IBM Opens Up POWER Architecture For Licensing

← Back to Stories (view on slashdot.org)

IBM Opens Up POWER Architecture For Licensing

Posted by Unknown on Wednesday August 7, 2013 @12:48AM from the finally-gonna-get-that-sweet-workstation dept.

New submitter HAL11000 was the first of many to write with news that IBM and others have formed a new consortium to license the POWER architecture to third parties "IBM puts up POWER architecture for licensing and announces the OpenPower Consortium with Google, Nvidia, Mellanox, and Tyan." Quoting El Reg: "The plan, according to McCredie, is to open up the intellectual property for the Power architecture and to allow customizations by licensees, just like ARM Holdings has done brilliantly with its ARM processors ... Nvidia is very excited about the prospects of marrying Power processors and Nvidia GPUs for both HPC and general purpose systems. ... Tyan will presumably be working on alternative motherboards to the ones that IBM has manufactured for its own use." There are mentions of the POWER firmware being "open sourced," but it is unclear if that actually means Open Source or something more like the Open Group's definition of open (vendors only).

131 comments

Min score:

Reason:

Sort:

Phones by Anonymous Coward · 2013-08-07 00:52 · Score: 0

Does this mean I can have a phone that starts out with Android and Maemo backwards compatibility, but I'll lose Android compatibility at the first annual warranty replacement? PS3YLOD!
1. Re:Phones by Therad · 2013-08-07 00:54 · Score: 1
  
  Only if it is a sony-made phone.
A Little Late? by Anonymous Coward · 2013-08-07 00:55 · Score: 5, Insightful

Shouldn't they have done this while Apple was still using the PPC? At number of developers and developer tools available for PPC back then has to be orders of magnitude higher than it is today. Better late than never?
1. Re:A Little Late? by Neo-Rio-101 · 2013-08-07 01:05 · Score: 1
  
  Better late than never..... but better never late.
  Are IBM hoping that people migrate to AIX or something? (good luck with that!)
  
  --
  READY.
  PRINT ""+-0
2. Re:A Little Late? by drinkypoo · 2013-08-07 01:20 · Score: 5, Informative
  
  PowerPC cores are incredibly more ubiquitous than you probably believe. They show up all over the place. Hell, Motorola Cellphones had old tired PPC cores in them for some time, since as a contributor Moto had a license to make embedded PPC chips. And of course, it's well-known that there's a tri-core PPC in the Xbox 360. There's also a castrated little PPC core in the front of the PS3's processor, where there was a MIPS core in the PS2's. And there's a ton of little MIPS-based portable computers out there, but in recent times their sales have been cannibalized by ARM. There's no reason to believe that there couldn't be a ton of little PPC-based portables out there, if PPC were licensed like ARM. Now, allegedly, it will be. Probably too little too late, though.
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
3. Re:A Little Late? by Anonymous Coward · 2013-08-07 01:21 · Score: 0
  
  No, Power was well behind Intel and AMD. Apple had to make the move if their computers weren't going to be significantly slower than entry level PCs and laptops.
  Furthermore, PPC developers rarely hit the metal directly, they use compilers, the tools available a decade ago are still around. Unless Power offers a significant performance advantage (it doesn't), is significantly cheaper (it isn't) or draws very low power (it doesn't), it really doesn't have a chance today. But letting others at the designs may offer interesting shrinkage of expensive and juicy mid-range boxen.
4. Re:A Little Late? by chowdahhead · 2013-08-07 01:51 · Score: 2
  
  I believe many automotive onboard computers are PPC based also.
5. Re:A Little Late? by bill_mcgonigle · 2013-08-07 01:57 · Score: 4, Informative
  
  Are IBM hoping that people migrate to AIX or something? (good luck with that!)
  A few days ago on the Fedora homepage was announcement of the full release of Fedora 19 for IBM Power, presumably with Linux 3.10. You can get RHEL 6 if you want support and certainly there are debian and netbsd ports in various states. If there's a market for the hardware, the software is ready.
  
  --
  My God, it's Full of Source!
  OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
6. Re:A Little Late? by bill_mcgonigle · 2013-08-07 02:03 · Score: 1
  
  And spacecraft because there are rad-hard parts available.
  The thing I wonder about is all the years of ARM driving for low power (Intel has done this too) while IBM Power (uppercase) focused on being really fast and powerful for server work. Either they expect to buy a company that has the expertise to reduce Power's power consumption or they expect one of these companies to license the design and do it themselves, though I'm not sure why Power's architecture is better enough than ARM's for such a company to do so.
  
  --
  My God, it's Full of Source!
  OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
7. Re:A Little Late? by Anonymous Coward · 2013-08-07 02:03 · Score: 3, Interesting
  
  It didn't help that Intel and AMD were burning 200 Watts and PPC was at about 15 Watts and people expected similar performance. Couple that with Motorola sucking wind and a major memory design flaw/feature on an embedded processor (I gave up at the 7447) and it was doomed.
  PPC developers rarely hit the metal directly,
  
  Could not be less true. People liked the PPC because you could get to assembly easily and it was not difficult to squeeze 90% performance out of it. The programming strategy was so easy compared to other options at the time.
  The problem IBM has is that it creates these alliances so it can go through alliance customer lists and get introductions to decision makers. I fear for NVidia and Tyan because they are going to lose business and customer confidence. Once IBM is in the customer's room they will uninvite the alliance members and divide a sale. I've seen it first hand.
8. Re:A Little Late? by Anonymous Coward · 2013-08-07 02:10 · Score: 0
  
  Ubuntu has had an up-to-date PPC distro for some time now.
9. Re:A Little Late? by Anonymous Coward · 2013-08-07 02:27 · Score: 0
  
  POWER was not behind Intel or AMD. In fact, for enterprise server workloads, POWER has and still is ahead of Intel and AMD's high end server CPUs since at least POWER4 (before then I'm not sure, I wasn't paying so much attention).
  Today, POWER7 can develop about twice the throughput per core than the Xeon EX CPUs from Intel, on transaction processing workloads (e.g., TPC-C, SAP-SD). It can scale, gluelessly, far higher, has much better RAS, and does so while having quite comparable energy use per unit of work.
  POWER8 on a more open platform actually could be an interesting development in the data center.
10. Re:A Little Late? by Carewolf · 2013-08-07 02:28 · Score: 4, Informative
  
  POWER has been been 64bit and massively out-of-order superscalar for years where ARM is only just beginning to enter the market. Simply put POWER is not in the same market as ARM, but in the same market as x86. Which means it is Intel (and AMD) who is killing them.
11. Re:A Little Late? by Anonymous Coward · 2013-08-07 02:29 · Score: 1
  
  And of course, it's well-known that there's a tri-core PPC in the Xbox 360. There's also a castrated little PPC core in the front of the PS3's processor
  Whoa there. If you're going to defend PowerPC, you should acknowledge that the PPC core in the Xbox360 is the same core as in the PS3. The PS3's main work is done by the 7 vector processors (alternately called "APU"s, "SPUs", "SPC"s, depending on who's talking), but the CPU core is the same repipelined Power4 with a VMX unit on it. My memory fails me, but there may be a minor change in the VMX for the 360. Microsoft didn't want to spend the time on custom vector units, so they just asked for enough cores off the shelf to meet their objectives.
  Refer to The Race for a New Game Machine for more details if you're truly curious. It was written by two of the lead designers/architects of both CPUs.
12. Re:A Little Late? by robthebloke · 2013-08-07 02:36 · Score: 1
  
  (alternately called "APU"s, "SPUs", "SPC"s, depending on who's talking)
  SPE = Synergistic Processing Elements.
13. Re:A Little Late? by mlts · 2013-08-07 02:47 · Score: 4, Informative
  
  POWER7 has some nice advantages over x86/amd64:
  1: A feature (mainly for database licensing) to turn off half the cores, let the cores working use the cache on the cores turned off, and crank the clock speed up. Performance in that mode is almost the same as turning on all cores, but these results can vary on what is bring run.
  2: Decent bang per watt.
  3: A different CPU architecture with a different set of bugs. This helps for secure applications, so if there might be a F0 0F-like bug lurking around, the bad guys would have to find it for IBM's architecture.
  4: More registers to use and abuse.
  5: Very good virtualization capability. Every POWER7 box thrown out is made from the ground up with a hypervisor built into both FSPs. One can just use a single machine with access to all hardware, or add VIO servers [1] and LPAR it out.
  [1]: VIO servers are small AIX [2] instances that pass disk I/O and networking through to the other VMs. On VMWare ESXi, they would be roughly equivalent to a VM appliance that does routing between virtual switches.
  [2]: More of a variant of AIX, called IOS... however, oem_setup_env gives you a root prompt if needed.
14. Re:A Little Late? by drinkypoo · 2013-08-07 02:52 · Score: 1
  
  Whoa there. If you're going to defend PowerPC, you should acknowledge that the PPC core in the Xbox360 is the same core as in the PS3.
  Not only am I not defending PowerPC or POWER (I give a shit if it lives or dies, which hopefully will happen on its own merits, ha ha) but the PPC core in the 360 is not precisely the same as the one in the PS3, though they're based on the same design. The one in the PS3 is stripped down further and then glued to the vector units. The core in the 360 is faster, there's three of them, they are symmetric and they have slightly more features.
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
15. Re:A Little Late? by Anonymous Coward · 2013-08-07 02:57 · Score: 1
  
  And what has any of that to do with the POWER architecture?
  PowerPC != POWER!
  I thought this was a geek site. Then again, that was a looong time ago.
16. Re:A Little Late? by Narishma · 2013-08-07 03:48 · Score: 1
  
  As far as I know, the only difference between the PPE and a Xenon core is that the later has a modified VMX (AltiVec) unit. They upgraded the vector register count to 128 per thread compared to 32 per thread on the PPE and they replaced a few vector instructions with others that are more useful in gaming.
  
  --
  Mada mada dane.
17. Re:A Little Late? by cant_get_a_good_nick · 2013-08-07 03:57 · Score: 1
  
  Back when Apple first came out with PowerPC (the PowerPC 601 based 6100, 7100, 8100) the Ford EEC controllers were PowerPC 40x cores.
  We had "PowerPC Inside!" marketing stickers (I worked for the campus Apple Reseller) that i wanted to stick on random Ford cars for the hell of it. I never did, didn't want to screw up their paint jobs.
  Im sure there still are PowerPC applications in cars. And think of the 68000/DragonBall. A family that came out in 1979 but was sold until the 2000's as an embedded controller.
18. Re:A Little Late? by bill_mcgonigle · 2013-08-07 04:09 · Score: 1
  
  Ah, good point. Intel is way far ahead in low-power ILP/OOE, but they don't license their architecture, so yeah, there could be a market.
  
  --
  My God, it's Full of Source!
  OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
19. Re:A Little Late? by Anonymous Coward · 2013-08-07 04:13 · Score: 0
  
  Show me the powerpc roadmap and we can talk about advantages. x86, hell even sparc has a decent roadmap these days, there is no roadmap for power because the architecture has no future. They open up the product now because they are aware of it too.
20. Re: A Little Late? by niftymitch · 2013-08-07 05:11 · Score: 1
  
  Spot on. The PPC core is well received in the embedded world. The ARM folk have powered up and moved into space once owned by PPC. With newer process the PPC could move down and put a lot of pressure on ARM.
  The 64bit ARM is a reach.
  Now if someone could build a five watt PPC Raspberry-Pi/BBB with 2x more RAM and GigE for five bucks less the landscape would shift in no time.
  
  --
  Truth is stranger than fiction, but it is because Fiction is obliged to stick to possibilities; Truth isn't. Mark Twain.
21. Re:A Little Late? by Anonymous Coward · 2013-08-07 05:13 · Score: 0
  
  but the CPU core is the same repipelined Power4 with a VMX unit on it.
  It's actually related to the Power6, not the Power4. Power4 was an out-of-order design optimized for instructions-per-cycle like the Power5, while Power6 and PPE/Xenon were optimized for high clock frequencies (a mistake copied from Intel's Pentium 4) and added hardware multithreading. Fortunately, Power6 fixed a couple of the mistakes made in the development of Cell, such as adding back out-of-order execution.
22. Re:A Little Late? by Anonymous Coward · 2013-08-07 05:45 · Score: 0
  
  There is also sparc, where you don't have to pay license fees at all, rendering your argument invalid. Firmware (OpenBoot, OpenBios, ...) is open source too btw. There is just no reason for power anymore.
23. Re:A Little Late? by Anonymous Coward · 2013-08-07 05:48 · Score: 0
  
  I own a PS3, and would not buy one of these again even if I was paid billions. Out of order execution was *NOT* what I saw. I had a 1.8 GHz Pentium4 that I was comparing it to, and the Intel chip (which I at that time considered obsolete), blew the pants off the POWER chip. The Power chip was weak and unable to perform well. These things have life cycles, and I am still stuck with it, having paid for it and waiting for it to die. Every day it becomes less useful though. Show me a cheap tablet and I will replace the over priced, under powered POWER based beast with the tablet, and will have a much better experience. IBM lost this battle about 10 years ago. Power is to IBM what Itanium is to Intel, or what surface is to microsoft. "You can't make me buy this!"
24. Re: A Little Late? by Darinbob · 2013-08-07 06:55 · Score: 1
  
  I do like POWER and PowerPC more than ARM. But ARM get the leg up by licensing to third parties. Still PowerPC still has the edge in higher performance (and power sucking) applications.
25. Re: A Little Late? by citizenr · 2013-08-07 07:15 · Score: 1
  
  Now if someone could build a five watt PPC Raspberry-Pi/BBB with 2x more RAM and GigE for five bucks less the landscape would shift in no time.
  Realtek makes RTD1186 750mhz mips + PowerVR SGX 531 + 1Gbit + USB 3.0 + hdmi + sata + pcie, and it costs $3
  That doesnt mean shit because Realtek is one of those companies (like Broadcom) that never ever give out documentation.
  
  --
  Who logs in to gdm? Not I, said the duck.
26. Re:A Little Late? by Anonymous Coward · 2013-08-07 07:19 · Score: 0
  
  The programming strategy was so easy compared to other options at the time.
  A million times this. Anyone who's seen the mess that is x86 assembler and who has seen PowerPC assembler knows which one is nicer; PPC every time. Are there quirks? Of course. But PPC assembly writing is leaps and bounds nicer than x86.
27. Re:A Little Late? by Jeremy+Erwin · 2013-08-07 07:20 · Score: 1
  
  1: A feature (mainly for database licensing) to turn off half the cores, let the cores working use the cache on the cores turned off, and crank the clock speed up. Performance in that mode is almost the same as turning on all cores, but these results can vary on what is bring run.
  How's that going to be useful in a free software context? It sounds almost akin to a bug.
28. Re: A Little Late? by Anonymous Coward · 2013-08-07 08:44 · Score: 1
  
  The PowerPC in the PS3 was very crippled. It was in-order, and a single core with hyper threading. The idea was that it would only be responsible for setting up the SPUs and making them do all the work. In practice, the core wasn't even up to that in some cases.
  So you definitely shouldn't take the PS3 PowerPC core as an example of a typical PowerPC.
29. Re:A Little Late? by Master+Moose · 2013-08-07 09:35 · Score: 2
  
  Better late than never..... but better never late.
  Better late than pregnant
  
  --
  . . .gone when the morning comes
30. Re:A Little Late? by unixisc · 2013-08-07 11:35 · Score: 1
  
  Wasn't POWER already an open architecture?
31. Re: A Little Late? by drinkypoo · 2013-08-07 12:04 · Score: 1
  
  Realtek makes RTD1186 750mhz mips + PowerVR SGX 531 + 1Gbit + USB 3.0 + hdmi + sata + pcie, and it costs $3
  Cheapest finished product on ebay is eighty bucks...
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
32. Re:A Little Late? by drinkypoo · 2013-08-07 12:08 · Score: 1
  
  Simply put POWER is not in the same market as ARM, but in the same market as x86.
  Well, POWER covers a lot of ground. PowerPC is derived from POWER, and there have long been embedded PowerPC cores. The problem is that they worked on an old-world licensing model where each new product required a new license, and that license pretty much had to be negotiated with Motorola because that's who was doing low-power PowerPC.
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
33. Re:A Little Late? by Anonymous Coward · 2013-08-07 12:11 · Score: 0
  
  You could have always put the stickers in the corner of a window.
34. Re:A Little Late? by Anonymous Coward · 2013-08-07 12:15 · Score: 0
  
  If the kernel detects that the load is low enough, it could start turning off cores to minimize power.
35. Re:A Little Late? by dfghjk · 2013-08-07 12:17 · Score: 1
  
  Apple's computers at the time WERE significantly slower than entry level PCs. That ship had sailed long before.
  PPC was very successful in embedded where developers do "hit the metal directly" so you have no idea what you're talking about. PPC was quite nice to program in assembly.
  You can't license an x86 core from Intel or AMD to integrate into a larger chip design. You can do that with ARM. Power offers a "significant performance advantage" in that respect. It seems you don't get it, so it's a good thing you post as AC. ;)
  Who says "boxen" any more and why did they ever?
36. Re:A Little Late? by Anonymous Coward · 2013-08-07 12:40 · Score: 0
  
  I'm from the other side of the railroad tracks. My guess is if you wanted to get really good performance on Cell or POWER6, you had to rearchitect your code to fit within those processors' constraints. Cell's constraint is to offload to SPE. POWER6 is to minimize stalls to take advantage of the high frequency. It just takes a lot of analysis ability to do so, which is not what you will generally find nor get free from a tool.
37. Re:A Little Late? by drinkypoo · 2013-08-07 13:08 · Score: 1
  
  As far as I know, the only difference between the PPE and a Xenon core is that the later has a modified VMX (AltiVec) unit.
  Well, in the Cell the SPEs are supposed to do the vector processing, so that makes sense. But as I've written here before, it's an extremely puzzling decision on Sony's part. Did they just believe some total bullshit from IBM about how great Cell would be? That would be fairly ironic given the impact of Sony's bullshit about how great PS2 would be on the Dreamcast, to add to the irony of following up a console for which developers complained about difficulty of development due to a wacky architecture (the PS2) with another console with wacky architecture. At least, as I predicted, they did not make that mistake again with the PS4.
  
  --
  "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
38. Re:A Little Late? by Anonymous Coward · 2013-08-07 13:14 · Score: 0
  
  Actually PowerPC!=POWER was a long time ago too, since POWER6 the books were merged and PowerPC is POWER, or rather POWER is the only book left.
  The ISA is the same in all new Power/PPC chips, though clearly the internal architecture is not (130W Server CPU vs. 100mw Digital Camera SoC).
39. Re: A Little Late? by citizenr · 2013-08-07 17:41 · Score: 1
  
  Chip is $3.
  Chinese are able to churn out $40 android tv sticks using $10 Rockchip chips, but are unable to make anything close to that price range using 3x cheaper Realtek - because Realtek hates open documentation.
  This is my point - availability is not enough, products wont just magically happen just because someone is selling cheap chips. Rasppi happened ONLY because people inside Broadcom invested their own time to be a buffer between bunch of closed source dicks and community.
  
  --
  Who logs in to gdm? Not I, said the duck.
40. Re:A Little Late? by Anonymous Coward · 2013-08-07 17:50 · Score: 0
  
  Yea, but why would you? SPARC is SLOW.
41. Re:A Little Late? by unixisc · 2013-08-07 18:27 · Score: 1
  
  But they've been losing a whole market segment that they had initially gained - at the expense of MIPS and Pentiums. Last time, they were there in Xbox 360, PS3 and Wii. Now, Xbox & PS3 are both going AMD, and I dunno about Nintendo. That's not good news for a platform that started out well w/ RS/6000 workstations, PowerMacs and PREP systems, then lost them, then won some game systems, then lost that, and is now there only in some high end IBM servers. Power.org does list some vendors who do make PPCs, but who uses them?
  RTFA, looks like Google will be using them in datacenters, where POWER has improved its power consumption, and will be getting the performance/W that it needs. In terms of the OS, Google can easily port Chromium OS or a server version of it there, and run them on POWER clusters. I'm glad to see a potential revival there - I'm not happy about the x64-ARM duopoly that currently exists. Hopefully, if that catches on, lower power POWER chips can be made to go into at least tablets, if not phones.
  As for 'open sourced', IBM could make some HDL models at older process geometries available - which is providing the source code. Any chipmaker can take that and start working on shrinks, and add logic that either keeps power consumption low, or increases either execution units or L1 cache on the CPU. Open source in these cases can mean that the original designs can have their behavioral HDL models published, w/o publishing their structural modeling, since that is implementation specific and some leeway can be provided to vendors to come out w/ any structural model whose behavioral model matches it. Remember, the HDLs are going to change going from fab to fab, or even different fabs within the same company. So it's good enough if Power.org publishes behavioral models as part of specifications, and then leaves the rest to vendors as implementation specific details.
42. Re:A Little Late? by unixisc · 2013-08-07 18:37 · Score: 1
  
  IBM actually promotes PowerLinux for server, datacenter & cloud solutions. IIRC, AIX is now just there for legacy sites that started off w/ it when it was the only option for the RS6k
43. Re:A Little Late? by Anonymous Coward · 2013-08-07 19:51 · Score: 0
  
  powerpc is still used in network appliances. freescale makes a power based CPU used in a few console servers, for example.
  Just not gonna find that shit in consumer hardware anymore.
44. Re:A Little Late? by Anonymous Coward · 2013-08-07 20:08 · Score: 0
  
  That's because the PS3 was does not execute instructions out of order. This was not a secret. Seriously, if you cared about these details, didn't bother to look them up, and still bought one, you only have one person to blame.
45. Re:A Little Late? by Anonymous Coward · 2013-08-07 20:15 · Score: 0
  
  WTF are you talking about?? Why can't you talk about advantages without referring to a roadmap? Are you some kind of nitwit?
  FYI, you can get access to detailed roadmaps if you are an *actual* customer or prospective customer and sign NDAs. If you aren't, why would they care about you?
46. Re:A Little Late? by Anonymous Coward · 2013-08-08 00:03 · Score: 0
  
  An instruction set is not 'SLOW', it's just the interface of how to control a microprocessor. If you refer to current SPARC microprocessors, well, they are among the fastest in the world so I have no clue what you are talking about. Read up on some numbers and while at it, also read up on I/O performance before you come back.
47. Re:A Little Late? by Plumpaquatsch · 2013-08-08 02:41 · Score: 1
  
  Show me the powerpc roadmap and we can talk about advantages. x86, hell even sparc has a decent roadmap these days, there is no roadmap for power because the architecture has no future. They open up the product now because they are aware of it too.
  Roadmaps are just security blankets for PHBs - all vaporware once showed up on a roadmap.
  
  --
  Of course news about a fake are Fake News.
48. Re:A Little Late? by StuffMaster · 2013-08-10 07:39 · Score: 1
  
  I'm pretty sure the Power architecture was well ahead of Intel and AMD. Apple switched because IBM couldn't/wouldn't get the heat and power draw low enough for laptops.
WAY Too late. by Anonymous Coward · 2013-08-07 00:56 · Score: 0

This has happened way too late.
So many people ditched POWER years ago.
Why do you always make backwards decisions, IBM? What is wrong with you?
POWER, hell, Cell, could have been brilliant if you done this years ago, but you let both of them die slow painful deaths.
Billions lost because of them.
Excellent news! by Anonymous Coward · 2013-08-07 00:58 · Score: 0

Excellent news! I wonder if something made around POWER can challenge x86-64 architecture of general purpose CPUs!
1. Re:Excellent news! by Cid+Highwind · 2013-08-07 03:34 · Score: 2
  
  Possibly, but it better have a snappy marketing name that conveys both it's hardcore POWER roots, and it's friendly personal computer approachability, and it needs to have a short acronym that can fit into the same column width as "x86" and "ARM" on benchmark charts.
  Maybe they could call it "PowerPC", or "PPC" for short...
  
  --
  0 1 - just my two bits
2. Re:Excellent news! by Anonymous Coward · 2013-08-07 04:23 · Score: 0
  
  Why would you need power for that? The hype is with ARM. If you really want another RISC why not go with SPARC? License free, standardized spec and the software ecosystem is equally in place (compilers and such). There are even open sourced IP cores around that should get you started. All that without paying a 3rd party fees ant whatnot.
Let's crowdsource and make a PowerBerryPy by Anonymous Coward · 2013-08-07 01:02 · Score: 0

Let's crowdsource and make a PowerBerryPy before it goes too closed architecture!
We could run Clasic Mac and Sillicon Graphics!
Or maybe Ubuntu Edge could use this awesome architecture!
1. Re:Let's crowdsource and make a PowerBerryPy by arth1 · 2013-08-07 01:34 · Score: 4, Informative
  
  We could run Clasic Mac and Sillicon Graphics!
  Classic Mac and old Silicon Graphics machines did not use PowerPC. They used MC68k CPUs. Later Macs used PowerPC, but SGI never used them, going to 64-bit MIPS CPUs instead.
2. Re:Let's crowdsource and make a PowerBerryPy by Anonymous Coward · 2013-08-07 04:04 · Score: 1
  
  There is OpenSPARC where you get IP cores, you also get the SPARC architecture license virtually for free and don't have to pay anyone if you actually should produce chips. ARM / Power are not open in the same way SPARC is.
3. Re:Let's crowdsource and make a PowerBerryPy by Anonymous Coward · 2013-08-07 06:34 · Score: 1
  
  "Classic" Macs did indeed use PowerPC chips, if by "classic" you mean "Macs using an architecture prior to the current one".
  The originals were Motorola 68k. No Macs used the 68010, AFAIK. Only the Lisa did. But several Macs used the 020, 030, and 040. The Mac III was the last of the non-PPC Macs. The 68k Macs are considered "classic".
  The "PowerMac" generations were PowerPC, and there were 5 generations.
  Those retconned into "G1" used the PPC601. These were the famed "pizza box" Macs. These are considered "classic", since there was a massive performance difference between the 601 and the subsequent generation of PPC chips. Also, NuBus.
  Those in "G2" used PPC603 (integer only, but with power management) and PPC604 (FP coprocessor built-in, power hog). The clones were all in this generation. Clones are all "classic" since the program ended so abruptly. Other "G2" Macs are considered "classic" because they predate the G-whatever naming convention.
  Then the PowerMac G3 came out. This used the PPC750, which was built on the principles of the 603, but included a scaled-back FPU. Between the beige and blue/white releases of these Macs, there was a change in the OpenFirmware spec that resulted in a divide between "new world" and "old world" Macs. This caused "old world" PPC Macs (G1, G2, beige G3) to become "classic". To this day, you can't install Debian on an "old world" Mac. The blue/white models became "classic" when they were the last model without SIMD instructions needed to make slow-ass Macs capable of running any kind of heavy-computation (mostly graphics processing) software.
  "G4" was the MPC7400, which was an overclocked G3 with some SIMD instructions tacked on. It was only made by Motorola. It became a "classic" several times. First, the "Yikes" G4 was a blue/white G3 motherboard with a G4 chip on it. It was instantly a black sheep and became "classic" (read: old) within a few months of release. Second, the single processor G4's became "classics" when the dual processor ones were released. Third, the dual processor ones became "classics" when the gigahertz barrier was broken. They all became "classics" when the Intel Macs were released.
  "G5" was the PPC970. It's "classic" because it's not x86.
  The first Intel Macs were based on the Core architecture, and were 32-bit only. They were "classic" on release day, since the G5 had already ushered in the 64-bits-are-better-than-you attitude a couple of years before.
  Everything after that is probably not "classic"... yet.
  There are a lot of "classic" Macs, and for various reasons. Long-term product support is not Apple's strong suit.
4. Re:Let's crowdsource and make a PowerBerryPy by Anonymous Coward · 2013-08-07 07:23 · Score: 0
  
  +10 internets to you. Accurate information, if not terribly relevant to the subject.
5. Re:Let's crowdsource and make a PowerBerryPy by Anonymous Coward · 2013-08-07 08:14 · Score: 0
  
  Those retconned into "G1" used the PPC601. These were the famed "pizza box" Macs. These are considered "classic", since there was a massive performance difference between the 601 and the subsequent generation of PPC chips. Also, NuBus.
  PPC601 was used in more than the Power Macintosh 6100, which is the thick pizza box form factor model. The was also the 7100 desktop and 8100 mini tower.
  
  Those in "G2" used PPC603 (integer only, but with power management) and PPC604 (FP coprocessor built-in, power hog). The clones were all in this generation. Clones are all "classic" since the program ended so abruptly. Other "G2" Macs are considered "classic" because they predate the G-whatever naming convention.
  PPC603 has hardware FPU. PPC604 has power management. Differences are in microarchitecture, speeds, etc..
  
  Then the PowerMac G3 came out. This used the PPC750, which was built on the principles of the 603, but included a scaled-back FPU. Between the beige and blue/white releases of these Macs, there was a change in the OpenFirmware spec that resulted in a divide between "new world" and "old world" Macs. This caused "old world" PPC Macs (G1, G2, beige G3) to become "classic". To this day, you can't install Debian on an "old world" Mac. The blue/white models became "classic" when they were the last model without SIMD instructions needed to make slow-ass Macs capable of running any kind of heavy-computation (mostly graphics processing) software.
  You can install Debian on an OldWorld Mac. You may need an old version of Debian and a kernel that was still supported by BootX.
  
  "G4" was the MPC7400, which was an overclocked G3 with some SIMD instructions tacked on. It was only made by Motorola. It became a "classic" several times. First, the "Yikes" G4 was a blue/white G3 motherboard with a G4 chip on it. It was instantly a black sheep and became "classic" (read: old) within a few months of release. Second, the single processor G4's became "classics" when the dual processor ones were released. Third, the dual processor ones became "classics" when the gigahertz barrier was broken. They all became "classics" when the Intel Macs were released.
  There were two fairly different microachitectures subsumed under the G4 moniker.
  Apple sold single processor and dual processor versions all along the Power Mac G4 product line (except on the first model). Dual-processor 1.25 GHz MDD, for example.
POWER PowerPC by Anonymous Coward · 2013-08-07 01:04 · Score: 0

So it's fine, there never was a community in the first place.
Compiler support good for general PPC? by Anonymous Coward · 2013-08-07 01:10 · Score: 1

I recently got computing time on a BlueGene/Q (PowerPC A2) and I needed to run my C++ program on it. The compiler support was atrocious. It uses OpenMP for parallelization. GCC is damn slow on it and LLVM does not support parallelization using OpenMP. The IBM in-house compilers are crappy too for anything besides Fortran or baseline C.
My question is: Is it like that also for the more "general purpose" PowerPCs? If yes, I really hope nobody licenses it. IBM supercomputers really do not deserve the TOP XXX titles they get, unless IBM commits to develop better compilers or better yet: scrap vacpp / xlc++ and just start fresh with LLVM.
Posting as AC, as I do want to get computing time on other clusters still.
1. Re:Compiler support good for general PPC? by elfprince13 · 2013-08-07 02:02 · Score: 2
  
  From the perspective of most supercomputer users, the problem here is that you want to use something which is not Fortran or C. Very little hardcore numerical stuff is done outside of those two languages. I'm not saying that's a good thing, but it is the reason why there's so little incentive to have good compiler support.
2. Re:Compiler support good for general PPC? by Anonymous Coward · 2013-08-07 02:10 · Score: 0
  
  Is it like that also for the more "general purpose" PowerPCs?
  
  The problem is not so much the compiler. PPC compile is really quite straightforward (I used to do embedded PPC software)
  The issue you are seeing in compile times is the linker trying to make sense of your C++ objects.
  I used to use PPC and the Greenhills compiler environment. The compiler worked fine and was pretty quick. The linker took *hours* for something gcc would take perhaps one minute on.
  Check your linker settings, especially if you are going for a compact executable.
3. Re:Compiler support good for general PPC? by Anonymous Coward · 2013-08-07 02:26 · Score: 0
  
  I used to use PPC and the Greenhills compiler environment. The compiler worked fine and was pretty quick. The linker took *hours* for something gcc would take perhaps one minute on.
  Check your linker settings, especially if you are going for a compact executable.
  You were probably having the Green Hills compiler do link-time optimization over your entire project.......
  You can make gcc do the same nowadays.........
4. Re:Compiler support good for general PPC? by robthebloke · 2013-08-07 02:39 · Score: 1
  
  The compiler support was atrocious.
  Visual C++ is the best compiler I've used for PPC. (It's a shame that's not available outside of the 360 devkit).
5. Re:Compiler support good for general PPC? by Funk_dat69 · 2013-08-07 03:54 · Score: 2
  
  You can get the 'advance' toolchain. It's basically an upstream gnu toolchain, with a lot more optimizations and support for ppc chips.
  Looks like LLVM is getting improved as well.
  
  --
  FUNK!
6. Re:Compiler support good for general PPC? by Anonymous Coward · 2013-08-07 04:52 · Score: 0
  
  Yes, now. But at the time gcc was quite awful on C++ linking. When embedded PPCs only have 32 MB of RAM and the c++ compiler with a few templates spits out a 15 MB file, we tried to optimize for size. Funny thing, the gcc compiler at 1.95 and linker *in C* without the templates put out a 2 MB file for the same code.
  Most people think the compiler/linker is a done deal but there are still tradeoffs.
7. Re:Compiler support good for general PPC? by file_reaper · 2013-08-07 07:09 · Score: 1
  
  The standard compiler for the PowerPC Architecture is xLC which is made by IBM.
  
  xLC is supported on AIX and Linux and it has all the optimization features needed for BlueGene/Q and other big iron (inter-procedural analysis, auto-vector optimizations etc...)
Nvidia... by MachineShedFred · 2013-08-07 01:22 · Score: 5, Funny

Nvidia is very excited about the prospects of marrying Power processors and Nvidia GPUs for both HPC and general purpose systems.
Nvidia hasn't quite figured out how to get their thermal energy per square centimeter to the level of a nuclear reactor, so I'm sure opening up the POWER series of chips has them quite excited on that front.

--
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
1. Re:Nvidia... by cold+fjord · 2013-08-07 02:28 · Score: 2
  
  Kirk: Do you think it will work?
  Spock: It will depend on what Mr. Scott can coax out of the systems.
  Kirk: Scotty, Spock thinks that if we can boost the precision of the sensors and overlay the data on the navigation computer we may be able to navigate through the interphase rift to escape the Tholian web. Can you do it?
  Scotty: Aye Captain. With that last maintenance overhaul at Star Base 11 our computers were updated with the new Multitronic GPU processors. For once I have the power.
  Power, the next frontier. These are the stories of the GPU maker Nvidia. Its 5 year mission: to boldly show what no graphics board has shown before.
  Brought to you by: Interplanetary Business Machines. When the label says Power by IBM, you'll know the performance will be out of this world.
  
  --
  much of left-wing thought is a kind of playing with fire by people who don't even know that fire is hot - George Orwell
2. Re:Nvidia... by Anonymous Coward · 2013-08-07 02:37 · Score: 0
  
  Waiting for that 10000 pin package of POWER/Nvidia CUDAness. DIY requires an industrial presser for mounting the CPU on the motherboard. Perfect for that bedroom gaming/hpc/render machine. Every morning will be like waking up half cooked.
3. Re:Nvidia... by Desler · 2013-08-07 05:16 · Score: 1
  
  Since when are two numbers "comparable" when one of them is 54 to 92 percent larger?
4. Re:Nvidia... by iroll · 2013-08-07 06:10 · Score: 2
  
  When you're using a log scale?
  
  --
  Repetition does not transform a lie into the truth. - FDR
5. Re:Nvidia... by Desler · 2013-08-07 06:12 · Score: 1
  
  LOL
6. Re:Nvidia... by Anonymous Coward · 2013-08-07 08:19 · Score: 0
  
  If I understood correctly, the GP said the CPU that uses twice the power also is twice as fast.
7. Re:Nvidia... by Anonymous Coward · 2013-08-07 13:31 · Score: 0
  
  When the one that is 54-92% larger has twice the throughput? Actually I'm just restating the GPs post, because you have reading comprehension issues.
  But if what he states is true: 200W with twice the throughput vs. 130W, the Power is 30% more efficient than the Xeon. Real numbers may differ.
8. Re:Nvidia... by Anonymous Coward · 2013-08-07 20:29 · Score: 0
  
  "comparable energy consumption per unit of work", you fucking nincompoop.
  In the amount of time you thought you were so smart calculating the WRONG percentages, you could have just bothered to actually read the post. Then, instead of outing yourself as a moronic, useless, nitpicking retard, you might have actually learned something.
  You are what is wrong with this site.
IS OS\2 NEXT ?? by Anonymous Coward · 2013-08-07 01:23 · Score: 0

Say, say, say !! 16 bit can live on !!
1. Re:IS OS\2 NEXT ?? by cold+fjord · 2013-08-07 02:53 · Score: 1
  
  Actually OS/2 was 32 bit for quite some time before IBM discontinued it and Serenity Systems picked it up as eComStation.
  
  --
  much of left-wing thought is a kind of playing with fire by people who don't even know that fire is hot - George Orwell
2. Re:IS OS\2 NEXT ?? by unixisc · 2013-08-07 18:49 · Score: 1
  
  OS/2 work is already underway as osFree, which uses the L4 microkernel at the bottom, and includes a PM personality. It can be a good OS for a POWER based computer.
CPUs by Google... by peppepz · 2013-08-07 01:31 · Score: 2

...that spy^H anonymously profile your behaviour at the microcode level? I'll pass, x86 SMM is already evil enough for me.
1. Re:CPUs by Google... by bunratty · 2013-08-07 02:34 · Score: 1
  
  If you wanted to subvert a processor in that way, it would be easiest to add a secret knock that would allow the attacker to run any code at all on it. And any CPU you buy can have such a secret knock built into it. I don't see any particular reason to trust one CPU manufacturer more than another. If you have some evidence that one is untrustworthy, please share it.
  
  --
  What a fool believes, he sees, no wise man has the power to reason away.
2. Re: CPUs by Google... by peppepz · 2013-08-07 09:12 · Score: 1
  
  My comment was tongue-in-cheek. All current mainstream processors already have operating modes that allow the firmware to run code without the OS knowing (or having a way to know). But then no CPU vendor is, yet, in the market of collecting and analysing personal information.
"Open Systems" by khundeck · 2013-08-07 01:32 · Score: 1

I use a lot of IBM software and hardware on a daily basis. I /really/ feel like this is more of a 'corporate alliance' than an 'opening up' of their 'intellectual property."
I guess I just dislike the fact it's called the "OpenPower Consortium". Somehow I feel it dilutes the word "open", which has a lot to free/libre.
KPH
1. Re:"Open Systems" by serviscope_minor · 2013-08-07 01:50 · Score: 2
  
  Somehow I feel it dilutes the word "open"
  You're thinking about it the wrong way. Consider instead:
  IBM are open to the idea of taking your money.
  That is the fundemental idea of openness behind it.
  
  --
  SJW n. One who posts facts.
2. Re:"Open Systems" by iggymanz · 2013-08-07 02:42 · Score: 1
  
  no, it's more like the open API of Unix, open specs of Sparc. IBM finally doing something Sun had success with decades ago but not useful marketing ploy now. Too little way too late.
3. Re:"Open Systems" by Anonymous Coward · 2013-08-07 04:07 · Score: 0
  
  There was/is OpenSparc where you can actually get GPL'ed IP cores. I agree that the buzzword 'open' is quite misleading here.
Crawley wasn't wrong by Anonymous Coward · 2013-08-07 01:34 · Score: 0

Maybe that's why they're furloughing the entire Server and Technology group for the last week in August... they'd rather have have others make their processors than make them themselves
The AIM alliance is back! by MacColossus · 2013-08-07 01:39 · Score: 2

I can't wait for the return of Motorola Starmax, Umax Supermac and Power Computing's Power Tower Pro. I remember my Power Tower Pro was upgradeable to 1 GB of ram in 1997! Shut up and take my money!
1. Re:The AIM alliance is back! by Virtucon · 2013-08-07 03:17 · Score: 1
  
  Great hardware until Apple killed them.
  
  --
  Harrison's Postulate - "For every action there is an equal and opposite criticism"
2. Re:The AIM alliance is back! by unixisc · 2013-08-07 18:51 · Score: 1
  
  Actually, when Apple killed that business, they should have ported BeOS to those boxes, and sold them. At that time, they still had a chance, given that Apple was still running System 7 based OSs on the PowerMacs. Instead, they simply folded completely. Another disappointment was Be discontinuing the BeBoxen
Power Licensed by LoRdTAW · 2013-08-07 01:42 · Score: 4, Informative

The headline and summary are confusing, Power is licensed and Power based chips are produced by third parties. Applied Micro (AMCC) along with Freescale make power core based CPU's/SoC's for embedded use and Xilinx has power cores in their high end Virtex 5 FPGA's. A-EON uses the AMCC Power CPU on mATX motherboards for modern Amiga systems. What they mean is that IBM is making it easier for others to license and adopt Power for their needs. Though the Gamecube, Wii, Wii-U, Xbox 360 and PS3 use power processors, they are all made by IBM like the Apple Power CPU's.
Its good to see more RISC architectures that have been around for a while becoming more popular. The mobile market pretty much bought RISC back into the spotlight and is giving x86 a run for its money. And more interesting are the partners and the task Power is looking to solve: the cloud (I feel dirty using that phrase). Intel better watch out, with everyone pushing software as a service and mainfr^H^H^H cloud computing, companies are looking to create hardware targeted towards those tasks while also reducing power.
1. Re:Power Licensed by afidel · 2013-08-07 02:01 · Score: 1
  
  Is this about POWER or PowerPC? There's a significant difference and I seriously doubt IBM is going to risk that sweet, sweet large system revenue by allowing others to produce POWER based CPU's.
  
  --
  There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
2. Re:Power Licensed by SoupIsGood+Food · 2013-08-07 02:01 · Score: 2
  
  I believe the intention is to make the high-end POWER chips more ubiquitous in the server room - heavy duty RISC/Unix(Linux) server platform. The intention is to squeeze x86 out of the datacenter with AMD systems at the low end, and POWER-based gear for the serious number crunchers.
  IBM hopes that by bringing competitors into their platform, they can use economies of scale to make their systems more cost-competitive, and name recognition to separate themselves from the other POWER platform providers. Reduce the cost advantage of Intel-based gear and eliminate the single-source disadvantage of POWER, and they'll put some more daylight between themselves and HP and Dell, their closest rivals.
3. Re:Power Licensed by LoRdTAW · 2013-08-07 03:22 · Score: 2
  
  Its confusing because different Power architecture versions may support both PowerPC and Power ISA's or PowerPc or Power only. Power 1/2 evolved into PowerPC which was renamed to Power ISA. Both Freescale and AMCC call their processors Power processors and support the Power ISA v.2.03 spec which also supports PowerPC. Newer Power ISA versions are called both PowerPC and Power, e.g. CPU's which comply with Power ISA v.2.05 are called POWER6 and the PowerPC 476. The latest power spec, Power ISA v.2.07, does not have a PowerPC name. So it can be confusing. Bottom line is PowerPC and Power are very close and in some cases interchangeable depending on how the code was compiled.
  From Wikipedia:
  
  PowerPC (short for Performance Optimization With Enhanced RISC â" Performance Computing, sometimes abbreviated as PPC) is a RISC instruction set architecture created by the 1991 Appleâ"IBMâ"Motorola alliance, known as AIM. PowerPC, as an evolving instruction set, has since 2006 been renamed Power ISA but lives on as a legacy trademark for some implementations of Power Architecture based processors.
  and:
  
  PowerPC is largely based on IBM's earlier POWER instruction set architecture, and retains a high level of compatibility with it; the architectures have remained close enough that the same programs and operating systems will run on both if some care is taken in preparation; newer chips in the POWER series implement the full PowerPC instruction set.
4. Re:Power Licensed by Anonymous Coward · 2013-08-07 06:40 · Score: 0
  
  Apple's PPC CPU's weren't all made by IBM. I had a beige G3, and distinctly remember being disappointed that my particular model had the Motorola-manufactured chip in it, which had aluminum traces. That meant it couldn't be overclocked as much as the IBM-manufactured ones with copper traces.
  In general, Motorola (now Freescale) was the crappy PPC manufacturer.
  Also, Freescale (formerly Motorola) doesn't have a "license" to the POWER or PPC architectures. They were part of the design team and hold "equal" (big fat air quotes there) terms to the technology. It's technically as much theirs as it is IBM's.
How is this any different from by Anonymous Coward · 2013-08-07 01:42 · Score: 0

1. power.org
2. any of the existing u-boot / Linux / glibc / etc
3. there was the PowerOpen association 20 years ago, so I guess they had to call it OpenPower this time around?
No... by Anonymous Coward · 2013-08-07 02:14 · Score: 0

It's about POWER. That's the big announcement. I think IBM is realizing that such an ecosystem really won't be much a threat to their bread and butter since that largely hinges upon AIX, IBM i, and related. The POWER architecture is decent and all, but clones aren't exactly going to poach the niche still willing to pay for IBM in that space. It may, however, improve relevance of IBM's P hardware to more markets.
I actually think this could be a smart move for IBM, if not an ominous sign for IBM employees.
Day late, .... by nbritton · 2013-08-07 02:20 · Score: 0

POWER support is dead on all enterprise Linux distributions, Red Hat dropped support with EL5. Furthermore OpenPower boxes are contractually prohibited from running AIX.
You've got a box of hardware with nothing to run on it and it can only deliver half the performance of comparatively priced Intel equipment. If you outsource support to IBM, their support specialists in the delivery centers will accidentally nuke your whole frame during routine maintenance, and you could be down for days. If you can manage to find an engineers and administrators capable of maintaining the systems, they don't want anything to do with it because their smart enough to know the platform is walking dead.
1. Re:Day late, .... by the+eric+conspiracy · 2013-08-07 03:19 · Score: 1
  
  Hey, I still have some Mac System 9 floppies lying around. This should run those, right?
Build a micro-itx motherboard with one soldered on by Marrow · 2013-08-07 02:28 · Score: 1

You want to start a revolution, make the damn equipment simple and available.
Build it and they will come.
WRONG! by Funk_dat69 · 2013-08-07 02:46 · Score: 4, Informative

Impressive. You are wrong on just about *everything* you wrote:
>>POWER support is dead on all enterprise Linux distributions, Red Hat dropped support with EL5.
Nope and nope and nope
>>Furthermore OpenPower boxes are contractually prohibited from running AIX.
You are confusing this announcement with a previous attempt at the Linux market that was also called OpenPower. Those systems only ran Linux and could not run AIX. This announcement is about opening up the entire platform and licencing out parts or whole cores of the actual high end chips to companies like Google, who recognize that the single most expensive component in servers is the CPU - and they want choice and customization.
>>You've got a box of hardware with nothing to run on it and it can only deliver half the performance of comparatively priced Intel equipment.
The recently released Power7+ chip running Linux is the fastest thing on the market right now.
>> If you outsource support to IBM, their support specialists in the delivery centers will accidentally nuke your whole frame during routine maintenance, and you could be down for days
Umm..ok I'm stopping now

--
FUNK!
1. Re:WRONG! by Anonymous Coward · 2013-08-07 16:01 · Score: 0
  
  >>You've got a box of hardware with nothing to run on it and it can only deliver half the performance of comparatively priced Intel equipment.
  The recently released Power7+ chip running Linux is the fastest [sap.com] thing [spec.org] on [spec.org] the market right now.
  The key phrase you neglected to comprehend is "comparatively priced". POWER might be twice as fast. *Might* be, but also costs 10x for an equivalent system (based on equivalent RAM sizes on x86 and POWER services, webprices). It is not 10x faster, and so in no way can be considered equivalently priced. Also all the references you quote are for database transaction performance, where there is no evidence that the database is the same codebase or has the same level of optimisation on different platforms.
  Further anyone running SAP is on a mission to fail. I wouldn't take anything they say seriously, they are liars and they are responsible for more business failures than any other company in the world.
Is POWER dying? by Anonymous Coward · 2013-08-07 02:56 · Score: 0

How late is POWER 8 again? Is there even a roadmap?
Companies like IBM would not open up products when they can or want to control a market. We remember the OpenSPARC project which basically marked the end of Suns success with that architecture. The delays of POWER 7+ and 8 may be due to the fact that strong competition with the SPARC T4 and T5 was unexpected and the designs had to be moved back to the drawing boards. Is POWER in trouble? Even SPARC looks much more promising at the moment, Oracle has a solid roadmap and proved surprising leaps in the past three years. The attention they got showed that these improvements were completely unexpected. With their software portfolio and their insane reliability there are even lots of customers that buy SPARC for a reason. There is no reason to choose the powerpc architecture over x86, ARM or SPARC anymore which makes this move look like the last resort.
1. Re:Is POWER dying? by Virtucon · 2013-08-07 03:11 · Score: 1
  
  Just like Sparc, the Power architecture is dead and Itanium will follow shortly thereafter. Why? Because the X86 architecture and X86-64 specifically has won the marketplace in terms of compatibility and cost. Sure, I may get 8 billion threads on a Sparc III but it's slow and for the price I can get a few few X86-64 boxes. ARM is now knocking on the door of the Data Center and we then may see a rush of specialized, disposable servers that are suited for a small number of purposes, highly optimized and which point X86-64 architectures may still have a niche but not as big of one moving forward.
  
  --
  Harrison's Postulate - "For every action there is an equal and opposite criticism"
2. Re:Is POWER dying? by Anonymous Coward · 2013-08-07 03:42 · Score: 0
  
  x86 is miles away from reliability and the remote capabilities that Sparc platforms offer. As for ARM it will take a long time to be taken serious in this market. There are in fact many areas like finance, telecom., ... where Sparc will stay in business for a very long time and will remain a heavy hitter. Not to mention the countless shops that cannot simply switch away from it due to software requirements. Sparc is declared dead every year by some folks and each year it continues to surprise us. Oracle is pushing hard with the T6 in the pipe and Fujitsu is also still pumping money into it. With a true open model (no license fees at all) there are even lots of possibilities for unforeseen surprises. If NVidia really wants to do such a chip, go for Sparc and you don't have to pay anyone for a license...
3. Re:Is POWER dying? by Anonymous Coward · 2013-08-07 04:14 · Score: 0
  
  Just like Sparc, the Power architecture is dead and Itanium will follow shortly thereafter. Why? Because the X86 architecture and X86-64 specifically has won the marketplace in terms of compatibility and cost. Sure, I may get 8 billion threads on a Sparc III but it's slow and for the price I can get a few few X86-64 boxes
  Yeah, and you're a fucking moron... back to the basement with ye. As a sibling to this post states, "86 is miles away from reliability and the remote capabilities that Sparc platforms offer".. and I'd extend that to Power as well. In organizations where money is no object and reliability is key, x86 is a wannabe. For instance, I have an acquaintance who is a sixty year old field engineer still supporting these, and the companies using those systems have no intention of retiring them. Hell, they're having trouble finding a non-Wintel weenie that wants to learn the architecture despite the fact that they're willing to pay a shitload for somebody to do so.
4. Re:Is POWER dying? by Anonymous Coward · 2013-08-07 08:48 · Score: 0
  
  As someone fairly familiar with the nonstop market, to say its still around is to ignore the fact that its dying fast.
  HP probably could have turned into into the next zseries, but instead refuse to invest in it, and are doing their best to send mixed signals about its future. In the end customers are bailing from it as fast as they can port those legacy apps.
5. Re:Is POWER dying? by Anonymous Coward · 2013-08-07 16:17 · Score: 0
  
  Bullshit.
  Reliability: I've had thousands of hardware failures in my time, and I've barely seen a CPU fail (from any Vendor). Since CPUs are the most reliable part of every computer, the architecture of the system has no effect on system reliability. Yea, there are dodgy system vendors, and yea Sun used to make good reliable hardware (BTW, they made good reliable x86 hardware too), this has nothing to do with SPARC or x86 or POWER.
  Remote capabilities: Again, bullshit. Sun x86 and SPARC servers for years have had the same ILOM system (which sucks, btw). Dell, Fujitsu, IBM, HP all have remote platform management. For what it's worth, Dell has the best remote management system I've seen, and they only make x86 servers.
  Also SPARC chips are SLOW, they were well behind the curve when Sun crashed and burned and threw Opensparc out the window before they hit the ground. Now the architecture is so behind Intel, AMD and IBM that you would have to be the most retarded kind of fanbois to even consider buying SPARC (from Fujitsu, because doing any kind of dealing with Oracle is business suicide).
  Unfortunately, in the enterprise space, there are huge numbers of fanboy-type managers who will carry on buying obsolete, uncompetitive and dead technology out of some sick nostalgia or corporate loyalty. This is the kind of disease that keeps sales of FiberChannel up, kept Token Ring and FDDI around long after it started smelling bad. These types are also extremely unethical "professionals", in that they put the interests of their pet tech vendor ahead of their own employer, the organisation they are supposed to be working for. I have personally worked for this type, and seen the kind of reckless budget expansions and ultimate business failures their fanboy-reasoning and rejection of objective reality, metrics and measurements leads to.
6. Re:Is POWER dying? by unixisc · 2013-08-07 20:49 · Score: 1
  
  SPARC has a rich suite of legacy Solaris applications, but Linux support for SPARC has dropped - RHEL no longer supports it, even OEL doesn't - one has to look to the likes of either Debian or OpenBSD. As a result, SPARC has been losing support, except for Oracle only houses.
  As for POWER, it's primarily an IBM platform for not just AIX but also i and other mainframe platforms. But even IBM has embraced Linux, promoting it for their servers, datacenters and cloud computing. Does IBM's POWER based HPC solutions use AIX or Linux? Either way, IBM's bets are hedged - AIX is still there for their legacy customers, while Linux is there for current users. If only SCO had been as wise and maintained OSE & UnixWare alongside Caldera OpenLinux.
  But then again, SPARC is more open in terms of no license fees needing to be paid to Oracle or anyone - something that OpenPower would have to do to catch up. MIPS too has to figure out a way to regain its popularity - in fact, it's the platform best capable of challenging ARM on the low power turf.
7. Re:Is POWER dying? by unixisc · 2013-08-07 20:51 · Score: 1
  
  HP's mistake was acquiring the likes of Compaq & Tandem. Or, when they acquired Compaq, they should have spun off the DEC portion of it, so that they could have at least serviced their existing clientele and kept the OVMS/Alphas alive. Deciding to migrate to OVMS was the worst decision they made.
8. Re:Is POWER dying? by Anonymous Coward · 2013-08-07 23:59 · Score: 0
  
  You cannot seriously use 'Dell' and 'reliable' in the same sentence. All jokes aside you simply cannot get the I/O performance of Oracles SPARC infrastructure with anything x86. As for 'SPARC chips are SLOW', apparently you are also unaware of the SPARC T4, T5, M5, ... systems of the past 24 month. 3 Years ago I too said SPARC is slow and dead, but it's a totally different game now.
IBM won't ever change! by aglider · 2013-08-07 03:12 · Score: 1

It's loo late, dude.
25 years too late.

--
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
Re:How is this any different from power.org by Funk_dat69 · 2013-08-07 03:15 · Score: 2

Power.org manages the ppc ISA.
The ppc ISA is and has been completely open. You can design and create your own chips based on it and people do.
This is about opening up, or decomposing, the development of the high end POWER chips that IBM develops. Large data center companies have an increasing desire for customized chips. Customizing chips is not what Intel is good at or want to be good at. The only game box win that Intel had was the original XBOX and that was a massive failure, partly because of the inflexibility of Intel and their precious margins. I really can't think of any other custom wins that Intel has had since.
If Google wants to cobble together a small 2-4 core Power Chip with exactly the parts they need, based on licensed pieces from IBM, and then go fab it at wherever is cheapest, I've got to think that will save them money versus being a mountain of retail Intel chips.

--
FUNK!
What are you smoking? RHEL6 supports IBM Power by Chirs · 2013-08-07 03:30 · Score: 1

http://www.redhat.com/products/enterprise-linux/for-ibm-power/
Does it mean by Anonymous Coward · 2013-08-07 04:24 · Score: 0

the sparc architecture is on its deathbed?
1. Re:Does it mean by Anonymous Coward · 2013-08-07 04:36 · Score: 0
  
  look at the roadmaps of power and sparc, you'll see that power is dead and sparc not so much.
YUM, JFS, NUMA, DB2 by emil · 2013-08-07 04:43 · Score: 2

A few words about Linux technologies that originated from solid positions within the IBM camp...
YUM (Yellowdog Updater, Modified) Yellowdog is a well-known RPM-based Linux distribution for the POWER architecture. JFS The OS/2 native filesystem was incorporated into Linux and released to production in June, 2001. NUMA IBM's acquisition of Sequent eventually led to NUMA code releases for the kernel which have been particularly appropriate for Hypertransport and QPI - high-performance Linux ows much to IBM. DB2 While not a free product, the UDB database is likely the largest competetor/option to Oracle on Linux.
Linux owes a great deal to IBM.
Re:Build a micro-itx motherboard with one soldered by Anonymous Coward · 2013-08-07 05:07 · Score: 0

how will that compete with Intel and AMD?
Re:Build a micro-itx motherboard with one soldered by swamp+boy · 2013-08-07 06:28 · Score: 1

and affordable!
POWER, Power, and PowerPC by Guy+Harris · 2013-08-07 06:54 · Score: 1

IBM and company's use of variously-capitalized forms of "Power" can be a bit confusing. When the RS/6000s first came out, IBM described the instruction set architecture they implemented as POWER, for "Performance Optimized With Enhanced RISC"; see the "IBM POWER Instruction Set Architecture" Wikipedia article and its references. Starting with the second-generation RS/6000 processor, they started naming the processors "POWERn" as well.
PowerPC was an instruction set architecture based on the POWER ISA; a few instructions were removed, and a number were added; more were added to PowerPC over time. The POWER3 processor implemented the full 64-bit version of PowerPC, and I think it also implemented some of the POWER instructions removed from PowerPC. PowerPC ended up getting renamed "Power ISA" - not to be confused with the all-caps "POWER ISA" mentioned earlier - as part of the "Power Architecture".
I don't know what stuff this consortium is dealing with. There's already Power.org for the Power Architecture, including the Power ISA. I'm guessing that this is for licensing the microarchitecture of the POWERn microprocessors; that seems to be what some of the articles are saying. Then again, some articles are calling it OpenPOWER and other articles are calling it OpenPower, so who knows?
1. Re:POWER, Power, and PowerPC by file_reaper · 2013-08-07 07:04 · Score: 1
  
  Power.org is responsible for the Power Architecture ISA. (ie architecture)
  OpenPOWER will be the consortium which will license out the actual POWER Cores and associated peripherals.(ie microarchitecture and implementation)
2. Re:POWER, Power, and PowerPC by unixisc · 2013-08-07 19:00 · Score: 1
  
  Why not have just one org - power.org, w/ its pithy URL - handle all of that? Have divisions within that org handle the stuff, if you need
Amiga desktop? by Duncan+J+Murray · 2013-08-07 09:10 · Score: 1

Anyone want to surmise whether we'll get a desktop machine anytime soon?
Quite fancy a 5Ghz desktop beast running Amiga OS 4.
Just imagine - Full - motion - video. Less than 0 second shutdowns. Deluxe paint loading quicker than you can thumb a floppy in.
Or you could run ubuntu and have the dash load up in the time-frame your short-term memory works in.
D
1. Re:Amiga desktop? by Master+Moose · 2013-08-07 09:44 · Score: 1
  
  My 1st thoughts were to the Amiga.
  . . and while this news makes me hopeful, as an Amigain, I have learned not to get disappointed :)
  
  --
  . . .gone when the morning comes
Re:How is this any different from power.org by Anonymous Coward · 2013-08-07 17:38 · Score: 0

If Google wants to cobble together a small 2-4 core Power Chip with exactly the parts they need, based on licensed pieces from IBM, and then go fab it at wherever is cheapest, I've got to think that will save them money versus being a mountain of retail Intel chips.
Or more likely fab it at IBM's fab in East Fishkill, as a lot of the special sauce of POWER is IBM's well tuned manufacturing process.
MIPS is dead. by Anonymous Coward · 2013-08-08 00:14 · Score: 0

MIPS was interesting until they gave up on non-interlocked pipeline stages. I still don't understand why they did that because it was what this architecture was all about. When they did, they lost their single most important distinction from all other architectures and with it the reasons for why someone would want to use a MIPS.
Power PC by Anonymous Coward · 2013-08-08 04:22 · Score: 0

Does this mean that there may be more computers and appliances running AmigaOS? Or perhaps port HAIKU back to PPC?