Slashdot Mirror
AMD Launches Fastest Phenom Yet, Phenom II X4 980
MojoKid writes "Although much of the buzz lately has revolved around AMD's upcoming Llano and Bulldozer-based APUs, AMD isn't done pushing the envelope with their existing processor designs. Over the last few months AMD has continued to ramp up frequencies on their current bread-and-butter Phenom II processor line-up to the point where they're now flirting with the 4GHz mark. The Phenom II X4 980 Black Edition marks the release of AMD's highest clocked processor yet. The new quad-core Phenom II X4 980 Black Edition's default clock on all four of its cores is 3.7GHz. Like previous Deneb-based Phenom II processors, the X4 980 BE sports a total of 512K of L1 cache with 2MB of L2 cache, and 6MB of shared L3 cache. Performance-wise, for under $200, the processor holds up pretty well versus others in its class and it's an easy upgrade for AM2+ and AM3 socket systems."
I just bought a 6-core AMD chip a week ago. Where is the x6 version of this baby?
Censorship is obscene. Patriotism is bigotry. Faith is a vice. Slashdot 2.0 sucks.
I'll be waiting for the dust to clear with Bulldozer before I make a commitment for my next build. No reason to buy a $200 Phenom II X4 980 now when there is no application that needs that much power. If you buy a Sandy Bridge or a higher-end AM3 board/processor now, your average gamer or office worker won't be able to max it out for years -- unless he does video editing or extensive photo shop or if he has to get his DVD rips down to a 10 minute rip vs a 15 minute rip per feature film...
Might as well wait for the dust to clear or for prices to fall.
Anyone notice that weird benchmarks TFA uses for the gaming performance evaluation? TFA compares several processors against the X4 980 by running a pair of games at low quality with minimal quality to "isolate out the graphics card:"
we drop the resolution to 800x600, and reduce all of the in-game graphical options to their minimum values to isolate CPU and memory performance as much as possible. However, the in-game effects, which control the level of detail for the games' physics engines and particle systems, are left at their maximum values, since these actually do place some load on the CPU rather than GPU.
I find it hard to believe that the guys at "hothardware.com" know enough about 3d game architecture to have any understanding of what places a load on the CPU and what places a load on the GPU. Anyone have any thoughts?
the toads are wireless! i repeat, the toads are wireless!
It's pretty well known that most games max out the GPU at higher resolutions.
But is it also well known that increasing resolutions and knocking off all the effects dose nothing to the CPU performance?
This [arstechnica.com] thread has some interesting information on possible BD performance.
.....
This is 301 posts with back and forth that looks basically to be speculation. Prove me wrong by quoting specific statements of those that have benched the [unreleased] bulldozer. Because otherwise, this link is basically a bunch of AMD fanboys fighting against Intel fanboys. But prove me wrong...
AFAICT, the thinking is that running a resource intensive game at GPU-indifferent settings will allow you to focus on CPU usage.
(fwiw i'm not into benchmarking/overclocking, but I think this is how the benchmark is to be interpreted - which I think is a fairly standard way of doing things for a CPU)
10 years ago it was confusing enough with what could be seen as reliable product APUs fro AMD with 1.4 ghz here, and 1.23 ghz there, name changes to meaningless marketing numbers and names. So, I'll stay ignorant and simply ignore these 'breakthrough' numbers and buy product instead of specifications.
It's not uncommon. Other benchmarking sites do it too.
Visit the
Once the GPU is maxed-out, there's nothing more for the CPU to do. If you're running at 30 FPS at high-res, the CPU might be at 30%. At that point, any number of different CPUs will have identical benchmark results. When you drop the load off the GPU, the CPU hits 100% usage and you can compare 150 fps to 160 fps, for example. This is a very simple and typical way to benchmark CPUs for gaming perf. Reviews and reviewers (such as myself) have been doing this for 10+ years, since the very first 3D accelerators came to the gaming market.
Bah.
My ten-year-old CPU does 3100 megahertz. Things have slowed dramatically since the the 80s and 90s, when speeds increased from approximately 8 megahertz (1980) to 50 megahertz (1990) to 2000 megahertz (2000). If the scaling had continued, we would be at ~20,000 by now. Oh well. So much for Moore's Observation.
My AC stalker: " I personally agree with your posts most of the time, but that won't keep me from modding you troll"
For testing CPU in a video game, it's traditional to generally run the benchmark at a low resolution in order to help ensure that the CPU is the bottleneck and not the graphics card. Compared to the processor, more strain is placed upon the graphics card/gpu as the resolution is increased.
It has been this way for a very, very long time.
Is anyone else disappointed that AMD's fastest desktop processor can barely keep up with Intel's midrange Sandy Bridge Core i5 processors in most applications? Sure, AMD's processors are still a great value, but it seems like they fall further behind with their performance parts every year.
I just hope that the performance improvements for Bulldozer are all they're cracked up to be.
That seems like a stupid way to benchmark. It encourages people to be misinformed by thinking that they can get better frame rates by buying a faster CPU even though under real world conditions the game will be GPU bound and the CPU is irrelevant. Why not stick to benchmarking using applications that are actually CPU bound under normal usage?
Who cares? All you're concerned about is removing the video card as a factor in the benchmark. It gives you a baseline from which to compare one processor to another, even if the absolute performance values themselves are otherwise meaningless. That's the entire concept of a synthetic benchmark.
Ummm, against what, my obsolete Phenom (I) X4 9850? Funny how true fanbois can read the same review as an objective person and walk away with entirely different conclusions, eh?
The AnandTech review was even less forgiving of AMD's underdog status, and basically recommended passing and either waiting for the allegedly awesome new Bulldog line or jumping ship for Intel. Hell, when Sandy Bridge both outperforms AND underconsumes (power), you oughtta be seriously questioning that underdog affection. I certainly am.
Read this excerpt from an AMD management blog:
"Thanks to Damon at AMD for this link to a blog from AMD's Godfrey Cheng.
We are no longer chasing the Phantom x86 Bottleneck. Our goal is to provide good headroom for video and graphics workloads, and to this effect “Llano” is designed to be successful. To be clear, AMD continues to invest in x86 performance. With our “Bulldozer” core and in future Bulldozer-based products, we are designing for faster and more efficient x86 performance; however, AMD is seeking to deliver a balance of graphics, video, compute and x86 capabilities and we are confident our APUs provide the best recipe for the great majority of consumers. "
People, read between the lines.
What he is saying is that they can no longer compete with Intel on speed and have decided to concentrate on a balance at the low end priced points.
The days of the cpu wars are in fact over and Intel has won with Sandy Bridge.
Yes, I have been an AMD only fan for years but you have to face the reality that times have changed permanently in Intels favor and AMD's days are numbered.
Why else do you think Bulldozer is over a year late!
Oh, and AMD is prime for a buyout right now and there are rumors.
When AMD fails Intel will have a monopoly and the consumer will loose in the end.
Sad but true.
Not stupid at all. It shows that if your video is not a factor or you upgrade to an adequate video card when one is available,the better cpu to buy is X.
This is common practice that has been used for at least a decade now.
furthermore - its a SYNTHETIC BENCHMARK. no one wants to play crysis at 800x600, but its a tool we can use to measure cpu performance. no one wants to buy a PC soley to sit and calculate prime numbers all day either, but stressprime and other stuff is used for benchmarking also.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
That seems like a stupid way to benchmark.
Why would you use a GPU-limited benchmark when comparing performance of different CPUs? That would be retarded.
Not stupid at all. It shows that if your video is not a factor or you upgrade to an adequate video card when one is available,the better cpu to buy is X.
Assuming that GPUs are available that are so fast they move the bottleneck back to the CPU, the way to do it would then be to use one of those GPUs and use normal resolutions. If even the fastest GPUs still result in the GPU as the bottleneck, just find a different benchmark. There is no lack of synthetic benchmarks that someone could use without misleading people into thinking they need a super fast CPU for a heavily GPU-bound game.
This is especially true now that integrated GPUs are becoming respectable. Someone on a budget may be very interested to know whether AMD CPU + integrated AMD GPU is faster than Intel CPU + integrated Intel GPU, but you don't get an accurate picture if you skew the settings to make it CPU bound.
From the post of "AMD Launches Fastest Phenom Yet, Phenom II X4 980",I know that the garden pop up gazebo is really a good equipment for us to own one, the same with the party tent, for it can bring us much more convenience.
That's the point. Find the thing people actually do which is CPU-bound, don't just fudge a GPU-limited thing until you get different numbers for different CPUs.
They released their fastest processor? Wow-- unusual for chip makers to make improvements to speed and design.
Show me Bulldozer and then we'll talk.
I told you i made you famous! http://zepski.net/joequote.png
That's the point. Find the thing people actually do which is CPU-bound, don't just fudge a GPU-limited thing until you get different numbers for different CPUs.
Nothing that the average person does is CPU-bound if they have a fast CPU; most of the time it will be idling. The closest they'll get is gaming when not GPU-bound, which they may not do today, but they will when they replace their GPU in two years.
Ultimately if you want the fastest CPU then you want to run things that are CPU-bound. If you just play crappy console game ports and run them so they're GPU-bound then you'll do fine with a dual-core in most cases.
It has been standard practice for the 10+ years I've looked at benchmarks. And it works.
e.g. Using the same CPU, the differences between two video cards get smaller as you drop the resolution.
Generally true; handbrake is the only app I routinely use that maxes out all 8 cores on my box.
Well yes, if such GPUs were available today, sure. They're not. However the gaming benchmark is not useless, because its a real-world mix of code in a typical app the CPU might be used to run. You're looking at the benchmarks expecting them to be some absolute result. They're not. You have to use your head and interpret the results, as with any experiement. A benchmark isn't a "you need to buy this cpu for this game" statement. Its a performance indication on a particular code path.
Synthetic benchmarks might give you a number to compare CPUs with, but if they're not doing real world tasks (or even better, the exact tasks you intend to use the box for, such as running a game if you're a gamer) they are easily cheated on - the cpu vendor can simply spend transistors on optimizing for the instructions most commonly used on the benchmark.
In summary: without thoughtful analysis, all benchmark results are useless. Also, no one benchmark should be taken as "authoritative". Compare multiple benchmarks, pay special attention to those that run similar code/apps was you will be running, and make your choice that way.
Don't rely on a single number to do it for you.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
Nothing that the average person does is CPU-bound if they have a fast CPU
Exactly. With a trailing edge GeForce 210, an Athlon 64X2 4000 and proper drivers, even CPU hogs like Flash on Firefox don't burden the system. Mostly the system is waiting for me to type or the network to respond.
"I don't know, therefore Aliens" Wafflebox1
No more 125W processors... Those days are over for me. Buying 'tock' processors (the die shrink phase) that run cool and quiet.
Comment removed based on user account deletion
Depends on your principals. A) Preferring to pay more to lower your energy usage or B) Not willing to pay a company your dollar for doing things more than worthy of anti-trust investigations and/or anti-trust lawsuits on a regular basis. Over a year 140 dollars is nothing even for someone in poverty (which I am with less than 15000 a year). Its also a small pittance of a fraction of the KWh being used by unclean energy. Most the unclean energy you use is from the products you consume including gas, plastics and food. There has to be something you dropped money on worth more than 140 dollars this year that was actually more useless than giving your 140 to the electric company over choosing AMD. Mine would probably be beer.
I remember when Intel had a virual monopoly in the 90's and early 200's. Prices of chips were expensive. We complain about $10 on a $200 processor when 10 years ago the entry level proc was over $400 (now they are a scant above $50, take that with inflation). Only when AMD bought out their Athlon 64 proc's out did Intel start to take them seriously with a complete arch redesign (the core series), Intel was so far entrenched in monopoly that before the Core series was released AMD had their X2 series out. Of course Intel didn't complete the arch changes until the Core 2 series, Core Duo's were "enhanced" Pentium M's.
So the tool who modded this flamebait needs their head adjusted with $400 over-hearted Pentium microprocessor.
BTW: _NOT_ an AMD fanboy, I own several models of both brands of Proc's, including a few examples of the quite expensive Pentium III's and Xeon 5500 series as well as a cheap Core 2 Duo.
Calling someone a "hater" only means you can not rationally rebut their argument.
For normal usage (read: not mission and time critical applications), there is no such thing as computing power need. It's not like a specification: "I need my notepad to start in less than 300ms. If it starts in 301ms my computer is useless".
If it starts in more than 30 seconds you can still do your work.
If it starts in 2 picoseconds it can still use an upgrade so it starts in 0.5 picoseconds, or in the awesome wtfbbq 0.1 picoseconds!
You pay more, you get more. Its as simple as that.
I have recently done some (minor and unscientific) tests of a P4 vs. a single core Atom. In per clock performance, they seem about equal and the P4 has a higher clock speed. Of course, the Atom is NOT optimized for speed in the first place. Its strengths are that it is cheap to manufacture and light on power consumption.
A better comparison is P4 versus any modern quad core. The quad might have power consumption similar to the P4, but waay better performance.
Apple mac books: 280- 520 USD Iphone 4: 260 USD Ipad 2 64gb + wifi + 3G : 330 USD New Ipod touch 64gb: 120 USD Dell Alienware M17x: 700 USD Dell Alienware M15x: 500 USD MacBook Pro ( MC024 LL/A )17-inch 2.66GHz Intel Core i7: 510 USD MacBook Pro ( MC373 LL/A )15.4-inch 2.66GHz Intel Core i7: 485 USD BlackBerry Pearl 3G 9105: 350 USD Nikon F 6 - SLR camera - 35mm: 685 USD Nikon D3000 (with 18mm-55mm and 55mm-200mm lens): 315 USD Nikon D3X : 985 USD Canon EOS 5D Mark: 565 USD Playstation 3 PS3 Metal Gear Solid 4 80GB Bund: 220 USD
For about two years (2004 to 2006), AMD's Athlon 64 clearly beat the aging Pentium4. Especially after the dual-core Athlon 64 X2 was introduced in 2005.
Intel took the lead again with the Core2Duo in 2006, albeit at much higher prices than AMD. Since that time, Intel has usually offered the fastest processors at the high end, while AMD usually offers better performance per dollars in the budget range. Recently, the introduction of Sandy Bridge has put more pressure on AMD, but a new AMD processor generation is approaching release. Lets compare that to Sandy Bridge in another two months.
Finally (and that's why I buy AMD), AMD offers ECC Ram with all except the cheapest CPUs. Intel supports that only in the Xeons for the server market, and those are really expensive.
C - the footgun of programming languages
I'm not interested in overclocking, but in the past when I wanted to test my cooling, I've just opened a terminal, started bc (arbitrary-precision calculator program), set the decimal points to 1000000000000 and then entered "2/7" or something. I suppose this is just stressing the ALU?
Intel had good entry level CPUs 10 years ago a lot less than the $400 you're quoting. During the second wave of the pentium II (when they went back to sockets), the pentium III 700 was a very solid CPU that was fairly inexpensive. I bought two of them for like $170 each.
Previous to that, the celeron 366 was even cheaper and could be OC'd into a fairly strong cpu.
previous to that the celeron 300a was possibly the most famous overclocker of all time, it sold for like $125 and would consistently OC to be faster than intel's top of the line mainsteam processor.
Even the lowly celeron 266 (no l2 cache) was a fairly good gaming processor.
While AMD (and to a lesser extent Cyrix) had some early success with cloned 386 processors in the low-end/replacement cpu market, really the price war you enjoy today started with the K6, not the Athlon64. The K6, especially in K6-2 garb, was the first alternate cpu to really start eating into intel's market dominance in the mainstream arena. Its follow up, the original Athlon, took a HUGE bite into not just marketshare, but mindshare, as it was the first time Intel was outperformed in the mainstream arena in not just clock speed but work-per-clock as well.
Honor the K6. we owe our current market to it.