"The long awaited Tegra 2 will start shipping in smartphones and tablets in early 2011"
Tegra 2 is already in tablets See page 6 of your Holiday Smartphone Buyer's Guide (written by Vivek Gowri and Brian Klug) http://www.anandtech.com/show/4047/holiday-smartph... They may be shitty tablets that don't have android marketplace and are running Android 2.2, but Tegra 2 is currently out.
Good that the Tegra2 phones are coming in a few months. Tegra2 tablets are shipping already though - Viewsonic GTab and Advent Vega. The Motorola Olympus + Honeycomb looks awesome too.
Exciting news, but I wonder what the increase in power consumption is when you double the number of cores AND increase clock frequency even with a transition from 65nm to 45nm. If I had to choose between a phone that runs 50% faster than my existing phone (HTC Incredible) or a phone with 50% longer battery life, at this moment in time I would choose the latter.
Isn't one of the benefits behind having faster processors in mobile devices the fact that you can return to low power states faster? OoO should greatly help there, and having an additional core should help on multi-threaded apps greatly when CPU bound.
Given how good companies are getting at idle power consumption, you could see improved battery life with these new processors. Gotta wait to see though.
If the workload is large enough to full load two cores, be that at 1.0 or 1.5GHz, a dual-core will complete said workload in half the time of a single-core and will spent the remaining time idling.
In either case there's a net win in power efficiency.
If it was an identical dual core vs single core, that would not be true. I don't know how this idea got started in the press.
Energy consumption for a given task will remain the same whether you're peaking at 1A for 100ms or 500mA for 200ms.
Dual core is a win because to complete the same task as an identical single core, you don't have to use 2x the power due to the non-linear way frequency scales with voltage (and therefore, power).
You're only looking at the power draw of the SoC however, in reality that's only going to make up part of the actual power draw.
If calculating the workload slower means more time with the display or various radio units active a faster execution time will end up as a net gain in power efficiency.
Agreed. But I would question how that scenario comes about in real-world usage. When a person is using a smartphone, the vast majority of time spent viewing with the display on is when the page has loaded and the processor is for the most part idle. The actual load time for today's 1GHz parts is trivial.
Is the 1s difference between page loads (I'm being generous here) going to be a blip on the radar of reading an article?
I don't think any of the other components with today's 1GHz parts will really be bottlenecked by the processor either. Hell, in most cases, the SoC has dedicated units to handle talking to those external devices.
I agree that back in the days of ARM11 iPhones, upgrading to a Cortex A8 and more memory definitely helped overall battery life. But we're pretty much at a point of diminishing returns for that now.
I'd say that one of the other selling points of dual-core SoCs is enabling usage scenarios that we haven't really thought of yet though.
As for the upgrade path it's probably worth considering that a rather large part of existing Cortex A8-based devices are also manufactured at 65nm so there's the improvement in manufacturing to take into consideration as well.
What work load? Web browsing? Gaming? We met those years ago.
Processors with more transistors require more power to idle. A faster processor's negative impact on battery life is a bigger problem than most anything else a phone would typically be used for.
I rarely use my phone for websurfing, rarely for gaming or anything else. Why pay $500/year to surf the web when I can use a more versatile laptop or desktop virtually everywhere I want to connect?
I do, however, use my phone for CALLING people. When the battery is dead, because of a marketing decision to use the most power hungry processor currently available and I can't make a critical phone call - that bothers me.
" I do, however, use my phone for CALLING people. When the battery is dead, because of a marketing decision to use the most power hungry processor currently available and I can't make a critical phone call - that bothers me. "
If only there were competition in the phone space so that people could trade off what was most important to them...
Seriously, WTF are people worrying about? The chip hasn't even shipped yet and already we have people panicking that this means, in a bizarre reversal of phone trends to date, that EVERY manufacturer next year is going to be selling nothing but devices with very short battery lives.
I don't understand why this is even an issue then, considering that you don't even seem to be in the market for a smartphone.
Both web browsing and gaming stand to benefit from faster SoCs. Or, conversely, faster SoCs would enable more advanced gaming and web browsing technology.
My 5 year old Nokia dumbphone still lasts about two weeks on a single charge but that doesn't mean I consider it the future of mobile devices.
and if I had to choose between 50% longer battery life and 50% smaller size with same performance (in volume), I would gladly choose the latter. This is the reason it is impossible to please everyone.
For me, the holy grail of mobile phones would be the current 1ghz cpu's inside something the size of Nokia 2730, even if we had to live with a smaller screen.
Hopefully the will fix the problem with overheating in smartphones, cause it was reported on one of the tech sites (can't remember which) which stated that they had run test on triple-core processors on a prototype phone and the freaking thing actually caught on fire and melted. I don't know about you, but I wouldn't want to have that burning sensation in my pants. OUCH!! But seriously, all joking aside, if TI and Samsung do get their act together and put this tech in phones hopefully by mid-2010, then we'll really start seeing the competition heat up. It will be just like the PC processor old days. Can't wait!! - BLUEBOY
So when do we get a 4430 or 440 in a Gumstix? I can't wait to make super small video projects that actually can process HD video (pandboard is too big)
That and a 5MP+ HD able camera going directly on the card with provided drivers and SOC would be so awesome.
lol. That's why I stll have my ancient and simple mobile phone. Weighs like half than all these smartphones, much smaller and battery life is still very good after like 4 years especially compared to these smartphones. Ok, and I don't need the functionallity they offer. I mean I have a pc at work and at home and when I'm not I sure don't need to be gaming or surfing too. And if it is unavoidable it can still do email and run opera mini. ;)
So OMAP4430 we'll see well in advance of OMAP4440. Expect OMAP4430 devices right after the new year (early 2011) and OMAP4440 will be the drop-in upgrade from that point (mid 2011).
Yes, absolutely. A Nehalem does a lot more reordering in a lot more situations than f.e. a pentium pro, and that is a huge part of what makes it faster. However, that is not the case here.
> Weren't they already OoO?
Nope. Cortex A8 is superscalar -- or it can execute multiple instructions per clock, as long as they don't depend on each other. However, if the next instruction in the stream depends on some half-completed result, it didn't bypass it and try to execute the next one. Cortex A9 does this -- this doesn't help well-optimized compute-heavy code that much, but is a *huge* win on branchy code that does a lot of loads. You know, pretty much everything that would be sane to run on a phone. :)
Even still, OoO is *not* the big deal here. The memory interface is. I've done some profiling on a Cortex A8, and on it, damn near everything is bandwidth limited, even many things you'd really expect not to be. A single pathetic lpddr interface just wasn't that good of an idea, especially as it was shared with the GPU and CPU. OoO is just icing on the cake -- and it also helps hide memory latencies.
I expect this to be *at least* twice as fast as a similarly clocked A8 in single-threaded games.
It absolutely helps compute-heavy code. Especially since both NEON and ARM core only has access to 16 registers each (only 14 for ARM core, since one is used as the PC and the other the link register).
OoOE is one of the best ways to get around register starvation. On an architecture with more registers (say, IA-64 or PPC), it's less of an issue.
Even so, being able to reorder low latency instructions (VADD, VAND, etc.) around high-latency instruction (VMLA) is very advantageous given limited execution resources.
Great point on the single LPDDR interface. This is one of the advantages of the OMAP4430 / OMAP4440 which has dual interfaces supporting 6.4GB/s that allow sustained HD multimedia performance without stalling the dual-A9s. Others that don't support have degraded performance - I have seen the data that shows a dramatic dropoff without this approach.
"and the SoC supports up to two 12-megapixel camera sensors (potentially for use in high resolution stereoscopic 3D photography)"
So you think the 2 camera sensors would be for 3D photography vs the front and back cameras that are currently in place for snapshot and video conferencing?
SoC manufacturing have generally been very late in adopting new processes, Qualcomm just made it to 45nm for example.
Are we now to expect the big names jumping to 28nm along the same time as, for example, next generation AMD graphics cards are expected?
I'd love to see that but it seems rather far-fetched.
Not that it's likely to matter for me personally as it seems to be mostly Motorola favoring TI chips and their devices very rarely make it to my end of the globe.
Well, to be fair, it's the time to market for devices from OEM's that's the major delay. 45nm OMAP4's have been available for quite some time, for example, but it takes a while for handset makers to design a product around them and then to get it past both the FCC and the carriers.
Compare this with AMD selling a reference-designed board with its latest and greatest chip.
To have the ability to download a 1080p stream from your phone, and then if you wish, put that video stream on the TV wireless. This would be the future of phones as we know it. Who knows, we might see it when Android 5 comes around? The whole idea of doing what you wish on a laptop you could do on your smartphone is really exciting. Just be sure to put a line-out on what Apple has on their smartphones for heaven's sakes! Hearing the tinny speakers on a smartphone while listening to music is really annoying to say the least! - BLUEBOY
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
39 Comments
Back to Article
Kensei - Wednesday, December 8, 2010 - link
Thanks for the update. Speaking of dual-core smartphone chips... any news on the Atom Z600? Looked promising based on your article back in May.Roland00Address - Wednesday, December 8, 2010 - link
"The long awaited Tegra 2 will start shipping in smartphones and tablets in early 2011"Tegra 2 is already in tablets
See page 6 of your Holiday Smartphone Buyer's Guide (written by Vivek Gowri and Brian Klug)
http://www.anandtech.com/show/4047/holiday-smartph...
They may be shitty tablets that don't have android marketplace and are running Android 2.2, but Tegra 2 is currently out.
Raghu - Wednesday, December 8, 2010 - link
Good that the Tegra2 phones are coming in a few months. Tegra2 tablets are shipping already though - Viewsonic GTab and Advent Vega. The Motorola Olympus + Honeycomb looks awesome too.geniekid - Wednesday, December 8, 2010 - link
Exciting news, but I wonder what the increase in power consumption is when you double the number of cores AND increase clock frequency even with a transition from 65nm to 45nm. If I had to choose between a phone that runs 50% faster than my existing phone (HTC Incredible) or a phone with 50% longer battery life, at this moment in time I would choose the latter.IntelUser2000 - Wednesday, December 8, 2010 - link
"when you double the number of cores AND increase clock frequency"Unless you know, its not going to do the latter.
"I’d expect the 4440 to be used in tablets while the 4430 seems more like a smartphone SKU."
mpschan - Wednesday, December 8, 2010 - link
Isn't one of the benefits behind having faster processors in mobile devices the fact that you can return to low power states faster? OoO should greatly help there, and having an additional core should help on multi-threaded apps greatly when CPU bound.Given how good companies are getting at idle power consumption, you could see improved battery life with these new processors. Gotta wait to see though.
strikeback03 - Wednesday, December 8, 2010 - link
Nvidia is claiming that power consumption will indeed drop for that very reason, but of course we'll have to wait and see with actual shipping phones.MySchizoBuddy - Wednesday, December 8, 2010 - link
Nvidia is claiming power drop if the dual processors are running at 500Mhz each not 1GhZ and certainly not 1.5GHz. Check the Nvidia chart again.MySchizoBuddy - Wednesday, December 8, 2010 - link
Sorry it says for the same load, the processors will run at 500Mhz, to see the power drop.http://www.engadget.com/2010/12/08/nvidia-touts-th...
Exodite - Wednesday, December 8, 2010 - link
Irrelevant.If the workload is large enough to full load two cores, be that at 1.0 or 1.5GHz, a dual-core will complete said workload in half the time of a single-core and will spent the remaining time idling.
In either case there's a net win in power efficiency.
Exodite - Wednesday, December 8, 2010 - link
Bah, typos. :(metafor - Wednesday, December 8, 2010 - link
If it was an identical dual core vs single core, that would not be true. I don't know how this idea got started in the press.Energy consumption for a given task will remain the same whether you're peaking at 1A for 100ms or 500mA for 200ms.
Dual core is a win because to complete the same task as an identical single core, you don't have to use 2x the power due to the non-linear way frequency scales with voltage (and therefore, power).
Exodite - Wednesday, December 8, 2010 - link
You're only looking at the power draw of the SoC however, in reality that's only going to make up part of the actual power draw.If calculating the workload slower means more time with the display or various radio units active a faster execution time will end up as a net gain in power efficiency.
metafor - Wednesday, December 8, 2010 - link
Agreed. But I would question how that scenario comes about in real-world usage. When a person is using a smartphone, the vast majority of time spent viewing with the display on is when the page has loaded and the processor is for the most part idle. The actual load time for today's 1GHz parts is trivial.Is the 1s difference between page loads (I'm being generous here) going to be a blip on the radar of reading an article?
I don't think any of the other components with today's 1GHz parts will really be bottlenecked by the processor either. Hell, in most cases, the SoC has dedicated units to handle talking to those external devices.
I agree that back in the days of ARM11 iPhones, upgrading to a Cortex A8 and more memory definitely helped overall battery life. But we're pretty much at a point of diminishing returns for that now.
Exodite - Thursday, December 9, 2010 - link
Perhaps.I'd say that one of the other selling points of dual-core SoCs is enabling usage scenarios that we haven't really thought of yet though.
As for the upgrade path it's probably worth considering that a rather large part of existing Cortex A8-based devices are also manufactured at 65nm so there's the improvement in manufacturing to take into consideration as well.
sleepeeg3 - Wednesday, December 8, 2010 - link
Relevant.What work load? Web browsing? Gaming? We met those years ago.
Processors with more transistors require more power to idle. A faster processor's negative impact on battery life is a bigger problem than most anything else a phone would typically be used for.
I rarely use my phone for websurfing, rarely for gaming or anything else. Why pay $500/year to surf the web when I can use a more versatile laptop or desktop virtually everywhere I want to connect?
I do, however, use my phone for CALLING people. When the battery is dead, because of a marketing decision to use the most power hungry processor currently available and I can't make a critical phone call - that bothers me.
name99 - Thursday, December 9, 2010 - link
"I do, however, use my phone for CALLING people. When the battery is dead, because of a marketing decision to use the most power hungry processor currently available and I can't make a critical phone call - that bothers me.
"
If only there were competition in the phone space so that people could trade off what was most important to them...
Seriously, WTF are people worrying about? The chip hasn't even shipped yet and already we have people panicking that this means, in a bizarre reversal of phone trends to date, that EVERY manufacturer next year is going to be selling nothing but devices with very short battery lives.
Exodite - Thursday, December 9, 2010 - link
I don't understand why this is even an issue then, considering that you don't even seem to be in the market for a smartphone.Both web browsing and gaming stand to benefit from faster SoCs. Or, conversely, faster SoCs would enable more advanced gaming and web browsing technology.
My 5 year old Nokia dumbphone still lasts about two weeks on a single charge but that doesn't mean I consider it the future of mobile devices.
marc1000 - Wednesday, December 8, 2010 - link
and if I had to choose between 50% longer battery life and 50% smaller size with same performance (in volume), I would gladly choose the latter. This is the reason it is impossible to please everyone.For me, the holy grail of mobile phones would be the current 1ghz cpu's inside something the size of Nokia 2730, even if we had to live with a smaller screen.
Exodite - Wednesday, December 8, 2010 - link
Seeing as the display is by far the largest power drain in a modern smartphone you might just be able to get both. :)blueboy_10 - Thursday, December 23, 2010 - link
Hopefully the will fix the problem with overheating in smartphones, cause it was reported on one of the tech sites (can't remember which) which stated that they had run test on triple-core processors on a prototype phone and the freaking thing actually caught on fire and melted. I don't know about you, but I wouldn't want to have that burning sensation in my pants. OUCH!! But seriously, all joking aside, if TI and Samsung do get their act together and put this tech in phones hopefully by mid-2010, then we'll really start seeing the competition heat up. It will be just like the PC processor old days. Can't wait!! - BLUEBOYzoubi - Wednesday, December 8, 2010 - link
So when do we get a 4430 or 440 in a Gumstix?I can't wait to make super small video projects that actually can process HD video (pandboard is too big)
That and a 5MP+ HD able camera going directly on the card with provided drivers and SOC would be so awesome.
wasnt - Wednesday, December 8, 2010 - link
Doesn't even matter... with battery life falling with ever update, these should get about 1.5 hour talk time in a smart phone.Battery tech updates is what I want to see! This coming from an HTC Evo 4g user.
beginner99 - Wednesday, December 8, 2010 - link
lol. That's why I stll have my ancient and simple mobile phone. Weighs like half than all these smartphones, much smaller and battery life is still very good after like 4 years especially compared to these smartphones.Ok, and I don't need the functionallity they offer. I mean I have a pc at work and at home and when I'm not I sure don't need to be gaming or surfing too. And if it is unavoidable it can still do email and run opera mini. ;)
R3MF - Wednesday, December 8, 2010 - link
is it only the 4440 SKU that is due 2H 2011, or the entire Omap4 family?bad news if the latter, because Omap4 was expected to be in production Q1 2011.
in short, can we expect the 4430 any sooner?
Brian Klug - Wednesday, December 8, 2010 - link
So OMAP4430 we'll see well in advance of OMAP4440. Expect OMAP4430 devices right after the new year (early 2011) and OMAP4440 will be the drop-in upgrade from that point (mid 2011).-Brian
o.mar - Wednesday, December 8, 2010 - link
" It also enables some amount of out-of-order execution. "Weren't they already OoO?
Are there different levels of OoO?
Tuna-Fish - Wednesday, December 8, 2010 - link
> Are there different levels of OoO?Yes, absolutely. A Nehalem does a lot more reordering in a lot more situations than f.e. a pentium pro, and that is a huge part of what makes it faster. However, that is not the case here.
> Weren't they already OoO?
Nope. Cortex A8 is superscalar -- or it can execute multiple instructions per clock, as long as they don't depend on each other. However, if the next instruction in the stream depends on some half-completed result, it didn't bypass it and try to execute the next one. Cortex A9 does this -- this doesn't help well-optimized compute-heavy code that much, but is a *huge* win on branchy code that does a lot of loads. You know, pretty much everything that would be sane to run on a phone. :)
Even still, OoO is *not* the big deal here. The memory interface is. I've done some profiling on a Cortex A8, and on it, damn near everything is bandwidth limited, even many things you'd really expect not to be. A single pathetic lpddr interface just wasn't that good of an idea, especially as it was shared with the GPU and CPU. OoO is just icing on the cake -- and it also helps hide memory latencies.
I expect this to be *at least* twice as fast as a similarly clocked A8 in single-threaded games.
Tuna-Fish - Wednesday, December 8, 2010 - link
Apparently I really need to learn to proofread -- s/Cortex A8/OMAP3430/ for the last two paragraphs.metafor - Wednesday, December 8, 2010 - link
It absolutely helps compute-heavy code. Especially since both NEON and ARM core only has access to 16 registers each (only 14 for ARM core, since one is used as the PC and the other the link register).OoOE is one of the best ways to get around register starvation. On an architecture with more registers (say, IA-64 or PPC), it's less of an issue.
Even so, being able to reorder low latency instructions (VADD, VAND, etc.) around high-latency instruction (VMLA) is very advantageous given limited execution resources.
FutureMobile - Thursday, December 9, 2010 - link
Great point on the single LPDDR interface. This is one of the advantages of the OMAP4430 / OMAP4440 which has dual interfaces supporting 6.4GB/s that allow sustained HD multimedia performance without stalling the dual-A9s. Others that don't support have degraded performance - I have seen the data that shows a dramatic dropoff without this approach.vol7ron - Wednesday, December 8, 2010 - link
"and the SoC supports up to two 12-megapixel camera sensors (potentially for use in high resolution stereoscopic 3D photography)"So you think the 2 camera sensors would be for 3D photography vs the front and back cameras that are currently in place for snapshot and video conferencing?
A 3D camera is an interesting idea though...
Exodite - Wednesday, December 8, 2010 - link
Truly?SoC manufacturing have generally been very late in adopting new processes, Qualcomm just made it to 45nm for example.
Are we now to expect the big names jumping to 28nm along the same time as, for example, next generation AMD graphics cards are expected?
I'd love to see that but it seems rather far-fetched.
Not that it's likely to matter for me personally as it seems to be mostly Motorola favoring TI chips and their devices very rarely make it to my end of the globe.
metafor - Wednesday, December 8, 2010 - link
Well, to be fair, it's the time to market for devices from OEM's that's the major delay. 45nm OMAP4's have been available for quite some time, for example, but it takes a while for handset makers to design a product around them and then to get it past both the FCC and the carriers.Compare this with AMD selling a reference-designed board with its latest and greatest chip.
Exodite - Wednesday, December 8, 2010 - link
I really, really tried to make a clever reply that commended your point but the spam filter hates me.sleepeeg3 - Wednesday, December 8, 2010 - link
Time to end the MHz battle - can we finally get on to focusing on battery life?R3MF - Thursday, December 9, 2010 - link
this is what i suspect, given that the blackberry tablet is due end of Q1, i am hoping for the 4430 in the lead Meego device from nokia too.blueboy_10 - Thursday, December 23, 2010 - link
To have the ability to download a 1080p stream from your phone, and then if you wish, put that video stream on the TV wireless. This would be the future of phones as we know it. Who knows, we might see it when Android 5 comes around? The whole idea of doing what you wish on a laptop you could do on your smartphone is really exciting. Just be sure to put a line-out on what Apple has on their smartphones for heaven's sakes! Hearing the tinny speakers on a smartphone while listening to music is really annoying to say the least! - BLUEBOY_CommManagerJ - Thursday, May 4, 2017 - link
Testing 1 2 3 Testing 1 2 3