AMD Releases Dual-Core FX-60 Processor

mikemuch writes "AMD just released their new Dual-Core FX-60 processor which is basically two FX-55s strapped together. Unfortunately, the FX-60 doesn't blow away Intel's recently announced Pentium 955 Extreme Edition, and it's actually slightly more pricey. It gets a slight edge in games and runs cooler, as Loyd Case found when he put the FX-60 through ExtremeTech's battery of benchmarks. From the review: 'AMD now ships a dual-core CPU that's essentially the equal of Presler, while generating far less heat. In terms of performance, however, this means that AMD no longer commands the same type of lead it once did when Intel only had the somewhat anemic 840 Extreme Edition. In fact, AMD is now more expensive, at $1,031 (quantity 1,000), versus the 955 Extreme Edition at $999 (quantity 1000).'"

Comments

Other Reviews

[hattig]

Tech Report (FX60 beats out 955 in most of the benchmarks, if not by a large margin then at least consistently).

Also check out AMDZone, AnandTech, Björn3D, FiringSquad, HEXUS, HotHardware, LostCircuits,
PC Perspective, t-break, and TrustedReviews who all have reviews as well.

Re:Other Reviews

[metarox]

I just can't wait to see how well AMD will do once it goes 65nm and changes to DDR2. Power consumption will probably drop by a significant amount proving once again that the AMD design is better. They actually are better with 90nm and DDR memory in most benchmarks.

Re:Other Reviews

[Stalks]

Something is wrong about the reviews that are out at the moment. Perhaps individual chips vary a lot in performance, because I just checked out the article, your linked review, AmdZone and AnandTech's reviews and they all have cross references for the same benchmarks yet the results show a varying difference between the 2 processors. What is the underlying factor that can make one review look like a benchmark is similar on both architectures, and then a different review and same benchmark, AMD is 60% faster? It makes me wonder if the reviewer tweaks the results to show his/her biased view. Whether it be Intel or AMD.

Re:Other Reviews

[ReaperEB-Moo]

I've also seen other reviews that show that the FX-60 out performs the 955 in most of the marks. Yeah the 955 does come out on top in a few. It's pretty impressive that a FX-60 non-overclocked at stock 2.6Ghz can beat the 955 EE that's overclocked to 4+Ghz. Go figure.. There are a few review sources that I trust, because they have no bias towards the products or vendors. Along with the sites mention by hattig, check out the last two issues of MaximumPC, they usually beat the hell out of machines when testing, and even if they love something, they'll still have something bad to say about, but usually in a positive way.

Re:Other Reviews

[Glock27]

AMD has peaked, it is downhill from here for them. The new Intel line will start to dominate.

You mean when it actually ships? Intel's desktop lineup (such as the EE processor mentioned) is still Netburst architecture, just at 65 nm.

AMD should still have a great story to tell when it hits 65 nm. and supports newer memory architectures.

Intel still doesn't have an integrated memory controller, or an answer to Coherent Hypertransport. Even in its upcoming new architecture.

Competition is a great thing, and Intel has a long way to go before it stops losing marketshare to AMD.

Re:Other Reviews

[hattig]

It takes a year or two for these things to filter through to end users buying machines. Notice how long it took AMD to start regaining market share with the K8.

The goodness that is Conroe won't factor into peoples perceptions until late 2007. The best thing that Intel can hope for is that they are far enough ahead of AMD for enough time that they'll change people's perceptions.

At the worst, AMD will have a competitively fast processor when they move to 65nm that has roughly the same power consumption figures and cost.

In the shorter term, AMD's F Stepping on 90nm is meant to reduce power consumption further (or increase clock speed), and adds in DDR2 (667MHz and 800MHz) which some people say will increase performance by between 5% and 15%. However Socket AM2 is a good reason to not buy any high-performance AMD processor at the moment, because better is coming relatively soon (and what games coming out this quarter will really benefit from a brand new rig anyway?).

Re:Other Reviews

[kesuki]

DDr2 ram isn't being used because of it's abysmal timings. http://www.ocztechnology.com/aboutocz/press/2005/1 48
compare that to the 2-3-2-2 timings one can get on DDR modules.

The reason why there is such a huge discrepency between performance between some review sites and others is that some sites are using abysmal 3-3-3-3 timings DDR memory for the FX-60 while others are using the better timed DDR chips. For gaming there is a HUGE advantage to having 2-3-2-2 timings because the entire content of the ram can be dumped almost twice as often as 4-4-4 timed DDR2, which because of it's better frequencies can pump more data at a slower rate.

Mind you AMD will need DDR2 support in the future, unless they somehow decided GDDR3 was better, because in about a few years DDR2 modules will be coming down to the 2-2-2 timings level, and will blow away the standard ddr modules. i mean technically if you look at video cards with ddr2 and ddr3 memory there is no engineering reason why someone couldn't make a ddr2 or ddr3 memory that worked awesome today, but there is plenty of 'marketing' reasons why they nead to have a 'clear' roadmap into the future.

DDR memory still has a lot of years of life left in it if you get the good timings stuff, like ocz or patriot. too bad ddr2 is 240 pin and ddr1 is 184 pin, so one can't make them pin compatable.. and no doubht ddr3 and ddr4 won't be pin compatable when they come out either.

ah well, tought to say, but if i was at AMD and trying to think of a way to 'counter' the DDR2 solution intel is using i'd instead opt for the simplified GDDR3 as main system memory. At least i'd consider the viability of doing so. the high end memory card market overnight decided to drop agp support and ddr2 support and go all pci-e with gddr3, because they were simpler more elgeant and properly working designs. agp is, was and remains a kludge to work around a problem that a better solution hadn't been thought of and ddr2 is full of legacy design needs from it's legacy heritage too.

Anyways, I'd rather see an AMD system (on 65 nm core) with GDDR3 modules than DDR-II modules.

Re:Other Reviews

[Vigile]

http://www.pcper.com/article.php?aid=197&type=expe rt
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2668
http://www.hothardware.com/viewarticle.aspx?articl eid=767&cid=1
http://www.xbitlabs.com/articles/cpu/display/athlo n64-fx60.html
http://www.lostcircuits.com/cpu/amd_fx60/
http://www.bit-tech.net/hardware/2006/01/10/amd_at hlon_64_fx-60/1.html

Re:Other Reviews

[mcbevin]

I just can't wait to see how well AMD will do once it goes 65nm and changes to DDR2. Power consumption will probably drop by a significant amount proving once again that the AMD design is better. They actually are better with 90nm and DDR memory in most benchmarks.

Doing a search-replace on your message gives, I think, an equally valid but equally pointless form of wishful thinking / trolling:

Yes, well I can't wait to see how well INTEL will do once it improves its design to match AMDs. Power consumption will probably drop by a significant amount proving once again that the INTEL manufacturing process is better. They actually are better with their crappy Pentium IV design and no integrated memory controller in most benchmarks. Lets just stick with comparing what actually is available. I'm no great Intel fan - I bought a dual-core AMD just a month back, but if Intel had had a better option at the time, I would have gone with Intel rather than waiting to see how AMD did with the 65nm process or whatever.

Re:Other Reviews

[Surt]

DDR2 is coming shortly (the new socket M2), in fact the FX-60 may well be the last (highest performing) DDR processor release.

Re:Other Reviews

[default luser]

Mind you AMD will need DDR2 support in the future, unless they somehow decided GDDR3 was better, because in about a few years DDR2 modules will be coming down to the 2-2-2 timings level, and will blow away the standard ddr modules. i mean technically if you look at video cards with ddr2 and ddr3 memory there is no engineering reason why someone couldn't make a ddr2 or ddr3 memory that worked awesome today, but there is plenty of 'marketing' reasons why they nead to have a 'clear' roadmap into the future.

AMD needs DDR2 support in the future because, even with the latency issues, DDR2 at 667 MHz still outperforms DDR 400 with fast timings.

In addition, DDR2 uses a much lower voltage (1.8v), as opposed to DDR (2.6v), giving it significantly lower power consumption. If AMD wants to compete on the mobile front, they'll need to move to DDR2 pronto...Intel already made this move with Dothan and the mobile 915.

As for video cards, thay're not a good comparison. They get exceptionally good usage out of even high-latency memory becuase they perform non-random reads and writes. Graphics engines work on blocks of memory at a time, so they can take full advantage of the burst read modes (reads / writes multiple columns in one continuous burst) offered by DDR, DDR2 and GDDR3; this helps hide the high latency of these speedy parts. What, you didn't think that cards boasting 600-700 MHz DDR1 were doing it with low latency, did you? Fast DDR2 and GDDR3 parts are also very high latency

Sure, you can leverage burst reads on a PC, but only so much. If you have a lot of randomly-accessed data, burst reads will not be able to cover up the latency of reads / writes, and will simply mean more data to clog up your cache.

Also, GDDR3 is still not shipping at nearly the same density levels as DDR1 and DDR2. DDR1 is available with 2Gb density, and DDR2 is available in 1GB density. The best GDDR3 can do is 512Mb density chips, and these are currently only available through Samsung. The price of uptfitting a system with 1GB ram using these high-end 512Mb GDDR3 chips would be astronomical.

Re:Other Reviews

[level_headed_midwest]

For some strange reason, I'd not expect AMDZone to have a completely unbiased view on the AMD chip.

Re:Other Reviews

[thePowerOfGrayskull]

"I know you are but what am I?" Sorry, wrong thread... or is it?

Re:Other Reviews

[Agripa]

The latency issues between DDR and DDR2 have never been clear to me so maybe someone can explain it.

If you check http://www.lostcircuits.com/memory/ddrii/2.shtml and http://www.lostcircuits.com/memory/ddrii/6.shtml they show DDR2 using an I/O clock at twice the frequency of the DDR I/O clock and both using same core clock. The objective here is to double the I/O bandwidth using twice the I/O clock and doubling the width of the core. Given identical core timing, DDR2 would have twice the latency IN CLOCKS as DDR but this would actually equate to the same amount of time.

So excluding the differences in command timing and using the same core clock, would not the CaS latency in both technologies be the same in nanoseconds? CaS 4 in DDR2 looks like it should be equivalent to CaS 2 in DDR in the timing diagram.

Re:Other Reviews

[kesuki]

Well, at least my point about why sites were getting widely varied benchmarks was left intact ;)

I was just drooling over the posibility of ddr3 as a system memory, because it simplifies the bus, so even though it's higher latency it has fewer 'hops' to run through. not grounded in reality sigh, but a man can dream can't he? plus DDR3 can read and write simultaneously across the bus, previous generations (ddr and ddrII) cannot perform that amazing feat. Just that fact alone compensates for a huge amount of latency, and allowed 512MB cards to finally start out performing 256MB cards in more benchmarks. DDRII is six years old already we need better tech :/

Re:Other Reviews

[maraist]

or increase clock speed), and adds in DDR2 (667MHz and 800MHz) which some people say will increase performance by between 5% and 15%.

And I've read other reviews that show an actual decrease in performance by moving to DDR2. Sorry for not providing the research here, but I'll simply say with that caveat that I've heard that DDR2 has greater latency, so the extra "bursting" speed is barely able to compensate. Note that memory clock speed has almost nothing to do with performance, and often times, you get the exact same core memory of the slower speed technology, just strapped on with a higher-speed BUS. The result? Same memory with many many more "wait" cycles, but really fast burst (of 4 successive clocks to fill a cache-line). That's the difference between "value" memory and over-priced memory.

But what is significant is that if you compare said value to premium memory for simple DDR, you still don't see a tremendous performance boost.

Re:Other Reviews

[hattig]

That's why AMD waited for DDR2-667 rather than starting with DDR2-400 and DDR2-533, like Intel. The latency tradeoffs aren't as bad. The 5-15% increase is probably with DDR2-800.

But indeed, you can probably simulate the expected performance increases with very high end DDR. However you can't simulate core enhancements that may, or may not be, present in the next AMD stepping.

Re:Other Reviews

[Agripa]

Cas 4 spends more time compaired to what at what clock frequency?

Is not Cas 4 at 200 MHz ( 4 x 5 nS ) exactly the same amount of time as Cas 2 at 100 MHz ( 2 x 10 nS )? There may be something about DDR2 that makes it slower then DDR but having twice the Cas at twice the frequency does not appear to be the cause.

Re:Other Reviews

[Tekzel]

Dare I say it? Of course I dare. Only an AC would say something so ridiculous.

Kinda INtel Biased

[Unixfreak31]

Maybe I am an amd Droid but the last i read the 4800+ and the 955 were almost identical in most benchmarks expcept those that intel always does well in (Specproof etc) and from the article the fx-60 does slightly better than the 4800+ which makes sense granted the fx is slightly faster so I would say AMD has a slight edge here but any who wants to spend 4 digits on cpu period INTEL or AMD??

Re:Kinda INtel Biased

[gol]

whoa! no more caffeine for you

Re:Kinda INtel Biased

[Anonymous Coward]

please try to use some comma's or dots in your sentences because otherwise people will think it's one big flow of words and will stop reading even if you have something interesting to say like you obviously did however i did not catch it entirely because when i was reading your one sentence i forgot the start of it.

price difference

[ShaneThePain]

you really think 32 bucks is going to make a difference to me if im going to buy a THOUSAND dollar processor?

Re:price difference

[Sj0]

It's a condition I call being "Pathologically poor".

You'll see it in people nitpicking and haggling and generally making an ass of themselves as if they're a step away from the poor house, even though they're doing things which obviously don't qualify them as poor.

This is to be discerned from "Smart shopping" from the desperate nature of it. "Those AMD bastards are so EXPENSIVE!!" for 32 bucks on a 1000 dollar processor is a goof example. However, keep in mind that some people using AMD processors these days were the pathologically poor people of yesteryear who wanted to save a buck at any cost. :P

Re:price difference

[click2005]

Especially when you'd probably save about $32 in electricity costs from running the AMD over the Intel for 3 months.

Re:price difference

[Sique]

It might make a difference for assemblers who try to put out machines priced under a certain limit. If they are trying to build something like a $1599 machine, they have $30 more headroom for the other components, leading maybe to the next better graphic card or a an additional 512 MByte of RAM.
In a market where specs for the components are everything, the prices are made to fit unter certain arbitrary limits, and the balanced choice of components takes a backseat, such $30 may be the deciding factor for choosing the processor architecture.

Re:price difference

[pla]

However, keep in mind that some people using AMD processors these days were the pathologically poor people of yesteryear who wanted to save a buck at any cost.

I agreed with you right up to that...

In the "old days", AMD chips cost a LOT less than Intel (like a third to half the price), for 80% of the performance. When you can pay $150 or $400 for basically comparable chips, you can't accuse someone of acting "pathologically poor" for going with the AMD chip.

Recently, AMD has held a small but steady lead over Intel. And they still sold for less, for comparably performing chips... Not half the price, but more than 10% less.

And now... The Athlon 64 has a real competitor. I would tend to call the FP just a tad biased (since another test found the Presler inferior to the 4800, which one might expect the FX-60 to beat). And AMD charges a small premium for it. Not acting as an apologist, just observing a trend... Personally, I think AMD may have made a mistake in judgement there, because it will push away some of their underdog-loving fans.

As for me... I've made the switch to Athlon 64s, primarily for their power and heat edge over Intel, but also because (at least until now) they do perform significantly better, dollar-for-dollar. Very little chance I'll rush out and buy an FX-60... This may very well drop the 4800 to a price at which I will buy it, however.

Re:price difference

[Ruger]

Well, it's $32 more, but the delta we're typially used to (AMDvsIntel) is negative...AMD being considerably less. So being $32 more isn't really the issue. If you expect the AMD to be $100 less, it's really overpriced by $132.

Re:price difference

[wild_berry]

I agree with your concept but not the specifics: someone building $1599 system couldn't throw $999 on the processor, what with their markup having the parts cost somewhere between $900-1300. I'd believe it for a $2500 or $3500 system.

Re:price difference

[asuffield]

This was inevitable really. Once Intel woke up and realised that they couldn't just coast along without opposition any longer, they had to trim the pork from their operation and start turning out cheaper processors. AMD didn't have any fundamental advantages over Intel, they were just trying harder. It's one of those rare occasions where the free market demonstrates actual competition that results in positive effects - or at least, it appears to be.

Re:price difference

[laffer1]

You're right, but I also think AMD needs money. They have to compete with intel's new marketing campaign somehow. So far AMD has done well with smaller ad campaigns and word of mouth. If they want to get into the business sector and my local best buy, they need to tell people they exist. Maybe AMD can drop their lawsuits with intel and compete by paying for Dell and HPs ads like intel does.

Re:price difference

[rvw14]

I would say that by making their chip more expensive than the Intel chip, AMD is marketing their chip to the person who equates $$$ spent to quality. If you believe your chip is better, why not charge more for it?

Re:price difference

[InsaneGeek]

Idle the Intel 955 uses 189 watts
Idle the AMD FX 60 uses 148 watts
Difference of 41 watts
41/1000 = .041 watts per kWh .041 * $.078 = $0.003198 difference in cost per hour to run (7.8 cent average)
$31 / .003198 = 9693.558 hours = Electricity break even point ~403 days

Under load the Intel 955 uses 286 watts
Under load the AMD FX 60 uses 225 watts
Difference of 61 watts
61/1000 = .061 watts per kWh .061 * $.078 = $0.004758 difference in cost per hour to run
$31 / .004758 = 6515.34258 hours = Electricity break even point ~271 days

Re:price difference

[Dastardly]

Not to mention it seems the AMD parts tend to retail for closer to or less than the 1000 unit price than Intel.

Re:price difference

[Deliveranc3]

However AMD uses a cheaper motherboard system because of competition instead of Intels massively overpriced MOBO chipsets.

Coupled with AMD's cheaper memory (DDR is still cheaper than comparable DDR2 mainly because DDR2 was supposed to be a similar performing but lower priced successor, and because there isn't much call for enthusiast ram for Intel mobos.

If system manufacturers are "willing" to use AMD's cooler then the cost savings will be significant.

One problem is that they have been using the grey market OEM AMD processors, which saved consumers about 5-10% on any AMD processor which could be spent on a custom cooling solution.

AMD should really consider offering an OEM solution even if they are worried about counterfitting, we'd simply appreciate it too much for the cost to be a waste.

Re:price difference

[maraist]

Dude... The words FX and value should never go together.. Just like Intel's EE editions.

You most definitely do NOT get your value from the incremental extra performance the extra cache (and BUS) provides.

The point of this line of CPU's is called sucker takers. Those that want something similar in performance to tomorrows CPU's from today's CPUs.

It's like getting extra performance out of an economy car by putting expensive booster rockets on it.. Yes, technically it goes faster, but only by putting yester-year's over-priced technology on it (e.g. large caches, more pins).

Don't get me wrong, the FX and EE line are great for certain types of applications. But value has almost nothing to do with it. Most applications can better have the money spent elsewhere; HD-RAID, 10k RPM drives, more memory (to disable swap file), instant-on, optimally recompiled software (in the Linux/BSD world), using a 64bit OS (to utilize the hidden registers and more optimal virtual memory archetecture), and in this case, a MUCH more expensive vid card ($500 vid cards are insane, until you compare that to deciding between a $300 CPU and $1000 CPU). The idea that 'the CPU is a black-box' hurts purchasing decisions, because we ask the question, 'how much money for the fastest possible black-box?' CPU companies can get really creative at answering that question; and they'll love you for it.

That being said, AMD has held a tremendous $ + heat + PWR + performance edge over Intel for a long long while (though not always in every category). The real issue is that very recently, they've been able to match Intel's stability, though definitely not in terms of 3'rd party hardware integration stability. Historically motherboard driver support has sucked ass (MB makers don't devote as many resources to maintaining their AMD lines as they have less volume). Likewise for video card vendors (though I can't tell if this is the fault of the MB or the vid card). All too often, I read message boards where the same vid card is rock-solid on an intel board, with a caveat of 'yeah, it crashes on this or that AMD-based board'. This isn't AMD's direct fault, and the success of AMD's current line has helped MB manufacturers devote more resources.

And if you're dealing with servers, it's probably more cost effective to go with the opteron line (though a price premium definitely exists there as well). More often than not, a server application is multi-tasked/mult-threaded, and would rather have a greater number of slower CPUs and even more interestingly, independent memory banks for each CPU. Here the extra money is more proportionate to greater performance. But definitely not for video gaminig of photo-shopping. (What are two dual core CPU's going to do for the 2 or 3 threads of video conversion?)

Re:price difference

[cloudmaster]

And since you can plug them both into identical motherboards with identical memory technologies, this is valid. Wait a minute... Sure, it's a small difference, but a difference all the same.

Anyway, the national average rate for electricity is 8.2 cents per KW/hr. That's 245 days under load. In NY where it's 14.52 cents, that's 145 days. 110 days in 19.23 cent average Hawaii. So that's about 3.5 months to break even in Hawaii running a seti client. And that assumes you're paying the average rate - for something to be average, there generally will be lower and *higher* rates. I'll bet that someone in New York City (or some other "costs a lot but provides so little" city) could break even in 3 months. :)

Re:price difference

[Paradise Pete]

It's a condition I call being "Pathologically poor".

Interesting idea. I guess I'm one of those people. But my motivation is (at least I think it is!) playing the game. I have to make the best "move." Same thing when I (for instance) have to frequently drive to the same place. I'm compelled to figure out the absolute best route to take, even if the difference is only seconds.

By the way, congratulations on getting healthier. You might think about a new picture though. The one on the right looks kinda scary.

Unfortunately? or not?

[digitaldc]

Unfortunately, the FX-60 doesn't blow away Intel's recently announced Pentium 955 Extreme Edition, and it's actually slightly more pricey.

Althought I understand that some people do not like Intel, I think that this will just make AMD work harder to make a faster processor. The competition between these two chip makers will ultimately benefit everyone by creating better/faster technology. That being said, $995 is a bit pricey.

Re:Unfortunately? or not?

[astralbat]

AMD have done well in the past to catch up and even get further ahead than Intel, but I think Intel's future looks bright - particularly in the mobile market with the current dual core Yonah processor and the future Merom processor which is a complete redesign and extremely power efficient. It looks like its AMD who have to play catch up now.

The bottom line is that I think Intel are right when they say it's now about performance per watt.

Re:Unfortunately? or not?

[dubiousx99]

So now Intel is still right when they change their tune to say performance per watt? Was AMD wrong when Intel was touting clock speeds and they were saying its not about clock speeds, it's about performance?

32 dollars

[Quick Sick Nick]

It may be a whopping $32 dollars more expensive, but that's better than replacing your motherboard and memory which would be necessary to switch from intel to AMD or vice versa.

It isn't always a matter of, "this is the slightly better processor," unless you're building a new system.

Re:32 dollars

[mwvdlee]

Since most people don't upgrade $1000 CPU's every year, it really doesn't matter much considering sockets change all the time.
Every single CPU upgrade I've done (once every 2 or 3 years) also demanded new motherboard and memory, and I've always used AMD CPU's. I think Intel switches sockets even more than AMD does.

Re:32 dollars

[WuphonsReach]

Yeah, CPU replacements are generally a dead-end. Unless you initially buy the absolute minimum speed CPU that the motherboard supports and later upgrade it to the maximum speed CPU that the motherboard will support. But that's a somewhat losing proposition because the upgrade is going to only be marginally worth it.

Out of the dozen or so machines that I've built over the last 10 years, I can't think of any that managed to get just a CPU upgrade. They almost always ended up requiring new MB/RAM in addition to the CPU.

And it's not even always just a socket issue. You can run into issues where a motherboard only supports certain speeds for a particular socket (or is particular about FSB). So you have to make sure to get the right "step" of CPU chip.

Price isn't really going to matter

[njvic]

I don't think price is going to make that much difference.. Let's face it, AMD now has a pretty keen fanbase. I know I feel far more loyal to AMD than I ever did to Intel.

Re:Price isn't really going to matter

[Hal_Porter]

I'm not loyal to either of 'em. I just buy the best processor in each generation in the budget I can afford.

I.e. mostly non Intel upto the Celeron 300, then Athlon XP, and probably a Merom next. But if Merom ends up sucking badly benchmark wise, I'll get a Athlon 64.

I honestly don't understand the whole fan boy thing. Processor companies aren't sports teams, and it doesn't matter to you who wins the battle. Hell, most of the people I know don't even get that way about sports teams.

As far as the article goes, this is significant because the AMD brand has strengthened to the point where they don't need to underprice. Which they deserve actually, the design choices they made with the Athlon and '64 were much better than on the P4. With a bit of luck, Intel will make the right choices with Next Generation Microarchitecture, and get back on track with their manufacturing processes, and produce something that leapfrogs the AMD 64. If they manage that, I'll buy it and if they don't I won't.

Oh screw it

<fanboy>
Woohoo, Intel are gonna put the Megahurts on AMD with Merom. Squeel piggy, Squeal. Yeeehaw!!!!
</fanboy>

Re:Price isn't really going to matter

[kypper]

I agree with you.
I love AMD as the underdog, but when it comes down to it, I buy the processor for my needs. I needed hyperthreading when the dual cores weren't out, and the Prescotts overclocked nicely (contrary to what some people say about heating, if you do a proper cooling system, you can overclock about a gig). AMD was also not significantly cheaper for my target system.

Re:Price isn't really going to matter

[yourlord]

I've supported AMD in every generation except during the p5 era (amd had no answer to the p5 for most of it's life). My 386 was an AMD, my 486 was an AMD, my pentium was Intel, my K6 was AMD, my K6-2 was AMD, my Athlon was AMD, and now my Pentium 4 is Intel only because someone threw a 2 month old HP in a dumpster and I grabbed it for the parts.

I bought AMD for 2 reasons, price/performance, and because I know what Intel will do to us if we lose the only competitor on the market. My money going to AMD was more a vote to keep competition alive than anything else.

I'll buy whatever gives the best bang for the buck. In the pentium classic era that was Intel, and intel alone. AMD released the K6 which stepped all over the Intel chips in everything but games, and I stuck with it more to support the only company capable of keeping Intel in check. If AMD had folded or left the market back in those days we would probably all still be running Pentium 3's that cost $1.5k each.

Better for games, still.

[Ruger]

AMD usually out performs Intel in game tests...which is the only reason to own one of these bleeding edge processors, right? It was interesting how much cooler the FX-60 ran compared to intel's 955 Extreme Edition. 15C less is huge. Cooler case, less power required for fans...so quiter too. AMD has always been the better value too, but in this case it's MORE than the intel. AMD should do it's best to avoid this becoming a trend.

Re:Better for games, still.

[Vlad_the_Inhaler]

My understanding is that virtually all current games are single-threaded. Having a dual-core processor only allows other software to run on the second core, it would be rather amazing if it allowed a game to run faster. There is actually a negative here - the processor only goes into power-save mode if both cores are idle.

This processor is designed for use in a server, not for games.

Re:Better for games, still.

[wild_berry]

Th Ahtlon64 FX line is AMD's enthusiast chip. It's clock multiplier is unlocked for overclockers; the FX-60 is intended for gamers as patches arrive for games enabling them to make use of multi-threading capabilities of a dual-core processor. I understand that the games may be little more than hacks to offload some of the work to the second core and to make sure that the primary game thread has a core to itself, but this wouldn't be possible with only a single processing engine in your computer.

Re:Better for games, still.

[default luser]

Virtually all games are single-threaded, but a surprising number of them released recently are multithreaded.

Quake 4 and Call of Duty 2 are both multithreaded via patch, thanks most likely to their development on the XBOX 360.

Serious Sam 2 also ships with multithreading support.

That's a whole lotta big-name multithreaded games for winter 2005. I expect many more next year.

amd vs intel

[chrisranjana.com]

Initially AMD's selling point was price, now they have proved quality.. price comes second

article way biased

[akhomerun]

this summary is so biased, it's rediculous. by being a cooler processor and faster in practically every benchmark, doesn't the FX-60 deserve to cost $30

like that's going to matter when you are buying a $1000 processor. i'd gladly pay the extra 3% for a cooler processor that performs, then my cooling solution could cost $30 less.

I'd also be interested in what the retail prices will be. Yeah, the 1000 quantity intels are cheaper, but what if the markups on the intels are higher once they hit retail? I mean it's not to say that the Intel will be more expensive or the AMD will be more expensive at retail prices, I don't know, but I'd say that there's a good chance that those prices will even out a little when you are buying 1 processor from a retailer.

Re:article way biased

[akhomerun]

oh crap...corrections

subject should be "summary way biased" ..."doesn't the FX-60 deserve to cost $30 more" - forgot the word "more"

yeah...i know
should have hit the preview button

Re:article way biased

[cgenman]

like that's going to matter when you are buying a $1000 processor. i'd gladly pay the extra 3% for a cooler processor that performs, then my cooling solution could cost $30 less.

Or, as the processor is unlocked, overclock it until it runs at the same temperature. At that point it will be faster than the Intel.

Re:article way biased

[Bodrius]

Did you read TFA? I'm all for criticizing slashdot summaries for the sake of AMD fanboy-ness, but how exactly is this 'so biased, it's rediculous'?

The summary makes three points regarding the results from the article, or rather, rephrases the single point in 3 ways:

- The FX-60 has a slight edge over Intel's equivalent and runs cooler
- It doesn't 'blow away' Intel's equivalent anymore
- 'AMD no longer commands the same kind of lead'

IOW: yes, AMD still wins, but it doesn't have the obvious price+perf advantage that it often had. You may actually have to think for a sec before choosing where to spend 1K in CPU bucks. Imagine that.

I'm not sure how exactly you read the article, but I found the summary surprisingly accurate for /.

Re:article way biased

[timeOday]

The processor costs $30 more. The writeup says "slightly more pricey." If you somehow find that false, I suggest you're a fanboy. If you want to justify the slightly higher price, such as reduced power or cooling costs, fine. But don't slam the writeup for being factually correct, or engage in speculation about how the street price of the AMD might actually be less when the opposite could just as well be true.

Re:article way biased

[professionalfurryele]

That would be the case except the AMD chip doesn't overclock so well, they tend to throw up error even running a dash over their usual clock speed. The Intel one can (if you can supply the juice) be seriously overclocked on the other hand. However I don't believe the article when it says the Intel Chip is faster. I've read what I would consider a much more reliable review which had the AMD chip either infront, or way infront for most benchmarks (especially when it came to multithreaded tasks). Maybe there is some variation in the chip qualities, me no know. However, for the time being at least, I more willing to believe results which give AMD the nose ahead than Intel. Granted I'm bias.

Re:article way biased

[Tango42]

Why preview? Just re-read what you typed in the text box. Preview is for formatting errors, not typos.

Re:article way biased

[MythicalPuma]

It's not that this article is biased it's that this article is comparing two very different processors(obvious I know but trust me it gets better). These two processors are set for different release dates http://news.zdnet.com/2100-9584_22-6018135.html(th anks for the link Glock) and so that means we're comparing brand new technology to technology that won't be out on the market for two or three more quarters. According to AMD's buisness plan they should be kicking the pants off this processor by the time they get to the third quarter of this year, which is when pentiums processor is set to release.

Re:article way biased

[Guysmiley777]

It IS faster than the Intel processor! It just doesn't "completely blow away" the Intel CPU. The submitter is asking how AMD can dare to charge more than the Intel product because it's only somewhat faster.

Well, since Intel's "branding" is worth jack more than AMD's these days, charging slightly more for slightly better performance seems to make sense.

anyone with any sense

[PowerBert]

Is going out to buy a AMD64 X2 4800+.

Re:anyone with any sense

[Bobsledboy]

Anyone with any sense is going out to buy any of the X2 range.

One-thousand dollars!

[wetfeetl33t]

Sooo many new, powerful, expensive processors. Most of which are unaffordable. I think I'm getting left behind...

But does it have HDCP?

[Anonymous Coward]

This has got me worried, with all the shift going to high definition in PC's and Intel chips having DRMs like HDCP, makes me wonder if AMD has this as well?

As you know Intel created HDCP, so is AMD licensed to use it? Will AMD PC's not be able to view HDCP High Def?

Re:But does it have HDCP?

[level_headed_midwest]

Let's hope that AMD gets real popular here when Vista comes out. If a bunch of people's AMD computers lack the TPM DRM chips and HDCP and can't play media, like hell they're going to buy a new computer just to play a song or a DVD. They just won't buy the song or DVD...and the DRM crap will abate for a while.

Now I'm not for pirating, but a lot of the DRM crap gets in the way of legitimate use, such as making a backup copy of discs so you don't have to buy another when they get scratched, ripping CDs to play from your hard drive, or even reinstalling applications that require activation or key codes.

1,000 dollar processor perfect for gaming?

[cgenman]

No offense, but when did people start spending 1,000 dollars for just the processors in their gaming rigs?

People! Nothing takes advantage of that yet! And by the time things do, the processor will cost 1/8th of what it does today. I've been running an AMD 2400+ for a few years now, a simple 100$ processor, and I STILL haven't found a game that it can't run solidly.

Yeah, if you need a mission-critical server that you desperately need to be as fast as possible... distribute the load.

Basically the top end is for bragging rights and pure-profit silicon. Neither AMD nor Intel can claim bragging rights at the moment. And that's fine, they both should be working hard to push processor design further and further along, and a leadership question will only help that.

But no matter which is the faster processor, please don't buy one. If you really want the ultimate gaming experience, buy three gaming rigs for that price and invite some friends over. You'll be glad you did.

Re:1,000 dollar processor perfect for gaming?

[v1]

Some things you have to pay more if you have certain performance requirements. If you need a fast jet, you buy a fast jet. You cannot just buy two slower jets and "stack" them. Computers, for high end purposes anyway, can usually be paralelled. At a grand a processor, I can see that CPU getting whupped by a cluster of 4 CPUs running at 1/2 the speed, at less than 1/4 the cost each, yielding better performance at about 1/2 the cost. If you can parallel it for cheaper, I see very little point in buying the heavy guns.

Re:1,000 dollar processor perfect for gaming?

[NewbieProgrammerMan]

If people want to drop $1000+ on these processors even when they don't need them, I say more power to them. Intel & AMD can take that money and use it to design more powerful chips that will benefit those that do need them (and I'll have to pay less for an FX-60 when I actually need one). :)

Re:1,000 dollar processor perfect for gaming?

[m50d]

People! Nothing takes advantage of that yet! And by the time things do, the processor will cost 1/8th of what it does today. I've been running an AMD 2400+ for a few years now, a simple 100$ processor, and I STILL haven't found a game that it can't run solidly.

See how long your end-of-turn wait is 3/4 of the way through a civ 4 game. I'm on civ 3 on my 2600+ and the wait is annoying. Not enough to make it unplayable by any means, but annoying.

Re:1,000 dollar processor perfect for gaming?

[Ziviyr]

You gotta power those extra whole-systems, that costs money over time.

Re:1,000 dollar processor perfect for gaming?

[Achromatic1978]

I have an Asus laptop, 1.86GHz Pentium M (Dothan? The one with 2MB cache). I play Civ IV with all civs, on large maps, and even come 1500AD I'm only waiting 2-3 seconds.

Perhaps the kicker here is that I also have 2GB of RAM?

Re:1,000 dollar processor perfect for gaming?

[travail_jgd]

Gaming isn't the only thing that fast processors are good for. I'd love to have a fast dual-core system, just to eliminate delays and interface slowdowns when my system is compiling*, running spamassassin, etc.

While that kind of load is easily split among multiple machines, that increases the chance of mechanical failure (fans or drives), the cost of electricity, the need for cooling in summertime, and reduces the space I have in my apartment.

For the majority of people, a slower processor will suffice. If I used my home PC for work, I wouldn't hesitate to put in a much faster processor.

If that answer doesn't work for you, look at it this way: faster processors push down the price of slower ones.

* Yes, I'm running Gentoo

Re:1,000 dollar processor perfect for gaming?

[WuphonsReach]

Even a slow dual-CPU or dual-core system would help you on responsiveness. (In fact, with a responsive system, you start to care less about a compile takes.)

Re:1,000 dollar processor perfect for gaming?

[Perky_Goth]

shut up. let them blow their money. how do you think the price comes down?

Re:1,000 dollar processor perfect for gaming?

[level_headed_midwest]

I tend to buy Intel processors because I have to use a laptop. A few years ago, almost no laptops (and I mean *laptop*, not a 10-pound portable desktop)had AMD chips in them. With the relatively recent introduction of the Turion 64s, there have been a few more, but they are still few and far between.

Re:1,000 dollar processor perfect for gaming?

[level_headed_midwest]

Hear, hear. I generally make it a rule to not compile anything that has a source much bigger than a few MB as it takes forever. I hunt down RPMs if at all possible if it's something big. (My rig is a Pentium 4 2.2GHz, 845MP chipset, 2x512 DDR266 running SuSE 10.0) Compiling a recent 2.6 series kernel and modules takes about 75 minutes, what does it take on your Gentoo box?

Re:1,000 dollar processor perfect for gaming?

[Darby]

Compiling a recent 2.6 series kernel and modules takes about 75 minutes, what does it take on your Gentoo box?

Damn, dude, that sounds really slow.
I have a Dell Optiplex gx260 (P4 2.8, 1 Gig RAM) and it takes around 20-25 minutes.
My new server at work is a dual Opteron 246, 2 Gigs of RAM and it took just over 5 minutes.

Re:1,000 dollar processor perfect for gaming?

[shird]

How much did your 2400 cost "a few years" ago though? The only reason it has stood the test of time is because you bought the top of the range a few years ago. If you had bought mid range a few years ago, youd be sitting on something which would have no chance of running a modern OS and games adequately.

Re:1,000 dollar processor perfect for gaming?

[Blnky]

I agree. I use my home system for development as well as gaming when I want a break. I find that the needs of my work system (compilation, servers, rendering, network stuff, etc) tend to overlap with a good gaming system. Delays caused by multiple resource hungry applications really tend to add up at the end of the day. Typically when I upgrade I end up with both a development and a gaming system. I push the memory a little more for the development and I push the video card a little more for the gameing. With respect to the processor, it is the same for both*. There also seems to be a good bit of overlap with pc based server systems.

*Though dual processors is more suitable for development, it does help a little with keeping a game on one cpu the other junk on the remaining cpu.

Re:1,000 dollar processor perfect for gaming?

[Magius_AR]

I've been running an AMD 2400+ for a few years now, a simple 100$ processor, and I STILL haven't found a game that it can't run solidly.

Try F.E.A.R at 1600x1200 resolution w/ all details on high.

Believe me, software has most certainly outpaced hardware. Now you yourself might be happy playing all your games at 800 x 600 with medium detail, no shadows, and no anti-aliasing, but some ppl like the immersive experience. They're the same kind of people who are into HD everything and multi-speaker surround sound.

Now, while I agree spending this kind of money on a processor is utterly ridiculous (I have a 3000+ myself), I also don't believe newer games wouldn't take advantage of faster processors (or faster GPUs for that matter...they sorta go hand-in-hand).

Pathologically indeed

[Lonewolf666]

Someone who worries about 32 bucks on a 1000 dollar processor probably should not buy either the Pentium EE or the AMD FX-60. The 4400+ would be a much smarter choice, maybe 15% slower but much cheaper.
The last 20% of performance are always disproportionally expensive. Unless you really need them or are really rich, buy a bit smaller.

Clarification

[rellix]

I'm in need of a bit of clarification - I've seen the term used before but ignored it until now. At the end of this article:

In fact, AMD is now more expensive, at $1,031 (quantity 1,000), versus the 955 Extreme Edition at $999 (quantity 1000).

What exactly is "quantity 1000" referring to? Is that the number they've produced? Should I completely ignore this term?

Re:Clarification

[c0l0]

These numbers refer to the per-unit price you're going to pay when shopping for at least 1000 (one thousand) units directly at the manufacturer.

Re:Clarification

[TheCowardofAnonymous]

It means that is the price for 1000 of the CPUs - bulk cost, in other words.

Re:Clarification

[neochubbz]

This simply means that that is the price when you purchase 1000 processors. This is not necessarily the retail price you would find at newegg, best buy, etc.

-Chubbz

Re:Clarification

[Andy Dodd]

It means that if you buy 1000 processors at once, you get that price.

If you buy a single unit, it will cost (usually significantly) more. I'm not sure if AMD or Intel will sell quantities less than 1000 directly to anyone. If you buy from someone who bought 1k units from AMD or Intel, you're definately going to pay more than that price.

Expect these to be available to consumers at a 10% markup or more.

If you buy more than 1000 units, you may get a lower price.

I've seen some ICs go for $5+ each in quantities below 100, drop to less than half that for order quantities over 100, and then go sub-$1 for quantities over 1000.

Re:What kind of video card?

[Neoprofin]

Parent was modder down for being offtopic but his first thought was the same as mine. Even if you get a cheap processor like the 2400+ for $100 you're still going to blow another $100 atleast to get a videocard capable of running the latest batch of games, if not closer to $200 if you want the latest eye candy.

This isn't the first time I've seen the latest processot priced around $1000, the same thing happened when the first AMD64 FXs came out. If you want to see a real batch of money and powerhogs look at videocards. Only $2200 for cards in you new quad SLI rig!

Its no opteron

[Anonymous Coward]

According to toms, in most cases the 955 couldn't even wax the X2 4800 in most benches.

http://www.tomshardware.com/2006/01/05/the_65_nm_p entium_d_900s_coming_out_party/page20.html

Even though it is 65 nm it still can't even beat the X2 for power consumption either.

And to the person who said go out and buy an X2 you've got it all wrong (well somewhat). The most cost effective cpu right now is the 939 dual core opteron for its legendary overclockability. My 170 was installed yesterday and I had no problems bringing it up to 2.4ghz running cool. X2 4800 performance for half the price and I'm not even pushing it at all. I've got no doubts that 2.6 is easily attainable. All for maybe quarter the price of an fx-60.

Those of you looking for a $1000 cpu might be wise to look into the 940 dualcore opterons that can be dualed on a board for 4 cores. Whilst you might pay a few more hundred dollars nobody can deny that 4 core is going to beat the pants off anything 2 core.

Oh but of course most games don't support threading so you're better off with a single core still if you are a gamer.

Hope that helps

Re:Its no opteron

[Dun Malg]

Oh but of course most games don't support threading so you're better off with a single core still if you are a gamer.

I would say you have little to gain with dual core if you're a gamer. You're never better off with single core vs. dual core.

Re:Its no opteron

[PhraudulentOne]

You're never better off with single core vs. dual core.

Not to nitpick, but if price is your main concern, then you are better off with a single core. If your looking for performance, and cost is no object, then yes, you are correct.

PC Gamer Magazine

[Neoprofin]

PC Gamer reviewed the EE955 and the FX-60 in their Febuary issue and stated:

"The FX-60 trounced the Pentium Extreme Edition 955 in test spins with Quake and F.E.A.R. Even more humiliating in F.E.A.R. the FX-60 came out ahead of the PEE 955 overclocked to 4ghz by 25FPS." ExtremeTech ran plenty of benchmark programs, but in real application tests there was no competition, The FX-60 showing to be around 30% faster in every benchmark.

Re:PC Gamer Magazine

[Neoprofin]

So I missed the page where they test 5 games, the results are still in AMDs favor but not by the same margins found in PC Gamer.

PC Gamer also tested F.E.A.R. with Nero transcoding a DVD in the background and it still defeated the PEE running only the game.

Re:What kind of video card?

[cgenman]

I agree that it is a valid question. The CPU is just one part of a computer, and depending on what you do it may not even be the most important.

I have a ATI 9600 / 128. It was about $120 dollars when I bought it, which seemed reasonable at the time. High-end graphics cards go for $400 dollars, and can be doubled up for an $800 gaming graphics system. I haven't seen quads, but they must be right around the corner.

Oddly enough the one bottleneck that kept coming up during developing the one PC game I've worked on was RAM. If you ever ran out of RAM, your system went through the floor. If you could speed up your RAM bus, your game performance went up in almost direct proportion. There is just so much data in modern games (real or superfluous), that faster and better can be really helpful, and if you ever fall below a certain threshold (my game was about 512MB) your performance is screwed.

YMMV, but my gaming rig came in at about 700 dollars with specs that most people would be quite pleased with... 1GB RAM, 200GB silent HDD, a motherboard that didn't crap out in the first week, a 300W psu / "silent" antec sonata case, video capture card, 3.5" and 5.25" floppy disk drives, a DVD+-RW dual layer drive... Basically, there is no need to spend more than 1,000 dollars to get a really solid home computer, including monitor. (If you're in the market for a server, you have special needs which may not be covered here)

more expensive...

[l3v1]

...with $0.032 per processor.

Ok, you can mod me offtopic :D

Only $32 more?

[Dsm0nd]

REPOST:

Well, I read that if I am throwing in $1000 for a processor I don't care spending $32 more...

Not sure. Those prices are for 1000 units. That means manufacturers.

Now suppose they sell the whole computer for exactly the same price. No matter which processor they use. If they sell 10K units (not a big deal) they generate $320,000 more revenue. Not exactly 32 buck saving, right?

Manufactures try to save to the cent.

Headline is misleading

[EmagGeek]

Sure, it might cost slightly more, however, if you look at the long-term cost of ownership compared to the Intel, the AMD is far cheaper. Why? ELECTRICITY. If the AMD uses 30 fewer watts than the Intel (didn't feel like digging up an actual number), assuming it's 70% utilized over its lifetime, that's 21W less on average. 24/7 for a year, that's almost 184kWh less per year, which is about $40 worth of electricity at commercial rates (Philadelphia area).

Also, there is a cost associated with cooling the datacenter, which will probably increase those annual electricity savings by about 50%.

I'd pay $30 more to save $60/year. If the lifetime is 5 years, then that's $270 less over the lifetime of the CPU.

Re:Headline is misleading

[hattig]

What if you buy the lower power processor AND turn off that 60W lightbulb that you leave on all the time for no reason at all? What if the cost of electricity goes up (which with all those oil and gas power stations it will inevitably do)?

That's if you have any old-style lightbulbs still, except in rooms where the on-off frequency is quite high (bathrooms).

And to the parent's parent poster, the FX60 system used 60W less than Intel system fairly consistently - idle and under load. If you overclock the Intel system to gain comparable performance then it was more like 100W difference at load though.

I do think that taking running costs into account when buying something is sensible.

Re:Intel's going to own the next gen of processors

[the eric conspiracy]

Not until Intel gets the memory controller on the CPU, which isn't until 2007.

Will this mean other X2 prices will go down?

[kalirion]

When a new "top of the line" CPU is introduced, older ones are supposed to decrease in price, right? I really how the price for the 4400+ drops - I plan to purchase one within the month for a new system, and it's been at $497 on newegg for what seems like an eternity now.

Re:Will this mean other X2 prices will go down?

[Weaselmancer]

Here's hoping they do, but it probably won't happen within the month. I bit the bullet and bought a 4400+ X2 around the beginning of last December and paid about $520. They're not coming down in price very fast.

But that being said - it's worth it. X2 is so good you'd swear it's alien technology.

Re:What kind of video card?

[QMO]

"I haven't seen quads"

Wasn't some of the Voodoo series quad capable, about 10 years ago?

I don't remember clearly since I generally run on hand-me-down video cards.

Re:Intel's going to own the next gen of processors

[hackstraw]

Not until Intel gets the memory controller on the CPU, which isn't until 2007.

Huh? http://www.xbitlabs.com/news/cpu/display/200506152 32538.html

But Intel still charges $2,000 to $3,000 for the things. Currently, AMD Opterons are the best all around chips in terms of price, performance, and power consumption. I can't wait until massive multi-core processors are commonplace, x86 finally dies with disco, and BIOS goes away with Reagonomics.

Re:Only $32 more?

[corngrower]

Ah, but suppose that board for the intel chip requires some support chips for the processor that are more expensive than those for the AMD chip, say $15 more, that would be only $170,000 more revenue if they sold 10K units of intel based computers. But supposing also that you used the AMD chip, and got more units in sales, because most gamers know that the AMD is a better chip for them and a $20 price difference when you're paying $2000 for the machine is squat. How much potential revenue are you losing? By my reasoning they'ld only need to sell less than 100 more units of the AMD to make that additional revenue, which I'm sure they would do.

waiting

[fireiceviperhotmail.]

I just think it isnt smart to spend $1,031 for a cpu thats running on the dead socket
939 i mean everyone knows socket m2 is coming...

Julien http://free.hostdepartment.com/8/81fortune/index.h tm

Re:Only $32 more?

[Dsm0nd]

Completely agree.

It's just a task for correct marketing. And in any case, the article clearly states that Application performance benchmarks showed better performance for the AMD, so anyway gamers would buy the AMD option.

Re:not a bad price

[Glock27]

You can get a 939 board, fx-55, ddr400 dual channel ram, cpu cooler, 250 gig hd, combo burner, ghetto case, 400w ps and 7800 gt video card and still be right around $1800.

At the moment, I'd rather get an Opteron 165 or 170, and save about $500. For most (video card limited) games you'll see exactly the saem framerate, but you'll have the extra core for those games that take advantage of it, or other compute loads like Ventrilo. Not to mention what a system like that could do with serious work...

(Also, I'd spend a bit more on a quality SLI motherboard, that seems a better way to go... :)

Another increasingly modest gain

[deuterium]

I'm getting kind of bummed about the meager gains we're seeing in recent years with CPU speed. Sure, we're going to multi-core, but there's only so much a programmer can do to wring performance gains out of threading without making things too complex and unmanageable. I want to know when the next real quantum leap is coming, like optical processors or quantum computing... something that gives us an order of magntitude increase instead of another 8%. Looking at Intel's timeline for the next 10 years isn't encouraging. A lot of fuzzy ideas about the Internet, "user experience", and parallelism. I wonder where we'll be performance wise in 30 years, in comparison to the difference between now and 1976.

Re:Another increasingly modest gain

[corngrower]

I'm sure that software developers will find a way to improve their products with multiple threads of execution. Better animations and simulations for one thing. I don't see clock speeds increasing very quickly for the next 5-10 years either, but I can envision ways in which those additional transistors on the new chips would be used to provide more functionality with less power consumption. I can see processors supporting hardware level parsing, making compilers & interpreters faster, effectively making interpreted code nearly as fast as compiled code. Better support for speech recognition and machine learning applications could be built in as well.

Cooler equals faster

[ryty]

Now that I have read this article, a cooler, close in speed processor matching with a faster, hotter processor sounds logical to me. Bump up the multiplier, take the beast for a ride, and if you have any problems, mod a pentium 955 extreme cpu cooling fan.

wtf is industry strength?

[Ayanami Rei]

Is that like "commercial grade"?

Did anyone read the review?

[Anonymous Rockstar]

The only benchmarks that the fx-60 got beat in was sisoft sandra and some speech recognition app that they only got beat by tenths of a second. In real world benchmarks the 955 extreme edition couldn't touch the fx-60 in other benchmarks. I get a feeling the poster didn't read it either.

Re:Bias?

[level_headed_midwest]

It's called the Conroe. 65nm dual core, 64 bits, ~14-stage pipeline, based on the Core Duo laptop chip architecture- NO MORE NETBURST!, 2x2MB L2 cache, don't know the FSB yet. Probably going to debut at about 2.3 to 2.5GHz.

2400 to run anything?

[Fei_Id]

My 2200 a couple years ago DEFINITELY would have trouble running games solidly today. You've got delusions, friend.

How about: Doom3, FEAR, and Quake4? If you think your processor will run them "solidly" (solid is full quality settings except for AA and AF) then you probably havent played them :) Because my A64 3200 won't play FEAR at full specs; and if I turn on AA and AF at full resolution in the other two games, it bogs down.

Maybe you've got a different idea of "solid" than everyone else; but to REALLY run new games, a 2400 isnt going to cut it.

Re:2400 to run anything?

[Blnky]

Just thought I would reply to this as I currently have both a 2200 and Doom3. It is indeed solid while playing. No crashes, smooth play, etc. Very enjoyable. I can even crank everything up and it still stays solid. Though, at that point, I can go get a cup of coffee between each frame. The grandparent post is probably correct in its use of "solid". What you are talking about is smooth. With that, you are correct. At high settings, the 2200 (and a medium graphics card) is as smooth as a heart monitor showing flat-line and about as responsive. :P

Re:2400 to run anything?

[Fei_Id]

Yeah I figured it was semantics pretty much :)

You say 'tom-AY-toe' I say 'tom-AH-toe' ;)

Always shop at the knee

[Weaselmancer]

If you go to pricewatch.com or any other place that lets you look at a bunch of computer gear all at once, you'll see what I mean. The price goes up a dollar or two for every 100MHz or so until it hit a certain point. Then it starts to go up $10 or so per 100MHz for 3 or 4 processors...then *wham*, prices shoot through the roof. $100 to $500 per 100MHz. The graph, if you plot (roughly) dollars on the y-axis and performance on the x, will have a knee in it.

Shop there. Best bang for your buck.

That being said, it's another noticeable thing that the top 5 or 10 processors are past the knee. Whenever another new processor appears as the very most expensive, one of the ones above the knee tends to drop into the knee.

So everytime a new super-expensive part comes out, a new candidate drops into the knee. And that's a great thing if you keep your eyes open for a bargain two-weeks-ago-it-was-high-end cpu.

Dual cores...

[Cyno01]

How many of these apps utilize dual cores? Most games dont. So if this thing is just 2 FX55s, that means that an FX55 is beating intels newest chip.

Re:Dual cores...

[kryocore]

It's not just about Games. Dual cores don't need a program designed for multithreading for them to shine. If you are a multi-tasker... most people are, then you will see the benefits. I ofter have 10+ windows open at a time, and dual core benefits here. Just like Intel's hyperthreading helps with multi-tasking, dual core helps even more so.

Re:Dual cores...

[jafuser]

Dual-core processors do help single-threaded games to a small degree. All those other tasks that are running in the background can use one CPU while your game hogs the other CPU for itself.

Of course, as more people get dual-core systems, you can expect games to take advantage of it.

More interesting story: Another Dell rumour!!

[edxwelch]

Nevermind the FX60, there is a far more interesting story:
http://www.theinquirer.net/?article=28889
Anyone like to partake in some pointless speculation weither the new Dell rumour has any substance or not?

full list of reviews here

[ruiner5000]

http://www.amdzone.com/modules.php?op=modload&name =News&file=article&sid=4680&mode=thread&order=0&th old=0
We overclocked to 3.2GHz. Also we have links to stories on the release.

http://www.amdzone.com/modules.php?op=modload&name =News&file=article&sid=4685&mode=thread&order=0&th old=0

Re:Where is the Dual core Turion, please?

[level_headed_midwest]

Yeah, I am looking for the same thing. My old P4-M Gateway is getting long in the teeth and is needing to be replaced. I have heard rumors about a dual-core Turion 64, code-named Taylor, but it's release is expected to be in "2006." That's all I have seen except that it might run on a new 600-some-odd pin socket. Nobody's even sure as to if it's going to be a 65nm or 95nm chip. I guess AMD is taking a page out of Apple's book- no info about a future product until it comes out.

Real poor people overclock

[Dr. Spork]

What the proper poor person does is this: find AMD's best processor (now FX-60), see what core it uses, buy the cheapest chip with that core (Toledo, X2-4400 $500), huge heatsink, overclock past FX-60 levels.

If that's still too much, I'd recommend the X2 3800+ Manchester core, only $322 at newegg. People report getting these running cool and stable well past 2.6GHz with a good heatsink.

Proper cooling is required for the FX series

[Anonymous Coward]

One of the main points of the FX series of processors is ability to overclock. The new Opteron 939's are able to pull 600mhz+ easially (mine is OC'd 1.1ghz) but they have a problem with very cold temps, ie chilled water and phase change. This is where the FX series excels. You can take a FX-57 to around 3ghz on air and decent water before it tops out, but with chilled water or phase change, you can bring it up to 3.4+. So in reality you are paying for a chip you can OC far beyond anything out there right now.

The few OC's to the FX-60 I've seen bring it up in the 3.3 - 3.4ghz range, which for a dual core that preforms better than every other processor out there, is insane.

Re:not a bad price

[synth7]

My first self-bought computer (circa fall of 1990):

386SX-16Mhz
4MB (everyone else that I knew has 1, maybe 2MB... totally jealous)
VGA card /w 256K (I think)
40MB hard drive
BOTH a 5.25 and a 3.5 floppy.
14" color vga monitor
DOS 5.0
Original Soundblaster PLUS the add-on CMS chips for synthesized music
Dual serial and single parallel card

Dropped about $2300 on it.

Anyone remember shopping around for SOCKETED serial ISA cards, so you could pull off the crappy UARTs and put on the 16550s?

AMD or Intel?

[Kyrokushen]

I dont know which to choose, I like AMD a bit more than intel but I find that both release great product. I myslef have a pentiun 4 processor in my computer but would much prefer a AMD64FX....poor me. How about you?

Available when?

[Targon]

What I find funny is that no one has mentioned availability when they compare AMD to Intel.

Can you buy the latest from either AMD or Intel at this point? If I can't find a processor on sites like pricewatch and pricegrabber, then from my perspective it's not available for purchase.

So, the big question you should all be asking is when we will see the Athlon FX60 or the Pentium 955EE available for sale.

If there are only enough processors available to ship out to review sites, but not enough to sell in volume, then it's just a paper launch. My own feeling is that we will see the FX60 available sooner than the 955EE. Selling through only one vendor(such as Dell) doesn't mean it's available because Dell will only sell the new chips in systems and not for individual sales.

Re:Intel's going to own the next gen of processors

[the eric conspiracy]

I don't understand the Huh? - the article says 2007.

Bias? We don't need no stinking Bias...

[Some_Llama]

So the AMD chip is a dual core (Intel 955 is single core), performs just as good if not better in most benchmarks, AND runs cooler? I can totally understand the articles complaints about a 32 dollar price difference.. WTF?! Nice slant there buddy...

Re:Bias? We don't need no stinking Bias...

[Some_Llama]

Yah, stupid me, I realized this after reading a second review elsewhere. When I read the main article linked here I didn't see the dual cores mentioned...

Why two FX-55s?

[Ndkchk]

Why wouldn't AMD just "strap together two FX-57's?"

Re:Why two FX-55s?

[be-fan]

The thing is, it isn't two FX-55's. It's a Toledo core chip. Thus, basically its a 4800+ clocked to 2.6 GHz.

Re:The upgrade is worth it!

[WuphonsReach]

True, dual-core does change the equation quite a bit. There you're actually going to get a significant increase.

One of the scenarios that I encountered was with an AMD AthlonXP motherboard that would only take a limited range of chips. The fastest chip supported was only 50% faster then the slowest. Plus, memory speeds had gone from PC2100 to PC3200, so you would've been shackling a 50% faster chip to a slow PC2100 memory bus.

Re:Only $32 more?

[Dsm0nd]

Correct me if I'm wrong here...but

I'm afraid you are...

AMD = $1032/unit in 1000 unit packs.
Intel = $999/unit in 1000 unit packs.

The difference is $33 (I'll assume $32 to round to your/my previous numbers) per chip. So if they sell 10K PCs -> 10K processors. And 10K processors @ $32 makes $32K.
You were wrong, but so was I. I misplaced a 0. It's $32,000 more revenue (not $320,000, but no way $320)

...I think...

Missing the point?

[MaximuMK]

"AMD just released their new Dual-Core FX-60 processor which is basically two FX-55s strapped together. Unfortunately, the FX-60 doesn't blow away Intel's recently announced Pentium 955 Extreme Edition, and it's actually slightly more pricey. It gets a slight edge in games and runs cooler, as Loyd Case found when he put the FX-60 through ExtremeTech's battery of benchmarks. From the review: 'AMD now ships a dual-core CPU that's essentially the equal of Presler, while generating far less heat. In terms of performance, however, this means that AMD no longer commands the same type of lead it once did when Intel only had the somewhat anemic 840 Extreme Edition. In fact, AMD is now more expensive, at $1,031 (quantity 1,000), versus the 955 Extreme Edition at $999 (quantity 1000).'" The argument that the AMD is not better by much based on its standard clock speed and voltage setting is kind of a misnomer. Admittedly, the AMD runs far cooler than the Intel Extreme 955. This means that someone who knows what theyre doing, can overclock that dual core AMD high above its retail specs depending on heat tolerances, and make it worth the extra buck. Saying that the AMD chip is near the same speed as the intel is deceptive if youre basing it on preset clock and voltage settings. The FX series of chips ARE NOT MEANT for people on a budget. They are sold by AMD for use in high end gaming machines, and machines running true 64bit OS apps. If you look at lower priced chips from AMD and Intel, its much more obvious who is cheaper.