Quote:

That's all speculation of course. At first they started with a Cell 3 design (which by IBM's word it should have 32 cores) and now we're down to 10, if at all.

Quote:

How good is a CPU if you can't buy it in a system?

Quote:

Yes, I realized that after I wrote the post. It was obviously a mistake :)

Quote:

Well, think about it. If your #1 customer (Sony) decides to drop your product and your partner (Toshiba) is in hard financial troubles, what would you (IBM) do with Cell? I never meant R&D on technologies used in Cell, I meant the Cell CPU itself. Would you spend hundreds of million dollars to continue research on a costly product? I don't think so.

Quote:

Now you're asking me to do it for you. Hilarious.

Quote:

Indeed someone is talking nonsense, but it's not me apparently. I almost never ask for sources when I have a casual discussion. If I was to prove something I would provide sources without your asking. In this case, I mentioned something that I can't prove. I merely concluded as much, from what I read and from discussions I had with people that know more, and I believed them without asking for "sources" like you do here. And yes, when people that know more say something I tend to believe it -at least if it's not outright ridiculous. Whethere I do my own research afterwards, that's another topic.

In fact, we could do a much better conversation if instead of "source?" you said "hey, why do you say that?" or "what makes you say that?" or whatever.

Quote:

I'm sorry, I will not do that work for you. If you don't believe me, be my guest, but it's your decision to go and search google. Btw, posting a huge number of URLs -which seems your habbit- might make you believable, but it still doesn't prove -in the end- what you claim.

Quote:

"Mild"? :)

Quote:

Well, you are right here. After all, "stopped" is too harsh, and unless I was IBM myself -which I'm not- how could I back this up? But I "can" read the signs here, so do many others. Cell is a money-hole, however good it is.

Quote:

a scaled down Cell with two more cores, is hardly the expected result of 4 years R&D (when PS4 is released, 2012 est.). By that time, who knows what the competition will offer. Perhaps it's not a sign of a stop in R&D, but it definitely is not a sign of continuous R&D.

Quote:

eh? What good is a plain CPU? Sure, I probably could order 1k 8610 CPUs myself right now, what good would these do without a system to put on? Btw "You and I" never discussed about anything, my comments were to VelcroSP, and all were focused on 5121/LimeBook vs ARM/Atom. Sorry, I wasn't talking about CPUs, I was talking about CPUs INSIDE Systems.

Quote:

Again, no, read above. the discussion started from 5121, which was quite a specific model/market. I just merely pointed that while other platforms have a plethora of choices and furious development happening, PowerPC has stagnated to just 2 CPU lines (Power7 and QorIQ).

Quote:

Sorry, I'll believe Titan when it's actually released.

Quote:

Indeed. In fact, I know that there is some kind of a clearance going on in FixStars, some Cell blades are being sold quite cheap (for Cell blades that is).

> these lose greatly to pretty much every modern Intel/AMD cpu

Quote:

I *meant* that these systems, lose greatly when compared to pretty much every modern Intel/AMD-based *system*. Better?

> That's all speculation of course.

That's the prospect of the sources you refered to, isn't it? You were (mis)taking them for fact that "Cell R&D" has stopped", and when I tell you what these sources *really* were prospecting, it's suddenly "all speculation". Smells like double standard...

> At first they started with a Cell 3 design

Source?

> which by IBM's word it should have 32 cores

Yes, 2 PPEs plus 32 SPEs for the PowerXCell 32ii and 4 PPEs plus 32 SPEs for the PowerXCell 32iv.

> and now we're down to 10, if at all.

8 plus 4 sums up to 12 in my book.

>> PS4 not using a better Cell than PS3 (or even no Cell at all)
>> wouldn't have to mean anything regarding Cell R&D itself. There
>> might be other purposes for a better Cell than a Sony's gaming
>> device. The existing PowerXCell 8i is proof of that.

> How good is a CPU if you can't buy it in a system?

No future Cell in a PS4 equals to no future Cell in any system? Your logic is twisted. Look at the mentioned PowerXCell 8i. It's not in a Sony's gaming device and probably never will be, but it's existing nonetheless and there are PowerXCell 8i systems out there which you can buy, e.g. from the vendor of the PowerStation you refered to.

> If your #1 customer (Sony) decides to drop your product and your
> partner (Toshiba) is in hard financial troubles, what would you
> (IBM) do with Cell?

As we very well know, Sony is *not* going to drop the Cell for PS4. So tell me, what do you think IBM is going to do with Cell then? And what about Sony's ZEGO? Is it cancelled?
Furthermore, the PowerXCell 8i got R&D'd (and produced) without any connection to any prospected Sony product, and so will be the PowerXCell 32ii and the PowerXCell 32iv according to IBM's roadmap.

>> "it's trivial to find the articles on google" for you to provide
>> me with the sources I asked you for.

> Now you're asking me to do it for you. Hilarious.

I really tried hard but couldn't come up with any search results backing up your original claim that "Cell R&D has stopped". So what else could I do than asking you to provide me with the source(s)?

>> Pure nonsense. You claim something, I ask you to back up that
>> claim. That's how discussion works and has alway been.

> Indeed someone is talking nonsense, but it's not me apparently.

You are, trust me.

> I almost never ask for sources when I have a casual discussion.

So you naively believe everything that is thrown at you? Even if it initially creates a cognitive dissonance for you (which your statement did for me)? Then you're definitely a bad exemption, fortunately.
(That's really nothing to do with casual or not, rather than the inner workings of a human's mind.)

> If I was to prove something I would provide sources without your asking.

Your twisted logic again. How could you know you are to prove something *before* someone contests or challenges your claim by asking? If you decide to *not* provide sources right along your claim, then nobody is allowed to ask you for them for not having to rely on your words alone?

> In this case, I mentioned something that I can't prove.

Then don't put it as fact. Simple as that. How could one conclude that "Cell R&D has stopped" is only your *personal* perception rather than a fact you read somewhere or concluded (as in "not just guessed") from other facts?

> I merely concluded as much, from what I read and from discussions I
> had with people that know more

And these knowing people told you the PowerXCell 32ii and PowerXCell 32iv were cancelled? Or how did you "conclude" this?

> I believed them without asking for "sources" like you do here.

There's a difference: These people might be affiliated with IBM or Toshiba (if not, you could as well believe anybody, but even then it would be just you), and I know you are not. That's why I called your claim a "factual statement about a third party".

> when people that know more say something I tend to believe it -at
> least if it's not outright ridiculous.

From my own perspective you don't count as someone who knows more, that's why I asked you for source(s), e.g. from people who (from my perspective) might know more.
Or should I really have believed you for no reason? Then I would have taken your non-factual claim for fact, as we now know. "Cell R&D has stopped" very much sounds "outright ridiculous" to me btw.

> we could do a much better conversation if instead of "source?" you
> said "hey, why do you say that?" or "what makes you say that?" or
> whatever.

So it's really about semantics? You're being a sissy now.

>> I didn't expext you to provide sources right with your statement
>> [...] But you should be able to provide them after being asked for
>> (just what I did) if you want to be taken seriously (by me, at least).

> I'm sorry, I will not do that work for you.

Yes, and we even know the reason by now: There simply are no sources backing up you claim that "Cell R&D has stopped".

> If you don't believe me, be my guest,

I didn't. And it has been the right decision, as proven by the outcome of our discussion.

> but it's your decision to go and search google.

As I already told you, I did.

> posting a huge number of URLs -which seems your habbit- might make
> you believable, but it still doesn't prove -in the end- what you
> claim.

At least it would prove that I didn't just pull my claim out of thin air. And if I claimed "according to [or concluding from] X, Cell R&D has stopped", then an URL to (the statement from) X would indeed prove - in the end - what I claim (presumed my conclusion is not just wild guessing, that is).

>> That's still no proof that "Cell R&D has stopped", rather a mild
>> indication.

> "Mild"? :)

Yes, very mild. Do you really think that Sequoia not having Cell processors to be a strong indication that PowerXCell 32ii and PowerXCell 32iv have been cancelled?

> unless I was IBM myself -which I'm not- how could I back this up?

By referencing appropriate statements from IBM or Toshiba (affiliates), for instance. To do so, such statements have to exist in the first place, of course. If they don't, you cannot state your claim as fact.

> But I "can" read the signs here

..signs you (mis)read as "Cell R&D has stopped" originally. Fine that you put your glasses on after I questioned your way of reading :-P

> a scaled down Cell with two more cores

It's supposed to be 4 more, not 2.

> is hardly the expected result of 4 years R&D (when PS4 is released,
> 2012 est.).

But a 3.8 GHz PowerXCell 32iv with 4 PPEs and 32 SPEs would be, wouldn't it? That wouldn't have anything to do with PS4, as I already outlined, but that's surely no requirement for a CPU to be a Cell processor.
And still, in regard to PS4's Cell, even very slow R&D is R&D, whereas a stopped R&D would mean no further R&D at all.

>> You and I were talking about Power Architecture CPUs, not systems.

> eh? What good is a plain CPU?

Ask that to yourself. After all, you were the one claiming the (Power Architecture) future will bring "only QorIQ and Power7", which are plain CPUs, not systems. So it was obviously your intention to talk about CPUs.

> I probably could order 1k 8610 CPUs myself right now, what good
> would these do without a system to put on?

Don't ask me. I merely answered your statements about some of the CPUs you mentioned.

> Btw "You and I" never discussed about anything

You talked about CPUs and I answered you, talking about CPUs as well. That very well classifies as "you and I were talking about" CPUs in my book.

> my comments were to VelcroSP

This is a public board. Private discussions do not exist in the public board area by definition. For such purpose there's a PM function implemented. No matter to whom you answer in a public board, everybody is authorized to answer your posting.

> all were focused on 5121/LimeBook vs ARM/Atom.

Your memory seems broken:

"AltiVec is pretty much dead, the e600 is also dead, there is no e700 coming, [...] Cell R&D has stopped (and PS3 will probably be the last PPC-powered Sony console, dunno about Xbox yet), only QorIQ and Power7"

...is definitely not about "5121/LimeBook vs ARM/Atom", but rather about e600, e700, Cell, QorIQ and POWER7.

> I wasn't talking about CPUs

Now you're being funny (again?).

> I was talking about CPUs INSIDE Systems.

As was I. After all, all (future) Power Architecture CPUs I mentioned in my posting I pointed you to are supposed to be built into systems, of course.

>> You and I were talking about Power Architecture CPUs in general,
>> not just desktop suiting ones.

> Again, no

Let's see: You mentioned QorIQ and POWER7. QorIQ is (according to Freescale) supposed to be a communications processor, POWER7 is supposed to be a server processor.
Do you really stand by the claim that you were not talking about Power Architecture CPUs in general, but only desktop suiting ones?

> the discussion started from 5121, which was quite a specific
> model/market.

I don't care wherefrom the discussion may have started when my intention is solely to hint you to a correction of a false factual statement of yours.

> I just merely pointed that [...] PowerPC has stagnated to just
> 2 CPU lines (Power7 and QorIQ).

It's true that you claimed this. But that doesn't render your claim true. I already pointed you to my posting where you can read up that your claim is false.

>> Contrary to your claim, QorIQ and POWER7 are *not* the only Power
>> Architecture CPUs supposed to come (see my link to my previous
>> statement).

> I'll believe Titan when it's actually released.

That would make exactly one less. What about the other ones I mentioned? Will you continue to deny that these CPUs and CPU lines are to be further developed?

> I *meant* that these systems, lose greatly when compared to pretty
> much every modern Intel/AMD-based *system*. Better?

Quite much :-)

I'll admit, I'm too bored to reply each and everyone of your statements, I'll just say a few things, take it as you like; I had private discussions with Freescale guys, where they *explicitly* told me that e600 + altivec was considered dead inside the company, LONG before it became publicly known elsewhere. Have you seen such an announcement from Freescale? No and you won't because it's bad publicity for such a company because they still sell this stuff.

I also had private discussions with financial folks (they see things in a different light than us geeks, it doesn't matter if a technology is cool, if it doesn't make money it will die), where they said that with the current situation with Sony/Toshiba, Cell's future is unclear. I'm definitely not going to provide "references of appropriate statements" from any such private discussion. Oh and, sure, go ahead and nitpick my choice of words (most of your replies are around this anyway), forest vs tree anyone?

Cell's absence in Sequoia also adds more to the rumours. As for the PowerXCell 32, dunno if you've ever seen the Cell roadmap (used to be found here, but not anymore, surprise?), it should be around available at 2010 (the roadmap was from 2006), but the PS4 is due out in 2012? 2 years delay, when PS3/original Cell release was almost simultaneous? Strange? Anyway, actually I'll be happy if I see the PowerXCell 32, I like my PS3 and the Cell and it would be fun to play with something more powerful, but I somehow think it will take quite longer than expected, if at all.

Also, in regard to the other CPUs you mentioned, I actually don't see anything other than embedded CPUs. They're great really for their purpose - I love my efika and the beagleboard is really fun to play with-, but they're not much use for a desktop or even a netbook. If you read my first mail again, I stated (amongst other) that e600 (the de-facto desktop-oriented CPU line, at least so far) is dead, and we're left with just server and embedded lines (QorIQ is just the most powerful, I don't really care for the rest, which are many, because they're embedded-only and too low-power).

So, what *new* CPU (-based system, ok) could an average powerpc fan choose? None. Which was *exactly* my original point, which you still seem to miss. It doesn't matter the model, fact is that PowerPC has not much to offer for this market (well there are plenty of existing CPUs for the job, but no commitment from the companies). Unless some bold investor decides to pick up, say the e600 core and revive the 86xx line, I don't see any product that would satisfy the average powerpc fan (and I mean desktop user/developer basically, not server/embedded they have no problem).

[ Edited by feanor on 2009/4/10 21:41 ]

> I had private discussions with Freescale guys, where they *explicitly*
> told me that e600 + altivec was considered dead inside the company,
> LONG before it became publicly known elsewhere. Have you seen such
> an announcement from Freescale? No and you won't because it's bad
> publicity for such a company because they still sell this stuff.

Fine, but where exactly is the analogy to you claims about Cell R&D? Did you have private discussions with IBM or Toshiba guys, where they *explicitly* told you that Cell was considered dead inside the companies? If so, then be frank about that. I won't dispute that. Giving private discussions with people who should know as source of your claims is better than giving no source at all, like you did with your Cell claim.

> I also had private discussions with financial folks [...], where
> they said that with the current situation with Sony/Toshiba, Cell's
> future is unclear.

I guess these financial folks are from outside IBM and Toshiba, so it's just opinion by their side. And "unclear future" isn't exactly the same as "stopped R&D". So it's obvious where your claim did *not* come from, at least.

> I'm definitely not going to provide "references of appropriate
> statements" from any such private discussion.

You referenced private discussion. That's sufficient. I didn't ask for more. You made very well clear that they never told you that "Cell R&D has stopped".

> go ahead and nitpick my choice of words (most of your replies are
> around this anyway)

...says the one who argued that "we could do a much better conversation if instead of "source?" [I] said "hey, why do you say that?" or "what makes you say that?" or whatever". Very funny.
My replies are not about your choice of words, but about your factual claims being false (or unproven at best).

> As for the PowerXCell 32, dunno if you've ever seen the Cell roadmap

Of yourse I have. Or where do you think I know this CPU from?

> used to be found here, but not anymore, surprise?

It's still there, surprise:

http://www-05.ibm.com/hu/news/events/2007/bladenap/pdf/Hofstee_Cell.pdf (page 16 dating 08/06, page 17 dating 04/07, document dating 11/07)
http://www-06.ibm.com/itsolutions/jp/deepcomputing/events/pdf/ibm.pdf (page 10 dating 03/08)
http://www-03.ibm.com/technology/resources/technology_cell_pdf_Cell_computing_platform.pdf (page 17 dating 05/08)
http://www-05.ibm.com/no/news/events/lsu2008/pdf/cell_overview.pdf (pages 15/16 dating 11/08)

> it should be around available at 2010 (the roadmap was from 2006)
> but the PS4 is due out in 2012? 2 years delay, when PS3/original
> Cell release was almost simultaneous? Strange?

No, not at all. As I told you numerous times already, the PowerXCell 32 isn't supposed to have anything to do with PS4 and never was. This alleged connection exists only within your twisted mind.

> I like my PS3 and the Cell and it would be fun to play with
> something more powerful, but I somehow think it will take quite
> longer than expected, if at all.

What about the already mentioned PowerXCell 8i then? It's more powerful than the Cell B.E. and it's out for quite some time already. Too expensive you say? Then forget the PowerXCell 32 instantly.

> in regard to the other CPUs you mentioned, I actually don't see
> anything other than embedded CPUs.

Yes, quite like the QorIQ you mentioned yourself.

> They're great really for their purpose [...] but they're not much
> use for a desktop or even a netbook.

I know. No need to educate me on this.

> I stated (amongst other) that e600 (the de-facto desktop-oriented
> CPU line, at least so far)

Dont't forget the PPC970 :-)

> is dead, and we're left with just server and embedded lines (QorIQ
> is just the most powerful, I don't really care for the rest, which
> are many, because they're embedded-only and too low-power).

Don't try to take me for fool. Why not simply saying "only server and embedded lines" when you mean "only server and embedded lines"? Instead, you claimed there would be "only QorIQ and Power7" (the latter one being no problem as POWER is the only server CPU family within Power Architecture and POWER7 being the upcoming member) in plain denial of the existance of the many other embedded lines within Power Architecture which are supposed to get further developed. Do you wonder yet why your ridiculous claim got challenged?
And for your argument why you're caring or not: the POWER7 is not suited for desktop applications either, yet you mentioned it.

> what *new* CPU (-based system, ok) could an average powerpc fan
> choose? None.

Right. I never disputed that.

> Which was *exactly* my original point, which you still seem to miss.

No, it's not that I missed it. It's just that I didn't address it, which *you* still seem to miss. In order to avoid any misunderstandings I even quoted the exact statements of yours I was addressing, which you seem to have missed as well.

Quote:

I'm glad you find it amusing. Pity you missed the point. Again.

Quote:

Nice job, thanks, I'll keep you in mind when I need to find sth on Google.

Quote:

Ok, now you're talking BS. A company like IBM won't spend *hundreds of millions* of USD on R&D for a CPU if it only ends up in a couple of their supercomputers and a few thousand boards sold here and there. They have never done that. Even the 440 and 450 used in BlueGene systems are widely sold as embedded CPUs (perhaps slightly modified). The Cell itself *was* funded by Sony and the end product was sold in millions of units and if it wasn't for Sony we would probably never see the Cell. For the same reason, the PowerXCell 32 *initial* purpose was to go in the PS4. Just as the PowerXCell 8i is a bigger/better version of the Cell BE (which came *after* the Cell BE), so the 32-SPU Cell would find its way to the PS4. At least that would make perfect sense, before Cell proved a $1B money-hole for Sony. Go and do some reading first, before you post tons of links, where you probably just skim the title and post the URL, I'm tired of this.

Quote:

Yes, but unlike the QorIQ they're not powerful enough. FWIW, I know that QorIQ was even considered by Genesi as basis of a ppc-based system.

Quote:

I admit I was tempted...

Quote:

I have to say, I do wonder yes. After all this discussion, you still miserably fail to see the obvious. The PowerPC line is withering and limiting itself to just a few models/markets, whereas the competition has a plethora of choices for whatever need. You try to justify your reasoning with comments like "hey, you forgot the Titan and the IBM 472 and the A2 and all the -unrelated to the topic- embedded CPUs!". Considering the netbook target, the only *new* CPU fit for the job would be the QorIQ actually (being multi-core and able to reach high frequencies), which is why I mentioned it. And I also mentioned Power7 because it's the most important line currently (and the only one that is able to beat Intel/AMD offerings). I like the PowerPC and have been programming it for ages (have you?) but it JUST ISN'T THE BEST CHOICE ANYMORE. Why is that so hard to grasp?

Quote:

Oh, there you're wrong. It would make a fine super desktop, faster than most PCs out there. It even has VSX (altivec-like SIMD unit). If only one could afford it; and the electricity bill.

[ Edited by feanor on 2009/4/11 11:53 ]

> Pity you missed the point. Again.

I think I got the point, which is: You're full of double standards.

>> the PowerXCell 32 isn't supposed to have anything to do with PS4
>> and never was. This alleged connection exists only within your
>> twisted mind.

> now you're talking BS.

I don't think so. There has *never* been any statement neither by IBM nor Toshiba nor Sony that the PowerXCell 32 would be a candidate for the PS4. The only people assuming such connection were some rabid PS fans and random clueless analysts. You can convince me of your claim being true by presenting just *one single* official statement that goes along your view. Until then, you're the BS talker here.
Btw, the PowerXCell 8i wasn't supposed to power the PS4 either, yet it got R&D'd. Care to explain?

> A company like IBM won't spend *hundreds of millions* of USD on R&D
> for a CPU if it only ends up in a couple of their supercomputers
> and a few thousand boards sold here and there.

Maybe they had plans for their Cells that never came to fruition, who knows. Fact is, you (and me) know nothing about IBM's internal plans for the coming PowerXCell 32. At least, there's not one single document out there proving that R&D of PowerXCell 32 has anything to do with PS4. So you're obviously talking out of your behind again.

> They have never done that. Even the [...] 450 used in BlueGene
> systems are widely sold as embedded CPUs (perhaps slightly modified).

No, the 450 is not sold on the free market. It's only used within BlueGene systems. So, they indeed have done that before already in an even stricter way. The PowerXCell 8i is not only used in their RoadRunner but also on server blades and PCIe boards for the free market. Why shouldn't they do the same thing they already did with the PowerXCell 8i and (in an even stricter way) the 450 with the PowerXCell 32 as well?

> the PowerXCell 32 *initial* purpose was to go in the PS4.

Your claim is worth nothing without proof. Why has the PowerXCell 8i been R&D'd then?

> Just as the PowerXCell 8i is a bigger/better version of the Cell BE
> (which came *after* the Cell BE)

Yes, of course. That's been my point the whole time.

> so the 32-SPU Cell would find its way to the PS4.

From the PowerXCell 8i being a better version of the Cell B.E. you conclude that the PowerXCell 32 was intented for the PS4? What's that kind of logic called?

> At least that would make perfect sense

Really? The PowerXCell 8i didn't end up in a Sony's gaming device, yet it got R&D'd and gets produced and is sold on the free market as part of certain IBM systems today. Care to explain?

> Go and do some reading first

URLs please. You're telling me over and over to do some reading, yet you keep failing to provide me with even *one single* URL to do so.

> before you post tons of links, where you probably just skim the
> title and post the URL, I'm tired of this.

You're tired of me posting tons of links? Let me recommend you to get some sleep helping your tired and confused mind then. Else your delusions might reach a dangerous state.

> unlike the QorIQ they're not powerful enough.

You know the power of future 460s and future e500s other than QorIQ? Please tell.

> I know that QorIQ was even considered by Genesi as basis of a
> ppc-based system.

Genesi's considerations remind me of the caprices of a menstruating woman ;-)
Seriously: I expressed my feelings towards the QorIQ regarding desktop use (especially running MorphOS as this is MorphZone after all) already on page 2 of this very thread.

> After all this discussion, you still miserably fail to see the
> obvious. The PowerPC line is withering and limiting itself to just
> a few models/markets, whereas the competition has a plethora of
> choices for whatever need.

How do you conclude that I'm failing to see this? I'm quite *with you* on this observation! It's just that I didn't address this in any way, as I told you already. You simply *cannot* conclude any position of mine regarding a certain issue from me *not* addressing that issue!
Obviously, you're assuming too much. Please stick to the things I actually write, not to some strange fantasy assumptions in your head.

> You try to justify your reasoning with comments like "hey, you
> forgot the Titan and the IBM 472

It's the 476.

> and the A2 and all the -unrelated to the topic- embedded CPUs!"

Indeed, I justify my reasoning that you forget these CPU lines with comments like this. Any objections?
As for the unrelatedness of embedded CPUs: Why is the QorIQ related, but other future e500s and future 460s are not?

> Considering the netbook target, the only *new* CPU fit for the job
> would be the QorIQ actually (being multi-core and able to reach
> high frequencies), which is why I mentioned it.

And other future e500s and future 460s certainly won't fit netbooks?

> I also mentioned Power7 because it's the most important line
> currently

Most important regarding which aspects exactly?

> and the only one that is able to beat Intel/AMD offerings

Regarding performance, yes. Yet it's offtopic in any discussion about desktop use for the reasons you're going to name:

>> the POWER7 is not suited for desktop applications either

> there you're wrong. [...] If only one could afford it; and the
> electricity bill.

Thanks for presenting the reasons for the POWER7 not being suited for desktop applications. So, thanks for proving my point.

To sum it up: You claimed the QorIQ and POWER7 to be the only future Power Architecture CPUs and now justify your denial of the other lines with something like thread context, which is ridiculous especially regarding your mentioning of POWER7. You could as well have mentioned the e200 line instead for it's probably the most used Power Architecture CPU line of all. That would have been as much within this thread's context as POWER7. Your justification is composed of inconsistent reasons you now come up with to conceal your claim being false.

Btw, you still keep failing to back up your claim that Sony "first [...] started with a Cell 3 design" for the PS4.

Quote:

Well, sure, PowerXCell 8i has 5-8 times *more* double precision floating point math performance (actually that's the theoritical limit, usually it's more like ~4x faster), and allows up to 16GB of RAM. It's *targetted* for scientific simulations, where floating point is almost useless, and lots of RAM is of vital importance. The plain Cell is fine for games (32-bit floats are fine for a game, but they just don't cut it when it comes to scientific calculations, they're almost useless there). But I guess you never had to do that so you wouldn't know.

Quote:

Well, sure, let's see. Amidst the insults, I see a challenge. So, if we have an 8-SPU Cell used in PS3 good at 32-bit fp math and shortly after a HPC-targetted PowerXCell 8i, still 8-core but ~4x faster at 64-bit fp math, used in huge clusters and supercomputers, and we have an announcement of a 32-SPU Cell successor (2 versions, 32-SPUs each, 2/4 PPEs resp.). Now, why does that sound familiar? Oh wait, it might be because one model, one possible explanation might be that the 2-PPE one would be strong only on 32-bit fp and the 4-PPE strong on 64-bit fp as well, just like the Cell and PowerXcell 8i. Of course that's just a guess, because not much specs are available on either CPU model, in fact, it's only a rumour that there are two versions. Of course you wrote it as a fact, falling in the same trap you accuse me of.

Quote:

Oh, now you have inside info of the IBM CPU sales? Fantastic. However, I think this is funny:

http://www-03.ibm.com/press/us/en/pressrelease/7433.wss

check the part where it says "A third synthesizable processor core, the IBM PowerPC 450, targeted for the networking and communications segments, is expected to be available in 2006."

Availability for what? demos only? And you would know this, how exactly?

Quote:

Because the 450 is cheaper to R&D and produce and sell than the PowerXCell 32 by a big factor? Especially, when Sony or someone else is not interested in funding the R&D? What I mean is, if Sony is not sharing R&D costs, of what benefit would it be to IBM to take it up on itself to develop a 32-SPU Cell, only to put in a few super computers and sell a few thousand blades/boards? If you add up the costs, it doesn't make financial sense.

Quote:

Please read above.

Quote:

Please read above.

Quote:

Thank you that was a nice advice, I can see more clearly now that my tired and confused mind wasn't playing tricks on me. You firmly established my opinion of you.

Quote:

Sure, when they're here, they might, "future" being the key word here. QorIQ at least has already been shown and demoed, and available in mid-2009 -or so they say anyway. Same argument used as for the Titan, let's see it first, even as a demo and then we can talk all day about the merits of each CPU.

Quote:

Brand recognition. Most people know Power6/7 and its -allegedly insane- performance, but very few know names and specs or even existence of the other models. I guess you don't disagree there.

Quote:

You know, I've never met anyone so obsessed with proving the other one wrong. You remind me of http://xkcd.com/386/. SO WHAT if I neglected a few unrelated CPU models? I mentioned real and existing CPUs, which are the main focus points of the two biggest PowerPC companies, and you mentioned future ones. And you call me inconsistent? Let's wait first ok?

Quote:

http://www.businessinsider.com/2009/1/rumor-sony-ps4-plans-leaked-and-it-targets-casual-gamers-sne
http://www.gossipgamers.com/leaked-ps4-details-flatly-denied-by-sony/

These are ...

Bah, I just felt an strong earthquake now and I realized just how much I'm wasting my time replying to you. Last message on thread, enjoy your meaningless triumph.

[ Edited by feanor on 2009/4/12 2:59 ]

[ Edited by feanor on 2009/4/12 19:42 ]

Quote:

Just to add.
ST-Microelectronics STn8820
http://www.beyond3d.com/content/news/586

@feanor

Like Ageia?s PhysX PPU, Toshiba is attempting to add their SpursEngine (quad SPE only) as a part of the X86 PC ecosystem e.g. on selected Toshiba laptops and WinFast (Foxconn) PxVC1100 PCI-E add-on card.

It would be nice if SpursEngine is available as an ExpressCard.

Quote:
~50 percent of PCs sold in 2008 includes ATI CAL or NV CUDA** hardware.

**SIMT(single instruction multiple thread) model. It?s like SIMD but with multi-thread data dispach.

Attention all, please:

IBM decided to realize its version of processors for netbook computers!

Infacts IBM associated to Chartered Semiconductor, GlobalFoundries, Infineon, Samsung, STMicro and Toshiba creating a society named Joint Development Alliance.

New CPU will feature 28nm technology and will be 40% more powerful than Intel Atom consuming 20% less than Atom processors...

http://www.vnunet.com/vnunet/news/2240525/ibm-moves-28nm-chip-fabrication

[ Edited by Raf_MegaByte on 2009/4/20 12:00 ]

Quote:

...but this is a yet another ARM initiative. Doh.

If even IBM has lost interest in POWER architecture, who will make our favourite chips? We look more and more into a dead end.

The Amiga curse!

Quote:

We have two solutions:

1) Learn how to program ARM processors and migrate from PPC to ARM...

At least these processors are pure RISC and it not be should far more different from PPC RISC architecture...

[Question:

Are arm Big Endian or Low Endian??? IF ARM are LOW Endian then the porting should take a little (?) more...

...But we Amigans have a great skill in patience and waiting...

End of Question]


2) Continuing using PPC that are still manifactured from AMCC as those used in SAm440...

[ Edited by Raf_MegaByte on 2009/4/20 15:21 ]

@Raf_MegaByte

No.

Quote:

As Pilatus said in Jesus Christ Superstar famous Musical:

- What do you mean by that? This is not an answer!


1) "No" is intended to my question LOW or BIG Endian?

2) Better "No" to passing to Arm?

3) Better "No" to using still PPCs manufactured by AMCC?

@Raf_MegaByte

See my previous post.

|||




[ Edited by Velcro_SP 19.04.2011 - 16:19 ]

Just a little update in the whole CherryPal scam..

3643040463_a7fdddb535.jpg

@ ironfist

Yeah, I read the documents BBRV talks about in their latest blog:

http://bbrv.blogspot.com/2009/06/warningto-all-interested-parties.html

Very interesting...

|||

[ Edited by Velcro_SP 19.04.2011 - 16:12 ]

Quote:

So, at least one amigan has a Cherypal? I didn't notice about it, and would find very valuable your opinions, if you haven't posted them already. Please tell us, is it as bad as some people say?
Haev you started hacking it yet? Is it remotely possible?

|||

[ Edited by Velcro_SP 28.05.2011 - 06:29 ]

@Velcro_SP

OK, you don't like what I say, but am I wrong?

|||

[ Edited by Velcro_SP 27.04.2011 - 06:58 ]

Velcro, I agree with all your comments here. Even those "against" takemehomegrandma. Count me also in that category, as I usually do the same praising towards Genesi from time to time. Somehow, I feel in debt with them, perhaps some more people here feel like that (but wouldn't admit it in public).

I've red your blog entry about the Cherrypal ("LimePC" from now), and recall doing so in the past. Sorry for fooling myself, as I obviously red (and commented!) your MorphZone thread also, but had forgotten about it.

Your idea of "picking up LimePC" makes sense, at least for an ignorant like me. but it's hard to tell how difficult it would be. Being such a little compuiter, it should be VERY easy to do, or it wouldn't be viable, we're talking low cost computing.

In the past, I thought of an initiative consisting in buying a bunch of LimePCs and give them to the MorphOS Team, for them to check the possibility of a port. Hell, they are cheap! It's not like buying a bunch of PowerMacs G5! (yeah, right, as I'm dumb and can't do it myself, I pass that task to tme Team, as if they weren't busy enough).

The MacMini port is a good thing. Mostly because it's the ONLY possible thing. Doh.

But it's a pity the LimePC can't enjoy a decent operating system (and community!). Because LimePC and MorphOS deserve each other. Knowing that the original plan was about MorphOS only hurts more.

So, if we had the guarantee (yeah, right) that freescale would be manufacturing the MPC5121e for five more years, and THTF would do so with the motherboard... But those are, precisely, two companies known for NOT delivering on their announcements. Anyone here in China willing to reach THTF's offices for an agrement over LimePC? Wait, Genesi did so, and were screwed...

It's just a shame that the LimePC, indeed, does exist, and that "being born" for MorphOS, we can't do anything about it.

Or "yes we can" (TM)?