Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

When can I buy a laptop with specs similiar to these?

  1. Dec 1, 2005 #1
    5-10 GHz Clockspeed processor (Possibly IBM cell) (possible 2 proscessors)
    4-10 GB Ram
    1-2 TeraByte Hard Drive Space
    Double layer Blue Ray writer
    High Definition
    TV in (coax,rca,s-video,component,fiber audio)
    Two TV Outs simultatiniously (same formats as video in)
    Voice recognitions
    Windows Vista
    Blue ray drive Backward compatible with (SACD,CD ROMCDR,CDRW,DVD+R,DVD-R,DVD,DVD AUDIO)
    64-bit archetecture

    Do you thinkn there will be a laptop on the market by Q3 2008 with these specs? What do you think?
  2. jcsd
  3. Dec 1, 2005 #2
    With the rate that technology changes I'd say it's fairly hard to speculate a date.
  4. Dec 1, 2005 #3
    I think that I that a computer with roughly the specs I described will be on the marked late 2008. Do you think that sounds reasonable? THat is when I plan on buying my next computer.
  5. Dec 1, 2005 #4
    Well, standard laptop hard drive capacity is ~80gb, processor speed ~3GHz, DVD DL writers fot laptops came out, so did 64bit arc, and the voice recognition and 802.11g WiFi. Vista "is sceduled to come out" in 2006, but it'll come out in 2007/2008. your processor might be developed around 2007, your hard drive might be out around 2009, and the RAM, around 2008. when your laptop is released around July 2009, it'll be around $8K, though it may vary from $4K to $10K
  6. Dec 1, 2005 #5
    I dissagree, livingpool. Although a laptop with those specs in late 2008 (if is is out at that time), it would think it would be more like 3500 max. Look at the top of the line computers now, they are about 3500.
  7. Dec 1, 2005 #6
    The best way to figure it out is likely taking a look at the last ten years of the gaming world and how it's demanded the increase in technology. Without gaming we'd still be using PIIs.
  8. Dec 1, 2005 #7
    Sorry about the grammer above, I was taking on the phone while typing.
  9. Dec 1, 2005 #8


    User Avatar
    Staff Emeritus

    I would say it could be possible with a desktop pc, but not a notebook. The trend with notebooks has been on conserver power and battery life than high speed computing.
  10. Dec 1, 2005 #9
    True, but alienware and now dell with the gen xps have released high powered laptops and will continue to do so as gamers move more to mobile solutions.
  11. Dec 1, 2005 #10
    You make a good point dd. I agree with you now that I think about it. But what about mores law. I know that is says that roughly every 18 months th number of transistors on a microprocessor doubles. Is this still happening? I mean has this happened in the last 18 months. Becuase all I really know about is clock speed and about 16 months go there was 3.4 GHz processors on the market, and I think the fastest now is like 3.8. Why has it not doubled? If the number of transistors on the chip is not directly related to the clockspeed, would you be so kind to explain to me what the number of transistors matters in terms of performance? If you don't want to answer the question just tell me and I will look it up. I just kind of would like to have short two way conversation on the subject.
  12. Dec 1, 2005 #11
    And if you all don't mind could you briefly explain to me what is the most important factors of performance in the specs of a CPU, because it is obviously not just clockspeed. I think clockspeed is just easy to advertise, so thats why everyone knows about it. Again, if I am bugging you with questions please just say so and I will shut up.
  13. Dec 1, 2005 #12
  14. Dec 1, 2005 #13
    Thanks friend!
  15. Dec 1, 2005 #14

    When I read the thread title I thought it said where, so consequently I burst out laughing when reading the specs.

    Forget Cell and Vista, not gonna happen at all. I'd be surprised to ever see an MS OS run natively on the Cell. I think they're too much in bed with Intel and x86 for that.

    That said, you're looking more 2009 at the soonest. You might get a desktop like that by then, but not a laptop. As for 5-10 GHZ, prolly not ever gonna happen. We seem to have it the limit of practical clockspeed increases at about 4 GHZ. Its simply much easier and cheaper to double the number of processors than double the clock speed anymore. Look for parallel computing, not faster computing.

    As for voice recognition, why? Keyboard input is far, far faster. I can easily type faster than I can talk, so I don't any use in voice recognition really, other than reducing carpal tunnel. Not something I'm interested in personally.
  16. Dec 2, 2005 #15
    Not to mention the cost of that laptop. At such a price you'll be better off with a high end gaming machine with 2 30-inch Apple displays and 400 watt logitechs.
    Last edited: Dec 2, 2005
  17. Dec 2, 2005 #16
    Well if this laptop is not out by 08 then I will build it from scratch. Could someone give me some tips on how I can make a microprocessor with a sewing machine, my solding iron, and a hot glue gun?
  18. Dec 2, 2005 #17
    In a laptop form factor, I think probably never, unless you expect to carry a backpack around for a power source, battery energy density increases dramatically or you keep it permanently plugged into wall power. HD storage density would have to increase by an order of magnitude to squeeze a TB into a 2.5" FF drive. CPU speeds seem to have plateaued for the moment at 3.2-3.8 GHz, with Intel and AMD starting to focus more on increasing performance/watt rather than performance/GHz. Most laptops only have room for 2 memory slots, so memory chip density would probably have to double to reach your specs.

    The rest of it is already doable or is probably not very far away.

    Naturally, I expect my prediction to be proven in 2-3 years.
  19. Dec 2, 2005 #18
    Are we not getting to a point now, where it is becoming increasing harder to get higher clock speeds with the technology we use?

    Anyway, why on earth would anyone need something that powerfull.. I cant think of any reason on the workstation side of things to need 10G of mem.. Of course on servers running some enterprise DB it is already in use..

    The Storage space is the only thing I can see that is warrented
  20. Dec 2, 2005 #19


    User Avatar
    Science Advisor

    There's something called the "speed of light limit". If the clock is too fast, then, of two consecutive ticks, the second tick might occur before the first tick has had time to propagate across the whole chip. You can visualize this with a very large CPU chip, 300,000 Km long for instance. If that chip's clock ticks more than once a second, then before the first tick reaches the outer limits of the chip, some components of the chip have already received and acted on the second tick (the ones closer to the clock). Computers are generally designed with a synchronous model in mind (that's why we have the clocks on the chips) so this isn't good. For a 6*6 cm synchronous chip, light would need at most 2*10^-10 seconds to travel 6 cm. So the clock shouldn't tick more than 10^10 times a second. That's what, about 10Ghz? I should probably check this stuff online, but a 10Ghz synchronous processor might not be even possible on a 6 cm chip, (you'd have to make much smaller, keep the clock in the center, or delay the components a bit depending on distance from the clock)).
    In order to get really fast we will eventually need asynchronous designs, probably with multiple processors, or chips with multiple sub-processors.
    Before we ever get to 10Ghz it would probably be must faster and cheaper to have 5 2Ghz processors, or 2 5Ghz processors (if the motherboard is well designed). There's always some overhead in producing a really fast processor, the architecture might need to change significantly, so in the end the CPU might be able to perform instructions really fast but, on average, the # of instructions per high level commands or # of operations per instruction might be much higher. Of course there's also some overhead in the implementation of multiple processors. For one, storing/retrieving memory contents for use in instructions becomes more tricky, but also there will a slowdown from a software perspectivee because, to prevent deadlocks or data corruption the OS will have to restrict the amount of parallelelism that is actually used.
    However, there's no "speed of light limit" with asynchronous machines and you would be able to use currently available processors, the difference would be in the motherboard and OS.
    I also think that, in the future, clock speed will become less important because, as wireless networks expand and become more powerful, we may be getting close to the point where PCs will be simple machines with an internet connection where most of the processing is actually done in very powerful servers off somewhere.
    Last edited: Dec 2, 2005
  21. Dec 2, 2005 #20


    User Avatar

    Staff: Mentor

    Some quick reseach shows

    -Intel released the 500mhz P!!! sometime in mid-1999
    -Intel released the 1ghz P!!! in mid-2000 (approximately 12 months)
    -Intel released the 2ghz P4 in August-2001 (approximately 14 months)
    -We're still waiting for 4ghz (84 months and counting)

    I'd say Moore's law is pretty much dead.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook