26 Desktop Processors Compared

theraindog writes "The number of different CPU models available from AMD and Intel is daunting to say the least. The Tech Report's latest CPU review makes some sense of the landscape, exploring the performance and power consumption characteristics of more than two dozen desktop processors between the $999 Core i7-975 and more affordable sub-$100 chips. The article also highlights the value proposition offered by each CPU on its own and as a part of the total cost of a system. The resulting scatter plots nicely illustrate which CPUs deliver the best performance per dollar."

Comments

Seems pretty clear:

[fuzzyfuzzyfungus]

AMD has pretty embarrassing performance on the high end, which makes your choice there downright trivial; but is an excellent value in those niches where they have an entry.

Re:Seems pretty clear:

[sanosuke001]

So a $1000 i7 Extreme is an easy choice over a $250 Phenom II 955? AMD in a heartbeat

Re:Seems pretty clear:

[mdm-adph]

Not embarrassing, just not as fast. However, a Ferrari is vastly more powerful for the money, but a Corvette is still the much better deal -- do I really need to explain this on here?

Re:Seems pretty clear:

[caerwyn]

No, but the $300 i7 920 surely is...

Re:Seems pretty clear:

[sanosuke001]

Forgot to add...

The only thing I'd get the i7 over the phenom ii's for is triple channel ram. That's the only upside I see with Intel chips.

Re:Seems pretty clear:

[mdm-adph]

Well now, they are faster. They're just not $800 faster. Nowhere near it.

Not to mention that the corresponding hardware you have to buy with a Core i7 (the expensive motherboard, the expensive triple channel ram) makes it even more not worth it.

Re:Seems pretty clear:

[Anonymous Coward]

Plot isn't that clear to me at all.

Price/performance is slope on that graph, and AMD looks pretty good at first look, occupying a band mostly on the top-left.

They ought to have plotted performance on the independent axis vs. price/performance on the y axis. And put that sucker on a log-log plot. Then we'll see who's got the best price/performance for various performance levels.

Re:Seems pretty clear:

[hansamurai]

I don't know about that, the AMD X4 955 is on par performance and price wise with the Intel Q9550. Of course, the i7 line is probably what you mean by high end, but you're paying at least $200 - $900 more for a 25-75% performance gain.

Re:Seems pretty clear:

[mister_playboy]

The Corvette ZR1 has more horsepower and is less expensive than any current Ferrari.

I love Ferrari... but the Corvette needs no apologies at all!

Re:Seems pretty clear:

[characterZer0]

It does if you want to turn.

Or if you want to not look like a redneck.

Re:Seems pretty clear:

[Anonymous Coward]

Not if you factor in the motherboard price difference - about $100 for x58 over AM2+. The i7 is brilliant for heavy, multi-threaded workloads but I question how much of that performance is really worth it for an average consumer or even Slashdot reader.

Re:Seems pretty clear:

[mister_playboy]

A quick glance through some magazine tables shows the ZR1 with more grip (1.05 g or more) on a skidpad than any Ferrari, as well. Only the Viper ACR has done better.

Rednecks don't drive cars that cost more than 100K USD.

Re:Seems pretty clear:

[Rewind]

So a $1000 i7 Extreme is an easy choice over a $250 Phenom II 955?

AMD in a heartbeat

Why exactly are you comparing an extreme model to a Phenom? The Extremes are sort of an outlier. Also it is all relative. For a 'high end desktop processor' I would take a $180ish E8400 over a $200ish X4 940. Then again if you are really using a lot of applications that would benefit from quad the tables could easily be flipped. Really all depends on what you are running etc. Either way I think most people consider Intel's extreme models out of the realm of desktop PC, gamer or no. They obviously sell or Intel wouldn't have them, but they have never really been a great value to most imo.

Re:Seems pretty clear:

[athakur999]

The Phenom II X4 955 beats the i7 920 in 3 out of the 4 games they tried. The only one it lost was Cryis Warhead and it was a narrow loss (48 vs. 46 FPS).

These price difference of these two chips is about $35 on Newegg. I think for gamers, getting the X4 955 and putting that extra $35 towards the video card will net better results. This isn't counting the additional cost of DDR2 vs. DDR3 memory which has minimal effect on performance right now but still has a big price difference.

Re:Seems pretty clear:

[fuzzyfuzzyfungus]

For somebody motivated primarily by performance, not price/performance, yes.

I am not one such; but they are hardly nonexistent.

Re:Seems pretty clear:

[Znork]

Personally I don't find it that trivial; never before has the best choice been so extremely dependent on what you need the performance for.

Say you've got some eminently parallel task, like ray tracing. With the huge price/performance difference between low end and high end you might beat the high end in performance by buying two cheap systems rather than one expensive. Look at the i7-940; it's just barely twice as fast as the cheapest CPU, yet it costs six times as much. That price would easily accommodate a cheap motherboard and memory and you'd still be shelling out less.

On the other hand, say you have very few parallel tasks, then you may still be as well off with a cheap CPU. Without several tasks you're not going so see much difference between a dual core CPU with good per-core performance and a quad core. Not an entirely rare situation when talking about desktop systems.

Or you might be in the sweet spot of latency sensitive parallel tasks, possibly applicable to some games, in which case a more expensive quad core CPU might definitely be what you're looking for.

And add to that the need for actual application specific benchmarks to determine actual performance as opposed to generic benchmarks... well, I wouldn't call the choice trivial.

Re:Seems pretty clear:

[Rockoon]

I'd also question its price/longevity scale as well, since the sort of folks who are going to drop $700+ on a basic (no video card, no monitor, no hd's) high performance system are probably the same sort of folks who are going to scrap that box as soon as Larrabee hits.

My AMDx2 3800+ is starting to show its age, but there is no way I am going to buy a high end part with Larrabee so close.

Re:Seems pretty clear:

[hairyfeet]

Not to mention that for most folks that "ludicrous speed" is simply overkill. I've been building my customers a mix of Pentium Duals and AMD 7550s and all they can talk about is how blazing fast they are. Because for most of the stuff folks are doing today, even games, the CPU is rarely the bottleneck and passed "good enough" a few years back.

So at least for my customers and myself (I liked how the last AMD 7550 ran so quiet so I built one for myself) it is more about "bang for the buck" and when I can build a nice AMD 7550 with 4Gb of RAM, a 780 board along with a 300Gb Sata 2 and a nice black case for $281 shipped it is just nuts to blow all that extra cash in this economy to get a bigger epeen. For most of us even the bottom of the line dual cores are sitting idle a good 80% of the time, so why blow the extra cash? I just checked process Explorer and with 9 tabs plus Comodo Internet Security on XP X64 I've still got 3.2Gb of RAM free and am only using an average of 1.4% CPU, plus I got another 1Gb of RAM on the $50 HD4650 for watching videos.

Considering my first machine was a VIC20 where I had to peek and poke everything all this extra power is nice, but I doubt I'll be saying "you know what? I need more power!" for quite awhile yet. So for a box that cost a hair over $500 counting the OS and GPU I'm a happy little camper. While having ludicrous speed might be good if I was doing major CAD or graphics work, for the things that myself and my customers do with their PCs there just really isn't a point. Hell I gave my nearly 5 year old 3.6GHz P4 to my oldest and he is blasting zombies in Left4Dead even as we speak, so even that CPU is "good enough" for what he wants to do with it. The "bang for the buck" you get nowadays even on the low end is frankly unreal!

Re:Seems pretty clear:

[sznupi]

I think that by choosing Corvette you made you car analogy inadequate (and that would be inexcusable! ;p )

You were looking for any Porsche or Audi R8...things you can actually use for daily driving.

Re:Seems pretty clear:

[Latinhypercube]

Wait until your kid is 3d rendering in Maya and you will be screaming for 8 core Xeons.

Re:Seems pretty clear:

[EastCoastSurfer]

Only the Viper ACR has done better.

So they finally fixed the Viper? I remember that used to be one of its biggest drawbacks - tons of power, but awful handling if you needed to do anything other than drive in a straight line.

Re:Seems pretty clear:

[jgtg32a]

I thought that the phenoms could do triple channel if you put them on an AM3 board, IIRC Intel has the memory controls on the chip itself and and AMD puts them on the motherboard

Re:Seems pretty clear:

[Anonymous Coward]

I got my i7 920 for $199.99 from Microcenter (in-store only). I have the receipt right here (still need to do the rebates for other comp. bits)

1/3 off the CPU was certainly a major decision for this last build.

Re:Seems pretty clear:

[poetmatt]

That's just the motherboard, let alone the processor or ram cost differences. Not to mention they are about to change the socket again rather soon, if I recall correctly (aka 1 year).

AMD changes sockets a lot more gradually, letting people actually, you know, upgrade.

Re:Seems pretty clear:

[geekoid]

WHen I can I buy the highest end I can and use it for 5 years.
I might upgrade a video card. After 5 years, it gets deprecated into the house hold pool.
I have 5 year old computers running games like team fortress 2 and wow just fine. Not the highest end graphic performance, but fine enough.

When I build my next box, It will have 16Gigs ram, minimum. Hopefully 32. This will probably stretch it's use out to 8 years.

Re:Seems pretty clear:

[DAldredge]

The consumer i7 chips do not support ECC RAM while the AMD ones do. That could be important to some.

Re:Seems pretty clear:

[Anonymous Coward]

Rednecks don't drive cars that cost more than 100K USD.

True, but some of us do have 100K lawn ornaments!

Re:Seems pretty clear:

[LordVader717]

That's one way to go, but the most economically sound way to do it has always been (slightly) more frequent upgrade cycles and lower range hardware.

Re:Seems pretty clear:

[iamhassi]

"AMD has pretty embarrassing performance on the high end"

But it was interesting to see all the lower end AMD models being of equal performance and just a few dollars cheaper than the Intel counterparts.

My issue with the graph is someone needs to take a class on "how to make a graph". 0% performance and $0 cpu.... why? Was there a $0 cpu? Did any of the cpus get a 0%? So instead of a nice, full graph that explains everything easily they have big empty white spaces at the x-y axis, then a cluster of ~20 plot points within a small area. Over half the graph is empty: from 0% to 275%, 0-100% is empty, and 250% to 275% is empty, so nearly half the graph is empty white space. Fail

Re:Seems pretty clear:

[Chabo]

Hell I gave my nearly 5 year old 3.6GHz P4 to my oldest and he is blasting zombies in Left4Dead even as we speak

I read this sentence, and at first I thought you were saying you let your "nearly 5 year old" kid play Left 4 Dead!

Re:Seems pretty clear:

[Sj0]

I agree so much I'm going to take it to the next level:

Almost all hardware today is ridiculously powerful for ridiculously cheap. Review sites implicitly agree by reviewing at full detail 1600x1200 or higher.

Re:Seems pretty clear:

[bangthegong]

There are some niche markets where price/performance or raw speed matter, but for the majority of buyers it's irrelevant. I've recently convinced my baby-boomer parents to get a new computer, and in the store they kept asking about the speed as though they would have any clue what 2.4GHz meant anyway. I finally convinced them it was the features other than speed they should consider, and as long as the computer has a modern CPU it was going to work for them just fine. Good enough is good enough for most people nowadays! There are some cases where it matters even for the average user (such as the differences between netbook processors) but for anything with a dual core or quad core, forget it. It is all "fast".

Re:Seems pretty clear:

[dr_wheel]

I know that some people hate to hear the "yea, but if you overclock part X" argument, but here goes...

You can pick up a core 2 duo E8400 wolfdale ($168@newegg) and an arctic freezer 7 pro hsf ($37), and perform a very, very modest overclock from 3.0 to 3.33ghz for a total of $205. Hell, you could probably perform this overclock with the stock cooler with no issues and save a further $37.

Now, you have the equivalent of the $270 E8600 c2d which also rates high in their gaming benchmarks (beating the phenom ii 955, i7 920, and even the i7 940 in their hl2 and crysis warhead benches, and only slightly losing to the 955 and i7 940 in farcry 2).

What's the point of all this? If I'm building a gaming PC on any kind of budget, I'm still looking at the easily overclockable and basement bargain c2d, leaving even more money for a faster gpu. I have yet to see a compelling argument for 3-4 cores for gaming. For example, look at this price/performance data:

Unreal Tournament 3 - 1680x1050
http://www.tomshardware.com/charts/desktop-cpu-charts-q3-2008/Unreal-Tournament-3-1680x1050,819.html
Q9650 - $325 @ newegg - 149.7 fps
E8500 - $190 @ newegg - 133.7 fps

71% increase in cost for an 11% increase in performance.

Crysis - 1680x1050
http://www.tomshardware.com/charts/desktop-cpu-charts-q3-2008/Crysis-1680x1050,818.html
Q9650 - $325 @ newegg - 132.4 fps
E8500 - $190 @ newegg - 134.8 fps

71% increase in cost vs. a 2% decrease in performance.

Sure, you could make the argument that future games will likely utilize 4 cores more effectively. But when? I'd wager a guess that we aren't really going to see a significant advantage to quad-core gaming for a few years yet. Just about enough time to plan the next pc upgrade!

Re:Seems pretty clear:

[Ossifer]

I upgrade on the order of every four years. This means CPU, MB, RAM and graphics card. No way around it.

That being said, I plan on upgrading to i7 920. My current system is a desktop, mail-, web-, file-, print-, dns-, and dhcp-server, router, firewall, and a few more things I am forgetting. [Of course I run Linux]. My current P4 2.8Ghz (space heater that also does some computation) can't handle much when it I run a simple multi-threaded desktop app. The cpu has 1 core and two "hyperthreads". I look forward to 4 cores and 8 hyperthreads--I could use them. I don't see why other people need a desktop and a server on today's hardware.

I generally think that it's better to get a reasonably priced base system on the latest platform, and then you can do a cheap partial upgrade to the high end version of the platform (off eBay) in two years...

Re:Seems pretty clear:

[jon3k]

"Now, you have the equivalent of the $270 E8600 c2d which also rates high in their gaming benchmarks (beating the phenom ii 955, i7 920, and even the i7 940 in their hl2 and crysis warhead benches, and only slightly losing to the 955 and i7 940 in farcry 2)."

Maybe I'm missing something but the links you posted do not show the E8600 beating the 920, 940 or 955, except where it beats a 920 by less than 1 frame per second. All that tells me is that games are not CPU bound.

"71% increase in cost vs. a 2% decrease in performance.

Sure, you could make the argument that future games will likely utilize 4 cores more effectively. But when? I'd wager a guess that we aren't really going to see a significant advantage to quad-core gaming for a few years yet. Just about enough time to plan the next pc upgrade!"

I'm not a gamer, but remember, your computer has a process scheduler that will thread processes to each core. So, as opposed to a synthetic benchmark, in real life, you may be doing several things when playing games. Virus scan, listening to an mp3, downloading a torrent. All of those things would be offloaded to idle cores, without affecting your gaming at all.

With that said, if all you do is play video games, your best bang for your buck is probably an AMD processor followed closely by something like an E8X00, no doubt. Remember, most of the time you're much better off just dumping any extra money into a better graphics card. Of course you'll get a lot more longevity out of upgrading to something like an i7 920 and upgrading the entire platform.

Re:Seems pretty clear:

[jon3k]

Skidpad numbers doen't translate directly into handling. That said, the zr1 is an exceptional value and an amazing automobile.

Re:Seems pretty clear:

[Hatta]

Not embarrassing, just not as fast. However, a Ferrari is vastly more powerful for the money, but a Corvette is still the much better deal -- do I really need to explain this on here?

Yes. In my experience, computer nerds and muscle car freaks are groups that seldom overlap. As far as I'm concerned Ferraris and Corvettes the same thing, impractical expensive cars for those going through a mid life crisis. I have no idea what you are trying to convey with this analogy.

Re:Seems pretty clear:

[ceoyoyo]

In science we call that honesty. You can make a tiny little difference look REALLY BIG by cropping your graph so that it only shows a very tiny range. By showing the origin tiny differences look, well, tiny.

Re:Seems pretty clear:

[steveha]

My issue with the graph is someone needs to take a class on "how to make a graph". 0% performance and $0 cpu.... why? Was there a $0 cpu? Did any of the cpus get a 0%?

Probably because the person who made that graph for The Tech Report wanted all the proportions to be honest.

Did you ever read the book How to Lie with Statistics? Or the book How to Lie with Charts? Or a nice, short blog post called Graphs That Lie?

When you chop out some of the "wasted space" in a graph, you distort the graph. Unless people are careful and check where your axes begin, and then mentally visualize where the axes go, they'll get a misleading idea of the data from the graph.

Suppose the bottom part of the graph was sliced off, at the 90% line, to make you happier. Imagine what it would look like. The AMD X2 6400+, sitting at the 100% line, would have very little white space under it; and the Intel i7-920, sitting a bit below the 200% line, would now appear to be ten times faster than the AMD X2 6400+. The numbers would be the same, but the visual impact would be that the Intel chip totally blows away the AMD chips.

The graph is good the way it is.

I'll meet you halfway, though: it wouldn't have hurt for them to have put in a second chart, zooming in on just the most crowded areas.

steveha

Re:Seems pretty clear:

[skyride]

I have to agree with you here. Just tonight, Im back from sorting out a 30 year old with her new laptop. She picked up for the bargain price of £370. Specs are as follows: Core 2 Duo 2.33GHz (don't recall the model number but it was Penryn based). 3GB RAM 250GB HDD Intel Integrated Graphics 15.6" 16:9 1366x720 Now, people, seriously. That is EXTREMELY cheap for a machine of that spec. Thats actually faster than my own laptop which I am currently typing this on. After buying it she took it straight to me, didn't even open the box. Given her needs, I thought Ubuntu with Hamachi on there for the purposes of remote assistance would better suit her (and given her response, it seems to be a winner). After having her bloatware filled, 4 year old XP system, she constantly kept remarking how fast it was, and the thing is it was bloody fast! It was at the desktop in under 30 seconds, and considering the most heavy app she is likely to run is brasero for copying CD's, theres no way thats needed. As a gamer myself, I have to say that gamers in general are a pretty silly lot. They seem to care about how many FPS they get in crysis or whatever, but the point I always make is that they don't spend 20 hours a week playing crysis, they spend it playing Team Fortress 2 which isn't in the same league in regards to resources used. As a previous poster stated, Team Fortress 2 can run happily on a 5 year old PC. Fast computers are only required to run more bloatware courtesy of the OEMs (not microsoft, the OEMs). Windows may seem slow versus a finely tuned install of Slackware, but just try it running a million and one 90 day trials.

Re:Seems pretty clear:

[skyride]

I thought that the phenoms could do triple channel if you put them on an AM3 board, IIRC Intel has the memory controls on the chip itself and and AMD puts them on the motherboard

Not true, AMD have had memory controllers on the chips since AM2 came out. Intel simply followed in that entirely logical pattern with 1366 socket.

Re:Seems pretty clear:

[citizenr]

The Phenom II X4 955 beats the i7 920 in 3 out of the 4 games they tried.

AMD always leads in games when you compare similarly priced chips ... until you overclock them :(

Re:Seems pretty clear:

[Falconhell]

The most economically sound upgrade path for me has been sell after 6 months, getting 95% of purchase price back-my people like buying my 6 month old systems, and I like buildiing new ones!

Re:Seems pretty clear:

[BikeHelmet]

Notice that we are making some assumptions here that may not be entirely valid. For instance, we've priced the Socket AM3 Phenom II processors on a Socket AM2+ motherboard with DDR2 memory, though we tested most of them with DDR3 memory. As you may have noticed, memory type didn't make much difference at all to the performance of the Phenom II X4 810

and we expect the story will be similar for the rest. In the same vein, we priced the Core 2 processors with DDR2 memory, though we tested them with DDR3. Our goal in selecting these components was to settle on a standard platform for each CPU type with a decent price-performance ratio, not to exactly replicate our sometimes-exotic test systems.

"and we expect the story will be similar for the rest."

You expect that, huh? Based on what? My experience with Intel processors has always been that they're memory starved. Core i7 may be a huge improvement, but please provide me proof. After all, DDR3 is more expensive, Core i7 CPUs are more expensive, and Core i7 boards are more expensive. It adds up to a 50% jump in cost. (~$500 -> ~$750+)

I'd like to know whether that was factored in - but the quote right there makes me think maybe it wasn't?

By their own admission, their results may be skewed in Intel's favour, because they couldn't be bothered benching with the hardware they priced out.

Re:Seems pretty clear:

[TheKidWho]

The point of getting an i7 920 is to overclock it to 3.6-4ghz+... At that point it blows the Phenom II out of the waters.

Re:Seems pretty clear:

[crispin_bollocks]

WTF? There's no turns on a strip - drag or Main. And the chicks have no clue waht a Ferrari is.

Re:Seems pretty clear:

[dr_wheel]

"Maybe I'm missing something but the links you posted do not show the E8600 beating the 920, 940 or 955, except where it beats a 920 by less than 1 frame per second. All that tells me is that games are not CPU bound."

The e8600 beat all of those cpu's in TFA, not the links I posted.

Re:Seems pretty clear:

[Fulcrum of Evil]

Skidpad is only a measure of static lateral grip - slalom speed is probably a better performance metric there. Corvettes are fast and cheap (and feel like it too), but I prefer things like the Lotus Elise.

Re:Seems pretty clear:

[TheKidWho]

Some people want fast computers to do work on... Like CAD/FEA/Video Editing/3D Modelling/Scientific Research/Code Development/etc.

Besides the fact, that while someone might get 120fps in crysis today, they will be barely breaking 30fps when the next game comes out.

Re:Seems pretty clear:

[BZ]

Fast computers are still needed if you're actually using them for work...

Just as an example, compiling+linking a nontrivial program with profile-guided optimization (say Firefox) using MSVC++ takes several hours on decent modern hardware. I'd love for that to happen in, say, seconds instead. ;)

Re:Seems pretty clear:

[funkboy]

The Corvette ZR1 has more horsepower and is less expensive than any current Ferrari.

I love Ferrari... but the Corvette needs no apologies at all!

Especially when used as a donor car for something fast!

http://flyinmiata.com/V8/

Re:Seems pretty clear:

[funkboy]

Not embarrassing, just not as fast. However, a Ferrari is vastly more powerful for the money, but a Corvette is still the much better deal -- do I really need to explain this on here?

Yep, and the "better deal' includes the door chime from a Chevy Cobalt when you leave your key in the ignition with the door open.

Re:Seems pretty clear:

[Anonymous Coward]

Well the Xeon i7s aren't too much more expensive and they support ECC. If ECC is that important you could spend the $20 or so more to get the W3520 instead of the 920.

Re:Seems pretty clear:

[Anonymous Coward]

Meh, that's not long enough to even be new tech in 6 months. Not worth the money you lose. I seriously doubt you're getting 95% for a used system. Then factor in the inconvenience, shipping (if any), plus your labor and it's actually costing you a lot more than you think.

Re:Seems pretty clear:

[Falconhell]

I dontship, I have a large enough friends/family/coworker group to easily find a sale locally.

My labour, meh, 1 hour to build and image the system.

In fact most of the time I get back more than I paid at wholesale level.

Love those XP patches that allow boot in any
new machine!

Re:Seems pretty clear:

[afidel]

I want full detail at 1080P60 which is the native resolution of my LCD. Unfortunately I don't want to pay for either the components or the power to achieve that. I get closer to 1080P30 with an Athlon x2 4200+HE and 9600GSO, my whole system uses about a third the power of just the graphics cards in a top end system today.

Re:Seems pretty clear:

[Anonymous Coward]

My issue with the graph is someone needs to take a class on "how to make a graph".
<snip>
Over half the graph is empty: from 0% to 275%, 0-100% is empty, and 250% to 275% is empty, so nearly half the graph is empty white space. Fail

You fail. This is an appropriate way to make a graph. Starting a graph at 100%, skipping over some "empty" spots etc (which seems like it's what you want to do) is going to do nothing but misrepresent the relative values on the graph. Do you work in marketing?

Re:Seems pretty clear:

[Anonymous Coward]

Maybe in the future... Specifically, 2015. And just think, by that time we'll all have Jitz monitors! ;)

Re:Seems pretty clear:

[Klintus Fang]

The only significant difference between Intel and AMD processors on the memory BW front is that AMD moved to the integrated memory controller two generations before Intel did. If your experience is that Intel processors have always been memory starved, then your experience is limited to the time between when AMD integrated the controller and when Intel integrated the memory controller.

On the question of memory BW it is that simple. The company that integrated the memory controller first was ahead on deliverable memory BW for a while. With i7 now both companies have integrated memory controllers, and that memory BW benefit which AMD had over Intel is now gone.

Re:Seems pretty clear:

[poetmatt]

high end version? do you understand intel's processor path? It's a mandatory upgrade every new socket. Maybe you might want to look into AMD if you want to not waste cash on an i7 920 which is being discontinued, by the way. Mind you, I use an i7 920, but only because I got it free. Had I not, I'd be running AMD.

Re:Seems pretty clear:

[deathguppie]

Here! Here! Mod that man up as insightful.... in small yet functional increments.

Re:Seems pretty clear:

[BikeHelmet]

Oh, I'm sorry - I didn't realize the E6300 and E8400 had integrated memory controllers. ;)

Here I thought they'd get a performance boost from the benchmarks being DDR3!

Re:Seems pretty clear:

[Almahtar]

2 years ago I got an Athlon FX-2 6000+, a motherboard that supports 16 gigs of ram, a GeForce 8600 GTS (512MB), 4 gigs of ram, and a 256 GB SATAII hard drive for about $5-600 total and just threw it in an existing case with other peripherals. At today's prices I can upgrade the ram to 16 gigs and the vid card to something more recent pretty dang cheap.

The CPU is a bit dated but not much of what I do is CPU bound anyway. I run a lot of concurrent VMs that mostly just sit around and play a game here or there.

Just gotta shop around a bit when you buy. When I bought that stuff it was hardly high end - it all cost less than $600.

Re:Seems pretty clear:

[Almahtar]

Amen, brother! Over here in Idaho I get *so* sick of all the corvettes I see. They're EVERYWHERE!

Re:Seems pretty clear:

[beelsebob]

The price/performance graph makes it look like there's a big difference in price between those chips, yeh. In reality, the Q9550 is only $15 more than the X4 955, with a gap that small, I'd put that down more to choice of other components than to "wow, the AMD chip is so much cheeper for the same performance, I'd better run out and get one.

Re:Seems pretty clear:

[Klintus Fang]

True. Those Intel chips do not have integrated memory controllers.

All the chips from both companies potentially get a boost from DDR3 memory over DDR2 memory. But, as the article demonstrates on the other pages, for the benchmarks they are looking at, the sensitivity going from DDR2->DDR3 is small.

I only talked about it in terms of integrated memory controllers because you brought up the idea that Intel parts have always been "memory starved". That statement is what I was contesting. Intel parts have only been memory starved on some workloads, relative to AMD, when AMD had an integrated memory controller, and Intel didn't. The blanket statement that this has always been the case is not true.

As to your point that you want to see data justifying the claims that DDR2 vs DDR3 in their study isn't all that relevant, the article provides that data. One may question the workloads they chose, but the data justifying the claim within the scope of the workloads that they did choose is there.

Re:Seems pretty clear:

[hairyfeet]

Hey, I said for CAD/CAM and other "heavy lifting" that there was a purpose in having the ludicrous speed, I was just pointing out that for a good 80-90% of the home and SMB markets it is simply overkill as their PCs will be sitting idle most of the time due to the incredible processing power of even the lowest model AMD or Intel dual compared to the work they have for it.

BTW it is a damned shame it takes that long, or else I'd be asking you about compiling a 64bit Songbird. We have a 64 bit Firefox which frankly rocks and having a native 64bit Songbird would probably rock for those of us with large databases. Oh and as far as compiling goes? Obligatory XKCD reference.

Re:Seems pretty clear:

[Anonymous Coward]

Well the Xeon i7s aren't too much more expensive and they support ECC. If ECC is that important you could spend the $20 or so more to get the W3520 instead of the 920.

And a non-consumer motherboard, which costs much more. (price is important to some)

Re:Seems pretty clear:

[BikeHelmet]

I only talked about it in terms of integrated memory controllers because you brought up the idea that Intel parts have always been "memory starved". That statement is what I was contesting. Intel parts have only been memory starved on some workloads, relative to AMD, when AMD had an integrated memory controller, and Intel didn't. The blanket statement that this has always been the case is not true.

You should've been fighting my conclusion - they couldn't be bothered benching with the hardware they priced out. I mentioned the i7's improvement in my first post, but I'd still like some numbers on just how big a difference it makes.

As to your point that you want to see data justifying the claims that DDR2 vs DDR3 in their study isn't all that relevant, the article provides that data. One may question the workloads they chose, but the data justifying the claim within the scope of the workloads that they did choose is there.

QX9770(DDR3) vs QX9775(DDR2)?

Mem Latency +60%
Crysis FPS Avg -21%
Crysis FPS Min -32.5% (below 30FPS; expect stutters/jolts with VSYNC enabled)
FarCry2 FPS Avg -35%
FarCry2 FPS Min -34%
UT3 FPS Avg +13%
UT3 FPS Min +4.5%
HL:EP2 FPS Avg -7%

Particle Sim Bench FPS Avg (?) +13.8%

Worldbench - FF + Media Player -8%
Worldbench - Nero CD Authoring -7%

I'm too tired to check the rest. I left out a few that were nearly equal.

My point? If they are going to rate the best value based on certain hardware, they have to bench with that hardware, or their chart is a bunch of hooey.

I will agree though that the AMD processors didn't take any serious hit from DDR2.

the sensitivity going from DDR2->DDR3 is small.

I guess I don't consider 30-35% in a few games to be "small". Crysis and Farcry2 are pretty good indicators for how future games will be affected, so I do feel it should move the CPU around on the value charts.

Re:Seems pretty clear:

[AVee]

The graph is correct in starting at 0, but claming to visualize performance per dollar it stail fails because the axes aren't scaled properly.
Their own remark about the graph is "To oversimplify slightly, the best CPU values tend to be located closer to the top and left edges of the plot."
Yet the plot the 0-250% performance on the same space as the $0-$1000 price. This leaves the i7-920, closer to the top-left corner then the baseline X2-6400+, suggesting that (nearly) twice the performance for (nearly) three times the money is somehow more performance per dollar. It also makes the i7-975 look an acceptable option instead of ten times the price for not even three times the performance.

The scale should have been comparable, twice the price and twice the performance should cover the same distance, not four times as much. It's a pity they failed to get this right because it's a very usefull chart otherwise (appart from all the obvious issues with determining the actual numbers).

Re:Seems pretty clear:

[Anonymous Coward]

Didn't AMD's sockets 939 & 940 have the integrated memory controller first?

Re:Seems pretty clear:

[Sj0]

Isn't it crazy that you're able to get full detail at 1920x1080 anywhere near playable frame rates with a 100 dollar video card?

Re:Seems pretty clear:

[Anonymous Coward]

As did the socket 754...

Re:Seems pretty clear:

[Ginger Unicorn]

i always love mentally picturing what people describe when they use the word "screaming" on slashdot. now imagine all these people are asking for 8 core xeons, and ask yourself would anyone ever want an 8 core xeon to the extent that they would make that kind of noise?

Re:Seems pretty clear:

[mdm-adph]

I have no idea what you are trying to convey with this analogy.

That they're both almost equally fast, yet one costs almost 3 times the other. I tried to pick two very, very common cars.

Re:Seems pretty clear:

[mdm-adph]

Well, the pre-customized racing versions (like the ACR, I think) were always great on the track, if you wanted to spend $100k+.

Re:Seems pretty clear:

[mdm-adph]

Was actually late versions of the original Socket A's, I think!

Re:Seems pretty clear:

[BZ]

Oh, for home users totally agreed. ;)

I don't have a 64-bit windows setup, sadly...

Re:Seems pretty clear:

[mdwh2]

It also means it's harder to see the actual differences.

It's perfectly legitimate to have a cropped graph, so long as you indicate it as such with the broken axis squiggle (I forget the technical name for it) (unfortunately many people don't, which is a problem, but that's another matter).

Re:Seems pretty clear:

[MarkGriz]

"do I really need to explain this on here?"

Please. Do you have a car analogy?

Re:Seems pretty clear:

[ceoyoyo]

It depends what you're doing. It's also perfectly legitimate to say in a TV commercial that your product outperforms all other products in its class, so long as you have the small print to inform the viewer that "outperform" is defined using a weighted set of selected metrics and "class" is defined solely at the discretion of the product producer.

It can still be very misleading, and it's usually considered bad form, particularly when you're showing something for which small differences have little to no impact.

Re:Seems pretty clear:

[mdwh2]

I'm not talking about small print or being misleading, I'm talking about what is a standardised practice for drawing graphs. I don't really mind the white space, but equally, it would have been okay if they'd have got rid of it, with the appropriate symbol for a broken axis.

Re:Seems pretty clear:

[dave420]

Heh. Build quality? Cornering? Styling? Acceleration? Transmission?

Horsepower and price are but tiny parts of the equation. Corvette needs all the apologies it can muster :)

Re:Seems pretty clear:

[ceoyoyo]

Read the original post I replied to. The poster was lambasting the benchmarker for his graphs, suggesting he "needs to take a class on 'how to make a graph'".

It is not wrong to make a graph showing the origin, as the original poster implied. You're also right, in some circumstances it is acceptable to show a graph that does not include the origin. However, almost all of the time in science someone is going to smack you down for doing so, unless there's a really, really good reason for it. Incidentally, that really, really good reason almost always involves error bars.

Re:Seems pretty clear:

[buddyglass]

Okay. Forget the 920 then. My reason not to get excited about the 955: Q9550:

1. Phenom II X4 955 - $245
2. Mobo w/ 4 DIMM slots, 16GB max, DDR3 (e.g. GIGABYTE GA-MA770T-UD3P) - $80
3. Total: $325

1. Q9550 - $230
2. Mobo w/ 4 DIMM slots, 16GB max, DDR3 (e.g. GIGABYTE GA-EP45T-UD3LR) - $90 (after rebate)
3. Total - $320

Performance is approximately equivalent with the AMD chip winning some and the Intel chip winning others. Only, the Intel chip has a TDP of 95W while the AMD has a TDP of 125W. Also, and maybe this is irrational, but I expect the P45/ICH10 to be better supported by "alternative" OSs than the AMD 770/SB710. About the only advantage I see to the 955, unless I really care about a particular application and it happens to perform better, is that the 955 is more overclockable. But since I don't care about overclocking that's not an issue.

Re:Seems pretty clear:

[buddyglass]

My home system, which is six years old, has 1 gig and I don't feel like I need more. Then again, I don't game. That's 1/2 the max supported RAM on my motherboard. If we use that as a guide, then theoretically 8 gig should last you ~6 years since most motherboards support a max of 16. It also lets you buy cheaper 2G DIMMs.

Re:Seems pretty clear:

[hairyfeet]

That's a damned shame. I bet the extra registers and memory would speed the hell out of compiling. Despite all the horror stories I heard about compatibility XP X64 has turned out to be a damned fine OS. I have been running even my ten year old disk catalog software without even a hiccup. The only catch I found was my 8 year old cheapo TV tuner doesn't have 64 bit drivers. Big whoop, the replacement is a whole $50. i built my AMD 7550 with 4Gb of RAM, a 300Gb Sata2, and a nice black case for a whole $281. Figuring in the OS and a HD4650 and DVD burner I got out of there at a hair over $500.

So if you get a chance give it a try. I still think MSFT offers a 90 trial version that lets you see if your hardware is supported. Since my new box supports Phenom II Quad and 32Gb of RAM I figure I can bypass Vista and Win7 if I want and not have to worry about the 32bit RAM limits. It really is a nice OS, which shouldn't have surprised me since it is just Server 2K3 with some extra compatibility on top, but from the horror stories you would have thought it was as bad as Vista!

Re:Seems pretty clear:

[BZ]

Actually, the #1 bottleneck to compiling Firefox in particular on Windows is performance of the OS process and file access APIs... In the case of PGO, the #2 bottleneck is the huge amount of memory the linker ends up touching.

64-bit might help with the latter, maybe. Maybe not.... Have to push more bits around.

Re:Seems pretty clear:

[hairyfeet]

I would think that the 64bit register and other advantages would be a help in compiling. i have noticed in those tasks that are heavily CPU geared, such as video editing/transcoding that the same machine gets the job done quicker in XP64 VS XP32. And looking at process explorer neither was maxing out the RAM so that wasn't the culprit so I have to assume it is having a 64bit wide data path and the extra registers.

So I'd say give it a shot if you got the HDD space to spare. I'm sure you can dig up a copy somewhere to give it a spin and see if it works for you. Unfortunately they have taken down the free trial page at MSFT for a page trying to push 64bit Vista(shudder...or as my customers say EEEEWWW!) but after going XP64 I just can't see myself going back. It just handles too damned nice. And I bet if your compiling software works on it that it'll give it a kick in the pants speed wise. I know it sure did on my video editing.

Re:Seems pretty clear:

[Agripa]

Well the Xeon i7s aren't too much more expensive and they support ECC. If ECC is that important you could spend the $20 or so more to get the W3520 instead of the 920.

How much do the Xeon i7 motherboards cost though? Most of the consumer i7 motherboards and all of the early ones are significantly more expensive than the ECC supporting AM2+ motherboards.

I build a Phenom II 940 system for this very reason. The money I saved on the motherboard, processor, and RAM while still having ECC payed for a fast hardware RAID controller.

Correlation is positive

[JoshuaZ]

The first take away I get is that there is an actual, substantial positive correlation between cost and performance. This is a good thing. If I were in a cynical mood I would have guessed that the correlation would have been small or non-existent. The other thing to note is that there are some CPUs that by this metric are clearly just not very worth it where their are cheaper ones that perform better. So, more expensive generally means better, but not always. So CPUs are sort of like wine?

Re:Correlation is positive

[Anonymous Coward]

You perhaps aren't cynical enough. I wonder how much this chart would change if the chips were overclocked. That would tell you if some of the slower processors were simply underclocked versions of the faster ones.

Re:Correlation is positive

[TinBromide]

yup, just like cars (Once upon a time, you could spend 36k USD for a pt cruiser), houses, internet subscriptions, insurance, meat, items on ebay, and just about anything that has a varied price range and is subject to advertising, brand loyalty, or outright cons.

I've always tended to get right behind the cutting edge, processors usually take a dive in price and speed, Like when i bought my processor, i paid 300 for a quad core intel last summer, the next step up was 395, then 600, then 1000, but the 1000 wasn't 3.3x faster, the 600 wasn't 2x faster, etc. However, the $150 intel chip was 40% slower, so the price/performance sort of follows a log function where the trick is finding the right place before the price starts to skyrocket in relation to performance. I'm sure that if you could find a bunch of old pentium 1 processors on the super cheap, in bulk, you might get the same price/performance as a new mid-level intel chip, but why would you want to run a beowulf cluster to get the same throughput? (Ignoring the price of all of the other components, since the article is price/cost of the cpu alone, that was the metric i went by as well).

I wonder what the price/performance ratio of a pentium 1 that cost $0.01 is.

Well, I suppose that running a beowulf cluster is it's own reward.

Re:Correlation is positive

[Anonymous Coward]

Aside from the possitive correlation, the other main point I would take away is the apparent 'knee' in the chart at about the $300 mark.

Seems that is the point of diminising returns.

Re:Correlation is positive

[suggsjc]

So CPUs are sort of like wine?

I get your underlying point, but wine has more subjective values (wine enthusiasts would debate this though) that come into play whereas CPUs have more quantifiable comparisons. That said, for both CPUs and wine I would think that most of the people who are buying "high end" products they are a little more discerning about their purchases.

Re:Correlation is positive

[Kingrames]

And perhaps you're too cynical.
Judging overclocked chips?

Are we talking about a 32 second test, or a test over several months?

Does the overclocked $250 system catch your house on fire after 6 hours of gaming?

Re:Correlation is positive

[Wrath0fb0b]

The other thing to note is that there are some CPUs that by this metric are clearly just not very worth it where their are cheaper ones that perform better.

Depends on how much you value the time it takes to complete $PROCESSOR_INTENSIVE_TASK. If I hired software engineers for $100,000 a year, that's $1 a minute and so it might very well be "worth it" to spend $1000 on the latest N-core processor (compilation being nearly infinitely parallel) if that save 5 minutes per compile over 200 compiles on the year.

The only thing the plot proves is that there are diminishing returns, not that any particular price/performance point is not "worth it".

Small graph

[Anonymous Coward]

That graph is so squished together I can hardly decipher some of it. There seem to be more dots than labels.

Re:Newt Gingrich: Traitor +1, Informative

[Itninja]

tl;dr Also, WTF is a "vicegerent"?

Only Intel compatible ....

[Alain Williams]

if they want to talk about desktop processors come back to me when they talk about decent choice, what about the ARM ?

I am fed up with all these people who think that all the world is Intel compatible -- when there are better CPUs out there.

Re:Only Intel compatible ....

[Microlith]

They'll cover ARM when someone sells a motherboard with a socket I can stick a 2+ GHz quad-core ARM in and get performance equal to an equivalent AMD or Intel chip.

As it stands, ARM is irrelevant outside the embedded/pda/non-x86 netbook scope.

Re:Only Intel compatible ....

[Rewind]

ARM is largely irrelevant to the regular desktop PC. Not totally, but pretty close. Why exactly do you think an ARM processor would be superior?

Re:Only Intel compatible ....

[Darkness404]

Um, ARM CPUs really aren't high end in the least. Sure, they might be decent enough for web browsing, but tell me whenever I can get an ARM CPU that can render, play games, etc. Until then I'm sticking with x86. All ARM-based CPUs are good for is low power systems like phones, etc.

Re:Only Intel compatible ....

[pankkake]

You mean there are powerful ARM CPUs suited for heavy desktop usage?

Re:Only Intel compatible ....

[Cyberax]

It makes little sense to compare ARM boards. They are usually power-optimized (with peripherals consuming substantial fraction of power), not performance-optimized. Also, end-user devices might be underclocked and undervoltaged. Besides, if you want performance with non-x86 CPUs, then look at MIPS CPUs (which are usually used for things like AV processing in set-top boxes).

Now, a comparison of end-used devices (netbooks/phones) with ARM/MIPS CPUs might actually make sense.

A better solution.

[LWATCDR]

What is the cheapest CPU that can playback 780P flash well?
That is probably a good CPU for 99% of the population. Flash is a resource hog and is likely to be most intensive thing that most people use.
The next step up would be to list several games and see what is the cheapest CPU that can play them at say 60FPS at good settings with a $99 video card.

If your a video editor, hardcore PC gamer, transcode a lot of video, or run CAD get the fastest CPU you can afford.
So hard core types should buy I7s and pretty much everybody else should buy AMDs once you take into account ram and motherboard prices.

Also if you are planing on running virtual machines AMD are often a better choice. Intel doesn't support virtualization on a lot of their CPUs while I think AMD does on their AM2 and up CPUs.

Re:A better solution.

[wjh31]

if all you want is 720P flash, pretty much anything on the market will do. My wifes laptop with a C2D T2080 @ 1.73GHz manages it fine. when comparing it against other desktop processors of similar performance on here: (http://www.cpubenchmark.net/cpu_lookup.php?cpu=Intel+T2080+%40+1.73GHz), and then looking it up on ebuyer.com, suggests that a £40 proccesor is fine for 720p flash.

Re:A better solution.

[LWATCDR]

That is kind of the point. If you want to step it up then go for one that supports 1080p and that will be good enough for most people.
And actually I think the Atom doesn't handle 720p all that well yet.

That is the interesting thing about CPUs today. The best price to performance ratio isn't always important.
If you have no need for the extra performance don't get it. As it is very few applications will use an I7 well. Most people just don't need 8 threads.

Re:A better solution.

[Anonymous Coward]

Flash sucks. Use HTML5.

Re:A better solution.

[jon3k]

You mean 720P flash I assume, or, less likely, 480P. 780P isn't a standard high-def resolution.

But, to answer your question, probably the new class of netbook like the Pegatron, which, interestingly enough is running a Freescale processor with an ARM based core. This little netbook also has flash based GPU acceleration (supposedly), is incredibly thin and sports I think 6 hours of battery life.

Re:A better solution.

[LWATCDR]

Sorry that was a typo. Yes I meant 720P
And yes I think one of those ARM notebooks would be very handy.

Re:A better solution.

[jackharrer]

I'm doing it on IBM T42 with 1.7GHz PIV and Radeon 9700. That's like 4 years old. We really don't need faster processors we have already for general use computing. Games on the other hand...

I use Ubuntu (with Compiz), but I suppose Windows should be also fine, probably even better as drivers are faster.

Re:A better solution.

[jon3k]

"And actually I think the Atom doesn't handle 720p all that well yet."

Incorrect. NVIDIA Ion based nettops are already playing 1080p video smooth as silk. It's more about the graphics adapter than it is the Atom.

Re:A better solution.

[jon3k]

It's a pretty compelling little device considering it's only going to cost $199.

Re:A better solution.

[LWATCDR]

It That is blue ray and not Flash. That is why I specified Flash.

Re:A better solution.

[jon3k]

No where in the post I responded to does the word "flash" even appear. I've got a hp mini 2140 with a 1.6Ghz Atom, 2GB RAM and the standard intel GMA 950 graphics adapter and have no problem whatsoever playing 720p encoded video via the VGA output port, including numerous h.264 encoded videos (which is the same format used by Flash)

"Flash Player 9 Update 3, released on December 3, 2007,[2] also includes support for the H.264 video standard (also known as MPEG-4 part 10, or AVC) which is even more computationally demanding, but offers significantly better quality/bitrate ratio.[3] "

So again, to answer your question - yes, the Intel Atom is easily capable of playing 720p, including Flash based h.264 video.

Re:A better solution.

[Cocodude]

Also if you are planing on running virtual machines AMD are often a better choice. Intel doesn't support virtualization on a lot of their CPUs while I think AMD does on their AM2 and up CPUs.

Not quite true. Quite a lot of Intel chips, including some Pentium 4 even, do include Vanderpool (or VT-x as it's now called). Wikipedia is your friend - http://en.wikipedia.org/wiki/X86_virtualization#Intel_Virtualization_Technology_for_x86_.28Intel_VT-x.29

There are many alternative virtualisation technologies around that don't require CPU support too, such as Xen (paravirtualisation) and VMWare (full emulation).

Re:A better solution.

[LWATCDR]

Well anandtech's tests say differn't
http://www.anandtech.com/weblog/showpost.aspx?i=602
From their posting.
"720p HD Completely unwatchable, CPU utilization > 90%
And that is for an Atom 330 using Nvidia's ION platfrom.
Even over clocking makes it just watchable. Flash even if they use H.264 seems to be slower than just a straight H.264 playback.

And my original post specified Flash 780p. I meant 720P but I was having a bad typo day.

Re:A better solution.

[Creepy]

Hardware Virtualization for both AMD and Intel is hit and miss in general: http://en.wikipedia.org/wiki/X86_virtualization

My laptop has it, but my desktop (with a newer, more powerful CPU) doesn't. Both AMD and Intel base it on marketing numbers and what the CPU is targeted for, but yes, in general AMD supports it on more platforms than Intel. I'm personally using Sun's VirtualBox at home on my desktop machine because it supports software virtualization.

Re:A better solution.

[LWATCDR]

AMD supports it on more CPUs than Intel does. Of the current generation I think only the Semperons don't support it.

tl;dr

[Anonymous Coward]

Honestly, I can't be bothered any more. If I need a CPU I'll just buy whichever one fits my budget & choice of motheboard (which frankly is far more likely to decide my choice of CPU than anything else).

Re:tl;dr

[Endo13]

Honestly, it doesn't matter much any more. Except for certain things like audio and video ripping/encoding, 3D design rendering, etc., most any modern dual-core CPU will do whatever a home user needs just fine. Yes, even for gaming. I'm running a socket AM2 Athlon 5600+ and it's not been a bottleneck yet. My video card could stand to be upgraded for some games though.

So your method of choosing a CPU makes more sense than any other right now.

Even better value with integrated gfx

[sznupi]

AMD would end up with even better value on low-end if the cheap AMD system was built on a mobo with integrated GFX...which is good enough for everything except recent GFX-intensive games.

Yeah, there are always Nvidia chipsets...but for some reason motherboards with them are definitely more expensive (at least where I shop) than comparable ones (both with AMD and Nv chipsets) for AMD platform.

Re:Even better value with integrated gfx

[PitaBred]

Seconded. I have a Radeon HD3200 and an Athlon 7750 in my Mythbuntu media center, and I couldn't be happier. Plays HD video like a dream and the total system including 4GB of RAM and 4 1TB hard drives clocks in at around $600 or so, $700 with the nice Antec case I got for it.

All those cpus are not something i really need

[holywarrior21c]

I spent about $300 to build a whole new system, not $300 for a cpu. I bought about dozen computers in the past year for work. I found those $300 sufficient to build adequate system for regular web browsing/office work with Vista running on those computers.
I think it is more important to compare combination of 3 different parts or the system as a whole. Just by adding large amount of Ram boosts performance. In some cases having SSD as a HDD helps.
In reality, i find that you can easily spend a lot of money to buy "outlier" parts just you want best of what you can get to help work done a bit faster. Few grand in difference doesn't seem like foolish investment when it comes to professional video editing and graphical design. Time really is money for many jobs. Few minutes add up and become hours over a week. That will alter work output by chunks.

Re:All those cpus are not something i really need

[PIBM]

With vista ? There must not be much money left for the hardware!

In my day

[lavacano201014]

We didn't have 26 processors to compare. We had two: The kind they put in computers, and our brains. And let me tell you, our brains overshot the computers.

Ancient CPUs (upgrade comparison)

[Anonymous Coward]

Do any of the CPU reviews use old CPUs? What I what to know is how much faster today's CPU is compared to my 3-6 year old CPU, but these hardware reviews typically have a low end much newer/faster than my current system. Practically, a 50% CPU edge is too marginal for me to upgrade, but if a new system was 3X faster than my current aging machine I would be tempted!

Re:Ancient CPUs (upgrade comparison)

[Colonel Korn]

Do any of the CPU reviews use old CPUs? What I what to know is how much faster today's CPU is compared to my 3-6 year old CPU, but these hardware reviews typically have a low end much newer/faster than my current system. Practically, a 50% CPU edge is too marginal for me to upgrade, but if a new system was 3X faster than my current aging machine I would be tempted!

tomshardware.com

Re:Ancient CPUs (upgrade comparison)

[Anonymous Coward]

Err, tomshardware.com doesn't usually go back that far. One CPU generation typically, sometimes two, so say 2-3 years. Do they do a yearly deep past comparison that I've missed?

Re:Ancient CPUs (upgrade comparison)

[dr_wheel]

Tom's cpu chart is a great tool, but they don't generally compare older chips to newer ones. They also change the testing credentials from time to time, so there's no real way to directly compare old vs. new.

Anandtech has a new cpu benchmark site that compares everything from a single-core atom up to the top-of-the-line core i7. They've also recently added two pentium 4 chips to the mix so you really can directly compare them to the newer stuff. Check it out:

http://www.anandtech.com/bench/

Re:Ancient CPUs (upgrade comparison)

[ginbot462]

Most people just like to bash on Tom's or Anandtech or both. That was actually informative (for people like me that build a computer once in a blue moon). As opposed to me guessing which chart on Tom's he was even referring to. This one maybe?

Flash is a resource hog? Little red riding hood!

[G3ckoG33k]

Flash is a resource hog? Little Red Riding Hood for you!

http://vimeo.com/3514904

How can you call that a hog? It is a blast, on any CPU.

What I'm seeing...

[jd]

The performance scales sub-linearly with the price, and ends up almost flat at the extreme end. This means you need to examine the cost of SMP vs. a more powerful CPU. Two X2 6400+ chips in an SMP should give you about the same performance at the same cost as one i7-920, after you add in the extra for the upgraded chipsets and mobo.

More powerful low-end chips become more and more effective when SMPed versus their higher-end rivals. The other benefit of going SMP is that you have fewer cores sharing the same cache, therefore increase the number of distinct tasks you can perform in parallel effectively without cache-flooding.

Of course, you can't SMP forever - the largest SMP array you can make before the system slows down by more than the CPU increases performance is 16-way. Even before then, you lose linear scaling fairly early on. So you end up balancing the different CPUs against the different methods of arranging them to get the best performance for your money.

Re:What I'm seeing...

[Anonymous Coward]

No shit Sherlock. Please link to a motherboard that supports 2 X2s. Not Opterons, X2s.

What you say? There are none? Guess your whole post just became pointless.

Re:What I'm seeing...

[jon3k]

Dual and Quad socket motherboards are exceptionally expensive and typically require registered/ecc RAM. I would genuinely like to see a setup with comparable CPU features to say, a core i7 920, where you'd get comparable performance for the price using two processors. (no used prices obviously, since we need apples v apples).

Re:What I'm seeing...

[Anonymous Coward]

Yep, this would have all been true if not for Intel and AMD's unfortunate market segmentation. The original Pentium Pro could support 4 CPUs, the Pentium II and III only supported 2 CPUs, with Pentium II and III Xeons being required for 4 or more CPUs. Intel goes a step further with the Pentium 4, requiring Xeon DPs for even 2 and Xeon MPs for 4 or more. Eventually Xeon DP was renamed the Xeon 5000 series and Xeon MP was renamed the Xeon 7000 series when Intel adopted processor numbering for Xeons, with a new 3000 series for single CPU only Xeons. AMD followed when they adopted SMP with the K7-based Athlon MP for dual CPU, and later the K8-based Opteron 1xx for only one CPU, Opteron 2xx for two, Opteron 8xx for four or more.

Re:What I'm seeing...

[Anonymous Coward]

Quad cores are a *LOT* cheaper than ANY pair of dual cores:

Once you buy the $500 Supermicro dual socket board and a pair of decent Xeons or Opterons (like $200 or so for a basic dual core version)... It makes a $300 i7 920 look dirt cheap all of a sudden! And that's when you don't need a bigger case for a EATX board or even a server PSU! There's many nice quad cores under $200 too (there are some quads under $100 even)

Next payday I'm buying a Phenom II X4 940. $190 total expense, drop-in replacement for my existing CPU. The i7 920 is a little faster on average (10% or so), but it's $300 for the CPU, like $200+ for a decent motherboard, and I'd have to replace my perfectly good 8GB of DDR2 RAM for DDR3 too (more $), as in over 3x the price for something barely faster I wouldn't even notice.

Tech Report Slashdotted

[Xsydon]

Looks like Tech Report has been slashdotted.

Re:Tech Report Slashdotted

[Anonymous Coward]

Just curious. How does anything actually get slashdotted when no one actually RTFAs. Or maybe I'm just new around here and don't yet fully understand the term.

Re:Tech Report Slashdotted

[twidarkling]

easy, people click, then get scared of all the words, and run away. But they still access the page.

Too bad

[Sycraft-fu]

Because desktop PCs of the world DO run Intel compatible chips, and that is just how it is. Doesn't matter if you don't think it should be that way, that's how it is and that's what tech sites deal with. Windows is of course the by far dominant desktop OS with over 90% market share. Well Windows only runs on Intel and AMD chips. Most versions are x86 and x64 only, there are a few IA64 (Itanium) versions. No ARM, no Power, etc.

After Windows is MacOS. All new Macs sold for a number of years have been Intel chips only. Until recently they did provide support for their older PPC desktops, but no more. The latest MacOS is Intel only.

After those two, is Linux. Now Linux can, in theory, run on any CPU. Search hard enough and you'll find distros for just about anything. However we are talking desktop PCs. So who's the biggest there? Ubuntu. It's primary platforms? x64 and x64. There technically are PPC and SPARC versions, but no ARM. Also the SPARC and PPC support are very unofficial. PPC support was for PPC Macs, and officially ended at version 6.10.

Now all the OSes aside, there is the simple matter of hardware, which is what this site was talking about. Where do you go to buy an ARM desktop? I am not aware of any vendors selling them. So you can't very well do a price/performance comparison for a desktop that doesn't exist.

So please, less whining about irrelevant BS. If you like ARM that's wonderful, and ARM chips are used in many devices. ARM is extremely popular in the mobile arena, for example. However desktop computers are x86 and x64. That's just life. Don't get pissey with hardware sites when they deal with that reality. They aren't in business to push your viewpoint, they are in business to help people make buying decisions.

Re:Too bad

[sznupi]

Well, it seems somebody makes ARM PCs...
http://en.wikipedia.org/wiki/A9home
http://www.advantage6.com/products/A9home.html
http://www.cjemicros.co.uk/ - just 600 Pounds! ;>

But yeah, I agree with you.

Though I wonder what upcoming ARM netbooks will bring; with existing official Debian ARM port, be might even see the one true desktop Linux distro that you mention, Ubuntu...

/.'ed

[dr_wheel]

Apparently, the tech report should have benchmarked their web server before putting this article up.

Re:/.'ed

[Xsydon]

Looks like they use AMD. ;)

Re:/.'ed

[StormReaver]

> Apparently, the tech report should have benchmarked their web server before putting this article up.

It's probably more bandwidth related than computer related. When the County first put our sex offender registry online, we had a measly T1 which got saturated by the community within minutes of the local news stations carrying the announcement. Our (low-power at the time) Linux web server registered 0.00 0.00 0.00 loads most of time time. If it wasn't for the server logs expanding rapidly, I wouldn't have thought anything was happening.

The site got slashdotted.

[CEHT]

4:27pm EDT, the site is broken because this is being /.-ed. Power of social media!

Will they tell us what CPUs their DB are running?

[Lorens]

The site seems to be working with an empty database after being slashdotted.

Article doesn't really talk about features

[vux984]

Its too bad the article doesn't talk about things like Execute Disable, Virtualization support, etc. For a power user audience like /. these are important considerations.

For me not being able to install Xen, or Windows 7 XP mode, etc are complete deal killers. I want CPUs with those features, especially when shopping "value CPUs".

Getting something like an E8190 is a mistake that will bite a /. power user in the ass eventually even if it is a few bucks cheaper than an E8200 and delivers the same performance, at the same wattage, etc...

Re:Article doesn't really talk about features

[memnoch37]

Its too bad the article doesn't talk about things like Execute Disable, Virtualization support, etc. For a power user audience like /. these are important considerations.

For me not being able to install Xen, or Windows 7 XP mode, etc are complete deal killers. I want CPUs with those features, especially when shopping "value CPUs".

Getting something like an E8190 is a mistake that will bite a /. power user in the ass eventually even if it is a few bucks cheaper than an E8200 and delivers the same performance, at the same wattage, etc...

I completely agree. I recently put together a new pc for myself and I almost went with the recommended CPU from this article, the Q8400. Then I realized Win7's XP mode won't even run without hardware virtualization, which that chip lacks. The Q9400 is only slightly more expensive, has a larger cache, and has hardware vt, making it a no-brainer for me.

Re:Article doesn't really talk about features

[Anonymous Coward]

My work did the same thing. We had the E8400, which features VT, so logically the Q8400 would have it too? Bought 5 or 6 of them before we actually tried to run a 64 bit VM. So now we spend a bit more and buy the Q9400.

Re:Article doesn't really talk about features

[Anonymous Coward]

Go with AMD. They don't try to artificially segment their market by disabling CPU features on random processors like Intel does.

i7 920 - the most cost effective?

[CEHT]

Haven't finish reading the report before the site gone under, but the i7 920 seems to be the most cost-effective according to their statistics.

Intel Atom 330

[Eil]

I've been something of an AMD fanboy ever since the Athlon came out, but I just bought an Intel Atom 330 for a lightweight file server, and I have to say I'm thoroughly impressed. 64-bit, dual-core, virtualization extensions, and low-power to boot for around $80 which includes the motherboard. Simply unbeatable.

Also wanted to mention that these guys have easy-to-read benchmark charts of a wide variety of CPUs. Certainly more than the 26 in TFA. Benchmarks don't tell the whole story of course, but it's a good start for quick-and-dirty comparison.

Re:Intel Atom 330

[Anonymous Coward]

You could have bought an AMD X2 4850 plus a 780G motherboard for $80-90 total and gotten much better performance at load, with comparable power draw at idle / standby. You haven't done your homework.

Re:Intel Atom 330

[Anonymous Coward]

I just bought an Intel Atom 330 for a lightweight file server, and I have to say I'm thoroughly impressed. 64-bit, dual-core, virtualization extensions, and low-power to boot for around $80 which includes the motherboard. Simply unbeatable.

I'm not sure I agree. Fry's regularly advertises low power $60-80 AMD CPU and motherboard combinations that have much better performance than the in-order Intel Atom processor. I have one of their cheap Sempron systems that idles at 35 watts.

By the way, the Atom 330 doesn't have virtualization extensions. (athlons do)

I just priced out two mid-level gaming systems

[mandark1967]

$1025.93

All Parts from Newegg.com

$52.99 HDD -- N82E16822148262 - (Seagate 500GB SATA3G)

$159.99 Motherboard - N82E16813188043 - EVGA 780i

$219.99 CPU - N82E16819115131 - Q9400 C2Quad 2.66Ghz

$47.98 RAM - N82E16820104072 - Kingston HyperX 4Gb (2x2Gb)

$139.99 - PSU - N82E16817139009 - Corsair 850W

$99.99 - Case - N82E16811129021 - Antec 900

$189.99 - GTX260 - N82E16814130372 - EVGA 896-P3-1262

$25.99 - DVD+-RW - N82E168271361532 - LG DVD+-RW w/ Lightscribe

The other was an AMD-based system that costed about the same.

Both give similar performance and both will suit their owners needs just fine.

Screw a whole bunch of $1000 CPUs. I'll take $1000 systems anyday.

Re:I just priced out two mid-level gaming systems

[Spatial]

Criticism: they could've been $900 systems if you dropped the quads and got 500W Corsair PSUs.

The computers are only going to use around 350W at max load. Games don't utilise more than two cores (with some rare exceptions) so you could've got better real-world performance for less money with a dual core CPU. The E8400 for example.

Re:I just priced out two mid-level gaming systems

[Anonymous Coward]

$1000 for mid level ? WTF are you smoking, and will you share with the rest of us?

For me, $400 is a decent gaming system, and $1000 is a super-high-end uber-elite gaming rig.

Re:I just priced out two mid-level gaming systems

[psycho12345]

Mid-level? Are you kidding me? CPU is overkill, and in no way you need more then maybe a 500W PSU with single GPU. You could roll that a lot cheaper.

Re:I just priced out two mid-level gaming systems

[drizek]

You are overpaying for the case. You can get a good Antec case for about $60 during one of their very frequent sales. Sometimes they come with the excellent Earthwatts PSUs, save another 140 bucks. Video card is overkill too, you can get a very good card for a hundred bucks. HDD is excellent choice, as is DVD burner. A good AMD motherboard would save you quite a lot of money, same on the CPU, a hundred dollars combined.

Basically you can get a similarly capable system for about 7 or 8 hundred if not less.

Figures

[Anonymous Coward]

Surprisingly enough a site comparing Procs is down once posted on /.

GBA, DS, and DSi have ARM CPUs

[tepples]

tell me whenever I can get an ARM CPU that can render, play games, etc.

Games? Nintendo has put ARM CPUs in its handhelds since mid-2001.

Windows Mobile for subnotebooks

[tepples]

Windows only runs on Intel and AMD chips.

Unless it's Windows CE. I seem to remember that some early examples of what everyone but Psion now calls "netbooks" ran Windows CE.

Where do you go to buy an ARM desktop? I am not aware of any vendors selling them.

There used to be desktop computers that run RISC OS, but I get your point.

Re:Windows Mobile for subnotebooks

[drinkypoo]

Windows CE is ballocks, it's barely acceptable for a smartphone or PDA. I have a tablet with it (WebDT 360) and it's a nightmare trying to use it as a real OS. (I'm working on building Angstrom Linux, have got a build, now need a customized kernel and X server. Fun times.)

Re:Newt Gingrich: Traitor +1, Informative

[The Archon V2.0]

Also, WTF is a "vicegerent"?

It's the guy who becomes gerent after the incumbent gerent is shot.

All kidding aside, "gerent" and "vicegerent" are words, no matter how much it looks like someone typoed "regent":
http://www.merriam-webster.com/dictionary/gerent
http://dictionary.reference.com/browse/gerent
http://www.merriam-webster.com/dictionary/vicegerent
http://dictionary.reference.com/browse/vicegerent

Too bad the ATI Driver Sucks Balls

[DG]

I recently built a similar system - Radeon HD3200 integrated graphics, Phenom II Quad 940, Ubuntu Jaunty. Box absolutely screams (it replaced an Athlon 3200+) and I'm very happy with the Ubuntu user experience (recent convert from Fedora)

But Lord of Guns and Butter, the ATI 3D drivers SUCK. Crash crash crash, where the NVidia drivers Just Work.

In retrospect, I wish I had skipped the integrated graphics and just bought an NVidia card.

DG

Re:Too bad the ATI Driver Sucks Balls

[PitaBred]

Yeah. I'm not worrying about 3D right now in it... I'm using the open-source radeon driver that ships with 9.04 because it has tear-free xv in it. The fglrx/catalyst drivers suck donkey cock at doing video. If you're just looking for a media center or non-3D stuff, I highly recommend sticking with the open source drivers.

On the plus side, ATI is devoting some serious resources to open-source drivers. Take a peek over at the Phoronix forums for some ATI devs posting the latest and greatest, answering questions about what direction 3D is going, and so on. Good stuff, IMHO. I paid for a 4670 since I wanted to support a company that's doing open-source 3D better than Intel.

Performance/$ vs. CPU chart

[Anonymous Coward]

What would be more interesting is a Performance/$ vs. CPU plot.

Upgrading a component at a time

[msobkow]

I just ordered a new motherboard that I think is compatible with my old P4 570 CPU. (It's 800MHz FSB, so it should work, I hope!)

The one thing I've learned over the years is to buy a good motherboard that lists your current CPU as one of the oldest ones it supports -- that way you've got plenty of room for future upgrades if you need them. The simple truth is that if I end up upgrading my CPU in the near future, it'll be because the new mobo isn't compatible, not because I need more speed.

Re:Upgrading a component at a time

[hairyfeet]

That's what I did with myself and my customers. For the office guys I maxed out the boards at 4Gb, because they all have legacy apps and can't stand Vista. For myself and the home users i got boards that would take at the minimum 4Gb, most do 8Gb and mine does 32Gb ;-)

But all support the latest quads. Mine can take any Phenom 2 Quad that is 95 watts, and honestly? I doubt i will ever get to the point i actually need quad. Most of the games I play are like Bioshock or Sacred Gold and simply don't need the mega horses of a quad core. My cheapo $50 HD4650 decodes DivX and most of the other codecs and plays my games smooth. So why blow cash on a bigger epeen? By sticking with the AMD 7550 my PC is so quiet I have actually accidentally turned it off before because I walked into the room and didn't think it was on. It does what I want it to do and does it fast, and with XP X64 I can sit out Vista and even Win7 if I desire and not have to worry about the 4Gb RAM limit.

As someone who has worked PC repair and retail since the days of Win3.xx I can say that in the past there WAS a reason to keep up with the latest CPUs. In a five year period I went from a P100 to a P2-233 to a P3-600 to a P3-1100MHz. In each case the jump was dramatic and noticeable. But working with customers who have to have the biggest epeen and getting to spend time with their uber quads I can honestly say for the things I do I wouldn't have noticed a difference. I have worked with just about every Intel and AMD chip and pretty much ALL the dual cores and above are "ludicrous speed". So unless you are buying a Netbook (which is like shopping for P3s) I just haven't seen anything to make the extra cash layout for an uber CPU worthwhile.

It is better IMHO to spend that cash on RAM and storage space and GPU. Those will make a noticeable difference in the customers experience, whereas pretty much any modern dual core and my customers are going "Oooohhhhh....It's just so fast!" which is why they keep sending their family and friends to me ;-)

Re:Upgrading a component at a time

[mdm-adph]

Wish I could mod this up. That's the only component I never skimp on -- the motherboard. It's trivial to add a new CPU in the future if you need more speed -- you don't even have to reinstall Windows.

Try doing that (reliably) with a new motherboard.

Re:Upgrading a component at a time

[WuphonsReach]

My rules of thumb:

- For a basic office machine, 2 or 4 cores is plenty, individual core speed doesn't matter much. It's usually whatever CPU gives me the most cores in the lowest power factor and costs less then $100. The 45W CPUs are my favorite. Given RAM prices now, the choice between 2GB or 4GB is a coin flip for XP installs. (I'll generally go with 4GB to avoid an upgrade down the road.)

- For a game box, the speed of an individual core does matter. So I'll spend more on the CPU to get a faster dual, triple, or quad core. Quad is a bit overkill, but sometimes dual isn't enough. But I still won't pay more then about $300 for a single CPU, and preferably closer to $200-$240.

I really like low power CPUs for office machines. It means quiet operation with low fan speeds. We may even switch over to 2.5" laptop hard drives soon to shave another 8W or so off of the power usage. Maybe... (the cost to mount a 2.5" drive inside a box designed for 3.5" drives tends to negate the power savings).

When I rebuild my game box next year (probably using Win7), I figure 8GB or 16GB with quad-core will be the minimums.

Re:Upgrading a component at a time

[hairyfeet]

Why not just go with the WD green drives? I have used those in the office builds where a customer is going to be working in a quiet office and the temp/noise levels are much better IMHO than a standard drive while not being nearly as expensive as the 2.5s. From the sounds of it the WD green is just what you are looking for. This way your users have the power when they need it, and have the quiet and the power savings when they don't.

If it were me and I could get the company to sign off I would get a couple of kill a watt meters and place one on a standard drive equipped and one on the WD green equipped PC. Have them both in the same department so the loads should be identical and then compare the power savings and the noise from the users perspective after a week. This should let you know if the switch makes sense from a dollar standpoint, but I bet from a power savings the green drives would probably pay for themselves if you have a large group of PCs and intend on keeping them for any length of time. After all, they are gonna be using their PCs one way or another, why not save the company the electricity if it makes financial sense to do so? With the green drives the lower price should make up for the extra power they might use, and as I said this gives your users the ability to have the extra speed when required. Seems like the best of both worlds to me.

Re:Prediction

[Anonymous Coward]

Ah a trio of concurrent pathetic weenie assholes.

Oh and personally, I think AMD are crap. I have had more wierd problems with 10 AMD systems I have worked on than all of the thousands of Intel systems. They are rubbish!

Re:Prediction

[therufus]

Fair enough. You're entitled to your opinion. I don't agree though. We build about 15 AMD based PC's to every intel. I am slightly biased to AMD, as I don't believe in propping up marketing companies that make processors as a side project.

That being said, I cannot blame any issues with PC's that we have had on any platform EXCEPT for one intel system. I ordered in a high end Gigabyte mainboard for an E8500 system I was building and was matching it with 8Gb of DDR2. Thing is, the particular chipset (which escapes me now) couldn't boot with 8Gb of RAM, even though the mainboard manufacturer claimed the board would. After much research, I discovered that the chipset was simply not up to current standards. This was only 2 months ago.

I switched the customer over to a Phenom X4 810 system and it works fine with all 8Gb.

What about 1080i?

[Anonymous Coward]

In my experience with making budget PCs to run mythtv from ATSC broadcast, the biggest load is for software deinterlacing due to the number of bizarre formats that get broadcast like 1080i or even the 720i and 480i I see to encounter periodically. It requires CPU horsepower because of all the odd cropping, deinterlacing, and scaling that needs to go on due to interlaced source material being embedded into strangely letterboxed broadcast frames.

Re:Flash is a resource hog? Little red riding hood

[Anonymous Coward]

Not on my 1GHz Via Nehemiah processor.

Calculate cost on a per year basis including Watts

[turing_m]

I upgrade on the order of every four years.

I've got an AMD Athlon X2 4850e. The time before that I lasted 6 years. So far I'm having a hard time seeing the value proposition in replacing the 4850e in the foreseeable future. There is still nothing better that I can see is the best combination of performance, low power consumption and price.

In fact, it would make more sense to calculate a cost per year to own rather than the outright purchase cost, since this is what it really costs a person. Probably most people on slashdot have their computer on half the time, idling most of the time. A simple $1/year/Watt*0.5 (the 0.5 is for the time on) and straightline depreciation down to say, zero after 4 years, would be a good way to calculate it. So if the CPU is $100, the power usage is 45W at idle (or whatever it is supposed to be, I can't remember and can't be bothered googling properly), then the cost per year is $100/4 + $45 * 0.5 = $25 + $11.25 = $36.25/year. Power is 1/3 the cost, even with a power frugal CPU. For a lot of other CPUs this percentage is only going to increase, until you get to the very high end where the purchase price is so large that it dwarfs the power consumption.

There will be one blindingly obvious upgrade I will make within 4 years, and that will be an SSD.

Re:Newt Gingrich: Traitor +1, Informative

[jonbryce]

At my school it was what we called the deputy headmaster.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Wow, people still do comparisons like this?

[Junks Jerzey]

My main desktop PC is a single-core 3GHz Pentium 4 from 2004. Yes, it's over 5 years old. And with that purchase my worries about performance and chasing the high-end came to an end. I haven't played a 3D PC game since DOOM 3 (which ran beautifully on my system when the game first came out), so I don't really know much about that end of things, but I haven't even had vague worries about performance. I process raw digital photos, program in a lot of interpreted, high-level languages, put together complex documents in vector drawing and page layout packages. Was there a time when PCs were slow? :) I've also got a dual core Mac Book, which is heading toward three years old. I have yet to do anything to make it break a sweat. As far as I'm concerned, I have infinite computing power in front of me.

The most economically sound way is to not upgrade

[gosand]

I am still amazed by the progress made with systems, I just cannot keep up with it.

My first PC in 1990 was a 386DX-33 w/2 MB RAM and an 80MB HD. I paid $2200 for it.

I am definitely on the slow end of the upgrade cycle. I mean embarassingly slow. I am running a Duron 1.3 with 768 MB RAM. I just upgraded my video card to a Radeon 9600 AGP that I got for free. I am outside any upgrade curve, I have to ditch it and go with a whole new system - which is why I haven't done it yet. My 4 year old daughter has a faster computer, it's an AMD Sempron system that someone had set out in their bulk trash. I grabbed it thinking I could use some parts, and it was a working system.

I run Kubuntu, and my system is still OK. I wish it could be a bit faster at times, but overall it does what I need it to do. I *have* found that it does run slower over the past year for certain websites... e.g. Slashdot takes a long time to load, and navigates a lot slower than it used to... and my CPU pegs. It didn't used to do that.

Why don't the testers get, that...

[Hurricane78]

...A Core i7 never is a good choice?

They argue, that it gives you the best performance, and if you need it, money does not matter.

But what they completely conceal, is that for that price, you can get up to four other CPUs! With will give you way more power than those 230%. (Roughly around 350%-450%.) For the same price.

What we need in the consumer area, are those server boards, that take 4 CPUs (or more?). Then nobody cares for top speed per die anymore.

Re:The most economically sound way is to not upgra

[LordVader717]

Absolutely correct. What many people forget, but is the most important purchasing decision is "what do I need?". The majority of PC users use their systems almost exclusively for browsing or other low-performance applications.
Others may have specific needs that can be met with certain hardware, but where more doesn't make it any better, like for playing HD Movies or another specific application/game.

One fundamental problem is that laymen won'r know how to adress their specific needs, so a lot of people buy whole new systems just because their hard drive's fragmented or full up, or their DVD drive is broken.

Who cares?

[jotaeleemeese]

In all honestly, who actually gives a hoot about this?

My strategy for the last 10 years has been to spend £500 to replace my desktop.

I go out there, get the best all around deal (last time it was an LCD screen, wireless mouse, more memory, bigger disk) and I don't even check the CPU details.

The CPU is going to be faster and cheaper, and for a desktop, where the most demanding task could be a game (which I don't play in my desktop anyway), anything currently in the market will be faster and better value for money than my current machine.

Home users should look at their wallet rather than at CPU cycles...