with xmp through a good external DAC to bother making an AWE32 work. Last time I heard most of this music, I was using an SB 1.5 on a slow 386SX. The difference is striking.
OTOH, the AWE32 cards do have an SP/DIF output straight from the Ensoniq DSP... =)
Moby Dick, by Hermen Melville, may be freely downloaded as zipped ASCII text at ftp://ibiblio.org/pub/docs/books/gutenberg/etext91 /moby.zip, courtesy of Project Gutenberg.
Enjoy.
Perhaps my view is somewhat skewed. I've had limited experience with DSL, and I haven't read the RFCs. If anyone with more of a clue cares to correct my notions, I invite them to do so.
/29 or somesuch, have your ISP provision you one. If they won't, you again have a social problem, not a technological problem.
In my area, Ameritech offers two kinds of end-user equipment. One is a DSL "router" which uses DHCP, performs NAT and includes a 4-port 10mbps ethernet hub, from Efficient. The other is a DSL "modem" which talks PPPOE to the customer's PC, and is made with the Westell mark.
The router works just as one might expect. It has a reasonably complete command line interface, allowing a fair amount of creativity with forwarded ports and other simple router tricks. I do not know what is involved with its initial setup, as Ameritech does this themselves. I also do not know if it is possible to use it without any NAT at all. And I've got no idea what the transport layer back to the DSLAM consists of.
And I just don't care. It works well.
The DSL "modem" is just like any other modem. POTS into one jack, serial into another. Except in this case, that serial data is encapsulated into ethernet frames. So what?
It talks PPP, quite obviously. I'm using the Roaring Penguin software, with superb reliability. On 160/768 ADSL, I see a few percent of CPU usage on an absolutely horrible Cyrix MII firewall/NAT box under one of the 2.2 linux kernels. If such a meager amount of CPU time can't be spared (and, according to some of you, it's impossible), I guess you've already got your priorities in line and need a router of some form to offload the task. Nevermind that the cute perl script you whacked together to paste together fragments of porn from usenet, the cause of all the bandwidth usage, is already eating more CPU than that.
This latter arrangement also sports a dynamic IP address. Who cares? It was trivial to have the address updated automagically on one of the dynamic DNS services. It's also at the whim of the ISP - I'm sure that I could get a static IP, if I wanted one. It's certainly not an issue of the capabilities of PPP, but more of a social thing. I've had a static IP PPP dialup nailed up at home for years, without an ounce of trouble. If the connection drops, it reconnects. Things then continue where they left off, like it never happened.
PPP also supports routing entire networks, or subnetworks, or whatever. If you want a
Which is to say that PPPOE is just as capable as anything else DSL, given some sort of router. Is the need for a router at home something new to this world of supposed geeks? Stop whining and fire up ipchains, natd, ipfwadm, or a pretty box that says Linksys, D-Link, Netgear, ZyXEL, or some other such name.
People complain that PPPOE links aren't always-on. Of course they aren't. NOTHING in this world is.
Deal with it. The PPPOE-connected linux firewall does justfine handling such things as bringing the connection back in the event of it dropping. It takes seconds. It happens infrequently enough that I don't notice unless I'm looking at logs.
This stuff works over a pair of amazingly thin copper wires, strung crazily around the streets at the whim of city leaders, and operated by the telephone company. Nevermind that the design specification for these wires is 300-3,000Hz, and that DSL in any incarnation is an ugly fucking hack to begin with. It's amazing it works at all. And you want perfection?
This change is so fucking insubstantial to the way things work in practise that it's absolutely laughable to see so many people upset about it. When's the last time you had a network problem, and found PPP (over -any- medium) to be at fault?
Get a life. Stop whining. Even if PPPOE -does- incur a real performance hit, you'll never notice. And if you do notice, and still care, and still feel like whining about insignificant things, find something else. Vote with your wallet. Show those money-grubbing assholes as SBC just how you want to be treated.
You can have your T1 (and pay for it, too).
Resampling digital audio always has detrimental effects. Doing it twice (44.1 to 48 and back) makes these issues worse. Three times (44.1 to 48 to 44.1 and 48 again on playback through an SBLive!) is obviously quite a bit of comb-filtering bit-munging.
Better to just buy a card with a non-resampling assortment of SP/DIF I/O, such as the plethora of "pro" cards from Lexicon, M-Audio and the like.
Or, a $30 Zoltrix Nightingale (or about any other card [including some motherboards] based on the CMI8738 chip) will do the trick nicely with coax or toslink. Also works well as a hardware format converter, and an SCMS stripper.
While I'm on the subject, the error correction of a CD player takes place well before the bits reach the digital output.
While I'm on the subject, it occurs to me that such things as SafeAudio lend a hand toward legitimizing filesharing services. "Well, your Honour, I didn't have any way to utilize Fair Use and use this CD in the MP3 player that came with my new Mazda, so I downloaded the files from someone else who was able to figure it out."
Check local surplus houses. Mendelson Electronics (www.meci.com) is near enough to me that I can swing down in an afternoon, and pick up rack of almost any size for less than $100-150. Some of these are nice, half-sized racks with glass doors. They say things like "Compaq" on them.
Buy a road case from someone like Anvil, Starcase, or SKB. They're durable. You can take them with you to a LAN party. And your equipment will be safe in transit, or as luggage on a 747, or whatever.
Alternatively, build a rack. Parts for serious road cases and other racks (aluminum extrusions, hasps, heavy steel corners, plywood with colorful vacuum-laminated fiberglass, pre-tapped rails, etc) can be found at TCH (the URL escapes me). They seem to be the same to road cases as Black Box is to networking gear.
Rails are also available at Parts Express (www.partsexpress.com), for cheap.
Whatever you do, be sure to compare the depth of the equipment to the depth of the rack. It might not be a big deal if the back of your server hangs out of the cabinet a few inches, unless it bothers you to look at. But it would be somewhat troublesome if the back cover for your new road case doesn't fit once it's loaded with equipment.
Companies like Starcase and Anvil are completely willing to build custom projects, so if you -really- want something special...