Wotao popped up (long ago) on the KS comments thread and basically said "I'm an investor but I can't really talk about anything". He also replied to direct emails to his uni email, confirming his involvement. One person who contacted him, reported him (not seen the email) as admitting he's not yet seen a working device.
I think he probably got pulled in as an investor (and to add credibility) and probably now rues the day he said yes. We'll see. Either way, at this point Google searches won't forget his involvement.
The thing also needs (most of its) power to run a Bluetooth receiver (+accelerometer), which typically uses 10-15ma @ 2V (=20-30mw) when in _receive_ mode. The duty cycle is short but waking up every 5 seconds to see if the tag is being pinged isn't typically something you'd power with leprechauns (unless they were made of lithium).
Wetag's? Sure, they're pretty sketchy. Practically no hard info except two wack pseudo-science documents and some nice photoshop work. No pics, no demo, no nothing (so far).
Mine? Errr... well, there's a ton of links to source material throughout, there's links to other people who think the same way (and did the math also), there's facts, figures, references, etc. If you didn't click any of the links, do any Google searches, look at the iFind site or pay attention in math or physics class in highschool, then.. yes I suppose you find them unverifiable. Tell you what, I promise, cross my heart, it's all true. Does that work?
What you say is not plausible; how do you think any radio-activated device receives the "I lost you" signal? In practice it's far from "dormant 24/7"; it's completely deaf to any signal until it turns on its high gain radio which takes typically 10-15ma. The tag must be constantly waking up and consuming power (even if, as with BTLE, it's possible to set the gaps to e.g. 5-10seconds and the "on" time is very short).
Without trying to be unduly rude, you're 25% informed yet 75% confident in your knowledge, which I suspect comprises large portion of backers of this project.
Either do a little research (e.g. look up the math and/or read the datasheets for BTLE chips - all provided in the RTFA's google doc) or... go order a pack of 10 iFinds. .:-)
It's not (just) for speed, if at all, it's BECAUSE YOU HAVE PORTABILITY WITH iOS (and other platforms) If you use Java you're hosed. If you use regular C you can compile on both platforms, with a shim to interface to either iOS or Android as required. The GLES code can easily be compatible. The UI stuff not so much but (high end) games generally implement their own UI in GL for specifically this reason.
It's not pretty but it's what most pro game developers have been doing since at least 2010, and it's a _hell_ of a lot better than having to totally rewrite your Java app in ObjectiveC or vice versa.
The extra performance is sometimes useful in some places but it's almost always about compatibility with iOS or (more rarely) with existing C libraries e.g. video encoding or whatever.
You people are weird, I've been on 4x 1920x1200 monitors for years now (2x portrait, 2x landscape; It's fucking marvellous). I have a fifth but no desk space (or interest in turning head to view) so it's a spare. This cost... I dunno, peanuts, even a few years back. Super strongly recommend at _least_ a pair of portrait monitors plus at least one landscape. It's practically impossible to go back.
...you say that like that's a bad thing? This is Vegas! Just make sure you have some potential customers with you and it's your card behind the bar (or in the stripper's hand, etc)
My scope HELL yes I use it! It's crappy - analog, 30Mhz(!!) and it cost about $60 on ebay but it's still very useful for all sorts of "WTF is going on". Wouldn't be without a scope for anything.
MUST-HAVE'S: ** Dual-output bench DC PSU (Instek) - can't BELIEVE I ever lived without one - what was I thinking? Ass-saving to have current display & esp current limiter(!) **
Logic Analyzer is an Intronix Logicport which does up to 500Mhz and has adjustable threshold voltage which is required for 3v3 / 1v2 designs; I have a Saleae16 but it's much less useful IMO; recommend to everyone to get a Logicport as well.
Fluke287 multimeter which is expensive for a multimeter but does microamp current measurements which you need for modern battery powered stuff (just don't blow the fu$e$ cos they ain't cheap)
Signal/protocol generator - I made my own! (a $12 Cypress FX2 board plus $6 AD9850 clock generator = output any protocol you can imagine on up to 16 channels @ 0-25Mhz with 0.001hz programmable clock resolution, super fucking handy)
Beagle USB protocol analyzer (although software-only ones are useful too)
Cheap chinese iron+hot-air combo (Kendal 852D) which is great
Love my FTDI quad-UART boards! - many things have multiple CPUs to keep track of nowadays. FTDI FTW...and whatever programmer/JTAG/etc the current CPU needs.
I find it very useful to run the logic analyzer and other USB tools on a separate PC & monitor to my "work" PC rig; makes them more like standalone tools.
...then they could use directional antennas on cell base-stations and divide up each cell into slices!! Hey they could call them "sector antennas" and... oh never mind.
Java? Casual? That's like saying the US Tax code is good bed-time reading. After realizing I was spending half my frickin' life compiling, reloading, and waiting... waiting... (I'm looking at _you_ Tomcat) I switched to Python and never looked back.
"Touch surfaces don't want to be vertical" absolutely completely obviously true to anyone who's used one for any length of time. After 15 seconds you're already thinking "uh oh" and after five minutes it become torture.
I hear Chuck Norris uses one while hanging completely inverted suspended only by his toenails.
"Professional" mobile games (i.e. by commercial dev companies) are almost universally written in straight C/C++ with minimal ObjectiveC / Dalvik wrappers to get to the phone hardware. If you have a hit title, do you -really- want to have to rewrite the whole thing from top to bottom to port it to other platforms?
I spent several months a few years back working hard to convince my employer (a certain US carrier) that going ahead and launching a J2ME-based mobile platform (in the last 00's - this is post-iPhone, people) was would elicit nothing more than mockery (and, at best, shovelware) from the developer community. My employer subsequently canned the idea, and I like to think that my steely knives helped kill the beast. My main argument was that forcing developers to rewrite significant portions of code almost guarantees you won't get major titles, regardless of your hardware lineup.
One of the smartest things Google did with Android was the NDK; I recently ported a top-10 iPhone 3d game (written 99% in straight C/++) to Android NDK and including my getting-to-know-you time I was done in 3 weeks. Was scorchingly fast on the Galaxy Tab compared to iPad.
The frank reality is that iOS is very obviously the largest mobile platform for developers, and others (Android, WP7, WebOS etc) must make it as easy as possible to port titles over. Google did a marvellous job of adding this capability; NDK gives you plenty enough bare metal to port easily from other platforms. I've not looked at WebOS;-) but it appears they were smart enough to provide a plain-vanilla C++ and OGLES environment for games.
Android and iPhone can handle running native code apps just fine. If WP7 can't make itself a viable (easy!) porting target like Android, it's going to be spending a lot of Saturday nights at home watching TV waiting for the phone to ring.
Glad you like the idea, tho I strongly doubt I'm the first person to think of it.
Additional (somewhat crazy) suggestions for faster swap speed; 1) Reserve two extra specific Hibernate-space partitions (one OSX, one Win) on the HD and keep them completely defragged and unused for anything else etc. 2) Some kind of insane ramdisk driver solution where you reserve say 512MB from the OS at boot, use it as a compressed ramdisk to hibernate to. Getting the necessary 'reload hibernate file from my ramdisk' driver to be available to the OSes at the right point would probaby be heinous, I've no idea. This is silly solution but would be speedytastic. 3) USB Flashdrives are kewl and handy dandy but slower than HD. (28MB/s read, 15MB/s write for fast ones)
It seems that one might well be able to subvert the 'standby' or 'sleep' modes of both OS's to provide fast OS switching; hit a key to get the system to slumber (i.e. save system state) then add a 'system state swap' hack where you can switch over into the slumber mode of the other OS and reawaken.
AFAIK both OS's have both 'light standby' and sleep modes, presumably sleep involves swapping the ram out to disk and even reinitialising hardware on wake, so may just be the ticket.
If this can be made to work and tweaked for speed it would seem that you'd be able to ALT-TAB between OS's with a sub-10 second delay. That'd do for me.
That is a dumb reply. The asserts() make things stop instantly when the test suite touches a sore spot, hence making debugging much simpler. They are supposed to make things break, that's the point! - and break in a clean, logged, maximally reproducible way. Essentially they just indicate that things have already broken, and make it as obvious as possible to everyone, including the test harness (that does not have the benefit of such inside information as to the possible brokenness of things).
FUD? WTF? See the thing in my post that starts with "http://"... that's part of a secret code that only me and my brotherhood the Justified Ancients Of Mu will understand. It communicates to them - and them alone - the secrets of which I speak.
You know what it is, this Quantum stuff; scientists have finally discovered a pointer error in physics.
From the article: These atoms were each spinning clockwise and counterclockwise at the same time. Moreover, like miniature Rockettes they were all doing whatever it was they were doing together, in perfect synchrony. Should one of them realize, like the cartoon character who runs off a cliff and doesn't fall until he looks down, that it is in a metaphysically untenable situation and decide to spin only one way, the rest would instantly fall in line, whether they were across a test tube or across the galaxy.
Any halfway competent C programmer can easily see this is simple pointer aliasing. Physics was clearly written in C++ - albeit with a very high precision floating point library. What is happening is that, to save memory on the galaxy, most of the different atoms we perceive are actually just the same one, aliased using pointers. There is some neat code in Physics.cpp which detects when an atom is modified and makes a mutable copy as required. Clearly in this case (with the atom, cat, whatever) something is fux0red in the code and it's not making a copy; hence modifying one atom modifies the perceived value of several. Fortunately as soon as a human observes it, atom->View() is called, and a stable copy of the atom is created and from then on the bug kinda disappears (all the atoms go about their business as normal).
As a bug, it probably got noticed in beta, but was considered low priorty, however now there's such a fuss about it, I'd expect an online patch to stop the scientist hax0rs exploiting "the Quantum effect" any millenium now, so don't go writing it into your world view.
(Incidentally, this is why people die, it's to avoid problems with them knowing too much and causing stack overflows, but that's another story)
Slashdot Mirror
Wotao popped up (long ago) on the KS comments thread and basically said "I'm an investor but I can't really talk about anything". He also replied to direct emails to his uni email, confirming his involvement. One person who contacted him, reported him (not seen the email) as admitting he's not yet seen a working device.
I think he probably got pulled in as an investor (and to add credibility) and probably now rues the day he said yes. We'll see. Either way, at this point Google searches won't forget his involvement.
The thing also needs (most of its) power to run a Bluetooth receiver (+accelerometer), which typically uses 10-15ma @ 2V (=20-30mw) when in _receive_ mode. The duty cycle is short but waking up every 5 seconds to see if the tag is being pinged isn't typically something you'd power with leprechauns (unless they were made of lithium).
err... whose claims are unverifiable?
Wetag's? Sure, they're pretty sketchy. Practically no hard info except two wack pseudo-science documents and some nice photoshop work. No pics, no demo, no nothing (so far).
Mine? Errr... well, there's a ton of links to source material throughout, there's links to other people who think the same way (and did the math also), there's facts, figures, references, etc.
If you didn't click any of the links, do any Google searches, look at the iFind site or pay attention in math or physics class in highschool, then.. yes I suppose you find them unverifiable. Tell you what, I promise, cross my heart, it's all true. Does that work?
What you say is not plausible; how do you think any radio-activated device receives the "I lost you" signal? In practice it's far from "dormant 24/7"; it's completely deaf to any signal until it turns on its high gain radio which takes typically 10-15ma. The tag must be constantly waking up and consuming power (even if, as with BTLE, it's possible to set the gaps to e.g. 5-10seconds and the "on" time is very short).
Without trying to be unduly rude, you're 25% informed yet 75% confident in your knowledge, which I suspect comprises large portion of backers of this project.
Either do a little research (e.g. look up the math and/or read the datasheets for BTLE chips - all provided in the RTFA's google doc) or... go order a pack of 10 iFinds. . :-)
I did, see Google doc. So did others (better than me). Results posted in doc.
You say: "no idea, did not check it".
Your ideas are intriguing to me and I wish to subscribe to your newsletter.
It's not (just) for speed, if at all, it's BECAUSE YOU HAVE PORTABILITY WITH iOS (and other platforms)
If you use Java you're hosed. If you use regular C you can compile on both platforms, with a shim to interface to either iOS or Android as required.
The GLES code can easily be compatible. The UI stuff not so much but (high end) games generally implement their own UI in GL for specifically this reason.
It's not pretty but it's what most pro game developers have been doing since at least 2010, and it's a _hell_ of a lot better than having to totally rewrite your Java app in ObjectiveC or vice versa.
The extra performance is sometimes useful in some places but it's almost always about compatibility with iOS or (more rarely) with existing C libraries e.g. video encoding or whatever.
...in which case a Google Bus is going to appear out of nowhere and park between you and the moon, completely blocking your view.
You people are weird, I've been on 4x 1920x1200 monitors for years now (2x portrait, 2x landscape; It's fucking marvellous). I have a fifth but no desk space (or interest in turning head to view) so it's a spare.
This cost... I dunno, peanuts, even a few years back.
Super strongly recommend at _least_ a pair of portrait monitors plus at least one landscape. It's practically impossible to go back.
...you say that like that's a bad thing? This is Vegas!
Just make sure you have some potential customers with you and it's your card behind the bar (or in the stripper's hand, etc)
My scope HELL yes I use it! It's crappy - analog, 30Mhz(!!) and it cost about $60 on ebay but it's still very useful for all sorts of "WTF is going on". Wouldn't be without a scope for anything.
MUST-HAVE'S:
** Dual-output bench DC PSU (Instek) - can't BELIEVE I ever lived without one - what was I thinking? Ass-saving to have current display & esp current limiter(!) **
Logic Analyzer is an Intronix Logicport which does up to 500Mhz and has adjustable threshold voltage which is required for 3v3 / 1v2 designs; I have a Saleae16 but it's much less useful IMO; recommend to everyone to get a Logicport as well.
Fluke287 multimeter which is expensive for a multimeter but does microamp current measurements which you need for modern battery powered stuff (just don't blow the fu$e$ cos they ain't cheap)
Signal/protocol generator - I made my own! (a $12 Cypress FX2 board plus $6 AD9850 clock generator = output any protocol you can imagine on up to 16 channels @ 0-25Mhz with 0.001hz programmable clock resolution, super fucking handy)
Beagle USB protocol analyzer (although software-only ones are useful too)
Cheap chinese iron+hot-air combo (Kendal 852D) which is great
Love my FTDI quad-UART boards! - many things have multiple CPUs to keep track of nowadays. FTDI FTW. ..and whatever programmer/JTAG/etc the current CPU needs.
I find it very useful to run the logic analyzer and other USB tools on a separate PC & monitor to my "work" PC rig; makes them more like standalone tools.
...then they could use directional antennas on cell base-stations and divide up each cell into slices!!
Hey they could call them "sector antennas" and... oh never mind.
You picked Flex _and_ PHP as winners? Damn dude, don't ever go to Vegas.
Java? Casual? That's like saying the US Tax code is good bed-time reading.
After realizing I was spending half my frickin' life compiling, reloading, and waiting... waiting... (I'm looking at _you_ Tomcat) I switched to Python and never looked back.
This is why the ./ guest account is called 'anonymous coward', of course.
"Touch surfaces don't want to be vertical" absolutely completely obviously true to anyone who's used one for any length of time.
After 15 seconds you're already thinking "uh oh" and after five minutes it become torture.
I hear Chuck Norris uses one while hanging completely inverted suspended only by his toenails.
"Professional" mobile games (i.e. by commercial dev companies) are almost universally written in straight C/C++ with minimal ObjectiveC / Dalvik wrappers to get to the phone hardware.
If you have a hit title, do you -really- want to have to rewrite the whole thing from top to bottom to port it to other platforms?
I spent several months a few years back working hard to convince my employer (a certain US carrier) that going ahead and launching a J2ME-based mobile platform (in the last 00's - this is post-iPhone, people) was would elicit nothing more than mockery (and, at best, shovelware) from the developer community. My employer subsequently canned the idea, and I like to think that my steely knives helped kill the beast.
My main argument was that forcing developers to rewrite significant portions of code almost guarantees you won't get major titles, regardless of your hardware lineup.
One of the smartest things Google did with Android was the NDK; I recently ported a top-10 iPhone 3d game (written 99% in straight C/++) to Android NDK and including my getting-to-know-you time I was done in 3 weeks. Was scorchingly fast on the Galaxy Tab compared to iPad.
The frank reality is that iOS is very obviously the largest mobile platform for developers, and others (Android, WP7, WebOS etc) must make it as easy as possible to port titles over. ;-) but it appears they were smart enough to provide a plain-vanilla C++ and OGLES environment for games.
Google did a marvellous job of adding this capability; NDK gives you plenty enough bare metal to port easily from other platforms.
I've not looked at WebOS
Android and iPhone can handle running native code apps just fine. If WP7 can't make itself a viable (easy!) porting target like Android, it's going to be spending a lot of Saturday nights at home watching TV waiting for the phone to ring.
http://www.msnbc.msn.com/id/39516346/ns/us_news-life/
Skip to 5:20 in.
Gene Cranick (interviewed sitting in front of his incinerated house) is politely presented faced with many of the above arguments. ...comedy genius
The interviewer does his best to be nice... you can actually hear his brain start to boil at 8:50
He went to *Israel* and offered them dishwater-grade intelligence?
Well it saves Akamai the hassle of doing staff intelligence tests.
for a Burning Man project, using a spherical screen and a 6-channel sound system;
http://frickinlaserbeams.com
http://www.youtube.com/watch?v=-5muJOcAd7c
Various construction vids at
http://www.youtube.com/view_play_list?p=EAEC79EF55D409D2
Works great but green's really where it's at. Blue and Red are much, much dimmer and have shorter persistence.
Alas my shit was not together in time to work properly on the playa... I'll be back this year with the real thing...
Glad you like the idea, tho I strongly doubt I'm the first person to think of it.
Additional (somewhat crazy) suggestions for faster swap speed;
1) Reserve two extra specific Hibernate-space partitions (one OSX, one Win) on the HD and keep them completely defragged and unused for anything else etc.
2) Some kind of insane ramdisk driver solution where you reserve say 512MB from the OS at boot, use it as a compressed ramdisk to hibernate to. Getting the necessary 'reload hibernate file from my ramdisk' driver to be available to the OSes at the right point would probaby be heinous, I've no idea. This is silly solution but would be speedytastic.
3) USB Flashdrives are kewl and handy dandy but slower than HD. (28MB/s read, 15MB/s write for fast ones)
FL
It seems that one might well be able to subvert the 'standby' or 'sleep' modes of both OS's to provide fast OS switching; hit a key to get the system to slumber (i.e. save system state) then add a 'system state swap' hack where you can switch over into the slumber mode of the other OS and reawaken.
AFAIK both OS's have both 'light standby' and sleep modes, presumably sleep involves swapping the ram out to disk and even reinitialising hardware on wake, so may just be the ticket.
If this can be made to work and tweaked for speed it would seem that you'd be able to ALT-TAB between OS's with a sub-10 second delay. That'd do for me.
Hope so!
FL
That is a dumb reply. The asserts() make things stop instantly when the test suite touches a sore spot, hence making debugging much simpler.
They are supposed to make things break, that's the point! - and break in a clean, logged, maximally reproducible way. Essentially they just indicate that things have already broken, and make it as obvious as possible to everyone, including the test harness (that does not have the benefit of such inside information as to the possible brokenness of things).
FUD? WTF? See the thing in my post that starts with "http://" ... that's part of a secret code that only me and my brotherhood the Justified Ancients Of Mu will understand. It communicates to them - and them alone - the secrets of which I speak.
Thanks man, that is exactly the kinda concrete answer that was needed to lower the noise level. ;-)
From the article:
These atoms were each spinning clockwise and counterclockwise at the same time. Moreover, like miniature Rockettes they were all doing whatever it was they were doing together, in perfect synchrony. Should one of them realize, like the cartoon character who runs off a cliff and doesn't fall until he looks down, that it is in a metaphysically untenable situation and decide to spin only one way, the rest would instantly fall in line, whether they were across a test tube or across the galaxy.
Any halfway competent C programmer can easily see this is simple pointer aliasing. Physics was clearly written in C++ - albeit with a very high precision floating point library. What is happening is that, to save memory on the galaxy, most of the different atoms we perceive are actually just the same one, aliased using pointers. There is some neat code in Physics.cpp which detects when an atom is modified and makes a mutable copy as required. Clearly in this case (with the atom, cat, whatever) something is fux0red in the code and it's not making a copy; hence modifying one atom modifies the perceived value of several. Fortunately as soon as a human observes it, atom->View() is called, and a stable copy of the atom is created and from then on the bug kinda disappears (all the atoms go about their business as normal).
As a bug, it probably got noticed in beta, but was considered low priorty, however now there's such a fuss about it, I'd expect an online patch to stop the scientist hax0rs exploiting "the Quantum effect" any millenium now, so don't go writing it into your world view.
(Incidentally, this is why people die, it's to avoid problems with them knowing too much and causing stack overflows, but that's another story)