Please don't.... a yellow mustang? Never... save it for the Hummer that has yet to be discovered (or perhaps its bones... aren't they extinct? Or Chinese now?). I'll take the Mustang in red or blue, thank-you.
Google is a member of the Clear alliance, the Wi-Max guys. Sprint owns 51%, the former Clearwire shareholders own 27%, and the rest is owned by a consortium of Comcast, Time Warner, Intel, Google, and Bright House. So Google's already in the wireless business, albeit as a minority shareholder. And they did bid at the 700MHz auction, though it was Verizon and AT&T that actually won the best RF blocks available. But they did lock in open access rules -- you'll have to pay AT&T or Verizon to get hooked in to their 700MHz LTE networks, but they can't dictate the type of device being used.
Actually, Comcast is one they're pretty comparable to:
Comcast FY2008: revenue = 34,256 million, net income = 2,547 million Google FY2008: revenue = 21,795 million, net income = 4,226 million
More interesting... Comcast's income has doubled over the last six years. Google has slowed down to the point where they're only doubling every two years or so. That's why Google has nearly five times the market cap of Comcast (another way to measure corporate size, since that's largely a measure of the expected future of the company).
Most Blackberries have the same problem as the iPhone... the screen is just too damn low-resolution. The web is questionable enough on any pocket device, but at 480x320, you're just dreaming about a great web browsing experience. Ok, it's every-so-slightly better than my old Treo (320x320), but my new DROID is the first pocket device this side of the better Nokia Linux web devices (which are pocket-sized only if you're wearing cargo pants) with enough actual screen for real web pages to work as real web pages. Even Slashdot is totally readable at 856x480.
I would also add "keyboard" to the list of demands... whether it's Blackberry/Palm style with tiny nubs, or a slide out like the DROID and the G1, it's really a must have for any extended "computer-like" use of these pocket computers we still insist on calling phones.
Most of those audio + USB connectors are bog-standard USB... they just sense if you're connected to USB or not, and switch the audio signals to the connector when you don't have an active device connected there (eg, the headset is a passive device, no power provided). This is old hat -- Motorola's RAZR did this.
All the new phones use the micro-B USB jack, not the mini-B. It's supposed to be about 20x more long lived.
I will, once they stop being single-function, dedicated devices, and become general purpose computing tools.
Like many, I've had to deal with smart or semi-smart phones with functions crippled intentionally by the carrier, in order to make it easier to nickel and dime customers together. I don't think it's even necessary to flash your own OS... just the fact that you have the ability to flash the OS yourself will prevent this kind of tampering.
I think you'd also look for this ability much more readily if these other electronic devices were similarly compromised. If my microwave oven refused to cook food from unapproved manufacturers, if my washer only worked with "Tide" detergent, if my television down-rezzed content from non-Viacom sources, if my car only accepted fuel from Exxonmobile, you can bet that my next purchases would ensure I had the ability to work around those sort of artificial limitations.
And yet, the phone manufacturers have pretty much fixed this. All the new phones have moved to the micro-B USB connector as the basis for charging. Ok... it's probably not a cable you already have (I had B, mini-B, micro-AB, but no micro-B in my vast collection of cables... but they're $0.65 at monoprice.com), but it's a start. My old Verizon RAZR had some Verizon evil in there to reject charging over that phone's USB other than from "Authorized" chargers... the new DROID charged up just dandy from some old USB adaptor I had left over from some random need to change an MP3 player, years back.
They're of course still pushing all the expensive accessories in the store. But maybe they're starting to think about things I might actually want... I actually did spend $20 on a car mount for the DROID, which is more than I'd ever spent on in-store accessories before. Largely because that seemed a useful thing at a reasonable price, not something that should be in the box as a freebie, or a $5 charger selling for $30.
You can't likely reprogram the radio baseband processor -- yes, the local regulatory boys (like the FCC in the USA) do know that the radio code you're running affects the degree to which your radio device follows or does not follow the local rules.
On the other hand, pretty much every phone these days, and all smart phones, have separate application processors anyway -- that's where the phone OS runs: Android, iPhoneOS, Symbian and Maemo (both Nokia these days), WebOS, whatever RIM calls their BlackBerry OS, etc. And there's no reason you can't change that one... at least in the case of Android, Maemo, and the more recent Symbian, which are open source.
Of course, open source has the potential to cut both ways... carriers have always been able to get customized versions of a phone OS, but may have been limited in what they could change. These days, they're able to do all kinds of intense customization if they like... to the point where, on Android, there's a "Google Experience" branding to indicate which versions have not been altered in any way by the manufacturer. Some of that altering has proven fine... there are some pretty good alternate "home" programs for Android (the "desktop", if it were a desktop computer), but there's certainly the potential for evil there. But less so than in the past.
As far as open networks go, at least in the USA, there's only so much openness. GSM voice is definitely open... largely because the carriers have no way to block it.... you can't tell an AT&T issued SIM card from a T-Mobile issued SIM card. Unfortunately, for CDMA systems, the network ID is built-in on the phone, and you can tell who's is who's... so Sprint will refuse to allow your Verizon phone on their system, and vice-versa. The laws were cleverly written, I suppose, probably by lobbyists for Verizon... it's illegal to refuse to unlock a phone (though they can charge for it), but there's no law requiring any compatible phone to be accepted on any given network.
And beyond that, the issues with frequency allocations (particularly on GSM, since they need much more bandwidth for 3G than on CDMA) and other network configuration issues make it dicey at best to move any sort of smart phone onto a network other than its network of origin. This would ideally be something all sorted out in the 4G world (well, other than the fact there's already competing standards, with Sprint (Clearwire, Intel, Google, Comcast, etc) in the WiMax camp and everyone else waiting to roll out LTE starting next year, and on new frequencies (Verizon and AT&T both have 700MHz spectrum, though Verizon has twice the bandwidth). Someone, I'd wager on "business as usual".
Google's web-based word processor is different because it already has 2 million customers. Not as big as MS-Office, certainly, but that does suggest that it has past a certainly usability level. Most web-based tools of the past simply failed because they sucked.
Right. And hey, maybe it was a prototype, maybe not. But I saw the photo, and as a life-long carpenter, I can assure you, those were machine screws, not wood screws. The screw head wasn't deep enough for a proper wood screw (wood screws and sheet metal screws have to be self-tapping, machine screws thread into a tapped bolt).
It's not so much the stress on the GPU as the stress on the CPU. GPUs came along to do things faster than a PC can do them.
Since HD came along, one of those things is video. You need a pretty decent dual core PC to play high end HD video (1080/60p, for example) using 100% of both CPUs. Throw a decent GPU doing video acceleration in there, and that'll drop, maybe to 50%, maybe lower. Sure, that same GPU might do the 3D pipeline 25x faster than that same CPU, but that doesn't negate the value of GPU video acceleration.
GPUs are also starting to help out in video rendering. This is a place where a 25x improvement over the CPU (rather than the maybe 2x improvement you're seeing now, with the best rendering accelerators so far) would be considered "a nice start". So don't think gaming is the only value here (well, ok, gaming and mechanical CAD, if you have a high-end enough GPU to do a properly accurate OpenGL).
Re:Android WILL take over.
on
Less Than Free
·
· Score: 1
No, no one made all such claims about Linux. The big missing one would have gone something like "Compaq is putting Linux on 50% of its computers", "Dell has 350 people working on Linux", "all major computer companies except Apple and Microsoft are actively working on Linux computer systems"... something like that.
What you see today that really makes you think isn't Open Source, isn't Free, isn't "Runs on any hardware". It's the fact that every major phone manufacturer except Apple, Palm, and RIM (far as we know) are working on it (even number 1 smart phone manufacturer Nokia, though so far, only on one of their web tablet devices, not a phone.. yet). HTC has it on half of their phones. Motorola has 350 people working on Android related things. Dell's moving into the smart phone market (and China) based on Android. So are a bunch of other PC makers and CE companies not usually associated with cellular phones.
Intel already had a power dominance in the CPU industry (still do... about 51% of all semiconductor profits come form Intel's x86 chip line). The went to their OEMs and said, basically, if you buy this other, competing product, we're going to hurt you. We're going to charge you more for our products, and you may have a problem getting the latest ones. The big reason this was illegal, is the same reason it was possible: Intel is so big, they have monopoly powers. If they didn't, they couldn't strong-arm anyway, they'd be told to go pound sand. Or silicon. This is the same level of control Microsoft still enjoys over the majority of the computer industry. It's no illegal to have that kind of control, it's illegal to use it to artificially restrain the market from other competitors. That's precisely what Intel was doing.
Google is not doing anything like that.. they're simply saying, "if you put the 'Google Experience' on your Android phone, we'll share revenue from that phone with you". Why not... they're not yet a major force in Smart Phones. And it's not even a requirement to use Android... many of the phones out there are taking the "roll your own" approach, like Motorola's Blur interface, Sony's new Rachael, and the HTC thing.
The big reason is phone companies and networks alike are loving Android is that this is the opposite of their previous treatment. You want Windows Whatever on a phone, you pay Microsoft the $25. And as usual, MS has done little to promote that... hey, they get their $25 buck for every phone, whether it sells or not. Or take Apple... they went to AT&T, and demanded a little of their action -- you pay us, and we'll let you carry the iPhone. If not, we'll take it elsewhere... apparently, AT&T was one of those elsewheres, after Verizon said "no". Apple taking money from Verizon is no more or less abuse of monopoly powers than Verizon now taking money from Google. Verizon is far more powerful in the phone industry today than Google, but this isn't abuse of monopoly powers, since neither is any kind of monopoly.
The big win for Google is ultimately expanding this. While Apple's still trying to get a deal in China with the number 3 carrier there, Dell just struck a deal with China Mobile.. the big 500,000 customer China Mobile, to supply them with Android phones. The simple fact is that Android is liberating the possibilities of the mobile networked computer we still call a phone in ways even better than MS-DOS and Windows did with the PC, since no one is ultimately in total control of the platform.
This also does offer consumers a level of assurance that never existed before. If you're worried about your carrier cutting off features, you know you can root your phone, install open source versions of Android, etc. You can buy a "Google Experience" model like the DROID... carriers are not permitted to mess with those. Ok, sure, Google is, but they have yet to be shown to be interested in locking me out of functions. In fact, they keep handing me new ones, free. No gun to the head... I don't have to use Google Navigation... but it's SO cool...
Technically, you need infinite resolution to record a perfect 22kHz square wave. If you were to record this at 44.1kHz sampling, what you actually get is a pretty nice 22kHz sine wave on playback.
You should mention RIAA equalization, too. Much of the reason you get bass at all on LPs is because, before a perfectly reasonable audio signal is put on an LP, it run through RIAA "pre-emphasis". This cuts bass at 20Hz by 20dB, and boosts treble at 20kHz by 20dB (there's a detailed curve one must follow ). When you play it back, you boost the lows and cut the highs. This keeps the grooves small enough on the LP to allow it to carry its required 22min or so per side, and cuts down on high frequency hiss... effectively dropping hiss by up to 20dB. Same basic principle as Dolby Noise reduction, in fact. This technique was invented in the mid 1920s at Bell Laboratories.
This was originally an ad-hoc thing, but eventually standardized by the RIAA (and you thought they just sued 13-year-olds for downloading MP3s). A few early CDs were put on disc with the RIAA pre-emphasis applied, and naturally, there was no de-emphasis on playback -- they sounded horrible. Part of the mad rush to get material on CD.. eventually people actually started listening to the masters before they went on disc.
The intent of the gold contacts... ok, well, today, the intent is to make consumers say "oooh... goooollllddddd" and buy these cables, rather than the cables without gold plating for half the price sitting next to them. But the real functional purpose is simple: gold doesn't corrode. Same reason there's gold plating on the fingers of your PCIe cards, or on many types of computer connector.
Of course, the gold is so thin, it's not totally useless, but short-term useful. If you plug once, the gold's still in place, and your connector won't corrode, or react with the metal of the mating connector, assuming that's not also gold. But cycle this a little, and the gold's worn off... and chances are, what's under that gold is something evil.
This is why professional audio cables, as used in studios, on stage, etc. typically have much thicker nickel plating. Not as attractive as gold, but also not corrosive, and it lasts through thousands of plug/unplug cycles.
As far as what's important, you left out impedance... cable impedance is very important, for long runs of cables. When it's a video cable, or worse yet, a digital cable, "short run" is probably a matter of an inch or few... so an impedance matched cable is life of death. Important for audio, too... at least when you have a cable run of a mile or two. Over practical lengths, it's not an issue.. the frequencies are way too low.
Similar is the "skin effect" that audiophiles often yammer about (the wire's self-inductance, plus an AC signal, creates a small reverse current, which manifests as an increased resistance, the deeper one goes into the wire). For a radio signal, a gigabit digital link, etc. it is actually true that most of the signal rides very close to the surface of the connector.. effectively, the cable has more resistance than the gauge would suggest. At audio frequencies, not so much. A 10ft run of 12ga solid wire would have a natural resistance of about 0.033 ohms. At 20kHz, you might see an additional 0.011 ohms... not significant. This drops off with frequency, of course. And it's not terribly likely that anyone's going to use 12ga solid core wire... oh wait, that's the coat-hanger speaker wire again! So there actually is a tiny but measurable difference between that and the [stranded] Monster cable.
Actually, that's an important point -- re-encoding. The device you started out with might be MP3, but most folks change devices more often than they re-rip entire CD collections. I started out with MP3 players. I had an iPod once (it was a gift) that could handle AAC, which certainly has a higher coding efficiency that MP3, so (particularly on a 10GB device) you want the AAC. Then it was a Zune (30GB for $75 was a good deal, and the brown matches my guitar, so I don't want to hear it), and more recently, a DROID, which can play OGG as well as MP3 and AAC.
The one thing I liked about the Zune software... you could let the sync program worry about re-coding whatever you had on HDD for device use. Typically, the device quality is limited by my environment, by ear-buds rather than studio headphones, and by the device itself (particularly that iPod...). So no need for FLAC on the device, yet. Why not encode your audio for the device being used, today.
Only, with lossy compression, you run the risk of greatly reduced sound quality. It was a factor, in early compression algorithms, that they suffered significant generational loss. Early Sony MD players suffered really noticable loss after a few generations (later ATRAC encoders could do dozens without noticable loss). But that's accelerated when you mix and match psychoaccoustic models. So you may find that 64K AAC file made from a 256K OGG or MP3 file is dramatically lower quality than, say, a 64K AAC file made from a 256K AAC file. Keeping around uncompressed audio, no issue here... each transcoding will sound as good as if ripped directly from the original CD.
Oh... also, take a close look at those cable ends. Doesn't that look like nothing more than gold plating over PLASTIC.
This wouldn't be the first time that companies try to fool consumers using the sensibilities of the past. Like trying to convince consumers that better digital cables of any kind will add improved quality to their audio/video. Sorry... it's pass/fail, pretty much.
Hah... yeah, that's a good one! Nearly as funny is that this cable, like all consumer TOSlink (from Toshiba-Link... Toshiba's answer to Sony & Philips RCA-based standard) cables, is made of plastic fiber optic material, not glass. So it's actually pretty lossy over an 8ft run... if your device doesn't have a high gain receiver, you may be getting substantial bit errors. Technically, the interface is supposed to be good for 10m... some devices handle even much longer runs, others, not so much. You're generally much better off with a decent RCA cable on the old S/PDIF connector. Or a modern HDMI cable, also carrying digital audio over good old copper.
Best Buy, in general, is scamming people on cables, like most A/V resellers. They generally have to compete pretty realistically on price versus other retailers on the main item. I bought an HDTV at Best Buy some years back, after investigating various alternatives, including online ordering from a place like B&H Photo Video (good prices, great service). But even with the lower price, Best Buy came out on top once you factored in postage vs. tax.
But right next to the HDTVs and other gear, they're selling $100+ HDMI cables and all kinds of other interconnects, at prices that would make a Radio Shack guy self-conscious. That's a big profit center for them.. these days, they can make nearly as much profit on selling you accessories as on the TV itself. And the thing is, with those fancy cables, you may not even know what you're getting. What is the gauge of the wire inside? Far, far better off buying cables at a place like monoprice.com.
I think some of this stuff started in good faith, but got to the crazy "Audiophoole" level of today through good, old-fashioned greed.
When you're talking strictly analog signals, there's at least a decent notion that you can do something to somehow improve the quality of the signal, since there's always some loss. So, start with the coat hanger speaker wire, and work up from there... you just MIGHT make something that really, honestly, sounds better. The problem becomes that, next week, you competitor produces a better-still one, complete with full explanation as to why it's better, and they're selling it for more than yours. So you must escalate, and so on.
That, sadly enough, has lead to this fictional world of $10,000 per 10ft speaker cables, with all kinds of science-fiction written up about their special properties (eg, "Any significantly advanced scientific explanation is indistinguishable from a scam"... some of these guys ought to be writing "Star Trek" screen plays... they'd never come up with something as silly as "red matter"). There's so much obvious BS, any lingering truth is hidden.
On a more practical point, there's the original recording. Yes, some recordings are made using guitars that cost several thousands, pianos that cost more than your car, mics that run into the thousands, etc. Others, less so. Either way, the wires being used... not so expensive. You might pay $50-$100 for a mic cable, but that's not to get some magical property in the wire other than "copper", but to get a cable that's going to last through the typical abuse of studio or stage. It's hard to believe anyone telling me you need to spend 100x more on wire to play back music than you needed to create it. Plus, you don't find musicians arguing about the "color imparted" by the mic cable, as if discussing a fine wine. Any sound engineer overhead discussing the subtle harmonics or enhanced tonal clarity from the wires between his console and his monitor is either a poser, or he's been in to the brown acid again.
That's interesting, but not unexpected. We tend to favor the familiar, and we tend to adapt to it.
I found myself doing this with digital video. I started working with digital video in the early 1990s, the MPEG-1 days. Anyone who recalls MPEG-1 on VideoCD and early digital satellite TV probably has nightmares about the quality, and rightly so. But part of that wasn't quite so simple... it wasn' t that VCD has any more or less distortion than typical consumer VHS, but -- once you factored out things like bad encoding and all -- it was simply that most of us had spent the last 10-20 years adapting to VHS, which itself was pretty similar to plain old television (a bit worse, but bad in exactly the same ways).
So enter digital, and the early digital was awful.. but a big piece of that was psychological... we hadn't learned to ignore the noise. Over the years, I got more in digital video, got TiVo for my house and worked on various early PVR stuff (satellite receiver recorder with internet and other cool features back in the late 1990s), and before you knew it, I didn't mind TiVo video so much. Sure, I could still the noise, but not all that shabby. But just as interesting, analog video was starting to look much worse to me... I was losing my adaptation to the analog forms of noise.
Since then, of course, I've moved on to HD, first in my own video stuff, then on TV, and now all that older stuff looks like crap.
It's pretty impossible for a generation of listeners to love their iPods and yet not adapt to the sound of the iPod. If that's the normal for you, I do believe you'll in time find the the preferred sound. Which is kind of sad, given that's a drop in quality versus what's attainable... but not necessarily versus what most folks generally listened to.
I mean, most of old farts probably adapted just the same to LP and FM radio, which is one reason CDs were just so fantastic when they first came out. More hidden, we also adjusted to the realities of analog production of the day, various distortions that audio engineers just had to live with... to the extent that, even when they're gone, we're still adding them back in... tape and tube emulators are among the most popular plug-ins for DAWs. I myself have this wonderful little pre-amp that contains a DSP that can model all kinds of tube and other analog preamps -- works great on mic and guitar, and at least gets me a step closer to that sound I grew up with.
Makes you wonder... will kids today start rejecting any music WITHOUT Auto-tune abuse on it?
If you're mixing down directly from 24-bit to 16-bit MP3, that in itself is probably bad news. Unless you have a fairly smart MP3 encoder, it's probably just truncating the 24-bit samples to 16-bit, which is usually going to result in very noticeable quantization error (essentially, you're magnifying the existing quantization error in the 24-bit audio).
You really want a decent dithering and noise shaping conversion when you drop from 24-bit to 16-bit. I use those built-in on Sound Forge myself, and there are plenty of other audio tools that do this. You're adding what amounts to a small bit of analog noise, in order to effectively remove the very annoying digital noise. There are a variety of different algorithms... if you're not doing this, give it a try. This is a standard function in most modern DAW tools... I've heard even the open source Audacity has dithering for bit depth conversion. Go to mastering plug-ins like iZtope's Ozone for more advanced approaches. You might also the the free dithering plug-in in the MDA pack: http://mda.smartelectronix.com/
I'm not sure the big argument when CDs came out was about the dynamic range (eg, 16-bit vs. something better). You might argue for or against 44.1kHz vs. 48kHz vs. 96kHz or whatever. No one's going to argue that 24-bit is a waste vs. 16-bit... I think it was just the tech limits of the day that estabilished 16-bit as the CD standard. After all, even today, you'll usually find most 24-bit gear is actually 18-22-bit gear with these in there as "marketing bits". Of course, when you're doing a mix from enough sources, that won't matter anyway... you'll have a true 24-bit product when you're done.
The CD controversy has always been around the claim that most people can't hear beyond 20kHz (true), versus the claim that you can't hear the effect of frequencies beyond 20kHz in music (fals). So for CD, they figure you're pretty safe with the Nyquist frequency a bit over twice that (enough to maybe-kinda-sorta allow for good anti-imaging and anti-aliasing filters based on 1980's technology). Obviously they were wrong.. but in reality, no one will EVER need more than today's high-end specs, 24-bit at 192kHz. Trust me on this:-)
I think that's a good way of say it... "the quality is quite good". This comparison tends to suggest so, too, and that's not the first time.
When MP3 was originally created, the main idea was to be able to schlep audio feed for radio broadcast over ISDN lines... 128kb/s digital, rather than some kind of expensive dedicated line. MP3 is clearly "transparent" (which means you can't tell the difference between it and uncompressed) for voice at 128kb/s, and for the purposes of FM radio, just about anything else. Once you're listening direct, yeah, it's not too hard to pick out a 128kb/s MP3 vs. uncompressed, if you know what to listen for. WMA was worse, at least when it first came out... it had really obvious pre-echo... but maybe that's been refined. Certainly, today's LAME in VBR at 256kb/s or so, I can't tell the difference, maybe on some rare tracks where the psychoaccoustic model breaks down. Maybe when I was 18, and had only been to a handfull of rock shows:-)
I think that Amazon has the right idea... high quality MP3, and they do essentially acknowledge it's somewhat less of a thing than a glass mastered CD. Amazon MP3 albums vary from $1.00 below the CD to the occasional $1.99 for the whole album. It's easier to accept something like that... it's theoretically lower quality, I can't likely tell the difference, AND it costs less. Though I'm still going to buy "In the Court of the Crimson King" 40th Anniversary edition with 5.1 mix on DVD-Audio. From analog tape, circa 1969..
Eventually, if there's a "good flaw" from the analog world, someone will add in via a plug-in to the digital world. Things like tape saturation, tube harmonics, etc. are commonplace these days as studio plug-ins... only, with very precise control.
Why are these "good flaws"? Well, that's a fairly involved topic, with different camps. One set of advocates are hardcore adherents -- they'll claim that tape saturation is something ideal, or tube harmonics and saturation, etc.... versus, say, other types of intentional compressors. Others look at it as a historical thing... we musicians of today grew up listening to classic stuff made on this now-vintage gear, so we're turned into that particular characteristic sound, and want to re-create that.
I'm kind of in both camps... some of this stuff really does sound good, but at least part of why I judge that "good" is because I heard the same effect on a Springsteen or Dylan album I grew up with.
And there are plenty of analog tape artifacts most folks can live without. Most of the time, you don't want the hiss, wow, flutter, or spectral non-linearities you get with tape... you really just want the "good stuff", and then, only when you want it. About the only time I really miss analog tape saturation, for example, is when I'm recording something just a little too hot. In digital, you get those missing peaks -- can't go over 0xFFFF or 0xFFFFFF no matter what you do (assuming unsigned samples, of course). On tape, it would probably sound fine... I just shot a video last weekend, and one mic was just a bit too hot... had to toss that out entirely (fortunately, I had three other mics and a mixer feed to work from).
Slashdot Mirror
Please don't.... a yellow mustang? Never... save it for the Hummer that has yet to be discovered (or perhaps its bones... aren't they extinct? Or Chinese now?). I'll take the Mustang in red or blue, thank-you.
Google is a member of the Clear alliance, the Wi-Max guys. Sprint owns 51%, the former Clearwire shareholders own 27%, and the rest is owned by a consortium of Comcast, Time Warner, Intel, Google, and Bright House. So Google's already in the wireless business, albeit as a minority shareholder. And they did bid at the 700MHz auction, though it was Verizon and AT&T that actually won the best RF blocks available. But they did lock in open access rules -- you'll have to pay AT&T or Verizon to get hooked in to their 700MHz LTE networks, but they can't dictate the type of device being used.
Actually, Comcast is one they're pretty comparable to:
Comcast FY2008: revenue = 34,256 million, net income = 2,547 million
Google FY2008: revenue = 21,795 million, net income = 4,226 million
More interesting... Comcast's income has doubled over the last six years. Google has slowed down to the point where they're only doubling every two years or so. That's why Google has nearly five times the market cap of Comcast (another way to measure corporate size, since that's largely a measure of the expected future of the company).
Most Blackberries have the same problem as the iPhone... the screen is just too damn low-resolution. The web is questionable enough on any pocket device, but at 480x320, you're just dreaming about a great web browsing experience. Ok, it's every-so-slightly better than my old Treo (320x320), but my new DROID is the first pocket device this side of the better Nokia Linux web devices (which are pocket-sized only if you're wearing cargo pants) with enough actual screen for real web pages to work as real web pages. Even Slashdot is totally readable at 856x480.
I would also add "keyboard" to the list of demands... whether it's Blackberry/Palm style with tiny nubs, or a slide out like the DROID and the G1, it's really a must have for any extended "computer-like" use of these pocket computers we still insist on calling phones.
Most of those audio + USB connectors are bog-standard USB... they just sense if you're connected to USB or not, and switch the audio signals to the connector when you don't have an active device connected there (eg, the headset is a passive device, no power provided). This is old hat -- Motorola's RAZR did this.
All the new phones use the micro-B USB jack, not the mini-B. It's supposed to be about 20x more long lived.
I will, once they stop being single-function, dedicated devices, and become general purpose computing tools.
Like many, I've had to deal with smart or semi-smart phones with functions crippled intentionally by the carrier, in order to make it easier to nickel and dime customers together. I don't think it's even necessary to flash your own OS... just the fact that you have the ability to flash the OS yourself will prevent this kind of tampering.
I think you'd also look for this ability much more readily if these other electronic devices were similarly compromised. If my microwave oven refused to cook food from unapproved manufacturers, if my washer only worked with "Tide" detergent, if my television down-rezzed content from non-Viacom sources, if my car only accepted fuel from Exxonmobile, you can bet that my next purchases would ensure I had the ability to work around those sort of artificial limitations.
And yet, the phone manufacturers have pretty much fixed this. All the new phones have moved to the micro-B USB connector as the basis for charging. Ok... it's probably not a cable you already have (I had B, mini-B, micro-AB, but no micro-B in my vast collection of cables... but they're $0.65 at monoprice.com), but it's a start. My old Verizon RAZR had some Verizon evil in there to reject charging over that phone's USB other than from "Authorized" chargers... the new DROID charged up just dandy from some old USB adaptor I had left over from some random need to change an MP3 player, years back.
They're of course still pushing all the expensive accessories in the store. But maybe they're starting to think about things I might actually want... I actually did spend $20 on a car mount for the DROID, which is more than I'd ever spent on in-store accessories before. Largely because that seemed a useful thing at a reasonable price, not something that should be in the box as a freebie, or a $5 charger selling for $30.
You can't likely reprogram the radio baseband processor -- yes, the local regulatory boys (like the FCC in the USA) do know that the radio code you're running affects the degree to which your radio device follows or does not follow the local rules.
On the other hand, pretty much every phone these days, and all smart phones, have separate application processors anyway -- that's where the phone OS runs: Android, iPhoneOS, Symbian and Maemo (both Nokia these days), WebOS, whatever RIM calls their BlackBerry OS, etc. And there's no reason you can't change that one... at least in the case of Android, Maemo, and the more recent Symbian, which are open source.
Of course, open source has the potential to cut both ways... carriers have always been able to get customized versions of a phone OS, but may have been limited in what they could change. These days, they're able to do all kinds of intense customization if they like... to the point where, on Android, there's a "Google Experience" branding to indicate which versions have not been altered in any way by the manufacturer. Some of that altering has proven fine... there are some pretty good alternate "home" programs for Android (the "desktop", if it were a desktop computer), but there's certainly the potential for evil there. But less so than in the past.
As far as open networks go, at least in the USA, there's only so much openness. GSM voice is definitely open... largely because the carriers have no way to block it.... you can't tell an AT&T issued SIM card from a T-Mobile issued SIM card. Unfortunately, for CDMA systems, the network ID is built-in on the phone, and you can tell who's is who's... so Sprint will refuse to allow your Verizon phone on their system, and vice-versa. The laws were cleverly written, I suppose, probably by lobbyists for Verizon... it's illegal to refuse to unlock a phone (though they can charge for it), but there's no law requiring any compatible phone to be accepted on any given network.
And beyond that, the issues with frequency allocations (particularly on GSM, since they need much more bandwidth for 3G than on CDMA) and other network configuration issues make it dicey at best to move any sort of smart phone onto a network other than its network of origin. This would ideally be something all sorted out in the 4G world (well, other than the fact there's already competing standards, with Sprint (Clearwire, Intel, Google, Comcast, etc) in the WiMax camp and everyone else waiting to roll out LTE starting next year, and on new frequencies (Verizon and AT&T both have 700MHz spectrum, though Verizon has twice the bandwidth). Someone, I'd wager on "business as usual".
Google's web-based word processor is different because it already has 2 million customers. Not as big as MS-Office, certainly, but that does suggest that it has past a certainly usability level. Most web-based tools of the past simply failed because they sucked.
15 years ago, it was a 68040 Amiga at 25MHz or an Intel Pentium at less than 100MHz running Windows 3.1.
My "phone" is more powerful than these today... don't worry about the desktop.
Right. And hey, maybe it was a prototype, maybe not. But I saw the photo, and as a life-long carpenter, I can assure you, those were machine screws, not wood screws. The screw head wasn't deep enough for a proper wood screw (wood screws and sheet metal screws have to be self-tapping, machine screws thread into a tapped bolt).
It's not so much the stress on the GPU as the stress on the CPU. GPUs came along to do things faster than a PC can do them.
Since HD came along, one of those things is video. You need a pretty decent dual core PC to play high end HD video (1080/60p, for example) using 100% of both CPUs. Throw a decent GPU doing video acceleration in there, and that'll drop, maybe to 50%, maybe lower. Sure, that same GPU might do the 3D pipeline 25x faster than that same CPU, but that doesn't negate the value of GPU video acceleration.
GPUs are also starting to help out in video rendering. This is a place where a 25x improvement over the CPU (rather than the maybe 2x improvement you're seeing now, with the best rendering accelerators so far) would be considered "a nice start". So don't think gaming is the only value here (well, ok, gaming and mechanical CAD, if you have a high-end enough GPU to do a properly accurate OpenGL).
No, no one made all such claims about Linux. The big missing one would have gone something like "Compaq is putting Linux on 50% of its computers", "Dell has 350 people working on Linux", "all major computer companies except Apple and Microsoft are actively working on Linux computer systems"... something like that.
What you see today that really makes you think isn't Open Source, isn't Free, isn't "Runs on any hardware". It's the fact that every major phone manufacturer except Apple, Palm, and RIM (far as we know) are working on it (even number 1 smart phone manufacturer Nokia, though so far, only on one of their web tablet devices, not a phone.. yet). HTC has it on half of their phones. Motorola has 350 people working on Android related things. Dell's moving into the smart phone market (and China) based on Android. So are a bunch of other PC makers and CE companies not usually associated with cellular phones.
Disruptive indeed.
Intel already had a power dominance in the CPU industry (still do... about 51% of all semiconductor profits come form Intel's x86 chip line). The went to their OEMs and said, basically, if you buy this other, competing product, we're going to hurt you. We're going to charge you more for our products, and you may have a problem getting the latest ones. The big reason this was illegal, is the same reason it was possible: Intel is so big, they have monopoly powers. If they didn't, they couldn't strong-arm anyway, they'd be told to go pound sand. Or silicon. This is the same level of control Microsoft still enjoys over the majority of the computer industry. It's no illegal to have that kind of control, it's illegal to use it to artificially restrain the market from other competitors. That's precisely what Intel was doing.
Google is not doing anything like that.. they're simply saying, "if you put the 'Google Experience' on your Android phone, we'll share revenue from that phone with you". Why not... they're not yet a major force in Smart Phones. And it's not even a requirement to use Android... many of the phones out there are taking the "roll your own" approach, like Motorola's Blur interface, Sony's new Rachael, and the HTC thing.
The big reason is phone companies and networks alike are loving Android is that this is the opposite of their previous treatment. You want Windows Whatever on a phone, you pay Microsoft the $25. And as usual, MS has done little to promote that... hey, they get their $25 buck for every phone, whether it sells or not. Or take Apple... they went to AT&T, and demanded a little of their action -- you pay us, and we'll let you carry the iPhone. If not, we'll take it elsewhere... apparently, AT&T was one of those elsewheres, after Verizon said "no". Apple taking money from Verizon is no more or less abuse of monopoly powers than Verizon now taking money from Google. Verizon is far more powerful in the phone industry today than Google, but this isn't abuse of monopoly powers, since neither is any kind of monopoly.
The big win for Google is ultimately expanding this. While Apple's still trying to get a deal in China with the number 3 carrier there, Dell just struck a deal with China Mobile.. the big 500,000 customer China Mobile, to supply them with Android phones. The simple fact is that Android is liberating the possibilities of the mobile networked computer we still call a phone in ways even better than MS-DOS and Windows did with the PC, since no one is ultimately in total control of the platform.
This also does offer consumers a level of assurance that never existed before. If you're worried about your carrier cutting off features, you know you can root your phone, install open source versions of Android, etc. You can buy a "Google Experience" model like the DROID... carriers are not permitted to mess with those. Ok, sure, Google is, but they have yet to be shown to be interested in locking me out of functions. In fact, they keep handing me new ones, free. No gun to the head... I don't have to use Google Navigation... but it's SO cool...
Technically, you need infinite resolution to record a perfect 22kHz square wave. If you were to record this at 44.1kHz sampling, what you actually get is a pretty nice 22kHz sine wave on playback.
You should mention RIAA equalization, too. Much of the reason you get bass at all on LPs is because, before a perfectly reasonable audio signal is put on an LP, it run through RIAA "pre-emphasis". This cuts bass at 20Hz by 20dB, and boosts treble at 20kHz by 20dB (there's a detailed curve one must follow ). When you play it back, you boost the lows and cut the highs. This keeps the grooves small enough on the LP to allow it to carry its required 22min or so per side, and cuts down on high frequency hiss ... effectively dropping hiss by up to 20dB. Same basic principle as Dolby Noise reduction, in fact. This technique was invented in the mid 1920s at Bell Laboratories.
This was originally an ad-hoc thing, but eventually standardized by the RIAA (and you thought they just sued 13-year-olds for downloading MP3s). A few early CDs were put on disc with the RIAA pre-emphasis applied, and naturally, there was no de-emphasis on playback -- they sounded horrible. Part of the mad rush to get material on CD.. eventually people actually started listening to the masters before they went on disc.
The intent of the gold contacts... ok, well, today, the intent is to make consumers say "oooh... goooollllddddd" and buy these cables, rather than the cables without gold plating for half the price sitting next to them. But the real functional purpose is simple: gold doesn't corrode. Same reason there's gold plating on the fingers of your PCIe cards, or on many types of computer connector.
Of course, the gold is so thin, it's not totally useless, but short-term useful. If you plug once, the gold's still in place, and your connector won't corrode, or react with the metal of the mating connector, assuming that's not also gold. But cycle this a little, and the gold's worn off... and chances are, what's under that gold is something evil.
This is why professional audio cables, as used in studios, on stage, etc. typically have much thicker nickel plating. Not as attractive as gold, but also not corrosive, and it lasts through thousands of plug/unplug cycles.
As far as what's important, you left out impedance... cable impedance is very important, for long runs of cables. When it's a video cable, or worse yet, a digital cable, "short run" is probably a matter of an inch or few... so an impedance matched cable is life of death. Important for audio, too... at least when you have a cable run of a mile or two. Over practical lengths, it's not an issue.. the frequencies are way too low.
Similar is the "skin effect" that audiophiles often yammer about (the wire's self-inductance, plus an AC signal, creates a small reverse current, which manifests as an increased resistance, the deeper one goes into the wire). For a radio signal, a gigabit digital link, etc. it is actually true that most of the signal rides very close to the surface of the connector.. effectively, the cable has more resistance than the gauge would suggest. At audio frequencies, not so much. A 10ft run of 12ga solid wire would have a natural resistance of about 0.033 ohms. At 20kHz, you might see an additional 0.011 ohms... not significant. This drops off with frequency, of course. And it's not terribly likely that anyone's going to use 12ga solid core wire... oh wait, that's the coat-hanger speaker wire again! So there actually is a tiny but measurable difference between that and the [stranded] Monster cable.
Actually, that's an important point -- re-encoding. The device you started out with might be MP3, but most folks change devices more often than they re-rip entire CD collections. I started out with MP3 players. I had an iPod once (it was a gift) that could handle AAC, which certainly has a higher coding efficiency that MP3, so (particularly on a 10GB device) you want the AAC. Then it was a Zune (30GB for $75 was a good deal, and the brown matches my guitar, so I don't want to hear it), and more recently, a DROID, which can play OGG as well as MP3 and AAC.
The one thing I liked about the Zune software... you could let the sync program worry about re-coding whatever you had on HDD for device use. Typically, the device quality is limited by my environment, by ear-buds rather than studio headphones, and by the device itself (particularly that iPod...). So no need for FLAC on the device, yet. Why not encode your audio for the device being used, today.
Only, with lossy compression, you run the risk of greatly reduced sound quality. It was a factor, in early compression algorithms, that they suffered significant generational loss. Early Sony MD players suffered really noticable loss after a few generations (later ATRAC encoders could do dozens without noticable loss). But that's accelerated when you mix and match psychoaccoustic models. So you may find that 64K AAC file made from a 256K OGG or MP3 file is dramatically lower quality than, say, a 64K AAC file made from a 256K AAC file. Keeping around uncompressed audio, no issue here... each transcoding will sound as good as if ripped directly from the original CD.
Oh... also, take a close look at those cable ends. Doesn't that look like nothing more than gold plating over PLASTIC.
This wouldn't be the first time that companies try to fool consumers using the sensibilities of the past. Like trying to convince consumers that better digital cables of any kind will add improved quality to their audio/video. Sorry... it's pass/fail, pretty much.
Hah... yeah, that's a good one! Nearly as funny is that this cable, like all consumer TOSlink (from Toshiba-Link... Toshiba's answer to Sony & Philips RCA-based standard) cables, is made of plastic fiber optic material, not glass. So it's actually pretty lossy over an 8ft run... if your device doesn't have a high gain receiver, you may be getting substantial bit errors. Technically, the interface is supposed to be good for 10m... some devices handle even much longer runs, others, not so much.
You're generally much better off with a decent RCA cable on the old S/PDIF connector. Or a modern HDMI cable, also carrying digital audio over good old copper.
Best Buy, in general, is scamming people on cables, like most A/V resellers. They generally have to compete pretty realistically on price versus other retailers on the main item. I bought an HDTV at Best Buy some years back, after investigating various alternatives, including online ordering from a place like B&H Photo Video (good prices, great service). But even with the lower price, Best Buy came out on top once you factored in postage vs. tax.
But right next to the HDTVs and other gear, they're selling $100+ HDMI cables and all kinds of other interconnects, at prices that would make a Radio Shack guy self-conscious. That's a big profit center for them.. these days, they can make nearly as much profit on selling you accessories as on the TV itself. And the thing is, with those fancy cables, you may not even know what you're getting. What is the gauge of the wire inside? Far, far better off buying cables at a place like monoprice.com.
I think some of this stuff started in good faith, but got to the crazy "Audiophoole" level of today through good, old-fashioned greed.
When you're talking strictly analog signals, there's at least a decent notion that you can do something to somehow improve the quality of the signal, since there's always some loss. So, start with the coat hanger speaker wire, and work up from there... you just MIGHT make something that really, honestly, sounds better. The problem becomes that, next week, you competitor produces a better-still one, complete with full explanation as to why it's better, and they're selling it for more than yours. So you must escalate, and so on.
That, sadly enough, has lead to this fictional world of $10,000 per 10ft speaker cables, with all kinds of science-fiction written up about their special properties (eg, "Any significantly advanced scientific explanation is indistinguishable from a scam"... some of these guys ought to be writing "Star Trek" screen plays... they'd never come up with something as silly as "red matter"). There's so much obvious BS, any lingering truth is hidden.
You know they're intentionally hiding it, to keep their house of cards intact, for the simple reason that none of these guys do double-blind A/B testing. There are, finally, actual Engineering resources showing up to put the Amazing Randi on some of these guys: see http://www.roger-russell.com/wire/wire.htm and http://www.edn.com/blog/980000298/post/830048683.html.
On a more practical point, there's the original recording. Yes, some recordings are made using guitars that cost several thousands, pianos that cost more than your car, mics that run into the thousands, etc. Others, less so. Either way, the wires being used... not so expensive. You might pay $50-$100 for a mic cable, but that's not to get some magical property in the wire other than "copper", but to get a cable that's going to last through the typical abuse of studio or stage. It's hard to believe anyone telling me you need to spend 100x more on wire to play back music than you needed to create it. Plus, you don't find musicians arguing about the "color imparted" by the mic cable, as if discussing a fine wine. Any sound engineer overhead discussing the subtle harmonics or enhanced tonal clarity from the wires between his console and his monitor is either a poser, or he's been in to the brown acid again.
That's interesting, but not unexpected. We tend to favor the familiar, and we tend to adapt to it.
I found myself doing this with digital video. I started working with digital video in the early 1990s, the MPEG-1 days. Anyone who recalls MPEG-1 on VideoCD and early digital satellite TV probably has nightmares about the quality, and rightly so. But part of that wasn't quite so simple... it wasn' t that VCD has any more or less distortion than typical consumer VHS, but -- once you factored out things like bad encoding and all -- it was simply that most of us had spent the last 10-20 years adapting to VHS, which itself was pretty similar to plain old television (a bit worse, but bad in exactly the same ways).
So enter digital, and the early digital was awful.. but a big piece of that was psychological... we hadn't learned to ignore the noise. Over the years, I got more in digital video, got TiVo for my house and worked on various early PVR stuff (satellite receiver recorder with internet and other cool features back in the late 1990s), and before you knew it, I didn't mind TiVo video so much. Sure, I could still the noise, but not all that shabby. But just as interesting, analog video was starting to look much worse to me... I was losing my adaptation to the analog forms of noise.
Since then, of course, I've moved on to HD, first in my own video stuff, then on TV, and now all that older stuff looks like crap.
It's pretty impossible for a generation of listeners to love their iPods and yet not adapt to the sound of the iPod. If that's the normal for you, I do believe you'll in time find the the preferred sound. Which is kind of sad, given that's a drop in quality versus what's attainable... but not necessarily versus what most folks generally listened to.
I mean, most of old farts probably adapted just the same to LP and FM radio, which is one reason CDs were just so fantastic when they first came out. More hidden, we also adjusted to the realities of analog production of the day, various distortions that audio engineers just had to live with... to the extent that, even when they're gone, we're still adding them back in ... tape and tube emulators are among the most popular plug-ins for DAWs. I myself have this wonderful little pre-amp that contains a DSP that can model all kinds of tube and other analog preamps -- works great on mic and guitar, and at least gets me a step closer to that sound I grew up with.
Makes you wonder... will kids today start rejecting any music WITHOUT Auto-tune abuse on it?
If you're mixing down directly from 24-bit to 16-bit MP3, that in itself is probably bad news. Unless you have a fairly smart MP3 encoder, it's probably just truncating the 24-bit samples to 16-bit, which is usually going to result in very noticeable quantization error (essentially, you're magnifying the existing quantization error in the 24-bit audio).
You really want a decent dithering and noise shaping conversion when you drop from 24-bit to 16-bit. I use those built-in on Sound Forge myself, and there are plenty of other audio tools that do this. You're adding what amounts to a small bit of analog noise, in order to effectively remove the very annoying digital noise. There are a variety of different algorithms... if you're not doing this, give it a try. This is a standard function in most modern DAW tools... I've heard even the open source Audacity has dithering for bit depth conversion. Go to mastering plug-ins like iZtope's Ozone for more advanced approaches. You might also the the free dithering plug-in in the MDA pack: http://mda.smartelectronix.com/
I'm not sure the big argument when CDs came out was about the dynamic range (eg, 16-bit vs. something better). You might argue for or against 44.1kHz vs. 48kHz vs. 96kHz or whatever. No one's going to argue that 24-bit is a waste vs. 16-bit... I think it was just the tech limits of the day that estabilished 16-bit as the CD standard. After all, even today, you'll usually find most 24-bit gear is actually 18-22-bit gear with these in there as "marketing bits". Of course, when you're doing a mix from enough sources, that won't matter anyway... you'll have a true 24-bit product when you're done.
The CD controversy has always been around the claim that most people can't hear beyond 20kHz (true), versus the claim that you can't hear the effect of frequencies beyond 20kHz in music (fals). So for CD, they figure you're pretty safe with the Nyquist frequency a bit over twice that (enough to maybe-kinda-sorta allow for good anti-imaging and anti-aliasing filters based on 1980's technology). Obviously they were wrong.. but in reality, no one will EVER need more than today's high-end specs, 24-bit at 192kHz. Trust me on this :-)
I think that's a good way of say it... "the quality is quite good". This comparison tends to suggest so, too, and that's not the first time.
When MP3 was originally created, the main idea was to be able to schlep audio feed for radio broadcast over ISDN lines... 128kb/s digital, rather than some kind of expensive dedicated line. MP3 is clearly "transparent" (which means you can't tell the difference between it and uncompressed) for voice at 128kb/s, and for the purposes of FM radio, just about anything else. Once you're listening direct, yeah, it's not too hard to pick out a 128kb/s MP3 vs. uncompressed, if you know what to listen for. WMA was worse, at least when it first came out... it had really obvious pre-echo... but maybe that's been refined. Certainly, today's LAME in VBR at 256kb/s or so, I can't tell the difference, maybe on some rare tracks where the psychoaccoustic model breaks down. Maybe when I was 18, and had only been to a handfull of rock shows :-)
I think that Amazon has the right idea... high quality MP3, and they do essentially acknowledge it's somewhat less of a thing than a glass mastered CD. Amazon MP3 albums vary from $1.00 below the CD to the occasional $1.99 for the whole album. It's easier to accept something like that... it's theoretically lower quality, I can't likely tell the difference, AND it costs less. Though I'm still going to buy "In the Court of the Crimson King" 40th Anniversary edition with 5.1 mix on DVD-Audio. From analog tape, circa 1969..
Eventually, if there's a "good flaw" from the analog world, someone will add in via a plug-in to the digital world. Things like tape saturation, tube harmonics, etc. are commonplace these days as studio plug-ins... only, with very precise control.
Why are these "good flaws"? Well, that's a fairly involved topic, with different camps. One set of advocates are hardcore adherents -- they'll claim that tape saturation is something ideal, or tube harmonics and saturation, etc. ... versus, say, other types of intentional compressors. Others look at it as a historical thing... we musicians of today grew up listening to classic stuff made on this now-vintage gear, so we're turned into that particular characteristic sound, and want to re-create that.
I'm kind of in both camps... some of this stuff really does sound good, but at least part of why I judge that "good" is because I heard the same effect on a Springsteen or Dylan album I grew up with.
And there are plenty of analog tape artifacts most folks can live without. Most of the time, you don't want the hiss, wow, flutter, or spectral non-linearities you get with tape... you really just want the "good stuff", and then, only when you want it. About the only time I really miss analog tape saturation, for example, is when I'm recording something just a little too hot. In digital, you get those missing peaks -- can't go over 0xFFFF or 0xFFFFFF no matter what you do (assuming unsigned samples, of course). On tape, it would probably sound fine... I just shot a video last weekend, and one mic was just a bit too hot... had to toss that out entirely (fortunately, I had three other mics and a mixer feed to work from).