New optical CPU 1000 times faster than regular CPU´s
Posted by: [NT] on: 10/30/2003 05:57 PM [ Print | 12 comment(s) ] · 5027 views
An israeli startup company, called Lenslet has developed the world's first commercially available vector DSP with an embedded optical core.
The EnLight is a fixed-point processor comprised of three elements: a Vector Matrix Multiplier (VMM) capable of performing vector-matrix operations at a rate of up to 8000 Giga MAC operations per second; a Vector Processor Unit (VPU) capable of performing 128 Giga operations per second and an industry standard DSP (TI TMS320C64xx) for control and scalar processing (SPU). The EnLight256? could be used either as a system-embedded accelerator or a standalone processor for computational intensive types of applications.
« Neue CPU tausend Mal schneller als herkömliche CPU´s · New optical CPU 1000 times faster than regular CPU´s
· 3 neue Microsoft Sicherheitspatches »
Comment
|
Lithorus Unregistered |
#49041 Posted on: 10/30/2003 06:11 PM
A post from slashdot : "This is NOT a Harvard architecture part - this isn't fetching instructions from RAM and executing them, like a regular DSP would. Think of this more like an FPGA - you have a device that is configured for a specific processing algorithm, and data is fed in at wire rate and processed at wire rate. An example of how a device like this might be used may be in order: I'm trying to find a radar pulse buried in the noise coming in from my receiver. I want to know the phase delay of the radar pulse - how long from when I sent it till I got it back. Now, I know what my radar pulse looks like as it goes out. I know that any reflection is going to consist of versions of that pulse shape, delayed and of varying strengths. So what I do is called a correlation - the easiest way to think of this is to imagine having 2 transparencies, one of my outgoing pulse, and one of the incoming signal. Now, I hold them up to the light, and slide the incoming signal across the reference pulse until things match up - that's the point of maximum correlation, and that give me the delay of the signal. A real correlation function is a bit more complicated as you have to allow for the signal level to be changed - if I am looking for a signal of N samples in a received data stream of M samples, I have to do M*N multiply and add operations to get my correlation. Now, for a radar signal I might be sampling at over a billion samples a second, and looking for a chirp of a 100 ns would give me over 100 billion MAC operations a second. There are ways to do that with conventional DSPs, but they are a galloping ***** to do (you basically make a cluster of DSPs, and each DSP takes a part of the signal. Synchronising that is a *****.) This device would work by having the shape of the outbound pulse represented in the structure of the device itself, and the MACs are done by taking the incoming data stream and projecting it on the structure - thus you do all your processing in parallel, and at wire speed. You get a pulse out when the incoming signal matched the signal you ar looking for." If it sounds too good to be true it probably is... |
Comment
|
DarthBinky Unregistered |
#49042 Posted on: 10/30/2003 06:18 PM
But where's the link to the Quake 3 benchmarks? |
Comment
|
BetrayerX Unregistered |
#49043 Posted on: 10/30/2003 06:37 PM
LOL! Probably Nvidia "optimized" for them already so, what's the use? Man, that was funny. |
Comment
|
Rosco Member Posts: 24 Joined: 2005-02-09 |
#49044 Posted on: 10/30/2003 07:17 PM
"Man, that was funny." No, no it wasn't. It's getting old and tired. Try something a bit more constructive next time. The ATI/nVidia bashing is getting very old and not doing a damn bit of good. |
Comment
|
NT Unregistered |
#49045 Posted on: 10/30/2003 08:12 PM
I have deleted some comments. Disputes please via Private Messages and not here. Thank you |
Comment
|
Devourer Unregistered |
#49047 Posted on: 10/30/2003 08:29 PM
your an idiot |
Comment
|
Rosco Member Posts: 24 Joined: 2005-02-09 |
#49049 Posted on: 10/30/2003 09:02 PM
Thanks NT, a lot of people who read this site on a daily basis appreciate it. It might also help if you start cancelling some of the accounts of those that are always spamming it with anti ATI/Nvidia garbage all the time as well. |
Comment
|
Devourer Unregistered |
#49051 Posted on: 10/30/2003 09:34 PM
how about no1 is allowed to post any comments but just you? Then you you can have the entire site all to your self. |
Comment
|
razor_ro Member Posts: 53 Joined: 2002-07-03 |
#49055 Posted on: 10/30/2003 11:31 PM
HAHA,.... i thought it was funny too,.... maybe the other guy is a little touchy about the "bashing" cause he probably works at one of those places,... LOL |
Comment
|
Tuckwit Member Posts: 65 Joined: 2002-09-26 |
#49058 Posted on: 10/31/2003 12:34 AM
Good one. Place for news, not fools. |
Comment
|
BetrayerX Unregistered |
#49062 Posted on: 10/31/2003 02:12 AM
You should learn how to read properly. When I said it was funny I meant the Quake 3 Benchmark remark.....that's why I replied his specific comment. The NVidia vs ATI is your own comment. I posted a joke following the Q3's....some might like it......some not. |
Comment
|
BetrayerX Unregistered |
#49063 Posted on: 10/31/2003 02:17 AM
LOL! I can't believe such a stupid joke hurt so many feelings. Devourer: If your proposal happens, we are all doomed. Plz rethink that!  |
