Broadcast audio and video! Who would have ever thought? I love technology!
Re:Hello Shitty Quality
on
Review: EyeTV
·
· Score: 1
> I have to confess that your point about "it doesn't matter what Mbps the signals are sent to me at" was completely lost on me. Huh?
The data is going through several steps there: MPEG2 Data-> In to TV: First it's uncompressed to digital full-frame RGB values. Then it is converted to an analog signal and displayed. You hooking a capture device up to the RCA video-out on the TV doesn't give you access to the original compressed stream at all- all you see is the analog full-RGB signals. Sorry- I don't know how to explain it any better than that. Use this as an example- imagine that the input stream to the TV was the lowest-bitrate, most blocky and shitty 28.8k stream youve ever seen. If you were capturing the output, you would *still* be capturing the full uncompressed analog frames at the exact same rate- regardless of the original "source" bitrate.
> That is, after all, what we're talking about here. TV signals: 480 lines at 59.94 fields per second. That's all the data you're ever going to get out of an SDTV signal, no matter how you slice it. This particular encoder down-rezzes the picture to half-D1, but that's to comply with the VideoCD spec, not a limitation of USB per se.
Great- but that's not all we're talking about- Modern DV cameras produce resolutions higher than that. Additionally, when reencoding from an analog signal, you want to get the highest resolution possible to make the end result the best as possible. This device is *severely* limited- period.
Re:Hello Shitty Quality
on
Review: EyeTV
·
· Score: 1
You're missing some MAJOR points here-
It doesn't matter what Mbps the signals are sent to me at- The TV is decoding them to full-resolution uncompressed analog video. That's what's going to the input of this thing.
And look at the capture resolution, my webcam does better: (and yes, USB can handle that ridiculous resolution at 30fps just fine, but thats TV quality, and looks really crappy on a computer monitor) FWIW- HDTV and DVDs are at a much higher resolution as well.
"The EyeTV hardware captures at a fixed resolution of 352 by 240 pixels and compresses both video and audio to MPEG-1. The frame rate is full NTSC 29.97 fps. There are two quality settings: These do not change the capture resolution but affect the data capture rate. Standard quality captures at 170 kB/sec. Using this setting the captured video can be burned to a Video CD using Toast Titanium. High quality captures at 340 kB/sec. Both standard quality and high quality recordings can be exported as QuickTime movies."
MPEG1? USB1.1 does not have the bandwidth to capture that at any decent framerate/resolution. In my experience, the only way this can be done is with devices that have build-in MPEG2(dvd) encoding chips, and a USB2/firewire interface. See the Dazzle DVC150 or the Adaptec Vide-oh DVD.
Read what he writes, he's just "some guy" like you and me. And for all his talk, he still doesn't explain 2 very obvious things about this picture:
http://www.hq.nasa.gov/alsj/a11/as11-40-5874.jpg
1) I understand why the flag appears to be waving, but he has no explanation as to why neither the flag nor the flagpole cast any sort of shadow whatsoever. I could buy that because the flagpole is so thin that its shadow gets lost, but as you can see- the flag itself is BRIGHTLY illuminated. This means that you should very easily see a shadow for it.
2) The non-parallel shadows- he explains this in another picture as a low camera angle combined with the perspective of the camera- and you can plainly see that in the other pic it's a valid explanation. There's a lot of distance between the 2 objects. But in this pic, the 3 visible shadows (the lander, the rock, and the astronaut) are *very* close together, and the shadow angles vary *greatly*. You can imagine that the light source would have to be *very* close to make that much of an angle difference between those 3 things.
I dunno- I consider myself fairly sane and rational, and I know the government has done things *far* beyond covering up moon landings- but like someone else said- whether or not the US landed on the moon doesn't change any desicions *I* am going to make- so fuck it:)
>Can anyone give me a non-religious argument why, at some stage in the possibly distant future, that the workings of the brain won't be entirely comprehensible to humans?
Sure-
A system as a whole can not be described using only components of that system itself. Components of a system can not know about the system, otherwise they would be *outside* the system.
The brain will never be able to understand the brain:)
The result is the ability to allow any existing assembly instruction to pull-data-from and write-data-to any alternate register. And this without having to modify or extend the x86 instruction set in any way shape or form.
I propose the introduction of two new 32-bit hardware registers and four new assembly instructions. Simple, isn't it?:)
Not really- that "sensitive to IR" statement applies only to CCD (digital) cameras. And 99% of those already have an IR filter built in, because the images look pink without it. Go ahead and take your webcam apart- find the red piece of glass and remove it. Witness it's newfound ability to see in the near-dark...
Let me quote for your feeble comprehension skills:
If you get FreeBSD 4.7, it is exactly the same as anybody else's FreeBSD 4.7 in terms of included software. There's no RedHat FreeBSD, SuSE FreeBSD, Debian FreeBSD, etc. It's just FreeBSD.
With DIVX5 on it's default quality (2 notches I think?), and a 768kbps clip ripped from DVD, Windows Media Player just *barely* chokes. For super-high-bitrate stuff like that, I use the Divx Player 2.0 (that comes with Divx)- it works perfectly. For everything else at lower bitrates, it works just fine in all cases.
I've got one (A Fujitsu P2040 specifically)- thought I'd chime in-
I love the thing- much more than my old Sony Z505. I like it cause it has good battery life, a DVD *built-in*, and it's tiny. The only thing that attaches with a dongle is the VGA-out. But it is a little *slow*...
In my experience the CPU itself isn't the problem here- for integer CPU-intensive stuff it's plenty fast-
Disk I/O seems to be the sticky bastard. *not necessarily the disk itself tho*
I routinely copy divx movies from my server to it, and it can't keep up with a 100TX-FD network. The CPU jams DPC-time when accessing a large file. I don't know if the problem is with the disk itself just being slow as shit, or if its because the CPU is doing some weird "emulated PIO" or some crap. My tuning has been focused on trying to make it use the disk at little as possible.
Example: if Mozilla has been minimized for a while, and get's paged out- it literally takes 10-15 seconds after clicking at the bottom (Win2k) for it to show up. Oh and if you want some REAL fun try copying 2 large files at a time. Watch the "estimated time" go from 12 minutes to 58 minutes... (That points me at disk-seek?)
The CPU does some other weird things too- I think there's still al ong way to go in the optimization of the code-morphing software. It seems to do some things nice and quick, but others horrendously slow.
Overall, it's fine for me tho- I've got used to and adjusted for the slowness in certain places...
One weird property of light is that it doesn't follow the "baseball" effect-
Meaning- If you're in a car traveling 60mph, and you throw a baseball in that same direction at 60mph. Relative to the earth, the baseball is moving at 120mph. It works the same in reverse.
But light does not behave like this- light behaves more like a propagation through a field. This is why you can't "break the speed of light" simply by firing a laser from something that's moving.
That said, motion of the light-emitter is irrelavent. Once emmited, it propagates at C- Motion of the light *receiver* (us, that telescope) is ENTRELY relevant.
Adjust your calculation to factor in *only* the motion of the earth (expansion in only one axis) and I beleive that will be correct:)
Beyond that- you get all bent when you start thinking about "if light travels as a field-propagation- and if the field is the entire universe- then was the field itself expanding?" "was light *actually* travelling at C, or was it slowed by the field current against it?" "relative to the field it was always travelling at c, but externally we can see the motion of the field..." and then your head explodes.
Then there are 'hollywood' diet plans that offer people the chance to lose gobs of weight in only weeks, but what they don't mention is that you'll gain all that weight back and then some.
Yeesh- all through the article they're pointing out that Win2k has been riddled with security holes-
"Most notable are the Code Red and Nimda worms, which exploit a vulnerability in the operating system."
Last I checked, IIS was not part of the "Win2k OS"-
Doesn't this bother anyone? It's kinda like someon finding holes in apache, or sendmail, or OpenSSH, (all come fairly standard on all distros) and then saying "Multiple security issues found in Linux!"
The idea of hiding a gold nugget in a sea of crap is nothing new.
Except here you can just erase all that crap by just sniffing for TCP on 80- Well gee- now what you gonna do? Fake a slew of web requests so someone sniffing can't see the real ones?
Well right- but I think "eventually" is the key term there-
At some point, you decide that something has enough of a market presence that to switch away from it would be too painful- so you start trying to charge from it.
I dunno- mp3 is THE standard. I would guess even more of a standard than IE as a browser...
But yah- I myself will be converting all my 80gb of mp3s to ogg/vorbis- or... I *would* be- if silly ogg/vorbis would release an mp3-decode/ogg-encode utility...
(look at their website, they specifically are *not* going to do this- and they say exactly why)
Slashdot Mirror
Doesn't anyone think that there *must* be a more palatable word for an online journal than "blog"?
Sounds like something I would see on ratemypoo.com
Broadcast audio and video!
Who would have ever thought?
I love technology!
> I have to confess that your point about "it doesn't matter what Mbps the signals are sent to me at" was completely lost on me. Huh?
The data is going through several steps there:
MPEG2 Data-> In to TV:
First it's uncompressed to digital full-frame RGB values.
Then it is converted to an analog signal and displayed.
You hooking a capture device up to the RCA video-out on the TV doesn't give you access to the original compressed stream at all- all you see is the analog full-RGB signals.
Sorry- I don't know how to explain it any better than that.
Use this as an example- imagine that the input stream to the TV was the lowest-bitrate, most blocky and shitty 28.8k stream youve ever seen.
If you were capturing the output, you would *still* be capturing the full uncompressed analog frames at the exact same rate- regardless of the original "source" bitrate.
> That is, after all, what we're talking about here. TV signals: 480 lines at 59.94 fields per second. That's all the data you're ever going to get out of an SDTV signal, no matter how you slice it. This particular encoder down-rezzes the picture to half-D1, but that's to comply with the VideoCD spec, not a limitation of USB per se.
Great- but that's not all we're talking about-
Modern DV cameras produce resolutions higher than that.
Additionally, when reencoding from an analog signal, you want to get the highest resolution possible to make the end result the best as possible.
This device is *severely* limited- period.
You're missing some MAJOR points here-
It doesn't matter what Mbps the signals are sent to me at-
The TV is decoding them to full-resolution uncompressed analog video.
That's what's going to the input of this thing.
And look at the capture resolution, my webcam does better:
(and yes, USB can handle that ridiculous resolution at 30fps just fine, but thats TV quality, and looks really crappy on a computer monitor)
FWIW- HDTV and DVDs are at a much higher resolution as well.
"The EyeTV hardware captures at a fixed resolution of 352 by 240 pixels and compresses both video and audio to MPEG-1. The frame rate is full NTSC 29.97 fps. There are two quality settings: These do not change the capture resolution but affect the data capture rate. Standard quality captures at 170 kB/sec. Using this setting the captured video can be burned to a Video CD using Toast Titanium. High quality captures at 340 kB/sec. Both standard quality and high quality recordings can be exported as QuickTime movies."
MPEG1?
USB1.1 does not have the bandwidth to capture that at any decent framerate/resolution.
In my experience, the only way this can be done is with devices that have build-in MPEG2(dvd) encoding chips, and a USB2/firewire interface.
See the Dazzle DVC150 or the Adaptec Vide-oh DVD.
Read what he writes, he's just "some guy" like you and me.
g
:)
And for all his talk, he still doesn't explain 2 very obvious things about this picture:
http://www.hq.nasa.gov/alsj/a11/as11-40-5874.jp
1) I understand why the flag appears to be waving, but he has no explanation as to why neither the flag nor the flagpole cast any sort of shadow whatsoever.
I could buy that because the flagpole is so thin that its shadow gets lost, but as you can see- the flag itself is BRIGHTLY illuminated.
This means that you should very easily see a shadow for it.
2) The non-parallel shadows- he explains this in another picture as a low camera angle combined with the perspective of the camera- and you can plainly see that in the other pic it's a valid explanation.
There's a lot of distance between the 2 objects.
But in this pic, the 3 visible shadows (the lander, the rock, and the astronaut) are *very* close together, and the shadow angles vary *greatly*.
You can imagine that the light source would have to be *very* close to make that much of an angle difference between those 3 things.
I dunno- I consider myself fairly sane and rational, and I know the government has done things *far* beyond covering up moon landings- but like someone else said- whether or not the US landed on the moon doesn't change any desicions *I* am going to make- so fuck it
Yeah... enter jboss
Suddenly the ratio swings-
Lats face it, intitial software cost is still a *large* part of TCO...
>Can anyone give me a non-religious argument why, at some stage in the possibly distant future, that the workings of the brain won't be entirely comprehensible to humans?
:)
Sure-
A system as a whole can not be described using only components of that system itself.
Components of a system can not know about the system, otherwise they would be *outside* the system.
The brain will never be able to understand the brain
The result is the ability to allow any existing assembly instruction to pull-data-from and write-data-to any alternate register. And this without having to modify or extend the x86 instruction set in any way shape or form.
:)
I propose the introduction of two new 32-bit hardware registers and four new assembly instructions. Simple, isn't it?
5-Interesting?
Not really- that "sensitive to IR" statement applies only to CCD (digital) cameras.
And 99% of those already have an IR filter built in, because the images look pink without it.
Go ahead and take your webcam apart- find the red piece of glass and remove it.
Witness it's newfound ability to see in the near-dark...
Did you not read what was written?
Let me quote for your feeble comprehension skills:
If you get FreeBSD 4.7, it is exactly the same as anybody else's FreeBSD 4.7 in terms of included software. There's no RedHat FreeBSD, SuSE FreeBSD, Debian FreeBSD, etc. It's just FreeBSD.
"FreeBSD" != "BSD"
Without a doubt, yes.
With DIVX5 on it's default quality (2 notches I think?), and a 768kbps clip ripped from DVD, Windows Media Player just *barely* chokes.
For super-high-bitrate stuff like that, I use the Divx Player 2.0 (that comes with Divx)- it works perfectly.
For everything else at lower bitrates, it works just fine in all cases.
Tell me you weren't making orgasmic moaning sound effects after going here and clicking "action" after rotating around to the sides... :D
Hey-
I've got one (A Fujitsu P2040 specifically)- thought I'd chime in-
I love the thing- much more than my old Sony Z505.
I like it cause it has good battery life, a DVD *built-in*, and it's tiny.
The only thing that attaches with a dongle is the VGA-out.
But it is a little *slow*...
In my experience the CPU itself isn't the problem here- for integer CPU-intensive stuff it's plenty fast-
Disk I/O seems to be the sticky bastard.
*not necessarily the disk itself tho*
I routinely copy divx movies from my server to it, and it can't keep up with a 100TX-FD network.
The CPU jams DPC-time when accessing a large file.
I don't know if the problem is with the disk itself just being slow as shit, or if its because the CPU is doing some weird "emulated PIO" or some crap.
My tuning has been focused on trying to make it use the disk at little as possible.
Example: if Mozilla has been minimized for a while, and get's paged out- it literally takes 10-15 seconds after clicking at the bottom (Win2k) for it to show up.
Oh and if you want some REAL fun try copying 2 large files at a time.
Watch the "estimated time" go from 12 minutes to 58 minutes...
(That points me at disk-seek?)
The CPU does some other weird things too- I think there's still al ong way to go in the optimization of the code-morphing software.
It seems to do some things nice and quick, but others horrendously slow.
Overall, it's fine for me tho- I've got used to and adjusted for the slowness in certain places...
...subject says it all...
...that one day RMS is going to snap.
He's just the type- intelligient, but edgy, and living in his own world.
Mark my words, one of these days he's going to let loose on a city bus with an AK.
That's not quite correct-
:)
One weird property of light is that it doesn't follow the "baseball" effect-
Meaning- If you're in a car traveling 60mph, and you throw a baseball in that same direction at 60mph.
Relative to the earth, the baseball is moving at 120mph.
It works the same in reverse.
But light does not behave like this-
light behaves more like a propagation through a field.
This is why you can't "break the speed of light" simply by firing a laser from something that's moving.
That said, motion of the light-emitter is irrelavent.
Once emmited, it propagates at C-
Motion of the light *receiver* (us, that telescope) is ENTRELY relevant.
Adjust your calculation to factor in *only* the motion of the earth (expansion in only one axis) and I beleive that will be correct
Beyond that- you get all bent when you start thinking about "if light travels as a field-propagation- and if the field is the entire universe- then was the field itself expanding?"
"was light *actually* travelling at C, or was it slowed by the field current against it?"
"relative to the field it was always travelling at c, but externally we can see the motion of the field..." and then your head explodes.
Then there are 'hollywood' diet plans that offer people the chance to lose gobs of weight in only weeks, but what they don't mention is that you'll gain all that weight back and then some.
Not with Jimmy Tango's Fatbusters
http://www.amazon.com/exec/obidos/tg/detail/-/0964 350416/qid=1031612316/sr=8-2/ref=sr_8_2/002-205001 3-9568050?v=glance&s=books&n=507846
:)
It's a Japanese textbook translated in to English-
As such- you get a bunch of goofy Japanese cartoons in there too
Yeesh- all through the article they're pointing out that Win2k has been riddled with security holes-
"Most notable are the Code Red and Nimda worms, which exploit a vulnerability in the operating system."
Last I checked, IIS was not part of the "Win2k OS"-
Doesn't this bother anyone?
It's kinda like someon finding holes in apache, or sendmail, or OpenSSH, (all come fairly standard on all distros) and then saying "Multiple security issues found in Linux!"
hahah sweet name :)
Innovative? not really...
The idea of hiding a gold nugget in a sea of crap is nothing new.
Except here you can just erase all that crap by just sniffing for TCP on 80-
Well gee- now what you gonna do?
Fake a slew of web requests so someone sniffing can't see the real ones?
yeesh...
Well right- but I think "eventually" is the key term there-
At some point, you decide that something has enough of a market presence that to switch away from it would be too painful- so you start trying to charge from it.
I dunno- mp3 is THE standard.
I would guess even more of a standard than IE as a browser...
But yah- I myself will be converting all my 80gb of mp3s to ogg/vorbis-
or... I *would* be- if silly ogg/vorbis would release an mp3-decode/ogg-encode utility...
(look at their website, they specifically are *not* going to do this- and they say exactly why)
duh
http://www.vorbis.com/
Hahaha take a look at your memory usage when jre1.4 is running :)
And speedwise it still can't compete with the MS JVM-
And oh- go to a lovely web page with any flash or shockwave animation.
Open up multiple windows of that same page.
Watch your CPU% scream.
Then do the same in IE.