Will Bobcat be the home run AMD is looking for?

[extra]

http://techreport.com/articles.x/19937

More information on Hudson's power consumption.



So if we compare N550 atom vs Ontario we would have the following power consumption totals:

N550 atom (8.5 watts) + 2.1 watts for NM10 Express chipset= 10.6 watts

Ontario (9 watts) + 2.7 watts for the lower Hudson chipset= 11.7 watts

Total TDP difference= 1.1 watts

Yeah, seems like power consumption falls into the "plenty good enough" category. It won't compete with the new integrated atom stuff designed for phones and whatnot, but it's not designed too. The real key will be pricing imho. If they are cheap (and I think they will be) we'll see these flood low end laptops and netbooks that are in places like Walmart and Bestbuy.

The thing that people seem to be missing is there is one HUGE advantage that will ensure these sell like hotcakes:

They let the OEMs market the systems as having Directx 11.

Yeah, it's definitely not fast enough for DX11 games at any good quality, but that's not the point. OEM's love features, and DX11 is a big one. The average consumer that sees two laptops or netbooks, one with DX10, one with DX11, otherwise similar in price/features, is going to pick the DX11 one every time imho. This is why I believe these will sell by the bucketload to OEMs.

 

[CTho9305]

@CTho9305: Good post - the x86 architecture is complicated enough that you can find strange things everywhere you look (although op code prefixes and variable length opcodes up to 15 bytes [theoretically at least] still win in my book)
But I think you agree that you can handle most of it while decoding (and making sure that modern compilers use only instructions that are heavily optimized and get around the compability overhead) and I think Intel handles most of it that way as well. And one thing's sure I wouldn't want to be the person having to verify their decoders.

I actually disagree. The decoder can't protect the backend from any of the issues I mentioned, because there are things you can't figure out at decode time.

You could conceptually limit compilers to hardware-friendly operations (emulate the rest, 10X slower), but the risk is that your competitor's compiler will suddenly start emitting operations your chip is deoptimized for and you'll look really bad on the benchmarks they're involved with. People here are welcome to argue that a compiler writer would never be evil enough to do that...but in the real world that concern comes up.

I think the problem in the desktop/laptop space is, that what do you prefer: A 5% more expensive CPU that can run all your programs you already have or a marginally cheaper CPU for which there are only a handful of programs available? I - and I'm pretty sure the majority of people - would take the first choice. On the other hand who'd write a program for a ARM cpu that doesn't target a smartphone?
With the advance of bytecode interpreters that problem should become smaller and smaller, but atm we're not there I think.
The reason why people can port games from consoles to PCs is that there's already a framework that handles most of the complexity - for other applications these don't exist.

I don't think everybody can jump to ARM tomorrow, but I see a trend. I didn't even consider bytecode.

As for "programs I already have", much of what I do is in a web browser; outside the browser, I watch TV (well, when it's Hulu that's in the browser too), play games, and use TuxGuitar and Audacity (both are already cross-platform). Browsers exist on every platform, and as previously discussed many games are developed on PowerPC. I think the growing power of web browsers is actually a relatively large threat to Windows and x86. Once it works in Firefox, OS and instruction set become irrelevant.

Power users here may have more applications that are problematic, but as people like me jump to new platforms (e.g. a $200 impulse-buy tablet/netbook/whatever), some will be developers, and they'll solve the "missing application" problems. Really old legacy applications can be addressed with emulation. Heck, Microsoft already uses that solution even for x86 on x86 because software compatibility is imperfect.

Does anyone have information on how x86 decoding scales for smaller cores like atom and bobcat?

Would it be safe to assume less x86 decoder area is needed for a smaller cpu core?

I don't think I can give a satisfying answer because I don't think there's sufficiently-detailed public information to work from. However, I stand by my statement that there's overhead beyond the decoder (and a lot of that doesn't scale down much).

I keep wondering why atom has such a decreased performance per watt compared to other laptop and desktop processors? IO power budget scaling differences vs. x86 decoder scaling differences vs. other differences?

Given the perf-per-watt of Bobcat versus Atom, it would seem that Intel did something wrong for Atom. It's plausible that they started the design aiming at sub-watt operation, realized they blew the budget, and decided to crank the frequency and push it into netbooks in hopes of making money until they could fix the design. When you take an architecture well beyond its design target (either by pushing a server chip into a laptop, or a laptop chip into a server), you end up with horrible perf/watt.

 

[Khato]

Given the perf-per-watt of Bobcat versus Atom, it would seem that Intel did something wrong for Atom. It's plausible that they started the design aiming at sub-watt operation, realized they blew the budget, and decided to crank the frequency and push it into netbooks in hopes of making money until they could fix the design. When you take an architecture well beyond its design target (either by pushing a server chip into a laptop, or a laptop chip into a server), you end up with horrible perf/watt.

I'm not certain what information is being used to draw that conclusion? In comparison to Bobcat, Atom tends to be equal or better performance per watt on multi-threaded content. Single threaded is likely a different story, but it's hard to say due to the fact that there aren't any power consumption numbers available for that load. Now in comparison to CULV it certainly loses out on performance per watt, but there's a very good reason for that...

For both Atom and somewhat Bobcat, power consumption of the rest of the system plays a larger part than that of the processor and IO hub. That's the unfortunate part in many of the previews that decided to report power consumption figures, especially since pretty much all of them compared the mobile based demo platform of Bobcat against nettop implementations of atom/CULV, which frequently have over twice the idle power consumption of mobile platforms.

 

[cbn]

I'm not certain what information is being used to draw that conclusion? In comparison to Bobcat, Atom tends to be equal or better performance per watt on multi-threaded content. Single threaded is likely a different story, but it's hard to say due to the fact that there aren't any power consumption numbers available for that load.

How do we know the performance per watt of Atom CPU vs Bobcat CPU?

From my layman's point of view it appears Ontario's 9 watt TDP mostly comes from GPU, whereas N550 atom 8.5 watt TDP comes mostly form CPU.

EDIT: I looked through the preview benchmarks and couldn't find how you came up with your conclusion about atom multi-threaded performance per watt? Which preview are you going by? According to PC Perspective Zacate's multi-threaded performance per watt is higher than atom D510.

EDIT 2: My mistake for bringing up Atom N550 and Ontario when we obviously have no benchmarks and power consumption figures. Perhaps we will see a reversal of multi-threaded performance per watt if N550 is found to have a large enough jump in efficiency compared to atom D510. (13 watt TDP for a 1.66 Ghz Atom D510 dual core sounds like a bad trade-off in power consumption compared to 8.5 watts for a 1.5 Ghz atom N550 dual core)

 

[Concillian]

For both Atom and somewhat Bobcat, power consumption of the rest of the system plays a larger part than that of the processor and IO hub. That's the unfortunate part in many of the previews that decided to report power consumption figures, especially since pretty much all of them compared the mobile based demo platform of Bobcat against nettop implementations of atom/CULV, which frequently have over twice the idle power consumption of mobile platforms.

That's the issue that has kept me from an Atom, really. The idle power is not in line with what I'd expect.

hoping that the ITX Brazos boards have at least close to similar idle performance as the mobile platform they're pimping around now. Anything is going to be less than the S939 A64 I'm using for a fileserver now, my needs are so low for that machine that idle power and cost is about all I'm looking at. Right now, nothing makes sense yet. All the Dxxx Atoms are pretty poor on idle. There's an N550 Atom ITX board available that does pretty well for idle power... for $200... umm... $200 for an Atom + mobo? wat?

 

[Khato]

How do we know the performance per watt of Atom CPU vs Bobcat CPU?

From my layman's point of view it appears Ontario's 9 watt TDP mostly comes from GPU, whereas N550 atom 8.5 watt TDP comes mostly form CPU.

EDIT: I looked through the preview benchmarks and couldn't find how you came up with your conclusion about atom multi-threaded performance per watt? Which preview are you going by? According to PC Perspective Zacate's multi-threaded performance per watt is higher than atom D510.

EDIT 2: My mistake for bringing up Atom N550 and Ontario when we obviously have no benchmarks and power consumption figures. Perhaps we will see a reversal of multi-threaded performance per watt if N550 is found to have a large enough jump in efficiency compared to atom D510. (13 watt TDP for a 1.66 Ghz Atom D510 dual core sounds like a bad trade-off in power consumption compared to 8.5 watts for a 1.5 Ghz atom N550 dual core)

If you want to go by the PC perspective figures, then the multithreaded benchmark in use is Cinebench 11. Thanks to giving both platform power consumption numbers for idle and load during that benchmark, we know that the E-350 CPU power consumption is 8.8W + idle power, whereas the D510 is 3.7W + idle power. Meanwhile performance for Cinebench 11 Render Test is 0.6 for the E-350 and 0.55 for the D510 - a 10% performance advantage to the E-350 (in line with other benchmarks.) The difference in idle power on the processor itself is impossible to say, but on a platform there is at least some information to be found. One good resource for mobile system power consumption numbers would be notebookcheck, which has the D525 powered Asus 1215N with ion2 idling at 11.1W, or if you go to N550 based there's both an Acer D255 at 'medium' idling 6.3W, and Asus 1015PEM at 'medium' idling 10.8W. Which is the long way of saying that there really isn't any proof that either CPU beats out the other when it comes to idle power draw, we'll have to wait until actual product is available.

Anyway, once you acknowledge that the idle CPU power consumption of Atom and Bobcat are roughly equal, then the only hard number we have left is the dynamic power consumption. Of course, can't just use that for determining a performance per watt, so let's just say that both processors have a 4 watt idle power draw (I expect it's far lower than that though.) That at least brings the Atom down to only 1.5x the performance per watt of Bobcat on multithreaded applications. I wouldn't be surprised if in single threaded applications it's knocked down to being basically equal on performance per watt, just with far lower performance, haha.

 

[Khato]

That's the issue that has kept me from an Atom, really. The idle power is not in line with what I'd expect.

hoping that the ITX Brazos boards have at least close to similar idle performance as the mobile platform they're pimping around now. Anything is going to be less than the S939 A64 I'm using for a fileserver now, my needs are so low for that machine that idle power and cost is about all I'm looking at. Right now, nothing makes sense yet. All the Dxxx Atoms are pretty poor on idle. There's an N550 Atom ITX board available that does pretty well for idle power... for $200... umm... $200 for an Atom + mobo? wat?

First off, it's not the Dxxx Atom chips themselves that are poor on idle, they're really not much different from the Nxxx chips - they just didn't necessarily bin as well. The main difference is that mini-ITX boards based off the Dxxx chips assume prioritize cost over power - they figure it's low enough and why care about a few extra watts if it's plugged in anyway. That N550 based ITX board meanwhile is targetted towards the crowd that want lower power consumption, and hence they prioritize power over cost... and then charge a hefty premium for what is quite the niche product. Sadly, I wouldn't expect that trend to change really on Brazos based ITX boards, but at least then you're getting more performance for a minimal increase in total system power.

Anyway, I can definitely see the Brazos platform being popular for home/small business file servers thanks to its IO hub having 6 SATA ports compared to the NM10's 2. Though if pricing is similar then there would be no incentive for OEMs to design new products for that market when the atom based ones are 'good enough'.

 

[cbn]

Anyway, once you acknowledge that the idle CPU power consumption of Atom and Bobcat are roughly equal, then the only hard number we have left is the dynamic power consumption. Of course, can't just use that for determining a performance per watt, so let's just say that both processors have a 4 watt idle power draw (I expect it's far lower than that though.) That at least brings the Atom down to only 1.5x the performance per watt of Bobcat on multithreaded applications. I wouldn't be surprised if in single threaded applications it's knocked down to being basically equal on performance per watt, just with far lower performance, haha.

Ontario and Zacate have relative large GPUs on die and unless AMD figures out a way to completely shut them down isn't this going to cloud the picture of the true CPU idle?

 

[cbn]

If you want to go by the PC perspective figures, then the multithreaded benchmark in use is Cinebench 11. Thanks to giving both platform power consumption numbers for idle and load during that benchmark, we know that the E-350 CPU power consumption is 8.8W + idle power, whereas the D510 is 3.7W + idle power. Meanwhile performance for Cinebench 11 Render Test is 0.6 for the E-350 and 0.55 for the D510 - a 10% performance advantage to the E-350 (in line with other benchmarks.) The difference in idle power on the processor itself is impossible to say, but on a platform there is at least some information to be found. One good resource for mobile system power consumption numbers would be notebookcheck, which has the D525 powered Asus 1215N with ion2 idling at 11.1W, or if you go to N550 based there's both an Acer D255 at 'medium' idling 6.3W, and Asus 1015PEM at 'medium' idling 10.8W. Which is the long way of saying that there really isn't any proof that either CPU beats out the other when it comes to idle power draw, we'll have to wait until actual product is available.

How are you coming up with Zacate having worse multi-threaded performance per watt than Atom D510 according to PC perspective tests? (This is exactly the opposite of what the author of that article is saying).

If anything I would actually expect Zacate's CPU performance per watt to be even higher than what is concluded. Realize PC perspective is going by total power consumption during the R11 test (this would include some of Zacate's large GPU idling as well). In contrast Atom D510 has only a very small GPU idle power consumption contributing to the total power consumption figure.

Of course, We could also look at dynamic power consumption (as you have pointed out) That would eliminate Zacate's large GPU as a variable, but then this is confounded by D510 high CPU idle (which is due to lack of Speed Step).

 

[Concillian]

First off, it's not the Dxxx Atom chips themselves that are poor on idle, they're really not much different from the Nxxx chips - they just didn't necessarily bin as well. The main difference is that mini-ITX boards based off the Dxxx chips assume prioritize cost over power - they figure it's low enough and why care about a few extra watts if it's plugged in anyway.

It's more than just binning. Intel disables speedstep and the idle C-states for the Dxxx Atoms:

http://ark.intel.com/Product.aspx?id=49490

Notice the
idle states: NO

I have no idea why, as it's basically the only reason I don't currently own an Atom. AMD hopefully leaves those idle power reduction functions in place, if they do that and are cost competitive at all with Atom ITX boards, there will be zero reason for the Dxxx atoms to exist.

 

[cbn]

It's more than just binning. Intel disables speedstep and the idle C-states for the Dxxx Atoms:

http://ark.intel.com/Product.aspx?id=49490

Notice the
idle states: NO

I have no idea why, as it's basically the only reason I don't currently own an Atom. AMD hopefully leaves those idle power reduction functions in place, if they do that and are cost competitive at all with Atom ITX boards, there will be zero reason for the Dxxx atoms to exist.

Thanks for pointing out the differences in idle C-states. (I learn something new on this forum everyday)

Just wondering if anything else is also contributing to Atom D510s higher idle:

Intel GMA specifications

I noticed Atom D510's GMA 3150 GPU runs faster (400 Mhz vs 200 Mhz) than the Netbook Atoms 3150 GPU.

Does anyone have information on possible idle power consumption differences for these Graphics processors?

 

[Khato]

How are you coming up with Zacate having worse multi-threaded performance per watt than Atom D510 according to PC perspective tests? (This is exactly the opposite of what the author of that article is saying).

If anything I would actually expect Zacate's CPU performance per watt to be even higher than what is concluded. Realize PC perspective is going by total power consumption during the R11 test (this would include some of Zacate's large GPU idling as well). In contrast Atom D510 has only a very small GPU idle power consumption contributing to the total power consumption figure.

Of course, We could also look at dynamic power consumption (as you have pointed out) That would eliminate Zacate's large GPU as a variable, but then this is confounded by D510 high CPU idle (which is due to lack of Speed Step).

I come up with Zacate having worse multi-threaded performance per watt than the Atom D510 because the PC perspective tests completely disregard the platforms that they're running on. Put a Zacate into the same class of system as the D510 and then it'll be the one with way higher idle power consumption. After all, just take a look at a good review of the Asus 1215N that at least give a run-time during light usage - despite having a D525 and ion2 it still typically manages around 340 minutes, which with its 56W/h battery equates to basically 10W power consumption. Which makes sense given that systems based on the N550 are typically around 7-8W power consumption at idle.

Primary point of all that is... When you're talking about relatively small differences in power consumption, the quality of the platform plays an ever larger role... Which actually could turn out in AMD's favor, but I doubt it. Why? Because one of their PR slides included in a few of the reviews, titled "Brazos Core Power", states that the average idle core power is 2.71W, while average active core power is 6.5W - 3.8W of difference. Since they don't explicitly state that that's for dual core (if it was, you could be sure marketing would have put it there), then that's likely per core, which happens to roughly match up with the dynamic power difference between idle and 100% load on the PC Perspective review - 7.6W for CPU and 1.2W for memory/IO = 8.8W. That would put idle power for the cores at 5.4W(probably a bit less, depending upon what all they included in the 2.71W figure), which is actually rather reasonable when you consider that the GPU should be drawing 3W or less at its reduced 2D speed. After all, AMD marketing slides when the 5450 was released stated total card power draw was 6.5W, of which it's reasonable to assume half or more would go to DC-DC losses, memory, and memory controller.

Oh, that brings me to another interesting point. I do wonder where the PC Perspective power measurements were made - at the AC plug or after the conversion to DC? After all, their readings are a fair bit lower than other reviews were - 9.3W idle, 19.1W CPU load, 28.8W full load, compared to 11W idle and 31W full load at Hot Hardware, 26W CPU load and 33.6W full load at Tech Report, 15W idle, 25W CPU load, and 30W full load at Anandtech. Which is one way of saying... there's an awful lot of variation going on here, haha. Makes me look forward all the more to the actual release and getting to see what it's really like instead of trying to estimate comparisons with so many unknowns.

 

[cbn]

After all, just take a look at a good review of the Asus 1215N that at least give a run-time during light usage - despite having a D525 and ion2 it still typically manages around 340 minutes, which with its 56W/h battery equates to basically 10W power consumption. Which makes sense given that systems based on the N550 are typically around 7-8W power consumption at idle.

We'll have to wait and see how Brazos compares on the Netbook platforms.

AMD marketing slides when the 5450 was released stated total card power draw was 6.5W, of which it's reasonable to assume half or more would go to DC-DC losses, memory, and memory controller.

I might have overestimated the contribution of the AMD Brazos GPU to the total TDP:

(Looking back on one of my old threads I found this)

http://www.amd.com/US/PRODUCTS/NOTEBOOK/GRAPHICS/ATI-MOBILITY-HD-5400/Pages/hd-5430-specs.aspx

http://www.amd.com/US/PRODUCTS/NOTEBOOK/GRAPHICS/ATI-MOBILITY-HD-5400/Pages/hd-5470-specs.aspx

I noticed the TDP varies by a quite a large amount for these 80 stream processor mobility chips.

In the first link we have the 5430 mobility radeon at 550 Mhz core clock and 1.6 Gbps GDDR3 memory for 7 watt TDP.

In the second link the 5470 mobility radeon with the 750 Mhz core clock and 32.2 Gbps GDDR5 scales to 15 watts TDP.

How much energy would a 500 MHz Zacate GPU consume?

 

[Eug]

Core i3 330UM laptops (with 8-hour battery life) are now under $600. I wouldn't be surprised if Core i3 laptops hit the < $500 price point by the time Zacate launches.

They'll be slower than Zacate for gaming, and that's Zacate's one advantage, but that's a small minority of the very low power market. For CPU speed, Core i3 330UM 1.2 GHz will likely be faster than Zacate dual-core 1.6 GHz.

This will have to be factored into the price of Zacate. If dual-core Zacate laptops launch at greater than $600 at the low end, they'll sell but I don't see them flying off the shelves.

 

[cbn]

For CPU speed, Core i3 330UM 1.2 GHz will likely be faster than Zacate dual-core 1.6 GHz.

It is faster.

http://www.anandtech.com/show/4023/the-brazos-performance-preview-amd-e350-benchmarked/3

Recently Anandtech added the Core i3 330UM to the Zacate CPU benchmarks.

 

[dangerman1337]

Core i3 330UM laptops (with 8-hour battery life) are now under $600. I wouldn't be surprised if Core i3 laptops hit the < $500 price point by the time Zacate launches.

They'll be slower than Zacate for gaming, and that's Zacate's one advantage, but that's a small minority of the very low power market. For CPU speed, Core i3 330UM 1.2 GHz will likely be faster than Zacate dual-core 1.6 GHz.

This will have to be factored into the price of Zacate. If dual-core Zacate laptops launch at greater than $600 at the low end, they'll sell but I don't see them flying off the shelves.

Seeing how the Nile platform is priced right now and the target for Bobcat/brazos/zacate, i woudln't be surprised if it goes around the $400-500 (hopefully for Zacate)

 

[Hacp]

Core i3 330UM laptops (with 8-hour battery life) are now under $600. I wouldn't be surprised if Core i3 laptops hit the < $500 price point by the time Zacate launches.

They'll be slower than Zacate for gaming, and that's Zacate's one advantage, but that's a small minority of the very low power market. For CPU speed, Core i3 330UM 1.2 GHz will likely be faster than Zacate dual-core 1.6 GHz.

This will have to be factored into the price of Zacate. If dual-core Zacate laptops launch at greater than $600 at the low end, they'll sell but I don't see them flying off the shelves.

Yet the AMD chip owns intel in graphics performance. I think AMD is aiming for a price range of 400-500 dollars for their chips. Remember, they made the bobcat chips as cheaply as they could so they could mass produce them and put them in cheap laptops.

 

[cbn]

I think AMD is aiming for a price range of 400-500 dollars for their chips. Remember, they made the bobcat chips as cheaply as they could so they could mass produce them and put them in cheap laptops.

I know bobcat was meant to be a low cost APU, but how much does that low cost factor into the total bill of materials for a complete 10.1" or 11.6" laptop?

 

[Hacp]

I know bobcat was meant to be a low cost APU, but how much does that low cost factor into the total bill of materials for a complete 10.1" or 11.6" laptop?

A significant amount I would assume. How else would 10.1 inch atom netbooks be 300-400 bucks?

 

[cbn]

http://www.pcper.com/article.php?aid=1039

By keeping the Brazos platform as a lower power level AMD is aiming squarely at the netbook/notebook markets as well as the super-budget desktop machines. During my brief time with the development platform I was very impressed with out how cool the APU was during pretty strenuous testing - in many cases the fan on the tiny heatsink wasn't even spinning the temperatures were so low.

When is the last time you have seen temperatures that low on your CPU?

Comments on the temps? How does this compare to atom?

I am just wondering if we will see more OEMs come out with overclocking options for Bobcat netbooks (particularly Ontario APU).

For example, ASUS has a ultility called Super Hybrid drive. More information here.

MSI has also been know to provide overclocking through the Bios. More information here

(This might be particularly attractive for people that want to use a netbook as a primary computer when docked to AC power and connected a larger TV or LCD monitor).

 

[OCNewbie]

CNN always alerts me that the poll I'm participating in isn't scientific. I feel a little insecure participating in your poll without knowing if it's scientific or not. So, is it a scientific poll?

 

[Khato]

Comments on the temps? How does this compare to atom?

I'm quite disappointed with myself for having not thought to check Tom's Hardware for their preview. I mean, it's actually comparing the Brazos platform against notebooks, and they have a temperature picture. Granted, the comparison to the D525+ion on the temperature picture is kinda weird given the difference in heatsink, but it still hows 101F max for the E-350 and 117F max for the atom. But then if you take a look at the article, they actually recognized that difference and went ahead with taking the heatsinks off both and running bare die for a short amount of time, at which point the E-350 max was just shy of 180F, while the D525 was just shy of 155F... The debate being whether the red line on the E-350 is the memory controller side of the chip or the CPU side of it.

As for the power consumption numbers, they just have a single set of 'typical system use power' under relatively matched conditions between the systems - they even measured DC consumption for the Brazos system rather than AC. The amusing result being that the SU7300 based system beats out all the rest at 10.4W, while the D525+ion basically ties the Neo K125 at a touch over 14W, and Brazos is at 18.6W. Also interesting is their speculation that the GPU portion might have been running at full speed all the time, which would bring that number down a fair bit.

As for their benchmarks, one point of annoyance is that there's no mention of hard drive for each of the systems. Given the benchmarks where the E-350 is suddenly neck and neck with the SU7300, I kinda expect that the E-350 had the 128GB C300 mentioned in other reviews while the SU7300 had a normal laptop spindle drive.

Regarding pricing... According to ark.intel.com the 1kU price for D510 is $63, while an N550 is $86, and the NM10 is $20. And yet somehow you can buy the Intel mini-ITX motherboard with a D510 in it for $80, haha. So who knows exactly how much the CPU + chipset portion of a netbook is, but you can be sure that it's under $60, probably more like $50 - the motherboard itself costs quite a bit after all. One key difference on the pricing aspect that seems to be frequently disregarded is that while Brazos does have a die size advantage, its competition (atom + CULV) is likely cheaper to manufacture. Why? First up, they're just keeping Intel's N-1 45nm fabs busy - I'd imagine that by now the equipment has already paid for itself, so keeping it in use longer just means more profits. Same goes for the northbride/IO hubs, though there it's 65nm for the CULV and I think 130nm for the NM10/ICH. Now on the other side, AMD has to pay TSMC for each chip - 75mm^2 on the current generation 40nm process and then however big hudson is on their 65nm process (as a sidenote, I find it amusing that AMD decided to do their IO chip on 65nm, there's a reason why Intel has stuck with 130nm there.) Point being that while AMD might have designed a chip that can be 'stamped out for cheap', it's still likely going to cost them more to manufacture than it'll cost Intel to produce their competition.

 

[cbn]

The thing that people seem to be missing is there is one HUGE advantage that will ensure these sell like hotcakes:

They let the OEMs market the systems as having Directx 11.

Yeah, it's definitely not fast enough for DX11 games at any good quality, but that's not the point. OEM's love features, and DX11 is a big one. The average consumer that sees two laptops or netbooks, one with DX10, one with DX11, otherwise similar in price/features, is going to pick the DX11 one every time imho. This is why I believe these will sell by the bucketload to OEMs.

I believe that will help gamer sales of Windows 7 starter netbooks, but what about the multitude of potential netbook owners that are non-gamers?

Wouldn't they rather connect their Ontario netbook to a projector, TV and/or Larger standalone LCD monitor for multi-media and/or web surfing? (MS has disabled these functions in Windows 7 starter)

How about Ontario coupled to Google Chrome OS? (AFAIK, Chrome OS allows connection to a projector, TV and/or LCD monitor). Furthermore, the Wikipedia article on Chrome OS lists "offline media player", How would Ontario match up against ARM for that purpose? (I know the ARM Cortex A9 cores are pretty strong, but what about 1080p video capability?)

In July 2010, Google CEO Eric Schmidt said a Google-branded Google OS netbook was unlikely, despite Google's having previously negotiated with a couple of hardware manufacturers to produce it, and despite an earlier Google-branded device, the Nexus One Android phone. "Let's see how well those partners do first. My guess is we won't need to. The PC industry is different from the phone industry. The PC industry is used to working with Microsoft, whereas the mobile industry was not used to working with software".[35]

In early November 2010 Digitimes reported that Google was set to release a Google branded Chrome OS notebook later in the month. The notebook will be ARM-based, manufactured by Inventec and initially ordered in numbers of 60-70,000 units. The same source indicates that Acer and Hewlett-Packard will launch smartbooks manufactured by Quanta Computer in December 2010.[36]

EDIT: 11/21/2010 Google Chrome OS has been delayed-->http://www.eweek.com/c/a/Desktops-a...t-No-Chrome-OS-Netbooks-for-Christmas-178567/ (Speculation on why this is happening?)

Opinions?

How about Meego and Ontario?

 

[Idontcare]

CNN always alerts me that the poll I'm participating in isn't scientific. I feel a little insecure participating in your poll without knowing if it's scientific or not. So, is it a scientific poll?

Yes, absolutely it is, without a doubt.

All polls on public forums are to be assumed as being scientific polls unless explicitly stated otherwise.

We are, after all, of comparable caliber to that of other corporate media giants such as CNN, MSNBC, etc. So it only stands to reason that your expectations regarding the merits of their polls and ours are exactly as you presented in your post.

 

[OCNewbie]

Yes, absolutely it is, without a doubt.

All polls on public forums are to be assumed as being scientific polls unless explicitly stated otherwise.

We are, after all, of comparable caliber to that of other corporate media giants such as CNN, MSNBC, etc. So it only stands to reason that your expectations regarding the merits of their polls and ours are exactly as you presented in your post.

Phew, thank you. This was really eating away at me. The massive wave of relief that I'm feeling right now is just what I needed.