Intel's Silverthorne (Ultra-Mobile Processor) Previewed

[MacRumors]

http://www.macrumors.com/images/macrumorsthreadlogo.gif (http://www.macrumors.com)

At this year's International Solid State Circuits Conference, Intel is giving additional details regarding its upcoming Silverthorne ultra-mobile processor (via News.com (http://www.news.com/8301-10784_3-9859616-7.html) and AppleInsider (http://www.appleinsider.com/articles/08/02/04/intel_dishes_new_details_on_apple_bound_silverthorne_chip.html)).

According to the reports, Silverthorne will include dual in-order pipelines in order to keep power consumption at a minimal .5-2 Watts. Most modern chips use out of order pipelines for performance benefits, but at the cost of power. Nevertheless, AppleInsider says that the chip is expected to be as powerful as early Pentium M processors. Additionally, the processor will use the Core 2 Duo instruction set and will support hyperthreading.

Apple has been rumored (http://www.macrumors.com/2007/12/21/apple-to-adopt-intel-ultra-mobile-platform/) to be adopting Silverthorne for their next generation iPhone. Silverthorne is expected to ship in the first half of 2008.

Article Link (http://www.macrumors.com/2008/02/04/intels-silverthorne-ultra-mobile-processor-previewed/)

[Eidorian]

Hyperthreading is back on ~Core 2! Why can't my other processors have it?

[calabi]

great news! 3g in the iphone now as well please!

[Muzzway]

I wonder what would be able to be accomplished with a more powerful iPhone.

[FJ218700]

So the question is, will adopting the Silverthorne processor offset the increased power consumption by 3G, allowing the present battery life to persist or even increase for the 2nd gen iPhone?

[brooker]

Quick, someone start a "Waiting for Silverthorne" Thread!

[QCassidy352]

wow, a processor equivalent to an early pentium M in a phone... pretty cool stuff.

[yoman]

very promising. I wonder how much faster would it really be in the real world.

[GFLPraxis]

Hmm...faster processors might be able to decode HD?

Does this mean HD iPhone next Tuesday? :D

[Sandfleaz]

I'm drooling already, hmmm the next generation iPhone!
.

[2A Batterie]

I wonder what would be able to be accomplished with a more powerful iPhone.
For one, it would get me to whip out the Visa. Does anyone have an idea of when an updated iPhone will be released? Is this announcement an indicator?

[Eidorian]

For one, it would get me to whip out the Visa. Does anyone have an idea of when an updated iPhone will be released? Is this announcement an indicator?FCC documentation would be an indicator. I don't expect to see HD on the iPhone even with Silverthorne.

Does anyone else remember the hyperthreading rumors for Penryn? Anyone?

[lazyrighteye]

I wonder what would be able to be accomplished with a more powerful iPhone.

Basically, a more robust experience/functionality.
Think more laptop less smart phone.

Why can't I shake this hunch that the iPhone is headed more towards a small, tablet-ish size/device than maintaining current 'smart phone' dimensions?

[happydude]

iphone 2 on it's way . . . sometime in June at WWDC?

now all it needs is 3G, ichat, GPS . . . watching the iphone evolve is like watching paint dry while on ecstasy - still is excruciatingly drawn out but so fun and exciting all at the same time!!!

[PlaceofDis]

thats pretty awesome. intel seem to enjoy working with apple and i wonder if apple put any ideas into their heads with this one.

[Amdahl]

That is somewhat interesting that the chip has hyperthreading, but not out-of-order execution. They are essentially the same thing, so I wonder how much Intel is really saving by dropping OOE, but still having hyper-threading.

[liberty4all]

So the first 1st iPhone is really an Albatross, as no other iPhones will be on the same mobile CPU?! Great, thanks Apple for another dead-end...

First 68k
Second PPC
Third ARM
Fourth TBD (x86/x64?! But I have to wonder if they will migrate to Itanium?!)

[sterlingindigo]

3G, Flash, 16GB, iPhone!

[products7074740]

I was going to buy an ipod touch this week. Is it a bad time to buy? I would hate to buy it and then just a few months later see it get a big update. I don't like to wait and don't usually play that game, but if it is very imminent I would consider it. any opinions?

[QCassidy352]

Why can't I shake this hunch that the iPhone is headed more towards a small, tablet-ish size/device than maintaining current 'smart phone' dimensions?

If by that you mean that the form factor will gravitate towards a larger, more tablet-like shape, I strongly disagree. The appeal of the iphone is that it's your phone. You can carry it as easily and automatically as you would your phone. Make it bigger and more tablet-like and I for one would likely leave it behind frequently.

[sparkyms]

Quick, someone start a "Waiting for Silverthorne" Thread!

Or ask if it'll make safari snappier?

[sterlingindigo]

Basically, a more robust experience/functionality.
Think more laptop less smart phone.

Why can't I shake this hunch that the iPhone is headed more towards a small, tablet-ish size/device than maintaining current 'smart phone' dimensions?

9" Macbook Air light

[yoman]

Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU like Mac OS X; en) AppleWebKit/420.1 (KHTML, like Gecko) Version/3.0 Mobile/4A93 Safari/419.3)

will it be clocked down like the current iphone CPU to save battery life

[Eraserhead]

So the first 1st iPhone is really an Albatross, as no other iPhones will be on the same mobile CPU?! Great, thanks Apple for another dead-end...

I don't think it will matter really. Apple will just create Universal binaries again ;).

With regards to 68k, the Intel chips had far inferior performance in the early 80's, Intel is still big today because it was included in the IBM PC, who were the big player in the computer industry at the time.

PPC may have been a mistake when Apple switched, but PPC was a better chip at the time. Only against the G4 was Intel consistently better than PPC before the Intel switch, and that wasn't really true with the laptops until the Pentium M became good.

[tirerim]

So the first 1st iPhone is really an Albatross, as no other iPhones will be on the same mobile CPU?! Great, thanks Apple for another dead-end...

First 68k
Second PPC
Third ARM
Fourth TBD (x86/x64?! But I have to wonder if they will migrate to Itanium?!)

Apple is already using x86 on all Macs; I'm sure they would love to have all their products (or at least all that run some version of OS X) on the same platform. Doesn't mean that first generation phones will be "albatrosses", though, any more than my G5 is. I can sort of see why you might care what architecture your computer uses, but honestly, why would you care at all for your phone, as long as it works? It's not like there's vast quantities of software for the iPhone that would have to run in emulation if they switched.

[gnasher729]

That is somewhat interesting that the chip has hyperthreading, but not out-of-order execution. They are essentially the same thing, so I wonder how much Intel is really saving by dropping OOE, but still having hyper-threading.

Hyperthreading and out-of-order execution are completely different things.

Out-of-order execution is really difficult to implement: When the processor uns into instructions that cannot be executed yet because they wait for results that are not there yet, it has to examine instructions that should be executed further down the line, and not only find whether they are ready, but also prove that the instructions in between don't affect the result. The difficulty is rearranging the instructions.

Hyperthreading is much easier: It duplicates a small part of the processor (the part that reads instructions) to simulate two processors. It checks whether processor A has any instructions ready for execution. If it doesn't, it checks if processor B has any instructions ready. Instructions from two processors never interfere, so mixing them causes no problems. Much simpler than out-of-order execution.

[MBX]

3G, Flash, 16GB, iPhone!

Indeed.

I wouldn't mind 32gb (&+) though

[Abstract]

I wonder what would be able to be accomplished with a more powerful iPhone.

You can check your calendar.....but faster than ever. Phone calls will be faster. People will talk faster.

[tirerim]

I was going to buy an ipod touch this week. Is it a bad time to buy? I would hate to buy it and then just a few months later see it get a big update. I don't like to wait and don't usually play that game, but if it is very imminent I would consider it. any opinions?

There won't be a major update in the very near future, certainly; Silverthorne hasn't even been released yet. In "a few months", though, who knows. It's pretty much impossible to predict these things that far in advance short of Apple actually making an announcement, especially for products that are only in their first generation. There will probably be some advance warning of an iPhone update, given the FCC requirements, which would likely go along with any Touch update, unless somehow something to do with 3G holds up the iPhone.

[johnnyjibbs]

I wonder what would be able to be accomplished with a more powerful iPhone.
More power for less energy is always better. Imagine smoother animations but with extra battery life.

Once the iPhone has 3G and bigger capacity, I'll be much more ready to get one. :D

[numbsafari]

I wonder what would be able to be accomplished with a more powerful iPhone.

VoIP... on the hacked versions at least... and possibly real support for Flash...

[tirerim]

Indeed.

I wouldn't mind 32gb (&+) though

Yeah, 32 GB is the point at which I would definitely get an iPhone. My guess is 2009 for that, though. Hopefully my 3rd gen. iPod will last until then...

[numbsafari]

I wonder what would be able to be accomplished with a more powerful iPhone.

Oh yeah... and a real, multi-threaded Safari...

[noodle654]

Pretty sweet stuff. .5-2watts is nothing. I love how technology gets better every few weeks.

[gnasher729]

So the first 1st iPhone is really an Albatross, as no other iPhones will be on the same mobile CPU?! Great, thanks Apple for another dead-end...

1. It will be quite a while until this Intel chip will be ready. I don't think anyone would suggest Apple should have delayed the iPhone for 18 months or longer to wait for this chip. 2. At this moment in time, Apple doesn't care very much what kind of a processor is in any bit of equipment. It is just a compiler switch. If the MacBook Air had a quad core ARM processor, it wouldn't take any time to get any universal software to run on it. 3. This chip is reasonably powerful and low power. Guess what an ARM processor is. I have run tests ARM vs. Pentium 4 about two years ago, and an ARM processor was about four times faster per clock. And ARM hasn't stood still.

[EagerDragon]

thats pretty awesome. intel seem to enjoy working with apple and i wonder if apple put any ideas into their heads with this one.

Intel told us previously that Apple has been talking to them about how to look at Intel offerings differently. I think the Intel CEO said something similar to that.

When will this baby be available?

[kwood]

Or ask if it'll make safari snappier?

It does make Safari snappier!!! I just downloaded Silverthorne now and boy does it make webpages fly.;)

Maybe I have the wrong thread...:confused:

[numbsafari]

Basically, a more robust experience/functionality.
Think more laptop less smart phone.


Frankly... I hope they don't go that route... I mean... I love my iPhone, but trying to do anything that involves writing even short messages is a real PITA.

I hope they continue to keep it smart and simple.


Why can't I shake this hunch that the iPhone is headed more towards a small, tablet-ish size/device than maintaining current 'smart phone' dimensions?

I hope they keep the iPhone where it's at... it's a good size for a phone. I don't want to hold a tablet PC up to my ear when making calls.

[Amdahl]

Hyperthreading and out-of-order execution are completely different things.
Oh really?

Instructions from two processors never interfere, so mixing them causes no problems.

If the instructions never interfere, then I guess they are running on separate hardware, RAM, OS, and data?

[numbsafari]

Oh really?



If the instructions never interfere, then I guess they are running on separate hardware, RAM, OS, and data?

Rather than getting into a pissing match over something you clearly know nothing about, I recommend looking at Wikipedia or Ars Technica.

Hyperthreading is more less moving an OS-level concept (multi-programming) into the CPU microcode itself. This is an idea that Intel has been promoting for some time. Hyperthreading is a limited form of this idea.

Out-of-Order Execution is a completely CPU-level concept. If the CPU can determine that certain instructions are not related then it can process them out of order in such a manner that improves overall instruction throughput within the processor. This can be applied to the linear commands within a single process. Hyperthreading is merely taking independent processes and running them concurrently. OOE is MUCH more complicated.

[wildcardd]

How many watts does the current iPhone Arm processor draw?

[mrxak]

Every time I hear about a new chip that uses less power, I think to myself, are they going to use this to make things smaller, or are they going to use it to make things have a larger battery life? They aren't mutually exclusive of course, but there is a tradeoff involved. Personally I think laptops and the iTouch/iPhone are thin enough. What they are all lacking across the board is exceptional battery life. Let the size stay pretty much the same, and let the extra space be used for a larger battery. If you want to go back to making things smaller after my laptop has 10 hours of juice and my iPod has 20, feel free.

[numbsafari]

Every time I hear about a new chip that uses less power, I think to myself, are they going to use this to make things smaller, or are they going to use it to make things have a larger battery life? They aren't mutually exclusive of course, but there is a tradeoff involved. Personally I think laptops and the iTouch/iPhone are thin enough. What they are all lacking across the board is exceptional battery life. Let the size stay pretty much the same, and let the extra space be used for a larger battery. If you want to go back to making things smaller after my laptop has 10 hours of juice and my iPod has 20, feel free.

Agreed.

[mrxak]

The MacBook Air is just another example of sacrificing power or battery life for thinness. Don't get be wrong, the MacBook Air is very sexy, and Apple is all about the sexy. It's a big part of the reason why I like their products. I just wish that battery life was more sexy so they'd pay more attention to it.

[Jackal-Head]

Hyperthreading is back on ~Core 2! Why can't my other processors have it?

How effective hyperthreading is depends on factors like the pipeline length and whether or not out-of-order execution is used. Intel designed the later Pentiums with one goal in mind: Higher clock speeds (at virtually any cost). The only way to achieve this was to use a very high pipeline length. The downside were more frequent and longer stalls than would occur on a processor with a shorter pipeline. Consequently, the utilization of the CPU's integer and floating-point units was sub-optimal. This means when Intel introduced hyper-threading to the Pentium 4, there was a need for a way to increase processor throughput by making more effective use of its integer and floating-point units.

AMD, on the other hand, wisely chose not to compete in this rat race. High clock speeds may be good for marketing to the technically unaware, but a better measure for a processor's speed is throughput: Clocks per second multiplied by the amount of work that, on average, can be completed per clock. Thus, in their Athlon series, AMD preferred to improve the CPU design to combine moderate clock speed increases with great increases on the work per clock side. This allowed AMD to use a shorter instruction pipeline, meaning less, and less severe stalls, and consequently better utilization of the processor's work units even when additional units were added. Also, the longer clock cycles allowed AMD processors to perform more complex calculations per clock. The net result was that Athlons would, in real life applications, often outperform Pentiums that operated at significantly higher clock speeds, attracting all sorts of tech savvy customers to AMD. It is obvious from the Athlon's short pipeline design that hyper-threading would have been much less effective on Athlons than on Pentiums, so it is unsurprising that AMD did not choose to introduce it on their CPUs.

Enter Paul Otellini whom I highly respect. He may not have been the one who started the "Core revolution", but he definitely supported it. But let me explain what I mean by this term. By 2005, it was becoming more and more obvious that Intel would have massive difficulties further increasing their processors' clock speeds. It was also obvious that even in the mid term, many of the problems were insurmountable. Plus, the power consumption of the Pentiums reached absurd levels even before Intel introduced the shame de la shame of the entire line - the Pentium 4 "Prescott". The NetBurst architecture had reached a dead end. Intel finally did the right thing and redesigned its processors to use shorter pipelines and perform more work per clock - just like AMD had done for years. Intel even did this at the expense of clock speed. In doing so, Intel managed to overcome the efficiency problem while at the same time improving processor throughput. At the same time, hyper-threading became largely obsolete - while allowing more than one thread to run simultaneously was still desirable, this would require increasing the number of work units as well rather than just allowing two threads to share the same set of work units like in hyper-threading. Thus, the Core Duo was born, and the rest is history.

In a sense, Intel finally realized it was on the wrong track and has now chosen to abandon NetBurst and out-AMD AMD instead. They had a lot of success doing this. This should also explain why now there are Intel CPUs with 4 cores but no hyper-threading.

But why would Intel then want to reintroduce HT in Silverthorne? Simple - while Silverthorne as a Core CPU could effectively saturate its work units by out-of-order execution alone, Intel chose to remove OOO execution in order to conserve power. But this means Silverthorne might suffer from the old work unit utilization problem, and hyper-threading becomes interesting again, at least for this particular processor line.

As an interesting foot-note to a long post, I want to add that hyper-threading can, in extreme cases, actually hurt processor throughput. This can happen when both threads taken together require more cache memory than is available (while there would be enough cache for just one thread). In this case, they will cannibalize on each other's cache, and the number of RAM accesses will skyrocket, causing frequent stalls and thus slowing down the processor's operation.

[twoodcc]

I wonder what would be able to be accomplished with a more powerful iPhone.

im wondering that also. it seems the current iPhone is pretty powerful.

looking forward to it either way

[Eidorian]

How effective hyperthreading is depends on factors like the pipeline length and whether or not out-of-order execution is used. Intel designed the later Pentiums with one goal in mind: Higher clock speeds (at virtually any cost). The only way to achieve this was to use a very high pipeline length. The downside were more frequent and longer stalls than would occur on a processor with a shorter pipeline. Consequently, the utilization of the CPU's integer and floating-point units was sub-optimal. This means when Intel introduced hyper-threading to the Pentium 4, there was a need for a way to increase processor throughput by making more effective use of its integer and floating-point units....

...But why would Intel then want to reintroduce HT in Silverthorne? Simple - while Silverthorne as a Core CPU could effectively saturate its work units by out-of-order execution alone, Intel chose to remove OOO execution in order to conserve power. But this means Silverthorne might suffer from the old work unit utilization problem, and hyper-threading becomes interesting again, at least for this particular processor line.

As an interesting foot-note to a long post, I want to add that hyper-threading can, in extreme cases, actually hurt processor throughput. This can happen when both threads taken together require more cache memory than is available (while there would be enough cache for just one thread). In this case, they will cannibalize on each other's cache, and the number of RAM accesses will skyrocket, causing frequent stalls and thus slowing down the processor's operation.I'm sorry for cutting your very informative post but I lived through it. ;)

I'm sure other users would benefit from the history lesson. I'll just have to wait for Nehalem to see what Intel has in store for hyperthreading on x86 there.

[SthrnCmfrtr]

....

Quite possibly the most informative and comprehensible post I've ever read on the Intel-AMD wars. Thank you.

[dicklacara]

Yeah, 32 GB is the point at which I would definitely get an iPhone. My guess is 2009 for that, though. Hopefully my 3rd gen. iPod will last until then...

Hmmm.... 32GB SDHC cards are just becoming available for videocams. hey are about the size of a postage stamp in a approx 1/16 inch package.

These are targeted to HD Video capture.

I wonder if this technology is fast enough, etc. to be used in an iPhone and tablet.

Thoughts?

[Amdahl]

How effective hyperthreading is depends on factors like the pipeline length and whether or not out-of-order execution is used.
Hey gnasher & numbsafari! Better let Jackal-Head here know that he doesn't know what he's talking about!


But why would Intel then want to reintroduce HT in Silverthorne? Simple - while Silverthorne as a Core CPU could effectively saturate its work units by out-of-order execution alone, Intel chose to remove OOO execution in order to conserve power. But this means Silverthorne might suffer from the old work unit utilization problem, and hyper-threading becomes interesting again, at least for this particular processor line.
Hmm.. very interesting. So there is some kind of relationship between OOE and Hyperthreading? Guess I'd better hit the Wikipedia so I can stop pissing on myself!

As an interesting foot-note to a long post
In all seriousness, actually an excellent post. The decision to keep hyperthreading and drop OOE, must be based on some very app-specific factors, or the embedded space in particular, where hand optimization can be assumed.

[Amdahl]

I'm sure other users would benefit from the history lesson. I'll just have to wait for Nehalem to see what Intel has in store for hyperthreading on x86 there.

My guess is nothing on the desktop. Hyperthreading & OOE should both be going away (as core counts rise). That's why it is curious that this chip only got rid of one, when they both use the same hardware on the chip.

[Eidorian]

Hey gnasher & numbsafari! Better let Jackal-Head here know that he doesn't know what he's talking about!

Hmm.. very interesting. So there is some kind of relationship between OOE and Hyperthreading? Guess I'd better hit the Wikipedia so I can stop pissing on myself! From my skimming they both want to avoid letting stalls prevent the usage of execution resources.

In all seriousness, actually an excellent post. The decision to keep hyperthreading and drop OOE, must be based on some very app-specific factors, or the embedded space in particular, where hand optimization can be assumed.I'm going to brush up on my OOOE and hyperthreading knowledge as well. From my skimming they both want to prevent stalls from slowing down the usage of execution components to complete a task.

My guess is nothing on the desktop. Hyperthreading & OOE should both be going away (as core counts rise). That's why it is curious that this chip only got rid of one, when they both use the same hardware on the chip.http://www.google.com/search?q=hyperthread%20nehalem

It should trickle down from the Xeons just like on Core 2 and Netburst.

[Hattig]

Apple has been rumored (http://www.macrumors.com/2007/12/21/apple-to-adopt-intel-ultra-mobile-platform/) to be adopting Silverthorne for their next generation iPhone. Silverthorne is expected to ship in the first half of 2008.

Article Link (http://www.macrumors.com/2008/02/04/intels-silverthorne-ultra-mobile-processor-previewed/)

Yeah, rumoured by people who haven't got a single clue in the universe about what Silverthorne is competing with in the ARM camp. I've written it all before - too large a chipset, too power hungry still, not enough integration, ARM is well understood by Apple, ARM is good enough, etc.

An Apple tablet? Yes, maybe Silverthorne will be good there. Or a real Apple subnotebook. Or the next Apple TV hardware revision. Not the iPhone. Not the iPod Touch.

[Hattig]

That is somewhat interesting that the chip has hyperthreading, but not out-of-order execution. They are essentially the same thing, so I wonder how much Intel is really saving by dropping OOE, but still having hyper-threading.

SMT (the real term for Intel's HyperThreading) works really well on in-order CPUs - lots of bubbles in the pipelines to fill.

See Sun's Niagara CPU for a rather extreme implementation. Niagara 2 will have 8 in-order cores with 8-way SMT and can do 64 threads simultaneously and achieve very good figures. Not so hot in single-threaded tasks though.

[Hattig]

How many watts does the current iPhone Arm processor draw?

0.05W to 0.5W including cache, integrated 3D graphics (Imagination I expect, so that's a Dreamcast in your iPhone - it's PowerVR essentially but modernised - this is very simplified).

All this is also in a package that's a centimetre or two on a side - indeed I believe the iPhone CPU also has its 128MB of RAM in the package.

Primary power draws in the iPhone are probably the display and the wireless.

I think the iPhone has 3 ARM CPUs inside - the main processor (ARM11), the wireless (ARM9), and the baseband.

[Jackal-Head]

Rather than getting into a pissing match over something you clearly know nothing about, I recommend looking at Wikipedia or Ars Technica.

Hyperthreading is more less moving an OS-level concept (multi-programming) into the CPU microcode itself. This is an idea that Intel has been promoting for some time. Hyperthreading is a limited form of this idea.

Actually, Amdahl has a valid point, and the tone of your response is inadequate. Just as data dependencies exist between different instructions in the same thread, so do they exist between separate threads. It is the fact that these data dependencies are much easier to resolve that makes hyper-threading easier to implement (and consequently require less die space as well): As each thread has its own register set, data dependencies in between threads necessarily involve memory addresses. So one easy solution is to make both thread pipelines perform all memory accesses through the same, synchronized cache. Actually, the cache does not even need to be fully synchronized, i.e. a read-modify-write operation does not necessarily have to lock the cache line until it completes. It is sufficient to synchronize the cache at so-called memory barriers (Intel usually calls them memory fences) but that is another and potentially long story so I'll leave it at that.

By the way, contrary to what your usage of the word "process" suggests, hyper-threading can be used to execute two threads of the same process as well. Processes are an OS-level concept anyway - the CPU does not care which threads (or "tasks" in conventional Intel parlance) belong to which process.

[Masquerade]

out of order execution was invented to prevent idle states on processor when cache misses occurr (the processor asks data to the memory controller and this is a longer path than their neighbourhood caches).
in OOE the instructions are stacked when waiting for that, and thus, new instructions with no direct implication can be processed through the pipeline.

theory on OOE drop:
well, in the new range of mobile devices the processor is very close to the memory chips, the memory is faster and the tipical behavour of these operating systems don't require too many context switches along time (saving and restoring registers when switching processes)

[gamac]

looks like (almost) everyone agrees:

OOE != HT

[MacFly123]

So the question is, will adopting the Silverthorne processor offset the increased power consumption by 3G, allowing the present battery life to persist or even increase for the 2nd gen iPhone?

Somebody help me here, but didn't I hear something about Intel offering options of including WIFI, Graphics, & 3G built into these chips?

[MacFly123]

Basically, a more robust experience/functionality.
Think more laptop less smart phone.

Why can't I shake this hunch that the iPhone is headed more towards a small, tablet-ish size/device than maintaining current 'smart phone' dimensions?

I can see them adding an additional product like that to serve more as the Next Gen Newton, but I don't think people want their phones any bigger. I would like them to make the iPhone thinner and smaller, as long as the screen stays the same size.

[MacFly123]

I was going to buy an ipod touch this week. Is it a bad time to buy? I would hate to buy it and then just a few months later see it get a big update. I don't like to wait and don't usually play that game, but if it is very imminent I would consider it. any opinions?

You're fine. I don't think it will see any hardware changes until the Holidays. The iPod Touch and iPhone are meant to last longer and be updated with software mostly until there are obvious hardware benefits coming.

[Rocketman]

An Apple designed Intel Silverthorne product is for me.

Rocketman

[BoyBach]

You can check your calendar.....but faster than ever. Phone calls will be faster. People will talk faster.


And all at the same time! ;)

[bigwig]

Hyperthreading is back on ~Core 2! Why can't my other processors have it?
Hyperthreading was mostly a dud performance-wise. Why do you want it?

[ClassicMac247]

So should we be expecting a new iPhone as early as April 08 or will we have to wait for the silverthorne processor?

[AidenShaw]

Hyperthreading was mostly a dud performance-wise. Why do you want it?

Actually, HyperThreading was always a win on a single CPU (that is, a single core system). It would sometimes be quite a bit faster, more often a little bit faster, but never slower.

Hyperthreading on a dual-socket system, however - not so much a good thing. If you always had lots of threads computable, it would help (server-land).

If you had two computable threads on a dual-socket single-core system (two sockets, two HT chips, four logical CPUs), then HT could hurt if the OS scheduled the two threads on the same socket.

When that happened, it would be slower than if they had been scheduled one on each socket.

[chicagostars]

So should we be expecting a new iPhone as early as April 08 or will we have to wait for the silverthorne processor?

I wouldn't bet on seeing a new iPhone by April. We'll likely see more incremental software updates to go along with the first wave of SDK produced goodies. Also thinking of the s***storm that was kicked up when the price drop occurred, I'd hate to hear the noise if a new version of the iPhone was released before the early adopters were even one year into their contracts without the offer of a VERY attractive upgrade path. They won't hang back and let it get too stale, but I think Apple is going to tread somewhat lightly as far as iPhone revisions go . . . perhaps a little too lightly for people waiting on an updated iPhone. :) I'm guessing we'll more likely see something in the second half of the year. Again, I'm only guessing.

[MattInOz]

Yeah, rumoured by people who haven't got a single clue in the universe about what Silverthorne is competing with in the ARM camp. I've written it all before - too large a chipset, too power hungry still, not enough integration, ARM is well understood by Apple, ARM is good enough, etc.

An Apple tablet? Yes, maybe Silverthorne will be good there. Or a real Apple subnotebook. Or the next Apple TV hardware revision. Not the iPhone. Not the iPod Touch.

0.05W to 0.5W including cache, integrated 3D graphics (Imagination I expect, so that's a Dreamcast in your iPhone - it's PowerVR essentially but modernised - this is very simplified).

All this is also in a package that's a centimetre or two on a side - indeed I believe the iPhone CPU also has its 128MB of RAM in the package.

Primary power draws in the iPhone are probably the display and the wireless.

I think the iPhone has 3 ARM CPUs inside - the main processor (ARM11), the wireless (ARM9), and the baseband.

So in other words Apple would shoe horn in a fourth ARM if the power was needed before they would pop a sliverthorn in to the iPhone.

I thought the silverthorn was to replace the chip intel currently offer for UMPC or MIB applications and most of those are what 7inch and what 4 times the volume of the iPhone.

I think your right if apple use this it will be in something between and iPhone and a MacBook Air. Then again maybe it's not the processor that is the shortfall. What is the best resolution 7inch screen around?
You could see them waiting for till they can get 180, 200dpi.

[Krevnik]

That is somewhat interesting that the chip has hyperthreading, but not out-of-order execution. They are essentially the same thing, so I wonder how much Intel is really saving by dropping OOE, but still having hyper-threading.

They achieve the same result, but are very different in implementation and cost.

OOE is expensive in circuitry, as the processor has to decide what can and can't be out of order. This isn't exactly easy to build in fixed logic design. Easy in software, hard in EE.

Hyperthreading, on the other hand, is easy in implementation. The downside is that you now have to track two sets of registers on the die, but the upside is that by doing this, your insertion logic is very simple.

For both the PS3 and 360 CPUs, they had hyperthreading (under a different name), and lacked OOE... but suddenly freed up quite a bit of space on the die for specialized cores, more cores, etc.

[Krevnik]

Actually, HyperThreading was always a win on a single CPU (that is, a single core system). It would sometimes be quite a bit faster, more often a little bit faster, but never slower.

It would never be slower /overall/, but with Hyperthreading turned on, and how it balanced between processes... it did some noticeable perf damage. Single-threaded apps like games took a performance hit because there was other code balanced onto the same core taking up cycles it would normally use.

Gamers tended to turn off Hyperthreading or get chips without it to prevent background processes from siphoning off performance from the single process.

[Michael CM1]

I was going to buy an ipod touch this week. Is it a bad time to buy? I would hate to buy it and then just a few months later see it get a big update. I don't like to wait and don't usually play that game, but if it is very imminent I would consider it. any opinions?

Unless you desperately need one (read: don't have an iPod), I would wait until a 32GB model comes out. With the prices that USB flash drives are at, I'm really surprised that the iPhone and iPod Touch don't have more memory already. Don't they all use the same type of memory?

I will also tell you that cnet said the Touch's display is nowhere near the uber-high quality that the iPhone's is. Might be better to wait for the 2.0 version. First-gen products usually have plenty of bugs to work out. It took Apple until either the 4th or 5th generation to put the click-wheel on iPods.

[bretm]

Unless you desperately need one (read: don't have an iPod), I would wait until a 32GB model comes out. With the prices that USB flash drives are at, I'm really surprised that the iPhone and iPod Touch don't have more memory already. Don't they all use the same type of memory?

I will also tell you that cnet said the Touch's display is nowhere near the uber-high quality that the iPhone's is. Might be better to wait for the 2.0 version. First-gen products usually have plenty of bugs to work out. It took Apple until either the 4th or 5th generation to put the click-wheel on iPods.

I still have the 3rd gen iPod, which is the first iPod touch really. It has no moving parts (externally). Not even a click. Just a touch. I love it. Not sure why they went back to the wheel.

[Evangelion]

So the question is, will adopting the Silverthorne processor offset the increased power consumption by 3G, allowing the present battery life to persist or even increase for the 2nd gen iPhone?

um, I think that Silverthorne consumes MORE power than the current iPhone-CPU does....

[Evangelion]

Hyperthreading was mostly a dud performance-wise. Why do you want it?

Not really. Like others have said, it never really lowered performance (I think I saw benchmarks that showed 2-3% reduction at most, but that might be statistical noise). At best, HT improved performance by 10-20%. And in many ways, HT makes more sense today, since multi-core CPU's are the norm, so more software is being written that is multithreaded, so the benefits might be more apparent. Now, the downside of Hyperthreading was that it increased power-consumptiosomewhat, but Intel might have that under control these days.

And, we must not forget the fact that hi-end processors routinely use Hyperthreading (well, that's Intel's brand-name for it, but still). Suns Niagara uses it, Itanium uses it etc.

[Evangelion]

So the first 1st iPhone is really an Albatross, as no other iPhones will be on the same mobile CPU?! Great, thanks Apple for another dead-end...

First 68k
Second PPC
Third ARM
Fourth TBD (x86/x64?! But I have to wonder if they will migrate to Itanium?!)

I'm sorry, but your argument makes no sense. Yes, Apple has changed architectures in the past. So? How exactly is 68k a "dead-end"? How long did Apple use it? How is PPC a "dead-end"? They used it for a LONG time.

Yes, things change. But that does not mean that they things they used before is suddenly a "dead-end". Hell, Intel has for all intents and purposes moved from x86 to x86-64, does that mean that x86 was a "dead-end"? And even if current iPhone/iPod touch was the only device Apple ships that uses Arm, so what? How exactly would that change in CPU affect you?

And like I said: I don't think that Silverthorne is destined for the iPhone. It consumes more power, and the one benefit it has is that it uses x86. But in devices like iPhone, that feature is more or less useless. I'm thinking devices like AppleTV and maybe a device that sits between iPhone and a laptop.

[Evangelion]

Unless you desperately need one (read: don't have an iPod), I would wait until a 32GB model comes out. With the prices that USB flash drives are at, I'm really surprised that the iPhone and iPod Touch don't have more memory already. Don't they all use the same type of memory?

I would guess that the issue with storage is due to the number of chips. There propably isn't enough room for more flash-chips. So even if prices come down, we might not see an increase in capacity until flash-chips with higher capacities become available.

I will also tell you that cnet said the Touch's display is nowhere near the uber-high quality that the iPhone's is. Might be better to wait for the 2.0 version. First-gen products usually have plenty of bugs to work out. It took Apple until either the 4th or 5th generation to put the click-wheel on iPods.

Wasn't that with the very first units that shipped? They had issues with displaying blacks. I can say that my week 40 touch has perfect screen

[k2k koos]

Imagine that, a powerful computer in your pocket, in the shape of iPhone or similar. Puts that Casio pocket calculator of 25 years back to shame doesn't it :-):apple:

[gnasher729]

Actually, HyperThreading was always a win on a single CPU (that is, a single core system). It would sometimes be quite a bit faster, more often a little bit faster, but never slower.

There have been some unfortunate cases where a CPU with HyperThreading did more work, but achieved less. A simplified view is that a single core CPU with hyperthreading can work either as _one_ CPU running at 100% speed, or as _two_ CPUs running each at 60% speed. If you run Handbrake, for example, two CPUs at 60% speed are 20 percent faster in total.

But things went wrong for some applications that had one task that should have used 90 percent of available CPU, and another task that should have used 10 percent of the available CPU. It could happen that each task got one complete CPU for itself, so the first task got one CPU at 60% speed, and the second task got that as well.

And on a multi-core system obviously the OS must know about this, otherwise if you have two threads and run them on the same core instead of separate cores, you lose lots of performance (instead of two threads running at 100 percent, you could have two threads running at 60% speed, and two threads doing nothing if the OS is stupid).

[gnasher729]

Hey gnasher & numbsafari! Better let Jackal-Head here know that he doesn't know what he's talking about!

Jackal-Head wrote: "How effective hyper-threading is depends on [lots of things]". He is absolutely right. I talked about something entirely different: Hyperthreading is not very difficult to implement, out-of-order execution is difficult to implement.

What is quite interesting is that, as Jackal-Head said, hyperthreading is quite effective to fix a messed-up architecture like Netburst (Pentium IV), not very effective for a well-designed architectures like AMDs over the last year and Core / Core 2, and _very_ effective if you leave out out-of-order execution on purpose and use hyperthreading to get the same performance gain at lower cost.

[AidenShaw]

But things went wrong for some applications that had one task that should have used 90 percent of available CPU, and another task that should have used 10 percent of the available CPU.

It could happen that each task got one complete CPU for itself, so the first task got one CPU at 60% speed, and the second task got that as well.

You are right, I'm wrong in that there would be cases where code written explicitly for a single CPU would do worse with dual HT CPUs than with two full CPUs.

I think that most software engineers would agree, however, that depending solely on thread priority to schedule the CPU(s) is an inherently broken design, and inherently single processor.

Even with two cores, using priority as the "tool" could lead to problems. Since the "low priority" thread could get 100% CPU, it could cause I/O or other bandwidth issues (net/memory) that could affect the primary thread.

[winterspan]

nevermind, I've sort of answered by own post in a newer post..



Unless I'm mistaken, I think this article is somewhat misleading... And many people here are perpetuating this idea of a Silverthorne in an iPhone.
Unless I've missed a huge technological breakthrough, even using 45nm and dropping OOO execution is not going to get an x86 CPU like this anywhere close to the power consumption of the Samsung ARM in the iPhone.
Don't the modern ARM chips have much smaller power consumption at Idle vs Silverthorne? Is it possible for the Silverthorne CPU to all but shut down while the phone idles away waiting for a phone call??

Could someone with much more knowledge than I comment on the power consumption at load/idle of the Iphone ARM vs the 45nm Silverthorne???

[bigwig]

The ARM uses a lot less power because it's a lot simpler. The x86 will always be an energy hog no matter how many billions of dollars Intel spends. Note too that MIPS (despite its demise in the mainline-CPU space) is still very popular in embedded applications, precisely because it's low power and has good performance.

[winterspan]

Unless I'm mistaken, I think this article is somewhat misleading... And many people here are perpetuating this idea of a Silverthorne in an iPhone.
Unless I've missed a huge technological breakthrough, even using 45nm and dropping OOO execution is not going to get an x86 CPU like this anywhere close to the power consumption of the Samsung ARM in the iPhone.
Don't the modern ARM chips have much smaller power consumption at Idle vs Silverthorne? Is it possible for the Silverthorne CPU to all but shut down while the phone idles away waiting for a phone call??

Could someone with much more knowledge than I comment on the power consumption at load/idle of the Iphone ARM vs the 45nm Silverthorne??

Ok, yes, I am responding to my own post. lol. But I have taken some time to research this myself on Google to find out more details on Silverthorne vs ARM.
Based on a number of different articles and Intel presentation slides, it looks like the Silverthorne will have a PEAK TDP of between 0.6W -> "sub-2W" depending on model, which are supposed to range between 1Ghz and 2Ghz. The power consumption at Idle is said to be between 0.01W -> 0.1W depending on model, with these figures EXCLUDING chipset. Also Silverthorne is said to have a new low-power state, called C6.

From Arstechnica, "In order to alleviate some of the power difference between its chips and ARM's, Intel has equipped Silverthorne with a new low-power state, called C6. When Silverthorne is in C6, the only components that it leaves turned on are the SRAM that saves the existing processor state and some circuitry that can wake up the processor again when it's needed. (Getting out of C6 takes about 100 microseconds.) Intel claims that their testing indicates that Silverthorne can spend as much as 90 percent of its time in C6; if that's true, then that will bring the chip's average power dissipation far below its stated TDP. So Intel is counting on a combination of sleep-enabled lower average power and support for the full, awesome expanse of the extended x86 instruction set architecture to make Silverthorne a compelling basis on which to build a generation of mobile internet devices."

For the first generation, Silverthorne will initially be combined with the "Menlow" platform which has a seperate chipset and other components. Second generation Silverthorne chips will be integrated with the "Morrestown" platform which is a System-on-a-chip (SoC) design. This means that all the components needed to run the platform will be integrated onto one chip, including the processor, chipset/memory controller, Wifi/Ethernet, USB, etc. Possibly even graphics acceleration and video decoding DSPs will be embedded as well.This should allow the platform to further reduce power consumption.

Silverthorne specs:
- performance equivalent to first generation Pentium M (Banias)
- 45nm
- 47M transistors
- 25mm^2
- 533MT/s FSB
- In-order execution with hyperthreading

Competition
Silverthorne is going to have some very strong competition from the ARM SoC world. The most likely direct competitor looks to be the next gen ARM Cortext series, specifically the Cortext-A9 (out-of-order core) which is based on a 65nm process and will run in a 0.25W TDP. I was unable to find info on the Idle state power consumption of this chip, but it is assumed to be as low as or lower than Silverthorne based upon earlier generation ARM SoCs.

This is a different ARM core than what is currently on the iPhone (iPhone has an ARM11) and one which will be more powerful, reaching speeds of 1Ghz or more. ARM even claims the Cortext-A9 will have the same power consumption as the iPhone "ARM11" chip, but increase performance by a factor of FOUR. Also, there are both single-core and multi-core versions of the Cortext A9.

Based on all that information and what the analysts say, it looks like the 1st generation Silverthorne (Menlow platform) may be relegated to UMPC and Intel's new "Mobile Internet Device" (MID) type devices and not quite ready for the smartphone sphere. But this will probably change with the SoC "Morrestown" platform which will reduce power consumption by integrating all the components into one chip.
Here is how Jon Stokes of ArsTechnica ends his article from today (Tues. 5th)
http://arstechnica.com/news.ars/post/20080205-small-wonder-inside-intels-silverthorne-ultramobile-cpu.html)

" Having tried a few of Intel's Silverthorne-based prototypes, I must say that I wasn't particularly impressed. I own a Nokia N800 and an iPhone, both of which are ARM-based and both of which give a nearly complete Internet experience in a smaller form factor than Silverthorne will ever fit into. Indeed, at one point during a sit-down with Intel the rep told me that the warm, bulky prototype I was holding would give me the "full Internet in your pocket." I started chuckling, pulled out my iPhone, and said, "I already have that." He gamely responded that the iPhone's browser doesn't support Flash (in my opinion that's a feature, not a bug), but my point was made.

So Silverthorne is really a transitional product; it's Intel's first, slightly awkward foray into a market that it intends to eventually dominate by doing what it always does, and that's produce ever smaller, cheaper, and faster chips that support the world's most popular ISA. This recipe may ultimately work for Intel in the embedded market the way that it has worked elsewhere, but that day won't come just yet.

... So Silverthorne is just the start of something, and to ARM, MIPS, and the other established chipmakers who currently own the embedded space, it's Intel's way of saying "game on."


For more info on Silverthorne, see:
http://arstechnica.com/news.ars/post/20080205-small-wonder-inside-intels-silverthorne-ultramobile-cpu.html
http://en.wikipedia.org/wiki/Silverthorne_%28CPU%29
http://blogs.zdnet.com/Ou/?p=987
http://www.news.com/8301-10784_3-9859616-7.html


For more info on the ARM Cortext-A9, see:
http://en.wikipedia.org/wiki/ARM_architecture
http://newlc.com/arm-unveils-cortex-a9-processors-scalable-performance-and-low-power-designs
http://www.arm.com/products/CPUs/ARMCortex-A9_MPCore.html
http://www.linuxdevices.com/news/NS2917028234.html

[sterlingindigo]

You can check your calendar.....but faster than ever. Phone calls will be faster. People will talk faster.

Fast talkers...that's right! I'm a low talker, would it help me? I mean, be louder?

[AidenShaw]

http://www.news.com/Intel-picks-Atom-as-name-for-new-chip-family/2100-1006_3-6232776.html


Intel has picked "Atom" as the new brand name for its latest microprocessor, the world's largest semiconductor company said.

The Intel Atom processor is the name for the new family of low-power processors, the brains of digital devices, that will power mobile Internet devices and ultra low-cost and small notebook and desktop personal computers.

Intel sees a big market for the Internet-connected devices that can fit in one's pocket and for what it is calling the netbook, a low-cost PC costing around $250.

The Intel Atom processor is based on a new microarchitecture designed for small devices and low power consumption, Intel said. The chip is less than 25 square millimeters, and 11 of the chip's dies--the slivers of silicon with 47 million transistors each--would fit in an area the size of a U.S. penny.

The new chips, previously code-named Silverthorne and Diamondville, are made on Intel's 45-nanometer chipmaking technology and slated for introduction toward the middle of this year.

"Diamondville and Silverthorne both represent an attempt by Intel to sell chips profitably for a whole lot less," said Nathan Brookwood, an analyst at market research firm Insight 64. "This is the first new processor design coming out of Intel since the Pentium Pro in 1995."