Will Intel or AMD sell a cpu faster than 4GHz in the next five years?
 

Who here thinks Intel or AMD will sell a processor within the next five years that's faster than 4GHz at stock voltage? And how the heck do I get the polling function to work(checking the box doesn't seem to work)

"Moore's law" has nothing to do with clock speed.

Five years is waaay too conservative. Should be two years (three at the most) before 4 GHz is done-and-dusted, most likely by /both/ Intel & AMD.

By that time, both the core count as well as GFLOPS rating/core would have also tripled. By five years, some kind of GPU-on-chip would've brought TFLOPS capability to the CPUs.

Man, can I predict them or what! :geek:

I think so, but before that, I think we'll see a popular 8 core PC

[COLOR="DarkOrange"]Entirely irrelevant. Doing existing instructions in a fraction of the amount of clock cycles it takes now is much more promising. Besides, LL tests can be very easily done in parallel by doing multiple LL tests at the same time. The Core 2 Duo has less cycles / second than the Pentium IV and does more work at lower electricity costs. The real question should be: shall we see processors running at lower speeds than the Core 2 Duo doing more work in the same time?[/COLOR]

[quote=tha;102690][COLOR=DarkOrange]Entirely irrelevant. Doing existing instructions in a fraction of the amount of clock cycles it takes now is much more promising. Besides, LL tests can be very easily done in parallel by doing multiple LL tests at the same time. The Core 2 Duo has less cycles / second than the Pentium IV and does more work at lower electricity costs. The real question should be: shall we see processors running at lower speeds than the Core 2 Duo doing more work in the same time?[/COLOR][/quote]
I don't think it's irrelevant at all. The original question essentially asks essentially for opinions on Moore's law, such as whether it will begin to fail now. Your question essentially asks whether processors will improve.

[QUOTE=dominicanpapi82;102779]I don't think it's irrelevant at all. The original question essentially asks essentially for opinions on Moore's law, such as whether it will begin to fail now. Your question essentially asks whether processors will improve.[/QUOTE]
I considered this possibility, actually. I decided to limit things to one choice a vote, since the first and second choice contradict each other. Also, there's more than one way to interpret the question. I did indeed realize that the future probably has more instructions per clock in store for us.

Perhaps I should have left out the last choice.

[QUOTE=jasong;102836]I did indeed realize that the future probably has more instructions per clock in store for us.[/QUOTE]

From the Intel briefing mentioned in an article in [URL="http://www.mersenneforum.org/showthread.php?t=7681"]another thread[/URL] in this subforum it looks to me as the next big step for us is the four bit divider versus today's one bit divider.

[QUOTE=tha;102888]From the Intel briefing mentioned in an article in [URL="http://www.mersenneforum.org/showthread.php?t=7681"]another thread[/URL] in this subforum it looks to me as the next big step for us is the four bit divider versus today's one bit divider.[/QUOTE]
Looks like your pimping a pimping(metapimping?), but at least it's on topic. ;)

I don't think there will be a 4ghz chip for many years to come. Maybe never, there seems to be a gremlin or troll under the 4ghz barrier.

Man, I'm enjoying digging up old threads...
 

It's been nearly three years since the original post, so I guess I'll bump this sucker up.

Right now, there are only two years left before we'll know the answer to the question in this poll, and it looks like it'll come down to the wire.

So far, the fastest stock processor still has not beaten the 3.8GHz record set by the Pentium 4 in 2005. According to the usually-reliable wikipedia, Intel will break that record with a 3.866 GHz processor, the core i5-680, which will be coming out a few months from now:

[url]http://en.wikipedia.org/wiki/List_of_future_Intel_microprocessors[/url].

The 3.866 GHz speed is only available with turbo boost, but I'll assume that counts because nothing in this thread says that a 4GHz computer has to have that speed sustained when all cores are used. However, the Wikipedia article doesn't mention processors coming out after Q3 this year, so the one and a half year period between then and April 2012 will be filled with wild guesses and speculation. Here are my thoughts:

2011 will be the year that diminishing returns from increasing cores begins to play a major role. For 99% of people, there's little to no difference between 4 and 6 cores because most programs aren't multi-threaded. Because of that, it makes more sense to increase clock speed than to increase the number of cores, but the main question is when Intel and AMD will finally realize that.

Anyway, the poll has 20 "yes" votes and only 4 "no" votes right now. If it were possible, would anyone like to change their votes?

I saw water-cooled processors for PCs running well over 4 GHz last year...

Luigi

[quote=axn;102659]Five years is waaay too conservative. Should be two years (three at the most) before 4 GHz is done-and-dusted, most likely by /both/ Intel & AMD.[/quote]Lessee ... "01 Apr 07" plus three-at-the-most-years is ... next Thursday.

[quote]By that time, both the core count as well as GFLOPS rating/core would have also tripled.[/quote][quote]Man, can I predict them or what! :geek:[/quote]I love youthful optimism! :smile:

- - -

And, to have the energy of your avatar, axn ... !

Well I missed the boat on this one... but starting from NOW my answer is no, if Turbo Boost doesn't count. If Turbo Boost counts, then my answer is Yes.

[QUOTE=Historian;209755]For 99% of people, there's little to no difference between 4 and 6 cores because most programs aren't multi-threaded. Because of that, it makes more sense to increase clock speed than to increase the number of cores, but the main question is when Intel and AMD will finally realize that.[/QUOTE]But having your virus checker running 24/7 in the background, the OS, and various other programs all at once (IE and a media player and whatever else), the number of cores will be noticed. Everything will run faster and smoother.

[QUOTE=Historian;209755]
it makes more sense to increase clock speed than to increase the number of cores, but the main question is when Intel and AMD will finally realize that.
[/QUOTE]
Is it your position that Intel and AMD [i]could[/i] give us a 5GHz processor now, or at least soon, but because of their naivete actually [i]choose[/i] not to?

Robert Colwell was the chief architect of the P6 design team at Intel, and in his book 'The Pentium Chronicles' writes, about his leaving Intel after helping design the Willamette,
[code]
I felt that as one of the principals who had led the company to a high-
clock-rate x86 strategy, I should have been able to lead it away from that
strategy when it became necessary (and we knew from the beginning that
eventually it would). But it seemed to me that that time came around 1998,
and over the next two years I was unable even to make a dent in the
product road maps
[...]
Beyond all of that, however, was a looming thermal power wall that was
no longer off in the distance, as in P6, but instead was casting its long, ugly
shadow directly over everything we did. That experience was primarily why
I was so sure I did not want to work on any high-clock-rate chips beyond
Willamette. I just did not think that there would be enough end-user
performance payoff to justify the nightmarish complexity incurred in a
high-performance, power-dominated design.[/code]
See also [url="http://groups.google.com/group/comp.arch/browse_thread/thread/e0416fdf5a4b7fb8/de9dc2712561a03f?hl=en&ie=UTF-8&q=%22forrest+curve%22#de9dc2712561a03f"]the Forrest Curve[/url]

[QUOTE=Historian;209755]
The 3.866 GHz speed is only available with turbo boost, but I'll assume that counts because nothing in this thread says that a 4GHz computer has to have that speed sustained when all cores are used. However, the Wikipedia article doesn't mention processors coming out after Q3 this year, so the one and a half year period between then and April 2012 will be filled with wild guesses and speculation. Here are my thoughts:

?[/QUOTE]

Go read my paper from IEEE Computer: Exposing the Mythical MIPS Year.

Note that my I7 processor as 2.4GHz is faster per core than my 3.4GHz
older P IV.

Merely looking at clock rate and assuming that faster rates
means faster processors is just plain stupid.

[QUOTE=R.D. Silverman;209798]Note that my I7 processor as 2.4GHz is faster per core than my 3.4GHz older P IV. [/QUOTE]That is i7 not I7.

[quote=jasonp;209796]Is it your position that Intel and AMD [I]could[/I] give us a 5GHz processor now, or at least soon, but because of their naivete actually [I]choose[/I] not to?[/quote]Speaking for myself (Historian may differ):

" naivete "? No ... and No to the first part, too.

[quote]Beyond all of that, however, was a looming thermal power wall[/quote]That's a real problem, not naivete.

[quote]See also [URL="http://groups.google.com/group/comp.arch/browse_thread/thread/e0416fdf5a4b7fb8/de9dc2712561a03f?hl=en&ie=UTF-8&q=%22forrest+curve%22#de9dc2712561a03f"]the Forrest Curve[/URL][/quote]... and that's another non-naive problem.

The reasons for being stuck at sub-5GHz nowadays are real (but won't last forever).

[quote]But having your virus checker running 24/7 in the background, the OS, and various other programs all at once (IE and a media player and whatever else), the number of cores will be noticed. Everything will run faster and smoother.[/quote]
Are 2 cores better than one? Yes.
Are 4 cores better than 2? Yes.
Are 6 cores better than 4? Probably not, for most users.
Are 8 cores better than 6? No, unless you're running a server.

For most people, the law of diminishing returns starts to kick in above 4 cores. Think about it, if it were not for DC projects, how often do you actually need to use more than 4 cores? It's pretty rare for people to have more than 4 programs running at once.

[QUOTE=jasonp;209796]Is it your position that Intel and AMD [i]could[/i] give us a 5GHz processor now, or at least soon, but because of their naivete actually [i]choose[/i] not to?
[/QUOTE]
My position is that Intel and AMD cannot give us a 5 GHz processor right now or in the near future even if they want to. 4GHz? It depends*. But if Intel can, it won't release a 4GHz processor not because they're naive, but because of business reasons. Their main competitor, AMD, isn't a big threat yet, so there's no need to release a faster processor at the moment - you'd basically be competing against your own products at the high end.

*Yes if they continue developing dual and quad cores, no if they're trying to increase the number of cores beyond 6 cores instead of bumping up clock speed.

The crowd that visits here is in the odd position of belonging on the horizontal axis of the Forrest curve; give everyone on Mersenneforum a 32-core processor and they'll all run Prime95, or some other computationally demanding application, 31 times. Likewise, give me a processor with lots of cores and I'll find something to do with it.

The Forrest curve post appears at regular intervals in the usenet archives, back into the late 1990s, and your concern that real-world performance eventually won't improve has always followed it. If you literally only use email, browse the web and use office applications then you haven't needed to buy a new computer for the last decade (my wife's computer dates from 2001), and the number of cores you 'need' is irrelevant, it just has to be nonzero.

On the other hand, if you do have demanding computational needs, and rely on commercial software to address those needs, then you should indeed hope that the people who develop that software can incorporate multithreaded or parallel primitives into updates or new products, or at least use standard libraries (BLAS, FFTW, MKL, Accelerate, Intel's compiler tools) that do this already. If they do not, and the incremental single-thread performance gain that the next generation of processors provides is not enough for you, that's a business opportunity and not a liability.

It has always been true that software which is computationally demanding cannot be written once and then expected to accelerate forever. Creating demand for innovative software products is something the software industry as a whole should welcome; why are they getting paid otherwise?

Of much more interest to me are other issues: suppose that the amount of compute power and memory you have to solve your problems is suddenly much larger than you are used to. How does that change the way you solve the old problems? Also, what sort of currently impossible solutions become feasible, opening up new application domains?

Filesystems that never delete anything? 3-D raytraced home movies that are rendered in the background? There was a lot of hand-wringing about how the internet would make all these desktop computers obsolete because they are just platforms for a browser most of the time. Is that really true? Nobody can find anything to do with all this storage and low latency computation that couldn't be done through a web interface?

[quote=jasonp;209836]give everyone on Mersenneforum a 32-core processor and they'll all run Prime95, or some other computationally demanding application, 31 times.[/quote]
Only using 31 of 32 cores constantly... :huh:
What a bizarre concept...

[quote=Mini-Geek;209837]Only using 31 of 32 cores constantly... :huh:
What a bizarre concept...[/quote]

Not a good concept - Windows tries to distribute the workload to all processors. It's better to give the 32 cores to Prime95. It won't be able to perform that well with it anyway (because of memory access constraints).

I like the new 12 core opteron - get 4 of them and you got an 48 core system. Would be nice to play around with this guy.

[quote=axn;102659]Five years is waaay too conservative. Should be two years (three at the most) before 4 GHz is done-and-dusted, most likely by /both/ Intel & AMD.

By that time, both the core count as well as GFLOPS rating/core would have also tripled. By five years, some kind of GPU-on-chip would've brought TFLOPS capability to the CPUs.

Man, can I predict them or what! :geek:[/quote]

The cpu manufacturers have problems increasing the speed of one core (because of physical constraints). That's the reason they are more likely building more cores than more speed for less cores.

[quote=Historian;209827]Think about it, if it were not for DC projects, how often do you actually need to use more than 4 cores? It's pretty rare for people to have more than 4 programs running at once.[/quote]Oh? Task Manager says I have 13 processes going ... and that's just for my current Windows login, without including prime95 independently executing under a different login.

It's true that the Firefox browser is the only one with a user interface I'm using right now, but the other 12 [I]are[/I] there, and at least my antivirus software is using 1% on its own without any indication but its tray icon. I'm fairly sure they aren't all executing on CPU 0.

[quote=cheesehead;209862]Oh? Task Manager says I have 13 processes going ... and that's just for my current Windows login, without including prime95 independently executing under a different login.

It's true that the Firefox browser is the only one with a user interface I'm using right now, but the other 12 [I]are[/I] there, and at least my antivirus software is using 1% on its own without any indication but its tray icon. I'm fairly sure they aren't all executing on CPU 0.[/quote]

I've got 52 processes running but that doesn't matter. Most of them are sleeping (because of I/O-queuing, ...) so you get along really well with one or two cores. The reason is that cpu scheduling and the round robin method is time slicing your processes ... Prime95 on the other side tries to get as most cpu cycles as possible (if no other process needs it).

I contend that, as more "smart" software is developed to perform auxiliary information services for users, making more use of multithreading, the average numbers of cores actually in use simultaneously will continue to rise. The average user will not necessarily be aware of how much computing is going on in the background, just as he is not aware of how much more computation it takes to produce and maintain a fancy graphics interface than it did to show only text on the old green screens. But a few-CPU system will become less satisfactory under Windows LXIV than it was under Windows XP, and not just because it's from Microsoft.

I'm willing to agree that this won't be noticeable in just a couple of years, but it's coming. Doing all that fancy future graphic interaction, as (supposedly) depicted in "Minority Report" (and other recent movies) will burden more and more cores, I think. It's a consequence of the desire to have more "natural", intuitive and informative displays.

[QUOTE=Historian;209827]Are 2 cores better than one? Yes.
Are 4 cores better than 2? Yes.

[/QUOTE]

What compels people to make dogmatic statements about a
technical subject in which they are relatively ignorant??

Whether more cores is better depends upon many factors,
and the answer iS [b]NOT[/b] always yes.

Such things as cache contention, inter-thread communication, and
memory bottlenecks (among others) can actually make more cores
SLOWER.


[QUOTE]
For most people, the law of diminishing returns starts to kick in above 4 cores.
[/QUOTE]

Irrelevent. The people in [b]THIS[/b] forum are not "most users"

[QUOTE]

My position is that Intel and AMD cannot give us a 5 GHz processor right now or in the near future even if they want to. 4GHz?
[/QUOTE]

Unless you work for Intel or AMD, you have no information upon which to
base the above assertion. Which makes it pretty much worthless.

[QUOTE=Mini-Geek;209837]Only using 31 of 32 cores constantly... :huh:
What a bizarre concept...[/QUOTE]

It leaves one free to do interactive work; surf the net, read email,
etc.

[QUOTE=R.D. Silverman;209916]It leaves one free to do interactive work; surf the net, read email,
etc.[/QUOTE]I don't see how that is useful. I thought that was why we have priorities, so that interactive tasks can still happen for us even when other long term tasks are running. Please do elaborate as to why we need to "save" one CPU for such things.

[quote=retina;209919]I don't see how that is useful. I thought that was why we have priorities, so that interactive tasks can still happen for us even when other long term tasks are running. Please do elaborate as to why we need to "save" one CPU for such things.[/quote]

No it isn't usefull. The cpu scheduler (at least from Linux) tries to distribute it fairly to all cores. CPU affinity may be another thing but I don't see a great advantage in that concept.

[quote=R.D. Silverman;209915]What compels people to make dogmatic statements about a
technical subject in which they are relatively ignorant??

Whether more cores is better depends upon many factors,
and the answer iS [B]NOT[/B] always yes.
[/quote]
I was speaking in the general case. Of course, there are going to be some exceptions. If you still can't get the point, try this analogy:

"Are women shorter than men? Yes."
Of course, the answer iS [B]NOT[/B] always yes. Some women are taller than some men, but in general, the average woman is shorter than the average man.

[quote]
Irrelevent. The people in [B]THIS[/B] forum are not "most users"
[/quote]So what? Intel and AMD make processors that target the general population, not just mersenneforum users.


[quote]
Unless you work for Intel or AMD, you have no information upon which to
base the above assertion. Which makes it pretty much worthless.[/quote]Both a friend of mine and a distant relative work for Intel, so I do have a bit of inside info. Also, I have a big motive to know what's going on in there since I have several thousand shares of Intel stock.

[quote=cheesehead;209862]Oh? Task Manager says I have 13 processes going ... [/quote]
Most of them aren't CPU intensive, though. I have 54 processes going, but the CPU Usage is almost always below 5% when my DC programs are turned off.

[quote=Historian;209935]Most of them aren't CPU intensive, though. I have 54 processes going, but the CPU Usage is almost always below 5% when my DC programs are turned off.[/quote]

Sure most processes are sleeping (mean they are waiting for a user input or some data from a harddisk/nic) ...

[QUOTE=Historian;209933]I was speaking in the general case. Of course, there are going to be some exceptions.

[/QUOTE]

Hey moron. Your bold assertion made NO allowance for exceptions.
The rest of us can read.



[QUOTE]
Both a friend of mine and a distant relative work for Intel, so I do have a bit of inside info. Also, I have a big motive to know what's going on in there since I have several thousand shares of Intel stock.[/QUOTE]


Horseshit. You are just trying to cover your ass and trying to
imply that you know something that others do not. If your soi-dissant
"friend" and "relative" are conveying inside Intel information, then they
are most assuredly breaking insider trading laws.

<plonk>

[quote=R.D. Silverman;209953]Hey moron. Your bold assertion made NO allowance for exceptions.
The rest of us can read.
...
If your soi-dissant
"friend" and "relative" [/quote]
The exceptions are understood; nobody else seemed to mind that statement. BTW, what's "soi-dissant"? It's spelled with one s, not two:
[URL]http://www.thefreedictionary.com/soi-disant[/URL]

Are you too lazy or retarded to use your spell checker today?

[quote]
Horseshit.
[/quote]I thought this forum was supposed to be kid friendly? Whatever, I can care less whether or not you shut the fuck up* because you are now the first person to be on my ignore list.

[quote]You are just trying to cover your ass and trying to
imply that you know something that others do not. If your soi-dissant
"friend" and "relative" are conveying inside Intel information, then they
are most assuredly breaking insider trading laws.
[/quote]I don't even know why I'm even bothering to respond to this, but they haven't given me specific information like exact clock speeds for future processors and release dates; they've just given me a general overview of how things are going. Besides, if I really did want to bullshit things, I would have said that I am a current or former employee there, and I'm not.

*if I'm banned for that, so be it. Maybe Silverman is the one rotten apple that spoils the rest, but this site doesn't seem to be too friendly to newbies and people who're just trying to express their opinions.

edit: Just for clarification, when I said that "I do have a bit of inside info", I don't mean that I know their trade secrets or know what products are going to be released and when. I meant that the information I know is public but not well known among people who are not insiders.

In inverse order. Colour formatting mine.

[QUOTE=R.D. Silverman;209953]Hey moron. Your bold assertion made [COLOR="Red"]NO[/COLOR] allowance for exceptions.[/QUOTE]
[QUOTE=Historian;209933]I was speaking in the [COLOR="#ff0000"]general[/COLOR] case.[/QUOTE]
[QUOTE=R.D. Silverman;209915]What compels people to make dogmatic statements about a technical subject in which they are relatively ignorant??

Whether more cores is better depends upon many factors,
and the answer iS [b]NOT[/b] [COLOR="#ff0000"]always[/COLOR] yes.[/QUOTE]
[QUOTE=Historian;209827]Are 2 cores better than one? Yes.
Are 4 cores better than 2? Yes.
Are 6 cores better than 4? [COLOR="#ff0000"]Probably[/COLOR] not, for most users.
Are 8 cores better than 6? No, unless you're running a server.

For [COLOR="#ff0000"]most[/COLOR] people, the law of diminishing returns starts to kick in above 4 cores. [/QUOTE]
[QUOTE=Historian;209755][COLOR="#ff0000"]For 99% of people[/COLOR], there's little to no difference between 4 and 6 cores because most programs aren't multi-threaded.[/QUOTE]
As you will note Historian said 99% of people initially and has continually used "most". He was not specifically referring to forum users.

Sorry Bob, you lose.

Historian,

Please stop trading insults with Silverman for a while. Please read my latest post in the "Feindlich, Widerlich, Abstoßend, Undiplomatisch?" thread in Soap Box.

[quote=Historian;209963]you are now the first person to be on my ignore list.[/quote]That's a really good idea.

After I put the most irritating people on my ignore list, my forum experience became more enjoyable. There are hundreds of worthwhile people here; don't overlook them on account of your irritation with a few.

[quote]I don't even know why I'm even bothering to respond to this,[/quote]It's because you're letting your irritation with Silverman override your sense of whether you're contributing anything useful.

[quote]Maybe Silverman is the one rotten apple that spoils the rest,[/quote]You have control over your own reactions to Silverman (indeed, over whether or not you even see his posts). You don't have control over Silverman himself.

[quote]but this site doesn't seem to be too friendly to newbies and people who're just trying to express their opinions.[/quote]You're drawing a wrong conclusion because of your irritation with Silverman. Don't discard the benefits of this forum on account of just another member or two -- that you can add to your ignore list, just as I've put a few folks on mine.

Ok guys, settle down. Chill out, or cut some slack, or kiss and make up, or whatever the current vernacular might be.

Misinterpretation of the others' statements is no excuse for having blazing rows in public. Do waht you want as long as you don't frighten the horses so take it to PM if you really want to fight.


Paul

Only four programs !?!
 

Has no one actually watches people work on computers? I spent a few years doing PC support and this is just dead wrong!!

Factoring while generally dismissed above as special case is not really a special case. To the contrary, while this is the special case for users of this forum that special case can be transformed into a magnitude of other special cases.

Developers using MS products will often have several projects open in Visual Studio 2003, 2005, 2008, Front Page, Web Expressions, or what have you. Each instance on its own can eat up all the processing power on a PC, especially when dealing with large projects.

Graphic designer face the same with programs like Photo Shop, Corel, Flash, etc. You can find many such jobs with similar situations. If you work in a office, take a walks around and just check out how many application people have open in their task bar.

And of course lets not forget about all of the gamers, enterprises, scientists, students, etc...

While a portion of users may only run three to four apps, I would say this is not the general case and varies by user.

For example, I currently have 3 instances of Outlook open (which hangs my pc every 10 minutes checking for new mail), 8 Windows explorer windows, 3 command prompts, 47 instances of Notepad, 2 Instances of windows calculator, Windows Task Manager, 5 instances of Firefox (each with 5-20+ tabs open - which is starting to become a resource hog), two instances of IE, 7 visual studio 2008 project, 4 Visual studio 2005 projects, 3 Visual Studio 2003 projects, An FTP Client, SVN Client, SQL 2008 and 2005 and 2000 Enterprise Manger, 4 Instances of Microsoft Word, 3 Instances of adobe reader, NotePad++.

Then in the background, the AV client, Anti-Spam, Windows Update Bits, and Firefox and Google Toolbar constantly querying the web for something.

Then on the factoring side I running ECM curves on two cores for 512+ bit number and poly selection on a C162.

Oh did I forget to mention that I have 5 terminal sessions open to different servers at remote locations and there 8 different computers running in my house (My laptop, my desktop, my wife's desktop, my brother in laws laptop, my son's, the family computer, my web server and a back up laptop) which we all constantly use and all of which could use more processing power.

All of these computers/server get bogged out rather regularly.

Bottom line is every needs more processing power, not just users of this forum.

<feed troll>
wow you are right, you are definitely the average user!
</feed troll>

[QUOTE=Historian;209827]Are 2 cores better than one? Yes.
Are 4 cores better than 2? Yes.
Are 6 cores better than 4? Probably not, for most users.
Are 8 cores better than 6? No, unless you're running a server.

For most people, the law of diminishing returns starts to kick in above 4 cores. Think about it, if it were not for DC projects, how often do you actually need to use more than 4 cores? It's pretty rare for people to have more than 4 programs running at once.[/QUOTE]

Frankly, I think that the benefit to typical users drops off rapidly after the second core. 1 slow core (netbook) => 1 fast core (P4) is an improvement, 1 fast core => 2 fast cores is also an improvement -- 1 core for whatever you're doing, plus one core to run all the background stuff. But a third (Phenom II) or fourth... not so much.

Of course there are power users who want more, and I expect ordinary users will eventually see benefit from manycore computers as the power becomes available. But we're just not there yet.

The question is not how many windows you have open at once but how many programs you have *running* at once. Most people don't actually use more than one program at once.

Background virus checkers etc should run at low priority. And a system that ran the current foreground window at higher priority than other windows would deal with Outlook checking for mail etc.

The big problem is that lazy programmers ignore performance and write code that runs amazingly slowly. But it's often disk or network I/O bound so a faster CPU won't help.

Chris K

[quote=chris2be8;210577]

The big problem is that lazy programmers ignore performance and write code that runs amazingly slowly. But it's often disk or network I/O bound so a faster CPU won't help.

Chris K[/quote]

Yeah, most people but not the special kind of breed we're representing *hehe* ...

Three years ago I had a very fast Celeron CPU and it was always overworked (and I wasn't using Prime95 in that time). With the upgrade to a quad Q6600 I was very happy (costs were around 280 Euro).

I can't get enough cpu power and memory ... unlike most of the people.

Forgive me but I think a more appropriate question would be a 5GHz in the next four years. :-)

[quote=RichD;210595]Forgive me but I think a more appropriate question would be a 5GHz in the next four years. :-)[/quote]

No way the cpus will get more parallelized but the speed won't be increased by much ...

5 GHz has been done, just not by the x86 people:

[url]http://docs.google.com/viewer?url=http%3A%2F%2Fwww-03.ibm.com%2Fsystems%2Fresources%2Fsystems_power_news_20090428_annc.pdf[/url]

[quote=Historian;209755]
2011 will be the year that diminishing returns from increasing cores begins to play a major role. For 99% of people, there's little to no difference between 4 and 6 cores because most programs aren't multi-threaded. Because of that, it makes more sense to increase clock speed than to increase the number of cores, but the main question is when Intel and AMD will finally realize that.
[/quote]
If this recently leaked roadmap is correct:

[URL]http://www.reviewguy.net/technology/core-i7-core-i5-processors.html[/URL]

then Intel will only be releasing dual and quad cores late this year or early next year. Like other posters said, that doesn't mean clock speed will be increased, it may mean that performance per clock cycle is maximized instead.

I'm withholding my answer to the poll until March 2012.

Personally, I'm a huge fan of my single-core 2 MHz RCA 1802 CPU.

For the record, the correct answer to the poll is "yes":

[URL]http://www.anandtech.com/show/4955/the-bulldozer-review-amd-fx8150-tested[/URL]

Processor: AMD FX 8150
Max Turbo: 4.2 GHz

[QUOTE=Oddball;274210]For the record, the correct answer to the poll is "yes":

[URL]http://www.anandtech.com/show/4955/the-bulldozer-review-amd-fx8150-tested[/URL]

Processor: AMD FX 8150
Max Turbo: 4.2 GHz[/QUOTE]

"The FX-8150, 8120, 6100 and 4100 are what's launching today." All have sub 4GHz base clock. Turbo doesn't count (well that's my opinion on the spirit of the original question). The only base clock above 4GHz is 4170 and that's not released yet.

Given the original question was asked in 2007, I'm amazed 4GHz hasn't been achieved in production base clock.

-- Craig

Agreed. There's a slight chance Ivy Bridge will do it, but I highly doubt it. Assuming non-Turbo speeds, where we are currently, AMD seems more likely, though admittedly neither is particularly likely to do it anytime soon. Whoever voted no should be feeling better and better about it.

I hate to see a good thread die ;)
 

I'm bumping this, well, because I can...

We appear to be approaching an atomic(?) wall, the supposed limit of atoms. I say supposed because atoms are made of their own tiny little things, you just have to give up on Newtonian physics to deal with them. Mind-blowing stuff to say the least.

I've seen suggestions of things that can be done to keep going further. 3d chips, materials other than silicon, new ways of cooling chips so they don't die early...probably countless other things.

It's been said that if Moore's Law continued at the pace it was going, and it's actually sped up lately, then the known universe would be 100% saturated with information within 500 years. So basically we'd be using every material available to store or process information.

Ray Kruzweil says that very soon(in the next 20-30 years) computers will become so advanced that they will surpass human beings in intelligence and create a runaway technological phenomenon that will appear to change the world overnight. Makes me think of the T1000, the morphing robot in T2, except without the programmed in urge to kill people.

So I want to know, do you guys think the world will go through mind-bending changes soon or will it be business as usual with a possible major slowdown coming up for the electronics industry? Utopia, dystopia or same basic world for the forseeable future?

As for the atomic barrier, what you're referring to is that at around <10nm, electron tunneling becomes a serious problem, so silicon transistors are nearing their end of life (still a few years left yet though).

As for the robots: Futurama [URL="http://en.wikipedia.org/wiki/A_Clockwork_Origin"]S6E09[/URL]. That is all.

[QUOTE=jasong;290404]So I want to know, do you guys think the world will go through mind-bending changes soon or will it be business as usual with a possible major slowdown coming up for the electronics industry? Utopia, dystopia or same basic world for the forseeable future?[/QUOTE]

Business as usual.

[QUOTE]Will Intel or AMD sell a cpu faster than 4GHz in the next five years?[/QUOTE]
Aren't they already? :shock:

[QUOTE=jasong;290404] very soon(in the next 20-30 years) computers will become so advanced that they will surpass human beings in intelligence [/QUOTE]
Aren't they already? :shock:

[QUOTE=LaurV;290415]Aren't they already? :shock:[/QUOTE]
Good point. A lot of people think the problem is simply lack of decent software for the computers to use to interact with the world.

[QUOTE=LaurV;290415][QUOTE=jasong;290404]very soon(in the next 20-30 years) computers will become so advanced that they will surpass human beings in intelligence [/QUOTE]Aren't they already? :shock:[/QUOTE]Not really. They only think they are.

[QUOTE=xilman;290535]Not really. They only think they are.[/QUOTE]

Agree. Computers still don't have common sense.

[QUOTE=em99010pepe;290536]Computers still don't have common sense.[/QUOTE]Nor do some people. So that criteria for judging sentience is out.

Quote:
Originally Posted by [B]em99010pepe[/B]
[I]Computers still don't have common sense.[/I]

[QUOTE=retina;290541]Nor do some people. So that criteria for judging sentience is out.[/QUOTE]

:razz: Heh heh. If "sense" is "common", why is there so little of it?

Just ran across this:
[URL]http://www.xbitlabs.com/news/cpu/display/20120222205455_AMD_Piledriver_Chips_to_Conquer_4GHz_Milestone_Thanks_to_Resonant_Clock_Mesh_Tech.html[/URL]

[QUOTE=bsquared;290560]Just ran across this:
[URL]http://www.xbitlabs.com/news/cpu/display/20120222205455_AMD_Piledriver_Chips_to_Conquer_4GHz_Milestone_Thanks_to_Resonant_Clock_Mesh_Tech.html[/URL][/QUOTE]

Very interesting, not least because of the arguments which follow the article.
Thanks!

[QUOTE=kladner;290559]Quote:
Originally Posted by [B]em99010pepe[/B]
[I]Computers still don't have common sense.[/I]



:razz: Heh heh. If "sense" is "common", why is there so little of it?[/QUOTE]

Indeed. Have you ever watched the TV show "1000 ways to die"???

One of the most idiotic: A man camping out wanted to get drunk. He had heard that
there was alcohol present in gasoline. So he siphoned some gas from his car and drank it.
Of course, he threw up almost immediately. He might have been OK except for the fact
that he was standing over an open campfire at the time....... Hey everyone! It's crispy
critter time!

Or another guy on the 80th (approx) floor of a scyscraper who wanted to prove
how strong the windows were. He crashed against one.... Oops!

[QUOTE=R.D. Silverman;290573]Indeed. Have you ever watched the TV show "1000 ways to die"???

One of the most idiotic: A man camping out wanted to get drunk. He had heard that
there was alcohol present in gasoline. So he siphoned some gas from his car and drank it.
Of course, he threw up almost immediately. He might have been OK except for the fact
that he was standing over an open campfire at the time....... Hey everyone! It's crispy
critter time!

Or another guy on the 80th (approx) floor of a scyscraper who wanted to prove
how strong the windows were. He crashed against one.... Oops![/QUOTE]

I'm honestly surprised that you would watch that show. When I turn it on, my wife has to leave the room. She just doesn't appreciate Darwin. :smile:

This reminds me of the [url=http://www.darwinawards.com/]Darwin Awards[/url] :smile:

[QUOTE=bsquared;290560]Just ran across this:
[URL]http://www.xbitlabs.com/news/cpu/display/20120222205455_AMD_Piledriver_Chips_to_Conquer_4GHz_Milestone_Thanks_to_Resonant_Clock_Mesh_Tech.html[/URL][/QUOTE]

Is it me or do they do a terrible job of explaining what clock meshing is anyway?

[QUOTE=kladner;290559]Quote:
Originally Posted by [B]em99010pepe[/B]
[I]Computers still don't have common sense.[/I]



:razz: Heh heh. If "sense" is "common", why is there so little of it?[/QUOTE]"Common sense" is one of the standard oxymorons.

Others include:[LIST][*]Military intelligence[*]Civil servant[*]Industrial action[*]Microsoft Works[/LIST]
The third of these is well-understood in the UK and a number of other places in the civilised world but I've had to explain it to Americans on a few occasions in the past.

[QUOTE=R.D. Silverman;290573]Indeed. Have you ever watched the TV show "1000 ways to die"???[/QUOTE]

I confess that I was not even aware of the program, but it sounds amusing in the Darwin Awards context.....at least in reasonable doses.

[QUOTE=kladner;290610]I confess that I was not even aware of the program, but it sounds amusing in the Darwin Awards context.....at least in reasonable doses.[/QUOTE]

I was, but I haven't watched it in a while. me and my mom watch untold stories of the ER. some rare conditions come up from time to time.

[QUOTE=rogue;290575]I'm honestly surprised that you would watch that show. When I turn it on, my wife has to leave the room. She just doesn't appreciate Darwin. :smile:[/QUOTE]

I watch it from time to time. I find it amusing. It reaffirms my faith
in human stupidity.

[QUOTE=R.D. Silverman;290573]Indeed. Have you ever watched the TV show "1000 ways to die"???

One of the most idiotic: A man camping out wanted to get drunk. He had heard that
there was alcohol present in gasoline. So he siphoned some gas from his car and drank it.
Of course, he threw up almost immediately. He might have been OK except for the fact
that he was standing over an open campfire at the time....... Hey everyone! It's crispy
critter time!

Or another guy on the 80th (approx) floor of a scyscraper who wanted to prove
how strong the windows were. He crashed against one.... Oops![/QUOTE]
The best one I liked was with the guy whose truck started making some noises and he wanted to see where the problem comes from. He asked his friend to drive the truck about 30 meters and he crawled under the truck and hanged on to it (it was a tall one, there was enough space under, and no danger to to hit the ground if/when the truck move) to check for the noises during the truck is moving. After one minute and about 30 meters of slow driving, his friend found him rolled around the (shaft?) under the truck...

BUMP after multiple months

If you read tech news, you've probably noticed that processing on graphics cards or duplicating the way GPUs process things in cpus is becoming WAY more popular. When you think of the way the human brain works it makes perfect sense. Unless I'm mistaken, human brain cells operate at kilohertz frequencies, it's the fact that there are so many of them, among other things, that creates human intelligence.

So, while the poll is still legit, in a sense, I think the future is WIDER, not FASTER.

Maybe at some point we'll have cpus with significantly fewer cores that do far more work for things like games and video because the registers are thousands of bits wide. So we could have a dual-core portion made the way things are made now, and one or more ultra-wide pipe cores that do stuff that can be parallelized.