Slashdot Mirror
4 Cores? 6 Cores? Do You Care?
An anonymous reader writes "Intel has updated its processor price list earlier today. Common sense suggests that Intel may not care that much anymore whether its customers know what they are actually buying. One new six-core processor slides in between six-core and quad-core processors – and its sequence number offers no clues about cores, clock speed, and manufacturing process. If we remember the gigahertz race just a decade ago, it is truly stunning to see how the CPU landscape has changed. Today, processors carry sequence numbers that are largely meaningless."
Would you want to have a 4 inch penis? Don't you think a healthy 6 or 8 inches might be better?
I have a quad core, which I'm confident will soon become the equivalent of a 4 inch penis. I'll have to upgrade my e-peen when it become affordable.
Seriously though, if you like to game on your computer there is no such thing as too much power.
Not long ago I worked with a ph.d. professor who would have insisted that I explain to him why that cpu had the sequence number it had, and would not settle for anything less than something that makes sense - one reason I'm not working in IT anymore
The average consumer just thinks "bigger is better" and by creating a mess of hard to understand sequence numbers they can make it harder for the semi-knowledgable customer to pick the right CPU. The same can be seen with graphics cards and many other products (if there is some kind of system behind your sequence numbers you do have to remember to change the system every now and then to further confuse everyone).
Greylisting is to SMTP as NAT is to IPv4
Correct! Six thousand hulls.
maybe they should just use flops in the sequence number, with power draw if they were feeling actually informative.
I do care when Intel ships more cores. The price of 'old" cores drop and I get better value for my $$$.
The headline asked a question, I answered it.
Anything can be found funny, from a certain point of view.
Some combination that measures both how many operations per second, and how much power it's going to take to do said operations (i.e. Watts/computing unit). I don't know if even FLOPS is sufficient anymore to describe current computing tasks. Heck, I'd be happy with any sort of standardization.
The sending of this message pretty much inconveniences everyone involved.
One core is sufficient for 99% of office workstations that only run a browser and MS Office applications.
Absolutely not. There are so many crappy applications that will max out a single processor doing stupid things (like rendering javascript on a webpage), that a 2nd core is very very useful.
Since most software still isn't multithreaded, a crappy application will only max out one core, allowing you to still get work done.
Multicore plus enough RAM is generally a lot better performance-wise than singlecore plus low amount of RAM and an SSD.
If you're having performance issues in everyday office use that go away when switching from a regular hard drive to an SSD you could just try accepting that these days you need 1+ gigs of RAM instead of trying to implement a bunch of workarounds that don't address the actual problem (that your computer keeps swapping out stuff to the hard drive because you're running out of RAM).
Unfortunately it's pretty common to see regular office desktops with fast multicore CPUs and ridiculously low amounts of RAM (I've seen C2D 2+ GHz CPUs coupled with 512 megs of RAM, it ran slower than a low-end P4 with 2 gigs of RAM).
Greylisting is to SMTP as NAT is to IPv4
I just don't give a crap about intel products...
They proved long ago they do not win on price/preformance. And hell. someone has to pay for retarded tv commercials. I'll pass on being one of them again.
And they are still the only company that ever sold me a defective chip that couldn't do math. And their response was? 'Oh well, buy our new one'.
Eventually they DID replace it. But the entire experience has put me off intel products forever. I wont spec or support intel chip based hardware.
The the question is what for?
If you are a typical user you will only need one or two to run an OS with Web Browser and Word Processor.
More cores if you are running a server.
Most cores if you are doing an Virtualization and in-particular running any Virtualized desktop or server.
So I want more cores and memory. My family only needs one or two until the eye candy catches up and has an improvement from the current system.
Our point-of-sale software runs acceptably on a single core 1.6Ghz Intel Atom machine with 1Gb of Ram in embedded mode (client software + database, PostgreSQL, on same machine), it runs better with 2GB of ram, and then there is no noticeable difference between a Dual Core Atom with 2GB of Ram vs. a Core 2 Duo with 2Gb of RAM. And if you are running the software in client/server mode with a seperate database server for multiple terminals, then the 1.6Ghz Atom and 1Gb of ram is plenty.
"The problem with socialism is eventually you run out of other people's money" - Thatcher.
F@(# Everything, We're Doing Five Blades!
Such as processing times for apps, possibly flops, but for the average user that won't mean much either.
ESX server licensing is on a per CPU basis but they restrict the number of cores to 6 (from memory) before you need another license. So yes, I would care how may cores I was buying on a server.
... overdue for its 2 year processor and motherboard upgrade. It is overdue because when I started to look at what processor met my ideal performance/cost ratio it was impossible to figure out.
I don't have time spare to sit with a spreadsheet and a matrix of 30 different processors to work it out so I won't be upgrading now until something breaks. You lost a sale Intel, and I will have to pay more attention to system requirements of games for a while.
I would guess I am not the only one choosing not to buy because its so unclear...
[The Universe] has gone offline.
I remember the clockspeed race and it was much simpler to decide what CPU you needed when looking at system requirements. Just a week ago I was looking at a game's requirements and had no idea if my CPU met them. If I were to upgrade, I wouldn't know which CPU would satisfy the requirements. I'm pretty handy with computers and I find picking a processor with today's marketing daunting, I can imagine being totally in the dark if I knew little about computers. Intel could do a better job indicating which CPU is better than the other and letting you know what you're buying.
Chewbacon
The Bible is like Wikipedia: written by a bunch of people and verifiable by questionable sources.
isn't just that the sequence numbers are out of order...
But that the differences in processor performance are largely irrelevant anymore. Who cares if it's 4/6/8 cores/hyperthreading/gigawhatzitz. The bottom line is that all of them are ridiculously fast. You would do far better putting your money into just about any other component.
The number of cores and the speed per core becomes vitally important when you start doing virtualization. Since Windows 7 has this out of the box and Macs use it all over the place and everybody and their cousin are running VMware (or insert your favorite VM environment here), yes, I think alot of people care. That's not even starting to talk about the server space where almost everything is virtualized these days and more cores can mean more VMs (especially on Hyper-V).
I don't want to leave the enthusiasts out, so I will just say for their benefit that seeing all those core graphs lined up in task manager is a major rush and should not be discounted as users look to buy processors (though I guess Intel has that covered with "hyper-threading":P
... by the ad-sponsored review magazines; not Intel.
I have a Q12345EXTREME!1!! Pentium, what do you have? See mine is bigger.
*so says Joe Consumer
**Intel has won.
I want to delete my account but Slashdot doesn't allow it.
Nvidia and ATI have been giving their graphics cards arbitrary numbers for years.
Is a 330m better than a 220m? maybe.
What about a 9600 vs an 8800? who knows.
Intel didn't invent the random product model numbering scheme, they are just joining the ranks.
If the chip can't run all the cores at full speed due to heat/power considerations and therefore either throttles back each core's speed or disables some cores under heavy load, than core counts are really just a deceptive pissing contest, aren't they?
I've abandoned my search for truth; now I'm just looking for some useful delusions.
Misleading Indicator of Performance Statistic was the worthless number we had back then, and we liked it!
Blingcores?
But seriously, folks. I know we're all a bunch of geeks here, myself included. But the truth is that it's not the CPU market that's changing. It's the nature of computing itself that is changing. Devices that can be called computing platforms are varied in size, function, and resources. An iPhone is essentially a mobile computing platform, but people wouldn't call it a "computer" in the conventional sense of the word because "computing" is an activity that has moved so far into the background, behind what the end user wants to actually accomplish with said device.
Likewise, CPU speed and number of cores don't give anyone, let alone the average consumer, a universal standard of performance... by this I mean even the geek cannot assume on CPU speed or cores in and of themselves how a device will fare to other devices with different architectures, operating systems, background applications and so on.
Watts consumed.
For justice, we must go to Don Corleone
I don't care how many cores it has.... can it stream HD video without chop at only 4 watts? (Without a video card that requires its own nuclear power plant to run)
Self Defense - A Human Right www.a-human-right.com
Have you considered that the reason the processor numbers tell you nothing is that ALL the chips are fabbed with 6 cores and the ones that have one or two bad cores in testing have 2 cores disabled and are sold as quads?
I've abandoned my search for truth; now I'm just looking for some useful delusions.
Do I care if my truck has a Hemi or some other engine? No. I care if it offers the right balance of strength, carrying capacity, looks, and gas mileage. Do I care if the airplane I'm flying in uses GE or Pratt & Whitney or Rolls Royce engines? No. I care if the plane will get me where I need to go comfortably, safely, and quickly. In the early days of computing, it was a thrilling thing to have hardware that could keep pace with software. I still have painful memories of Photoshop 3 screen redraws. These days it is a given that while there may be differences in response time, for anything but serious gaming, the hardware is going to keep up with the software. Keeping track of what processor is inside the computer is, for most people, akin to keeping track of which subcontractor supplies the tires for a car. Sure, some people care a great deal about the tires, but the rest of us could care less.
Read the EFF's Fair Use FAQ
One core is sufficient for 99% of office workstations that only run a browser and MS Office applications.
..assuming the OS shares time and assigns priorities in a sane way, the users is not an quant, the multimedia apps are hardware accelerated and the anti-virus package doesn't try to take over the machine, of course.
I would like my CPU cores to be assignable. If I want 1 of my 4 cores on background stuff
all the time, thats my business.
Id like to be able to have fun with my GPU cores without being a super-duper programmer.
A Beowulf Core?
I use AROS (Amiga X86 OS) as a hobby. It doesn't support SMP. There are various other apps I use that don't support SMP.
So while 8 cores at 3GHZ each is ~24GHZ. I wish the speed wars hadn't stalled as I'd personally have more use for a single core running 20+GHZ.
We do seemed to have sort of stopped at the 3GHZ mark and just gone to adding cores.
When I decide a new computer (usually because the current one is out of warranty) I just buy whatever the newest Mac laptop is that seems to fit my use case. I might look at the specs a bit but frankly I couldn't tell you what processor is in the one I currently have.
I used to care a lot about this. When I was in high school. I have a lot more interesting things to care about and I think 99% of the public does too. I'm not trying to diss anyone here. If being a processor geek is your thing, more power to you. But I think people decide for whatever reason that at some point they need a new computer and just buy whatever fits their price bracket and feature needs.
If I was say, building a huge server farm, or spec'ing out computers for a big group of people I'd obviously do a lot more homework. But those are edge cases in the grand scheme of things.
That's probably more power than my first laptop 7 years avo and it ran many things far better than acceptably.
Multi-core systems are *great* for software developers. "Make -j8" is your friend. It doesn't scale perfectly, but it's pretty good.
Thing is, you need an assload of RAM to do it. On even moderate size projects, a single G++ can grow to a gigabyte of memory, so make -j8 is going to push your RAM needs up a bit.
I guess the DCC folks love them too - rendering is embarrassingly parallel.
Gamers... beyond two, I'm not sure it does much for 99.9% of all games.
And with a quad core system, you can run 3 crappy applications and still have a responsive system! A hex core system will let you run an outrageous 5 crappy applications!
"Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
Nothing to see here, move along.
More cores + cuda = me not having to wait as long.
Luckily we have task schedulers that handle this without needing a second core
I have an i7 quad core with 8 threads that are visibly pegged at ~100% in Task Manager for hours on end. How? Try using the higher quality settings on Carrara or DAZ Studio. They will peg any cpu or box that will be produced in the next ten years.
I'm sure there are plenty of other compute-intensive applications out there that will bring any cpu to its knees and make it beg for mercy.
Anywhere you see a branding that works, you can be sure it will eventually change to something that doesn't if the marketing department isn't kept check. These people need to justify their existence, and that means always pushing change for the sake of change, rhyme or reason be damned. It's quite similar to what happens in the fashion industry.
Intel probably realizes that the average consumer doesn't care about what they're buying. That's why Apple is still in business.
yes I do ! OP is a faggot
I switched to the superior Motorola 68000 processors years ago.
If you don't know what the part number means, don't buy the part.
Intel has products that cover the whole range of core counts, core speeds, core types, cache sizes, power/temperature ratings, and mechanical form factors.
They do care what you buy, or they'd offer you far less to choose from.
Figure out what you want to do. Then ask them. They'll find you the chip you need, and then you'll wander off to buy something less capable for less money from the other guys.
If the packaging or advertising says "x core processor", I expect x cores. Anything else is sneaky business.
The reason Intel is pushing for more cores is because we've sort of reach the "apex" in terms of processor frequency. Increasing the frequency of the processor will not make the processor much faster at this point because the data literally can't travel fast enough through the ports in processors to keep up with the processor. Imagine the processor as a person searching for a book using an online database, which is extremely fast. But after finding the book he wants, the person must travel to the library/bookstore to pick it up. It doesn't matter how fast the person can search the book online (be it 5 seconds, or 1 second, or 1 millisecond), he's not going to get the book much faster by improving his search speed. Similarly, improving the speed of the processor won't make it significantly faster, because we've already developed the processor to be ungodly fast. The only problem is forcing the crapload of data to travel fast enough to/fro the RAM or other parts of the processor. In addition, power usage is a cubic function of frequency. So if you double the frequency of the processor, power usage multiplies by 8. If you triple frequency, power usage multiplies by 27. If you quadruple... power usage multiplies by 64, reaching a temperature that'll probably melt your legs (if your using a laptop). How do you solve that problem? Build a multi-core processor The point of multi-core processor is NOT actually to increase processing power, but to reduce the (physical) distance data has to travel. Each core in a multi-core processor is smaller and less powerful than a single core, but data in each core can travel significantly faster, thereby increasing processing speed (w/o increasing processing power). The combined processing power of the several "unpowerful" cores will actually have faster processing speed than an extremely powerful single-core processor. Less is sometimes more.
Anything past two isn't going to help me for what I use my computer for the vast majority of the time. I have much less time to play computer games these days and everything with Firefox or other common applications are single processes anyway. I'd rather spend my money on beer than extra cores.
Since most software still isn't multithreaded, a crappy application will only max out one core, allowing you to still get work done.
Not to mention letting you kill said shitty application/process without waiting five minutes.
Live today, because you never know what tomorrow brings
The whole point of marketing is to prevent, as much as possible, the customer from realising what the product really is.
Intel gets this. AMD doesn't with it's honestly named Athlon/Phenom X4 X6 etc.
After logging in slashdot still does not take you back to the page you were on. It's been that way for 20 years.
Because 6.40 cores is all anyone will ever need.
I try to buy the fastest single thread processor my budget allows. I go to
http://www.spec.org/cgi-bin/osgresults?conf=cint2006&op=form
and run a query sorting in the "Key" "Primary" by "Baseline" Descending and then choose the fastes one I can buy. Easy and works for me.
I prefer Baseline because plenty of program are poorly compiled, without any optimization flag.
Since we're comparing cores to schmeckles, I think we should make a correction. Core speed would be more equal to schmeckle size, and core count would be more equal to schmeckle count. And in this case, game equals Megan Fox. So lets say your Megan Fox only likes one schmeckle at a time (most games), then having 4 cores at 25~% less speed or lets say 6 cores at 30~% less speed (or 4 or 6 schmeckles that are 25-30% smaller), wouldn't it be better to just have one or two (dual core) really big schmeckle for your Megan Fox?
On the other hand though, with 6 cores you could totally be running Megan Fox, Scarlett Johannson, Angelina Jolie (10 years ago), Jana Defi, Jordan Carver, and Denise Milani all at the same time, provided you have enough RAM in you to pull that off :D
The guy with 2 big cores doesn't even have that option.
I guess in summary, buy the CPU that fits your needs the most. I for one multitask like crazy, and often run 'a few' games at the same time, sometimes the same game multiple times. (Two EvE accounts, Two Aion accounts). And sometimes I do that while watching a movie on the 3rd screen or maybe running some other tasks in the background.
Clock speed, bit length, bus speed, and processor count is all that matters. I guess I've taken for granted that their opcodes aren't erroneous( i.e. Add Instruction takes 50 clock cycles ).
Does anyone ever verify their stats? I guess I take their word for it on those too.
to make a witty comment about quantum computers and speed/# cores not commuting (thus leading to "core-speed uncertainty"), but I think it sounded wittier in my head.
Sadly, we dont.
There are many situations in which modern operating systems will gladly let a single process hog a CPU core (it's often not "pure" CPU loads but the CPU ends up pegged due to other issues and everything else grinds to a halt).
Greylisting is to SMTP as NAT is to IPv4
I let other people do the work. When I buy a new system or processor, I just look at performance charts (e.g., Tom's) and pick the best processor in my budget. It takes all of 5 or 10 minutes.
Actually, I (sometimes) use my quad-core to run a virtual machine on two cores, and the native OS on the other two cores. That means that both OSes can potentially run one crappy application and neither becomes unresponsive.
Any fewer than four cores, and it's iffy, for exactly that reason.
I thought they handled the fdiv issue quite well. I was sent a replacement chip along with prepaid packaging to send back the defective P90. Nevermind that for most users the bug was a nonissue, esp. with OS workarounds.
I'm surprised no one has mentioned the Windows Experience Index (WEI). It may not be as exhaustive as the benchmarks many of us read, but it is very easy to understand. I've yet to see any manufacturer or retailer advertise a WEI score, but it would be a great help to consumers if they all did. Anyone could easily compare offerings from Intel and AMD, or see the significance of discrete graphics or SSDs (without even knowing what they are).
So you're basically asking for them to release Larrabee?
I wonder what sort of craptastic bus Intel would be used to tie 50 Atom cores to one DDR controller?
“Common sense is not so common.” — Voltaire
Add a dedicated core for an overbearing antivirus like McAfee.
well, a decent single core system is still okay for said tasks, processor priority still has some usefulness for such tasks (e.g. process explorer always runs at a high priority so I can kill things which get stuck without waiting)
Dad - mom says to stop bragging on slashdot or she'll start bragging on facebook. She also says you haven't poked her in a long time - and I don't think she was talking about facebook either.
I run VMware ESXi for network simulations and penetration testing. Having a core for each VM makes a big difference.
Cheers,
Dave
They that can give up essential liberty to obtain a little temporary safety deserve neither safety nor liberty.
Ben
AMD vs. Intel. Is AMD still crappy these days? I went back to Intel after three AMD CPUs before Intel Q8200 CPU. I only upgrade for the latest games and still plan to do that for my upgrade at the end of this year.
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
I want 6 cores and if possible I'd like it unlocked for a reasonable price. Their current "extreme" 6core is actually looking attractive to me but I keep waiting for the price to come down. I had hoped that a new 6core would come in that would be reasonably priced and that even if locked could be clocked up pretty good. But at $880+ I dunno' - I will wait for the street price to hit before I get interested.
Why do I want 6cores? Because I compress video pretty often and it's an hours long chore while keeping the quality and resolution high - file sizes plummet though. Hi Def video compression is intensive on the CPU and I often see rates as low as 13fps when compressing. That's on a 920 clocked to 4.2ghz. On water this thing hits 80C with a good sized radiator and multiple fans - I'll be moving to a bigger radiator soon in hopes of solving that. A 6core would give me at least a 30% increase in speed if not more depending on if Hyperthreading continues to buy me anything (it does now). If this new CPU can hit speeds like the unlocked Extreme and hits NewEgg for say $750 I'll score one but not when it's within $100 or so of the unlocked Extreme.
Frankly, if there was decent code to chain multiple machines together to process video I'd try that but the last I saw of code to do that it was old and not worth my time. Since I also happen to be doing this on Windows chances of finding good code to slave machines together is even slimmer.
So yeah - I care and I agree this new number scheme SUX! But hey in the end it's the performance I care about and how high it will clock without melting down. These Extremes are sick fast but wow are they pricey :-(
P.S. Were it not for video processing I'd still consider a C2D just fine or maybe an overclocked i5. This 920 STOMPED my 3.8GHZ C2D though so was well worth the investment and it has also beaten a few dual XEON Macs :-)
Build it, Drive it, Improve it! Hybridz.org
It wasn't until recently when I had issues with Microsoft Virtual PC because my BIOS (which had already been upgraded once) was bugged and would not enable hardware virtualization that I realized that my CPU (an Intel Core 2 Quad Q6600) was one of the very few with hardware virtualization back when I bought it, as the processor models directly above and below this one did not have it, and I bought this CPU assuming that any "nodern" (2007) quad core CPU would have it, I chose this particular model based on price alone.
When it comes to rendering my architecture visualizations, many cores make a big difference. Ever since Autodesk (finally) allowed Revit Architecture make use of more than 4 cores, my 6 core AMD couldn't be happier. It also makes render farms that less expensive for render farm capable software like 3D Studio Max.
I always loved posting this pic for a forum friend who worked at Intel.
http://images.invisibill.net.nyud.net/intelmodelnumbers.jpg
Some PCI slots are still needed to "future-proof" your machine. For example if you get a machine now and next year find that most external HDs come with USB3 you may want some way to add a USB3 port to your machine.
I will have to pay more attention to system requirements of games for a while.
If the matrix of CPU clock speed/costs/cores is too hard to figure out now how long will it be before you cannot easily tell whether your machine meets the system requirements for a game? Those requirements will, eventually, be written in terms which make it hard to understand whether your machine meets the specs or not.
It's really rather simple in my opinion to educate an average person, though being a geek that opinion may be rather wrong, who knows.
Basically, if you're trying to make the computer faster, sometimes all it takes is a good reinstall of (usually) Windows. Usually, this means amongst family that I'm signing my own death warrant to somehow reinstall windows on a case that's been invaded by dust bunnies without spending hours just doing menial things like cleaning up everything or advocating replacements to every component because the power supply looks shot, the cd (yes, CD!) drive seems to be from the late 90s and the IDE cables make the mobo look like the gimp from pulp fiction. Sure, maybe I could get away with just a reinstall but I care too much about this shit, man! Plus, who knows if their house will catch on fire at some indeterminate point in the future and you'll be blamed for the dustbunny problem if you don't do something about it.
The key thing after installing windows is to go through the whole "STOP INSTALLING RANDOM SHIT" rant, hopefully with a nicer tone, and set them up with some free microsoft security essentials (I've felt like it has pretty low resource usage, am I crazy?). Everyone seems to want office, and then explain the wonders of VLC. iTunes is usually something worth explaining and installing, while I expect everyone will chime in saying how much of a bloated piece of crap it is, it is great for an average user. They probably already have it anyway.
After that, the next step would be to replace the hard drive with something faster, either pitching the old HDD or delegating it as a storage volume for things that aren't important. This would likely make the biggest impact on a system aside from a full overhaul. If you dont mind shelling out for an intel or indilinx SSD, then you wont regret it, it will change your computing experience.
Anyways, now to the on topic stuff. The main thing IMHO if someone wants a computer that's not already months (*GASP*) outdated is that dual core is the basic, quad is the middle and hexa now seems to be the upper end. If price is a concern, AMD should be the first stop. If max possible performance is necessary, obviously go with intel. I know I'm preaching to the choir here. Most people with a few year old computers that weren't high end back then should just give up on the idea of keeping the motherboard. I know it varies, but usually you either have the ability to go with an older intel quad or you should just upgrade the mobo. If you don't have sata, you should upgrade the mobo.
The power supply should just be upgraded, period. Get rid of old ones whenever you feel like not losing the whole system because you felt like cheaping out, which should be always. Of course, I'm talking to somebody who probably has no idea why the power supply is important to be of good quality, if you bought a good one that you know was good when you bought it then maybe you can keep using it, who knows.
So then, the major difference between the dual and the quad would be if you plan on doing anything in the background while surfing the internet (the major activity of the average, non-techy person). Even then, a great dual will be a better value than a decent quad. The quad is for when you want to speed through something like photoshop or cad, but you likely know a lot more about computers if you're like that. I've known a lot of "average" people who use photoshop though, so it could be an OK investment. If you show them handbrake or anything that uses ffmpeg, and they seem interested, a quad should get more attention.
Obviously if you want to max out your gaming potential, a quad would be the minimum for that kind of mindset. Yet still, a great dual would likely beat out a decent quad. Even starcraft 2 only uses two cores I think, and that is arguably one of the pinnacles of PC gaming right now. Source games, another pinnacle, are optimized well enough that duals are enough. Games that run better on quads like GTA
I use an OS that doesn't suck, I can in fact, have an app trying to use 100% of the CPU and STILL manage to get work done because it won't let it! Its called a 'pre-emptive multitasking OS'. Maybe you should try one. Not sure what OS you're using that doesn't do this but its gotta be pretty useless now days.
One core is more than enough for almost everyone. Office apps don't really use a lot of CPU, even Office 2010. What web pages do you use that you run so much JS that you notice it running? Contrary to what Mozilla and Google are ranting about, JS speed hasn't been an issue for years, even if its the only change they've made to their browsers recently worth mentioning.
Contrary to popular belief, most people aren't trying to run quake in javascript. Your argument is dumb as it stands.
You should have referenced flash. You're argument would still be dumb, but at least you'd come up with a reason to need more CPU.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
If you want misleading names how about Exabyte tapes? In terms of performance they actually stored about a billion times less than their name suggested. If Intel followed suit we'd have an Exahertz or Gigacore CPU range.
Doesn't really matter either way, for $500 you're still getting a kick butt quad core computer.
I'd be more exited if applications took advantage of the cores. Would be nice if Firefox used more then 1 core. Would be nice if World of Warcraft used more then 1 core. I know WoW uses the second core for sound, but that doesn't count.
So many games, and so many applications should really use these cores, but they're not. We even have OpenCL which should make this easier, but nobody is using it.
I thought Intel's website had a pretty good "joe consumer" cpu choosing wizard to help them pick out a CPU.
If you're buying a computer, you're probably replacing one. Why? Because the old one doesn't do something that you need done (the other possibility is that your present computer is junked up and it's easier to replace than to fix).
You will buy a computer that will do what you need done plus provide a bit of headroom for future applications. Past that your choice will depend on two things:
1 - the skill of the salesperson trying to maximize his/her commission.
2 - your ego.
There are lots of people who will buy the maximum cores they can get for the same reason they will buy the most expensive car they can get the bank to lend them the money to buy. Sadly, I suspect there are more of those people than the ones who will buy only what they are likely to need in the life of the computer.
I agree. There are also a lot of background tasks that the system needs to handle, and giving it an extra core or three to work off of really helps things to run smoothly. I have a four-core Mac Pro desktop from 2006 at home and it runs circles around my MacBook Pro dual-core laptop from 2009.
"Give a man fire, and he'll be warm for a day; set a man on fire, and he'll be warm for the rest of his life
How true... I'm actually in the same boat where I've been thinking about upgrading for the last 2 or 3 years but haven't actually pulled the trigger yet. Sure I can look at charts and reviews and get a general idea of what's got the best performance for some tasks, and then try to cross-reference with the deals at different retailers but who has time for that? In some ways I miss the days 10 years ago when you went with AMD or Intel and got the fastest MHz you could afford (I certainly don't miss the prices though!)
The sending of this message pretty much inconveniences everyone involved.
The computer industry should take a lesson from the auto industry. When you buy a car, they don't bog you down with the details of bore, stroke, compression ratio, number of cylinders, RPM of max HP or any of the other components of producing power in an engine. They have a simple to read MPG rating for each car, so you can purchase with confidence and then feel like a complete schmuck when your 36MPG-rated car only gets 22MPG for you.
I need trepanation like I need a hole in the head.
Well, I have 6 cores in my processor, and that's jolly nice. Means I can do make -j6 and game and virtualise. Isn't that nice?
Could y'all stop it with the penis jokes, I think you're all jolly immature. Now that's not so nice.
Incidentally, I appear to have turned into Ardal O'Hanlon. What do you think of that? Do ye like it? I hope ye do. 'Cos it's a lot of effort, y'know?
If you are a typical user you will only need one or two to run an OS with Web Browser and Word Processor.
Those are just the tasks that average users have in common. You'll find the vast majority of users do much more than just web browsing and word processing, but that same vast majority have those tasks in common.
They just represent a unique class of device production, to which there may even be revisions.
These model numbers DO map to an architecture type, core count, speed and all the other tech bits that havent' even been recorded yet.
Its just a table lookup, rather than having the data encoded in the model number. CPUZ for instance has a pretty complete lookup table of all the mappings.
Its not like the data isn't known, probably a website (www.intel.com) that would let you look up the models pretty easy, maybe even some tools to do it for you (http://lmgtfy.com/?q=intel+processor+identification).
Look, I realize it shouldn't be this way, but it is ... People reading websites like slashdot expect the 'editors' to know a little bit about what they are 'editing'. I realize that on the Internet, thats a really dangerous assumption to make, but never the less, people, myself included, expect that when we see something on websites devoted to 'news' that someone will have put at least a marginal amount of effort into weeding out the rubbish, the snake oil, the milkmaids asking questions about being a car mechanic, and the sensationalism.
Timothy does exactly the opposite of what is expected. Every story he promotes to the front page FAILS the most basic of journalistic tests.
You guys like him, I got that, I'm, sure he's a good guy, but maybe someone could teach him a little bit about how to determine what to promote to the front page and maybe encourage him to do a little (doesn't take MUCH with google around) background research before promoting things to the front page.
Slashdot is becoming a lot more mainstream thanks to Google News so its only natural that more silly stories are going to be submitted, but for the love of god can you please put a little effort into keeping it relevent to the Geeks who have been here for ages and like the old school 'news for nerds' slashdot.
Please.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
You're a few years out of date.
My current system is a C2D 1.8GHz E6300 that's now pushing 4 years of age, yet according to all the benchmarks I've seen by Anntech, Tom's Hardware and others, my performance results are less then 20 percent below the latest/greatest CPU's. This raises the question why in hell should I purchase/build a new system as yet when it still suits my needs?
First off the software I run is mainly single threaded. Very few apps depend on much more then my 1.8GHz clock speed to run well enough for my needs. Second I'm not a gamer or someone running heavy number crunching or other specific apps like Video Transcoding/Editing so that's not important to me.
All this means is that I don't need a faster system. What I do need is ever more storage space due to my growing collection of multimedia and porn. Other then that, I don't really need more then what I now have and with No-Script enabled in Firefox, I could easily get by with my old 700MHz Celeron and 512M of memory running Win-Me. Yep I still have a working system running Win-Me and it does it's job nicely - which is home work for the kids.
Mod me up/Mod me down: I wont frown as I've no crown
They proved long ago they do not win on price/preformance.
Assuming you're talking about standard desktop processors for Athlon V. Pentium4...you're right. For Phenom V. Core...you're wrong.
And they are still the only company that ever sold me a defective chip that couldn't do math.
Yes that Flag erratum - or whatever they called it - was bloody annoying, but so was the more recent AMD TLB issue.
I tend to prefer an non-religious approach and just buy whichever is best at the time.
I care a lot actually, because I need to know whether I can type make -j 200 or make -j 4
It useful when developing massively parallel build systems.
No, I don't core.
Which one is faster - 9600M GT or HD 5450 ?
Are they even comparable ? Same or different price range, or maybe one of them is mobile ? Both ?
A new customer has no idea already. And today, there are 100s of models on market, whereas 10 years ago it could be counted on both hands.
I have nothing to lose but my bindings.
http://sciencecastle.com/sc/app/webroot/img/experiments/158.jpg
Table-ized A.I.
an outrageous 5 crappy applications
Theoretically yes. But in practice one of them could hog the file system, or another might freeze the entire desktop.
Glad someone jumped in here supporting the argument for virtualized machines. Just want to add, the individual VMs all run faster when assigned only a single core. I didn't realize this at first, but learned this from a slashdot post years ago.
The idea as I understand it, is ideally you don't want to share real cores between virtual machines. And hopefully there's enough cores to spread around your virtual server farm.
You can't be ahead of the curve, if you're stuck in a loop.
Unfortunately it's pretty common to see regular office desktops with fast multicore CPUs and ridiculously low amounts of RAM (I've seen C2D 2+ GHz CPUs coupled with 512 megs of RAM, it ran slower than a low-end P4 with 2 gigs of RAM).
For an office machine 512 MB should be a ridiculously large amount of RAM. OS and application code has gone totally batshit crazy.
... also, I can kill you with my brain.
POS software that needs a gig of RAM? I'd like to hear the justification...
If I have more cores I can worry less about all the viruses running on my box stealing my speed right?
Attention to detail is what makes a good troll really pop.
Keep plying your art, my fine friend.
DRM: Terminator crops for your mind!
Boycott Intel, boycott Israel! Switching to AMD processors is too easy and sends a strong message that the 9 shot dead, the 5 shot dead and throw overboard and the 34 wounded. Then we had the media lie of forged video, $1M in stolen equipments, the refusal of the New Zealand lead UN investigation...
So stuff Intel and buy AMD.
Just wait till they start adding graphics cores. oh brother.
He's an AMD fanboy, of course he's a few years out of date.
Hey All, I can say from my 1st hand experience that rendering the same scene in Maya with Mental Ray on win7 has improved going from 4 core to 6 core. A good percentage in speed increase and saved time. If you don't use a "well designed" multi-threaded app then save your money I guess. For pro 3D more cores the better. Cheers J
I'm guessing the 1GB DIMM is the cheapest memory you can attach to an Atom motherboard these days. Not cheapest per GB, but overall cheapest.
Remember that the older/smaller/slower parts become more expensive after production has stopped and the technology moves on.
More data, damnit!
has enough power for me. I just want a fanless ARM ubuntu nettop (2 cores) with 2GB RAM, ATi 4800 graphics, and an Intel X-25V.
Since there are 1.5ghz processors and memory chips are about the price of poststamps, i tell all my friends and family to buy the *cheapest* computer they can find with the biggest harddisk. Everything from 1.5ghz and up is just 'good enough' to do anything a normal consumer will ever do. That's never failed.
Quack damn you!
"Should be," sure. Here in reality, it doesn't work that way, and RAM upgrades are often the best thing performance boost for an office-oriented PC.
And RAM is cheap. Cheaper than CPU and motherboard upgrades, cheaper than SSD, cheaper than anything for the amount of performance increase.
(I used to get a lot done with 640k, but it's not 1991 anymore...)
Kid-proof tablet..
Yes, and a brick could fall from the ceiling and break half of the keyboard. In practice a lot of other events not related to the practicality of having more cores with virtual machines can happen. So what's the relevance of your comment?
Even if only because "make -j8" would be frickin' awesome.
Visit http://ringbreak.dnd.utwente.nl/~mrjb/growingbettersoftware to download your free copy of the book
Depending on the sort of code you're running of course. It would be quite easy to see no performance increase going to from 4 to 6 (or 8) cores if the memory bus is already saturated.
Also, You must take into account the overhead of communicating between cores - which will require some sort of bus too - the AMD CPUs are using LDT (Hypertransport?) and if you look at their specs the number of inter-core links has increased as the number of cores has.
I have a wonderful book on OCCAM from the 1980's that goes into great depth on butterfly switches, bus muxing, hyper-spheres and hyper-cubes... still boggles my mind we're trying to catch up with Inmos from 30 years ago!
AC.. cos'.. well just because!
So weird to get reality checks like this post. I mean you assume the geeks were the smart, but socialy inept... yet you discover at any turn that the geeks aren't smart thus the social problems.
The CPU numbers were never clear. 486 100mHz it's not the same as Pentium 100mHz, which is not the same with 586 100mHz. And this is only in the same league. Enter different generations, different manufacturing processes, RISC vs CISC, different cores...
At least Intel is fair now and does not imply that there is a connection. And this just makes the stupidity surface - "I can't make any sense if there's no Hz count"
http://www.accountkiller.com/removal-requested
People want something they cannot have: A single "goodness" number to tell them if something is better than something else. It is from this misguided desire that we get the over focus on single standardized tests and so on.
Well, life is more complex than that, and the same is true of computer hardware. You can't have a single number to tell you which graphics card is best. It is more complicated than that. A new card might be a slower speed than an old one, but do more per clock. Or it might have less ROPs but more shaders. You cannot demand a single number to sum it all up. You have to look at what it is you want to do and then consider what is best.
Alternatively, you can just operate by money and buy something in a range you can afford every so often.
Whatever the case, it cannot be simply summed up. The numbers are often a best effort to try and help. In ATi's case the first digit represents generation and thus technology so a 5000 series is DX11, 40nm, a 4000 series is DX10.1, 40nm and so on. The second is the major performance number, like how many units it has, 8 being fast single GPU, 3 being slow GPU and everything else being inbetween. The last two are minor performance numbers and mostly tell you if something is clocked faster than others.
It's as good as you can do. A Radeon 5870 tells you a bit useful. Otherwise it would be a "Radeon 40nm 32 ROP 80 TMU, 1600 shaders model 5.0 @ 850MHz 1024MB GDDR5 256-bit bus @ 1.2GHz with 4 transfers per clock." Even that doesn't tell the whole story, in addition to being confusing as shit. there' sno single number that tells you anything useful.
So either:
1) Buy based on general numbers the companies set as guidelines to try and help you see what is faster than what. Accept that it is an imperfect simplification.
2) Buy based on price, as in "I buy a new graphics card every 18 months and I spend between $100-150 on it." Accept that you get what you get, but that price is a pretty good indicator of performance in the computer world.
3) Do some research and learn what component best meets your needs given the tasks that you do, and get that. Accept that you have to turn your brain on and spend a bit of time on it.
First of all, we were talking about applications, not virtual machines.
Second, My events happen frequently with crappy applications, while yours do not.
Last but not least, I really hate arguing with AC, especially those who think they are smarter than they really are.
On Intel's desktop CPUs, the clock speed isn't all that variable. I mean they throttle down when idle to save power, but their "normal"" and "turbo" operating states are very close. For example their 6 core i7-970 operates at 3.2GHz normally. It can bump cores up to 3.33GHz if thermal and electrical parameters are in spec (generally if other cores aren't getting hit too hard) or bump a single up to 3.5GHz in some cases. Not a whole lot of variability.
You see more in laptops, because thermal concerns are more pronounced. Generally speaking the dual core CPUs are a lot faster than the quad core ones, at least historically. However these days, the quad cores can be flexible. For example an i7-720QM is sold as a 1.6GHz CPU. However if only two cores are loaded it can throttle up to 2.4GHz, which is the kind of speeds you see out of the lower range dual core mobile CPUs. With a single core loaded, it can go all the way up to 2.8GHz.
Basically it offers the ability to have more cores, or faster cores as needed. A laptop lacks the electrical and thermal capacity to do both, so it can trade off as needed.
"Core overhead"
As soon as you've gone to SMP -- which all modern kernel have, the playing field is pretty much level.
2 core, 4 cores, 8 cores...there is no processor overhead unless you recompile your kernel -- and even then the difference between a few hundred bytes out of a few gigabytes is meaningless.
What the "less-cores-is-good" guys are ignoring is the extra cache you get with more cores.
4 Core systems topped out at 8Meg cache.
6 Core Systems top out at 12 Meg cache.
If you don't use the extra 2 cores -- you can shut them off in the bios -- you get
get 4Meg more of program and data cache -- or you can limit your progs at run time to only use the lower 4 cpus...which-ever.
Now who's the first idiot who wants to make the argument that your general singled threaded application won't benefit from .. 6 times the cache?
And um lower clock speed?
4 cores max speed @ 3.2GH (with 8M cache, or 3.4 @ 4M cache).
Now I have 6 cores @ 3.2 and 12M cache and it wasn't the fastest CPU.
You guys should check your figures before you try to make case: FAIL!
And a minimal amount of research will tell you this. Intel's 6 core chips come off their new 32nm lines, since space and power are a premium. Intel's 4 core chips come off their much more prevalent 45nm lines. They are completely different processes and thus one is not sold as the other.
Intel has pretty good yields, they traditionally have, and thus don't have a real reason to do that sort of thing. It is more economical to fab quads on the more available 45nm process than to make them out of any failed 6 units.
The process of choosing a CPU (or any component for that matter) has never really changed. This is what you do:
Any other spec is just marketing.
Pretty good is actually pretty bad.
Everything I see shows that modern OSes not only don't have an overhead with more cores, it helps things. Reason is what OSes really have is a heavy context switching overhead. If a processor is doing something, and the OS needs it to do something else, it has to generate an interrupt, push everything on to the stack, switch to the kernel, switch to the net process, etc. It is a hefty overhead. However that all goes away if instead multiple things run at the same time on hardware. They don't switch contexts, they just keep running.
This is the reason why web/DB heavy servers like to have lots of cores, even if less powerful. Sun's new chips are designed with that in mind. Each core can handle 8 threads in hardware, meaning it acts like a 64-core CPU though only having 8 actual cores. Why? Context switching. The tasks it normally deals with are not high load, but they switch around a lot. The more than can run side-by-side from the OSes perspective, the less overhead and the more efficient use of processor resources.
In a desktop the tasks are more intense so it is less useful to have lots of threads/CPU (currently 2 is the highest in the Core i3/5/7 series) but more cores are still quite useful. It allows for more things to happen at the same time, from an OS perspective, and lowers overhead.
You notice too, using a multi-core, multi-threaded system. Things are damn responsive.
Turns out in modern games, a lot of shit happens at the same time. While this was traditionally coded as a bigass while loop because systems were singe thread, it doesn't have to be. You can thread all that shit out and have the game engine do multiple things at once. It is still being worked on, but it is getting much better. Most very modern (as released this year or perhaps last year) games make extremely good use of two cores to the point that many require it. They can fully load both, no problem. A smaller number, but increasing amount, can make good use of 3 or 4 cores. Game designers are learning how to code in parallel, tools are developing to make this work better, etc.
Games are already parallel and are only going to get more so.
You mean it becomes a race between which memory-leaking app can make the host OS collapse faster.
Remember, if the guest VM needs to swap, that's swapping from VM memory to VM HDD to Host memory to Host HDD, and back again. Something's gotta give...
Finally had enough. Come see us over at https://soylentnews.org/
If I go to Intel's page for the i7-970, which is easy to find they list: # of Cores, # of Threads, Clock Speed, Turbo Frequency, Bus/Core Ratio, QPI Speed, # of QPI Links, Instruction Set, Instruction Set Extensions, Lithography, Max TDP, VID Voltage Range and a whole bunch of other shit. Everything you could want.
So, what is easier:
1) Call it the "Core i7-970" which gives you a bit of info about where it falls in the series, and an easy to lookup number for their site to find more info as needed.
or
2) Call it the "Core i7 3.2/3.46 GHz 6/12 core/thread 12 MB 4.8 GT/s QPI SSE4.2 AES-NI 32nm 130 W" which is extremely confusing, and doesn't fit in most product headlines?
You have access to this thing called the Internet. Go to ark.intel.com - it's been there a long long time. It tells you *all* the details of every chip they sell.
They do sell a huge array of chips with many subtle differences - it's not like the good old days when there were 4 speeds of 486 to choose between - bus speeds, chip speeds, integrated GPU, process node, cache sizes, VT-x extensions, SSE4.2, AES, power draw, number of cores, Hyper-threading, form factor, socket, lead-free, PCIe lanes, ebmbedded...
Many, many people don't care, and buy their PCs to a price, and based on how it looks. For those of us that do care, or need a particular feature, it's *really* easy to find out what's what.
It should be ; in reality the machines we have in the office eat 650MB of RAM just to boot because of all the corporate paranoiaware installed on them.
I'm still using a Thinkpad T23 with a 1.2 GHz PIII-m as my main system. The fastest thing in the house has a 2 GHz P4. Both processors are fast enough for just about anything I'll want to do with them, even though I have to wait a bit longer for my builds to run through and I don't really fancy building eg. OpenOffice with them. Speed is overrated. What is not overrated, and what will finally get me to look for something a bit newer than these ~10 year old systems is memory. These PIII and P4 boxes don't handle more than 1 GB. When memory was expensive that seemed a lot. Now that it is not I'm getting more and more envious of those 4GB+ machines which sit on the desk doing nothing at all while I'm building the software they run on my limited hardware. Even though it is intentional that I use limited hardware to create software - if it runs here it will will run like greased lightning on modern stuff - that trick is getting a bit old now...
--frank[at]unternet.org
"an enhanced lubrastrip and improved blade suspension system" for a smooth, close shave... wait, what are we talking about again??
I absolutely love the argument "let's buy faster hardware, so that we can run more stupid crap on it".
Meanwhile, Internet Uses 9.4% of Electricity In the US and Laptops catch on fire.
All I care about is that when I alt-tab from a game that's fullscreen whilst waiting for a level to load, my desktop comes back -instantly-. That never happens. So what gives? What's holding up the instant switch back to desktop? Why is 'reading something from disk' blocking the entire computer in the year 2010 with 4 cores (of which only 1 is being used) and 4 gig of ram (of which 2 is used)? Is DirectX or Windows just designed totally incorrectly or can a judicious adjustment of process priorities fix this?
I now want spend money on every bit of my machine. The main problem is all the games I want to buy have some sort of DRM. So there has been no need to upgrade as I WILL NOT buy drm games that would spark an upgrade. I’ve saved a fortune on games and parts so far. So DRM hurts hardware sale too.
Unless I'm timing them, I'm hard pushed to tell the difference between my personal computers. I have 2.0GHz C2D, 2.6GHz Core i7 deskop and 2.4Ghz Core i5 mobile. They all do everything I need.
Today, the graphics chip makes a bigger difference to me: I have two Macs with the same CPU but one has an ATI chip and the other Intel GMA. Guess what, the Intel GMA drives me crazy.
I guess I'm waiting for the next generation of CPU intensive killer apps.
"If you don't use the extra 2 cores -- you can shut them off in the bios -- you get
get 4Meg more of program and data cache -- or you can limit your progs at run time to only use the lower 4 cpus...which-ever."
There is no limiting factor to cache other than cost. Its much cheaper to do something like 6x2mb than 1x12mb.
Now, to utilize that cache, you need a processor to address the memory space there, so shutting down two cores won't let you use 12mb of cache, it'll let you use 4x2mb (8 mb) with the other 4 going to waste.
While a 96-core system will run Crysis!
Comment removed based on user account deletion
http://xkcd.com/670/
The handicapper general peered odiously over her spectacles at the latest product roadmap, her mood growing ever more dark at the clearly typed columns and performance ratios for common tests of the new devices. Why couldn't those damn engineers be more like marketing she thought, comparing the crumpled smudged, poorly written marketting copy to the neat engineering report. Why, even when the marketing copy was well typed, spelled and readable it still made no real sense. The engineers she realized had tricked her. "Bunny-Marks",what she had initially believed had been the wholehearted attempt by those engineers were a scam! She should have realized sooner regardless of the terms ' Number-of-paws', fur-fuzz' , 'ear size' and 'carrot factor' were clearly false names for real things. Even through her permanently oil smudged glasses she could see that there was a clear progression from slow bunnies to faster bunnies and this wouldn't do. Slowly, a smile slipped across her face as she realized now she would have the justification to depressurize the scientists lab another 3000 feet. Let's see how those engineers did at 14000 ft! Oh she would make all the bunnies equal, or her initials weren't DMG.
Yes, I do have many uses for my many cores (I need about the equivalent of an i7 920 that I have to work efficiently with my machine and indulge in my hobbies, right now) , and I also do find these product naming schemes the CPU vendors employ quite useless. But they can't really be useful - it is the specifications are the actually useful, since the presence / absence of various features is just the bare minimum to roughly determine how fast these high tech devices are.
CPU are highly complex devices that are attached to more highly complex devices and that perform highly complex tasks. Ultimately, they exhibit complex performance metrics. Gigahertz or the amount of cores are only very, very roughly indicative of what actually makes a CPU of the current generation good - usually its new co-processors or other tricks that may actually be the most important performance feature, conditional on your software using these.
Obviously, I doubt most people who don't know the marketing numbers would understand the specifications any better. But this difficulty in assessing performance is just the nature of the beast.
So how to deal with it? Simple, you can approach it by either a) buying more or less blindly, most probably a CPU(machine in the mid-price range), b) running casual performance tests at an assembled machine, or c) getting an informed opinion from a hobbyist or professional on how well a given CPU performs on your software / hardware setup vs how much it costs.
6 cores. Do You Care?
Written like someone who's never heard of 'make -j'. Seriously, anybody that compiles stuff wants more cores, and if you ever reach a point were disk IO is the bottleneck just throw in an SSD.
Random project on my box:
make clean; time make -j8
Real: 4.3s
make clean; time make -j1
Real: 14.7s
Compiling is an inherently parallelizable task.
___________________ I want to be free()!
How many buyers care about the engine in their car? Forget about "car guys" - I'm talking about the other 95% of the car market.
Most people I see buying cars look at size, style, and reliability. The engine choice is almost irrelevant for many buyers.
I recently asked my sister what engine she would pick for her next car. Her answer: whatever gets good fuel economy and can get her onto a highway at a decent speed. Forget about the more "exotic" options like diesel and hybrid drivetrains.
Computer purchases are a similar thing. How many people really care what CPU is in their iPhone?
-ted
I do care. I am still on a single core and struggle with various item on the net. Mainly flash, but that is a different story all together. I want more cores and the apps to match. Just my bit on this.
"The laws of science be a harsh mistress." --Bender
I have a 4-core, 8-thread system for a number crunching application, it works beautifully with a tightly designed .Net4 multi-core application for a nearly 4x speed increase. Would love to get my hands on a 6 core :)
Who cares how Intel numbers the CPUs - it comes down to Core-Cache-Clock relationship & your individual needs.
We need to switch to a system with lots of cores, all with their own local memory, and able to send each other messages.
You're describing the Immos Transputer - way ahead of it's time when it was released c. 1980.
Each Transputer core was an interconnected computing mesh node, with it's own memory, with specialized hardware-based communication channels connecting it to other nodes. It was designed hand-in-hand with a specialized language "Occam" (as in Occam's razor) that directly mapped onto the parallelism and communication/message-passing abilities of the hardware.
If you said a section of Occam code could run in parallel then at run-time it would be distributed across nodes if available. If you defined a piece of code to read from an input channel, that could/would be mapped onto one of the hardward communication channels where it would be connected to another "process" writing to a channel somewhere else on the Transputer mesh it was deployed on.
One nice thing about Occam was that the language parallelism and communication constructs, although designed to map directly onto the hardware, didn't have to, so you could run any Occam program on any sized Transputer network, and the run-time deployment would map it onto the available hardware. They had nice demos showing things like Mandlebrot generation or 3-D scene fly-thru just getting faster as it was run on bigger networks.
Of course you could do the same today with, say, pthreads - a heavily multithread app will just get faster as it is run on hardware with more cores, but the nice thing about the Transputer and Occam was that it was scalable. If you wanted 100 cores you could build it, while today you're limited to the processors Intel/etc build, and unlike Occam your programming model totally changes (pthreads -> network based) if you need more cores than fit on a single chip or on a single board.
So.. let's hope the future of computing is as advanced and well designed as it's past was!
I find it difficult to notice any recent gains in hardware or software. My home/work computer still generally runs just as slow as it did 10 yrs ago (probably slower). Playing some non-gfx intensive games (like civ) is just as slow and painful today as it was 10 yrs ago. All the cores and 64bit applications and well designed code still leaves me with a sense that at the end of the day, Moores law has nothing to do with user experience, which to me has been more or less steady for the last 10 yrs.
My next computer is going to be an econo model because I simply can't justify spending top dollar for performance I personally can't really notice.
It's called impulse purchases.
Six cores can churn a shared cache quicker than four cores. This doesn't occur in many circumstances but it is a very real concern for database deployment.
thats all i want to know what the total GHZ of the machine will be
i dont care if it has 50 cores or is the earth's core ( neat movie though ...)
There sure as hell is a limiting factor to cache. Cache is not "free" in thermal load, power load nor die space. The cache uses significantly LESS power and produces less heat than an active core, but it takes up a LOT more space on the die. CPUs have a limit, based on their package, to how big the die can be.
4 Core systems topped out at 8Meg cache.
I assume that the use of "topped" in the past tense was intentional, as there are now quad-core Intel chips with 12MB L3 cache (like the Xeon E56xx line).
I use an OS that doesn't suck, I can in fact, have an app trying to use 100% of the CPU and STILL manage to get work done because it won't let it! Its called a 'pre-emptive multitasking OS'. Maybe you should try one.
Yes, I'm aware of multi-tasking, and techniques to avoid process starvation. Some OSes do this better than others. Out here in the real world, employees don't always get to chose which OSes we get to work with.
One core is more than enough for almost everyone. Office apps don't really use a lot of CPU, even Office 2010. What web pages do you use that you run so much JS that you notice it running?
Well, I rarely notice it since I use noscript to avoid all the javacrap. But without noscript, it happens a lot.
Contrary to what Mozilla and Google are ranting about, JS speed hasn't been an issue for years,
Riiiight. So why does firefox have a check for javascript that it running for an extended period of time?
Contrary to popular belief, most people aren't trying to run quake in javascript. Your argument is dumb as it stands.
It's so dumb, that the devs who wrote firefox try to detect it.
... with a graph card that Will. Not. Run. Games. At anything above 12 FPS minimum framerate. Unplayable. $500 : wasted.
Making laws based on opinions that stem up from false informations leads to witch hunts.
How does this slow a program down if the CPU GHZ are the same?
As for shutting down 2 cores shutting off 4MB of core -- unless went backward on their design with 6 core design, in their latest 4 core design, all 4 cores could utilize any or all of the 8MB of core. It was Intel's first Quad core that was with more restrictions as I believe it was effectively 2 Duals on 1 chip.
So unless intel went backwards on their design -- all 6 cores should be able to access all 12 MB of cache memory so shutting down cores will enable the core to be used by the remainder. Also, in intel's higher end chips, shutting down cores will speed up the rest of the cores -- not be even close to a linear speed up, but if all you want is single processor power, you get close to getting the next stepping up in speed out of 1 core by shutting the rest down -- or such was true of their quad cores. Their latest gen all use lower power -- running 6 cores in nearly the same thermal-wattage envelope as the previous gen's 4 cores.
The point was simply that they may not be able to add more cache if they keep adding cores. As well, they are limited by the overall size of the die in how much cache they can implement. There was no talk about "GHz."
A HUGE problem with this discrepancy in core/mhz/performance numbers is that the real big and expensive software packages that run on our servers (Looking at you, Oracle) charge licensing fees based on the total number of cores or GHZ in the hardware no matter whether or not that figure actually gives meaningful or significant performance increases or not.
I'd be OK with disabling most of a multicore processor if it still gave me mediocre performace but saved a hundred grand in licensing fees, and I'm afraid most of my customers would be too.
Well, I don't exactly leave the VM open all day. And since the guest OS is linux based (Running 7 x64 as the host), it's generally pretty fast if I want to reset it.
And in the end, unresponsive (or leaky) programs aren't exactly the rule, so situations like that are only going to show up every now and again.
Yes, they are going to happen, probably. But not often enough or severely enough to dissuade me, especially when I'm only using the VM lightly.
I'm assuming you're talking about a quad core. What is faster, running 4 encoding threads or 8?
I'm curious how HT affects performance that way...
How much is your data worth? Back it up now.
o Ehm, close programs that use 100% CPU.
o For those programs still needed, set Low Priority
for those programs that arent interactive.
o If any of those 100% Programs are multi-threaded,
web-browsers are multi-threaded, arent they?
They will eat 100% CPU on all Cores. Last
chance to solve it is to set Affinity and force programs
to use less Cores, or just one.
I run Firefox with Adblock and NoScript, but there are enough sites where I've enabled Javascript that occasionally Firefox will still freak out and try to run away burning all the CPU. In the past, this was really annoying, as my computer would become a total dog and it would be hard to get FF killed off.
Now that I have a dual-core CPU, Firefox still freaks out occasionally, and burns 100% of one core, but Windows is still capable of navigating, and I can decide whether to kill FF or wait for it to stop whatever it's doing, which it sometimes does.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Who needs to be bothered by details like hardware specs? There is blogging and Starbucks awaiting.
"Under God" was primarily pushed by the Knights of Columbus, who were Catholics and likely to be Democrats, though the D.A.R. and other right-wingers also liked it.
Look at Wikipedia or Google the Treasury's blurb on "In God We Trust" - it's been on coins since the mid-1800s, some paper money since the 1930s, and the only change in the 1950s was making it mandatory on all US paper money and coins. More importantly, the coins used to be silver and the bills promised to give you silver in return - so if they weren't hypocritically putting "In God We Trust" on it, they'd have had to put "Trust Us"...
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
I honestly got to the point where I do not care. With 4 Gigs and a Core 2 Duo I am perfectly fine, running OS X. I can run editors, edit basic videos (they can take long, but I run a render once a week maybe at night). I can browse, look at HD videos, music and on my PC I can play games (that is a 2 core, some older SLI Nvidia, and most games run just fine).
When even games run OK on a Core 2 Duo, and all my work applications, why look at i5, i7? I have colleagues, who questioned my purchase of a Macbook pro 13'' (Core 2 Duo up to today), and I realized, that I had no urge to get anything faster AT ALL, but prefer a better battery life, and a smaller laptop (apple i5/i7 are 15/17 inch, Core 2 duo is 13 inches).
Just my 2c. Rendering, crunching happens on server in my world, browsing, and needed apps work just fine on the older ones for me :)
One core is sufficient for 99% of office workstations that only run a browser and MS Office applications. Dual-core is a bit more snappy, but I'd rather spend an extra hundred bucks on an SSD for the O/S volume than a more-core processor.
Gods no. Have you ever spent time using equivalent machines, one with dual-core and one with single-core? The dual-core blows the single-core out of the water in terms of responsiveness. Quad-cores might be overkill for office use, but only until their prices drop.
And it's not like multi-core is expensive anymore. Hell, NewEgg only sells (2) single-core CPUs now ($36 for AMD, $40 for the Intel). In comparison there are (42) dual-core CPUs listed, and they start at about $50. That's a far cry from "save an extra hundred bucks".
Nowadays, I usually shop by a combination of thermal design envelope (i.e. look for a 65W CPU) combined with "what can I get for $80". And that number used to be $150 for the CPU. With quad-cores as low as $80 now, it won't be long before we'll be putting quads on the desktop. Hopefully something with a 45W or 65W design so that we can keep them quiet and cool. The current quads are 95W or 125W for the most part. I wish AMD still made their 45W CPU line, those were great for office machines.
If they can get the 150GB SSDs down below $100, I'll strongly consider switching all of our desktops over. Once you head to multi-core and have enough RAM, HD speeds become the biggest bottleneck.
Wolde you bothe eate your cake, and have your cake?
i just want good old-fashioned giga-hurtz.