×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

OpenSPARC and Power.org, Who has it Right?

ScuttleMonkey posted about 8 years ago | from the fight-to-the-death dept.

125

Andy Updegrove writes "Last summer, IBM set up Power.org, to promote its PowerPC chip as what it called 'open hardware.' This year, Sun launched the OpenSPARC.net open source project around the source code for its Niagara microprocessor. But what does 'open' mean in the context of hardware? In the case of Power.org, Juan-Antonio Carballo said, 'It includes but is not limited to open source, where specifications or source code are freely available and can be modified by a community of users. It could also mean that the hardware details can be viewed, but not modified. And it does not necessarily mean that open hardware, or designs that contain it, are free of charge.' True to that statement, you have to pay to participate meaningfully in Power.org, as well as pay royalties to implement - it's built on a traditional RAND consortium model. To use the Sun code, though, its just download the code under an open source license, and you're good to go to use anything except the SPARC name. All of which leads to the questions: What does 'open' mean in hardware, and which approach will work?"

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

125 comments

Binary minds want to know. (4, Insightful)

jellomizer (103300) | about 8 years ago | (#15115327)

Who has it right?

I hate that question because it assumes that One is Right and the other is wrong.
It is like asking a student what is the Square root of 9

One student says 2 and the other says 5. Well there is no consensious so one of them has to be correct right? No both are wrong.

In an other class that asks the same question
One student says -3 and the other says 3. So one of them has to be wrong they are different answers. No both answers are correct.

Just because they are multiple view points it doesn't mean that there has to be a write or wrong answer for one of them.

Open your mind people!

Re:Binary minds want to know. (2, Insightful)

nharmon (97591) | about 8 years ago | (#15115343)

Generally when you ask for the square root of something, it is implied that you're asking for the positive square root.

Re:Binary minds want to know. (1)

mctk (840035) | about 8 years ago | (#15115523)

I disagree. I think it's generally inferred.

Re:Binary minds want to know. (1)

dgiaimo (794924) | about 8 years ago | (#15117102)

Not true. The square root of a positive real number, a, is, by definition, the positive real number, x, such that x*x = a. IAAMM (I am a math major.)

Re:Binary minds want to know. (1)

mctk (840035) | about 8 years ago | (#15117730)

Well, it depends on what we are talking about. First off, the square root of a number, a, is, by definition, the number x such that x*x=a. The definition of a square root says nothing about positive results. However, the definition of the square root function does specify only positive results (because it wouldn't be a function otherwise ). From Mathworld.com:

"For example, the principal square root of 9 is sqrt(9)==+3, while the other square root of 9 is -sqrt(9)==-3. In common usage, unless otherwise specified, 'the' square root is generally taken to mean the principal square root."

Generally taken: inferred.

Who has it right? (2, Informative)

ihatewinXP (638000) | about 8 years ago | (#15115369)

I do.

its .org - not ,org

This is the first sentence of the post -c'mon don't make it so easy - I swear I am not a spelling nazi.

Re:Who has it right? (1)

Bromskloss (750445) | about 8 years ago | (#15115412)

I swear I am not a spelling nazi.
*wakes up abruptly* Huh?! I heard something. I work for Godwin, and I heard "nazi".. eh.. oh, sorry. False positive. You're clear. Have a nice day. *falls to sleep again*

Re:Binary minds want to know. (2, Interesting)

timeOday (582209) | about 8 years ago | (#15115421)

I don't think arguing over some dictionary definition of Open is the point, but rather to figure out the costs and benefits of different aspects of openness to all the parties involved. Look at how BSD and GPL differ in their interpretations of openness, and how those differences have played out over the years, for instance with the fragmentation of BSD. These decisions do matter.

Intelligent minds at least try to be wrong. (0)

Anonymous Coward | about 8 years ago | (#15115461)

Ahh.. so because sometimes there isn't a right and a wrong, we aren't allowed to ask ever again.

Was Atilla a bad man? How dare you be judgemental!!!

It comes down to the same thing as ever. "Open" doesn't mean much at all. "Free" as in freedom is a moral thing, and is much more stable and safer. Once again, the "open" people will make it more difficult for the "free" people to start up, but inevitably the "open" projects end in commercial failure due to their inability to find a stable base to cooperate on.

Be scientific. Make judgements. Make predictions people can actually use, measure and follow. Dare to be wrong. It's much better to be wrong than lost. Don't just sit there and say "it's all okay with me man"..

"It's not even wrong." - Pauli.

Re:Binary minds want to know. (0, Flamebait)

Maset (190867) | about 8 years ago | (#15115560)

What a fucking redundant post!

Your rhetorical questions (which are then answered, you dolt!) are the worst kind of sophistry; you don't even attempt to anwer the question.

Stop trying to be clever and you might just find yourself being so.

Re:Binary minds want to know. (1)

monkeySauce (562927) | about 8 years ago | (#15115740)

You had opened my mind, until I read:

it doesn't mean that there has to be a write or wrong answer

and then it snapped shut again!

You are what you write... right?

Re:Binary minds want to know. (0)

Anonymous Coward | about 8 years ago | (#15116411)

Good points against the bogus dichotomy implied by the headline, but peeps whose minds are not (too?) autistic/ADD will find obvious that the question is how the two approaches are different and the consequent implications, not whether one is PC "open" and therefore right/wrong.

Binary minds answer: (0)

Anonymous Coward | about 8 years ago | (#15116737)

It is like asking a student what is the Square root of 9

11. Duh.

Re:Binary minds want to know. (1)

joe 155 (937621) | about 8 years ago | (#15117157)

Open your mind people!

You have inspired me, so I will now release the source code of my mind in order to have an "open mind"... currently I'm running on the v.20 firmware and my mind maps out a little like this... "beer---sex---...beer"

Well... (4, Informative)

nweaver (113078) | about 8 years ago | (#15115354)

Open source cores for full processors are actually old news.

The LEON 2 SPARC-compatible core has been around for years.

Anyone doing a real chip design, however, can afford to pay for a real supported core.

Re:Well... (3, Insightful)

georgewilliamherbert (211790) | about 8 years ago | (#15115420)

Hey Nick, it's been a while.

The difference here is that we're seeing current generation high end processors start to appear here... there's a huge difference between low end, embedded CPU class cores being available and a Niagra T-1 SPARC. T-1 isn't the fastest single thread CPU out there, but it may well be the fastest total multithreaded throughput CPU out there on the market today. Whether that's appropriate for most users / workloads or not, it is clearly a huge difference compared to embedded CPUs.

Re:Well... (1)

nweaver (113078) | about 8 years ago | (#15115877)

This actually makes the open SPARC less relevant.

Anyone using an FPGA would use a smaller hardcore built into the FPGA or something like a MicroBlaze softcore. You couldn't fit a synthesized, OOO SPARC on all but the biggest FPGA.

Anyone doing an ASIC which has a use for a high performance CPU would just buy the IP anyway, as it gets into the noise compared with all the other costs.

Re:Well... (2, Insightful)

georgewilliamherbert (211790) | about 8 years ago | (#15116279)

I think it's clear that they're not addressing the same market space. But my point was that this release by Sun is qualatatively different than prior embedded CPU releases under open source rules. Fewer people may end up synthesizing and using T1's than will use either FPGA or synthesized embedded CPU designs, but the availability of T1's design does open up interesting new possibilities.

For which, I agree, the "market demand" is unclear. At the very least I know that a lot of researchers are looking at it closely for ideas and comparisons. And given the licensing, someone who had a high-thread-count embedded application might find a use for. It's not entirely clear to me that the IP is in the noise... it's been a decade, but I was working around chip IP issues in the mid-90s, and a large chunk of the budget was the IP, and another was verifying the level of stuff which is already verified with the T1 design. You still need to synthesize and tape it out, but that's a large chunk of effort taken out.

I agree that the total number of shipped products from the T1 design release could end up being zero. But I wouldn't bet any serious money on that.

Re:Well... (1)

IvyKing (732111) | about 8 years ago | (#15119119)

I did see a mention of some small outfit that was making noises of doing a single core version of the T1 - presumably small and low power. I was not able to find out anything more of the company mentioned.

Would be nice to have a low power (say 5 to 10 watts) Sparc board to run Solaris - something like a Sparc equivalent to the Soekris boards.

"Real design" (2, Interesting)

MoxFulder (159829) | about 8 years ago | (#15116134)

Anyone doing a real chip design, however, can afford to pay for a real supported core.


I remember reading ~1992 that anyone doing real software development could afford to pay for a real supported compiler. They were downplaying the benefit of GCC and the other GNU tools. Well as hardware has gotten cheaper and faster and the Internet has expanded and more people have gotten tech-savvy, guess what? Lots of people are doing *REAL* software development with FLOSS software tools.

Back when you needed $10000 worth of hardware + OS licenses etc. to make software development feasible, paying $300 for a compiler was no biggie. But now you can get an awesome complete workstation for $600 and even the $100 Microsoft OS tax starts to seem like a pretty crappy deal.

I imagine that hardware design will increasingly go the same way. Obviously, there are a lot more hurdles to go before we'll be fabbing chips in our basements. But I work in electronics research in a physics department, and people are doing amazing stuff like printing integrated circuits with inkjet printers... commercial equipment to do that is now selling for $100k.

About being the first one... (1, Insightful)

Anonymous Coward | about 8 years ago | (#15118593)

I just stepped into OpenSPARC.net and read this:

For the first time in history, developers gain access to the chip multi-threading (CMT) technology unique to the UltraSPARC T1 processor, which will be released under the OSI-compliant GNU General Public License (GPL).


I can't help but wonder, when will be the second time in history developers can gain access to the chip multi-threading (CMT) technology unique to the UltraSPARC T1? Is that even possible as per definition?
Sun is big, they should know better than to let this bullcrap marketing business slang selling-point junk be the first thing a developer sees...

Re:Well... (1)

Horrus (968006) | about 8 years ago | (#15119425)

If you want to sell devices with Leon2 inside it you have to pay money to the owner of the code. Also its LGPL. All modifications have to be sent back to the owner of the code. But it is a fine piece of VHDL Code. Good for learning how to code and in the same code how not to code hardware

You keep using that word. (4, Insightful)

AKAImBatman (238306) | about 8 years ago | (#15115370)

All of which leads to the questions: What does 'open' mean in hardware, and which approach will work?

I think you're confused. "Open" has traditionally been shorthand for "Open Standards". Thus your hear terms like "OpenWindows", "OpenLook", and "Open Group". They're all referring to the standards being available to all, and not any sort of Open Source Software take on those standards. Open Standards make the world spin 'round, and are a key reason why we have so much compatibility in our daily lives.

What you're thinking of is "Open Source", also known as "Free (as in freedom and game show prizes) Software". This is a very different category of of openess that relies on a developer to give up some of his rights to support the greater good. This is a laudable goal, but it is often not shared by coorporations and businessmen.

For what its worth, Wikipedia has a fairly good article on the concept of Open Standards [wikipedia.org].

Re:You keep using that word. (0)

Anonymous Coward | about 8 years ago | (#15115550)

This is a very different category of of openess that relies on a developer to give up some of his rights to support the greater good.

The developer doesn't actually give up any rights at all. If you own the copyright, you own the copyright - if you decide to allow people to do something with your copyrighted work under an open source license, that in no way implies that you've given up any of your rights. You're just telling people they have to meet the terms of your license to do certain things that you, yourself, can do because you own the copyright - but they could not without some sort of express permission, because they do not.

Re:You keep using that word. (1)

geoffspear (692508) | about 8 years ago | (#15115748)

What you're thinking of is "Open Source", also known as "Free (as in freedom and game show prizes) Software".

Wait, you mean I'm required to pay tax on the value of the free software I recieved? It's a good thing I still have 5 days to correct my tax return!

Re:You keep using that word. (1)

maxwell demon (590494) | about 8 years ago | (#15116068)

Wait, you mean I'm required to pay tax on the value of the free software I recieved?

Ssh! Don't give the Minister of Finance ideas!

Re:You keep using that word. (2, Insightful)

hackstraw (262471) | about 8 years ago | (#15115961)


I believe "Open" in the sense of hardware means that you know how it works because its documented. NVidia graphics cards are NOT open. One of my microphone preamps is open. It has a pseudo-schematic that shows the signal flow through the device, so I know what control does what and where it is in the signal path. Without the schematic, I would still be under the assumption that the output knob adjusts the output on both the digital and analog outputs, but the schematic clearly shows me that its only on the analog outputs.

Oh, and those hippie Linux people have opinions regarding open hardware. Its located here: http://www.open-hardware.org/ [open-hardware.org] The site seems slow to me, so you may not be able to view it for long.

I'm a fan of open stuff. It makes my life better.

Re:You keep using that word. (1)

inaequitas (885724) | about 8 years ago | (#15115975)

Actually "Open" source is more along the lines of the standards approach rather than the FSF one. OSS != FSS as many activists on either side will tell you.

Opening the source allows people to expand on it. The Open hardware model lets you do the same, because OSes and drivers are, in a way, an extension on hardware much like plugins extend software functionality. By having more development focus around a certain hardware platform more demand can possibly arise. So SUN and IBM can make more money out of 'outsourcing' the software development to the community while still keeping the rights to manufacture the chips.

On the community's side this is a gain because they are now able to add any feature and functionality they want [well, within specs] and have better support for these chips in various OSes.

Apples/Oranges (0)

Anonymous Coward | about 8 years ago | (#15115389)

You have to understand the two companies to understand the differences in hardware licensing. Sun does not produce chips. Their business is to define a standard and have others implement the standard, which Sun then uses in their systems. IBM otoh DOES manufacture chips, as well as design them. Now they currently share production with Motorola on PPC, but it's always been a very tight and closed relationship.

Now given that, you can see how their respective views of "open" hardware have been formed, as they follow closely the respective companies buisness models.

Re:Apples/Oranges (5, Informative)

georgewilliamherbert (211790) | about 8 years ago | (#15115469)

Sun does not produce chips. Their business is to define a standard and have others implement the standard, which Sun then uses in their systems.

No, Sun has always designed the majority of the CPUs that they use in their systems. They are fabless; they don't own the chip fabrication factories... but they have buildings full of people working on chip design.


Others have designed SPARC CPU chips (Fujitsu / HAL; Ross) which Sun used, as well. But Sun engineers designed UltraSPARC-I, II, IIi, IIe, III, IIIi, IV, T-1, and going backwards the SuperSPARC, MicroSPARC-I and II, most of the Sun-4 and Sun-4c generation stuff.

Re:Apples/Oranges (0)

Anonymous Coward | about 8 years ago | (#15115776)

They are fabless

I think I'm still confused... did Sun write the one about the ant and the grasshopper?

Answer: NEITHER (5, Funny)

Anonymous Coward | about 8 years ago | (#15115398)

In my dream for America I would like to put control of computer hardware and software companies firmly in control of the government - exactly where it should be. At the moment terrorists have way too much access to the inner workings of computers and software - especially with regard to open source software.

Just imagine what could happen if terrorists used the freely and openly available source code in Linux for example to create some sort of super weapon. The results could be catastrophic.

I am pushing like crazy for the US government to take full control of all USA based computer hardware and software companies, effectively creating one large mother company. There would be no more OS wars as there would only be one. Consumers would have the benefit of knowing that their hardware and software was US GOVERNMENT APPROVED (TM) and terrorism free.

To take it perhaps one step further, the government could even enable monitoring devices within the equipment to further prevent any crime or terrorist attacks.

As the old saying goes: IF YOU HAVN'T DONE ANYTHING WRONG YOU DON'T HAVE ANYTHING TO WORRY ABOUT.

Re:Answer: NEITHER (4, Funny)

artifex2004 (766107) | about 8 years ago | (#15115455)

As the old saying goes: IF YOU HAVN'T DONE ANYTHING WRONG YOU DON'T HAVE ANYTHING TO WORRY ABOUT.

Which is why you felt free to post this under your account, Mr... Coward, is it?

Re:Answer: NEITHER (0)

Anonymous Coward | about 8 years ago | (#15118214)

Hey, I get a bit touchy about my name sometimes!

Re:Answer: NEITHER (1)

SuperGhost (952604) | about 8 years ago | (#15115669)

To take it perhaps one step further, the government could even enable monitoring devices within the equipment to further prevent any crime or terrorist attacks.


What makes you think they dont already do this?

another bloggers spouting off (0)

Anonymous Coward | about 8 years ago | (#15115457)

Sigh. Another numb-nutted blogger trolling for page views. Why do we see this crap on what used to be a great site? This guy is worse than a waste of time.

ScuttleMonkey, you are worthless.

Right/Practical (2, Interesting)

starseeker (141897) | about 8 years ago | (#15115465)

OK, obviously by the definitions most of the Slashdot crowd will go along with, the Sun release is the "open" one.

The more interesting question is "what use is an open core?"

Open source software has obvious utility in that it can be used by millions of people for a wide variety of jobs. All you need is a computer to get started.

Open hardware, on the other hand, is useful only for education or simulations unless you happen to have a fab plant.

If education and experimentation can be served by a "non-free" license then is there really any benefit to having a "free" license? I suspect by the time off the shelf technology is available to create CPUs based on current designs, they will be centuries obsolete. Even US copyrights and patents will have expired by then (unless they change the laws again) so it's a bit of a moot point.

Now I grant this might be a bit of a narrow viewpoint - for example some of the Lisp hardware designs would be very interesting to work with - but since the hardware costs of this sort of manufacture make the information needed to do it only one component of the (EXCEEDINGLY expensive) whole, I'm not sure the marginal benefit of having "free" cores will be very interesting, at least for something like a modern CPU.

Of course, there are non-economic considerations, but I don't really see overwhelming benefits for the "free as in freedom" model as opposed to the "free except for producing your commercial product based on them" model.

Re:Right/Practical (2, Interesting)

georgewilliamherbert (211790) | about 8 years ago | (#15115528)


Open hardware, on the other hand, is useful only for education or simulations unless you happen to have a fab plant.


Or MOSIS [mosis.org]....

A few thousand bucks and a working chip design will get you parts these days, in suprisingly modern fab processes (a few tens of thousand for 0.13u and 90nm).

Re:Right/Practical (1)

starseeker (141897) | about 8 years ago | (#15115607)

Cool! I hadn't heard of that before, but it sounds like a great idea!

I'm a bit out of it on the latest design requirements for CPUs - is the technology of these folks actually good enough to make a reasonably modern CPU?

Re:Right/Practical (1)

scheme (19778) | about 8 years ago | (#15115744)

I'm a bit out of it on the latest design requirements for CPUs - is the technology of these folks actually good enough to make a reasonably modern CPU?

You could probably get something in the P3 coppermine range with the 130nm tech. The 90nm will get you theoretically get you P4 or opteron systems but Intel and AMD use a lot of custom tweaks that MOSIS won't do. I think you'll need to assume that you need a generation better to match commericial cpu cores. E.g. a commericial cpu produced in 130nm would need 110nm or better design rules to reproduce using MOSIS.

Packaging could be a problem since I'm not sure MOSIS will be able to package the cpu in the correct packaging for you.

Re:Right/Practical (2, Informative)

georgewilliamherbert (211790) | about 8 years ago | (#15115772)

I'm a bit out of it on the latest design requirements for CPUs - is the technology of these folks actually good enough to make a reasonably modern CPU?
Yes. I believe that the best MOSIS process is the IBM 90 nm process [mosis.org], which is 7 metal layer, pretty flexible. The T-1 SPARC we're talking about (Niagra) is a 90 nm, 9 metal layer Copper wire fab design (see Sun's Specs [sun.com]). You can't quite fab a T-1 as Sun laid it out with IBM's process, but it's pretty close. You could produce a roughly the same size, slightly larger and/or slower version of the same chip with a new detail layout, using the same chip level "circuit diagram" but a different physical design with fewer layers of metal used etc. AMD uses 0.13 and 0.09 u (90 nm) processes for their current Opteron line [amdcompare.com], though theirs are Silicon on Insulator fab processes. Again, different design details, but the same general scale and capabilities. The newest Intel Zeon MP processors are at 65 nm processes, one step past the IBM 90 nm process (components on the average taking roughly half the surface area per step). But Intel still produces a lot of slightly older 90 nm and larger CPUs, and industry consensus is that the 90 nm AMD and 65 nm Intel chips are still roughly at equal performance.

Re:Right/Practical (1)

Momo_CCCP (757200) | about 8 years ago | (#15115831)

Having seen a couple things here and there about CPU design, I can tell you that transforming a 9 layer design into a 7 layer is far from trivial, so I guess you can forget about this.

Re:Right/Practical (1)

Forbman (794277) | about 8 years ago | (#15115549)

Well, say someone really liked the design of the SPARC (or PowerPC), and they wanted to develop their own version for use in environments, say, far different than desktop or server room environments, and perhaps with some value-add on the die itself as well. Say its a company that has already done work at hardening or ruggedizing silicon for different environments, so they've already sunk the capital into the processes etc to make it happen. Say AMD and Intel aren't interested in licensing their chip designs to this company (i.e., the license fees are rather high compared to past license fees). This company probably already has a contract fab facility at their disposal, they just want a cheaper way than Intel or AMD to make their chips. Both the PowerPC and SPARC "open" models could be good for them, and they'd probably go for the PowerPC model because the value of the extra stuff that comes with IBM's license and royalty fees is worth access to all that stuff that isn't available with OpenSparc.

Besides, at least at the Univ of Washington, more than a few projects involved developing silicon designs for projects. Now one has an open CPU design with OpenSPARC, and they can work within their project to try new things with the design as the basis, such as a RISC-based LISP CPU (or, better, Ruby CPU...).

Re:Right/Practical (1)

AuMatar (183847) | about 8 years ago | (#15115585)

THe same way Free Software works. Free Software tends to be higher quality due to its open nature, with fewer bugs per LOC. Hardware could recieve the same attention. In fact, it would be an even greater gain- software may take more man hours to fix once its released, but its still just manpower cost. Finding bugs in test silicon is expensive.

Open Hardware would also lead to growth in open hardware design tools, as companies won't be viewing their design tools as secrets that need to be kept anymore.

Beyond that- you don't need to own a fab plant to take advantage of this, you just need to be able to purchase the output. While its not economic to buy 1 CPU off an open design, it would be economic for an embedded hardware manufacturer, or a PC manufacturer to buy a lot of CPUs. These CPUs would be cheaper as there is no middle man in the picture.

There's also innovation to consider. Some people say open source software isn't innovative- bullshit. Show me a Windows with an O(1) scheduler. Show me one that has an SE Linux equivalent. Show me one that has virtualization support. Not all open source software is innovative, but much of it is. And its a lot easier to try a hardware innovation if you already have 90% of the CPU built and need to only replace your part, than if you need to build from scratch.

Re:Right/Practical (1)

starseeker (141897) | about 8 years ago | (#15115633)

Good points. Maybe it could be made to work, although I'm not sure how the changes could take place - in an industry with so much $$ tied into their current methods, it could be hard to get them to risk their crown jewels on a new development methodology.

Re:Right/Practical (1)

scheme (19778) | about 8 years ago | (#15115824)

Beyond that- you don't need to own a fab plant to take advantage of this, you just need to be able to purchase the output. While its not economic to buy 1 CPU off an open design, it would be economic for an embedded hardware manufacturer, or a PC manufacturer to buy a lot of CPUs. These CPUs would be cheaper as there is no middle man in the picture.

I don't see this at all. It costs a bunch of money (250+K) to create a mask to test your silicon. To produce a commercially viable cpu, you'll probably go through a few masks resulting in a fairly high cost for development.

A pc manufacturer or embedded hardware manufacturer would probably find it cheaper to just buy the cpus from a cpu maker like ARM or Intel then to finance cpu development. Besides any open source cores available would be pretty out of date due to time and financial constraints.

Re:Right/Practical (1)

AuMatar (183847) | about 8 years ago | (#15116427)

I don't see this at all. It costs a bunch of money (250+K) to create a mask to test your silicon. To produce a commercially viable cpu, you'll probably go through a few masks resulting in a fairly high cost for development.

A pc manufacturer or embedded hardware manufacturer would probably find it cheaper to just buy the cpus from a cpu maker like ARM or Intel then to finance cpu development. Besides any open source cores available would be pretty out of date due to time and financial constraints.


Either the fab could do it on spec in downtime (and increase his costs) due to it being an open core, or a group of embedded developers could go in for it. Perhaps the mask designs could also be open.

AS for out of date- this is embedded. Most cores are out of date by a decade, some of them are still pumped up microcontrollers. Older cores are used because the bugs are known and they don't need more power. A PC manufacturer might care, an embedded wouldn't.

As for open source cores even being out of date in the first place- I don't see that being true. Open source software isn't, why would open hardware be?

one example (1)

mmkkbb (816035) | about 8 years ago | (#15116013)

Having a CPU core stuffed into your FPGA along with the glue logic for your device means one less micro on the board. You could even lop off the core bits that don't get used if you're that concerned about space. You may think this is outlandish but Altera already does this with their NIOS II soft-core embedded processor.

Yes, it won't perform as well as a real processor but there are times when CPU performance is not the bottleneck. Electromechanical systems spend a lot of time waiting for the motor to get move the fribbus over the wotsit.

Re:Right/Practical (1)

MoxFulder (159829) | about 8 years ago | (#15116171)

Well, open cores certainly are of some use. For one thing, they lower the barrier to entry to a new company with some capital and some good ideas about how to expand an existing design. Just look at http://opencores.org/ [opencores.org] for examples of companies that have actually fabbed chips using their cores!!!

No, they don't yet make it possible for me to churn out SPARC chips in my basement. But I work in academic electronics research, and I've recently seen a talk on a machine that can be used to print (medium-scale) integrated circuits with an inkjet-like process. It costs $100k currently.

Clarifications (5, Funny)

Kohath (38547) | about 8 years ago | (#15115493)

- "Nigeria" is an African country. A prince there will soon be making me very rich.
- "Viagra" is a sex drug. I ordered some from a nice company that emailed me. It will also be here soon.
- Sun's chip is called "Niagara"

"Niagera" is none of these things.

Re:Clarifications (0)

Anonymous Coward | about 8 years ago | (#15116302)

Niagara also happen to be some very large waterfalls on the US/Canada border in Ontario.

Why not let people use the code? (2, Insightful)

ChrisGilliard (913445) | about 8 years ago | (#15115508)

I guess I don't get why IBM would have a problem with other people using their hardware specs for free. The barriers to entry are pretty big for one thing. It's not like your average Joe has a Billion dollar fab in his back yard and can use IBM's code to create a processor. The real trade secrets are in the manufacturing process. There's a big difference in making a chip and making millions of chips that cost less than $100 to mass produce.

e-iOpenBuzzword.com (2, Interesting)

fortinbras47 (457756) | about 8 years ago | (#15115650)

It used to be everything was e: etrade, e-mail, e-commerce, ebay ...

Then there was the i craze. iPod,iMac,ivillage.com, BMW's iDrive, ....

Maybe "open" is the new cool prefix to use. I'm sure anyday now someone will be selling OpenPods, sending openMail...

Re:e-iOpenBuzzword.com (0)

Anonymous Coward | about 8 years ago | (#15115947)

Then someone will come along and build a beowulf cluster of buzzwords, and it'll be e-i-e-i-Open. And on that server farm they'll@%$}{@£^"+++LOST CARRIER+++

Open Door, Brick Wall Behind It (2, Insightful)

postbigbang (761081) | about 8 years ago | (#15115703)

The phrase 'Open" means nothing. It implies many things, ranging from whether you're RMS to Steve Jobs. Developer programs have been mutating for years, starting way back in the '80s. The real depth of the programs, and their usefulness is pretty simple. Take an example: Intersil releases their specs for their chipsets for WiFi. These chipsets have more WiFi code in BSD and LinuxLand than any other, bar none. Proxim/Lucent/Terabeam/others have huge and cool software basis in the open source world. By contrast, others that mandate you swear fealty and pay staggering amounts of money for code, pragmas, instruction sets, timing info, and so on, get left in the dust.

If you RTFA, you'll find quite a contrasting amount of difference between two top vendors. But read the licenses carefully. Then, where lucky, look up code that others have done before starting to conjure up apps, drivers, and so on. This is the beauty of being open: code, reuse code, share code, improve code, make closed source knotheads look like the idiots they are.

Soft, Hard and Open (1)

Doc Ruby (173196) | about 8 years ago | (#15115712)

"Open" hardware is a design I can download to my FPGA. Covered only by GPL at most - no patents, other copyrights, or other restrictions on use and redistribution. That I can change and redistribute (possibly requiring publishing my changes, as per GPL). There's other kinds of open hardware, but I know the kind that I recognize: the same as open software.

Re:Soft, Hard and Open (1)

Arandir (19206) | about 8 years ago | (#15115935)

I know the kind that I recognize: the same as open software.

You're confusing firmware with hardware. If hardware can be programmed (as with an FPGA), then the firmware is indeed like software. But the hardware itself is still immutable. Most hardware components are not amenable to modification. No matter how much I try, I can't reprogram a SPARC as a PowerPC.

p.s. But I do want the hardware *specification* to be open. That's what this article is about.

Re:Soft, Hard and Open (1)

Doc Ruby (173196) | about 8 years ago | (#15116059)

That's not firmware. Firmware [google.com] is the software stored in ROM, like a BIOS. FPGA is reconfigurable HW, which is open hardware. I'm not going to get into a philosophical argument about whether FPGA HW is "immutable", or whether other HW like even CISC CPUs are "immutable". I pointed out a specific example of HW that is certainly open. Running a CPU on a FPGA, like a MicroBlaze "soft" CPU on Xilinx FPGA (running uCLinux) is "open hardware" by any reasonable definition. If I could download a SPARC or PowerPC specification to an FPGA, that would be really great - and really open.

Re:Soft, Hard and Open (0)

Anonymous Coward | about 8 years ago | (#15116567)

Or you could just run the PowerPC that's already inside that Xilinx (Virtex) FPGA and use the gates for something else! Perhaps that's why Xilinx is a Power.Org member.

Oh -- one more thing -- through Power.Org, there is a royalty-free license to a PowerPC core for academics. It's a real live core with real live support. That can be put onto FPGAs if you want.

Does it matter? (2, Interesting)

metamatic (202216) | about 8 years ago | (#15115798)

I don't honestly see what making SPARC or PowerPC "open" is going to achieve at this point.

SPARC and PowerPC are pretty clearly niche and/or legacy architectures now. IBM has ceded the mainstream desktop to x86, and SPARC lost that battle a long time ago. The only question most people care about now is whether their x86 system is 32 or 64 bit, Intel or VIA or AMD.

Right now, most open source software tends to be tested and hacked on to at least make it run on PowerPC, for the benefit of Mac users. As the PowerPC Mac users switch to x86, who's going to care about PowerPC compatibility? I remember what it was like running Linux on PowerPC before OS X, and it wasn't pleasant--lots of stuff x86 Linux users took for granted didn't work, or you were stuck with old versions, because nobody had bothered to port, test or debug it.

SPARC and PowerPC will continue in the high-end server niche, but I think that niche is increasingly going to be squeezed by x86 too. Why deal with the possible risk of having your enterprise application break on PowerPC Linux, or being stuck with old versions of software, when you could run it on a big x86 Linux system and run the same binary 90%+ of the app's users are relying on every day? Sometimes there's safety in numbers.

PowerPC has the embedded space, of course, and maybe that'll be enough to sustain it as a target for general purpose code. I guess video game toolkits and related libraries will continue to be ported to PowerPC, at least.

But to go back to "openness"--in the embedded and video games space, who cares if the design is "open" or not? All the PowerPC video game consoles are locked down proprietary systems, as are various other embedded PowerPC systems like TiVo and car computers. And in the high end server space, I don't know that anyone cares there either--System i and System z seem to do OK without having open standard CPUs.

[Opinions mine, definitely not IBM's, obviously... and I may be completely wrong, perhaps openness is important in those niches?]

Re:Does it matter? (1)

Mr_DW (894313) | about 8 years ago | (#15116069)

I'm not sure you should be touting the general purpose computer market as anything but niche. There are far more embedded processors then desktop/laptops/servers.

Re:Does it matter? (1)

daverabbitz (468967) | about 8 years ago | (#15118418)

I'm not sure you should be touting the general purpose computer market as anything but niche. There are far more embedded processors then desktop/laptops/servers.

Especially given that most pc's have several embedded computers built into them, think ATA, RAID, FDC (do computer's still have those?), system watch dogs, some lan cards, printers and tv tuners, for a start.

Embedded, Embedded, Embedded!!! (2, Insightful)

W. Justice Black (11445) | about 8 years ago | (#15116089)

SPARC and PowerPC are pretty clearly niche and/or legacy architectures now. IBM has ceded the mainstream desktop to x86, and SPARC lost that battle a long time ago. The only question most people care about now is whether their x86 system is 32 or 64 bit, Intel or VIA or AMD.

Unless we're talking about the 100x or so more machines in the embedded space. Just because the chip isn't in a PeeCee doesn't mean it's not a computer. And embedded designers DO care about this stuff.

Re:Embedded, Embedded, Embedded!!! (1)

metamatic (202216) | about 8 years ago | (#15116568)

Can you give an example of a chip design that has succeeded in the embedded space because of its openness?

Re:Does it matter? (0)

Anonymous Coward | about 8 years ago | (#15116193)

"I guess video game toolkits and related libraries will continue to be ported to PowerPC"
Um... PowerPC is the heart of Xbox360, Playstation 3 (Cell is Power), and GameCube (and Revolution). The port will be from Power to PC, not the other way around... unless you want the bad game performance on a PC to drive your game experience. And Sony has announced Linux on PS3.

System z is not Power. System p and System i are.

Hacker appeal? (2, Interesting)

MoxFulder (159829) | about 8 years ago | (#15116289)

I don't honestly see what making SPARC or PowerPC "open" is going to achieve at this point.


Sure, everybody just wants a nice cheap powerful x86 box these days, even Mac users. But there'd be a lot of hacker appeal to having the source for the processors available. Everybody would tinker with their processors and implement them on FPGA [wikipedia.org] (a cheap way of fabbing chips on a small scale, basically). Instead of people boasting about their tuned kernel, we'd be boasting about our tuned processors. "I got my OpenSPARC running on a Xilinx FPGA and I optimized out the floating point unit so I could add more cache."

And as we've seen with Linux, W3, Apache, etc., hacker projects can turn into BIG business down the road!!

Frankly, I'd switch over to OpenSPARC or OpenPOWER or OpenRISC if I could. I already can and do hack around with the source code to my software, and I'd love to be able to do the same with the hardware...

Re:Does it matter? (1)

backwardMechanic (959818) | about 8 years ago | (#15116678)

It's sad, but for the desktop and server world I'll have to agree with you. The strange thing is the timing. Different architectures have battled it out for years. Finally, with the spread of Linux, BSD, and OSS in general, the underlying architecture makes less difference - I can just port/recompile the bits I want, and my users see the same front. But all the other processors have disappeared, and we're left with x86.

Maybe the Chinese will design something crazy, and radically different...

I care. (1)

twitter (104583) | about 8 years ago | (#15117161)

Right now, most open source software tends to be tested and hacked on to at least make it run on PowerPC, for the benefit of Mac users. As the PowerPC Mac users switch to x86, who's going to care about PowerPC compatibility?

I'm hoping for more, not fewer PowerPC platforms. PowerPC continues to do more per watt than other hardware. It's better for the kinds of small, quiet systems most people really want. The Mac mini is a great example of the kind of system I'd like next. The same things make an attractive laptop. It would be nice to see IBM make Linux hardware and get back into the home market. They have what I want.

Re:I care. (0)

Anonymous Coward | about 8 years ago | (#15117400)

VIA is already doing all that. PPC is nice, but it's not needed. A loss leader for IBM so they probably won't be going there anyway.

I don't see where "Linux" has any importance here. The OS is irrelevant.

Re:I care. (0)

Anonymous Coward | about 8 years ago | (#15117448)

PowerPC continues to do more per watt than other hardware.

If you're talking about the G4/G5 chips Apple was selling, this is not true. Intel had them beat on both performance and watts.

Re:Does it matter? (1)

menace3society (768451) | about 8 years ago | (#15118876)

With Windows goes the x86. If Open Source software (not just operating systems but applications too) really takes off, what particular platform you have won't matter nearly as much, since you can compile your own binary anyway (this of course depends on things being written in a reasonably architecture-independent fashion, which we know doesn't always happen). Since SPARC is apparently free to use, that means that Texas Instruments, Motorola, or one of those guys could start building SPARC chips for calculators, phones, etc. Or a new company could put together a fab plant and pull an AMD on everybody but with a POWER chip.

I think this will begin to make more sense in terms of finances and opportunities as people expect embedded electronic systems to have more features, more power, and the ability to interact better with other items (could you imagine writing a TCP/IP stack with an 8-bit chip? Even 16-bit would be pretty sucky if you have to add in this like IPSec or SSL). Manufacturers will want to use a mature, tried & test 32- or 64-bit design, and if they can license SPARC (or eventually POWER) for free they'll probably go with that. Is it a guarantee? No. But it's a very interesting possibility, one that's appealing enough to justify the move.

Re:Does it matter? (0)

Anonymous Coward | about 8 years ago | (#15119111)

Yes. Because having only 1 CPU ISA is the way to go. One day a critical flaw is discovered and, suddenly, every computer on earth can be pwn3d. Thanks but no thanks. My server at home is a Sun Ultra60 running NetBSD 3.0, and I like it that way. Intel can take their obsolete and stupid ISA and shove it up their ass, I'll keep using reliable hardware in the meantime.

"show me the code" (-1, Flamebait)

Yonder Way (603108) | about 8 years ago | (#15115808)

Sun has contributed very little to the free software community. The only distros that do support Sun can only do so very weakly, due to lack of any material contributions from Sun. No hardware, no documentation, nothing. Aurora, CentOS, Debian, and so on can only half-ass support SPARC at this time.

IBM has a whole division [ibm.com] just for working with the Linux community and making sure that Linux runs well on all of their hardware. Yes, including Power architecture.

So all of you folks running RHEL or SLES (IBM's partners), check out your source RPM's and look at all of the contributions released under the Gnu GPL to make sure that Linux runs like a top on Power architecture.

Re:"show me the code" (3, Informative)

iggymanz (596061) | about 8 years ago | (#15115946)

well, I wish sun would give UltraSparc 3 and 4 docs to the open source OS too, but to say they've contributed very little to open source just isn't true. they've made huge contributions like opensolaris & staroffice (you might not like their licenses but that's another issue)

Re:"show me the code" (1, Interesting)

Anonymous Coward | about 8 years ago | (#15117478)

Uh, the license to OpenOffice.org is LGPL, something the slashdot crowd usually is okay with. Your statement could be interpreted as implying there is something wrong with CDDL; if you believe so, you should state it.

Regarding UltraSparc III and IV, those are older designs and aren't particularly interesting. In fact, they're probably less useful than the the CoolThreads design, because there are a bunch of other Sun-ish things it assumes exist (MP cores are far more about cache coherency and chipset then just straight-line execution). It takes resources to Open Source anything (think about the legal review alone), so it's not surprising that Sun would Open Source whatever it thinks will provide the most impact on the market.

Re:"show me the code" (0)

Anonymous Coward | about 8 years ago | (#15116091)

Sun has contributed very little to the free software community.

Well since Sun has attributed very little that you consider useful, please remove RPC, PAM, StarOffice, Java, Gnome, shadow passwords and any other code and concepts that Sun wrote or contributed heavily to.
Thank you!

Re:"show me the code" (1)

Homestar Breadmaker (962113) | about 8 years ago | (#15116751)

That would be nice wouldn't it. And then if you removed all the shit from GNU, and some more random cruft from who knows where, you might have a reasonable system.

Am I asking too much? (1)

Lord Kano (13027) | about 8 years ago | (#15115968)

All I really want is a simple PPC CPU+Motherboard with AGP and PCI that'll fit into an ATX Case. For less than $600 US.

I am NOT going to pay $2k for a reference board. It seems to me that for all of IBM's talk about openness with respect to the PPC architecture, it doesn't seem to have done an awful lot to bring it to the masses.

LK

Re:Am I asking too much? (0)

Anonymous Coward | about 8 years ago | (#15116805)

All I really want is a simple PPC CPU+Motherboard with AGP and PCI that'll fit into an ATX Case. For less than $600 US.

Then head to eBay: you can get a PowerMac G4 with AGP and PCI for under $200. If you don't like the case and power supply, it'll take you a few minutes with a screwdriver to gut it, and you can probably even resell the case on eBay.

It seems to me that for all of IBM's talk about openness with respect to the PPC architecture, it doesn't seem to have done an awful lot to bring it to the masses.

For all of the complaints about Apple's closed-ness, they've done an awful lot to bring computers to the masses.

Re:Am I asking too much? (0)

Anonymous Coward | about 8 years ago | (#15118890)

For all of the complaints about Apple's closed-ness, they've done an awful lot to bring computers to the masses.

Those two statements have nothing to do with each other. The average user does not care about how open the architechture is. A tinkerer and hobbyist like me does. Apple does not offer anything that suits my hacking needs, therefore I do not use them.

Apple maintains a VERY closed architecture. Steve doesn't want other people playing on his playground.

Re:Am I asking too much? (1)

soupforare (542403) | about 8 years ago | (#15117163)

Cannibalize some old macs.
AGP, PCI, upgradable cpus.
Best of all, when you want to actually get work done, it'll run OSX!

In related news, China now to produce redSPARK cpu (1)

poopie (35416) | about 8 years ago | (#15116036)

... and require that all government computers run RedFlag on redSPARK.

Please post your comments with your Redberry.

Re:In related news, China now to produce redSPARK (0)

Anonymous Coward | about 8 years ago | (#15117632)

"... and require that all government computers run RedFlag on redSPARK."

They can't use the term Sparc, so they are think it to turn it around and name it RedCraps.

Defining "Open" (1)

bradgoodman (964302) | about 8 years ago | (#15116186)

It is right on IBM's site:
Q: The PowerPC specification is open? How long has it been open?

A: It has been open since its inception; the architecture specifies an instruction set architecture (ISA) that allows anyone to design and fabricate PowerPC-compatible processors, and source code for software modules developed in support of PowerPC is freely available.
From: http://www-128.ibm.com/developerworks/power/newto/ [ibm.com] So if by "open", they mean "the architecture specifies an instruction set architecture (ISA) that allows anyone to design and fabricate compatible processors" - - Doesn't that mean EVERY architecture is "open"??

Both of them are right (2, Interesting)

NekoXP (67564) | about 8 years ago | (#15116400)

I work for a Power.org member, so maybe I am biased, but I think OpenSPARC is one of the best things Sun has ever done.

But it's no way as cool as the Solaris port to PowerPC.

Sun is involved in both, you see!

So who cares? They're both right.

Other Open PPC implementations (1)

nurb432 (527695) | about 8 years ago | (#15117254)

Sooooo since power.org really isnt free, as how most of us think of free, is there a free implementation out there somewhere like there is for SPARC ( leon2/3 )?

stop the "open" avalanche (1)

AlgorithMan (937244) | about 8 years ago | (#15118218)

nowadays everyone likes "open" thigs - thats because open-souce software is so nice to its users... but I see more and more products being called "open" although they are only partially open or completely closed, but use open file standards...

I say RMS should patent the word "open" in IT context and only GPLed stuff may be called "open"...

Point of power.org was to sell more CPU's (0)

Anonymous Coward | about 8 years ago | (#15118463)

When power.org started, the execs in charge were thinking to somehow build what Intel had built... a market where outside vendors made the motherboard and IBM just supplied the CPUs and IP to build the motherboard. PowerPC64 Linux was the OS that would be used on this power.org hardware.

However, PPC64 Linux (Bare Metal Linux exempted) is currently dependant on not only having a Open Firmware device tree, but on firmware abstracting much of the HW specific nuances of the system. The ugly nuances of the interrupt subsystem (Apple's U3), for example, forced IBM's recent PowerPC970 based systems into using the LPAR (Logical Partitioning, i.e. hypervisor) architecture. In general, to have a fast enough "time to market", firmware must abstract the funky HW issues from the OS's. Note that the major PPC Linux distributors have >8-12 month integration pipeline, whereas the integration (testing/review) pipeline for firmware on a given system can be 1 month.

A good chunk of the firmware on one of the PPC970 systems will be made public if not already. Theoretically, people could modify this firmare to boot-strap an Open Hypervisor like "Zen". Whether this free firmware actually springs a market, or just becomes another PPC toy dinosaur remains to be seen. I'm fairly skeptical, though, as I've seen that the market for "open systems" only leaps forward when the reference system has a decent price/performance point. IBM's going to have to work harder with it's vendor'd motherboard chipsets to achieve that.

The Premise is Wrong (0)

Anonymous Coward | about 8 years ago | (#15119025)

OpenSPARC is not a response to POWER.org. POWER.org is a response to SPARCInternational. SPARC International is an independent, non-profit organization founded in 1989 to administer SPARC instruction set licenses. SPARC International owns SPARC, not Sun.

OpenSPARC is a separate program, just like LEON is a separate program. OpenSPARC is run by Sun.

"What does 'open' mean in hardware?" (1)

kfogel (1041) | about 8 years ago | (#15119091)

What does 'open' mean in hardware?

Simple: patent-free, or at least patent-unencumbered. Hardware development is
such a minefield of patents that no small player can seriously participate
without getting big allies. Of course, there are many other reasons why
small players would have a hard time, patents are only one. But they are
determinant: a patent-laden "open" hardware spec is not really open. You
either have freedom or you don't, the rest is mere nuance.

Patants and royalties (1)

Horrus (968006) | about 8 years ago | (#15119339)

Hardware is never really free.
If you want to use something like CANbus or I2C you have to pay royalties.
Even MUX'es have to be paid (to generate data/strobe signals in high speed serial lines)

So, Code is free; but if you want to sell it you have to pay royalties.
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...