news asus ivy bridge will usher in 600 ultrabooks
News

Asus: Ivy Bridge will usher in $600 ultrabooks

Talk of ultrabooks seems to be on everyone’s lips lately, as systems from Acer, Asus, Toshiba, and others gear up for retail. Some of those machines will no doubt be solid deals, but the real bargains could come next year. As DigiTimes reports, Asus is planning to unleash ultrabooks priced as low as $600 next spring.

Word comes from none other than Asus CEO Jerry Shen, who quoted a price range of $600-900 for Asus’ future Ivy Bridge ultrabooks. Those laptops will reportedly be out in April 2012. For reference, DigiTimes says Asus has "tentatively" priced its impending batch of Sandy-based ultrabooks at $899 and up.

If next year’s ultrabooks indeed cost as little as $600, they could become dangerously tempting alternatives to cheaper, full-sized notebooks—especially if slim enclosures don’t induce unsavory tradeoffs in terms of performance and battery life. There’s definitely something to be said for an ultra-slim, ultra-light machine one can easily throw around on the couch or slip inside a messenger bag.

Perhaps this all has something to do with Intel’s recently disclosed plans to shoot for 10-15W thermal envelopes with its future mobile chips. Right now, standard-voltage Sandy Bridge CPUs have 35W TDPs, and Intel slaps a premium on chips with 25W and lower thermal envelopes. If today’s low voltage is tomorrow’s standard voltage, then tomorrow’s entry-level laptops will no doubt be slimmer and lighter.

0 responses to “Asus: Ivy Bridge will usher in $600 ultrabooks

  1. Um… yes we have. How did I keep missing that “best” comment.. while ridiculing [i<]your[/i<] reading comprehension? NeelyCam fail. +1; clean win, well done!

  2. So, in fact, if the highest binned 15W Ivy Bridge CPU featured a stock clock of 2.0GHz, it might be a little “meh”?

    I think we’ve come full circle.

  3. So if the high end 15W Ivy Bridge part featured a 2.0GHz clock speed, you would be satisfied?

    Yes, I understand that it would easily turbo over 3.0GHz, but that’s not the point.

    The question is, would 2.0GHz be an acceptable clock speed for the highest binned 15W IB CPU?

  4. [quote<]I'm rather hoping that the $600 is a real price point for a reasonable ultrabook.[/quote<] That, right there. That's my problem. You can't expect the price [b<]floor[/b<] to provide a reasonable product. It's like in netbooks, there were $200 netbooks, but they were rubbish. They existed and they were definitely netbooks, but $100 more got you better build quality and longer battery life. Ultrabooks will be the same. $600 might be a price point, but you'd be mad to expect a "reasonable ultrabook" at that price. For $700-800? Sure, but not $600.

  5. Because Intel already offers a 1.8GHz 17W CPU. If their highest binned 15W CPU ran at 2.0GHz, it would be a bit of a disappointment.

  6. I guess people didn’t want to pay a premium for lower performance but with lighter weight and better battery life. They also tended to be 14” or smaller although there were some exceptions so maybe people prefer bigger screens also!
    I’d be curious to see a breakdown of laptop sales by price ranges to see how much people are spending.

  7. I didn’t say anything about how good the products are. I was comparing the bottom-end SB ultrabook to a bottom-end IB ultrabook, to get a sense of how the cost of the platform changes from this generation to the next. Why are you bringing up a mid-range gaming desktop into this, and comparing it to a non-gaming desktop? That has nothing to do with my comment.

    Not only your reading comprehension is out of whack, but also your logic. Maybe get some sleep, and try again tomorrow.

  8. Reading comprehension? I said a 2GHz/15W CPU wouldn’t be a meh-chip, compared to SB or anything from AMD. Keep in mind that Ivy Bridge will have better graphics, and potentially idles lower and/or has better turbo.

    Edit: re-reading your original comment, I’m getting a feeling that you don’t understand what TDP means, and how Turbo is almost entirely independent from it. Just because SB can turbo to 2GHz doesn’t make it equivalent of a 2GHz baseline part at 2W lower TDP.

  9. [quote<]There's definitely something to be said for an ultra-slim, ultra-light machine one can easily throw around on the couch or slip inside a messenger bag.[/quote<] I'd think twice before throwing a thin laptop around or place inside a bag that I'm likely to throw around. Chances are the manufacturers had to sacrifice durability to make the thing thin and still affordable by not using exotic, durable materials.

  10. It’s the price [b<]floor[/b<], why would you expect a good product at the absolute bottom end? That's like saying you expect to get a mid range gaming desktop for $500 just because some computer is sold at $500. The $500 desktop is junk!

  11. You are aware that Sandy Bridge already offers a 1.8GHz 17W i7 CPU, right?

    So a 0.2GHz bump from i7 Ivy Bridge would be completely acceptable? Really?

  12. No kidding. 2.4GHz would certainly be acceptable, 2.0GHz would not.

    Oh and it’s poor form to double post. Edit your original post.

  13. So, speculation, but no one has actually heard anything–like a paper at Hot Chips or similar?

  14. I don’t know why you’re using “medium” prices for your math… Comparing the cheapest one to the cheapest one seems more reasonable.

    The $1000-$2000 seemed like Asus whining about Intel’s margins, in order to get everyone to freak out, with the ultimate goal to get Intel to cut the prices (instead of trying to cut margins itself). I would be very surprised to see Asus release the i7 version at $2000, when Acer is expected to sell one for some $1200-$1300:

    [url<]https://techreport.com/discussions.x/21578[/url<] I think more realistic apple-apple prices are $900->$600 for low-end, $1300->$900 for high-end.

  15. I never understood why CULV wasn’t successful… Real laptop performance in a netbook-like package, with great battery life and minor price premium.

  16. Well, some Intel slides describe it as “Tick+” which has been interpreted many different ways. The larger transistor budget could be spent in some [url=http://www.agner.org/optimize/blog/read.php?i=142<]notable areas of improvement[/url<] such as boosting the micro-op cache, changes in the L1/L2 cache to increase bandwidth and/or reduce bank conflicts, etc. None of those would constitute a major architectural change but could result in a nice (albeit single digit percentage) boost in IPC (especially on AVX and other bandwidth-heavy workloads). Of course they could be referring to the supposed IGP improvements (eg DX11/OpenCL support, increasing EUs) or something else (hardware random number generator) that has no bearing on CPU IPC whatsoever.

  17. Those are not my numbers, but the one from Techreport and other news sites.

    The first released of Asus ultrabook later this year are reported to be in the 1000 to 2000 range (Check the news page),
    and asus today stated that if you wait April this will drop to 600 to 900.

    Average both price range and what do you get ? asus medium ultrabook price in late 2011 : $1499
    asus medium ultrabook price in early 2012 : $749

    Now, whats is 1499$ divided by two? surprise!

  18. Well if the Ultrabook platform is as unsuccessful as CULV then we can expect to get bargain $600 laptops in 2012. I actually hope that it does well otherwise it will be another nail in the coffin of the hope for premium design laptops at reasonable prices.

  19. I’m rather hoping that the $600 is a [i<]real[/i<] price point for a reasonable ultrabook. I think it's a bit unrealistic, but I'd be very happy if it turned out to be true. I have a Core2Duo CULV ultraportable with an SSD in it that was well below $600... so I think it's possible.

  20. HP CEO recently told press:
    “Don’t buy an HP product now, we’re going to sell the whole division next year. Or maybe spin it off. Or maybe close it. Anyway, with all that uncertainty over its future, maybe you should just look elsewhere”

  21. I’ve assumed from the beginning and also hope that the $600 price point is bogus as I can’t see that a premium class Ultrabook with an Intel CPU at $600 is a serious proposition in only 7 months time. It’s possible that we’ll see cheap ass Ultrabooks at $600 that use a very compromised platform but that would make a mockery of the whole Ultrabook label surely!

  22. Either math fail, apples-oranges comparison or suspect information.

    $600 is 40% cheaper than $1000. Moreover, $600 is 33% cheaper than $900 (which is where the cheaper SB ultrabooks are expected to be).

    $2000 SB ultrabook sounds pretty ridiculous…. nobody’s gonna buy that. If they actually build it, I doubt one should compare it to a $900 Ivy Bridge ultrabook; a $2000 SB ultrabook would come with a 500GB SSD, a mad-highend display and a can of pixie dust. Those would push the IB version to at least $1300. (Whoa – there’s that 33% again!)

  23. [quote<]If the best we get is a 2GHz 15W part, then Ivy Bridge will be pretty meh.[/quote<] I'm not sure I would call that a meh, considering that it is quite a bit more capable (CPU- [i<]and[/i<] GPU-wise) than Brazos or even Deccan while having a lower "TDP" (although TDP doesn't really measure power efficiency that well..), and a lot of people don't call those 'meh'

  24. $600 would still be too close to an Ultrabook (which also has an SSD). If this $600 target for Ivy Bridge Ultrabook is real and achievable, that’ll automatically push Brazos/Deccan netbooks to a lower price point ($450-$500, and that’s with an SSD).

  25. Quoting Techreport on asus ulrabook:
    “Word is that UX-series variants with Core i5/i7 chips and solid-state drives will be priced in the $1,000-2,000 range”

    $600-900 in Arpil vs $1000-2000 in november.

    Thats not 33% and thats not 7 month.

  26. I think you’re onto something here…

    Price drop could be a result of multiple reasons. Part of it could be the SSD/memory/LCD prices (like smilingcrow said). Ivy Bridge could contribute to the savings because of lower power consumption: cheaper cooling solutions vs. Sandy Bridge ultrabooks, cheaper voltage regulation modules, smaller/cheaper battery with less capacity..

    The CPU price itself could be cheaper, as the area is likely to be smaller than SB CPUs. And although the price points have seemed ‘fixed’ in the past like Goty pointed out, those are fixed mainly for desktop CPUs people can by on Newegg etc. We as consumers don’t really have any visibility to what the [i<]actual[/i<] prices are that the OEMs like Asus are paying... I'm sure they have negotiated lower prices that have little to do with "list prices".

  27. Not if they use/abuse that TDP adjustment thinger. They don’t need to be in “ultrabooks,” anyways. There are plenty of very light and thin laptops that run 35w CPUs.

  28. “They don’t offer laptop quad-core i5s right now.”

    They could offer 25W quads with IB but that’s still too high a TDP for Ultrabooks seemingly.

  29. “It’s not about the CPU; it’s that the lower TDP will make cooling cheaper and the design less custom”

    They are going to be thin already so I can’t see that IB will impact that much. If they have to redesign and retool for thinner IB Ultrabooks that will impact on the cost savings anyway.

    SB is at 17W so unless they move to 10W I can’t see that it will have much impact on cooling. Even for this class of system the cost of cooling is fairly small and won’t even make a dent in this supposed $300 difference.
    Maybe the $300 price difference is based on the idea that TFTs, SSDs etc will have dropped in price significantly by April!

  30. It’s not about the CPU, it’s that the lower TDP will make cooling cheaper and the design less custom, because of how difficult is to cool a CPU in the very narrow body of this kind of laptop.

    A Macbook Air costs more to make than a Macbook almost entirely because of the low profile and the implications for cooling and fitting everything in.

  31. Asus CEO just told the press:
    “Dont buy an ultrabook this year, we have going to cut the price in half in a few month”

  32. “The 17W Sandy Bridge i5 2557M tops out at 1.7GHz.”

    My bad, I only looked at the UM parts so I missed the fact that they already have the 1.8GHz 17W i7-2677M.
    So if they manage a 50% bump in GHz/TDP we are looking at 2.4GHz at 17W which suggests that we can even expect reasonable performance at 10W.

    “Although an “i3” branded 2GHz 15W part would be impressive.”

    In practice if the 22nm process inherently supports 2.4GHz at 17W then the i3 will offer a similar level of efficiency regardless of whether Intel artificially labels them with a higher TDP as a way to create market segmentation. Look at the way that Intel uses the 65W and 35W labels on desktop CPUs. It suggests a much larger difference than actually exists to the uninformed.

  33. It’s certainly a great APU but it’s in a completely different class performance wise so I think a much smaller percentage of people would find it agreeable for use in a laptop. I can’t imagine paying $600 for a Laptop with such low CPU performance.
    But if you took a $400 Netbook with an E-450 and upgraded to an IPS screen and added an SSD to give a $600 Ultra-NetBook then I can see it makes sense at that price point.

  34. [quote<]that is powerful enough for a large percentage of users[/quote<] So is an E-450, and that really [i<]is[/i<] possible in a $600 machine - right now, too.

  35. They don’t offer laptop quad-core i5s right now, and they could throw out the dual-core i7s if they change that. That’s the only way I could see prices “dropping” on Intel’s end.

    It’s not as if they’re even going to care if laptops are using $20 ARM chips at that point. They have Atom to cover that territory.

  36. The problem with this is that Intel doesn’t change price points, they just offer more performance or more features at those price points. IB chips won’t be cheaper than SB chips, they’ll just be a bit faster and/or consume less power.

  37. These will be $600-$900 the same way that these first-gen ultrabooks are “under $1000″… after Intel foots the bill. Give me one of these with an E-350 (or Krishna/Wichita when they are released). That would drop the cost significantly and I’d be willing to pay a premium over most netbooks out there today just to get a decent chassis.

  38. In additiona to what others have mentioned, process shrinks make for cheaper chips on a per-transistor basis.

  39. Thanks. Even the top end i7 low power CPUs top out at $317 so there isn’t that much of a margin to reduce pricing. Regular i5 CPUs start at $225 and I doubt that i3 start at less than $150 so it’s unlikely that the entry level systems will save more than $100 in the shift to IB.

  40. SB Ultrabooks use ULV processors. Probably IB Ultrabooks can work with standard voltage processors – this and a bit of scale economy might lower the prices to 600$.

  41. Exactly, as Intel’s mobile dual cores top out at $346 so how can they knock $300 off the price of an entry level model?

  42. They currently have 1.46GHz at 18W so 2GHz at 15W offers ~64% more GHz per Watt which sounds impressive to me. What are you expecting then?
    On top of that there will also be IPC gains.

  43. If the [i<]best[/i<] we get is a 2GHz 15W part, then Ivy Bridge will be pretty meh. The 17W Sandy Bridge i5 2557M tops out at 1.7GHz and turbos well past 2.0GHz. I expect more than a 0.3GHz bump for i5 from a die shrink. Although a "i3" branded 2GHz 15W part would be impressive.

  44. Performance increases will allow i3-class parts to get the job done and power savings will allow cheaper cooling.

    $600 is the floor, so Ultrabooks at that price won’t be impressive, but a few might exist.

  45. I dont see how changing the CPU (to a newer, potentially faster one) will knock $300+ off the retail price…

  46. If their 22nm process is as power efficient as early reports suggests then it makes a lot of sense for Intel to move the mainstream mobile CPUs to a much lower TDP range. I think they should be pre-empting the forthcoming onslaught from ARM and making lower power CPUs at mainstream prices is a start.
    A 2GHz i3/i5 with a 15W TDP should be possible with IB and that is powerful enough for a large percentage of users; CPU wise anyway.