> No one knows if [...] the PS4 will be more powerful

I think we actually do know for a week already.

> we all know both are PC and PC´s can be expanded to be faster. Just not the
> PS4 and the XBox.

...which would make them non-PCs according to your own logic ;-)

Quote:

Actually Nintendo managed to do better, they managed to get people to buy Wii and NOT play with it. Marketing campaigns for Wii were the best I have seen for a long time, useless console with crappy games.. but hey, play crappy games on a useless console together with friends while you zip a drink and its fun!

I personally know a dozen people owning Wii (got or bought as a xmas present) and have barely used it as it's really not that entertaining.

Reminder:

> Hyperionmp says:
> "The way in which it will be implemented however is already clearly defined
> and was subject to peer review by other developers. Obstacles to an
> efficient implementation were removed (e.g. the use of Forbid) and replaced
> in many OS components over the years (e.g. DOS). The foundation for
> SMP support was put in place, a clear picture exists what needs to be done
> to accomplish it and how. I'm willing to take a bet that it won't take 2 years ;)"
> http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=34171&forum=33&start=40#627520

Three quarters passed :-)

Update:

> "Additional production runs will be scheduled should demand continue to exceed the supply."
> http://blog.a-eon.biz/blog/?p=1252

"Despite high and volatile CPU prices, A-EON has commissioned another limited batch of systems to fulfill the demand on our pre-order list."
http://www.amigakit.com/x1000/

"As of March 2013, we are pleased to announce that a further batch will commence production in April, with delivery in Q3 2013 until end of year. We have a waiting list for the next batch. We now have sufficient customers to fulfill this current batch."
http://www.amigakit.com/x1000/?webpage=faq

"due to overwhelming customer demand, we have commissioned another manufacturing run of AmigaOne X1000 systems with our suppliers. We expect to receive new boards from Varisys within 8 to 10 weeks. Matthew Leaman of AmigaKit, our official distributor commented, "[...] We also have a large backlog of customers who have already registered their interest and they will be serviced first from the new manufacturing batch.""
http://www.a-eon.com/news/02-04-2013.html

I wonder if these will still be called "First Contact" systems. And it will be interesting to know whether Trevor will continue to pay the 475 USD extra costs of the 975 USD PA6T CPU.


Edit: Added quote from A-Eon press release

[ Edited by Andreas_Wolf 04.04.2013 - 13:42 ]

So they are just collecting preorders and then produce another batch. Just like what eyetech did.

Update:

>> Hyperionmp says:
>> "The way in which it will be implemented however is already clearly defined
>> and was subject to peer review by other developers. Obstacles to an
>> efficient implementation were removed (e.g. the use of Forbid) and replaced
>> in many OS components over the years (e.g. DOS). The foundation for
>> SMP support was put in place, a clear picture exists what needs to be done
>> to accomplish it and how. I'm willing to take a bet that it won't take 2 years ;)"
>> http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=34171&forum=33&start=40#627520

> Three quarters passed :-)

"There is a clear concept defined for multiprocessing since quite some time which is now being implemented. In fact, on an X1000, it is already possible to boot two independent instances of AmigaOS each running on a different core."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=33223&forum=14&start=320#702744

Another production run of X1000 computers is surprising, given that A-Eon should be well along with their development of its successor, which will not use the PA6T CPU.

Almost as surprising as the fact that there are still enough more customers that wish to purchase such an expensive computer, given the slow pace and lack of concrete information being released by Hyperion regarding progress and time line for when AmigaOS4.2 will be released.

I have mixed thoughts and feelings regarding the additional production run of X1000 computers, as on the one hand, if no more X1000's are produced, my own X1000 remains more unique and rare. On the other hand, with more X1000's produced and sold, it is more likely that it will receive more attention and support by both A-Eon & Hyperion, and possibly some few third party developers, who might write software that can only run on the unique hardware found within the X1000. On the whole, I am happy that more X1000's will be produced, as the increase in numbers gives me more assurance that my X1000 will be supported longer and better.

Who knows, maybe if enough X1000's are produced and one or two of them make there way into the right hands, there could even be the possibility of a port of MorphOS3.x to the X1000, but I doubt it.

Then again, I never thought that any port of MorphOS3.x would ever be created, or released for any of the SAM motherboards, and I was proved wrong.

[ Edited by amigadave 31.03.2013 - 20:17 ]

That is not SMP.

> Another production run of X1000 computers is surprising

In October after AmiWest, Trevor said that additional production runs would be scheduled should demand continue to exceed the supply (which is obviously the case), right after he had stated on AmiWest that there wouldn't be another batch.

> given that A-Eon should be well along with their development of its successor

I'm not sure this is true, considering Trevor's statement from just 5 weeks ago:

"I prepare to leave the beautiful New Zealand Summer for a trip to the frozen lands of Northern Europe [...]. During my visit I am meeting with Varisys, the designers of the A1-X1000, to [...] review plans for future hardware development."
http://blog.a-eon.biz/blog/?p=2830

> if no more X1000's are produced, my own X1000 remains more unique and rare.

This should only be a point for people purchasing the X1000 as a store of value, and so they should better not use it but keep it safe in a bank vault under absence of air for worse times to come ;-) For an actual *user* this shouldn't be a concern, I think.

> with more X1000's produced and sold, it is more likely that it will receive more
> attention and support by both A-Eon & Hyperion, and possibly some few third
> party developers [...]. [...] the increase in numbers gives me more assurance
> that my X1000 will be supported longer and better.

Indeed. It's a pity that drivers for onboard audio and onboard Ethernet are still not completed. The less X1000 machines in existence, the lower the motivation to work on supporting those components I'd guess.

> who might write software that can only run on the unique hardware found within the X1000.

You mean the successor of the X1000 won't feature Xena/XCore and Xorro? To be honest, I don't think that would be a notable loss.

@Andreas,

I have not kept up on exactly what has been officially stated (or even unofficially stated) regarding the progress of A-Eon's next motherboard design, so I don't want to reveal anything that I have been told, or otherwise been made aware of, as that is not my place to reveal anything. But I think it is safe to say that A-Eon and Varisys have not been idle for the last 6+ months since I last talked to Trevor, which is why I am a little surprised, but happy for Trevor and the guys at AmigaKit.

Trevor is smart to not divulge information prematurely (this time) as that usually reflects badly on anyone guilty of such a mistake in this community and earns lots of criticism from several vocal members. Trevor learned that lesson well, during the delays that pushed the release of the X1000 further and further out from his hoped for initial release date(s).

I certainly do not expect Trevor to further contribute toward paying the increased cost of the PA6T CPU's, and would expect that any new production runs would only be possible if he has been able to obtain those parts at the same, or similar price as the first batch, or that he would ask a higher price for the next batch of X1000's. Of course I have no way of knowing what price he and AmigaKit have paid to obtain their parts.

I just hope that all is going well for Trevor and AmigaKit with their next endeavor(s), as they are great guys and I wish for them to be successful and prosperous. I also hope that some day A-Eon can come out with hardware that meets the requirements of the MorphOS Dev. Team (both in price and performance), as I would love to be able to support both teams (I mean the A-Eon/AmigaKit team and the MorphOS Dev. Team) on a common platform, as MorphOS3.x is still my favorite operating system of choice. That said, I don't regret purchasing my X1000 to run AmigaOS4.x, as I still retain my interest in all parts of the Amiga community.

>>> given that A-Eon should be well along with their development of its successor

>> I'm not sure this is true, considering Trevor's statement from just 5 weeks ago: [...]

> I don't want to reveal anything that I have been told, or otherwise been made
> aware of [...]. But I think it is safe to say that A-Eon and Varisys have not been
> idle for the last 6+ months since I last talked to Trevor [...]. Trevor is smart to not
> divulge information prematurely (this time) as that usually reflects badly on
> anyone guilty of such a mistake in this community and earns lots of criticism
> from several vocal members. Trevor learned that lesson well

Nice to know that the development of the X1000's successor may be way more advanced than Trevor's statement from 5 weeks ago makes it seem :-)

Btw, recent statement:

"we drove to High Wycombe to meet with Paul Gentle and Adam Barnes the Managing and Technical Directors of Varisys Ltd [...]. We reviewed the status of current developments [...]"
http://blog.a-eon.biz/blog/?p=2998

Addendum:

> Look who's sponsoring http://www.powerlinuxusersgroup.com :
> http://www.meetup.com/Servergy-Power-Linux-Users-Group/sponsors/ :-)

Some recent words from Trevor:

"Servergy, Inc, the people behind the PowerPC server technology have now formed the PowerLinux Users Group (PLUG), an international users group to control and shape the future of PowerLinux software development for the OpenSource community. [...] Power Linux may not be of interest to many Amigans, but that fact that Servergy are promoting open source PowerPC to global Power Linux developers can only be good for the future of PowerPC hardware."
http://blog.a-eon.biz/blog/?p=2998

Some more words (Google translation):

"Our interest for PowerPC Linux is twofold. L 'A1-X1000 already supports 9 PowerPC Linux distributions, as well as AmigaOS 4, which is the operating system for which it was originally designed. And' in our plans for all future products PowerPC-based hardware that will be developed for the Amiga OS 4 will also continue to support the PowerPC Linux distributions. There is also a large community of developers who are interested in Linux PowerPC PowerPC hardware platforms. We see this as a potential market for the development of hardware and selling."
http://translate.google.com/translate?sl=it&tl=en&u=http://deliriotecnologico.blogspot.com/2013/05/powerpc-back-to-future.html


Edit: PowerLinux Users Group (PLUG) seems dead now.

[ Edited by Andreas_Wolf 27.08.2014 - 10:58 ]

Andreas_Wolf wrote:,
Quote:

Only Trevor and his partners know exactly how much progress has been made since last Oct., or if there have been any delays, or interruptions in the work being done on his next motherboard design. I can only hope that he and his partners have not run into any serious obstacles that would delay that work by many months, but as we observed with the design and development of the X1000's Nemo motherboard, with the limited resources available within our community, development times are extended much longer than those within say the Linux community, or the mainstream PC & Mac communities.

From Trevor's most recent blog statements, perhaps he will have some new information to release in July when he and AmigaKit attend the computer show in the UK that he mentioned in his blog. I already have all the hardware I need to run each of the operating systems I am interested in, but I am still rooting for his success, for the good of the community.

Update:

> it will be interesting to know whether Trevor will continue to
> pay the 475 USD extra costs of the 975 USD PA6T CPU.

More details coming from AmigaKit:

"The pricing of the PA-SEMI CPU is volatile unfortunately- the last production run we absorbed half the price increase and Trevor kindly absorbed the other half to ensure that the batch was produced for waiting customers. This time we will once again work hard to keep the price static for customers."
http://www.amigans.net/modules/xforum/viewtopic.php?post_id=78258#forumpost78258

Update:

>> I wonder what, if any, involvement IBM has had with the development of the
>> Xbox 720 or PS4 cpus?

> Regarding the PS4 CPU, none I guess.
> http://en.wikipedia.org/wiki/PlayStation_4#Console
> http://en.wikipedia.org/wiki/AMD_Accelerated_Processing_Unit#Jaguar_architecture_.28Kabini_and_Temash.29

Same with the XBox One CPU:

http://en.wikipedia.org/wiki/Xbox_One#Hardware

So Cell based PPCs are a dead issue.

> So Cell based PPCs are a dead issue.

Yes, apparently. This became clear 3 months ago at the latest. Even if Microsoft had opted for PPC, it surely wouldn't have been of the Cell type.

Reminder:

> Hyperionmp says:
> "The way in which it will be implemented however is already clearly defined
> and was subject to peer review by other developers. Obstacles to an
> efficient implementation were removed (e.g. the use of Forbid) and replaced
> in many OS components over the years (e.g. DOS). The foundation for
> SMP support was put in place, a clear picture exists what needs to be done
> to accomplish it and how. I'm willing to take a bet that it won't take 2 years ;)"
> http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=34171&forum=33&start=40#627520

Only 3 months to go :-)

> I think it is safe to say that A-Eon and Varisys have not been idle

Apropos Varisys:

"Ultra announces that it has acquired Varisys Limited ("Varisys") for an initial cash consideration of £16m. Additional payments of up to £2m will be payable subject to performance over the next two years. [...] Varisys is a privately-held British company based in High Wycombe, UK, and was acquired from its founders, Paul Gentle and Adam Barnes; both will remain with the business under Ultra ownership. The acquisition of Varisys will be financed using Ultra's existing facilities and is expected to be earnings enhancing in 2013. Varisys will continue to operate from its current location as part of Ultra's existing Controls business within the Aircraft & Vehicles Systems Division."
http://www.ultra-electronics.com/press_releases/299_Ultra_acquires_Varisys_-June_2013.pdf

"Varisys Ltd has been acquired by Ultra Electronics and is now a part of the growing Group. Varisys will join the Ultra-Controls business unit whilst continuing to operate out of its office in High Wycombe. The move allows Varisys to utilise Ultra's specialist capabilities in order to build on its successes in the Military, Aerospace and Commercial markets. Varisys' increased capabilities will allow it to deliver on more complex and technically demanding projects whilst adding more value to customers throughout the development process itself. Varisys will continue to design and manufacture products for high performance embedded computing applications and offer commercial-off-the-shelf products through our partnerships with leading manufacturers including GE Intelligent platforms, North Atlantic and SECO."
http://www.varisys.co.uk/news/varisys-ltd-acquired-by-ultra-electronics

Furthermore:

"Before working at Varisys, Adam Barnes (my business partner) and I worked for a company called Transtech Parallel Systems. [...] Before that I worked for Ultra Electronics."
http://obligement.free.fr/articles_traduction/itwgentle_en.php


Edit: A-Eon press release:

"Following the news that Varisys, our hardware partner and designers of the Nemo motherboard, had been acquired by Ultra Electronics, we have been in close communication with regards current and ongoing development projects. Adam Barnes, Technical Director commented: "We are looking forward to continuing our relationship with A-EON and the Amiga community. We always hoped to develop IP for embedded applications through the A-EON projects, and that has not changed one bit." Paul Gentle, General Manager added: "Adam and I are staying with the business, and are excited by the future prospects and opportunities that will arise." A-EON Technology Director Matthew Leaman commented, "We wish to congratulate Varisys on their acquisition by Ultra Electronics and look forward to strengthening our close working relationship with them on future exciting hardware developments"."
http://www.a-eon.com/news/16-06-2013.html

[ Edited by Andreas_Wolf 16.06.2013 - 11:40 ]

Update:

>> it will be interesting to know whether Trevor will continue to
>> pay the 475 USD extra costs of the 975 USD PA6T CPU.

> More details coming from AmigaKit:
> "The pricing of the PA-SEMI CPU is volatile unfortunately- the last production
> run we absorbed half the price increase and Trevor kindly absorbed the other
> half to ensure that the batch was produced for waiting customers. This time we
> will once again work hard to keep the price static for customers."
> http://www.amigans.net/modules/xforum/viewtopic.php?post_id=78258#forumpost78258

"Apparently Trevor was able to get a per unit price of $600 on the PA6T for the latest run."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=37872&forum=16&start=20#711574

"Trevor told us most recent order of PA6T was secured for $600 per chip compared to the $1000 that was mentioned at Amiwest last year. Remember that he was swallowing the extra cost from his pocket so don't expect the X1000 to get any cheaper."
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=37872&forum=16&start=40#711600

"There is no shortage of PA Semi CPU ..if you are willing to pay the price. The original project was based on a $500 CPU. We have paid anything from $600 to $1000 each since Apple closed the manufacture. We (AmigaKIt and myself) jointly subsidised the CPU price increase and did not pass these on to the customer."
http://www.intuitionbase.com/trevor_irclog.txt

"Apple purchased P.A. Semi and closed down production and the prices of the CPU almost doubled. The initial costings were based on a CPU price of $500 each. Since Apple ceased production the CPU prices have varied between ~$600 and $1000 per unit. However, we have absorbed these price increases and have not passed them on to the customer."
http://obligement.free.fr/articles_traduction/itwdickinson2_en.php

"We had negotiated a standard supply agreement with P.A. Semi at a unit price of $500 each. However, after Apple took control they cancelled our credit terms and eventually shut down manufacturing of the CPU. Since then the maximum price we have paid to date is $1000 each with shipping and insurance. However, I along with AmigaKit jointly subsidised the increased CPU cost and did not pass it on to our customers."
http://www.retroplanet.gr/content/interview-trevor-dickinson-%CF%83%CF%85%CE%BD%CE%AD%CE%BD%CF%84%CE%B5%CF%85%CE%BE%CE%B7-%CE%BC%CE%B5-%CF%84%CE%BF%CE%BD-trevor-dickinson


Edit: Added more quotes.

[ Edited by Andreas_Wolf 29.01.2017 - 00:35 ]

Reminder:

> Hyperionmp says:
> "The way in which it will be implemented however is already clearly defined
> and was subject to peer review by other developers. Obstacles to an
> efficient implementation were removed (e.g. the use of Forbid) and replaced
> in many OS components over the years (e.g. DOS). The foundation for
> SMP support was put in place, a clear picture exists what needs to be done
> to accomplish it and how. I'm willing to take a bet that it won't take 2 years ;)"
> http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=34171&forum=33&start=40#627520

Only 6 weeks to go :-)

Meanwhile, ssolie has to say the following on the topic:

"I never liked calling the feature "SMP" and I prefer to call it simply multi-core support. My reason is that you always get some egghead piping up and tossing definitions around. In my opinion, users don't care if it is SMP or ASMP or WXYZ or whatever else as long as all the hardware they paid for is being utilized to run their software. The current status is that the multi-core feature is in development. Until it is finished we will be running in single core mode on whatever hardware is available. [...] We have been moving to support multiple cores for some time now. For example, Forbid/Permit are annoying with multi-core so there has been an effort to remove/reduce their usage throughout the OS. Another example is the new memory subsystem which has been designed for multi-core as well. We haven't made such drastic changes for the fun of it. "
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=37939&forum=32#713132
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=37939&forum=32#713133

"Whether it is a textbook implementation of SMP or not is still not clear to me yet. But that's just my opinion. I know Hyperion still calls it SMP and that's also fine."
http://www.amiga.org/forums/showpost.php?p=741831

"if we do tie 68K apps to a single core does that mean it is still SMP? [...] I could just call it "SMP" even though it may (but not necessarily) be delivered as a hybrid [...]. The goal is and still remains to be true SMP"
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=37939&forum=32&start=20#713304

Nice summary of ssolie's alleged dislike for the term "SMP":

http://moobunny.dreamhosters.com/cgi/mbmessage.pl/amiga/239743.shtml


Edit: Added another quote.

[ Edited by Andreas_Wolf 25.07.2013 - 02:17 ]

Quote:

I seriously doubt that AmigaOS4.2, or any version of AmigaOS4.x with multi-core support will be ready for release within 6 weeks time frame, but I am also not bothered by rough guesses being inaccurate when it comes to any release dates for Amiga/Amiga-Like operating systems, or software. Comes with the territory.

Update:

> still no more technical details on POWER8 available to the general public.

"The Power8 chip [...] will have on-chip PCI-Express 3.0 controllers -- something IBM has been sorely lacking this year with the Power7+ chips -- which will put it on par with Intel's Xeon chips. Interestingly, the Power8 chip will get rid of the GX++ InfiniBand links used to plug in external peripheral drawers for Power Systems machines and replace it them something called the Coherently Allocated Processor Interface, or CAPI for short. This is an overlay that will ride atop the PCI-Express 3.0 mechanicals to provide coherent memory addressing for CPUs and external coprocessors like Nvidia GPUs."
http://www.theregister.co.uk/2013/08/06/ibm_opens_up_power_chips_armstyle_to_take_on_chipzilla/

Furthermore, IBM and partners have founded the OpenPOWER Consortium to allow licensing of IBM POWER chip technology (and related technologies):

http://en.wikipedia.org/wiki/OpenPOWER_Consortium

"Google, IBM, Mellanox, NVIDIA and Tyan today announced plans to form the OpenPOWER Consortium -- an open development alliance based on IBM's POWER microprocessor architecture. The Consortium intends to build advanced server, networking, storage and GPU-acceleration technology [...]. The move makes POWER hardware and software available to open development for the first time as well as making POWER IP licensable to others, greatly expanding the ecosystem of innovators on the platform. [...] As part of their initial collaboration within the consortium, NVIDIA and IBM will work together to integrate the CUDA GPU and POWER ecosystems. "The OpenPOWER Consortium brings together an ecosystem of hardware, system software, and enterprise applications that will provide powerful computing systems based on NVIDIA GPUs and POWER CPUs," said Sumit Gupta, general manager of the Tesla Accelerated Computing Business at NVIDIA."
http://www-03.ibm.com/press/us/en/pressrelease/41684.wss

"Under the OpenPOWER initiative, IBM will license the core intellectual property for our POWER technologies to other companies [...]. Up until now, IBM primarily used the POWER design in its own servers. This new initiative makes it possible for cloud services and their technology providers to redesign the chips and circuit boards where computing is done -- optimizing the interactions of microprocessors, memory, networking, data storage and other components. As a result, they can get servers that are custom-tuned for their applications. [...] we are licensing the microprocessor technology to other companies openly -- meaning they get to look at the blueprints for the processor and the software that goes with it so they can take full advantage of its capabilities. The cloud service providers will be able to hire IBM or other companies to manufacture the processors and other related chips. [...] this announcement shows our commitment to aggressive investment in our POWER processors and servers. Other innovations will come from our partners."
http://asmarterplanet.com/blog/2013/08/open-and-collaborative-development-is-the-future-of-cloud-computing.html

"The plan [...] 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 [...]. "The way the industry is innovating is shifting," says McCredie. "[...] And so we are taking our Power IP, opening it up as well as decomposing it -- the processor, the firmware, all of the key pieces -- to enable people to innovate around Power platforms. [...]" There is a confluence of different factors that have compelled IBM to open up the Power architecture. McCredie said that in the past, system makers of all kinds [...] were content to get the performance or total cost of ownership goals they have for machines by innovating at the motherboard or system level. But now, this is shifting and companies want to innovate at the chip level, particularly as processors, networking, and storage functions are being brought down onto system-on-chip, or SoC, packages [...] with chip etching technologies now sufficiently small that these different components can be put into a single die or within a single package. [...] the basic idea is that IBM will control the Power instruction set, much as ARM Holdings does with the ARM instruction set. In days gone by, people might have wanted to innovate at the instruction set level, but McCredie says this is less important today, and being able to add functions to a Power core (or a collection of them with cache memories and such) is what people want to be able to do. [...] while IBM would love for potential future Power chip designers to use its wafer baker in East Fishkill, New York, that will not be a requirement. Without naming names, McCredie says that IBM will work with consortium members to find alternate fabs for Power chips. [...] Nvidia, which is an ARM licensee, is not about to become a Power licensee, says Sumit Gupta, general manager of the Tesla Accelerated Computing business unit. But Nvidia is very excited about the prospects of marrying Power processors and Nvidia GPUs for both HPC and general purpose systems. [...] "And I personally see this about IBM backing CUDA [...]. They bring a high performance CPU to the party, and we have a high-performance GPU. [...]" [...] The OpenPower Consortium is not restricting the licensing of any particular Power chip technology, but McCredie says that it is really focused on next year's Power8 chip and onwards."
http://www.theregister.co.uk/2013/08/06/ibm_opens_up_power_chips_armstyle_to_take_on_chipzilla/

"IBM [...] will soon begin offering licenses for its venerable Power architecture to other companies, allowing them to build their own Power chips for "servers, networking, and storage devices." [...] "The OpenPOWER Consortium brings together an ecosystem of hardware, system software, and enterprise applications that will provide powerful computing systems based on Nvidia GPUs and Power CPUs," said Sumit Gupta, general manager of the Tesla Accelerated Computing Business at Nvidia, in a statement. IBM and Nvidia will be working together to make the Power architecture and Nvidia's CUDA platform play nicely with one another. The first architecture to be available to licensees will be the upcoming Power8 architecture [...]."
http://arstechnica.com/information-technology/2013/08/ibm-to-begin-arm-style-licensing-of-power-cpu-architecture/

"IBM's Openpower Consortium is also an attempt to mirror British semiconductor firm ARM's business model, licensing its technology to system builders and chip suppliers but not manufacturering the chips itself, though the news doesn't mean IBM has plans to stop making its own Power chips. Like ARM it seems that IBM's new alliance is an attempt to break into a growing server niche of energy efficient chips that go into a new breed of servers that save space and minimise power consumption."
http://www.theinquirer.net/inquirer/news/2287376/ibm-google-and-nvidia-form-openpower-consortium-to-rival-intel-and-amd

"IBM will license designs of the Power microprocessor architecture to other companies including Google, in an effort to expand use of the architecture and reverse declines in its systems hardware business. Intellectual property of the chip design is being opened up as part of a development alliance called OpenPower Consortium that IBM announced with Google on Tuesday. Consortium members will be able to make Power chips based on architecture designs [...]. [...] The first products from alliance partners could be based on IBM's upcoming Power8 design, said Brad McCredie, vice president and chief technology officer at IBM's Systems and Technology Group (STG), in an interview. [...] In an email statement, Google said the OpenPower Consortium "has the potential to establish Power architecture as a viable option for applications running within Google's datacenters." Google designs its own servers, and the search company could design its own integrated chip -- also called system-on-chip -- based on the Power architecture, said Dean McCarron, principal analyst at Mercury Research. [...] Tyan will be the first company to release a server based on the Power architecture, and said in a statement that the system was projected as an alternative to x86 servers. IBM has also included a feature in Power8 for component makers to easily attach their intellectual property to the chip. More companies will join the alliance in the coming months, McCredie said. Third parties could release new system-on-chips based on Power in a few years, McCredie said, adding that the design cycle lasts two years or more. The Power IP is also being opened up to manufacturers, and IBM will continue to make chips for third parties. [...] IBM's Power architecture brings more reliability, processing power and longevity to servers in cloud deployments, McCredie said, adding that the company is targeting the growing Asia market through the new OpenPower alliance. The Power architecture and chips will continue to be developed for data centers, and likely not go in the direction of smartphone and tablet chips such as Intel's Atom x86 chips or ARM, McCredie said. IBM's Power chip design for servers is not the same as Power designs being used in microcontrollers from companies like Freescale Semiconductor. McCredie said that opening up the Power architecture is also complementary to the Facebook-backed Open Compute Project, which focuses mostly on server designs. IBM has been trying to push the Power architecture into other applications for many years, but rather than growing in opportunity, the market for the chip has been shrinking, said Jim McGregor, principal analyst at Tirias Research. [...] One of the more interesting aspects of OpenPower is IBM and Google partnering with Nvidia, which is tuning its CUDA parallel programming toolkit for Power processors."
http://www.networkworld.com/news/2013/080613-ibm-opens-up-power-design-272589.html

Update:

> still no more technical details on POWER8 available to the general public.
> And it seems its release has been re-scheduled to "early 2014" [...]
> the first (to my knowledge) mention of POWER9 by IBM, scheduled for the 2016
> to 2020 time frame:
> http://projects.csail.mit.edu/caos/2012-01-25-caos-bgq-v1-ed.pdf (January 2012, page 6)

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=9463&forum=3&start=5

Reminder:

> Hyperionmp says:
> "The way in which it will be implemented however is already clearly defined
> and was subject to peer review by other developers. Obstacles to an
> efficient implementation were removed (e.g. the use of Forbid) and replaced
> in many OS components over the years (e.g. DOS). The foundation for
> SMP support was put in place, a clear picture exists what needs to be done
> to accomplish it and how. I'm willing to take a bet that it won't take 2 years ;)"
> http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=34171&forum=33&start=40#627520

It's now 6 days past the "deadline" :-)