Slashdot Mirror
Open Source Finally Hits Real Silicon
pagercam2 writes "While Open Source software has many success stories, hardware and particularly chips haven't had as much. While there have been multiple Open Source projects, none have come to a final product until now. The OpenRISC 1000 has been implemented by Flextronics Semiconductor(a division of Flextronics, the contract manufacturer possibly best known for its production of many Cisco products) along with PCI, 10/100 Ethernet, serial, GPIO etc. ... Details and pretty pictures available at OpenCores.org, and it even runs uClinux. Good Job!"
>>Flextronics Semiconductor(a division of Flextronics, the contract manufacturer possibly best known for its production of many Cisco products).
m l
Flextronics would actually be best known for being the main manufacter of the Microsoft Xbox.
http://www.wired.com/wired/archive/9.11/flex.ht
"Can anyone seriously point out some practical applications of this processor?"
Sure.
For chips derived from this test SoC:
MP3player
VoIP hard phone
Network Router
Firewall
Wireless Access Point
DVD player
Car stereo
Cell Phone
PDA
For uClinux:
It's all around you, many of the products _you_ use every day run it. Just because you think Linux means servers and desktops doesn't mean that's the only place it's widely deployed!
J
Have open-source cores been available to implement those specs been available for a LONG time?
For quite awhile, as I understand. The Leon chip is an example of this. Other areas such as Fujitsu's processors and set top devices have been based on Sparc.
I'm not saying that OpenCores is a bad thing. I'm just refuting this "we were here first" bullshit.
Javascript + Nintendo DSi = DSiCade
Sure it could be done, up to and including the design verification using chip simulations, but actually making the chips and debugging the silicon process could get very expensive. I'm sure you could find a foundry in Taiwan or China to produce it, but would there be a market for it so you could get back all those startup costs? Do you know of some folks who have a few hundred K to invest against AMD, Intel, Motorola and IBM for a tiny slice of the market? Hardware has a lot of startup costs than software to get it to market. It's not like compiling the new code for your kernal fix. Maybe if it was specialized and optimized for embedded applications it might have a shot. I guess you could call it the "Penguin" chip since I'm assuming it would be optimized for Linux.
From the SPARC website:
Hell, it doesn't even look like much of an open standard. You need to license the instruction set in order to be able to implement it. This is like saying UNIX is open source, since anyone can implement POSIX and license the UNIX trademark, and because a lot of people have licensed the source code. That's not open source; it may be an open standard (although I'd argue that in order to be an open standard, you can't restrict who implements it with licensing agreements). So really, SPARC is in no way open source, and I wouldn't even consider it an open standard.OpenCores, on the other hand, is really open source. You get the full design of the entire chip; you could just produce the chip by sending the CAD files to a chip fab and having them produce it. All of the Verilog/VHDL/etc. are open and freely available for you to use and modify. Even if you license the SPARC ISA, you still have to design the chip yourself.
Hell, there are plenty of ISA's that you can license. The IA32 architecture is implemented by Intel, AMD, Transmeta, and others. PowerPC is implemented by IBM and Motorola. MIPS chips are produced by lots of people. Open ISA's are a dime a dozen. What's important about OpenCores is that the full chip design is completely open.