Slashdot Mirror
Intel Releases "Fastest Chip Ever"
mao che minh writes "From News Factor Network: Intel has released the world's fastest chip ever. The new P4 runs at 3.06GHz, at 3 billion cycles per second. Man, and I'm still squeezing the last bit of life out of my Pentium 233!" Tom's Hardware already has a review up about it, and it looks to live up to most of the hype.
if you read the article they say that it can use up to 100 watts of power when you are using the chip to the utmost. That's a lot of power. Much more then the AMDs. Anyone else think that's a little extreme? I mean I'm all for more speed, but cost aside, this seems to be a huge factor in actually getting one of these systems. You also have to get a new motherboard.
/. do so I think that running a system all the time (with SETI or whatnot) would be expensive.
For server applications it's not as useful because you can't build dense systems. Since server applications are by their very nature more multithreaded then workstation, I would imageine that they would get much hotter. You'd need a lot more cooling. Also, don't the chips SLOW DOWN automatically when they get too hot, thus negating any increase in speed you might get from them.
Notice that the new heat sink is larger as well.
Not trying to bash it, but it seems like the older chips are still going to be better until they get this whole heat issue under control. I run my system almost 24x7 like I'm sure many people on
Does this new chip have support for Digital Restrictions Mechanisms? Does it still have the universally reviled serial number feature? Can it still be shut off?
Schwab
Editor, A1-AAA AmeriCaptions
I remember when processors started breaking the GhZ benchmark, people were making jokes about how we're starting to get to the point where the things will be emiting microwaves since they are in the GhZ's.
anyone know how close we are now? will this new chip boil water from a distance?
even if we're a couple years off from that, are we going to need sheilding in our cases soon so that we don't cook our lower legs? if so, does anyone else thing that this would cause a lot of problems since compUSA won't take that into account when they do an upgrade?
Just some thoughts...
Blaze a trail to the New World
Does anyone know if Linux of FreeBSD sees any benefit from the 'hyperthreading' technology? All the things I am reading say that you OS needs to support threads, but how does the processor know what is a thread, and what is a process?
I'm sorry, but there are only two explanations. One is that half of users out there are running maxxxed out machines that can handle that load (yes, with winblows). In which case, why the push for new chips?
The other explanation is that users really are burning cd's while playing games, in which case, the RIAA can pack up and go home, because those hundreds of thousands of CD's are obviously ending up as coasters, not as pirate booty.
I know, I know... I show my age when I remember the days where you clicked "burn" and ran like hell. I still remember the setup I had that would coaster the disk if I moved the mouse during the TOC writing. Admittedly, it was a brand new 1x burner, but still....
And considering my ole Celeron 300a runs Win2k just fine, why in the blue blazes would I need a 3G? Seems computers have hit the plateau... the average user gets along just fine with what they have, it's only professionals and gamers who really snap up the new hardware.
I'm gonna start a bet... how long can my 300 run before it's finally too slow?
(and to stop your flames, RedHat goes on my 1Ghz. So there)
If I knew the wedgies I gave you back in 6th grade would have resulted in this . . . I might have taken a moments pause.
It is virtually useless unless they make it very cheap.
Does anyone know if Linux or FreeBSD sees any benefit from the 'hyperthreading' technology? All the things I am reading say that your OS needs to support threads, but how does the processor know what is a thread and what is a process?
Exactly. The thing is that the testing that is done on Intels latest chips is done in a subjective manner for marketing purposes. They test the chips on machines that are not Identicle - with software that is optimized for that chips instruction set - then push it out to people saying "Look at Intel's latest chip! The new $Pentium-X will runs your applications so much faster!"
but there is a threshold we will hit - on the consumer level - and that day (although still a bit away) is coming faster and faster with every release. It is the subjective speed threshold, where the Human is the bottleneck. Where the computer can do anything the user can so fast - that the computer is then waiting on input from the user.
All input from a human comes in little spurts - and therefore will be processed by the CPU before the next batch comes in.
The point is that there is a somewhat finite desktop market incentive for faster processors, in that, for the average user - there will be a time, sooner than later, where they find that the machine they have is fast enough, featureful enough and big enough (storgage) to meet their (rather long term) needs.
Compared to the average person I do intensive computation, and I feel no pressure to upgrade. For the average user the need to upgrade must be entirely generated by marketing--right now performance improvements in hardware is irrelevant. I wonder what's going to change--assuming anything does--to make us all hunger for faster systems as we used to. I can't think of anything compelling, but i'm unsure because intel etc are spending piles of cash figuring out how to reestablish the need for improvement.
Somebody on the overclockers' forum got his 1.8 Ghz P4 C1 stepping (from a Dell computer) to 3.5 Ghz. While it doesn't have hyperthreading (which doesn't neccessarily give you performance benefits), it does have a much faster bus rate.
I'd say that 2 fast Athlon MP's (or XP's that take about 3 seconds to mod for SMP) are still going to be significantly faster for anything that can take advantage of multithreading. I don't know what tasks you have in mind, but with the price/performance ratio of dual Athlons, I can't imagine that Intel uniprocessor systems will be able to compete for some time. Of course, benchmarks wil really tell, I'd suggest 2cpu (http://www.2cpu.com) for more info, the forums are also filled with people who possess a great deal more knowledge than I do about system performance, esp. concerning multiple CPUs.
Mod me down, and I will become more powerful than you can possibly imagine!
This news doesn't matter. In a few months Intel will have brought out another, even faster chip that can do more, do it faster and more efficient while (hopefully being cooler then it's precursor.)
Hate me!
if i had mod points, i'd vote this funny. i was just about to boot up my old 4.77 MHz IBM PC from 1981 this weekend just to see if it still works. lets see, it'll probably be about 22 years from 1981 when they get 4.77 GHz...that's (22years/18month) about 15 moore-time-units...2^15 is much larger than 1000 (about 32000). that's interesting.
I believe XP and the newest linux kernel are aware of this and have been somewhat optimized. this was a big deal when the Xeon's first started coming out. If your running and older OS, like NT 4 or an older kernal you could see a decrease. But the processor is executing both at the same time, each getting a piece of the massive pipeline.
You should read the articles on anandtech or arstechnica, THG has gotten kind of dumbed down as it's popularity has increased. Sometimes I read it first, then go on to other sites for more details. I've been writing multi-threaded apps for sometime, so this was interesting to me. It took me a while to fully understand it as well.
This chip is more interesting than just the normal megahertz hike. It's the first of the desktop hyperhreaded chips - previously only available in the Xeon range (well, from Intel anyway. Other manufacturers had them).
:-)
It doesn't help a lot, at the very most a 20% speedup, typically much closer to 0%. This iteration just isn't that effective, maybe next time round with better management of cache or something, they might get it working more like separate chips. Right now it just doesn't.
It's good to see this entering mainstream though. It provides an incentive to write the kinds of multi-threaded applications that can actually squeeze out the full 20% speedup, and in turn, those applications will automatically be able to take advantage of real multi-processor boards without further changes. So this process ends with everybody having SMP laptops
Have you got your LWN subscription yet?
Are you kidding? Do you know what kind of yield they get off those wafers? Well over 95%
"With the introduction of its Hyper-Threading technology, Intel has confirmed that constantly increasing the clock cycle is not the only way to skin the proverbial cat."
Yah, AMD has been saying that for years with their performance-ratings, and Intel's been saying that cycles-per-second was the measurement that the consumer truely understood, and was a good way to get a measure of the speed of the processor.
Wonder if Intel will adopt that, now that they have a CPU that, at lower speeds, can process more data.
20 January 2017: the End of an Error.
..sigh... Every time one of these articles come out. First, if you can't tell the difference between 2GHz and 600MHz, you're dead. My 2GHz machine is nowhere near fast enough, even just running Konqueror and KMail. Second, more people need the power than you'd think. I write C++ code with some very heavily templated libraries. G++ eats my processor for lunch (I've got enough RAM that it's not HD-bound). Add to that 3D rendering (messing around with Blender for some 3D work) and numerical computation (simulations, Octave, Mathematica) and I probably won't ever have enough CPU. And I don't even do gaming!
A deep unwavering belief is a sure sign you're missing something...
Accuse me of FUD all you want, but examine the evidence for yourself.
.NET .. Looks like PC Mag has some integrity.
:)
Exhibit A
Win NT beats Windows 2000 in SQL Server 7 Benchmarks
What? The new O/S is slower? Must be FUD, doesn't have anything to do with bloated code and forcing users into hardware upgrades.
Exhibit B
Red Hat/Samba far outscales Windows 2000 on identical hardware
Yes your honor, it's true, at a load level of 16 clients Windows 2000 filesystem throughput flat lines vs. Red Hat Linux with Samba which is still scaling up nicely with 28 clients.
Does Windows 2000 mask the true power of the Intel hardware? Examine the report and look at the benchmark graphs. Decide for yourself if it's FUD or FACT. Note: the source is PC Magazine which if you will refer to this months copy contains many advertisements for Microsoft
Shall I continue?
Want to see why TUX stomps IIS and Apache for serving static content?
I challenge you to find the FUD in any of this. In fact, many of you might wish to save these links for future TCO discussions within your local IT departments.
PROVE ME WRONG!!!! Show me how Microsoft is doing it faster and better compared to either a) A Previous Microsoft Server Product, or b) Linux. Wave your hands and shout FUD all you want, but be prepared to back it up.
I wish someone would back me up!
As for my 486, I wrote a user mode driver which allows me to access the data pins on the parallel port to activate a relay and ultimately switch A/C power. (Web page coming soon.) This device can be used to remotely reboot Windows servers that BSOD, or turn on Christmas Lights add/or Coffee Pots via cron or telnet. Did I mention it all fits on a floppy, runs on a 486, and is network accessible? I am trying to shoe-horn a webserver onto the floppy now.
actually a lot of chips that are marked for sale under slower speeds can be overclocked to over 3.06 ghz using *AIR* (yes, thats right, air) cooling... needless to say rather "impressive". the new C0 stepping 1.8 ghz northwoods are very impressive overclockers, somebody reported on the overclockers.com forum a 1.7 ghz overclock on one.. thats 95% damn impressive, and on air cooling too. PIV's are great chips and very scalable, Intel probably has working chips all the way up to 4 gHz+ right now its just not as profitable for them to produce them (it would drive down the price of there current chips and create a market saturation effect... plus they obviously cost more to make due to yield rates)...
Even if we narrow the scope to x86 desktop apps, it seems that based on preliminary benchmarks (with Hyperthreading enabled) AMD's AthlonXP 2800+ still reigns (albeit, by a very small margin) as the fastest chip available.
There is no longer anything that can be done with computers that is nontrivial and clearly legal. -- Paul Phillips
Signatures are for stupids.