Do you really want the VIA Nano in an HP Mini? I don’t think so
I’ve lost count of how many times I’ve seen the comment "I can’t wait for the HP Mini-note to get a VIA Nano inside!" or some derivative thereof. We reviewed HP’s netbook with the current VIA C7-M running at 1.6 GHz and the new Nano is a pin-for-pin match of the C7. Folks are making the leap that a simple CPU swap by HP will reap big rewards in terms of performance. They’d be correct, since the Nano is magnitudes faster than the older C7-M but you can’t forget the other side of the equation. It’s the one we always have to consider with mobile computing devices: the see-saw with processing performance on one side and power usage on the other.
Reading through some of the recent comparisons between the Intel Atom and the VIA Nano today (you can find them here and here)only underscores this point. Remember that the 1.6 GHz Intel Atomshrinks its TDP or Thermal Design Power rating to 2.6 Watts under load.The TDP for a VIA Nano at 1.8 GHz? It maxes out at 25 Watts. That’sdoesn’t imply that the Nano constantly uses ten times more batterypower than the Atom, but it’s a large difference that will surelyimpact runtime.
Let’s look at this another way: the current HP Mininote runs the 1.6GHz C7-M. Maximum TDP for that processor? 8W, or roughly a third of thenew Nano. Of course, the other components come into play for any mobiledevice as they all use a certain amount of power. But if HP decided toput the Nano inside the Mininote, they’d have two choices: increase thebattery size to maintain the 2.5 hours of runtime on a standard battery(double that with the 6-cell) or drastically reduce the amount ofadvertised runtime. For some folks, that would be fine: they’ll wantmore power than the C7-M offers. They just have to remember that whenthey sit on the performance side of the see-saw, the power usage sidewill above their head and shoulders.
I’m still very much looking forward to the new VIA Nano asit’s a step forward for VIA and brings more competition to the playingfield. However, I think it might be better suited to a standardnotebook that can handle a larger battery… not a netbook like the HP Mini-note. For a compromise between power and performance, HP is better to either stick with the VIA C7-M or consider an Intel Atom.
(via Engadget)
It looks to me like there’s a nice upgrade path with the 1.33Ghz Nano.
Performance would be much better than the 1.6Ghz C7 (lets say between 50-200% CPU performance increaase) for the same 8W TDP.
In addition, you’d be able to lock down the CPU to 1Ghz at times of mobile need to save energy, increase battery life and still have the same processing power as the 1.6Ghz.
Steve.
It looks to me like there’s a nice upgrade path with the 1.33Ghz Nano.
Performance would be much better than the 1.6Ghz C7 (lets say between 50-200% CPU performance increaase) for the same 8W TDP.
In addition, you’d be able to lock down the CPU to 1Ghz at times of mobile need to save energy, increase battery life and still have the same processing power as the 1.6Ghz.
Steve.
I’m very happy with my new Acer Aspire One and its Intel Atom processor. Good performance, runs cool and provides decent battery life on the 3 cell battery.
Kevin, what you need is the VIA Nano U2100 which at 1.3+GHz has a TDP of 8W…
The benchmarks you state are for the L2100 which, as you say, has a 25W TDP and is more suited for low end desktop PCs.
Check out: http://www.umpcportal.com/
PJE, tell that to the reviewers.
The only reason I highlighted the 1.8 GHz Nano is because that’s what the they pitted against the 1.6 GHz Atom. That particular Nano was around 20% faster but used far more power. Dropping down to the 1.3 GHz as you suggest is definitely a good idea, but the chip still isn’t as power efficient as the Nano and likely offers slightly lower performance than the 1.6 GHz Atom (although that’s simply an educated guess on my part at this time).
Having said that: how would the 1.3 GHz Nano be better than a 1.6 GHz Atom in the HP? Too early to answer that one I think, but my gut says that you’ll see longer battery life and generally same or better performance in a 1.6 GHz Atom system over a 1.3 GHz Nano system. We’ll have to get our hands dirty and do some testing of course.
Kevin, I think they chose the L2100 as this is probably the only CPU VIA is sending out at the moment…
My thought is that the Nano U2100 has to beat the C7-M. If you can get a performance increase over the older chip (say 50% faster) using the Nano in a system such as the HP Mininote, does it really matter if it’s as fast as a 1.6GHz Atom.
What’s important to me is can it play full screen video without stuttering and how it handles Skype and other web based applications…
…and of course, battery life!
I seem to remember reading somewhere that one of the problems with the Atom is it will only execute inline code and not random access. If that is in fact true then benchmarks would likely be the only way to compare them with the Nano.
I’ve been one of the folks enthusiastically waiting for the HP to get the Nano. But, I’ll admit my reasons are somewhat peculiar: the Nano, unlike the Atom, is a 64-bit processor, meaning I can run 64-bit Linux on it.
If you don’t care about that, then I don’t know that the Nano will provide many advantages over the current processor.
It looks to me like there’s a nice upgrade path with the 1.33Ghz Nano.
Performance would be much better than the 1.6Ghz C7 (lets say between 50-200% CPU performance increaase) for the same 8W TDP.
In addition, you’d be able to lock down the CPU to 1Ghz at times of mobile need to save energy, increase battery life and still have the same processing power as the 1.6Ghz.
Steve.
Steve, I think your that’s about the only likely upgrade path. You gain performance over the existing C7-M and don’t go up in power usage. So the question becomes: how does the 1.3 GHz Nano compare to the 1.6 GHz Atom. Can’t wait to find out!
The Atom has variants that run 64 bit code, the Z540 for example.
All the talk of how poor the processor was nearly put me off getting a Mini-Note. But I’m glad I did. I’m actually pleasantly surprised with the performance, especially after an upgrade to 2GB ram and even with an upgrade to Vista Home Premium. I have no problems running ripped DVD’s full screen and although HD digital TV stutters standard def digital TV fine which surprised me. It definately runs better than y Sony UX with XP on it. Much, much better.
It’s even inspired me to install Photoshop for light work in the field.
The only real issue is multitasking, but I expected issues there and it’s not as bad as I thought it might be.
So all in all I don’t care what could have been because the Mini-Note does everythin I expected and much more.
Gordon
Hi Gordon, did you reinstall the whole Vista to your mininote or using the built-in Vista? I use the built-in Vista but it seems running not very well. It takes me more than 3 minutes to start it up!
I think the question also lies on what sort of atom your comparing the nano too.
I think its been put back to september but thats when the dual core atoms get released, these I presume will be far better at multi-tasking how these compare to the nano, which I don’t believe goes dual core to some point in 2009
It’s great how you respond to comments on this site. I was reading a notebook review (I think laptopmag) and noticed this too, how the VIA chip uses much more power. But in the end, the battery life is still the same for most atom/nano netbooks that don’t have extended batteries. I dislike that the most.
-I would still take the msi wind over the hp mini-note. the hp has a fantastic keyboard, but I’ve had no problems with the wind and the HP is just so much more expensive!
Atom Z540 does *not* support 64bit !
Where did u get your info ?
I got mine from Intel:
http://ark.intel.com/cpu.aspx?groupID=35460
Only Atom CPU with 64bit Support is the el-cheapo Atom 230.
I read a review today comparing the Atom and Nano. The power difference between idle & flatout on the Atom was 4W and was 25W for the Nano. The ‘gotya’ was the fact that the idle power was around 56W for both, so the difference wasn’t that great. What was surprising was the performance gains that the Atom had in some areas, even though it is a simpler processor.
http://www.pcper.com/article.php?aid=597
Dave, I read the same review, but let’s dig a little deeper. You mentioned the 56W used when both systems were idle. My Samsung Q1UP UMPC uses about 10W… when in use! If my portable device used 56W when idle then the power would drain from the 6-cell battery with about an hour of idle time.
My point being: out of the 56W in use when idle, the Atom is using .5W or 1% of the power draw. (Atom’s TDP is .5w when idele and maxes out at 2.6W) Probably not a good reference system to use an Atom in a review, although it does highlight the processing power of the new Nano.
…on the other hand, the Nano is supposed to be pin compatible with the C7. Setting aside whether that would lead to a faster adoption time by companies currently using the C7 – would the more adventurous be able to swap out the chip ourselves? Is the CPU soldered on the board for the 2133?
The deciding point for me in favor of the Nano versus the Atom is the poor chipset Intel uses with the Atom. Via has also recently upgraded the drivers for the 2133, I would presume further refining from Via will make the Nano a suitable upgrade.
Nano does hurry up to go into idle state, Atom uses constant power consumption: http://www.pcper.com/article.php?aid=597&type=expert&pid=8 see the graphics below.
I don’t think VIA’s Nano does use much more power than Atom. A benchmark of an ULV version of the Nano would be great, to see if it’s worth it.
I would love to the the Nano in the HP. Then maybe I’ll buy one since it won’t be underpowered. *cough*atom*cough*