Android x86 phone

That's interesting. I guess that's part of the reason Google released x86 support for their emulator, which I've heard is smoking fast on an intel PC. I've yet to try it because they also upped the ARM emulation significantly.

It's interesting that Intel is pushing a single core with HT (and pretending that's a dual core). It'll be interesting to see how it does.
 
Gotta be honest here and say I hope that Intel fails on this one. Watching ARM come so far so quickly, only for Intel to wipe out yet another arch would be more than I could bare at this stage tbh.
 
I agree and disagree with the-leander. Certainly killing off a platform is not a good thing for the market. Though if competition is good it's a good thing that Intel and hopefully AMD will get into the mobile race. AMD was actually there at one time but sold their ARM technology to focus only on x86. I'm sure they're kicking themselves. And doesn't IBM have any x86 items left? Perhaps we could see them jump back in and/or approach the marketplace with a PowerPC solution.
 
Ya, it's hard to say. You should at least take some comfort in the fact that Intel appears to be a little scared. I mean, with Windows now fully embracing ARM, Intel should be a little worried. And Intel will face some very entrenched competitors in the market and companies like Samsung aren't likely to abandon their own line of ARM chips in favor of Intel. The other interesting thing about Intel is that they have suffered many humiliating defeats in the past whenever they've attempted to compete outside of the x86 world they tightly control, so by no means is their venture into the mobile market on solid ground.

I'm not an embedded programmer and haven't looked at assembler since I toyed with it on the Amiga, so I can't really say which CPU architecture I prefer. However, I do know that those familiar with ARM seem to really like it and I expect Intel chips to be burdened with a fair chunk of legacy crap.
 
It's a mature processor on a mature 32nm process with optimised software and it's ...competitive with *lasts year's* 45nm chips. It'll be more interesting when the 32nm and 28nm parts start appearing.
 
I agree and disagree with the-leander. Certainly killing off a platform is not a good thing for the market. Though if competition is good it's a good thing that Intel and hopefully AMD will get into the mobile race. AMD was actually there at one time but sold their ARM technology to focus only on x86. I'm sure they're kicking themselves. And doesn't IBM have any x86 items left? Perhaps we could see them jump back in and/or approach the marketplace with a PowerPC solution.
Competition is good, but it needs to be fair and Intel has had a long history of doing things that are "questionable", to put it politely. I'm sure TheLeander could elaborate on his feelings, but my guess is that's the primary reason he appears to be anti-Intel. And I think that's a valid view.

Aside from that, Intel competing with ARM should be interesting. Ignoring AMD for the moment, Intel has the ability to make some very quick changes and advancements to their architecture AND bring it to market quickly. ARM might be able to produce better designs over time, but bringing them to market may take longer as they are at the mercy of 3rd parties for that. Does ARM make reference chips? I suppose they could, but that's costly and may ruffle some feathers amongst manufacturers. So it'll be interesting to see how ARM reacts, but I imagine this is no surprise to them.

As for AMD, well, the other way to think of this is now Intel is opening a new front. That often means a change in focus. And as big as Intel is, that can create opportunities for it's competitors. AMD just seems to be in the wrong spot, their biggest asset is their GPU technology which makes them big winners in the home entertainment market, but the CPUs are totally not geared for the home market at all and are making sales only on low-balling Intel chips. But that's a losing business plan as Intel has been driving down the cost of their chips considerably and the left over Sandy Bridge stock (making way for Ivy Bridge) will be selling at Bulldozer prices. Sorry AMD, but OUCH! So for AMD's sake, hopefully Intel will make some very costly mistakes in the mobile market and vindicate AMD's decision to stay out of that fight. Good luck with that.
 
Gotta be honest here and say I hope that Intel fails on this one. Watching ARM come so far so quickly, only for Intel to wipe out yet another arch would be more than I could bare at this stage tbh.

That could prove rather difficult even for Intel - Even if Intel was to completely own the smartphone market ARM would still be very profitable.
 
The problem with intel entering this market is the reliance on many of the handset manufacturers (Asus, Samsung etc) on intel parts for their desktop lines, look at how they got spanked in the EU courts for cutting AMD out of the desktop sector through single vendor discounts a few years back?

I don't think for a second they wouldn't stoop to threatening to up the prices of desktop parts now that they have a viable chip for mobility.

Having effectively only one vendor offering everything computer wise is an inherently bad idea. I'm sorry, but AMD aren't even a blip any more and haven't been for at least half a decade at this point.
 
Competition is good, but it needs to be fair and Intel has had a long history of doing things that are "questionable", to put it politely. I'm sure TheLeander could elaborate on his feelings, but my guess is that's the primary reason he appears to be anti-Intel. And I think that's a valid view.

Pretty much bang on.

I'm anti monopoly. As you say, Intel have been curb stomped already in the EU for their practices against AMD in the desktop sector.

Whilst they might not wipe out ARM as an arch, they could rend its development to nothing as they did PPC and MIPS.

I for one would love to see how the A15 cores pan out, because the dual A9 (OMAP4) pad I have is a truly astonishing piece of kit.
 
Whilst they might not wipe out ARM as an arch, they could rend its development to nothing as they did PPC and MIPS.

Better comparisons might be Alpha or PA-RISC which they killed outright. These, like PPC and MIPS were all made in limited quantities and sold for high prices. Intel had huge volumes so they could do something that was "fast enough" but cost a lot less. The RISC vendors couldn't compete with that. Intel took the entire workstation market and a big chunk of the server market.

With ARM it's the other way around. Intel chips are made in small volumes compared to ARM (currently 1.9 billion per quarter) and cost a lot more.

Intel's big advantage now is process tech, they pour huge amounts of money into this and are the undisputed leaders. However all this costs money - that they get mainly from server chips.

This is the second front in the battle, ARM is moving into servers. If ARM can provide something that is "fast enough" and cheaper than Intel, then that could force Intel's margins down and that could really hurt Intel.

Intel relies on high margins, ARM doesn't.



BTW I should point out I am a little biased here - I happen to work for a certain processor IP vendor in Cambridge.
 
Whilst they might not wipe out ARM as an arch, they could rend its development to nothing as they did PPC and MIPS.
Seems to me PowerPC is doing quite well. There's over 200Million PowerPC users out there. And that console headcount didn't include the numerous PowerPC chips used for embedded microcontrollers. One can find PowerPCs in automobiles and jet fighters. For example, the MPC5xx can be found in Jags, Fords, and GM vehicles.
 
Almost all of that though is low end stuff with the exception of the consoles. In terms of raw performance, I'm not sure that PPC has moved much beyond the 970 used in the last gen of PPC macs.
 
True most is lower end. But 2007 brought the quad core blue gene supercomputer. So it can be made to scale. It's a bit tricky as the Power line implements the PowerPC instruction set. That being so the Power/PowerPC capable chips have the majorithese are of Unix. I wonder if the hexadecacore powerpc xbox3 rumors are true.
 
True most is lower end. But 2007 brought the quad core blue gene supercomputer. So it can be made to scale. It's a bit tricky as the Power line implements the PowerPC instruction set. That being so the Power/PowerPC capable chips have the majorithese are of Unix. I wonder if the hexadecacore powerpc xbox3 rumors are true.
The next xbox and PlayStation will be x86 based. Surprise surprise, the PS4 will likely be AMD based. Turns out CPUs aren't that important anymore, GPUs are.
 
With ARM it's the other way around. Intel chips are made in small volumes compared to ARM (currently 1.9 billion per quarter) and cost a lot more.
ARM is made in large quantities but by many licensees. ARM and MIPS are licensed all over the place to lots of people who use them as cores on integrated embedded chips with DSPs and peripheral controllers etc all on one die.

It really depends on the application. Maybe you want insane mips, maybe you want Methuselean battery life, maybe you want a one chip solution, maybe you just want cheaper than dirt. Sometimes you're not a big enough order for the giants to pay you any attention - lots of factors - plenty of room for competition.
 
The next xbox and PlayStation will be x86 based. Surprise surprise, the PS4 will likely be AMD based. Turns out CPUs aren't that important anymore, GPUs are.
Perhaps. Rumors are an AMD Playstation and a PowerPC Xbox. We know the WiiU is a PowerPC.

I suspect that the lower quality games of phones and tablets are making a dent in gaming sales. I wouldn't be too surprised to see consoles next gen have worse sales than this gen because of it. For example - the iPod increased lower quality music sales and the higher quality CD and Record sales were impacted because of it. It seems people value convenience over quality.
 
Rumors are an AMD Playstation and a PowerPC Xbox.

Next XBox rumors are all over the map. Personally, I think they have a few different mock-ups and are watching when they need to get the box on the market before they make the decision which one to pull the trigger on.

Also, in addition to the next conventional console, they may make an XBox Live box that is more market-focused than the full console. More like a cut down media box for the Netflix, Hulu, Zune, Arcade-style games, and other services currently offered through the XBox Live. So the specs for that potential machine get mixed into the talks, as well.

I suspect that the lower quality games of phones and tablets are making a dent in gaming sales. I wouldn't be too surprised to see consoles next gen have worse sales than this gen because of it. For example - the iPod increased lower quality music sales and the higher quality CD and Record sales were impacted because of it. It seems people value convenience over quality.

While you may be correct, I don't think it will be nearly that dramatic a shift. The time people spend playing with an iPhone or Android is a very different activity from sitting down to play a game of Madden, Call of Duty, or Skyrim. The X360 and PS3 are already squarely planted in the enthusiast market.

CDs were not at all an enthusiast market. Very few of the total CD sales were to music enthusiasts who sat down and made an activity of listening to them. So, when something easier came along, the masses went there. But those few hardcore enthusiasts are the ones still buying CDs and records.

In this case, the X360 and PS3 are already down to primarily enthusiasts. I don't expect them to drop much more. Though, it could well be that Microsoft is thinking like you, which is why they're floating the idea of the XBox Live box.

-----------

Back around to x86 on a phone. I'm not sure I've developed an opinion, yet. On one hand, having a more similar architecture from phone to tablet to notebook to workstation to server makes sense. You bring power savings from the bottom up, and performance from the top down. A nice neat little spectrum. Compilers could also be more optimized if dealing with fewer targets. But, really... I don't know if any of that matters much. So long as there is at least some competition between manufacturers and fabs, I think we'll continue to see performance progress, regardless of architecture choices.
 
Perhaps. Rumors are an AMD Playstation and a PowerPC Xbox. We know the WiiU is a PowerPC.
Ah yes, you're right. The next xbox is rumored to have an AMD GPU which got me confused there. Still, Sony's experience with the CELL processor is likely what's making them consider x86. But I'm not sure what makes them think any AMD chip is a good idea right now, unless they know something we don't. And the xbox is rumored to be getting a 16 core PPC. From what I know of game development, multi-cores usually isn't what you want. So that does surprise me a bit.

I suspect that the lower quality games of phones and tablets are making a dent in gaming sales. I wouldn't be too surprised to see consoles next gen have worse sales than this gen because of it. For example - the iPod increased lower quality music sales and the higher quality CD and Record sales were impacted because of it. It seems people value convenience over quality.
Yes, that is kinda unfortunate. But there's one other factor: cost. Modern games are expensive. BF3 costs $60 brand new. And then you gotta add all that BS download content (extra maps, unlock packs and if you bought it used, you get to buy an online-play unlock code for your online game). Mobile games so far have been dirt cheap. You could buy BF3 OR you could buy around 10 or so other games for your phone. Where do you see better value? Well, for me it's BF3 because I don't play mobile games at all, but I'm sure I'm in the minority. Still, if next gen consoles end up upping the cost of games, which they might, then the next gen game consoles might be the last gen.
 
Tbh given the graphical capabilities of Tegra 3 (faster than a PS3) and the rumoured capabilities of Broadcoms forthcoming dual A15 SOC, I do genuinely wonder if the next gen of console won't be a multi core A9/15 hooked up to a mobility Nvidia/ATI part. As was noted, the cpus these days are plenty fast enough to drive this stuff, where the action is these days is in the GPGPU.

Yes, that is kinda unfortunate. But there's one other factor: cost. Modern games are expensive. BF3 costs $60 brand new. And then you gotta add all that BS download content (extra maps, unlock packs and if you bought it used, you get to buy an online-play unlock code for your online game). Mobile games so far have been dirt cheap. You could buy BF3 OR you could buy around 10 or so other games for your phone. Where do you see better value? Well, for me it's BF3 because I don't play mobile games at all, but I'm sure I'm in the minority. Still, if next gen consoles end up upping the cost of games, which they might, then the next gen game consoles might be the last gen.

GTA3 for my android, which is a pristine conversion from the PS2 cost £3 or so. According to rockstar who produce it, they saw some truly pant wetting profits off of that release, well in excess of what they expected.

Also, there is the possibility of another player in the console market: Steam, who have now produced a client for Linux.
 
Back
Top