nVidia building the PS3 GPU in its "entirety"

[darkblu]

i hope you'll excuse me for playing on your fears, Dave.

What fears would these be?

I for one hope that we continue to pay in a transparent way rather than through hidden media fees - this is the primary reason why DiVX never took off in a consumer implementation and I thank god that it didn't.

fact is, i paid 90 CAD for a brand-name dvd/divx player upfront. whatever (divx) media fees may have been present, that price suits me fine as an average consumer.

Curiously my wife and I went shopping for my Christmas present yesterday which happened to be a Pioneer DV-575 DVD player with pretty much the same functionality – the old 515 has been here for several years and I figured it was time to put it out for pasture.

congratulations on that purchase. as i am pretty sure it will serve your needs better than whatever pc-based video equipment you may have had before (unless you're into serious video producing, that is).

Where have I suggested that this isn’t the case? Equally is a driven by a Cell like platform going to be anymore worthwhile?

and where have i suggested anything about cell's worth? for all i can tell, it's just an attempt in the right direction, how successful it will turn out to be we're about to see. what i have been saying, though, is that the wintel pc has spectacularly failed as a consumer product in meeting the bang-per-buck levels of the day. see, problem is that the wintel pc is only perceived as being a 'jack of all trades', where's actually it's a "jack of one trade", and much of the versatility it has achieved (mind you, not for the average pc consumer) is not so much because of the industry leaders, it's despite of the industry leaders.

Even though I’ve bought this player that still doesn’t mean that a Media Center PC isn’t attractive to me though – when the TiVo gives up the ghost it’ll be something that I’ll give some more serious thought to as with a nice high performance wireless network it can become a multimedia server for the rest of the house. Not everyone wants or needs that, but I don’t suggest that to be the case.

correct, far not everyone wants or needs a whole media network at his home. most people will be happy if they can pop in a medium in a device and get a decent audio-visual experience out of that. at a resonable price. now, if somebody can offer a home media network with a price/performance comparable to popping a disk in a player - hey, all power to them! sony say they can do it and we are yet to see what happens. ms have been continuously trying for some time now. unsuccessfully.

Now, turn around your comments: why do you think your $200 DVD player has all that functionality? If it wasn’t for the easy connectivity of PC’s would digital camera’s have taken off as quickly as they have thus there being a desire to have JPEG playback on consumer DVD players? Would CD/DVD-R/RW playback be as an attractive selling feature (yes, we’ve not got HD/DVD-RW combos, but long after PC’s were burning video to optical disks)? Would DiVX still even vaguely be here if it now hadn’t become an open source Codec that has become very popular in (a certain segment ) PC life? No, PC’s aren’t the sole driving factors in these, but they have most certainly been an acceleration factor.

aha. and you seem to be mistaking the wintel pc platform for the very concept of 'personal computing'.

 

[Dave Baumann]

i hope you'll excuse me for playing on your fears, Dave.

What fears would these be?

I for one hope that we continue to pay in a transparent way rather than through hidden media fees - this is the primary reason why DiVX never took off in a consumer implementation and I thank god that it didn't.

fact is, i paid 90 CAD for a brand-name dvd/divx player upfront. whatever (divx) media fees may have been present, that price suits me fine as an average consumer.

You get me wrong - I don’t care about DiVX now that its just an open codec. IIRC the initial DiVX implementation, as a competing format to DVD, was cited as a unit that has low disk purchasing costs but would dial-up and charge you if you continued to play back the disk; that type of costing structure makes me shudder. I far prefer a format where I buy the disk and own the disk to use without any worries of further fees.

[Edit]
http://www.dvdchannelnews.com/dvdvideo/?&doc=pages/editorials/dv_ed_006797yg&norecent

congratulations on that purchase. as i am pretty sure it will serve your needs better than whatever pc-based video equipment you may have had before (unless you're into serious video producing, that is).

I currently have a Pioneer DV-515, the preceding model by several years and generations.

and where have i suggested anything about cell's worth? for all i can tell, it's just an attempt in the right direction, how successful it will turn out to be we're about to see. what i have been saying, though, is that the wintel pc has spectacularly failed as a consumer product in meeting the bang-per-buck levels of the day.

And I’m not suggesting that isn’t the case, all I’m saying is that PC’s as a multipurpose computing device are versatile and will probably continue to evolve in order to keep its relevancy in its core market and continue to explore the boundaries beyond. So far its seen “reasonable†success as a gaming platform and we’ll have to see how it goes beyond that.

aha. and you seem to be mistaking the wintel pc platform for the very concept of 'personal computing'.

And in that instance it’s the wintel platform that has been a significant driver for all these formats to be recognised as a selling feature in CE devices.

 

[Megadrive1988]

kinda off-topic:

I myself have Dreamcast, PS2, Gamecube and Xbox. I enjoy all 4 consoles quite alot. moreso than I did with the mainstream consoles of the previous generation: Saturn, PS1, N64. I find that all of the current consoles offer good technology and have great games to make use of it. I'm very much looking forward to the entire nextgeneration. not just the consoles, but the handhelds as well. all 6 mainsteam platforms i will own. i will get these platforms regardless of what graphics chip is in each. although it very interesting to see these things develop and with whos technology. at the end of the day its about what games appeal to you.

p.s. the Nvidia PS3 GPU win is huge for Nvidia. it's bigger than the technology comeback Nvidia has made from the flawed NV3x-GeForceFX to the extremely solid NV4x-GeForce 6800.

 

[Lazy8s]

Crazyace:

The high alpha blended fill rate of the GS is it's strongest point by far.

Unfortunately, the more it flexes that strength, the worse dithering becomes due to its inability to maintain full color precision when blending.

 

[arhra]

You get me wrong - I don’t care about DiVX now that its just an open codec. IIRC the initial DiVX implementation, as a competing format to DVD, was cited as a unit that has low disk purchasing costs but would dial-up and charge you if you continued to play back the disk; that type of costing structure makes me shudder. I far prefer a format where I buy the disk and own the disk to use without any worries of further fees.

DivX the video codec and DIVX the short-lived DVD competitor are completely unrelated (divx-the-dvd-competitor used standard mpeg-2 video, i believe), apart from sharing a name (which is now a registered trademark of DivX Networks, Inc). IIRC, in the early days, divx-the-codec was 'officially' called 'DivX ;-)', or something, to avoid trademark issues with divx-the-dvd-competitor.

Actually, doing a bit of googling, it seems the real story is even more complicated... There was divx-the-dvd-competitor, 'DivX ;-)'-the-(legally-dubious)-hacked-MSMPEG-4-codec (aka DivX 3.xx), and eventually divx-the-legal-mpeg-4-codec (aka DivX 4.x+).

No wonder people get confused...

 

[Entropy]

Maybe, just maybe, x86 is king becuase people want x86.

I'm not sure I see what the fascination or issue with "x86" actually is myself, a P4 or Athlon 64 has evolved to the demands of the modern PC and they will continue to do so. I'm really not sure whether breaking out or not of "x86" is really going to be an issue, and frankly I don't really care.

I think the world is with Dave on this one.
More importantly than world+dog, I think Microsoft holds the same position. They saw no need for x86 for PDAs, they see no need for x86 for smartphones, they see no need for x86 for XBox2.

x86 is a crufty architecture and even Intel openly acknowledges the challenges that backwards compatibility represents in terms of controlling complexity and corner cases. I really can't see where MfA comes from when he said that he would like to see x86 extended into the massively parallell realm. At what point do we want to acknowledge that it may be a good idea to start with a clean sheet, rather than keep extending the 4004? Never? Does anyone really think x86 is the the pinnacle of processor design regardless of target application?

I seriously believe that it would be preferable if the industry got started with a new general purpose architecture geared towards parallellism, and the x86 was relegated to clerical computing, implemented in more and more energy efficient manner over time, suitable for administrative environments. I doubt it will happen, but how anyone without a vested interest (and of course there are a lot of those about) can think that x86 is a good candidate for a multiprocessing future is beyond me. Even Microsoft, who really have a vested interest isn't that loyal.

And what this discussion has to do with nVidia contributing to the pixel processing of the PS3 is anybodys guess.

 

[aaronspink]

DVD player? , CELL is the future not present

So was Communism!

Aaron Spink
speaking for myself inc.

 

[Mulciber]

DVD player? , CELL is the future not present

So was Communism!

Aaron Spink
speaking for myself inc.

wow.

 

[aaronspink]

[
OK. Speed is relative. You are still implying that it is fast enough. This I disagree with. I don't for a minute suggest that the industry doesn't adapt. I say it adapts too slowly.I still don't see any reason why it has lasted as long as it has other than its shear size and Microsoft's strangle hold on the global market.

Because it in fact does adapt fast. Everyting sony talks about with CELL you can buy today from multiple companies (Clearspeed being but one example). There are many experiments happening in the PC arena constantly. The PC space is generally where the new technologies are created and researched.

 

[aaronspink]

And you're telling me to get out of my hole? Welcome back to 2000 with that thinking.

Vince, when you show an understanding of the industry, I'll stop tellling you to get out of your hole.

Dr. Frank Soltis, PDF of Cell Processor's use in future IBM Servers: http://www-1.ibm.com/servers/uk/consultants/pdfs/Dr_Frank_Soltis_Interview.PDF

I can keep going... or you can shut up and do it yourself.

Well, first of all your link doesn't work. And second of all, bring it on.

Cell has about as much chance of being IBM's server processor as it does of eliminating x86 or taking over the PC. Sure they may do some publicity stunts, but that is all they are.

Aaron Spink
speaking for myself inc.

 

[aaronspink]

As for the PC having died already? As I have already mentioned, The industry is simply too large and entrenched to have suffered such a fate. There hasn't exactly been a lot of competition for it either has there? Certainly nothing that could compete with a complex monolthic entity such as the PC industry, that's for sure.

Not a lot of competition? Boggles the mind.

First you have IBM with their Microchannel strategy, abject failure.
Second, you have Compaq et al with their MIPS based strategy (at a time wen Compaq controlled 30+% of the market). It of course failed.
Third you have the imfamous AIM grouping of Apple, IBM, and Moto. Failure.
Amiga... Failure
Acorn... Failure
Network Computing... Failure
Sun's various attempts.. Failure

There hasn't been a lot of competition because the competition always fails.

Aaron Spink
speaking for myself inc.

 

[aaronspink]

4004? Never? Does anyone really think x86 is the the pinnacle of processor design regardless of target application?

Nobody believes that it is the pinnacle. The question is if you can sustain a minimum of 2x delivered performance advantage over the 3 generation life span in releation to x86. So far, no one has. And that includes the brightest minds in architecture, microprocessor design, and circuit design. Unless you can demonstrate that x86 isn't "good enough" then it will always be the prefered embodiement of the microprocessor.

Even Microsoft, who really have a vested interest isn't that loyal.

Umm, they are extremely loyal. Loyal to the tune of over 100% of their profit. When they port Office and fully support Office on another platform besides x86 then we can have a discusion about their loyalty

Aaron Spink
speaking for myself inc.

 

[j^aws]

Even Microsoft, who really have a vested interest isn't that loyal.

Umm, they are extremely loyal. Loyal to the tune of over 100% of their profit. When they port Office and fully support Office on another platform besides x86 then we can have a discusion about their loyalty

Aaron Spink
speaking for myself inc.

IIRC, that Compaq Mips consortium was known as the ACE consortium. They had a HAL that supported various incarnations of Windows NT on various processors, MIPS, Alpha, x86, PowerPC and various OS's e.g OS/2. Of course the free market decided the Wintel combination.

Also, fyi, Office is currently supported on PowerPC and Mac OSX.

And finally, that list of failures you produced, I've lived through but that doesn't change the fact as me, a consumer, is not getting what I want because of that CRUD that is teh Wintel box! Still there's hope yet as both x86 and windows haven't been invited to this upcoming show of the Linux/CELL party. We'll see what kind of impact it makes. Even a minimal uptake as a 'home computer/games machine' ala Amiga fame would be deemed a success in my eyes.

I've seen manier decent tech die in this industry and a personal recent fav was a 'clean sheet' state-of-the-art multimedia OS called BeOS but Mac OSX will suffice. And as a tech head and a consumer I hope there is a choice that I can continue to make without having teh Wintel box rammed down my throat.

 

[Entropy]

4004? Never? Does anyone really think x86 is the the pinnacle of processor design regardless of target application?

Nobody believes that it is the pinnacle. The question is if you can sustain a minimum of 2x delivered performance advantage over the 3 generation life span in releation to x86. So far, no one has. And that includes the brightest minds in architecture, microprocessor design, and circuit design. Unless you can demonstrate that x86 isn't "good enough" then it will always be the prefered embodiement of the microprocessor.

You neglect power draw as a factor - one of the major reasons x86 has no presence in smartphones. Is x86 a good platform to build massively multicore processors? Not really. Can it be done? Sure.
But - why should we do so?
It remains to be seen if x86, while retaining a compelling advantage for legacy codes, can scale nicely as the number of cores that can be squeezed onto a chip increases.
Apparently Microsoft felt that staying with x86 for the XBox2 would have been a loosing move, and Microsoft has everything to gain by keeping their console as close to the Wintel PC standard as possible.
(It should be noted that those "brightest minds" you refer to above who have failed to supply a sustainable factor of two performance advantage, haven't had access to the "deepest pockets". The Windows world is a huge market to spread investment dollars over, as you have pointed out yourself, and x86 has benefitted from that.)

Now I tend to agree that Cell is unlikely to take the world with storm, but I see that more as a sign of maturity and inertia of the personal computing market, rather than inherent advantages to x86.

Even Microsoft, who really have a vested interest isn't that loyal.

Umm, they are extremely loyal. Loyal to the tune of over 100% of their profit. When they port Office and fully support Office on another platform besides x86 then we can have a discusion about their loyalty.

Microsoft will do so when there is money in it. Thus, maybe never. This has nothing to do with x86 having any particular merit, even for clerical computing.
The point I made was that even when Microsoft wanted to leverage their Office advantage in the extremely portable segment, they merrily supported a more suitable architecture. For the XBox2, they again actively chose another architecture, believing it more suitable for the target application. Enough so that they felt it worth to distance themselves from the PC standard. Now, if the XBox2 turns out to be a success, and Microsoft truly makes a PC type version of it, why wouldn't they port their Office products to it? After all, they expand their market, and their share of the revenue.

Microsoft sits in an extremely profitable boat. Of course they do not want to rock it unless doing so holds the promise of greater profit still by expanding their market, or reserving a bigger slice of the cake for themselves, or preempting a challenger. x86 is merely a tool, and when found to be a liability to Microsofts overall goals, it has been dropped. In clerical computing there is little such impetus though, as long as the CPUs do not get too awkward to to handle in terms of thermal output.

We are probably fairly alike in how we regard the likelyhood of Cell to make an impact on the PC paradigm. (Very low.) But we seem to differ greatly in what we base that prediction on. Also, I'd contend that what we now percieve as computing niches outside that paradigm, with smart phones and gaming consoles serving as good examples, may continue to grow their share of total computing dollars spent.

 

[aaronspink]

You neglect power draw as a factor - one of the major reasons x86 has no presence in smartphones. Is x86 a good platform to build massively multicore processors? Not really. Can it be done? Sure.

Never neglected power draw. Performance is pretty all incompassing. For the mainstream market, there are x86 products that match both power and performance.

The main reason you haven't seen an ultra low power (< 1w average) x86 is there hasn't been a need. There is nothing technically that prevents it.

It remains to be seen if x86, while retaining a compelling advantage for legacy codes, can scale nicely as the number of cores that can be squeezed onto a chip increases.

why wouldn't it?

Apparently Microsoft felt that staying with x86 for the XBox2 would have been a loosing move, and Microsoft has everything to gain by keeping their console as close to the Wintel PC standard as possible.

Maybe the two major x86 vendors thought it wasn't a worthwhile design win. They can probbaly make more money selling on the open market and neither seem to have an issue with selling all that they make.

(It should be noted that those "brightest minds" you refer to above who have failed to supply a sustainable factor of two performance advantage, haven't had access to the "deepest pockets". The Windows world is a huge market to spread investment dollars over, as you have pointed out yourself, and x86 has benefitted from that.)

I'm pretty sure that the combination of IBM and Motorola have pretty deep pockets. Money has very little to do with it. The main problems are physics and market economics.


Aaron Spink
speaking for myself inc.

 

[Sonic]

Microsoft does not think any current or near future x86 design is good enough for Xbox2/Xenon. I think the logic of AMD or Intel not being interested in the project is highly flawed. It doesn't really matter if they are interested in the project or not, it is a matter of what MS wants. MS clearly thinks that IBM has a higher capability of designing and producing a more powerful chip than any x86 processor out on the market now or in the pipeline. Even if AMD or Intel weren't interested and MS wanted them to make a CPU it wouldn't be hard for MS to spark interest so to speak.

 

[ultimate_end]

This comment has me absolutely baffled, especially in relation to consoles since they spend several years creating the standard and then you are left with only that standard until the next one comes out. With the console environment you are stuck for 5 years and then they often end up bashing out an entirely new standard for the next one, leaving developers another couple of years scratching their head trying to get the best out of it (Sony anyone?). One the otherhand every single cent you pay for the PC is going directly into the pockets of the silicon vendors – you also have more of a choice to tailor the PC towards your needs.

Bashing consoles proves what exactly?

Anyway. I agree that the PC is configurable to users' needs. That is one of its strengths as well as one of its weaknesses (too many different types of components to buy, frequent upgrades, every component has to be profitable). I am glad that I don't have to pay for a GeForce 6800 Ultra if all I need is an FX5200. But then, how would a "wired household" be any different? At least I wouldn't be paying for x86 legacy support and all the rest of the PC baggage.

As for standards taking multiple years to bash out, that may well be the case, but, using DirectX as an example, the standards that are being hammered out now aren’t actually possible now, on current silicon processes, with any level of performance – they are generally timed such that the new features are available in a timeframe when they can sensibly be implemented at the same, or higher performance than the previous iteration. Whats wrong with that?

You're right. In the case of DirectX, I admit that it generally is a forward looking process, an example being Global Illumination support in future iterations. But if I can use a comment you made as an example, about nVidia not being keen on implementing Unified Shader Model in hardware. nVidia's opposition to this must surely hold up progress in deciding the direction that GPU are taking? In fact, why have we not seen USM implemented already? Surely it isn't constrained by current transistor budgets? Speaking of transistor budgets, if the overall PC was more media friendly, then maybe we might actually be further down the road, in terms of graphics technology.

And why have we seen the delay in DirectX 10? It's got nothing to do with my argument, I'm just curious.

I recently sat through a technology presentation with ATI and one of the things they were discussing is the latest high definition video formats and how they were going to resolve the quality and battery life issues for laptops with these new movie formats. Now, they know that the standards are either going to be Blue Ray or HD-DVD or both, but frankly they don’t care which it is since the platform is immaterial to the actual issues – they’ll be drives available for both on the PC pretty much as soon as either are available in consumer devices. What they are doing is looking at the issues of how to ensure that a laptop can play a HD movie at HD quality still on a battery charge, much as they can on current laptops. By the time HD players actually start hitting shelve with any great popularity these issues will more than likely be solved. This doesn’t look like its not a forward looking industry.

I don't dispute that the PC industry looks slightly ahead of itself sometimes. In the case of HD formats, yes the PC is adapting to current consumer demand, HD content is an outside influence. It's not exactly indicative of the PC pushing the boundries of technology though is it? I also find it ironic that you use HD-DVD/Blu-ray as an example, as it was again consumer electronics companies that created these formats. Could that possibly be Sony and Co dictating terms to the PC industry? Surly not.

There hasn’t been a lot of competition or just that the rest has failed? If you believe there hasn’t been competition before how do you know how quick or slow the the PC industry will adapt in the face of competition?

All I'm saying is there hasn't been much and few were particlurly viable. Some of the competition failed due to their own shortcomings, while some of the competition failed because it is perhaps far too ambitious to try and compete with the likes of the PC with its ubiquitous nature and its all powerful Dictator, Microsoft. Anyway, why should the PC industry have to always be reactive to competition? If the PC industry was proactive enough, there would never have been a need for that competition to arise in the first place.

Its is Apples inability, or lack of desire, to tackle the PC component vendors within the industry on such fronts – it’s the vendors and drivers of the PC industry that we’re really talking about when we’re talking about the future of the platform.

And who is Apple to confront the might of Microsoft? You mess with the component vendors, you mess with Microsoft, essentially.

This is the case wherever you go – look at the mess of the DVD specifications, now they really took years to hammer out and they they are full of specification wholes (the spec didn’t even decide on a single method of lay changing!) and we have a whole litany of license fees going from one vendor to the next. Are you happy that every disk has a Dolby soundtrack on there, whether its needed or not, freezing out some DTS translations, and that every disk is paying Dolby some IP?

Yes, there will always be some delays, that's obvious for anything other than strictly closed system. In the case of the DVD market, it's really quite similar to the PC industry, except it's worse because there is no one company that has any real overall influence. At least though, they only had one format to fight over, rather than countless potentially incompatible components that will be fought over until forever.

No I most certainly am not happy that we pay for every disc to have DD encoding when that may not be required. For me personally, It is even more of an issue considering my tendancy towards DTS encoding, which is not so commonly available. But the end result is no different from the PC industry, where Microsoft's dominance and anti-competition nature mean that consumers essentially pay for features in their OS that they may not need or want. It's not like the PC industry is 100% free from licence fees either. It may seem that I'm conceding your point here, but in fact I think the two industries are too similar to compare. It was a bad example IMO.

When you start looking at it from the consumer direction things are no different, especially when you have consortiums driving the groups and then competing consortiums trying to get the better standard (look at the HD Video format I mentioned earlier).

Such "format wars" occur relatively infrequently and again, it is too open of a market to use it a comparison against the PC industry.

In fact, frankly it can be worse in these scenarios since you have multiple vendors trying to compete in the same grounds; at least with PC’s there are actually only a few competing groups within each sector so there can actually be less infighting and each of the different areas generally trying to pull in the similar direction.

Pulling in the same direction or towing the same line? If it wasn't for Microsoft, I imagine that the entire PC industry would be brought to its knees. Even with Microsoft in control, The spectre of incompatability still rears its ugly head. Anyway, as we approach the time of true ubiquitous computing, we may find that consumer electronics companies adapt more quickly due to their simpler "common interest". They may even adopt a universal standard such as Cell and rely on true product differentiation to sell their products. There is not a lot of room for product differentiation in the PC industry other than a scale of general performance. Again we have Microsoft to thank for that.

PC vendors can equally be very flexible in the technologies that they support – they just make the platform scalable enough to support the general needs of future technology; again using the HD Video format the actual technology settled on, be it one or both, is immaterial to the companies in the industry as, if they have a high enough bandwidth platform already in place, they can support either.

Yeah the platform is scalable, but only in the short term. I will have eventually replaced every component in my PC within a relatively short space of time. In spite of the PC industries apparent obsession with backwards compatability over a multitude of products, I still find that the Socket 370 motherboard that I bought becomes pretty useless pretty soon. The various industries that we have mentioned all share this in common, but it is the PC industry that does it most inefficiantly in terms of technological advancement. If you want to talk about scalable, I suggest you look no further than something like Cell, which is truly dynamic and universal yet is not anchored to a pre-media-revolution age.

The PC is like some kind of ancient old castle upon which generation after generation of owners keep adding bits and pieces, in a desperate attempt to adapt it to their modern needs. Contrast this to a modern modular kitset home built from scratch. It might not have a lot of character, but at least it sure is efficient.

But we’re not talking about that, we’re talking about a closed standard driven by a few vendors and manufacturing companies with a vision to a widescale devices utilisation – in that scenario you can either end up with the situation of the other consumer vendors having to pay large IP fees to license the technology or they’ll end up fighting both other CE vendors as well as the PC industry.

Yes that is what this is all about. However, in the case of Cell for example, the licencing fees may not be that large, if any at all. Once Sony and Toshiba have profited from Cell for a while, perhaps even denting the PC space, STI can openly licence the technology to anyone who wants it. The investment will have been recouped by then and all the companies can focus on what is important to them, which is making great products. I have a feeling that in the future, CE companies will be worrying more about facing off against the PC than fighting amongst themselves. You mention paying huge IP fees? Remind me again how much money Sony or Philips make from the Compact Disc?

All that may not even be necessary if STI becomes the next Wintel. If this happens, then all companies will be unified in the same way that the PC is today. The irony of these scenarios is ofcourse that one unweildy super organism will be replaced by another one. But it might be worth it to finally get rid of x86.

As for the other scenario of companies fighting each other. IMO, in the future, we are likely to see two, at most three, major competing architectures. Cell will be one of these. This scenario brings together the best elements of the open vs closed platforms. The architectures can be easily changed and controlled because they are closed, but yet competition between architectures will force rapid technological improvement, which wouldn't happen so much in an entirely universal closed platform.

I get the impression that you are taking the side of the PC as we rapidly grow towards convergance (collision?) of the PC and consumer electronics industries. For my own reasons, I take the opposite side in this coming war.

But in the end, I guess it will be the consumer who ultimately decides the outcome and that is, ofcourse, the way it should be.

 

[ultimate_end]

As for the PC having died already? As I have already mentioned, The industry is simply too large and entrenched to have suffered such a fate. There hasn't exactly been a lot of competition for it either has there? Certainly nothing that could compete with a complex monolthic entity such as the PC industry, that's for sure.

Not a lot of competition? Boggles the mind.

First you have IBM with their Microchannel strategy, abject failure.
Second, you have Compaq et al with their MIPS based strategy (at a time wen Compaq controlled 30+% of the market). It of course failed.
Third you have the imfamous AIM grouping of Apple, IBM, and Moto. Failure.
Amiga... Failure
Acorn... Failure
Network Computing... Failure
Sun's various attempts.. Failure

There hasn't been a lot of competition because the competition always fails.

Aaron Spink
speaking for myself inc.

You're quite an entrenched PC fanatic aren't you?

BTW, "competition" isn't always quantified by the number of competitors.
If you can give me a list of reasons why all those attempts failed, then we will talk.

In the mean time, hows about you reply to my comments I made on your reply to my original post, instead of taking potshots at my comments to other posts.

 

[jvd]

Microsoft does not think any current or near future x86 design is good enough for Xbox2/Xenon. I think the logic of AMD or Intel not being interested in the project is highly flawed. It doesn't really matter if they are interested in the project or not, it is a matter of what MS wants. MS clearly thinks that IBM has a higher capability of designing and producing a more powerful chip than any x86 processor out on the market now or in the pipeline. Even if AMD or Intel weren't interested and MS wanted them to make a CPU it wouldn't be hard for MS to spark interest so to speak.

That isn't true .

While intel has the fab space to run a console line amd does not. I'm sure amd would have loved to have gotten an xbox2 contract for the a64s but the fact of the matter is they would have had to loose alot of thier long term profits to produce it . In 4 years when an athlon 64 3000+ or whatever they would have used was selling for almost nothing to ms for the console they would still need to devote fab space for it


Intel on the other hand already went through this . They are making 733mhz celerons for the xbox . They are using up fab space that they too can be using to make thier 900$ cpus that sell well .



Add this into the fact that the x86 chips in the pc now are geared towards general computations and being able to run word extremely fast while having aol running , a movie up and other things going on and not focusing on games and its clear why ms did not go x86.

I'm sure if amd and intel wanted to they could desgin a high flop x86 cpu . But the fact is they are making chips for another reason .



THe pc will allways be here . 2 years after the xenon and ps3 launchs the r700 and nv60 or 70 will come out and offer better performance and graphics than the consoles can offer .

 

[Vince]

Dr. Frank Soltis, PDF of Cell Processor's use in future IBM Servers: http://www-1.ibm.com/servers/uk/consultants/pdfs/Dr_Frank_Soltis_Interview.PDF

Well, first of all your link doesn't work. And second of all, bring it on.

First of all, The article that was linked to concerning Toshiba showed how faulty your thinking was concerning each memeber of STI's role. Secondly, That's odd that it's down; nAo, Mfa or Panajev can confirm the article. Ohh, and you crack me up -- 'bring it on', heh.

Cell has about as much chance of being IBM's server processor as it does of eliminating x86 or taking over the PC.

Ok, as I said, things are inevitable at this point...

Unless you can demonstrate that x86 isn't "good enough" then it will always be the prefered embodiement of the microprocessor.

Textbook case of an "individual" using argumentum ad ignorantiam. And I bet he wonders why people have problems with what he says. If your arguments aren't even logically self-consistent, which is pretty fundimental, how do you expect to get anywhere with them? You and Dave need to rethink the logic behind this argument.