VIA Nano CPU Benchmarked, Beats Intel Atom 279
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.
Misleading title? (Score:5, Insightful)
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? (Score:5, Informative)
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: (Score:3, Interesting)
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.
Unfor
Re: (Score:2)
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? (Score:4, Informative)
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? (Score:5, Insightful)
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).
And for the SI-challenged among us... (Score:3, Informative)
Re: (Score:3, Insightful)
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...
Comment removed (Score:4, Informative)
Re: (Score:3, Interesting)
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? (Score:4, Interesting)
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: (Score:3, Insightful)
Speaking of the 35xx, check out Pandora [openpandora.org] 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 h
Re:Misleading title? (Score:5, Insightful)
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: (Score:2, Interesting)
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 max
Re:Misleading title? (Score:5, Informative)
Re: (Score:2)
Re: (Score:3, Informative)
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? (Score:5, Funny)
I hate to actually read the article and then post, but
Get Out
Re: (Score:3, Interesting)
Re: (Score:3, Insightful)
Almost had me going there... (Score:5, Informative)
Re: (Score:2)
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: (Score:2, Informative)
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: (Score:2)
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 [eeeuser.com]
Re:Almost had me going there... (Score:5, Insightful)
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: (Score:3, Interesting)
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.
Not a fair power comparison (Score:2, Insightful)
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.
Uhmmm... measurements? (Score:3, Informative)
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? (Score:5, Insightful)
>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: (Score:2)
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
Re: (Score:2)
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: (Score:2)
Offtopic, but if s/,/./g "can be a major PITA", then you are in trouble indeed.
Re: (Score:2)
Re: (Score:2)
Wait for the next generation (Score:5, Interesting)
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: (Score:3, Insightful)
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 (Score:4, Insightful)
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.
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).
conspiracy (Score:5, Funny)
Re: (Score:2)
Re: (Score:2)
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 (Score:4, Interesting)
Re:conspiracy (Score:4, Funny)
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???
Interesting. (Score:3, Interesting)
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: (Score:2)
Would you feel more comfortable with it if they went back to calling their CPU Cyrix?
Re:Interesting. (Score:4, Informative)
Re:Interesting. (Score:5, Insightful)
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: (Score:2)
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: (Score:2)
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: (Score:3, Insightful)
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.
VIA has become a real threat to Intel (Score:5, Insightful)
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 [youtube.com] 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.
Target's Linux laptops (Score:3, Informative)
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.
It's a tie (Score:3, Insightful)
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 (Score:5, Informative)
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 [via.com.tw]. Compact Flash, mini-PCI and PCI Express. Damn. I like that board. That is the perfect board for the small, quiet home server.
Re: (Score:3, Informative)
The Intel board could be a lot better if they used mobile chipsets. I have an old Shuttle desk
Re: (Score:2)
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.
More benchmarks and analysis here - HotHardware (Score:5, Informative)
Re: (Score:2, Informative)
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... (Score:5, Informative)
...with the same findings.
http://arstechnica.com/reviews/hardware/atom-nano-review.ars [arstechnica.com]
Re: (Score:3, Informative)
Biggers news: PCMark 2005 skews memory tests (Score:5, Interesting)
Reading through that article, I found this [arstechnica.com]:
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...
x86-64? (Score:2)
It's a nice thing indeed, but it'd be like comparing apples and pears.
Re: (Score:2)
And the Atom has EM64T.
What?
Re: (Score:2)
Re:x86-64? (Score:4, Informative)
x86-64 and EM64T are the same instruction set.
Re: (Score:3, Informative)
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).
Performance for full day battery life (Score:3, Informative)
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 [hardocp.com]). 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: (Score:2, Insightful)
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: (Score:2)
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: (Score:2)
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!
The 4 watts comment (Score:4, Informative)
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 [pcper.com]
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 (Score:5, Informative)
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: (Score:3, Interesting)
With only 4 more watts... (Score:3, Funny)
Wrong Benchmarks?? (Score:2, Insightful)
explaination of energy efficiency (Score:5, Informative)
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 (Score:4, Informative)
Re:explaination of energy efficiency (Score:5, Insightful)
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" [washingtonpost.com] 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: (Score:2, Informative)
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: (Score:2, Informative)
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 ins
That Which We Call a Rose...? (Score:5, Funny)
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: (Score:2)
Uh, no. I think you missed something:
IOW, before they
Re: (Score:3, Funny)
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
...Would Smell... (Score:2)
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.
The biggest problem with the Atom... (Score:3, Informative)
... 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.
In-order hyperthreading? (Score:4, Interesting)
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: (Score:2)
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... dun
Transmeta? (Score:2)
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?
'Only' 4 watts? (Score:3, Interesting)
Am I supposed to be impressed?
The new Cortex-A8 ARM processor [arm.com] 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: (Score:3, Insightful)
Wake me up when Tegra joins the race (Score:3, Interesting)
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: (Score:2)
Re:Wake me up when Tegra joins the race (Score:4, Insightful)
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.
Platform choice (Score:5, Interesting)
I was liking both processors up until I found this gem in the article:
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.
Power consumption is rather high for both of them (Score:4, Insightful)
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.
Re:The chart shows 60 watts (Score:5, Informative)
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
The problem (Score:5, Insightful)
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: (Score:2, Informative)
> 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: (Score:3, Interesting)
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
Re: (Score:2)
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: (Score:2)
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.