I've read novels on my palm for years. It's particulary useful for Neal Stephenson books which are enormous.
www.ereader.com formerly known as palmdigitalmedia is where I get all my ebooks.
I also get normal files and convert them to the palm media format and use their ebook converter under wine on linux (DropBook).
I find my T3 + an SD card a great way to read books. I can bookmark pages, highlight sections of text and lookup words using whatever dictionary.
And I don't mind good DRM. It means I can get new release books on my palm without waiting too long (they can become available when the hardback comes out).
However - I'd like ereader on linux without using wine. Hmm.
AFAICT it works not by detecting pressure but by seeing how much of the finger is pressing down on the pad. As you increase pressure - more of your finger is squished into the pad.
This is just the way it was in the old days of serial terminals which always had a control key under your left little finger. This is of course the natural thing for emacs.
DVD+RW and DVD+R are formats that were never supposed to be supported in consumer players.
Gasically this is a beta vs vhs thing were the proponents of dvd+r and dvd+rw don't want to pay the licensing fees to the dvd-r guys.
DVD-R is the most compatible format.
DVD-RAM is the best for rewrites and so on. Completely outstrips dvd+rw for performance and longevity and has in fact been around for ages in jukeboxes etc.
They have renamed it to the Cray X1 (SV2 was the code name when SGI owned cray - in fact the X1 uses the same technology routers and internode interconnects used in the Origin 3000's and Origin 300s's)
This page has the video describing it (look for the tech overview) in detail.
This press release just talks about it. Look at the 350lbs of force required to mate the MCM to the board.
Re:This is a software solution
on
IBM's Deep View
·
· Score: 1
The 'SGE' is similar to SGI's compositor which is a product you can buy off the shelf which composits the output of multiple graphics cards in hardware.
The IBM research device does some other things that the SGI IP compositor doesn't - but at least the SGI compositor has been available for around a year - so you can use it today.
Not one rack of 8 PCs and one half rack of some graphics engine.
It was drawing the full display at about 30fps.
It was as easy to use as any other workstation rather than the 'interesting' mix of 8 pcs in a rack and some other half rack of graphics stuff.
The hardest thing was to read the tiny little fonts on the screen (the display is 200dpi IIRC) - you need a magnifying glass (or of course you could increase the font size).
The Octane2 can do this because you can install two V12 graphics engines each with a dual channel adapter.
http://www.sgi.com/workstations/octane2/dual_cha nn el.html
SGI software stiches the cards together transparently.
It is a beautiful display.
The amuzing/annoying thing about these sorts of announcements is that customers have been using SGI stuff to do this for the last 8 years or so.
If you were on the leading edge of this kind of work would you wait that long for some kludged together solution which might work if you have enough Duct tape to stick it together?
Or would you pay the extra cash for a solution that works and gives you a huge jump on your competition.
The sort of software layer that can be used to make these four channels (two channels from each graphics card) into one display is stuff like:
http://www.sgi.com/software/multipipe/sdk/
Oh yeah - and re your 'every table comment'. The Octane2 fits on one table - it doesn't surround the table like the IBM stuff.
I expect McKinley to be the best price/performance FLOATING POINT chip out there when it is shipped.
In the scientific market integer performance isn't the problem it is floating point.
IPF (Itanium Product Family) is specifically targetted to do very good floating point. The first generation Itanium in fact had terrible integer performance but very good floating point performance.
There will be no comparison between a PIV and McKinley when doing things like floating point matrix multiplies etc etc. This is the important measurement when doing scientific computing.
AMD will not be able to touch the floating point performance of McKinley - even hammer. They are aimed at different markets.
So... there are two reasons for McKinley - 64bit and excellent floating point performance.
And don't forget Sun's cheapest version of this workstation doesn't have the 10/10/10/8 bit colour card - it has a much worse one (the PGX). So be careful comparing prices.
And then there are the hardware accelerated image processing functions (ie convolutions) in the V10/V12 SGI graphics which the Sun card doesn't seem to have.
From what I saw it wasn't an antialiasing problem in the traditional sense. It was an aliasing problem in that the depth sensor they use has a low spatial resolution.
The demo I saw that was live used the depth as a key - but it wasn't used to blend the objects together, it was used to make a simple visible/not visible decision. What you are describing is using the depth key to vary alpha which would be interesting to look at.
One VERY cool demo they did was to fly a CG character through the arms of the live talent. Where the CG object and the real object intersected you could clearly see blocky edges.
At the time I thought the technology (at least in a live setting) was maybe good enough for a young kids show were you want high speed production and the production values don't matter too much. Once they get the depth sensor up to broadcast the resolution however it will be a VERY nice live tool. (Maybe they have already done this - I haven't seen it in two years and they must have improved).
Not new technology - saw this at NAB in 2000
on
Video with Depth
·
· Score: 1
NAB is the National Association of Broadcasters conference. The ZCAM was was being demoed then.
In the demos they had realtime keying so they could fly a 3D CGI character in front of and behind the live talent. There was only about a 40ms delay. This is impossible with normal keying (ie blue/green screen). (You can only put stuff behind the talent).
It's biggest limitation was the resolution of the 3D sensor was low - so you had rough edges (think jaggies).
They also demonstrated a 3D Realplayer and 3D Windows Media players (which you watched with stereo shutter glasses). These players were called 'deep players'. Pretty cool but definitely not new.
What you described was demonstrated with this tech at NAB in 2000. In the demo they had the camera with the audience wearing polarized glasses (which I kept).
I bought an SBLive on the promise that creative can upgrade the firmware etc etc.
In the entire time I have owned the card creative never released any new drivers for it at all.
I use amd and via chipsets and have had many problems with the card across two different motherboards (a7v and msi 6380ru). Everytime I had to shuffle the card around because the IRQ sharing wouldn't work properly.
I finally gave up when max payne would randomly crash and I swapped out my sblive for my good old trusty Diamond MX300 card and never had a problem since.
I then upgraded to a hercules game theatre XP and this has confirmed I will never buy creative again.
Hercules frequently updates their drivers and they add new features all the time. This is what creative promised and never delivered.
The hercules card is noiseless - unlike my sblive, and was half the price of a sblive platinum (which has a comparable feature set).
I almost gave into the evil creative and the audigy but it was twice the price and considering creatives track record of never continuing to release new drivers for their old cards I gave up.
I now have a totally stable system for the six games I am currently addicted to (Alien vs Predator 2, Castle Wolfenstein, Max Payne (I am replaying it at a high difficulty level), Grand Prix Legends and F1-2001 with the force feedback patch).
SGI can not be judged on Oracle support or other database support (although from personal experience Red Brick (a data warehouse database) kicks ass on an Origin 3000)).
SGI is specifically targeting the scientific, creative, analytical and technical users.
These users use floating point, or need very large single system image (NOT clusters) or need the graphics that SGI provides.
V6/8/10/12 graphics provides 12 bits per component with 10bit D-A's in an Octane. This is VERY unusual in the graphics industry. This has enabled human vision research not possible on any other machine.
SGI workstations (Octane2), servers and Onyxes have HDIO (Hidefinition TV I/O cards). You need a LOT of bandwidth to handle uncompressed 1020p I/O.
Onyxes have at the moment 256MB of texture memory per pipe. This enable volume visualisation impossible on any other graphics card. This is very important for some forms of medical imaging and for Oil/Gas exploration (BUT NOT ALASKA PLEASE!)
SGI has a 1024 processor single system image (NOT a cluster) running at NASA AMES and has shipped multiple 256/128/512 proc SINGLE SYSTEM IMAGE Origin 3000s.
Scientists like single system image machines. The quote from one of the nasa guys is it is just like using a desktop workstation, only a bit faster:-)
These are not ordinary machines and are not targeted at people running Oracle.
There are two factors to worry about:
Refresh rate - this affects the 'flicker' you can see in the screen. In a bright environment you will perceive flicker at higher refresh rates than in a dark enironment. . In a dark environment you can get away with 60Hz (one of the reasons for dark rooms when doing video editing etc).
Frame rates.
The actual issue is 'the amount an object moves between frames' - not the frame rate.
So if you do an 180 degree spin you might need 200FPS to not perceive stutter. If you are doing a really slow pan you can get away with a much lower frame rate.
There is no such thing as the 'maximum framerate a human can perceive'. You see stuttering if an object moves more than a certain amount between frames.
I asked this exact same question of one of our Chief Scientists (I work for SGI) and he referenced a paper on this and summarised it:
Reference:
VISUAL DETECTION OF MOTION
ANDREW T SMITH / ROBERT J SNOWDEN
0 12 651660 X
Comment:
In there it indicates that deltamax should be less
than 15 arc mins. That is, for perception of smooth motion, objects should not displace more than 15 arc mins. from one frame to another.
As a reference, 1 arc min. subtends approx 1.8mm,
at a distance of 6m. This is how we get the term
6/6 vision using the Snellen chart.
What is needed is a combination of a radial marking menu and a linear menu in the same display such that the number of items in the menu can be increased beyond eight items while still permitting rapid selection for the items of the marking menu using a marking or stroke pattern and selection of the items of the linear menu using a locational method.
Pie menus are not patented. What is patented is a way large (greater than 8) options can be selected with nested pies and the overflows handled by linear selection.
The title of the patent helps: Methods and system of controlling menus with radial and linear portions
As I read it it is the combination that is patented.
Slashdot Mirror
I've read novels on my palm for years. It's particulary useful for Neal Stephenson books which are enormous.
www.ereader.com formerly known as palmdigitalmedia is where I get all my ebooks.
I also get normal files and convert them to the palm media format and use their ebook converter under wine on linux (DropBook).
I find my T3 + an SD card a great way to read books. I can bookmark pages, highlight sections of text and lookup words using whatever dictionary.
And I don't mind good DRM. It means I can get new release books on my palm without waiting too long (they can become available when the hardback comes out).
However - I'd like ereader on linux without using wine. Hmm.
http://www.serverwatch.com/hreviews/article.php/34 36001
a search on google for nasa altix water back will give you lots of hits.
Also we have one for eval in our data center from IBM.
So it's already here really.
The question isn't which vendor will do it, it's which customers want to put in the water cooling infrastructure needed.
http://www.negativland.com/audio.html
http://www.deuceofclubs.com/write/negativl.htm
AFAICT it works not by detecting pressure but by seeing how much of the finger is pressing down on the pad. As you increase pressure - more of your finger is squished into the pad.
DVD+RW and DVD+R are formats that were never supposed to be supported in consumer players.
Gasically this is a beta vs vhs thing were the proponents of dvd+r and dvd+rw don't want to pay the licensing fees to the dvd-r guys.
DVD-R is the most compatible format.
DVD-RAM is the best for rewrites and so on. Completely outstrips dvd+rw for performance and longevity and has in fact been around for ages in jukeboxes etc.
They have renamed it to the Cray X1 (SV2 was the code name when SGI owned cray - in fact the X1 uses the same technology routers and internode interconnects used in the Origin 3000's and Origin 300s's)
This page has the video describing it (look for the tech overview) in detail.
This press release just talks about it. Look at the 350lbs of force required to mate the MCM to the board.
The 'SGE' is similar to SGI's compositor which is a product you can buy off the shelf which composits the output of multiple graphics cards in hardware.
e ch _info.html
:-)
http://www.sgi.com/visualization/onyx/3000/ip/t
The IBM research device does some other things that the SGI IP compositor doesn't - but at least the SGI compositor has been available for around a year - so you can use it today.
(Not that I am SGI biased or anything
I have seen this display driven by one Octane2.
a nn el.html
http://www.sgi.com/workstations/octane2/
Not one rack of 8 PCs and one half rack of some graphics engine.
It was drawing the full display at about 30fps.
It was as easy to use as any other workstation rather than the 'interesting' mix of 8 pcs in a rack and some other half rack of graphics stuff.
The hardest thing was to read the tiny little fonts on the screen (the display is 200dpi IIRC) - you need a magnifying glass (or of course you could increase the font size).
The Octane2 can do this because you can install two V12 graphics engines each with a dual channel adapter.
http://www.sgi.com/workstations/octane2/dual_ch
SGI software stiches the cards together transparently.
It is a beautiful display.
The amuzing/annoying thing about these sorts of announcements is that customers have been using SGI stuff to do this for the last 8 years or so.
If you were on the leading edge of this kind of work would you wait that long for some kludged together solution which might work if you have enough Duct tape to stick it together?
Or would you pay the extra cash for a solution that works and gives you a huge jump on your competition.
The sort of software layer that can be used to make these four channels (two channels from each graphics card) into one display is stuff like:
http://www.sgi.com/software/multipipe/sdk/
Oh yeah - and re your 'every table comment'. The Octane2 fits on one table - it doesn't surround the table like the IBM stuff.
Grip and acceleration (either by more power, less weight or more grip).
I expect McKinley to be the best price/performance FLOATING POINT chip out there when it is shipped.
... there are two reasons for McKinley - 64bit and excellent floating point performance.
In the scientific market integer performance isn't the problem it is floating point.
IPF (Itanium Product Family) is specifically targetted to do very good floating point. The first generation Itanium in fact had terrible integer performance but very good floating point performance.
There will be no comparison between a PIV and McKinley when doing things like floating point matrix multiplies etc etc. This is the important measurement when doing scientific computing.
AMD will not be able to touch the floating point performance of McKinley - even hammer. They are aimed at different markets.
So
SGI's V10/V12/V6/V8 all have 10bit DACs. The frame buffer is 12 bits per component. And that's 12/12/12/12 (ie full range alpha) unlike the XVR-1000.
Oh yeah and the XVR-1000 has about half the pixel fill rate of SGI's V10/V12s which is a serious problem for multipass rendering and so on.
(note that the DVI-D spec allows a dual link to get out > 8bits per component from the digital interface).
(Of course Onyx's with InfiniteReality are a completely different story - but you can't really call them a 'workstation').
(I am biased though - I work for SGI but am not representing their views).
And don't forget Sun's cheapest version of this workstation doesn't have the 10/10/10/8 bit colour card - it has a much worse one (the PGX). So be careful comparing prices.
u nf lash.20020314.2.html
http://www.sun.com/smi/Press/sunflash/2002-03/s
And then there are the hardware accelerated image processing functions (ie convolutions) in the V10/V12 SGI graphics which the Sun card doesn't seem to have.
Make that four.
Also the original version was much better paced than the long slow version.
Like others however I prefer the more ambiguous ending than the original version.
I like hard boiled detective styled things.
I really liked the pace of the original version.
From what I saw it wasn't an antialiasing problem in the traditional sense. It was an aliasing problem in that the depth sensor they use has a low spatial resolution.
The demo I saw that was live used the depth as a key - but it wasn't used to blend the objects together, it was used to make a simple visible/not visible decision. What you are describing is using the depth key to vary alpha which would be interesting to look at.
One VERY cool demo they did was to fly a CG character through the arms of the live talent. Where the CG object and the real object intersected you could clearly see blocky edges.
At the time I thought the technology (at least in a live setting) was maybe good enough for a young kids show were you want high speed production and the production values don't matter too much. Once they get the depth sensor up to broadcast the resolution however it will be a VERY nice live tool. (Maybe they have already done this - I haven't seen it in two years and they must have improved).
NAB is the National Association of Broadcasters conference. The ZCAM was was being demoed then.
In the demos they had realtime keying so they could fly a 3D CGI character in front of and behind the live talent. There was only about a 40ms delay. This is impossible with normal keying (ie blue/green screen). (You can only put stuff behind the talent).
It's biggest limitation was the resolution of the 3D sensor was low - so you had rough edges (think jaggies).
They also demonstrated a 3D Realplayer and 3D Windows Media players (which you watched with stereo shutter glasses). These players were called 'deep players'. Pretty cool but definitely not new.
What you described was demonstrated with this tech at NAB in 2000. In the demo they had the camera with the audience wearing polarized glasses (which I kept).
In the entire time I have owned the card creative never released any new drivers for it at all.
I use amd and via chipsets and have had many problems with the card across two different motherboards (a7v and msi 6380ru). Everytime I had to shuffle the card around because the IRQ sharing wouldn't work properly.
I finally gave up when max payne would randomly crash and I swapped out my sblive for my good old trusty Diamond MX300 card and never had a problem since.
I then upgraded to a hercules game theatre XP and this has confirmed I will never buy creative again.
Hercules frequently updates their drivers and they add new features all the time. This is what creative promised and never delivered.
The hercules card is noiseless - unlike my sblive, and was half the price of a sblive platinum (which has a comparable feature set).
I almost gave into the evil creative and the audigy but it was twice the price and considering creatives track record of never continuing to release new drivers for their old cards I gave up.
I now have a totally stable system for the six games I am currently addicted to (Alien vs Predator 2, Castle Wolfenstein, Max Payne (I am replaying it at a high difficulty level), Grand Prix Legends and F1-2001 with the force feedback patch).
Alpha is one way to do the smoke/fog effects.
Alpha is the transparency of a material/texture.
SGI can not be judged on Oracle support or other database support (although from personal experience Red Brick (a data warehouse database) kicks ass on an Origin 3000)). SGI is specifically targeting the scientific, creative, analytical and technical users. These users use floating point, or need very large single system image (NOT clusters) or need the graphics that SGI provides. V6/8/10/12 graphics provides 12 bits per component with 10bit D-A's in an Octane. This is VERY unusual in the graphics industry. This has enabled human vision research not possible on any other machine. SGI workstations (Octane2), servers and Onyxes have HDIO (Hidefinition TV I/O cards). You need a LOT of bandwidth to handle uncompressed 1020p I/O. Onyxes have at the moment 256MB of texture memory per pipe. This enable volume visualisation impossible on any other graphics card. This is very important for some forms of medical imaging and for Oil/Gas exploration (BUT NOT ALASKA PLEASE!) SGI has a 1024 processor single system image (NOT a cluster) running at NASA AMES and has shipped multiple 256/128/512 proc SINGLE SYSTEM IMAGE Origin 3000s. Scientists like single system image machines. The quote from one of the nasa guys is it is just like using a desktop workstation, only a bit faster :-)
These are not ordinary machines and are not targeted at people running Oracle.
There are two factors to worry about:
Refresh rate - this affects the 'flicker' you can see in the screen. In a bright environment you will perceive flicker at higher refresh rates than in a dark enironment. . In a dark environment you can get away with 60Hz (one of the reasons for dark rooms when doing video editing etc).
Frame rates.
The actual issue is 'the amount an object moves between frames' - not the frame rate.
So if you do an 180 degree spin you might need 200FPS to not perceive stutter. If you are doing a really slow pan you can get away with a much lower frame rate.
There is no such thing as the 'maximum framerate a human can perceive'. You see stuttering if an object moves more than a certain amount between frames.
I asked this exact same question of one of our Chief Scientists (I work for SGI) and he referenced a paper on this and summarised it:
Reference:
VISUAL DETECTION OF MOTION
ANDREW T SMITH / ROBERT J SNOWDEN
0 12 651660 X
Comment:
In there it indicates that deltamax should be less
than 15 arc mins. That is, for perception of smooth motion, objects should not displace more than 15 arc mins. from one frame to another.
As a reference, 1 arc min. subtends approx 1.8mm,
at a distance of 6m. This is how we get the term
6/6 vision using the Snellen chart.
Here is the guts of the patent from the detailed patent description
Pie menus are not patented. What is patented is a way large (greater than 8) options can be selected with nested pies and the overflows handled by linear selection.
The title of the patent helps: Methods and system of controlling menus with radial and linear portions
As I read it it is the combination that is patented.
IMHO of course.