VIA Nano CPU Benchmarked, Beats Intel Atom

Vigile writes "Back in May, when the Isaiah architecture was first disclosed, VIA declared a performance victory over Intel's upcoming Silverthorne technology. Since then, Isaiah has become the VIA Nano processor, and Silverthorne changed to the Intel Atom — and now we can finally see tests comparing the two technologies. The Nano's out-of-order super-scalar design is definitely an architectural leap over the Atom's in-order single-issue design, but with Intel including HyperThreading technology in their CPU the competition is closer than expected. The Nano does win the performance tests by a considerable margin, but what might be more impressive is seeing the Atom use only 4 watts of power under full load!" As reader Mierdaan points out, that's 4 watts more than at idle, for about 60 watts total.

Comments

Misleading title?

[pwnies]

Seems to me that the title is getting across the wrong message. These processors are meant to be placed in ultraportables, where battery life is a MAJOR factor. In that sense, the atom easily beat out the nano here, seeing as they used 4 watts and 18 watts respectively on a full load.
With that amount of difference in power required, it's pretty obvious the nano would beat the atom, but that's like saying a smart car with a V8 is going to beat one with a V4 when it comes to speed (except they should have been testing efficiency, where the V4 blows the other out of the water)

Re:Misleading title?

[jmorris42]

Well yes and no. Did you read the part where they measured total power expended to accomplish various tasks? Burning more power is OK if you can finish the job faster and get back into a low power state. When that is factored in the contest is a bit closer. Of course if the Nano does run for long it is going to bake your lap more than the Atom and drain the battery a lot faster.

Looking at the photos makes it plain where the problem now lies, the northbridge. If Intel can get theirs under control they will totally dominate the low power business. But since the Nano draws so much more power a low power northbridge won't help them as much, which bodes ill for the future. Intel has a lot more room for improvement while Via would have to pull a major rabbit out of their hat to cut much off their current power consumption numbers.

And for small computers that aren't running on batteries but do need to be fairly cool (i.e. quiet) the Nano will be the hands down winner just because Intel is playing marketing games.

Re:Misleading title?

[Sj0]

I4. There are very few V4 engines, probably because it wouldn't be any more balanced than an inline 4 configuration, but would require much greater complexity. You could make a V4 shorter than an I4, which is why it's used extensively in motorcycle, but it'll cost much more to design and build.

You're bang on though. An Intel Extreme quad-core will eat a Via Nano for breakfast. You could probably emulate the Via Nano faster on the Intel Extreme quad-core than the Via will even go. The thing is, the Nano will last for an hour on the energy the Intel Extreme quad-core will use in a minute.

Re:Misleading title?

[jd]

Well, for something like that, you want several benchmarks - actual performance per actual watt, effective performance per usable watt, and mean wattage consumed under stress. In other words, what performance will you get per watt of power actually consumed, what performance do you get per watt consumed over idle, and if you really push the processor to do the absolute maximum it is physically capable of (in terms of MIPS, FLOPS and as many other metrics as you care to use) what wattage can you actually get it to consume on average? (Peak usage isn't necessarily useful if it's not sustainable.)

Re:Misleading title?

[4D6963]

4 Watts? Wait.. doesn't a ARM Cortex-A8 only use 300 mW?

Re:Misleading title?

[wren337]

I hate to actually read the article and then post, but the article makes it clear that the VIA Nano uses less power to perform the benchmark tests than the Intel chip, by taking slightly more power and finishing much faster. Running with 10% less wattage and taking 30% longer to complete is no savings.

[QUOTE]
For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watt-seconds (Joules) of energy while the Intel Atom processor used 38,290 watt-seconds (Joules) of energy.
[/QUOTE]

Re:Misleading title?

[Anonymous Coward]

Any 90 degree V configuration of two cylinders is going to have better reciprocating mass balance than any inline configuration of two cylinders.

So a V4 (assuming 90 degree V) is going to exhibit better mass balance than an inline 4. It is purely the complexity issue where an inline 4 wins.

A V configuration averages mass acceleration rates between the two cylinders. While one piston is at the point of minimum velocity (standing still) and entering an acceleration phase, the other piston is at a point of maximum velocity and entering a deceleration phase. The energy transfers between the pistons, leaving the combustion process purely to drive torque to the crank. If you draw this out schematically, you will find this velocity/acceleration relationship throughout the rotation of the crank.

This is good for two things. 1) smoothness and 2) responsiveness to throttle input at high RPM. At high RPM the acceleration rates are much higher because of the higher peak piston speeds, and the torque reaction to the acceleration cycle begins to compete with the torque generated by the combustion process, reducing the responsiveness of the engine to throttle changes. It becomes a signal to noise problem.

This is probably less of an issue for your "daily beater" but can make a difference in competition driving.

Yamaha did some excellent investigation work on this where they fitted an inline 4 engine with an offset crank so that the mass velocity cycle mimicked that of a V4. The were other issues that they had to do to make it work in inline 4 format, but the results were very good.

Re:Misleading title?

[frieko]

I'm a little confused here. My full-sized Centrino laptop draws 22 watts at idle and 38 watts at full load. That includes the LCD. How are either of these an improvement?

Re:Misleading title?

[Steve Max]

Remember that the boards are using an ultra-fast and ultra-power-hungry HD.

Imagine that the HD uses 20W during the test. If the Atom takes 120s to complete it, that's 240 joules of the total that come from the HD; if the Nano takes 60s, that's 120 joules from the HD. My point is that, if you use a low power device, both lines would go down by the same amount, so the integral for the Atom would go down by more than the one for the Nano. This effect could be big enough to make the Atom more attractive.

Unfortunately, they decided to use a power hog to test the energy use of low power systems, making their test very flawed and, therefore, unreliable. I'll wait for a better test before making any conclusion.

Re:Misleading title?

[postbigbang]

Correlating these into duty cycles, it's dependent on the design of the device and its usage profile with the duty cycles as to how much the overall design consumes power.

If you extract the base design, the numbers for Nano are very good and should be lauded. Now it's time to reduce the transaction cycles of peripheral devices and complete the chain of efficient design.

Re:Misleading title?

[nabsltd]

Did you read the part where they measured total power expended to accomplish various tasks? Burning more power is OK if you can finish the job faster and get back into a low power state.

Their numbers are messed up, though.

The VIA system uses 17.4 more watts while under load than the Intel system. That means if they are under load for one second, the VIA system has used 17.4 more watt-seconds than the Intel. On the other hand, the Intel only uses 0.9 more watts under load than the VIA uses when idle. So, if the VIA works for one second, then drops back to idle while the Intel is still working to complete the task, it would take 18.33 more seconds before their total power usage becomes the same (17.4 + (-0.9 * 18.33) == 0).

This means that to use the same amount of power over the long term, the VIA has to be able to complete tasks in ~5% of the time that the Intel takes. Put another way, the Intel would take 19 times as long to finish any task.

But, their own graphs show that the Intel system only takes about 2 times as long (at most) to finish tasks. So, there's no way that the VIA uses less energy over the long haul, but that's what they claim.

Their problem is that they are comparing the total power used by the VIA system only while "active" on the task, and not looking at the power that the VIA uses on idle while the Intel is still completing the task. This is like saying that as soon as you finish that task, you immediately power off (or drop to standby).

Re:Misleading title?

[Anonymous Coward]

They are using a 20W hard disk in their tests, while your laptop uses a lower power disk (SSD?).

Re:Misleading title?

[mihalis]

I hate to actually read the article and then post, but

Get Out

Re:Misleading title?

[smallfries]

Although the theory is nice it assumes that you shut down the Nano after it finishes the task. Because the idle drain on the Nano is so close to the load drain on the Atom the theory breaks down. Assuming that the Nano machine sat idling for the remaining time it is still power consumption they did not include in the comparison.

Of course I'm still trying to get my head round the concept that I should buy Via for performance and Intel for power consumption...

Re:Misleading title?

[0xABADC0DA]

Burning more power is OK if you can finish the job faster and get back into a low power state.

A flash ad that runs for 10 seconds will run for 10 seconds on a slow, low-power processor and also on a fast, high-power processor -- but the fast processor will do more work and use more power. A word processor that lays out the pages in the background will do less total work on a slow, efficient processor since the user will have typed in more data per layout iteration (so it will settle fewer times).

Just for a couple of counter-examples... except in contrived benchmarks, a slower processor often does less total work to get the same results.

Re:Misleading title?

[smellsofbikes]

Another really nice thing about inline and horizontally opposed 4-cylinder engines is that they have a flat crankshaft, so it can be formed by forging, giving a very strong, light crankshaft. I believe V4's require a cast crankshaft, because it has 90 degree segments (unless you want it to run/sound like a Harley.) Castings aren't as strong as pure forged.

At least that's what I've read when I was rebuilding my Subaru engine.

Re:Misleading title?

[TheRaven64]

Unlikely to make a difference, since both will be transferring the same amount of data to and from the hard disk (and so there is no difference if it's done by a read-ahead DMA), but more importantly because you tend not to spin a disk down until it's been idle for about ten minutes or the extra power needed to spin it up plus the user-irritation caused by waiting for it to spin up remove most of the advantage.

Re:Misleading title?

[Anonymous Coward]

Get Out

Go easy on that helicopter!

Re:Misleading title?

[Anonymous Coward]

V4 aren't used extensively in motorcycles. A list of inline 4 cylinder motorcycles would go on forever, while a list of V4 motorcyles includes the Honda VFR series, and some bikes by Matchless before any of us were born. Other than that, there were few uses of the V4 engine layout on two wheels.

Re:Misleading title?

[hairyfeet]

The problem with the Intel is they have tied it to the 945GC chipset. If you look at the pictures here you'll notice the giant heatsink+ fan for the Atom.That is NOT for the CPU,that is for the chipset. That chipset has a TDP of 22 watts,which is 6 times more than the actual CPU! I just wonder if they are saving part of their market for the Celeron,otherwise somebody at Intel made a bad mistake with that chipset. But as always this is my 02c,YMMV

Re:Misleading title?

[billcopc]

Burning more power is OK if you can finish the job faster and get back into a low power state.

Ok then, how about four Atoms in the same power footprint as one Nano ? I'll take 4 Intels over one Via any day. Heck, I'd take one Intel over a Via!

Re:Misleading title?

[Locutus]

something I saw recently mentioned that the atom requires a 2nd chip( northbridge ) which is often over looked when compared to other UMP SoC chips. I didn't look if this applied to these item but it should be kept in mind for things like ARM SoC's, the Geode and any other chip which does far more than the Atom alone.

Another interesting bit was that Intel had to go to 45nm to get the Atom to its current power range while almost all the existing competitors are built on 65nm. That is important because it means the others have just a die shrink to jump down even further in power usage if they even need to in order to put up better numbers than Itel's Atom ant.

LoB

Re:Misleading title?

[Locutus]

The Via Nano is currently built on a 65nm process and the Intel Atom on 45nm so Via has some room if they move to the 45nm process.

Makes me wonder if building the Atom on 45nm is costing them production of high price multi-core chips in order to squeeze themselves into this UMP market before the sector solidifies more. And seeing things like the TI OMAP 35xx chips it makes me wonder when the UMP market moves off x86. There is already pressure to bump up the price of the UMP as to not take away from the lowend laptop market.

LoB

Re:Misleading title?

[KillerBob]

Well yes and no. Did you read the part where they measured total power expended to accomplish various tasks? Burning more power is OK if you can finish the job faster and get back into a low power state. When that is factored in the contest is a bit closer. Of course if the Nano does run for long it is going to bake your lap more than the Atom and drain the battery a lot faster.

When you're talking about an always-on device like an HTPC or even a low-powered server (I have a file & print server running on a C7 1.5GHz w/ 2GB of RAM, for example), the Atom wins out though... from TFA, the peak draw on the Atom was lower than the minimum draw on the Nano. Even though the Atom takes as much as 30% longer to perform a task, meaning that when it's multiplied out it's drawing more joules to perform the task, the fact that you don't go to a zero power state when you're not performing a task is what wins out.

I wouldn't put it in something that needs performance... but for something that's always going to be on, and going to be idle most of the time, I'd certainly go for the Atom if I were building that server again today. It isn't significantly more expensive, and it will use less power over the course of the year. It seems like a great choice for something like a small network file/print server, a house server, a car computer, or an mp3/video jukebox, etc. As I'm planning on building an off-the-grid house powered by solar, having a low-power computer that can do things like rotate the panels to track the sun, and manage power distribution to minimize consumption, those extra 4W at idle is important to me.

Re:Misleading title?

[Atario]

Their problem is that they are comparing the total power used by the VIA system only while "active" on the task, and not looking at the power that the VIA uses on idle while the Intel is still completing the task. This is like saying that as soon as you finish that task, you immediately power off (or drop to standby).

Or you go on to another task. And you continue till you finish all your tasks, then you power off. Which is essentially the same thing.

Re:Misleading title?

[LarsG]

That Raptor is not quite that bad - 9W at idle-but-spinning and 10W at read/write. Still, that's 7-8W more than a decent laptop drive.

Not to mention that the i945G chipset in the Atom PC is something like 20+W. The Nano motherboard idle power seems a bit high too.

If the goal of the benchmark was to test the limits of performance per watt on the two platforms, the choices they made seems silly. All the benchmark really shows is that it makes no sense to put an efficient engine in a heavy suv.

Re:Misleading title?

[LarsG]

It does make a difference because the benchmark claims that it compares performance per watt. If you compare the CPUs only, the Atom is 3+ times more efficient; the more power-guzzling components you add to the two test systems, the more you hide that fact.

Take the "Nano is 3% better at mp3 encoding" claim. Replace that Raptor with a 2W laptop harddisk and suddenly the Atom is better.

Re:Misleading title?

[LarsG]

Speaking of the 35xx, check out Pandora for one example of what people are building with that chip. Heck, even last generation's OMAP2420 had enough oomph to run Linux with acceptable performance (Nokia N800/N810) and the 35xx is expected to be about 4x the speed.

Unless Intel gets their act together soon with Moorestown, they might find that the UMP market is eaten by Cortex-based ARM SoCs; at least the part of the market where x86 compatibility isn't important. One can at least hope, the Cortex certainly has enough performance for most tasks one would want to do with a handheld.

Re:Misleading title?

[Brain Damaged Bogan]

the car analogy is a bad one, a turboed, bored and tuned 4 banger can outperform a v8, especially around corners, ever pitted an exige against a dodge viper on a twisty track?

Re:Misleading title?

[LarsG]

Replace the Raptor HD with a laptop HD, and the Atom suddenly comes out on top. Not to mention that their "benchmark" ignores that most PCs spend a lot of time at idle; those numbers are only valid if you power on - encode mp3s - power off.

Atom is not designed for the desktop, it is too slow for that. It is designed for laptops and subnotebooks where power efficiency is more important than speed. Not to mention that the 945G chipset in that Atom PC they built is not exactly low power - 20+W.

Re:Misleading title?

[Deadfyre_Deadsoul]

The big difference I see. Pin to Pin compatibility. With Intel, anything new cpu wise "always" requires a new motherboard (or so it seems). Its always been more pins to socket 939/940/945. But then again, that plays into the marketing aspect which Intel has always rocked on. Via is just trying to not only get a foot in the door, but break the door knob.

The kicker though, for what I see, is the hardware accelerated cryptographic security setup for hashing. While it doesn't seem like much now, everyone else is going to emulate it into their own new stuff. Kudos for Via.

The power consumption is only a bonus.

I think by xmas time, they should have a better version of these out, and I'll have to pick one up for the new year.

Re:Misleading title?

[LarsG]

Yeah, the chipset issue is important and is currently the big missing piece in the puzzle if Intel wants to get into the UMP / high end cell phone game. Complete ARM SoCs that include CPU, USB, 2D/3D acceleration, SD-card interface, memory controller, on-chip RAM, camera interface with JPEG/MPEG-4 encoder and perhaps even a kitchen sink use about as much watts as the Atom CPU alone.

It is even an issue on the desktop. In that "benchmark" they used a 945G chipset motherboard. That chipset pulls 20+W, which is kind of silly when matched with a 4W CPU. The only reason that the Nano eked out a slight win in work per watt was because of the high power draw of the other components of the PCs; even replacing that Raptor HD with a laptop HD would have put the Atom on top.

Re:Misleading title?

[repvik]

Socket 939 and 940 are AMD sockets, not Intel.

Re:Misleading title?

[Deadfyre_Deadsoul]

Yeah, quite true. I don't actually run any Amd boards now. Just intel boards. Im to tired to even remember what motherboard this is. Oh, sleepy land where art thou. The pin to pin ratio has been tossed up in the air ever since cyrix was making cpu's. Its the marketing theory on forcing people to upgrade. Dogbert would be proud.

Re:Misleading title?

[Jorophose]

You know why they did that don't you?

The atom has next to no features by itself. It really is a "dumb-terminal" amongst CPUs.

The Nano is capable of much more, even with its chipset. Namely, last I checked it's capable of playing h.264 at 1080p, and can play Crysis.

Re:Misleading title?

[Whiteox]

Remember that the boards are using an ultra-fast and ultra-power-hungry HD.

Don't forget that you could use low power SDHDs instead of conventional drives.
Personally I'm very eager for ultra-low power computing - a headless 5 watt system sounds like the way to go, that's if a 1 TB SDHD card becomes amazingly cheap!
Well I can dream can't I?

Re:Misleading title?

[ScreamingCactus]

When did they start making V-4's?

Re:Misleading title?

[Bert64]

The total power expended at idle wasn't a huge gap between the two systems...
At load the Nano was considerably higher, but also considerably quicker...
Via do have some headroom to down clock the Nano and still achieve comparable performance. There's also the issue of the particular motherboard that was tested being a higher end model than the one Intel sells.

Re:Misleading title?

[Bert64]

The nano motherboard was also a higher end component than intel's one, it has pcie instead of pci and more sata channels...
If you were to clock the nano down to be performance comparable, and reduce the other features of the motherboard the power use would be closer.

The atom being an in-order design also means it would run faster with code more optimized for it, in-order designs just shift the work of optimization to the compiler.

What i'd like however, is a low power system with pcie to run mythtv, a system that has the capability to shut off the videocard when not in use... I don't really care about small, i want a couple of pci and a pcie slot

Re:Misleading title?

[TheRaven64]

This is why Intel is trying to invent the MID ('Mobile Internet Device') market segment. The only advantage they have over ARM is x86-compatibility. If you run a Free Software OS (Linux, *BSD, etc) on the machine then you don't care. If you run Windows, then you are going to be paying around 50% more in the UMPC segment for the Windows license, and this is really hurting Intel since Windows is the only OS that still depends on x86 (except the variant that Microsoft are pushing in small / cheap devices, Wince, which runs on ARM, MIPS and PowerPC too).

Re:Misleading title?

[Mattsson]

Since the platform on which the CPU runs have such a large impact on the total score, one would have to make actual power consumption tests on the whole system under a series of identical scripted runs and see which comes out on top.

Re:Misleading title?

[T.E.D.]

The VIA system uses 17.4 more watts while under load than the Intel system. That means if they are under load for one second, the VIA system has used 17.4 more watt-seconds than the Intel. On the other hand, the Intel only uses 0.9 more watts under load than the VIA uses when idle. ...
This means that to use the same amount of power over the long term, the VIA has to be able to complete tasks in ~5% of the time that the Intel takes.

Your math is good, but I don't think the logic is. Many devices with a user interface spends the vast majority of their time idling. Perhaps it doesn't go for an MP3 player. However, how much time does your cellphone spend turned on, but clipped to your belt (on in your purse, or whatever) waiting for calls to come in? I'd think power at idle would be a *huge* issue.

Re:Misleading title?

[LarsG]

I'm personally crossing my fingers for ARM SoCs based on the Cortex core to show up in MID/UMPCs before Intel gets Moorestown out the door. Cortex should be good enough in the performance department for most tasks one would use a handheld for, and should have a quite decent advantage in power consumption.

In this corner, we have the mean lean LinARM. While not the fastest contender out there, boy does he have stamina. Some claim that his Finnish ancestry will cause language problems, but others think that it is a non-issue.

In the other corner, we have the current heavy-weight champion Wintel. After his recent diet, does he have what it takes to take home the title in the flyweight division or will the Windows tax on his shoulders make him too sluggish?

Re:Misleading title?

[LarsG]

What i'd like however, is a low power system with pcie to run mythtv, a system that has the capability to shut off the videocard when not in use... I don't really care about small, i want a couple of pci and a pcie slot

The the Atom is probably a bad fit. If you intend to run both myth frontend and backend on the box, I'd expect it to simply be too slow. Not to mention that once you start adding pci/pcie cards and HDs, the power consumption advantage of the Atom compared to other CPUs like the Nano becomes insignificant compared to the power consumption of the rest of the system.

Re:Misleading title?

[mollymoo]

Are you suggesting the Northbridge does some of the work normally done by the CPU? I thought Intel just hadn't got a very efficient chipset yet.

Re:Misleading title?

[a.ameri]

A bit offtopic maybe but I couldn't let this one pass.
V4?

V4s are sometimes used in motorcycles, but I've never seen a V4 in my life. Wikipedia tells me that Lancia used to make some up until the '60s.

Nearly all 4-cylinder cars use the arrangement known as inline-4 or straight-4. Indeed the V arrangement has many downsides and some manufacturers (including BMW and Porsche) do not even use V arrangement in their 6-cylinder vehicles (Porsche uses the flat-6 arrangement, BMW a more simple Straight-6)

With 8-cylinder, the games changes a bit. There are probably as many straight-8s around, as there are V4s.

Assuming all engines are arranged in V is as much a sin in car enthusiast community as thinking everyone uses Windows is here.

Re:Misleading title?

[Anonymous Coward]

The numbers are not messed up!

You have to consider the power drawn by the system, not the cpu (the cpu can do nothing without the rest of the pc after all, and vice versa). If you add the power drawn by the rest of the components there is a much smaller difference. The VIA system draws about 4/3 the power, but finishes in 3/4 the time and so is more efficient (4/3 * 3/4 = 1).

As for the power drawn after finishing the task, well, you can use this for the next job while the atom catches up.

To be honest though I think it is meaningless to say which is better as it depends on what you need.

Re:Misleading title?

[Jorophose]

They offloaded more or less everything onto the chipset. All those fancy video decoding features are in the chipset, for one, and and there's a whole slew of other functions there.

Not to mention the Nano and C7 have Padlock built in them. Atom doesn't.

And it's far from the end. As soon as the ULV mobile edition of Nano is out, it's goodnight Intel. From what I recall when VIA releases an ULV model the power requirements drop by about 15%-25%. And then there's the VX800 that should be out about the same time. Here's hoping the mininote gets it, and it plays solidly. (and if Dell's keyboard quality is any good, then you also have an atom counterpart... at least this way you have competition)

What's odd is that it seems to consume more than the C7. Anybody want to comment? Are we just all using ULV C7 parts right now? Or was it because of the Raptor?

(What I meant in my other post, is regardless of chipsets the Nano can do that much)

Re:Misleading title?

[default luser]

Yes, Intel is selling the Atom with a craptastic 945GC chipset, which doesn't help things. Intel is still manufacturing the 945GC chipset on an old 130nm fab, and is practically dumping these chips for the sub-$200 computer market. This makes the D945GCLF dirt-cheap, but also much less efficient than the CPU itself would imply.

As I have posted earlier, you can pair-up Atom with many other lower-power chipsets. Options available include the mobile GM945, which features a TDP about 10w lower than the 945GC. Or if you want absolute low power, there's always the Poulsbo chipset (part of Centrino Atom), which sips power at 2.3w TDP.

I have to agree with the author on one thing: if Intel is strong-arming the industry and locking-down Atom boards, this is bad for all of us. I've been really looking forward to the low-power performance Atom could bring to the MiniITX market, but so far that hasn't materialized.

Re:Misleading title?

[hairyfeet]

You,know,I was wondering about that.If you go back and read the original press releases for the Atom before the whole netbook craze took off,the Atom was sounding more like a cell phone chip that had a little extra juice than an actual CPU. What I want to know is why everyone seems to be jumping on the Atom when it really isn't saving that much power compared to the Via. hell after seeing the performance I'd much rather have an underclocked Celeron M than an Atom! Maybe they are letting the OEMs have it dirt cheap. Personally when it comes time for a netbook I'll be looking around for either a Via or a Celeron,because the Atom just doesn't look good to me. But as always this is my 02c,YMMV

Re:Misleading title?

[universalconstant]

Not quite. They're using the full desktop chipset on that board. The Atom also works with the lower power 945GSE chip that's used by the Acer Aspire One, MSI Wind etc. You can't currently get a motherboard that has the lower power chipset yet.

Re:Misleading title?

[Anonymous Coward]

I think you forget one fact: when the VIA finishes earlier than the Intel, you have to take into account the difference between the power spent with the VIA on idle and the Intel on load until it finishes.

Re:Misleading title?

[Bert64]

Well, one pcie videocard for rendering the video, and we can shut down the videocard when the system isnt playing video.
One tv tuner card for receiving video feeds.
You would need these devices in any case for running mythtv... The videocard is likely to be the most power hungry, so shutting it down completely when not in use would be good. It should shut down when you aren't using the frontend, like when you put a purpose built dvr into standby mode.

Almost had me going there...

[Rayeth]

The atom doesn't use 4 watts under full load. It just used 4 extra watts. I knew that was too good to be true.

Re:Almost had me going there...

[mcvos]

The atom doesn't use 4 watts under full load. It just used 4 extra watts. I knew that was too good to be true.

Even so, I'm kinda hoping the processor doesn't pull more than a Watt while idle. Otherwise it'd be extremely wasteful compared to other processors.

Re:Almost had me going there...

[Anonymous Coward]

A watt while idle? As in a single watt?
Try 60.

Re:Almost had me going there...

[SpinyNorman]

They don't break out the CPU power usage - those figures are for the **entire computer**.

Idle Nano=59.2 Atom=56.4

Load Nano=77.5 Atom=60.1

If we assume that the bulk of the load vs idle power difference is due to CPU power usage, then we have the Atom using approx. 4W more under load, and the Nano using 18W more.

Whatever it's total power draw, the Atom is evidently much more miserly.

Re:Almost had me going there...

[idealego]

That's the difference in power consumption between idle and load of the ENTIRE SYSTEM, as is clearly stated at the top of the chart.

The TDP on the most power hungry atom is only 4 watts.

Intel needs to pair the atom with an efficient chipset. Unfortunately any of the chipsets Intel currently has available to pair with the atom look like power hogs next to the atom.

Re:Almost had me going there...

[Anonymous Coward]

Next time, try understanding the difference between a processor and a computer.

Re:Almost had me going there...

[Anonymous Coward]

The specs for Atom N230 say that it is indeed 4W http://en.wikipedia.org/wiki/List_of_Intel_Atom_microprocessors

Re:Almost had me going there...

[ameyer17]

60 watts? That's not that impressive. The CeleronM-based EEEPCs (70x, 900, and either the 904 or the 905, I'm too lazy to check) pull about 25 watts while charging[1].

[1]Shamelessly stolen from http://forum.eeeuser.com/viewtopic.php?id=22855

Re:Almost had me going there...

[Anonymous Coward]

Intel has a low power chipset: Poulsbo http://softwarewiki.intel.com/mid/Poulsbo. This is marketed together with Atom Z5nn series as Menlow platform for handheld devices (MID - Mobile Internet Device).

TDP for Atom Z5nn is even lower: 0.65 - 2.5W http://en.wikipedia.org/wiki/List_of_Intel_Atom_microprocessors#MID_processors_.28UMPC.29

Re:Almost had me going there...

[LWATCDR]

For a lot of users one of the low power Semperons would be a better choice right now. The problem is they are not as sexy as the Atom.

Re:Almost had me going there...

[RightSaidFred99]

Indeed, actually it does only use 4 watts under full load. The power problem is with the chipset, not the processor.

Re:Almost had me going there...

[Jherek Carnelian]

For a lot of users one of the low power Semperons would be a better choice right now.
The problem is they are not as sexy as the Atom.

That's because you spelled it wrong sounds like simperon, the real name is a whole lot more sexy - sempron .

Re:Almost had me going there...

[Mattsson]

I thought that they used desktop chipset parts due to cost and performance reasons, not availability of a compatible laptop chipset per se. =/

Re:Almost had me going there...

[Mattsson]

Yeah, the name alone makes one want it! 8-D

Up and Atom, Atom Ant!

Re:Almost had me going there...

[CrazedSanity]

Seriously... the article doesn't even indicate how much power the CPU itself is using; the last link that purports 4 watt power usage doesn't say anything of the sort. There's a 3 watt difference in the two systems for the entire system, which still shows 56.4 to 59.2 watts... I'm sure they're nice systems, but the whole "using 4 watts of power under load" is a bunch of marketing crap .

Re:Almost had me going there...

[mollymoo]

It's crap if you put it in a 60W system (which is just plain stupid for an Atom as it's not a desktop CPU), but if you put in in a 20W system it makes more sense. Put it in a 10W system and it makes a hell of a lot of sense.

Re:Almost had me going there...

[default luser]

But that is an incredibly poor assumption. The chipset bears some responsibility for that power increase at load.

Let me try to illustrate why:

Have you ever watched the temperatures of your system go up while under load? Sure, the CPU temperature climbs, but so does the chipset temperature - SIGNIFICANTLY. On a desktop board with a desktop CPU, the climb in chipset power at load is small compared to the CPU, and is often lost in the noise. But when paired with a low-power CPU like the Atom, the chipset power rise at load becomes significant.

Really, do you think that the sudden load of memory accesses and routing of data has no effect on the chipset's power consumption?

I firmly believe that the 945GC is responsible for half of that 4w increase, which puts the Atom right around the promised 2w TDP mark.

Re:Almost had me going there...

[mcvos]

Next time, try understanding the difference between a processor and a computer.

Although I also think 60 W for an idle mini-ITX system with low-power processor is a bit much. If every other part of the computer wastes that much energy, it's a bit silly to put this much effort in reducing the processor's power consumption.

I like the Western Digital Green Power drives, which cut power consumption for hard drives in half (from 10 to 5 W). Many graphics cards use enormous amounts of power, but you shouldn't put those in an energy-friendly mini-ITX system. But how much power does the motherboard use? And can memory be made more energy friendly?

The chart shows 60 watts

[iminplaya]

How do you get 4 out of that? It's that danged new math, I tell ya!

Re:The chart shows 60 watts

[pwnies]

The 60 watts is what the entire system is using. That includes the hard drive, the RAM, psu, mobo, etc etc.
They get the 4 watts from observing that when idle the system is at 56 watts, but when at a full load it meanders over to the 60.1 watt range.

60-56 = 4

Re:The chart shows 60 watts

[evilbessie]

There be magic afoot, next you'll be telling me 1 + 1 = 2, crazy. You should be careful, there are fragile minds on here who may not be able to take these lies.

Re:The chart shows 60 watts

[GrumpyOldMan]

Heck, I want to know how they managed to use as much as 60W at idle from hardware this skimpy. My dual core AMD 4850e based system with an M3A-H/HDMI motherboard, 4GB ECC memory, and 3 750GB sata drives registers 58W when idle, according to my kill-a-watt.

Uhmmm... measurements?

[autocracy]

It seems they may be measuring the whole system load in comparing the efficiency of the processors, which is more than a little unfair. What sticks out more, though are numbers like "63,434 watts". Uhmm... no? Besides being a clearly invalid measurement, it should probably be expressed at watt-hours. No way either machine drew 63 kilowatt hours either.

TFA is broken.

Re:Uhmmm... measurements?

[eddy]

>It seems they may be measuring the whole system load in comparing the efficiency of the processors

I think that's more fair as it's what's relevant to me as a customer of the end product; the computer.

Problem with the Atom side is that the chipsets used are crappy and use more than twice the energy of the CPU itself. So while the CPU might look great on paper, the actual products that use it does not have the fantastic battery times that you'd like (10h+)

These CPUs should be compared together with a viable chipset and memory subsystem combination. Add that up and there's the number you're interested in.

Re:Uhmmm... measurements?

[Lord Pillage]

It's probably Europe based. A lot of European countries use "," in place of "." and "." in place of ",". Thus, 63,434 watts = 63.434 watts which makes a little more sense.

Re:Uhmmm... measurements?

[david.given]

What sticks out more, though are numbers like "63,434 watts". Uhmm... no? Besides being a clearly invalid measurement, it should probably be expressed at watt-hours.

Well, no; a watt-hour is a measurement of energy, being one watt (a joule per second) applied over one hour. It's precisely equivalent to 3600 joules. If you want to measure power, that is, the rate of energy consumption, then watts are the right unit.

(Although I wish people would just use joules and joules per second instead; it would save all this confusion. At least it's better than the motor industry, who have decided that watts are insufficiently confusing and have opted for horsepower instead, each of which are equivalent to approx. 746 joules per second. Sigh.)

(And besides, the article actually said 63434 watt-seconds --- i.e., one joule per second applied for one second, i.e. one joule. So why they didn't say so in the first place I don't know.)

Re:Uhmmm... measurements?

[Anonymous Coward]

You don't think they mean 63.434 watts? The comma is used instead of a point in French for example.

Re:Uhmmm... measurements?

[muszek]

Don't know about others from the "a lot" group, but in Poland we use "," instead of ".", as in "Pi is approximately equal to 3,14159". Sometimes it can be a major PITA, when you get some data in one format and have to manually convert it to the other, so that it's recognized as a number.

We don't use anything instead of your use of ",", though. American "I have 1,000,000 pencils" would be "I have 1000000 pencils" or "I have 1 000 000 pencils".

Re:Uhmmm... measurements?

[LehiNephi]

[H]ard|OCP is actually based in Dallas, so the comma is indeed a thousands separator. However, the unit is watt-seconds (Joules), not watts, so it's still correct.

Re:Uhmmm... measurements?

[datapharmer]

no, I thought about that too, but common laptops use in the 60+ watt range, so if you are talking ultra-portable it would be much lower. A single core macbook pro for instance uses 19-55 watts, so david.given's explanation makes sense. If you take the 63434 watt-seconds / 3600 (watt hour) = ~17.6, which would be more in line with expecting a peak of 4 watts higher or ~21.6. I recall that the preview announcements claimed these systems would draw about 22 watts, and that is almost dead on, even with rounding error. That means if this processor were used instead of a core solo it would almost double the battery life (screen size, wireless, etc. aside).

Re:Uhmmm... measurements?

[rduke15]

Offtopic, but if s/,/./g "can be a major PITA", then you are in trouble indeed.

Re:Uhmmm... measurements?

[autocracy]

Well, my complaint was reading their comparison of total power consumption for the time of the task. Since the time required for different processors was different, I was expecting something equivalent to joules or kWH. It appears the article says "watt-seconds (joules)" when I checked it now, so they either fixed it, or I really messed up reading it in the first place.

For those commenting on the "," and "." being interchangeable, yes... but why read watts out to that precision? Indeed, with the comment on joules, they didn't.

However you look at it, the advantage of the implementation of the systems may have given 3.8% to the VIA CPU system, but we know the processor itself scored differently. I wish that had somehow be taken into consideration.

Re:Uhmmm... measurements?

[Anonymous Coward]

Actually the article states "63,434 watt-seconds (Joules)". While 'watt-seconds' isn't exactly a measurement in itself, they do show it as Joules, which is the correct measurement. They should have kept using Joules after that point.

TFA is not broken, perhaps you are. ;)

Re:Uhmmm... measurements?

[ogl_codemonkey]

If only there were some standard unit of measurement that expressed power over time... oh well, we'll just have to post bogus maths 'till one is invented.

Wait for the next generation

[eebra82]

If you're not in a rush to get one of the Atom/Nano based computers, wait for the next generation. Although both CPU:s are excellent in performance, the next iterations will bring two cores and far better efficiency.

The first generation of any product line is usually fairly rushed and experimental. That does not mean the product itself is bad, but we should expect a big jump from the next generation.

Re:Wait for the next generation

[ArsonSmith]

The current generation of any product line is usually fairly rushed and experimental. That does not mean the product itself is bad, but we should expect a big jump from the next generation.

fixed it for ya

Re:Wait for the next generation

[Anonymous Coward]

In other words, never ever buy a computer.

Re:Wait for the next generation

[pla]

If you're not in a rush to get one of the Atom/Nano based computers, wait for the next generation

Although the Nano represents a new chip series for Via, they have dominated the "ultra low power x86 PC-class" market for almost a decade now. I wouldn't, therefore, call it a fair comparison to call them both a first gen-chip. The Atom, yes. The Nano... To put it in terms more familiar to an Intel-dominated market, the Nano as a "first gen" more closely resembles the P3 vs the P2... Same basic core with a few modest improvements and running at a higher clock.

The FP post provides a simple example of my point:


that's 4 watts more than at idle, for about 60 watts total.

...While I currently run two VIA-based systems at home that consume notable less than that combined when running flat-out.

A truly low-power (yet entirely usable) system depends on more than just an efficient CPU. If your chipset and GPU and RAM each suck down more than the CP under load, you may as well splurge a bit on the Watts and go for a beefier CPU, because you'll never really see the difference in terms of battery life (or the AC equivalent, UPS runtime).

Re:Wait for the next generation

[0xABADC0DA]

The current generation Atom processor is fine. I have it running 64-bit Ubuntu right now and it's fast enough. The next generation may be worth waiting for, but this generation is decent.

The real problem with Atom at the moment is the mobo and chipset. The mini?-itx board I have from Intel uses 30 watts to turn on and 25 watts of that is the mobo minus processor (according to 4w rating). I know because the processor doesn't even have a heatsink and the whatever-bridge has a heatsink and fan. My advice: if there's a fan at all don't buy it. Same probably goes for laptops, unless they are advertising like 10+ hours then probably the chipset is using the vast majority of the power, so might as well get a faster processor if it's just going to piss away the power anyway.

The dual-core AMD desktop I have is, what, maybe 20 times faster at only say 3x more power (not sure how many watts the graphics card draws). The atom could easily be more like 10x more power efficient... then it would make sense.

Re:Wait for the next generation

[LarsG]

Yeah, the Atom itself is perfectly fine; in terms of performance per watt they are even in the ballpark of the ARM cores you find in most cell phones and other low-powered devices. What is missing for the Atom to enter that space is that an equally watt-efficient chipset and other components are not there yet.

That is an uphill battle for Intel, as the ARM manufacturers have long experience in making low-power complete systems - typically by putting the cpu core, RAM, memory controller, 2D/3D, USB, SD and all the other glue on a single chip.

Re:Wait for the next generation

[sardaukar_siet]

Wait for the next generation? You're telling that to this crowd? :D

Re:Wait for the next generation

[Anonymous Coward]

Relative to the load required by pc's servicing dedicated device roles, this is exciting even for first gen.

If the device makes a good mythtv playback platform, I can have another bigger machine do the encoding.
Right now i'm using a P IV machine as my firewall. Atom or Nano if they can push the same number of bits per second why wouldn't I switch? Payback on energy consumed may be slow, but I'm also looking at cooling any heat dissipated. That's a real issue when I'm sitting in sweltering Texas heat, there's a P IV at my feet blowing hot air and my window unit barely keeps it comfortable in here.

Assuming I don't just put the firewall in an unairconditioned part of the house, first gen of these devices is pretty compelling. Maybe I'm a special situation.

Re:Wait for the next generation

[Anonymous Coward]

two cores!?! sweet!
I'm going to wait for the Intel Molecule. H2 is going to be a flop and unstable so better wait for He2.

Honestly, if these Atom processors are more power efficient than the Quad cores and Core Duos, then we could see multi-processor arrays being more popular on low-end budget systems if we can get idle power consumption even lower.

conspiracy

[ILuvRamen]

And how does it stand up agains a 2.6Ghz Intel Quad Core? Seriously does anyone else see this conspiracy? They couldn't make any faster CPUs so they decided the best business decision was to start over at like 400 MHz and then pretend be making amazing speed increases all over again when really all they're improving is power efficiency a little. Maybe Microsoft should do the same thing and toss 7 out the window and make a Windows 3.1 clone and then go to a 95 clone and soon they'll be back at an incredible re-release of XP!

Re:conspiracy

[Bryansix]

Whatever they have to do to get back to XP is fine with me. In fact I was happy with 98SE for a LONG time there.

Re:conspiracy

[x2A]

yeah, a 2.6Ghz Intel Quad Core would do great driving my smart phone... those 5 minute battery times and second degree burns to the hand just make talking to people or connecting to my ssh server from my phone a pure delight!

Erm... different job, different tools???

Re:conspiracy

[Wiseblood1]

Yes but at least you wouldnt get you hopes up like people did with vista

Re:conspiracy

[Relic of the Future]

s/Insightful/Off-topic

When you don't know anything about the topic under discussion, it's better to just not say anything at all. Clearly I need to start doing meta-moderation again.

Re:conspiracy

[Anonymous Coward]

Not really. They're using new technology to re-invent low-end cpus. I'll bet the Atom and Nano processors are leaps ahead in power efficiency than a comparable clocked processor from a previous era.

To me it makes sense. Those fancy small transistors and design techniques are not just useful for your screaming 3GHz 130W processors.

Look at something like the ARM series. They haven't changed much [longer pipelines, few more instructions] over the years, but they have stepped down from 250nm to 90nm and below. All the while getting smaller and more power efficient. Nobody is arguing that.

So why is a 250nm to 45nm reduction in the Atom class of processors considered a money grab?

Re:conspiracy

[somersault]

In fact I was happy with 98SE for a LONG time there.

Seconded. 98 + Firewall (even just an incoming hardware firewall) was fine. If I'd had firefox all along instead of IE then I probably never would have had a virus in Windows at all (I only started using Windows around 1998). I only upgraded to XP because I wanted to play Lego Star Wars and it refused to install on 98SE (while stuff like Half-Life 2 was perfectly fine).

Re:conspiracy

[cyfer2000]

Wasn't the Core architecture based on Pentium III?

Re:conspiracy

[Orange Crush]

The driver model was wretched in Win98. Plus it was still DOS under the hood. No, going NT-based with XP was a good move on Microsoft's part. Now if only they had gone ahead with Neptune instead of Windows ME, they would have saved themselves a major embarassment.

Re:conspiracy

[XcepticZP]

Please, 98 had nothing on XP. At the time, to me, XP was like Windows 2000, with all the fancy admin features plus it had better support for games. But that's just me, everyone else probably has their own story about it. Either way, you sticking with 98"SE" for a "LONG" time before switching over to XP was a brainless thing to do, in retrospect. But at the time you would have been considered normal, because it was "totally in" to hate Microsoft's newest OS. Oh wait, the same thing is happening now...With Vista! Mark my words, one day we'll look back at this Vista-hating stage and call it brainless too.

Re:conspiracy

[somersault]

I don't disagree that going NT based was a good move, I knew that NT was the better OS, but the lack of USB and DirectX were the main issues! I think when 2000 came out its version of DirectX was backwards compatible with NT.. can't remember. Basing 2000 on NT was an awesome idea, and I don't think XP has much over 2000 than a bit of an interface redesign and built in support for wireless.. still a good OS though. How did MS manage to screw up so badly after starting to head in the right direction!?

I have never heard of Neptune, here's an interesting snippet from a discussion at http://www.flexbeta.net/forums/lofiversion/index.php/t2363.html :

Well, actually Neptune was a lot more than just Windows 2000 with a different logon screen. This was orginally a developer's release for something that was supposed to be much better. Neptune was not canceled. All you have to do is browse through some of the court testimonials available on Microsoft.com and you can clearly see that. In addition, Neptune has a feature called Activity Centers. I am sure all of you have seen screenshots of them. If you haven't, click here. For some reason, people kept thinking that once Bill Gates became head of Windows that he canceled the project and formed Whistler, but that is incorrect. Neptune's ideas have been around ever since Cairo times (technologies that were originally destined for Windows NT 4, the Windows 2000), but they simply could not be relased because of their complexity for the time. So, after Windows 2000 was released, MS released Codename Neptune, which was the starter for all of the echnologies. Unfortunately, for whatever reason, Microsoft deceided that they should tone down on the project and try to combine the 9x kernel with the more stable NT kernel. This is where Whistler (XP) came in to existence. As you can clearly see, MS used the login screen from Neptune for XP. Click here for a screenshot of the login screen from Whistler build 2250. Then of course, it went through beta and was perfected and Windows XP was released. All this time, however, Neptune was still being worked on. As time went on, MS used ideas like WinFS (a concept of MS's from 1993) and kept adding them to Neptune, jumping it from core to core (Windows 2000 --> Windows XP --> Windows Server 2003) until it arrived to where it is now: Core version 6.0. Yes, if you haven't guessed, this IS LONHGORN! Neptune is very much alive in Longhorn, as you can see, because one example is the Activity Centers that are integrated into explorer.

Sickening how Vista was hyped up so much and in the end was just a load of cruft and bloat :(

Re:conspiracy

[Lisandro]

That sound you hear is a joke. Above your head. Breaking the sound barrier.

Re:conspiracy

[pdusen]

obligatory whoooooosh.

Re:conspiracy

[Bryansix]

I think there is something to hating the latest OS release but with Vista it's different. With XP you could tweak it from day 1 to act exactly like 98SE except with a more stable kernel and protected mode drivers and all that shit. With Vista it's a whole new game. Vista makes me want to vomit. Mostly because powerhouse computers still take total craps with it. The Indexing service has gone to shit. They had MAJOR bugs like the File Transfer bug that just got fixed recently. Still File Transfers take longer then on XP. Why? Then the stupid implementation of UAC. Lastly the running of apps with them disappearing and NOT FRICKIN' ERROR MESSAGE! Nothing! Just click, run, nothing! Right-Click --> Run as Admin ... WTF! (pounds keyboard and screams inflammatory statements at Bill Gates) and nothing! WTF! I will ALWAYS hate Vista and the bloatware it brought to the PC. I hope 7 lets me turn that crap off and fixes the MAJOR bugs.

Re:conspiracy

[x2A]

How can a sound break the sound barrier?

Re:conspiracy

[x2A]

you know, if you make the sound, you break the joke...

Re:conspiracy

[Lisandro]

It has to go really, really fast!

Re:conspiracy

[Anonymous Coward]

Yeah, but imagine a beowulf cluster of those.

You could fry an dinosaur egg on the top of those cpus!

Re:conspiracy

[ILuvRamen]

But you could alter your voice with extremely processor intensive effects in realtime! hehehehe. Oh and play Crysis rofl.

Re:conspiracy

[savuporo]

Just a minor nit, but they went NT-based on Win2K already, not with XP

Re:conspiracy

[catmistake]

No, going NT-based with XP was a good move on Microsoft's part.

I don't mean to troll, (though evil moderators are stalking my comments, so I am prepared for the -1)but this statement seems bizarre and mixed up to me. As far as I can tell, all there is there is NT, not some innovative ground-up OS based on NT... it IS NT. I don't think any consideration was made one way or the other. Its just like saying: "No, going Cheetah-based with Tiger was a good move on Apple's part" or "No, going RHEL5-based with Fedora8 was a good move on Fedora's part."

Re:conspiracy

[pdusen]

I don't think I want to take advice on jokes from the guy who missed it. Thanks, though. ;)

Re:conspiracy

[Orange Crush]

^I was referring to the consumer line, which didn't ditch DOS until XP.

Re:conspiracy

[Orange Crush]

My statement was in the context of the DOS-based Windows line through ME. NT is a kernel. Moving the consumer line from DOS to the much saner NT kernel was a good move for Microsoft.

It was also a good move when Apple bought Next and used NextStep as the basis for OSX rather than the long-in-the-tooth MacOS "9 and under" codebase.

Re:conspiracy

[marcosdumay]

I duno. I guess just sliglty faster than ordinary is enough.

Re:conspiracy

[x2A]

It?! There's only one joke in the universe?!! You think maybe you might've missed all the others?

Re:conspiracy

[x2A]

Ironically my post got modded funnier than yours *lol*

Re:conspiracy

[catmistake]

My statement was in the context of the DOS-based Windows line through ME. NT is a kernel. Moving the consumer line from DOS to the much saner NT kernel was a good move for Microsoft.

Again, this seems mixed up... and I apologize if this sounds nitpicky... but wouldn't it make more sense to say they simply abandoned DOS-based systems,, and moved NT to the consumer line, rather than vise versa?
I just think your statement makes their work sound more impressive than it was, like marketing-speak.

Interesting.

[Blice]

Not long ago I only knew VIA as a chipset maker and a maker of chips for some other devices/things that weren't really "brand name" items.

But lately I see them getting more and more into the CPU business. They start pushing out their own motherboards, with their own processors, graphics chips, everything. Not only that, but their processors are beating Intels and their integrated graphics are on par with Intels!

I think it won't be long until we see some real desktop processors for other motherboards coming out- I mean VIA CPUs for ASUS, Foxconn, Supermicro, MSI, etc., competing at the same level with AMD and Intel... I think it's about time, too. We have this underdog who we buy because they're our favorites and not because they're better, still on 65nm processors while Intel has released 45nm and is getting ready to push out 35nm and 25nm- It's time we get some real choice. Maybe we can have a favorite company who also makes the better products. Choice is good.

Re:Interesting.

[0racle]

Not long ago I only knew VIA as a chipset maker and a maker of chips for some other devices/things that weren't really "brand name" items. But lately I see them getting more and more into the CPU business.

Would you feel more comfortable with it if they went back to calling their CPU Cyrix?

Re:Interesting.

[Sockatume]

Engadget (or was it Gizmodo?) made an interesting observation that it's "pin-compatible" with existing C7-M devices, so a whole lot of currently-dodgy VIA-based mini-notebooks, like the HP Mini-Note (which desperately needs to run cooler and longer) can make the transition very smoothly. Well played!

Re:Interesting.

[MBGMorden]

Via's been churning out processor's for years now - it's just that they don't target the average american desktop user. They make slower, cheaper, low power chips that were traditionally intended for very low cost computers sold in Asia (where in some of the poorer economies performance isn't so important as getting the cost down as low as possible), and now in mobile devices where the power consumption is a big issue.

I don't think you'll see Via competing with Intel and AMD in the mainstream desktop business anytime soon - that's a real but increasingly less important venue. People are replacing their computers less often these days (desktop processors have been "fast enough" for a good while now), and there is a big focus on mobile devices these days.

It certainly would be nice to bring back choice in desktop processors though. I remember way back when the Pentiums were on the scene, a customer generally had a choice between Intel, AMD, IDT, Rise, or Cyrix (sometimes marked IBM) processors. Further back in the 486 days you didn't have Rise or IDT but TI made x86 processors back then. The upside too was that back then, everything used the same motherboards/sockets, so going from one chip maker to another was trivial.

Re:Interesting.

[PIBM]

Well, in '99 VIA bought Cyrix, which used to make reverse engineered 80386/80486 in the old time, so I guess it was to get out of via by now :)

Re:Interesting.

[wild_quinine]

Not long ago I only knew VIA as a chipset maker and a maker of chips for some other devices/things that weren't really "brand name" items.

Methinks you are a young man, sir. VIA used to be much more relevant than they have been in the last few years. However, it is very pleasing to see them picking up again and standing against the competition.

Re:Interesting.

[AmiMoJo]

They make slower, cheaper, low power chips

Cheaper? Have you seen the price of EPIA boards lately? Atom boards seem to be a lot cheaper.

That was always the deal breaker with Via for me. The extra cost far exceeded any potential savings in electricity until you hit the five year on-time mark, compared to say an underclocked/undervolted Sempron or A64. It gets far worse if you want one of those expensive little EPIA cases and PSUs to go with it.

Re:Interesting.

[LarsG]

Or Centaur WinChip, rather. VIA bought Cyrix too, but from what I remember they never manufactured any if the Cyrix CPUs.

Re:Interesting.

[Dusty101]

Mod parent up. As a neglected fan of the Apple 12" Powerbook (MacBook Air? Bah. Gimme a break...), the only thing that stopped me plunking down cash for a pretty HP Mini-Note running Linux/BSD was knowing that it had a wimpy processor that was pin-compatible with the new Isaiah chips... So come on, HP - go for it.

Anyone at Apple reading this feel free to pass this oft-repeated chorus on to your management: we wanted a 12" MBP. Oh yeah, & your 15" MBP's peel as well...

Re:Interesting.

[Daengbo]

Introducing ... the Via Cyrix III (C3)

Re:Interesting.

[LarsG]

I stand corrected (but in my defence, it was only for a short time though).

VIA has become a real threat to Intel

[Animats]

There's a clear future in little laptop machines that don't cost much and don't use much power, yet are powerful enough to do most of the things most users do with laptops. The low end x86 CPUs are finally good enough to power such machines.

The laptop manufacturers had a Detroit mentality of "more computer per computer". This kept laptop prices up and margins high. But, as it turns out, cramming enough CPU power into a laptop to run wind tunnel simulations isn't what users really need. Especially when the network connection is the bottleneck anyway. The actual uses for a 4-CPU laptop are somewhat limited.

The flood of low-cost laptops has just started. The EE PC set off a race for the bottom. In a year or two, laptops will come in blister packs at the drugstore, in the section with the calculators, electronic dictionaries, and other office supplies. From here on, it's all about lowering margins. Intel and Microsoft will be squeezed hard on price.

Re:VIA has become a real threat to Intel

[SaDan]

In a year or two, laptops will come in blister packs at the drugstore, in the section with the calculators, electronic dictionaries, and other office supplies.

This is pretty much already happening now that Target is selling EEEPC systems.

Re:VIA has become a real threat to Intel

[Tweenk]

The low end x86 CPUs are finally good enough to power such machines.

The story from yesterday hints that the outcome may be that neither Via, Intel or AMD wins, but the dirt cheap MIPS (or Loongson) chips from China which already have good power efficiency... Those machines would probably be running Linux anyway due to their limited hardware, so the choice of processor architecture is a minor concern.

Re:VIA has become a real threat to Intel

[Omestes]

I think there is room for different markets here. I'm typing this on what I consider a "desktop replacement" laptop. Its big, its heavy, and it hardly ever leaves the desk, but it is more portable than a PC, and I take it to my girlfriends on weekends, and occasionally to the coffee shop. I basically use it for everything I used to use my (defunct) desktop for, but its less bulky, and more portable.

I still want to get a nice EEPC though, for a completely different reason.

Two different beasts.

Re:VIA has become a real threat to Intel

[psychicist]

I was looking at Intel's Atom and Via's Nano as viable processors for low-energy but relatively high-performance systems, but it's clear to me now that Loongson 2E/2F based systems will have a considerable power consumption advantage whenever x86 is not a requirement, not to mention that it may outperform both of them when compiler optimisations hit future releases of MIPS distributions (such as my own, an inofficial Slackware port).

Re:VIA has become a real threat to Intel

[Zebedeu]

Same here.
I have a super powerful 17" laptop because I need the power for work and the mobility for presentations and what not (it's also nice to have power when I need it at home), but for personal use I'm considering one of those low-powered laptops because frankly I'm getting tired of lugging this beast everywhere just to check my email and browse Slashdot.

The only reason I don't have one yet is that the current crop are almost there, but not yet there. I think the Dell might be the one. The Acer one and the MSI are also interesting, but the Asus is a little bit too expensive, at least in Europe.

It's a tie

[Sockatume]

The 1.8GHz Nano setup gets about 25% more performance than the 1.6GHz Atom setup. However the Nano setup uses about 75W under load, while the Atom box uses about 60W. That's about 25% more power consumption/heat output. I imagine an Atom and a Nano setup of equal performance would use equal amounts of juice, or in other words this is a tie in terms of work-per-joule, which is what we're after in mobile processors.

What really bothers me is that the Atom setup seems to use as much power idle as under load. What's going on there? Did the benchmarkers forget to switch on power management or what?

Re:It's a tie

[Anonymous Coward]

Chipset sucks it all up.

So... does any of these reviews benchmark the Via against the Atom in AES-256 performance with a padlock enabled openssl? :-)

Re:It's a tie

[Anonymous Coward]

The current Atom chipset is an old one. There is supposed to be a new chipset, specifically designed for Atom, coming soon.

Re:It's a tie

[TopSpin]

I'd say it's rather far from a tie when you consider more than simply power consuption.

As the story points out Intel is restraining board designers from using desirable technology on the Atom platform. No PCI Express, no DVI, no second memory slot. Theory is "Intel appears to fear Atom will cannibalize its Celeron sales". Perhaps. I'll bet VIA is more than willing to cannibalize those sales if Intel is going to let them.

The reference board in this review is nice. There are two ethernet phys, one of which must be gigabit. Compact Flash, mini-PCI and PCI Express. Damn. I like that board. That is the perfect board for the small, quiet home server.

Re:It's a tie

[nxtw]

The 1.8GHz Nano setup gets about 25% more performance than the 1.6GHz Atom setup. However the Nano setup uses about 75W under load, while the Atom box uses about 60W. That's about 25% more power consumption/heat output. I imagine an Atom and a Nano setup of equal performance would use equal amounts of juice, or in other words this is a tie in terms of work-per-joule, which is what we're after in mobile processors.

The Intel board could be a lot better if they used mobile chipsets. I have an old Shuttle desktop system based on Intel mobile hardware (915GM chipset.) With a 2.13 GHz Pentium M, the system idled around 35w and peaked just over 51w.

Re:It's a tie

[naasking]

I've read that the power guzzling northbridge+southbridge on the Nano's reference motherboard is the cause of the majority of the power consumption. I think they need a better reference board to truly showcase the Nano's abilities.

Re:It's a tie

[hellwig]

What really bothers me is that the Atom setup seems to use as much power idle as under load. What's going on there? Did the benchmarkers forget to switch on power management or what?

According to Tom's Hardware, this is a problem with the outdated, innefficient Northbridge they use on their board. The board itself uses a lot more power than the Atom processor, so whether or not the processor is running, the board is sucking up a lot of power. This is why VIA has always been in the embedded market, cause they aren't just focused on their processor. The Nano requires more power as a processor, but the entire system doesn't use too much more because VIA makes very efficient controllers.

I remember back in the day when I wouldn't buy a motherboard without a VIA controller. The sytems have become so specific these days (AMD vs. Intel, nVidia vs. ATI) I guess they just couldn't stay in the consumer market without paying out their nose for licensing rights to the various technologies. Personally I like my C-7M powered Cloudbook.

Re:It's a tie

[ceoyoyo]

The power measurements are for the entire system. So if the Nano is getting 25% more performance using 25% more OVERALL energy, then it's getting it's ass kicked in the work per joule category.

The reason the Atom doesn't use much more power under load as unloaded is looks like it's because it doesn't use much power in either situation. The Nano appears to use lots of power when it's working.

Re:It's a tie

[tolgyesi]

As far as I heard (about CMOS), the frequency/watt may not be linear. If the Nano is underclocked to be at the same performance as the Atom, can it be more economical?

heh 32,000 watts of power??

[Anonymous Coward]

someone seriously failed out of engineering school.

from TFA:

"For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watts of power "

that's priceless.

morons.

Re:heh 32,000 watts of power??

[Vigile]

Yeah, sorry, that's been changed.

"For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watt-seconds (Joules) of energy"

Better?

Re:heh 32,000 watts of power??

[Anonymous Coward]

It's called a frackin Joule Science get.

Re:heh 32,000 watts of power??

[somersault]

I think he was more bothered about the comma. As others pointed out above, it's a cultural thing. Some places in Europe apparently use comma instead of period. Only some though, here in the UK we use period. How confusing.

Re:heh 32,000 watts of power??

[treeves]

No, they just integrated the power it took to do the task over the time it took to do the task, to give a more useful number than a straight average might have done. 37W for 1000s (37000 J) is better than 30W for 2000s (60000 J), even though 30W is 37W.

Re:heh 32,000 watts of power??

[treeves]

It ate my "less than" sign.
It should have read "30W is less than 37W".

Re:heh 32,000 watts of power??

[somersault]

Yeah I didn't say that it wasn't, but I see what the original poster was getting at now it should just have been joules rather than watts whenever time is involved.

Re:heh 32,000 watts of power??

[marcosdumay]

< leads to <

More benchmarks and analysis here - HotHardware

[MojoKid]

More benchmarks and analysis here: http://www.hothardware.com/Articles/VIA_Nano_L2100_vs_Intel_Atom_230_Head_to_Head/

Re:More benchmarks and analysis here - HotHardware

[eddy]

The Vantage AES test doesn't seem to use the AES instructions on the Nano.

That's like having a 3D Graphics test not use a 3D Graphics API. Pretty worthless. If you're a geek and you buy a CPU with padlock, you are going to use padlock-aware encryption an hashing libraries/applications. I know I do.

It's soooo frustrating seeing review after review missing this.

Another review at arstechnica.com...

[IYagami]

...with the same findings.

http://arstechnica.com/reviews/hardware/atom-nano-review.ars

Re:Another review at arstechnica.com...

[keithjr]

Actually, the Ars findings were quite different. The Atom seems to be paired with a very power-hungry chipset, which dwarfs the processor's power draw. So, at the board-level, the Nano won on both performance AND power consumption!

x86-64?

[VincenzoRomano]

Besides having more cache and higher FSB freqency, the Nano is also x86-64!
It's a nice thing indeed, but it'd be like comparing apples and pears.

Re:x86-64?

[Svenne]

And the Atom has EM64T.

What?

Re:x86-64?

[VincenzoRomano]

Like comparing apples and pears or, at your will, a Motorola 68000 and an Intel 80386. They are different, running different machine codes.

Re:x86-64?

[The Warlock]

x86-64 and EM64T are the same instruction set.

Re:x86-64?

[LarsG]

Nope. Green and red apples would be more like it.

For most intents and purposes, AMD64 and EM64T is the same instruction set. You are probably thinking of IA64 (Itanium).

Re:x86-64?

[bestinshow]

Intel claim Atom has EM64T.

But it's been disabled on every shipping Atom so far. Some of the reviews have been pointing this out as a disadvantage limiting Atom boards to 3GB of memory - except that the shitty northbridge that they use with Atom only supports 2GB anyway.

Basically, for a home computer that will be plugged into a wall, don't get an Atom based system - the Nano should be the minimum to get because it supports 4GB in the chipset. The mini-ITX Nano boards will also support PCI-e as well, unlike the Atom boards (Intel specifically disallows this).

Performance for full day battery life

[anss123]

The Atom is a bit on the slow side. Like using ARM chips for desktop computing; so why not simply use an ARM chip?

Still, if the Atom's paired with a super low powered chipset we might just finally have computers with more than 8 hours of battery life (while still being affordable/portable/small). Imagine taking your computer to work, and then leaving it on all day. A small detail, but makes a big difference.

The Nano is faster, but it also use about 8 watts more power (according to HardOCP). Those 8 watts is a big deal when it comes to battery life, but OTOH Atom is quite a bit faster than even the fastest Atom. The difference being big enough that HardOCP stated that Vista on Nano was notably more resposive - notable enough to be picked up on in blind tests.

So perhaps Atom trades off too much performace...

Re:Performance for full day battery life

[Anonymous Coward]

The Atom is a bit on the slow side. Like using ARM chips for desktop computing; so why not simply use an ARM chip?

x86 compatibility reasons?

Re:Performance for full day battery life

[NerveGas]

I dunno... running CentOS on an Atom isn't terribly slow or disappointing. I could get by with it quite well if it meant 8-hour battery life.

Re:Performance for full day battery life

[Anonymous Coward]

x86 compatibility reasons?

Now why is this important today? I don't want or need x86 compatibility. Maybe it used to be the only game in town when Microsoft dictated CPU architecture but that isn't the case anymore.

What I want and need is power-sipping CPUs like MIPS or ARM in a laptop. Does anybody know what the hold-up is on this front?

Re:Performance for full day battery life

[suggsjc]

but OTOH Atom is quite a bit faster than even the fastest Atom.

In that case I'll take the fastest Atom, which is faster than the fastest Atom, which is faster than the fastest Atom....

So long as the power consumption holds constant, after a few iterations this thing should be able to run circles around even the BlueGene/L...all in a sub-compact notebook/UMPC!

Re:Performance for full day battery life

[adolf]

Not to be Captain Obvious here, but:

Does anybody know what the hold-up is on this front?

Windows and other boxed, retail software.

(In case you haven't heard, the war is not yet over. Your mother's new Dell laptop from Wal-Mart came with Vista, not Linux, and your income tax software won't run with an ARM CPU. Also: Photoshop.)

The 4 watts comment

[Vigile]

For those of you interested, the Atom CPU really DOES use just about 4 watts at load. The 60 watts number is for the entire system including power supply, motherboard, DVD-ROM, hard drive, etc. Idle power on both of these parts is measured in milli-watts so you can see how much power each uses under load by looking at the power consumption graphs on page 8:

http://www.pcper.com/article.php?aid=597&type=expert&pid=8

The gap between rest (on the far left) and load (middle) is much greater for the VIA Nano processor than the Intel Atom - in fact you can barely tell the Atom processor has changed wattage at all.

Re:The 4 watts comment

[NerveGas]

Eliminating the hard drive and optical drive, the Atom board still draws 40-50 watts. On my board, I observe a 55-65 watt draw from the wall with just the board, which taking PSU losses into account, is about right.

The part that it entirely irking is that that board alone draws more than my entire laptop, which includes a Core Duo, hard drive and a *display*, for crying out loud. Not such a great way to show off a low-power CPU.

Re:The 4 watts comment

[the_humeister]

Which is kind of intersting. I bought the previous generation of this type of mini-ITX board (D201GLY2) that came with the Celeron 220 (Conroe-L based and a SiS chipset for some reason), and it draws 40 Watts at idle (with only a harddrive attached). They're definitely pairing the Atom with the wrong chipset if the power draw is that high.

Re:The 4 watts comment

[Anonymous Coward]

Yes, but that's because Intel moved a lot of logic to the northbridge. Notice the honking great big heatsink and fan? Atom's a good processor, but the whole setup is so 'backwards' it really does look like they're trying to fiddle the performance/Watt figures.

Re:The 4 watts comment

[bill_mcgonigle]

Eliminating the hard drive and optical drive, the Atom board still draws 40-50 watts

Really? Is this being measured accurately? I've got a Via C7 board here that I've measured at 22W and that's got a spinning fan and everything. The Atom is supposed to be a step forward on efficiency from all I've read.

Re:The 4 watts comment

[NerveGas]

It's not the Atom, it's the northbridge. The DDR2 memory controller and built-in video are from an older desktop chipset, and are not power-optimized at all.

Those numbers are from me plugging in ONLY the motherboard, and measuring power draw from the wall, which was 55-65 watts. Factoring in losses in the PSU, 40-50 watts for the board is pretty close for the actual draw.

I have a C-something machine (I don't recall if it's an Eden or a Nehemiah) where the CPU, board, NIC, two hard drives, and optical drive together pull 65 watts from the wall under load.

The problem

[John Meacham]

The problem is that they don't know how much of that idle load is due to the CPU. You don't know whether that 4 watt difference is due to an efficient high power mode, or a really inefficient low power mode.

for instance, the nano might be 1 watt unloaded and 19 watts fully loaded, while the atom might be 20 watts unloaded and 24 watts loaded. This is clearly not the case, but would be consistent with the results.

Re:The problem

[Anonymous Coward]

> while the atom might be 20 watts unloaded and 24 watts loaded
Just about right, the i945 Northbridge has a TDP of 22W, which kills the Atom's power efficiency.

Re:The problem

[LarsG]

Atom doesn't pull 20W at idle, TDP on the Atom cpu in that benchmark is afaik ~4W. Which means that pretty much all of that power drawn at idle is from the rest of the board.

One of the major culprits would be the i945 chipset, which comes in at about 20W. The rest would be RAM, disks, etc.

If we only look at the CPUs, notice that the Nano pulls about 5 times more wattage than the Atom but only manages to be 20-30% faster.

What this benchmark really shows is that putting a low-wattage CPU in a system is kinda silly if the other components in the system are watt-guzzling pigs.

It also means that a total system with an Atom and proper low-power chipset/ram/etc will draw about 16W less than a Nano paired with the same low-power components. Which means that the Nano simply can't be used in MIDs, while the Atom will do quite fine in that form factor when Intel gets a decent low-power north/southbridge out the door.

Re:The problem

[CrazedSanity]

It is a statistical probability that 80% of all statistics are 50% questionable, 30% hype, and 20% lies, while the other 20% are only 50% truthful. There is a 50% margin for error in these statistics.

Re:The problem

[MoxFulder]

Exactly right. This review is almost useless to anyone trying to design a complete system around these processors, since it can't identify which specific components are responsible for the greatest share of the power consumption. Quite disappointing.

According to wikipedia, the hyperthreading Atom Diamondville (at 1.6-2.2 GHz) has a TDP of 4W. Which suggests that its low-power mode is very efficient--if TFA is correct--given that they only saw a 4W increase in power consumption under load.

OTOH, the 2GHz Nano has a TDP of 25W.

IMHO, it is a mistake to think that the Nano and the Atom are really fighting for the exact same market. This seems to be based on the fact that they are both available on Mini-ITX boards. It seems to me that Atom is an extremely stripped-down processor designed for extremely low-power use... while the Nano is actually more "beefed-up" than previous Via processors like the C7... featuring more complex execution units and the full 64-bit instruction set.

With only 4 more watts...

[kungfugleek]

I could have had a first post, too. :(

Wrong Benchmarks??

[weaver4]

The real performance for this market should be: Processing Power per Watt and Processing Power per Dollar. Not which one has the most raw Processing Power.

explaination of energy efficiency

[Sebastopol]

The article writers don't seem to be very technical guys.

If we look at the energy efficiency, Atom hands Via it's bilblical arse.

Let's talk about Joules, the things a battery stores. Batteries are important. Especially for the target marget of these two CPUs. In fact, battery life is most likely THE most important factor in anything below a notebook.

Keep in mind a battery only has so many joules between charges, that's obvious, I know.

Now, an efficient architecture would only use as many joules as needed to get the job done in a timely manner. Joules per seconds are Watts, btw.

So lets look at how these two stack up in terms of Joule consumption and Performance based on this data...

The VIA requires about 17W of power to chug through MP3 encode, for about 460 seconds. That means the power supply had to deliver 17 * 460 = 7,820 joules.

Now the Atom crawled along 30% slower, about 600 seconds to complete. But it only needed a delivery rate of 4 J/s, so it ate 2,400 joules.

So for a 30% improvement in performance, VIA had to gobble down MORE THAN THREE TIMES the energy!

That means you could encode 3x as many MP3s on an atom, but it will take 30% longer. Imagine if this was an iPod. Who would trade 3x less battery life for such a tiny bump? That isn't something to brag about when you are targeting a market starving for battery life.

I'll be really surprised if Via goes anywhere other than a few cheap Asian design wins.

Re:explaination of energy efficiency

[CannedTurkey]

And yet, in the PCper review that did the exact same comparison for mp3 encoding and cinebench encoding, the Via used 2.5 - 3 % less energy...

Re:explaination of energy efficiency

[PetiePooo]

The VIA requires about 17W of power to chug through MP3 encode, for about 460 seconds. That means the power supply had to deliver 17 * 460 = 7,820 joules.

Now the Atom crawled along 30% slower, about 600 seconds to complete. But it only needed a delivery rate of 4 J/s, so it ate 2,400 joules.

And now for the elephant in the room: Why are you encoding MP3s while you're running on batteries?

The normal workload for an ultra-portable running on batteries is not producing MP3s. Its outputting static screens to a beamer during a presentation, or surfing the "series of tubes" via the local WiFi spot between classes (or waiting for the plane). If you've got numbers or MP3s to crunch, save it for when you've got the thing plugged in, silly!

Assuming the average CPU loading is going to be a paltry 2-3% (essentially idle), you're looking at a very capable system using an average of somewhere around 61W vs. a less capable system using only 56W. The difference in battery time between those two is barely significant, which makes the increased potential performance of the Nano a big win in my book. When plugged in, the Nano is a passable desktop replacement. The Atom, not so much.

Re:explaination of energy efficiency

[smussman]

I know this is /. and we're not supposed to RTFA, but to quote:

For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watt-seconds (Joules) of energy while the Intel Atom processor used 38,290 watt-seconds (Joules) of energy. That is a difference of just 2.5% indicating that even though the Atom processor is slower, it's not that much less efficient than VIA's Nano.

And later on the same page:

Using the same method to gauge the results of our CineBench 10 test, we find that the VIA Nano used 63,434 watt-seconds (Joules) of energy to render the scene while the Intel Atom used 65,893 watt-seconds (Joules) of energy - an advantage of 3.8% to the VIA CPU.

Re:explaination of energy efficiency

[cookie23]

I'm not sure where your getting this data but according to the article (http://www.pcper.com/article.php?aid=597&type=expert&pid=8) they do actually do as you suggest and measure the Joules:

For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watt-seconds (Joules) of energy while the Intel Atom processor used 38,290 watt-seconds (Joules) of energy. That is a difference of just 2.5% indicating that even though the Atom processor is slower, it's not that much less efficient than VIA's Nano.

Via's Nano may use more power under load, but it finished fast enough to actually save energy over the Atom.

Re:explaination of energy efficiency

[Bigjeff5]

Your measurements are correct if you look at the CPU and nothing else.

However, computers aren't made of just CPUs (at least not yet), so you must factor in the entire load of the computers, including the CPU, to get an accurate measure of power usage.

When you do that, guess which one wins? That's right, the VIA does, with about 3k fewer joules on an MP3 encode. AND it did it 30% faster.

Which would you choose? A slower machine that uses insignificantly more battery power? Or a faster machine that uses insignificantly less?

The time when the Atom will be a clear winner is when system power usage at idle drops to 25w or so. At that point the Atom, with its 4 extra watts, will only be using roughly 20% more power under full load, while the VIA, with its 17 extra watts, will be using around 80% more power. The tradeoff of 30% less power for 60-70% more battery life would be much easier to make than the current 30% less power for +-1% battery life.

So, while it's a step in the right direction, if it requires a power-hungry chipset then the advantage of the Atom completely disappears.

Re:explaination of energy efficiency

[rthille]

You have a valid point, but the article makes the point that the entire system for the Atom process consumes more energy encoding the MP3, because while the Atom is a more efficient processor, it's power use is a small part of the overall system, so having the system 'on' longer to do the MP3 encoding will consume more energy over all.

Re:explaination of energy efficiency

[Anonymous Coward]

Did you fail to notice the following paragraph?

"For our MP3 encoding test, the VIA Nano processor used a total of 37,323 watt-seconds (Joules) of energy while the Intel Atom processor used 38,290 watt-seconds (Joules) of energy. That is a difference of just 2.5% indicating that even though the Atom processor is slower, it's not that much less efficient than VIA's Nano."

Of course, PC Perspective's power consumption is based on the draw of the whole unit, NOT just the extra draw under load. You can't shut off the rest of the machine and just use 4 watts to encode an MP3. Let's think of it this way, taking into account the entire draw, you can turn on the NANO to encode a MP3 and shut it off 460 seconds later, whereas you'd have to leave the Atom on for 600 seconds. 460*77 600*60. Of course, you can still argue that in a passive state the Atom is more efficient, but the difference is not dramatic. While it's easy to say that the Atom consumes # of watts less than the Nano you can only get a rough gauge of CPU performance of one over the other.

Re:explaination of energy efficiency

[Anonymous Coward]

So lets look at how these two stack up in terms of Joule consumption and Performance based on this data...

The VIA requires about 17W of power to chug through MP3 encode, for about 460 seconds. That means the power supply had to deliver 17 * 460 = 7,820 joules.

Now the Atom crawled along 30% slower, about 600 seconds to complete. But it only needed a delivery rate of 4 J/s, so it ate 2,400 joules.

I don't know man, but getting an MP3 to encode with only the processor seems kinda impossible to me (i cannot manually wire my breadboard to the proc so i can load an mp3...). So what we need is... A motherboard, an DVD-drive to read the audiofile and a storage-device to store the created MP3. If we take the power draw of the whole systems as presented by PC Perspective, then the VIA-based system uses 77 joules per second over 460 seconds -> 35.420 joules. The Intel based system uses 60 joules per second over 600 seconds -> 36.000 joules. So if i am using my system for mp3 encoding, i am using less joules per mp3 and it takes me less time on the VIA-based solution. So on this perspective, the Atom isn't better, and to use your terms, VIA pets the Atom mildly on the buttcheeks.

Re:explaination of energy efficiency

[Anonymous Coward]

You do realize you need a lot more than a CPU to have a computer, right?

Redo your numbers up there with complete system load; we're talking 60.1W for the atom and 77.5W for the nano.

In fact if you would have RTFA you would have seen that the nano actually beat the atom on total power consumption by 2.5%~3.8%. The nano might spike at a higher temperature but it actually uses less overall battery power when you consider performance vs joules.

Re:explaination of energy efficiency

[Anonymous Coward]

the graphs in the article give the opposite impression though.

Re:explaination of energy efficiency

[Anonymous Coward]

Sure, you can calculate that on paper, but paper os not the real world. You can't just calculate the CPU alone, because CPUs cannot function alone. You need to take into account the Northbridge, which is something you aren't doing, and you accuse the writers of not being very technical. On initial glance at the graph it does appear that the Atom is more energy efficient, yet when they calculated the entire energy usage of the system for the length of the task, the Atom lost out.

It is also important to realise that the two chips are different fab technologies: 45nm vs 65nm. Once we see the Nano fabbed at 45nm (and hopefully the northbridge too), we will see a much more competitive performance.

Re:explaination of energy efficiency

[LarsG]

That number is misleading. It compares the complete platform instead of looking at the CPUs in isolation, and it also ignores power consumption at idle.

Compared to the Atom system, the Nano based system use 2.8W more (5%) when idle and 17.4W (29%) more when under load.

The Nano is roughly 30% faster than the Atom.

Which means that if you power on the system, encode a bunch of MP3s and then instantly powers the system down then the Nano-based PC will use slightly less total power than the Atom-based PC.

However, most PCs spend a considerable time at idle - most of us don't turn the PC off when the CPU load is less than 100%. And since the Nano PC use more Watts than the Atom PC at idle....

If we compare the CPUs only, the Atom completely crushes the Nano; it is more than 3 times better in terms of performance per watt. The only reason that the PCper benchmark shows them as roughly equivalent is that the rest of the components used in the PCs are watt-guzzling hogs. Once Intel comes out with decent low-power chipsets for the Atom (as opposed to that 945G that consumes a silly 20+W), the Nano will not compare so favourably any more.

Re:explaination of energy efficiency

[LarsG]

Actually, the only thing that this benchmark proves is that putting a hyper-efficient Atom CPU in a PC which has power-guzzling components like the 945G (20+ watts, for crying out loud!) is silly if your goal is to get good performance per watt.

Re:explaination of energy efficiency

[ceoyoyo]

Or watching a DVD on an airplane... oops, now you're actually doing some work, constantly, for a long period of time.

Re:explaination of energy efficiency

[Anonymous Coward]

Your math is off. You have to look at total power
consumed of the system. Go back
The nano is consuming ~77 * 460 = 35420 joules
The atom is consuming ~60 * 600 = 36000 joules

So during the compute cycle the nano actually uses
less energy!

Re:explaination of energy efficiency

[LarsG]

But that is only because they decided to fill their test PCs with powerpig components. If the benchmark was intended to compare low-power Nano and Atom PCs, they did it all wrong.

For example, the 945G chipset is like 20+W. They stuck a 9+W-at-idle high performance HD in there, instead of something like a 1W-3W laptop drive. It is like the entire benchmark was set up to mask the fact that the Atom is 3+ times better than the Nano in performance per watt.

Re:explaination of energy efficiency

[PetiePooo]

Or watching a DVD on an airplane... oops, now you're actually doing some work, constantly, for a long period of time.

The big energy user there is the dvd drive. The main use for the CPU is CSS decrypting and shuffling data around, as the MPEG decoding is done in the graphics chip.

BTW, that's why I rip it to the HDD first then play it from there when I'm running on batteries. Don't tell the MPAA... *wink* *wink*

Re:explaination of energy efficiency

[ceoyoyo]

Depends on the graphics chip. But to take that out of the equation, let's say you're watching a DivX movie. Don't tell the MPAA. No DVD drive, no graphics chip helping you out, and a pretty processor intensive codec to decode.

Of course, I NEVER do that, but I've heard some people do....

Re:explaination of energy efficiency

[Frol]

You can't compare just the energy usage of the CPU and then draw the conclusion that you can encode 3x as many MP3s using that system. A more fair way to compare them are the total system energy usage:

The VIA requires about 77.5 watt for 460 seconds to encode the MP3s, 77.5 * 460 = 35,650 joules.

The Atom needs 60.1 watt for 600 seconds or 36,060 joules.

This means that if all you were doing with the computer was encoding MP3s, you could actually encode slightly (very slightly) more MP3s on a VIA system.

Just my 5c to keep the comparisons fair.

Re:explaination of energy efficiency

[bestinshow]

And since when isn't the platform a required part of a computer?

A CPU on its own is completely and utterly useless - and even a Power6 will draw 0W sitting on a desk with no platform. It's Intel's fault that they're pairing a 4W CPU with a 22W chipset, and if VIA offer a (max) 25W CPU with an 8W chipset then since when is it fair to exclude the chipset from power consumption figures?

At idle the articles (all of them I've read) show that the Nano uses at worst the same amount of power as the Atom, a few had it a watt under. What matters now is the number of power usage steps the CPUs have for moderate use, e.g., 400MHz -> 800MHz -> 1200MHz -> 1600MHz vs 400MHz -> 1600MHz. Load power usage on many systems is a very rare occurrence in standard use, hence the more intermediary power steps, the better the power usage under normal use.

Re:explaination of energy efficiency

[LarsG]

And since when isn't the platform a required part of a computer?

I didn't say that it wasn't. What got my knickers in a twist was "Nano beats Intel Atom", when "Atom let down by power-pig chipset" would be a more accurate description.

It's Intel's fault that they're pairing a 4W CPU with a 22W chipset

Absolutely. The only reason I can think of for Intel to require mobo makers to use the 945 is that they can produce them for cheap and want to make as much money as possible on the platform. The Atom 2x0 should work fine on pretty much any chipset that supports a 533 FSB, so it is really inexcusable for Intel to bundle this CPU that delivers 3+ times the performance per watt than the Nano with a chipset that is pretty much the opposite.

Re:explaination of energy efficiency

[bestinshow]

On interesting thing is that VIA will offer a 1.6GHz Nano with a 17W TDP as compared to the 1.8GHz Nano with a 25W TDP.

Given the reviews, this will still outperform the 1.6GHz Atom in many areas.

However it will drastically reduce the difference in load power consumption.

It's a shame that Paulsbo, the low-power chipset for Atom, is designed for MIDs, something that no-one is actually interested in. It'll probably find a way into the 7" subnotebooks as well, but it's not enough for anything bigger.

Re:explaination of energy efficiency

[LarsG]

It's a shame that Paulsbo, the low-power chipset for Atom, is designed for MIDs

Some people are actually making SBCs based on Atom/Poulsbo. While certainly not fast enough to use as a desktop PC, I'm sure people can build a lot of interesting appliances with a board like that.

Re:explaination of energy efficiency

[Sebastopol]

I took the info from the graphs right out of the article.

I ignored the system power because a real Atom target system is a UMPC (or smaller) which will use less power than the core itself.

Re:explaination of energy efficiency

[Sebastopol]

The target Atom products will be UMPC or lower. Typically these platforms use less power than the core. Large cell phones requrie milliwats of power, UMPCs might approach a watt.

Measuring total system power on an ATX form factor is like trying to record a whisper on a roller coaster.

That is why I chose to ignore the system power: the board is mostly for eval and developers. I sincerely doubt anyone would want to build an ATX system with an Atom.

Target's Linux laptops

[Animats]

Target is getting serious about Linux laptops. Looking under "Laptops", the first screen of "Featured items" has 3 colors of the EE PC, Linux version, and some HP laptop. The XP version of the EE PC costs $100 more.

That Which We Call a Rose...?

[camperdave]

Back in May, when the Isaiah architecture was first disclosed, VIA declared a performance victory over Intel's upcoming Silverthorne technology. Since then, Isaiah has become the VIA Nano processor, and Silverthorne changed to the Intel Atom...

OK, let me see if I've got this straight. The VIA Isaiah beat out the Intel Silverthorne. Then the VIA Isaiah was renamed the Via Nano, and the Intel Silverthorne was renamed the Intel Atom. Now the VIA is still beating the Intel? So what you're telling me is that a name change has no effect on chip performance? Well, color me shocked!

Re:That Which We Call a Rose...?

[DragonWriter]

OK, let me see if I've got this straight. The VIA Isaiah beat out the Intel Silverthorne. Then the VIA Isaiah was renamed the Via Nano, and the Intel Silverthorne was renamed the Intel Atom. Now the VIA is still beating the Intel? So what you're telling me is that a name change has no effect on chip performance?

Uh, no. I think you missed something:

Back in May, when the Isaiah architecture was first disclosed, VIA declared a performance victory over Intel's upcoming Silverthorne technology.

IOW, before they were concrete products that had been benchmarked, VIA made claims. Now there is a test substantiating those claims. The news is not the fact that the code names now have product names and nothing else has changed, its that there is now some evidence to point to in support of the claims.

Re:That Which We Call a Rose...?

[Cajun Hell]

Well, color me shocked!

It's easy to say that in hindsight. But we know that having an "X" in one's name makes a thing faster (e.g. Xatom, Nanox) so the experiment had to be tried with changes less radical than doing that.

Isaiah sounds all biblical and stuff, and computer in biblical times were very, very slow (even slower than World War 2 computers), so it was thought that upgrading to a techie-sounding word might improve things, and "nano" is a very techie word (for a word without an "X" at least).

On the other hand, Silverthorne sounds like an internal code name, not nearly dumbed down enough for the mass market. Renaming it to "atom" not only sounds lamer, but implies it's from the "atomic age" (i.e. the 1950s).

By changing each name in an opposite direction, it was hypothesized that it might cause a performance delta on the order of adding an "X". So the experiment was run. Sure, now we know. But did you?

Re:That Which We Call a Rose...?

[Jesus_666]

So what you're telling me is that a name change has no effect on chip performance?

Intel also added "Type R" stickers and racing stripes to their CPU. I think VIA added two spoilers, though, which - as everyone knows - equals "pretty damn fast".

Maybe Intel can still win if they put a modkit on the Atom to make it look meaner. However, it still lacks out-of-order execution, which is crucial for drift-racing.

Re:That Which We Call a Rose...?

[mgblst]

Should have painted it red, or put a racing stripe on it.

The biggest problem with the Atom...

[NerveGas]

... is the chipset it's paired with. I recently bought an Intel motherboard with an Atom on it. Whilst the CPU is only 4-ish watts, the board draws around 40-50 watts. That's the board, not optical or hard drives.

That northbridge, with the non-power-optimized video card and memory controller, sucks up the juice. The heatsink on the northbridge is 4x larger than the one on the CPU. Furthermore, the heatsink on the northbridge has a fan, where the HS on the CPU has none.

Re:The biggest problem with the Atom...

[Anonymous Coward]

I totally agree, I don't like the fact the 945 chipset needs a fan. Isn't the 945 just an older chipset by intel, kind of hacked to make it work with the atom chip?

Another thing I dislike about the 945 is the 10/100 lan, no gigabit.

I am waiting for the following board to come out, which by the looks of it resolves both these issues, it doesn't require a fan, has gigabit lan, is based on a proper embedded chipset so uses very little power. It also has a DVI connector as a bonus.

http://apac.kontron.com/products/boards+and+mezzanines/embedded+motherboards/miniitx+motherboards/ktus15mitx++16.html

Re:The biggest problem with the Atom...

[NerveGas]

Wow. That is really, really pimp. When that comes out, I might just have to buy one...

Re:The biggest problem with the Atom...

[jabuzz]

At last someone doing a proper mini-ITX board with an Atom, and they have mostly got rid of all the legacy junk of the back as well.

What would be sweet is a version with dual 1Gb lan, and capable of being set up with a serial console redirection for a low power headless home server. For added bonus points some more SATA ports would be nice.

In-order hyperthreading?

[John.P.Jones]

I took my graduate level architecture class from Dean Tullsen at UCSD, who invented 'hyperthreading' although it was called Symmetrical MultiThreading (SMT) back then. As I recall the entire greatness of the architecture was recognizing that all the fancy hardware introduced to allow out-of-order speculative execution could actually be leveraged to allow the processor to drive multiple independent threads at the same time, without much additional overhead. So if intel's atom (haven't been following it) uses an in-order core and hyperthreading that just don't make much sense. Anyone care to provide an explanation?

Re:In-order hyperthreading?

[julesh]

So if intel's atom (haven't been following it) uses an in-order core and hyperthreading that just don't make much sense

I think single issue is an even more relevant problem. I can see how HT could help with an in-order dispatch dual issue core (the primary thread could execute the next two instructions if they were completely independent while the secondary thread could execute an instruction whenever the next two instructions in the primary thread weren't independent), but in terms of single issue... dunno.

I remember there was a processor in the early 80s (I can't remember whether it was a Cray or CDC processor, but it was one of those) that had two threads and dispatched instructions alternating between the threads so that it could have a deeper pipeline without waiting for dependencies. Perhaps that's what's going on?

Re:In-order hyperthreading?

[Wesley Felter]

In-order cores still stall (in fact they stall more), so there's still a benefit from MT.

Re:In-order hyperthreading?

[Macman408]

I believe you meant to say *simultaneous* multithreading. You probably confused the name with *symmetric* multiprocessing (SMP, the fancy name for multiple identical CPUs). And it's still generically called SMT, Hyperthreading is just Intel's brand for it.
I have no specific knowledge to back this up, but I'd bet that they're doing something similar to the Sun Niagara (aka UltraSPARC T1). The cores in the Niagara are single-issue in-order. However, the hardware supports multiple threads, so it can pick any of them to issue an instruction on any particular cycle - if there's a data dependency or cache miss, it can fill what would otherwise be a pipeline bubble with an instruction from a different thread.
Basically, the Atom can use hyperthreading to keep the pipeline full instead of out-of-order execution. Out-of-order execution, simultaneous multithreading, and superscalar design are all independent design features.

Re:In-order hyperthreading?

[LarsG]

Both ooo and smt are techniques to avoid putting NOPs in the pipeline.

In ooo, it is "Instruction x+1 needs the result of instruction x before it can execute, so it can't enter the pipeline earlier than n ticks after instruction x. Let's re-order the instructions so that we can put useful instructions in that gap instead of issuing nops".

smt is "Instruction x+1 in thread 1 needs to wait n ticks, let's fill that gap in the pipeline with instructions from thread 2 instead".

Both needs the logic to understand how long instruction x+1 needs to wait after instruction x was put in the pipeline, but if you do smt only you do not need the instruction reorder logic.

I'm not a cpu designer so I have no idea how many gates are needed for the instruction reorder logic, but it seems Intel decided to do smt only.

Re:In-order hyperthreading?

[Anonymous Coward]

SMT is not only for stalls due to failed speculative execution, it can also be used when one thread is waiting for the 2nd level cache, main memory or i/o. Depending on how fine-grained it is, it could also be employed when one thread runs only FPU-instructions and keeps the Integer execution paths idle.

Transmeta?

[Ichoran]

I'm confused. Aside from it being over five years later, from a company that is not going out of business, how are either of these better than the Transmeta processors?

Re:Transmeta?

[Anonymous Coward]

Because Linus hasn't given it his death touch :)

Viva VIA-gra!

[flahwho]

Call your doctor ...If you experience power surges greater than 4 watts.

-All generalizations are false, including this one.

'Only' 4 watts?

[david.given]

Am I supposed to be impressed?

The new Cortex-A8 ARM processor consumes 300mW; less than a tenth that of the Atom. The Atom is marginly faster, but not much so --- the figures I've found show that a Cortex develops about 2.0 Dhrystone MIPS per MHz, vs about 2.4 for the Atom. Plus, the Cortex is a CPU core, not a discrete chip; most actual products couple it with an on-chip OMAP DSP engine, which is ideal for doing things like video encoding or decoding or OpenGL. With Atom you end up having to couple it with a dedicated GPU...

Re:'Only' 4 watts?

[the_humeister]

Ok, now find me a mini-itx board with this ARM chip you speak of for about $70. Otherwise, there's no point in bringing this up.

Re:'Only' 4 watts?

[psychicist]

I'm not impressed either. I'm waiting for Cortex A8 (and A9) based processors, boards and devices to become more widely available. On these kinds of devices the architecture doesn't matter, so my first choice would be an ARM and second a MIPS implementation for low power consumption and accummulated energy costs, which would greatly offset an initially even higher purchase price.

Re:'Only' 4 watts?

[david.given]

Yes, indeed --- I'm rather excited by the new MIPS-based mini laptops that are just becoming available. Not for use as a laptop; for use as a server --- for under $100, you get an adequate processor, adequate memory, UPS, console, flash disk, decent I/O capabilities via USB, and all this in a miniscule form factor and using practically no power. The fact that it uses MIPS rather than ia32 or ARM would give you a small edge security-wise, too.

But I'd still really like an A9-based mini-ITX board.

Re:'Only' 4 watts?

[Paradigm_Complex]

Yes, you are supposed to be impressed.

You're seriously underestimating the market value of being able to run Windows and Win16/32/64 applications (without binary translation). Yes, there's other operating systems and software that run fine on ARM and plenty we can do with them, but at the same time there's plenty more people who are petrified of any word processor other than MS Word.

Show me a 300mW ARM processor that runs Crysis (as the Nano does) and I'll concede the point. Until then, shut up sit down and make fun of the "63,434 watts" comment like the rest of us.

Re:'Only' 4 watts?

[Anonymous Coward]

Links for the MIPS based mini laptops?

Not a fair power comparison

[Anonymous Coward]

The board that Intel is currently selling is paired with a 945G *desktop* chipset. Probably to get it out of the door and reduce stock. That chipset uses 22 watts while the cpu uses 4 watts. When Intel finishes their Atom chipset, there should be a considerable difference between the two as far as power consumption goes.

Wake me up when Tegra joins the race

[s_p_oneil]

I think all of this will be moot when the nVidia Tegra devices come out. That will be when I break down and buy either a mini-laptop or a hand-held device.

Re:Wake me up when Tegra joins the race

[BrentH]

Uhmm, Tegra is an ARM. The whole point of these low power x86 chips is to bring, well, x86 compatibility to low power devices. Which, wether you like x86 from an architectural standpoint or not, has some obvious advantages.

Re:Wake me up when Tegra joins the race

[s_p_oneil]

I don't think x86 compatibility is a very useful feature for hand-held devices, or even for mini-laptops. Windows is fine on my desktop. I don't want it hogging the resources on my hand-held. When it comes to BSD/Linux, it's easy to recompile apps for a different platform. It's less easy to customize the apps for the smaller screens and lack of input devices, but it's not like that would be any easier for a Windows app.

Ultra low power combined with powerful graphics, on the other hand, is so much more useful. I may be a bit biased because I'm a 3D graphics/game developer, but great games and audio/video playback are incredibly desirable features on devices like this. And it's not like it won't be able to run a web browser, mail client, etc.

Re:Wake me up when Tegra joins the race

[psychicist]

I would like to get my hands on an Nvidia Tegra based device, but unfortunately it's being restricted to Windows Embedded only at the time. So until Nvidia sees the light and supports other operating systems, it unfortunately isn't a viable choice for those wanting something else than Windows.

Re:Wake me up when Tegra joins the race

[Jorophose]

Personally I'd rather see Tegra in music/video devices, like an iPhone minus the phone but with a lot more features (movies, music, games, browsers, torrent clients, wireless N, etc.).

I mean, it would be nice to finally eliminate x86 from the mobile market, but at the same time that doesn't seem to play to Tegra's strenghts... And I wouldn't hold my breath to see Tegra in laptops. It would be sweet, yeah, but good luck. I just wish the laptop market would be as open as the desktop one.

I myself am waiting for Nano-based mininote, because it's the only reasonable chance for a nano-based "netbook". Sony has stated it will develop (at Computex?) an Openbook-based laptop, but it might be with the C7... And the thing is the C7 isn't necessarily a bad CPU. It's the Atom, but 32-bit and minus HT. It's just the Nano packs more of a punch.

Ultimately I hope we'll get a minimum of two types of subnotebooks: the ones based on low-performance chips (Tegra that focuses on video, Atom that focuses on god-only-knows-what) to get better batterylife for Asus eee 700 and 900-sized laptops (think 8hrs idle, 6hrs pushing it), and the ones based on high-performance chips (Nano that focuses on being a lower-watted Core Solo) to get performance and OK battery life (think 6hrs idle, 4hrs really pushing it on average). That way those who want what they want can get it.

And although there's not much of a difference between an Atom and an ARM Cortex (like the kind you'd see in tegra), Nano is a totally different beast. When the ULV and 45nm ones come out it should rock the boat quite well, because then you've got something that does 1.8GHz at 15W or less, or 1GHz at ~3W, idling at 50mW to 300mW. (guestimates all around!).

Re:Wake me up when Tegra joins the race

[s_p_oneil]

Yeah, MS has definitely been bribing/wooing nVidia heavily. For the first time ever with the GeForce 8800, nVidia did NOT lead with OpenGL demos and documentation (even though it meant most of their customers had to wait an extra year until Vista shipped to play with the card's new features). I'll bet MS bribed them heavily for that. Of course, when DX10 was left out of XP, it didn't make sense to use anything but OpenGL (assuming ATI will fix their broken OpenGL drivers, which they won't).

Platform choice

[Outland Traveller]

I was liking both processors up until I found this gem in the article:

If there is egg to be thrown in anyone's face from this article, it is on Intel for its locking down of the Atom platform. Since Computex this year I have been hearing complaints from board vendors on the amount of restrictions Intel is putting on them for Atom products. Vendors are not allowed to build Atom motherboards with PCI Express, digital video outputs or more than one memory slot. VIA on the other hand is openly courting board manufacturers to put as much technology on a mini-ITX design as they can - as long as they DO build one!

Here we see Intel, up to its obnoxious "You'll use our technology only as we prescribe" games. This is the same philosophy that leisurely milked the market for 33Mhz CPU bumps every 6 months, while they sat on years worth of better technology, until AMD lit a fire under their ass.

Don't be fooled again.

Re:Platform choice

[hellwig]

Puh-lease.

Intel isn't playing the "do as we tell you or we're going home and taking our ball with us" game. What they're doing is looking out for the consumer. Why would the consumer want a fully-functional PC running a slow, innefficient processor like the Atom? Why would an atom-based computer need more than 1 memory dimm, digital video (have you seen Intel's graphics chipsets, they really aren't the greatest), and PCI-express? I don't think the FSB the Atom uses would support PCIe bandwidth. Intel is only protecting the consumer from the inadequacies of the Atom system (that and the 5+ year old northbridge they're using doesn't support many of those technologies).

If you want PCIe, more than 1 memory slot, and digital video output, might Intel receommend a Intel Core 2 Extreme on a X48 motherboard running a couple ATI x4870's in CrossFire (TM) mode?

Re:Platform choice

[hellwig]

Oh, and what did I find when I went back to my google homepage:

Atom's now available for the desktop.

Re:Platform choice

[0xABADC0DA]

I bought the BOXD945GCLF and really it's much better for the price than the competition's mini-itx boards. But judged on its own merits it's a piece of crap compared to what it could be and what it should be.

The BIOS is slow and flaky and limited, the VGA output noisy unless the power supply is perfect, the chipset draws 8 times as much power as the processor, the ACPI tables are seemingly wrong somehow since I've yet to get frequency scaling to work (it works under Windows from what I hear). And that's just getting started... I've never had a BIOS config screen actually freeze the computer before because I typed too fast for it. Maybe I've just been lucky that way?!

It looks like I got suckered just Intel can milk their better processor for more by releasing boards that are just slightly worth buying -- but only because there are no competition not because they are quality products. This is the first Intel board I've bought, and it'll be the last for a long time -- even if Via and AMD can't compete, if just as a protest.

Re:Platform choice

[tknd]

Why would the consumer want a fully-functional PC running a slow, innefficient processor like the Atom? Why would an atom-based computer need more than 1 memory dimm, digital video (have you seen Intel's graphics chipsets, they really aren't the greatest), and PCI-express? I don't think the FSB the Atom uses would support PCIe bandwidth. Intel is only protecting the consumer from the inadequacies of the Atom system (that and the 5+ year old northbridge they're using doesn't support many of those technologies).

Because I don't need the CPU power, I need the efficiency. Let's consider a few target markets:

• Home email/web browser/word processor user: this person doesn't care about 3D games or high def video. They just care about checking email, browsing the net, and writing a document or two. They don't need a 3ghz quad core cpu with pci express. They just need enough cpu power to display web pages, type up a document, or check email.
• Home Server user: this person wants to have a home file server for archiving files and making media available to all other machines within the home. Most of the time this machine is just going to be sitting idle or copying data. CPU resources are not necessary, but efficiency can potentially cut down on the power consumption and noise pollution.

The truth is Intel isn't protecting the customer. That's not what they want. They're protecting their own product lines targeted at other markets from competing with themselves. For example if the atom had no restrictions, there would be little reason for a manufacturer to purchase things like celerons and low end Core2s or Pentiums when they could get the cheaper atom part with comparable specs to meet the need and pass off the savings to the customer in a more marketable price. Intel's true customer is PC makers, board manufacturers, and distributors. Each of these buy chips based on different needs in bulk. The second you offer a replacement to the others at a cheaper price is when you are effectively competing with yourself. The only way Intel can control what is being made with its parts is to put restrictions on how they're used.

Now let's consider for a second the possibilities of the atom. There was a clip of VIA showing their Nano platform running Crysis or some other fancy game with a nvidia 8600gt stuck to it. The game appeared playable. If the atom came with a pci-express x16 slot and you found you could stuff a 9600gt in there and get good performance on a game like World of Warcraft, why would you need to buy any of intel's more expensive parts including fancier motherboard chipsets and cpus? If your only goal was to play WoW you would put the least money into the CPU and dump it all on the video card. The dominance and necessity of the CPU has decreased as specialized processors have taken over the most processor intensive tasks and most CPUs (even the bottom end ones) are good enough for most applications.

What Intel really wants you to do is step up to the Core2 or even Quad core parts with nicer fatter margins than the cheapo low margin atom platform.

Personally I was hoping to build a FreeBSD ZFS home server out of an Atom platform because I was hoping that Intel would eventually provide a low power motherboard and cpu combination to which I could just add hard disks and hard disk controllers. But I guess I'll have to make new plans now that it is apparent Intel doesn't want my money. It's too bad too, because price isn't necessarily my biggest buying requirement, so they could've priced the atom just like the celeron on pentium parts and I would have still bought it because I know I'll never use the CPU's full computation power.

Re:Platform choice

[smoker2]

I bought the BOXD945GCLF ...

I thought that was an acronym, puzzled me unti I saw your username - 0xABADC0DA

Re:Platform choice

[LarsG]

Well, that review also shows that the Atom is simply not a good choice for your average desktop so you are at least half-way right. For some strange reason Intel is making Atom + i945G chipset systems, which makes absolutely no sense to me. Whatever you save in wattage on using the Atom is eaten instantly by that ancient power-guzzling chipset.

Atom, when finally paired with a decent low-power chipset and other system components (SSD/laptop HD, low-power RAM, etc), will have no problems beating the Nano in performance per watt. But that 3+ higher performance/watt advantage it has over the Nano is currently hidden by putting it on motherboards that are not designed for power efficiency.

Intel is probably doing it to cash in on the current Atom hype, but it really makes me cringe inside; it is like putting an efficient low-power high MPG engine in a SUV made of lead.

Re:Platform choice

[MoxFulder]

Puh-lease.

Intel isn't playing the "do as we tell you or we're going home and taking our ball with us" game. What they're doing is looking out for the consumer. Why would the consumer want a fully-functional PC running a slow, innefficient processor like the Atom? Why would an atom-based computer need more than 1 memory dimm, digital video (have you seen Intel's graphics chipsets, they really aren't the greatest), and PCI-express?

"Why would anyone need more than 640k of RAM??"

Biggers news: PCMark 2005 skews memory tests

[copponex]

Reading through that article, I found this:

My my. Swap CentaurHauls for AuthenticAMD, and Nano's performance magically jumps about 10 percent. Swap for GenuineIntel, and memory performance goes up no less than 47.4 percent. This is not a test error or random occurance; I benchmarked each CPUID multiple times across multiple reboots on completely clean Windows XP installations. The gains themselves are not confined to a small group of tests within the memory subsystem evaluation, but stretch across the entire series of read/write tests. Only the memory latency results remain unchanged between the two CPUIDs.

Whoops! I wonder what they'll have to say about that...

...Would Smell...

[camperdave]

But we know that having an "X" in one's name makes a thing faster

One of my former clients, who was a paint manufacturer, told me that they used to sell off-white paint. One day one of their marketing people decided to call it "Antique Satin". Same paint, different name. It started flying off the shelves to the point where they were having problems keeping up with demand.

My biggest dissapointment with VIA

[Hektor_Troy]

My biggest dissapointment with VIA cpus is the cost. Here in Denmark I can get a motherboard with a 1.6 GHz Atom CPU for roughly 77 US$ but the cheapest motherboard with a VIA cpu is an EPIA ML8000AG costing 160 US$. The cheapest C7 cpu I can find is 190US$, and we don't have any local dealers selling the Nano.

It's rather dissapointing to be honest.

After Intel and Via...

[mriya3]

Despite this, the problems I see with VIA solution are crappy chipsets/integrated graphics (crappy at least on Linux) that are shipped on their boards.

BUT I'm looking forward to see AMD' move... how long will it take to release a comparable low-watt solution on the market ? (I mean, not just a CPU, but a complete mini-itx low-power mainboard/solution)

For the CPU we're not too far away (Athlon 64 1Ghz at 0.9V, 8W, scoring better than an Atom 230 @ 1.6GHz) it seems, see http://66.102.9.104/translate_c?hl=en&sl=de&tl=en&u=http://www.tomshardware.com/de/athlon-2000-Atom-230-Undervolting,testberichte-240084.html&usg=ALkJrhhQnkhENd8LbnpHTJpwfugMUrwNLw (translated from german)

Power consumption is rather high for both of them

[Repossessed]

The energy star 5 standard requires 50 watts or less at idle, something the old C7-D platforms from via managed quite easily, are they including the display in the system power consumption perhaps, or just using shoddy parts in other spots?

Also, what is the power consumption of the lower end Nano processors? As I recall, the slower nano managed to (barely) outbenchmark the atom, it would be a much better platform to check power consumption against in this case.

I also point out that for these kinds of systems, 3D based benchmarks seem fairly useless, neither platform is targeted at gaming or graphics development.

x86-64 and EM64T are ALMOST the same

[Anonymous Coward]

There, fixed that for you.

Re:x86-64 and EM64T are ALMOST the same

[Anonymous Coward]

the parts that differ between the two of them are not used by anyone, anywhere, for that very reason.

Mandatory

[neokushan]

UP AND AT THEM!

China on top again

[heroine]

China beats Intel again. Wonder why those silly Chinese work so hard.

And for the SI-challenged among us...

[Aardpig]

...a watt-second is also known as a joule.

Re:Power consumption is rather high for both of th

[LarsG]

Yeah, it almost seems like they deliberately picked high power-consumption motherboards (and what is that high-performance 9W-at-idle Raptor doing in there?). The only thing this benchmark shows is that it is silly to put a low-power CPU in a PC that has no other low-power components.

Nobody's mentioned this yet, AFAIK:

[Ant P.]

The Nano also has Via's hardware-accelerated SHA+AES - for laptops, that means wi-fi WPA2 and some HTTPS. In other words the Nano would wipe the floor with the Atom for website browsing.

Yo! Dumbshit! 66 MHz is DOUBLE 33 MHz

[Anonymous Coward]

Intel did a 2x speed increase with those 33 MHz. If it weren't for MORE CORES, CPUs today would be hard pressed to match that rate in four years, let alone one.

Besides, ARM is the champ. Atom and Nano can't compete on ability.

Via Mini-ITX boards are widely used

[symbolset]

Mostly in thin clients where you can't see them unless you pull the thing apart. Wyse has quite a few popular units in their line.

The argument defeats itself

[jharel]

Since the "normal workload for an ultra-portable" isn't that much anyhow like you've said, then the performance difference is moot because it's not that big of a gap anyways. ...and by the same token, who in the world is going to use "ultra-portable" as "desktop replacement"? Pbbt.

One is here, one is not

[symbolset]

Another worthwhile observation: You can get the Atom motherboard. They're $83 delivered from Newegg. Via has a habit of announcing product and stalling. The are selling this cute little devil though. It's called the "pico-ITX" form factor.

The reviewed item though? Not yet. Until it's available we're comparing what's on the shelf to what's not. Of course the one that comes later is targeted at a slightly higher performance level. It seems likely that when it ships Intel could reply with a machine that has dual Dimms, PCIe X16, dual gigabit nics, and then up the ante with a little DVI goodness and dual core. Everybody knows the chipset supports those features already.

Maybe they'll also do a process shrink on the MCH as well to get the power down.

Personally I'll be getting both. I've played with the Atom board and I like it. These are both good enough boards for some purposes I have in mind and if more better stuff comes along later, well, that's just the way of IT isn't it? If you waited until nothing better was going to come out you might as well go back to pencil and paper.

1.8 GHz Nano not for netbooks

[jensend]

The 1.3+ GHz Nano consumes only 8W under load and would perform about the same as the 1.6 GHz Atom; you could fit three of those in the 1.8 GHz Nano's 25W power budget. With a NVidia chipset provided by NVidia's partnership with VIA, the Nano will also do a lot better on the video decode and gaming tests. I think VIA has a winner on its hands here and just hope they don't drop the ball as they've done with a few other promising technologies.

Re:Power consumption is rather high for both of th

[Repossessed]

They may not have had much choice in system boards, low power processors usually come as a bundle, and I only ev3er found one C7-D board last year for sale in the US (other countries have better selection). I bet they put giant (and completely useless) cooling fans in there too though. The whole test seems to scream that they used a process meant for testing gaming level components.

Re:Power consumption is rather high for both of th

[Jorophose]

Well, look at it this way, if you were looking for a Nano-based system, the most wattage it'll ever usage would be about 70W. It's all downhill from there.

Especially since there's a rather massive difference between two of the points, something like 5W and 400mW idle for a few hundred megahertz.

Erm, why not choose a closer chip to compare?

[Wdomburg]

Why are they comparing the Atom against the L2100? That's from the desktop line of Via Nano processors. Replace the L2100 with the U2400 and you drop from peak draw of 25W to 8W and idle draw of 500mW to 100mW.

I'm also curious about the idle draw. There are systems out there based on the same chipset (CN896) that pull 10-15W *under load*. Given the negligable idle draw of the processor itself it seems curious it should be that high, unless the reference board is simply horribly inefficient. But again, in the market the Atom is aiming the VX800U (with a peak draw of 3.5W) would seem a far more likely pairing. Of course for this test Intel is also pairing with a less efficient chipset, so it's not the fault of the reviewer.

Power consumption?

[Anonymous Coward]

Shouldn't an out of order CPU consumes 0.0 watt? Sorry Intel, you lose.

How abouut comparing like for like

[Anonymous Coward]

They tested a 1.8Ghz Nano against a 1.6Ghz Atom. The 1.6Ghz nano has an idle of 100mw against 500mw for the 1.8 and the tdp 17w less on the 1.6. It would be interesting if they compared the 1.6 nano against the atom perhaps the power usage would be similar, if not the same.

That's a lot of Watts...

[ElberethZone]

Looks to me like those two "low power" systems are eating a lot of Watts for the performance... I have built a small LAN server a few months back and used "normal" (ie: not low power) desktop CPUs: Mobo: Gigabyte GA-G31M-S2L CPU: Intel Pentium E2200 (2.2 GHz / 1 MB cache) RAM: Teamgroup Elite Kit 2x1GB PC6400 DDR2-800 CL5 HDD: Hitachi HDT72502 250GB SATA & Samsung SpinPoint F1 1TB 32MB SATA II PSU: Seasonic S12 II 430 W (too much, I am planning to change it for a 100-200W one with 80plus Gold certif) The consumption of the entire system at idle at the power outlet is 50W and in burn on both cores 66W... This is of course by undervolting the CPU at 1.1V but it is rock stable, and even if the stock voltage is kept this would not be a lot more (the chipset is really the power hungry part).

Well using a nonexistent chipset won't work

[Anonymous Coward]

because you won't even boot.

And maybe by the time the Atom chipset is out (if ever), the Via chipset will be cheaper and less power hungry too, meaning no change.

Hardware Encryption

[Anonymous Coward]

NOBODY mentioned hardware encryption. Where my paranoid hardware geeks at?

Intel's Atom has *no* hardware encryption or hashing build in. VIA's Nano has 2 RNGs, SHA-1 and SHA-256 hashing, and AES and RSA engines BUILT IN ON THE CORE!

If you factor this in, the Nano STOMPS the fscking CRAP out of the Atom for apps needing security.

wtf!

[Anonymous Coward]

I dont get it! They are all very happy to see low power system which consumes 60 wats of power!
Where is green in that!
Just look at mobile celeron in action with total power consumption incluing hdd drive of 28 wats!

http://resources.mini-box.com/online/MBD-I-D201GLY/intel-d201gly-power-consumption.html