Acer Ferrari One: An AMD Ultraportable

39 Comments

View All Comments

• Shadowmaster625 - Tuesday, June 8, 2010 - link

  Why would you even review a netbook with a 65nm CPU with a 48Wh battery that sells for $600? Clearly the price is in the name and Acer is bat$%^t crazy on this one.
  
  You still havent reviewed a sempron M100 ffs. With all these new AMD cpus coming out, why waste time on this?
  
  I can only think of one reason for this FUD: to get people to ignore the V105 and others so they get as little attention as possible. People will see the upcoming reviews and they'll think "Hey I just read about AMD's new mobile processors. They still suck." Mission accomplished. How much does intel pay for this favor??

• JarredWalton - Tuesday, June 8, 2010 - link

  Short of going out and buying the laptops ourselves, getting anyone to send us AMD-based laptops has been difficult at best. And I fail to see how the M100 is going to impress. Single-core, 25W TDP isn't looking too promising. Based on the other AMD laptops we've looked at, those TDP numbers are realistic, so you're looking at a 2GHz part that consumes as much power as a mobile Core i3/i5.
  
  What AMD can do right now on the mobile front is offer a less expensive (slightly) laptop with "good enough" performance. Unless the price is right, it's not a compelling argument, which is the case with the Ferrari One. I'm not sure what we're supposed to say here: AMD has laptops that underperform relative to Intel: Editors' Choice! If we did that, people would really lose faith. I think the bigger question is why you're so convinced AMD has a great mobile platform when all of our testing indicates that they don't.
  
  I'll have an i3-330M vs. Athlon II M300 match up later this week or early next; we also have a Turion M600 review coming. Nothing I've seen with either of those AMD laptops makes me think AMD is the better buy on any laptop right now. They're merely an inexpensive alternative, and in some cases the difference is as little as $25.

• Shadowmaster625 - Tuesday, June 8, 2010 - link

  Oh that's great. Match up a $550 i3-330M against a $399 AMD M300 (http://bit.ly/cUdS7h). I cant wait.

• JarredWalton - Tuesday, June 8, 2010 - link

  We'll have an M600 laptop reviewed as well in the near future, which costs more and still doesn't close the performance gap. It's the lowest end i3 CPU against a reasonable counterpart. Of course the laptop you link has an MSRP of $600 but is on clearance (apparently) for $400. That's not a fair comparison either. Most M300 $400 M300 laptops are severely trimmed back (2GB RAM, 250GB HDD, HD 4100); the "normal" M300 cost more like $540:
  http://www.mwave.com/mwave/SKUSearch_v3.asp?px=FO&...
  
  But you can pick your comparison:
  
  i3-330M vs. M300: 330M is quite a bit faster, gets better battery life, but the M300 can be found a lot cheaper.
  
  i3-330M vs. M600: M600 is nearly as fast, gets worse battery life, and costs more.
  
  You make it sound as though I'm trying to handicap AMD, but I'm not. At least with the M300 it has a major selling point of costing less. The M600 laptops don't improve battery life but bump the price up almost $200 (only $50 relative to the normal M300 setups). At $400, that Newegg deal is an excellent value. At $500 or more suddenly it's not so interesting. As far as the i3-330M setup goes, it has a Blu-ray combo drive and sells for $550. Knock off at least $50 (more like $75) for that, and it's a $475 system otherwise... though no one seems to do that inexpensive of an i3 laptop. Frankly, $550 for a Blu-ray combo + i3-330M is another great deal.

• Shadowmaster625 - Wednesday, June 9, 2010 - link

  I highly doubt that i3-330M notebooks will be priced similarly to M600 notebooks. It may look that way on paper, but I'm betting the M600 will be 25% cheaper. It doesnt make much sense to ignore discounts when the discounts are so widespread. Every time I do price checks, I have no trouble finding them.
  
  At any rate, they're both too power hungry for my interests. I would be happy with an atom N475 that has an undervolted i3 gpu built into it. But Intel will not release a product like that because they know it will eat up their revenue. So the only hope is an AMD V105 or something along that line.

• JarredWalton - Wednesday, June 9, 2010 - link

  i3-330M: http://www.google.com/products?q=i3-330m&oe=ut...
  
  Turion II M600: http://www.google.com/products?q=turion+m600&h...
  
  Gateway NV5933u (i3-330M): http://www.bestbuy.com/site/Gateway+-+Laptop+with+...
  
  Note that the Gateway M600 unit at TigerDirect is "not available", so the cheapest available laptop I can find off hand with M600 is $600 with the CA Internet Security Suite. A few places (that I wouldn't immediately trust) look like M600 Gateway is available for $550.

• matt b - Tuesday, June 8, 2010 - link

  I concur with the other comments about a late, late review of the Congo platform. It's been out a year, and the much improved replacement platform was announced last month. From the article, it seems that the author is ignorant of AMD's Congo platform and the AMD Nile platform. I've been searching for a review of AMD's Congo platform by Anandtech for some time now as I've been in the market for a 11.6 notebook. To get a review, this late, and especially a review of the L310 chip (the slowest by far dual core chip sold by AMD), is simply not helpful for the vast majority of buyers. The L310 is only available in what - this computer only? The MV-40 Neo single core (draws less power) and the L335 are much more widely used. The excuse? AMD and the manufactures didn't send us a sample? Really? Did you ask in the past year? Anandtech has had extensive CULV and Atom reviews in the same time. Did it never occur to compare the Congo platform?
  First, from a price comparison standpoint this is a terrible comparison. The Ferrari label obviously jacks up the price significantly over what would otherwise be an $299 netbook. The Thinkpad x100e (given away by Anandtech but never reviewed) is available with the faster L335 and L625 chips, with twice the build quality, for less than this dud costs.
  Second, the conclusions reached in the article don't match the performance graphs. It doesn't make sense. On general performance page, the author claims that Futuremark places this computer behind the 1201N (Atom and ION). Wrong. The AMD chipset beats the 1201 N in 5 out of 7 Futuremark tests that are on the page. In other words, the 1201 wins 2 out of 7. So how is it behind?
  Third, the gaming conclusions are equally skewed. The reviewer rightly points out that the AMD platform beats CULV and 4500 graphics and Atom. He then points out that it loses to the much higher priced solutions, and writes, "Atom with ION beats the L310 with HD 3200, and CULV with HD 4330 eclipses the ION laptops." Without pointing out that AMD Neo, single core Congo with HD4330 also eclipes the ION laptops in the same chart.
  The point is that the L310 chip competes with Atom on price (but of course not in the Ferrari model). It kills Atom in graphics, it has a faster processor, but loses in battery by a lot. This platform competes with Atom 10 inch netbooks.
  Finally, the conclusions reveals either a desire to skewer AMD or a lack of knowledge about AMD's Nile platform. Not once in the review is it mentioned that this is a year old platform, and the replacement platform has already been announced (with detailed % improvements). Instead, the author writes,
  "What's more, the new Arrandale CULV products should be showing up any time now, with improved CPU and GPU performance relative to the old Core 2 + 4500MHD, at a similar price point."
  No mention of AMD's nile platform that is already in preproduction models everywhere. Should a sentence also read, "The new Nile products show be showing up any time now, with improved CPU and GPU performance relative to the old Congo + 3200HD graphics, at a similiar price point." Why not? I will assume the best - that it is ignorance not malice behind the omission. Instead, we get this: "It's also possible that a 45nm version of the L310 (i.e. the K325) could make a better showing, but it will have difficulty surpassing the current CULV and upcoming i3/i5 ULV platforms." And what knowledge is this based upon? It will certainly be cheaper.
  Also not mentioned is that AMD has many more design wins with Nile than with Congo. I suspect with good reason. Also, the author fails to mention that after the Nile announcement and the design wins were announced, Intel made a 180 degree change and announced that they will now allow their more expensive platform in smaller (12 inch and below) form factors. This is a clear response to AMD's new design wins in that space.
  But don't just believe me. Here's what NVIDIA's CEO Huang said a few weeks ago here - http://blog.laptopmag.com/nvidia-ceo-netbooks-and-...
  
  What’s your take on dual-core netbooks? Aren’t they overkill?
  
  Here comes AMD with their 11.6-inch platform, which is kind of netbookish. So Intel is under an enormous amount of pressure to force Atom to go to dual-core. They kept that at single core for netbooks for the longest time, and now this AMD Nile is sweeping design wins everywhere. It’s put Intel on the defense. Now Intel has pre-announced that they will build a dual-core for Atom, and now the OEMs are putting Ion graphics on dual-core Atom so that they could present a better alternative to AMD.
  
  But will dual-core Atom netbooks with Ion 2 will be affordable?
  
  They will have to be competitive. They will have to be competitive because Intel has no choice but to be competitive. All of a sudden consumers’ expectations are starting to rise, and then all of a sudden competition comes in. Just look at what a netbook’s going to look like come next year. OEMs are going to have to start being creative or else the category will go out of business."
  
  Regards,
  Matt B.

• ggathagan - Wednesday, June 9, 2010 - link

  What part of these statements did you not understand?
  
  "Short of going out and buying the laptops ourselves, getting anyone to send us AMD-based laptops has been difficult at best."
  
  "We didn't choose it specifically, we reviewed what AMD gave us (because it doesn't seem like any of the manufacturers want to give us any AMD based systems). AMD apparently thought the Ferrari One was a good representation of their current mobile platform? Otherwise I don't see why they would have sent it to us."
  
  If you want Anandtech to review different AMD laptops, you will need to:
  A) Convince AMD to send them more current laptops.
  or
  B) Go buy one that you want to have tested and send it to them.
  
  The fact that AMD *doesn't* send them better models says a lot about AMD's opinion of their own products.

• Edgar_Wibeau - Wednesday, June 9, 2010 - link

  I've re-read the article and it points to Intel alternatives several times (i3 alone three times) in many different combinations (eg with ION), many of which being products not currently on the market, which is fine, but only ONCE points to an AMD alternative, and even with a big mistake in this short sentence:
  
  "It's also possible that a 45nm version of the L310 (i.e. the K325)"
  
  Well, the K325 is not a 45nm version, it's a K10.5 vs. a K8. It has better battery life expectation and a more power efficient chipset. Most annoying is that there are even two faster versions which are also not mentioned:
  
  http://products.amd.com/en-us/NotebookCPUSideBySid...
  
  K625 and K665 which, vs. L310, bump up the speed from 1200 all the way up to 1700 MHz, yes, have a slightly higher TDP (15W vs 13W for the L310), but a way better idle efficiency, according to AMD. And double the cache of L310 (which is not listed on products.amd.com). And it uses faster DDR3 (and its low voltage variant) which is not directly coupled to the CPU speed.
  
  Ignoring K625 an K665, because they're called "Turion" vs. "Athlon" is not valid when also mentioning CULV and i3.
  
  So if we assume only 10% higher speed clock-for-clock, we end up with 56% higher raw speed and most likely better battery life.
  
  Not worth mentioning that of course ;-/
  
  Graphics will still be too slow for current games and more than fast enough for 3D desktops and stuff.
  
  Erm, 1080p video was only tested with YouTube using flash? Funny.

• Edgar_Wibeau - Wednesday, June 9, 2010 - link

  Got a permanent error regarding L2-cache size, same for all AMD CPUs mentioned by me. sorry.