Slashdot Mirror
Core 2 Extreme 40% faster than Pentium EE 965?
Marc writes "As far as I know, this is the first time that Intel has talked about what we can expect from its new gaming CPU, Core 2 Extreme. For once, there is no word on power consumption on this new chip, but Intel talks about raw speed and a 40% gain over the current 3.73 GHz Extreme Edition 965 - which would be rather impressive and could indicate a problem for AMD. In this interview with TG Daily, Intel also claims that a Core 2 Extreme-based enthusiast PC will leave the pixel power of a Playstation 3 in the dust. Gamers, this appears to become the most exciting year for you in a long time!"
The demo system Intel is showing at E3 features a Core 2 Extreme processor, which, judging from past pricing strategy, will cost slightly over $1000, as well as a Quad-SLI graphics card (i.e. probably two dual Nvidia graphics cards at around $1000 each).
Now, when you build such a high-end system you probably wouldn't skimp on the case ($200), motherboard ($200 & up), memory ($300 & up), power supply ($100 & up) and peripherals, either, so let's allow another grand for these things and you wind up with a $4000 PC.
Put in a Blue-Ray drive (expected to cost around $1000 initially) and you just hit 5 grand.
I'm not a Sony fanboy, not by a long shot, but comparing a 5 grand PC to a 1/2 grand PS/3 does seem a tad unfair, now doesn't it?
And yes, a quad-SLI system with a Core 2 Extreme *is* expected to blow the doors off a PS/3. No surprise here.
Dedicated Linux servers (root access) $45 p.M.
Although one must wonder why AMD would be scared of a 5.2 gHz rather than a 3.7 when CPUs that fast are never, ever the system's bottleneck. Seems like a lot of posturing.
I'm sick of hearing about power consumption and yadda yadda yadda. I buy processors based on performance, not on how many watts it eats up. Sure, it's nice to have a cpu that draws less power but why should that have to be one of the main marketing points unless the cpu is for a laptop? I'd rather have all cpus rated by megaherts again.
Because the ugly x86 instruction set acts as a form of compression, x86 code is more dense and fits more easily into the instruction cache than RISC code. The overhead of translating x86 to internal RISC is basically fixed and is therefore getting smaller each process shrink. It's already negligble. For this reason, the ugly x86 instruction encodings are now an advantage! x86 also gained an additional 8 registers and a cleanup with AMD64.
I've read over the article albeit briefly and I find myself thinking that the quote in the summary is total hype for a chip, sure a PS3 will cost about 600, but I seem to recall those EE chips being as much if not more and given that this chip is newer then the P4ee's no doubt it will cost even more. And that's not counting the cost of video cards etc.
I'm not impressed even by their marketing numbers. When I bought my 386 it was way more than 2x the speed of my 286. My 486 was at least twice as fast as the 386, ditto the Pentium, K6 and Athlon.
40% faster? Who cares. Especially for games.
2 is always smaller than 3 - even for larger values of 2.
. . . until next year. : )
harmonious design
It doesn't matter what Sony or intel come out with, neither company is likely to convert PC gamers to console gamers or console gamers to PC gamers.
Then again, what do I know? I fall into the "has too much money/buys them all" camp.
The article summary states:
"Intel also claims that a Core 2 Extreme-based enthusiast PC will leave the pixel power of a Playstation 3 in the dust.
but then I also see in the article:
"[I don't know off the top of my head] the number of polygons it can draw versus a Cell, but I think it's going to be higher, because there's a lot more bandwidth on the quad system than on the Cell system."
That doesn't sound like much of a claim to me.
I would expect in actuality we would be seeing something like a 60-70% increase in speed. A company like Intel would probably estimate conservatively so as to not over-hype a new product.
..that using the word extreme should be illegal?
x86 isn't less efficient. In some cases its even more efficient- you need less cache on common instructions. And some very complex things can be done in silicon with 1 instruction, saving overhead of multiple instuctions. FOr example, memcpy and memcmp are single instructions.
x86 is more complex. Its much harder to write a decoder for, and more difficult to debug the hardware. That adds cost (and a lot of extra transistors in the decode phase). But its a matter of complexity and cost, not efficiency.
I still have more fans than freaks. WTF is wrong with you people?
Go ask a gamer sometime whether he'd rather have a 40% faster processor or a new $1000 video card. Go ahead.
Bull... I believe the wii controller and the ps3 controller look quite promising for fps games. It's not quite mouse keyboard, but it might have some real potential.
I'll be a convert soon. Sick of throwing money at my PC.
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
Morons. They account for roughly 99.9% of the US market.
I still have more fans than freaks. WTF is wrong with you people?
A tiny huffman module seems like it could do wonders for code compression and also take negligible silicon these days.
I remember some papers on that from the guys at Colussa (the guys that Microsoft bought to be their CLR) that seemed impractical at the time, but now.... hmmm... double your cache for free.
I used to think this, too, but Macintosh Universal Binaries regularly see the Intel side both have bigger code and use more RAM (gcc codegen for both sides). I don't know why this is, but I'm wondering if the suggested instruction ordering, alignment, and such to optimize for Intel's latest processors eliminates the old advantage.
E pluribus unum
Please post the link to the data that supports your claim of "power hog".
https://www.accountkiller.com/removal-requested
The closest to real numbers that I know of is a black box test that Anandtech did a little while ago. The chip they had was supposedly the high end but not extreme Core 2.
When I bought my 386 it was way more than 2x the speed of my 286.
:(
Depends on your software and your hardware.
Did you have a 386DX40 or a 386SX16? I had a 386SX and my friend had a 286, Wolf3D.EXE was the same speed on both computers
There's a bit more design elements going into a PS3 than just the raw pixel pushing. I still don't see many FPS games on a PC that can do let 4 players play on the same computer screen.
today is spelling optional day.
AMD's Athlon 64 is 36% faster than Pentium 965 EE in UT2004 http://www23.tomshardware.com/cpu.html?modelx=33&m odel1=238&chart=71&model2=329
Is Intel's new Core 2 Extreme only as fast as AMD's FX-57?
Who then is Sony hoping to sell the PS3 to?
Blu-Ray fanboiz.
Edith Keeler Must Die
Definitely a DX. If I remember correctly, the SXs were DXs that were screwed up in manufacturing.
given the investment that anyone makes in a computer system designed for gaming, how it is a "most exciting year" to be faced with the possibility of yet another set of continuing reasons to spend more money on yet more gear? wouldn't a really exciting year in gaming have nothing to do with new hardware and everything to do with cool, inspired and inspiring new games?
link
Some Core 2 Duo C processors will apparently sport an E designation, indicating consumption between 55 and 75 watts.
75W is comparable to an Athlon 64 FX, for a processor that was designed to be miserly that's a pretty terrible direction to be heading in.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
I'm glad the names for processors are getting a little more exciting.
'Core 2 Extreme' is a techy name, and will date rather quickly, but better than the mundane 486, pentium III, pentium M etc. I always liked 'Athlon', there is something inoffensively sporting about it.
Now all I need is a few more nouns.
Words like 'pulsar','shard','gyroscope','peregrine'....
Car companies take care to have an imaginative and memorable name for their new product, I'd like to see more tech companies doing the same thing.
The demo system Intel is showing at E3 features a Core 2 Extreme processor, which, judging from past pricing strategy, will cost slightly over $1000, as well as a Quad-SLI graphics card (i.e. probably two dual Nvidia graphics cards at around $1000 each). Now, when you build such a high-end system you probably wouldn't skimp on the case ($200), motherboard ($200 & up), memory ($300 & up), power supply ($100 & up) and peripherals, either, so let's allow another grand for these things and you wind up with a $4000 PC. Put in a Blue-Ray drive (expected to cost around $1000 initially) and you just hit 5 grand.
Let's not, I don't want Blue-Ray. I don't want $1000 video cards, you can compete against PS3 with far less. You are effectively creating a gold plated PC that no one really goes shopping for, a tactic once commonly employed by Mac advocates. It was a bogus tactic then, it still is now. Peripherals, are well peripheral. You components are inflated. You can do the job with a $2500 PC and that is with name brand components, Antec case/PS, Intel mobo, Plextor, etc, and of course that $1000 CPU. And of course using today's prices. If you wait for when PS3 ships you could probably do just about as well with a $300-$500 Intel dual core, so we're really talking about a $2000 contempolrary PC.
The PS3 still costs a lot less, but now it is a reasonable comparison. Now folks can argue about practical issues, like will they get $1500's worth of value out of the computer with respect to non-gaming activities.
You'd have to pry my mouse and keyboard out of my cold, dead hands.. but on the other hand, this Red Steel trailer does make the first person gameplay with the Wii controller look pretty damn fun. (Then again, having to freeze time to tag each location to shoot the pistol is a cool effect, but it just makes me think how with a mouse you'd have those 6 shots off before the Wii controller had even tagged one of those guys)
I don't get your complaint. It outperforms AMD for the same power. Sure, it's nothing compared to the 4W 486 days when we played Doom, but seems like the right direction to me.
How does their mobile part's power compare to AMDs?
https://www.accountkiller.com/removal-requested
No one buys a console because of the power of the hardware, they buy it for the games that are on it. Therefore the PS3 will be sold to those who play games that will be exclusive to the PS3.
"I use a Mac because I'm just better than you are."
Code size? Memory usage? Execution speed?
These all affect each other. The binaries are different between different arches, duh! The fact is that you invalidate fewer cache lines with smaller x86 instructions, and those Intel-based Macs are way faster than the PowerPC ones.
World of Warcraft on my friend's 15" MacBook Pro blows the doors off my 12" Powerbook G4 of the previous generation. We're talking 20fps average vs. 50!
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
Also expectations, apparently.
Back in the old days (did I just say that?) we'd look at a 20% performance improvement, realize we probably couldn't tell the difference without a stopwatch anyway, and wait for the next 200% improvement. Now 40% is the "best year ever?" Wow.
The thing is that 40% faster processor is going to be $500 tops. If you're processor limited, it's going to be a better value to upgrade the cpu.
Physics and AI are all done on the CPU and they are taking more and more a toll these days. At the end of the day, if a new cpu is going to give me more frames per second, than its going to have value.
Just because X86 can do something in 1 instruction does NOT mean it can do it in one clock cycle, though. X86 instructions can take a variable number of clock cycles to complete, which makes for a very convoluted architecture. This is why those instructions are quickly translated into something more manageable internally.
To say that the binaries should be more compact though, is correct. cache is cheap though - this is why loop unrolling is generally considered an optimization, not a hindrance these days.
Jeremy
I think the 386SX only had a 16bit memory bus (although still 32bit internally). Similar to the way the 8088 was an 8086 (16bit) with an 8bit bus. The 486SX was the one that was "screwed up" in the factory by having its math coprocessor disabled (available in the 486DX).
-matthew
"THERE IS NO JUSTICE, THERE IS ONLY ME." -Death
I guess it's easy to have a lot of money when you can save on rent from living in your mom's basement.
If we appear to seek the unattainable, as it has been said, then let it be known that we do so to avoid the unimaginable
I'm not a compiler guy, but IIRC PPC Mac OS X binaries are compiled with -Os (ie: optimized for size), or that was what people said. Is Apple using the same options for Intel, or are they even using the Intel compiler instead of gcc?
I've this personal conspiration theory that when Jobs realized that IBM was not going to care anymore about laptops and Apple in general CPUs anymore (the real reason why Jobs switched to Intel - read the second paragraph of this interview) he planned a swtich which would harm IBM and PPC as much as he could, and misoptimizing PPC binaries would be an option.
(of course this is just conspiration, Apple may have been using -Os precisely to get more performance)
However, if we look at instructions per cycle, those transitions were less impressive, especially the Athlon that got thrown in there. There was some added efficiency, but MOST of the gain was from the basic fact that a process shrink (and a core adjusted to make use of it) allowed higher frequencies. The frequencies are certainly possible today, but the added performance is too low to make sense at the enormous cost in power (and its closest friend, Joulian heat).
Yeah, but if you just would take two Banias cores, overclock and overvoltage them enough to reach the same frequency, you would still have LESS processing power and a comparable total heat dissipation. The architecture has been widened as well. I'm certainly looking forward to getting a Merom machine sometime next year, unless AMD manages to turn the mobile market around completely.
True, 1 instruction does not mean 1 clock cycle. However, 1 instruction means only 1 thing coming through the decoder into the micro-op cache, and more predictability in collisions. These are speed gains for the 1 instruction implementation.
Now older style CISC chips (we're talking really old here- early 80s style) didn't allow multiple cycle ops and the length of the cycle was defined by the longest instruction. RISC was a speedup then, because the instructions that did more tended to take longer. Thats probably where the idea of RISC=faster comes from.
I still have more fans than freaks. WTF is wrong with you people?
Let's see... the current P4 EE ++ super-great processor is listed at $1299.95 (Canadian) at the local discount place. So presumably this new processor, that's 40% faster, will be released at the same or greater price point. So actually MORE than $1000, but it's a nice round figure.
Physics and AI do happen on the processor (well, people are interested in doing the physics on the video card too), but show me an example of a game that is CPU limited on Intel's current top of the line processor (and thus would benefit from the extra 40% they're hyping).
Hm. I knew it was one of those.
Incidentally, the 386 DX 40 was the one AMD made. Intel was rather peeved at them for licensing the design and then making it run faster than the fastest Intel chip.
I wonder which of the 100,000,000 PS2 owners Sony hopes to sell the new console to? Only a quarter of them is more than Xbox's market share. Some forcasts suggest that Sony can lose 20% of it's market share. That's still more than Xbox AND Nintendo combined. I think that's who they hope to sell to.
I suspect these new processors will likely be as expensive as a PS3, JUST the processors. Who exactly is the group with too much money?
Keep in mind this represents their NEW products not current. If they get to retail first they'll enjoy that success [e.g. lower watts/mips] but they're still a ways out.
Tom
Someday, I'll have a real sig.
SSE/MMX vs. AltiVec. Since the instruction set isn't orthoganal, more conversion instructions are needed then otherwise.
Inventions have long since reached their limit, and I see no hope for further development.-- Frontinus, 1st cent. AD
cache is cheap? Hahaha. You know that 1000 dollar Opteron? Look at how much is cache :-)
As for RISC vs CISC that argument is long since dead. After the decoders the processor is effectively RISC. While decoders are not trivial the ALU, AGU, FPU, LSU and other units account for more logic than the decoder. Look at things like PPC or ARM. They're both RISC yet their IPC is LOWER than the K8 core. Even on PPCs which are superscalar.
Granted there is no reason why you couldn't map most of the macro-op control onto the PPC instruction set the market is just not there to spend that much money on it.
In the pre-686 days the processors were entirely CISC. This made for short pipelines and slow execute units that couldn't ramp in frequency. But that pretty much went to the wayside when decoders came about in the x86 world.
Tom
Someday, I'll have a real sig.
The 8088 was designed by Intel first, it had 16 bit internal bus and 16 bit external bus. IBM wanted a lower cost processor, so Intel shrank the external bus to only 8 bits while leaving most of the internals the same, thus creating the 8086. So while the 8086 the first to be used in an IBM PC, the 8088 was designed first and was superior. Incidentally, if Motorola had been faster making a cost reduced 68K then IBM PC's would be using Motorola!!
... and in the DRM, bind them.
Well, I don't know what individual apps are built with, but XCode defaults to -Os and uses gcc for both architectures. Using those settings, the .o files are actually about 20% bigger on PowerPC for small things like SimpleText, contradicting what I said earlier. But I've definitely seen apps where the ratio is opposite, and most Universal Binaries claim bigger RAM requirements on the Intel side.
I generally don't hold with conspiracy theories, especially since Apple wants to keep the sales going during the transition. But it's definitely possible that things will get even better on the Intel side as the tools improve and developers get used to it.
E pluribus unum
Actually, power = heat/time since heat is thermal energy (disordered kinetic energy) and power is energy per unit time.
You really should stay awake is physics class. After all, knowledge is power.
Or perhaps they're concerned with more than video game performance?
HT links for example give more than 40% boost to most SMP setups compared to a shared FSB setup.
Dual channel memory, etc, etc, etc.
I think you'd find that a 50Mhz K8 would achieve a higher IPC than a similarly clocked 80486 core. Heck I'd bet you could pit a 100Mhz 486 against a 50Mhz K8 and still have the K8 win, specially with FPU intense code.
Tom
Someday, I'll have a real sig.
It's (apparently) 40% faster than the next-fastest chip, which means that it probably is more than twice as fast as most people's current PC. It's certainly more than twice as fast as my PC, anyway, which is an Athlon XP 2100+ (although it might not be twice as fast as my Intel iMac).
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
The funny thing is that you can get a faster CPU that uses less electricty and puts out less heat resulting in a quieter system. You just need to do a little research.. I don't know what the specs for this new CPU, but I would guess it probably doesn't fit that bill.
The Wii controller might be good for shooting games, yes, but I think it'll be a different enough experience from a mouse and keyboard that the two control mechanisms will appeal to different groups of people.
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
RISC makes the whole concept of pipelining much easier to implement, though - its probably the main reason why X86 chips aren't internally CISC anymore.
Jeremy
next-generation X-boxes are a fantastic hype-o-thetical benchmarking tool for as long as they're unavailable. In the mean time, just to not confuse people with the xbox 360, benchmarks should be performed on the PS3. In related news, their benchmarking tool includes Duke Nukem Forever and its follow-up.
molmod.com - computing tips from a molecular modeling
Swap your 8086 and 8088... The 8086 was the 16-bit version that came out first, and the 8088 was the 8-bit external bus version. The 6 corresponds to 16 and the 8 corresponds to 8.
So what's your point? Even if it is a ways out, AMD has nothing to show for the next 1.5~2 years. AM2 is their next offering (which slipped a full year), and it's a dog. See anandtech's site.
https://www.accountkiller.com/removal-requested
The Intel Core Peregrine Duo versus the AMD Pulsaron64 X2 !!!
:-P
The most exiting round of CPU battles yet!
This sig rocks the casbah.
Many people decided to buy the PS2 in part because they needed a DVD player.
Saying Java is nice because it works on all OS's is like saying that anal sex is nice because it works on all genders.
"No one buys a console because of the power of the hardware, they buy it for the games that are on it. Therefore the PS3 will be sold to those who play games that will be exclusive to the PS3."
*Sigh* If only this were true.
What really happens is (some) people look at specs and think "this system will be cooler, longer." They're betting that some time before the system dies, a bunch of woweee-zowweee games will show up.
I'm not saying everybody buys systems this way, but I've read quite a few comments on several forums where people have based their purchasing decision on their interpretation of the specs. Oh, and there isn't universal agreement here, so the 360 and ps3 fanboys don't get along too well. Hehe.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
No one buys a console because of the power of the hardware, they buy it for the games that are on it. Therefore the PS3 will be sold to those who play games that will be exclusive to the PS3.
Not necessarily. I will buy a PS3 simply because of the power of the hardware, and I'll like to see the intense graphics and good gameplay on my HDTV.
My last console was an Atari 2600.
My point, is that there is a blend of hardware and software. Often, hardware leads software, but with OSS, this has changed.
Macs may of had a better GUI before OSX, but I could not ever get into them because the OS was unstable. Maybe if I spent some time and learned the workarounds like I do with current buggy Mac OSes (Tiger is buggy!), but with unprotected memory and having bomb error messages all the time, Macs did not appeal to me. Neither did Windows ever really appeal much to me aside from Win95. Honestly, that was a pretty decent OS at the time. I started using Linux in 94, but I used Win95 up to about 1997, and after the last time that Windows got hozed, I realized that Linux was better, despite some of the creature feature apps that was available for Windows.
Used Linux for a while, and then Apple started getting a stable release of their GUI with real UNIX under the hood (OSX 10.3).
Even at $500 or $600 a PS3 will be the best gaming platform that I could buy without spending $2k+, and then I would have to endure Windows. For $5-600, you get dual display HDTV resolution graphics, and good 3d graphics and good sound and very likely good software. A mid-range video card costs this (high end is currently between $1,000 to $1,600+). A moderate sound card costs this (high end is between $400 and $1k+). A good monitor costs this or much more.
I'm a strange "early" or "late" adopter. I early adopt stuff as soon as its either 1st to market or clearly the best bang for the buck for a multi-use system. These new PS3s are going to be amazing.
"Intel also claims that a Core 2 Extreme-based enthusiast PC will leave the pixel power of a Playstation 3 in the dust."
Even if technically true, given the 'lowest common denominator' style of PC games, would that really mean a flippin thing for a year or two?
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
Even Grammar Nazis have been outsourced. j/k
Nonsense bring it on and add extra adjectives. For example the new Gell-itte Super Mondo Xtra Extreme shaver now with 20 blades!!!!!!!
Tired of all the isms, don't exploit people as an employer, or a government, mmmmK?
Ok so the clock speed rocks. But does the rest of the system keep up? The big advantage I see with AMD is Hyper Transport and the newly ratified Hyper Transport 3.0. You can have a THz CPU but if you can't feed it data/instructions it's just going to waste most of it's potential.
I'm not familiar with any possible new bus technology coming out with the new Intel CPU's, but based on my current experience with the latest Dell boxes (Intel) and our new Penguin Computing and HP AMD boxes Intel has a lot of catchup to do to outperform AMD and their whole architecture.
We are using these boxes as MySQL database servers with each server containing 100+ 500 MB to 50 GB databases attached to fiber channel disk arrays. These boxes are mostly doing I/O, but a fair amount of CPU is used for sorting/math done at the database level. The AMD boxes smoke the Intel ones.
Unless Intel also releases a whole new architecture that can compete with Hyper Transport the extra speed will most likely be wasted.
Sometimes my previously abandoned house makes me wish I lived in my mom's basement.
I know. I killed the joke. Sorry.
I have one of those - $4400 actually; it's a PII450. Has 128mb & it's still all original, from the paint to the case screws.
Pi Ran Out
Well then, you're either the exception that proves the rule, or more likely, a Nintendo fanboy being a smartass.
Most games have a capped frame rate in multiplayer mode anyway, so even with these high end chips it won't matter (unless it's a major upgrade from your previous cpu). You're better off upgrading your graphics card over a cpu if you want to improve your games performance.
UT2004 is capped at 85fps in multiplayer mode. There's no cap in singleplayer.
Doom 3 is capped at 60fps in single and multiplayer IIRC.
There are other games out there too that have a framerate cap. With my system, Doom3 is always at the 60fps cap and UT2004 is always at the 85fps cap in multiplayer, upgrading my cpu, I wouldn't notice any improvement. I'm sure these chips are lighting fast and all, but are they worth it? A high end graphics card will improve game performance alot better than one of these chips would, and would only set you back maybe $500 instead of over $1000.
I just built a gaming rig (partial specs in sig) and the total cost was around $1200 Canadian after tax. So for the price of these high end chips, you could build/buy a complete system. Hardly worth the money IMO.
The best part of FPS games on the PC are not the keyboard and mouse, but the
Buy a console FPS, and you get that FPS, sans headaches and compatibility problems. But also without upgradability.
The ______ Agenda
Even Grammar Nazis have been outsourced.
But, hey, the grammar Nazi Nazi is still a good old boy. Which makes me a grammar Nazi Nazi Nazi. I'm not telling where I live.
How did outsourcing get such a bad name in the first place? Reproduction requires outsourcing. I thought most people liked it that way. I mean, insourcing is a great just-in-time tactic for a few years, but it starts to seem, I don't know, a little bit xenophobic and self absorbed?
Comment removed based on user account deletion
I suspect these new processors will likely be as expensive as a PS3, JUST the processors. Who exactly is the group with too much money?
These processors will cost ALOT more than the PS3. The article says this new Intel Core 2 Extreme processor will be 40% faster than the Intel Pentium Extreme Edition 965. The 965 costs over $1000, this new chip will be 40% faster, so you can imagine just how expensive it will be.
Usually, that's not what happens for us. Generally, when our performance goes up, so goes the power.
Oh yes, quite a claim there. You could limbo under the Prescott power budget on stilts.
Now write some games that don't suck.
> You'd have to pry my mouse and keyboard out of my cold, dead hands. ;)
That's exactly why I prefer console gaming
I don't want to have to pry the mouse out of my twisted, crippled RSI afflicted hands.
Not to say controllers don't have their own RSI issues, but at least then I'm varying the actions I'm performing with my hands between work and gaming.
Advanced users are users too!
Hey stop that! I trademarked that clue long before *you* got to it.
What's interesting is how the converse stupidity was established in the first place. Proponents of the RISC camp were so busy announcing their imminent victory they forget to extrapolate more than one or two die shrinks into the future. A design parameter that makes you pull your hair out today can evolve into a subtle advantage a few shrinks later.
I'd rip apart how x86 handles the flag register (partial register stalls and truly Byzantine retirement logic) long before I'd replace the x86 instruction encoding format (except perhaps the prefix bytes, which are super annoying for decode alignment). The stack based floating point register file and instruction set is one of the few x86 features that was beyond redemption at any scale (do I sound like Ralph Nader or what?)
The bottom line is that the flawed nature of the x86 instruction set had a much bigger impact on thermal performance (MIPS/watt) than it ever had on top speed. Now that thermal performance determines top speed, that old shoe is beginning to pinch in new places.
Even at $500 or $600 a PS3 will be the best gaming platform that I could buy
If you want to play Halo 3 or SSB3 it's not. It's only a good buy if it has games you want.
"I use a Mac because I'm just better than you are."
If you've already hit a framerate cap then upgrading your video card isn't going to do you much good either. If you're NOT at a cap though, then I agree, upgrading your video card is going to give you much better performance than having the bleeding edge CPU.
There was some added efficiency, but MOST of the gain was from the basic fact that a process shrink (and a core adjusted to make use of it) allowed higher frequencies.
This is one of the most bizarre intellectual shell games to ever make common currency. The reason this new rocket goes twice as fast as this old rocket is that we burn the fuel twice as quickly. All the rest of the expensive engineering was to ensure that the rocket doesn't blow itself up on the launch pad coping with all that thrust. But the real technology was burning the fuel twice as fast.
90% of a modern CPU design consists of circuit elements to conceal latency from the frequency obsessed execution units: the branch prediction unit, the parallel instruction decoder, the L1 cache, the L2 cache, and the out-of-order execution engine. It costs electrical power to feed all those circuits to hide those latencies. Which is why cycles/watt have increased far more slowly than cycles/second, despite the fact that power is reduced to a first order as a function of voltage squared.
Actually, back then you could see a 20% difference. I could easily tell the difference between a 486/25 and a 486/33 by how fast it listed the files when doing a DIR command, all other things being equal. But now give me a 2500Mhz system and a 3300Mhz system and I would hard pressed to tell them apart without resorting to benchmarks of some kind.
Of course, just because I could tell a difference didn't mean I immediately upgraded my hardware.
You are so backwards. The 8088 was the cripple-chip used in the first IBM PC.
http://en.wikipedia.org/wiki/Intel_8088
Wang made a PC clone with the 8086 for a while. It was lacking compatibility on a number of fronts, so it died of a not-so-slow terminal illness. It ran about 60% faster than an 8088 PC at the same rated clock speed.
It's amazing that IBM looked at the 8086 and decided it needed an additional whack to the kneecap. As if it wasn't hobbled badly enough already.
Here's one small defect that greatly enhanced the Propecia sales among a generation of early adoptors. The segment registers were designed with a four bit offset from the address registers, yielding a 1MB address space.
If that offset had been six bits instead, the machine would have had a 4MB address space. Each of the eight ISA slots could have had a dedicated 64K address mapped I/O space (consuming 512KB). A 4MB address space would have nicely bridged the evolution to 32-bit multitasking operating systems, which become viable somewhere between 4MB and 8MB.
That four bit offset cost the industry billions of dollars that could have been far better invested elsewhere. There was precious little advantage to a 16 byte segment alignment granularity that I ever noticed.
We'll have to wait for impartial benchmarks to be sure, but intel's tests showed a 20% boost in FPS on average in games, with FEAR benefiting the most.
I'm not an Intel bootlicker, but more frames are more frames, and it appears that there will be an advantage performance-wise with Conroe. I'll happily buy an AMD solution if trounces Intel.
In conjuntion, Intel's procs are typically cheaper than AMD's. Fortunately for AMD, they are doubling their manufacturing capacity so they can compete on price for a while until they work out their 65nm process.
But I totally see your point about GPU dependency these days. It's reached the point the only way to challenge these high-end systems is to output at ultra-high resolution. Talk about a need for innovative gameplay.
I'm surprised that a faster processor makes that much difference. I'd be thrilled if somebody came out with a game that used all the processing power for AI, but I suspect it's far easier (and sells better) to just up the polygon count.
Well, for one thing, high power consumption results in a lot of wattage dumping into the room. It also runs up the electricity bill. It demands beefier PSUs, which must be larger. Larger PCs take up more room, which may not be a problem, but may be, especially in smaller abodes. High power output also creates thermal hazard problems, as real as cooling failure resulting in components melting and catching on fire, etc.
Right now, Athlon 64 X2s wipe the floor with P-Ds, in both power consumption and performance. Why settle for less? I've got a 3800+ X2 @ 2.4GHz and it consumes very little power for such an amazingly fast chip. It runs off a 220W PSU in a low-profile case with no problems, fan speed on half. A Pentium D would struggle with such a setup, that I know for sure.
A conroe (Core 2) may wipe the floor with AMD. In fact, it probably will. But you know what? That's good, it pressures AMD into releasing K9/reverse SMT chips sooner. A little competition never hurt anyone.
$60 dual core chips? Sooner than you might think...
Do not downmod posts "overrated" simply because you disagree with them.
What do you live in, the 80s? The disks in this clunk de munk run at 7200
Do not downmod posts "overrated" simply because you disagree with them.
I'm glad I'll finally be able to play solitaire at 800 FPS.
"Gamers, this appears to become the most exciting year for you in a long time!" Games on the PC are for 99% of the time restricted by the video card, only in rare cases the CPU will actually be the bottleneck, if you run a game at higher details (AA/AF) and resolutions you will put more strain on the VGA card, and an increase in CPU power will not translate into a boost in performance worth mentioning.
You truly are a broken record. The K8 architecture is over two years old. Lots of things can happen in 2 years, including 1) Intel coming out with a processor with new power-saving and performance features, and 2) Intel moving to 65nm processes. Please stop comparing Intel's new processors with 2-year-old AMD processors as if AMD's were brand new as well.
I've already given the performance and performance/watt crown to Intel (and to you, in a response in a previous topic) for the next year or two. Will AMD come out with something competitive in the next 2 years? We'll have to wait and see.
This silly -Os conspiracy is starting to annoy me. -Os is actually quite a lot faster in most cases. For the total borderline cases where -O3 is faster, you're supposed to profile and change it manually. -Os has all the optimisations of O3, except for those which bloat the code unnecessarily, such as 16-byte alignment of loop headers. This type of "optimisation" bloats the code and makes it _slower_ in most cases.
The idea that Apple uses Os to make IBM look bad is totally ridiculous.
"Hell, it'll also fill my wallet with dust with it's pricetag ALONE!" lol :)
Us REAL gamers have become pre-market suckers.
Ever-vaunted realism is cycles per second, lapping at the shore of capability, GPU aside, what has DX done for us lately? If I could donate every wasted or pointless cycle to BOINC I bet the world would be a better place. I dunno, I'm no scientist, but I think I know when I've been ripped off. Direct X is a software manipulation, GL is a driver fabrication and "anything else" is new-keyword-bullshit. I think I'll just stop playing these silly games and go Linux, Ubuntu-style.
I'm not even sure about the more difficult to debug thing.
I'm a fan of RISC, really, mostly because their architecture is more sane and it's much nicer to write a compiler for them, but with increasing chip size CISC has several advantages.
The decoder is an ever smaller portion of the whole chip; the innards use basically the same optimizations as RISC chips do. As you say, instruction size is smaller, which is good. Most of all, even multi-cycle-x86 instructions get translated into micro-ops (just like on some RISC implementations) so in the end they ARE micro-ops. It's only a compression at the programming level. Deep down in the core the same stuff happens.
I think x86 (or maybe its 64bit extension with more nice registers) has won. Apple got that.
The exception, IMHO, would be embedded, or mobile applications, where the decoding logic is still important. ARM is a great architecture, and it also has "compressed" (THUMB) instructions, to reduce code footprint.
For desktops and servers expect nothing better than x86/64.
Hey, let's be fair. This is an exciting year. The Nintendo DS is becoming cheap so even gamers who'd laugh in your face if you asked them to shell out 200 bucks for a piece of gaming hardware can buy one. That's the first non-Gamepark console since the original Playstation that's actually worth buying!
;)
Let's face it: All serious computer gaming was done under MS-DOS, Direct3D is for weenies without imagination.
By the way, apparently I can't post anything with a subject line of "Score: +1, Retrogaming Elitism". A Slash bug?
USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
That should've read "Intel Yugo Dual Extreme Edition". Although an Intel Yubo would be cool, too... Powered by Karavan technology.
USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
Intel is hedging their bets. That's why they still produce so many Netburst based cores while researching a completely new core, etc, etc.
The reason for sticking with x86 is only because of legacy software. If it cost you 5% of your die space to get a fast decoder and the rest is cache/execution parts then its worth it to both the producer and customer.
Think about it. For your 150$ processor 5% of the cost is $7. For that you get an efficient processor that will run all of your existing software without having to rebuild it.
Granted in the OSS world that's moot. So long as gcc has been ported to the ISA one could port Gentoo to it.
Tom
Someday, I'll have a real sig.
DDR2 support will amount to more when DDR2-800 comes about.
... you don't think AMD doesn't have platform changes in the woodworks?
And even though next years cores are looking better from Intel they still lack HT links and independent memory controllers which are critical for any sort of MP setup.
Also
Tom
Someday, I'll have a real sig.
Ok I've been keeping abreast of the whole new Intel chip, but readin the article does anyone else get a de'javou feeling? I seem to remember AMD bringing out their AMD 1700-2200xp chips with their low clock rates, and telling us performance mattered not clock speed. Intel then got carried away upto 3.7Ghz, telling us we needed faster and faster chips that cost 3 or 4 times the AMD ones.
AMD is finally ahead and while I don't do any network service work or place my PC understrain I've found the AMD64 3700+ does everything I would want, it doesn't slow down it runs everything very quickly and every game I own runs maxed out wiuth a decent frame rate. The one bottle neck in my system now seems to be windows itself (xp64 is an awful OS) and the hard drive access times.
Has Intel got lost in the 'MHz = l33t' attitude again? Or have they come up with something that competes with Hypertransport yet?
Well outside the MAC, Servers and your elite gamer can anyone see this chip selling? Its unlikely your average joe is interested eith the price tag being so high. The elite gamer seems to be a dieing breed, and does anyone know its performance against a AMDFX 62??
As for the 'who will buy a Sony?' comment, I will be, once the top notch one drops to around £200. My PS2 cost that and the years of enjoyment its brought me, coupled with the out of box games like singstar,buzz off etc show sony are thinking about me and my mates. It comes with Blu-ray since the 360 seems to be trying to pass itself off as a entertainment hub its lack of next gen drive is very very poor. Sonys Blu-ray, and general 'hype' seem to sugegst it can be a true entertainment hub that happens to play fantastic games. No i don't have a HDTV nor do I plan to buy one, but I'm pretty certain that the Movie industry is going to force this on us so would rather save the cash now and get it in my console.
the 20'000 aren't necessary in the same room to make a difference.
:it'll lowers their global bill).
If the 20'000 hardcore gamer that are living in the same city as you use 50W CPUs instead of 150W one, maybe they'll get 3FPS less on the screen, but on the other hand will consume 2 MW less.
Which will put less strain on the power plant, will be more ecologically friendly and will contribute less to the global warming.
(That's why enterprise are also interested in more eco-friendly CPUs
Same reasonning also goes for the SUVs. Appart from having a bigger toy than you neighbours, what's the point in using a vehicle that consumes more gaz and is less secure in case of crash than everyone else ? You're ruining the environnement with your bullshit. Not you as a signle person, but you as a member of the greater group of dummies who feel obligated to drive the car that looks the biggest and most impressive.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
I'm so glad that AMD became a powerful player in the desktop PC and server market... not because I love AMD but because now we are really seeing some earnest competition and innovation. Before, we were happy with Moore's law, but then AMD beat Intel to 1GHz and the ensuing struggle for mind and market share has brought about some truly phenominal changes.
Keep up the excellent competition... maybe we can have a third player jump in with some new ideas? IBM? Sun? Let's see you what you have...
"...on paper, they can generate a lot more polygons with that product than we can with a single graphics product. I think some of these quad-graphics systems can get there now."
Its a non-story. A core2 with a quald sli setup they "think" can match consoles now? Lets see. Thats a minimum of a $4000 system compared to a $399-599 console.
The real question is why are we paying so much for the same power in a PC or getting so little power for the same price (you can't really game on the $399 PCs)?
I'm getting an EXTREME RASH from the use of the word EXTREME. But the other possible marketing buzzword candidate was taken.
The dangers of knowledge trigger emotional distress in human beings.
God isn't that the truth. I think the AI issue is what is holding up S.T.A.L.K.T.E.R. (http://www.stalker-game.com/index_eng.html). It's been delayed for almost 2 years now and is trying to create a real ecology of AI entities. They were trying to create AI that just lived and even interacted with other AI in a procedural way.
Because of this innovative goal, I think it must be cause of their numerous delays.
It's no wonder Intel is working closely with major graphics engine makers - they have to keep driving demand for multi-core to sustain sales.
as if AMD's were brand new as well.
I'm not. I'm comparing the Conroe with AMD's AM2. Both are the latest offerings from each rival this summer. Benchmarks scores for both are on AnandTech.
Please try to read people's posts before bashing them.
Wait, this is Slashdot. Nevermind.
https://www.accountkiller.com/removal-requested
Not true. The memory controller onboard HURTS MP performance, it adds extra latency (beyond 2~4-way) because MP systems have their own complex memory subsystems. Not a smart move from the current king of server chips.
AMD changes? Yeah, they do: they have a new arch that is 1.5~2 years out, but until then, they just have a diff. memory subsystem.
But you're right: they could have more and are just tight lipped. I mean, no one knew squat about this Intel thing until a few months ago.
https://www.accountkiller.com/removal-requested
Conroe is a next-generation architecture, built on a next-generation process. AM2 is K8. You totally missed the point I made, but that's okay, this is Slashdot.
"Not true. The memory controller onboard HURTS MP performance, it adds extra latency (beyond 2~4-way) because MP systems have their own complex memory subsystems. Not a smart move from the current king of server chips." ... flamebait ...
This is total bullshit. Go learn what NUMA is. Software written for NUMA can be much more efficient than that on a traditional FSB setup. Also the FSB approach doesn't scale. As you raise nodes the shared FSB has to get slower and further divided. Memory bandwidth on a 4-way Xeon is MUCH lower than that on a 4-way Opteron.
"they could have more and are just tight lipped"
Changes to the core don't happen overnight. Opteron has gone from 1.6Ghz 130nm cores to 2.6ghz [and higher] dual core 90nm cores. That alone is a huge process change and upgrade.
I can't really say anything about plans but just suppose that AMD isn't totally asleep at the wheel. If you were halfway objective you'd learn what makes the Opteron approach efficient instead of just spouting ignorance.
Tom
Someday, I'll have a real sig.
What Gamers? The Sims 2 Game will not play on Dual-Processor Machines.
My daughter has a Dual PIII 900MGz PC and SIMS will not play. It deadlocks right on start.
One good thing about it is that now she knows how to Go to Process-Explorer (Sysinternals replacement for TaskManager) and set the Affinity of SIMS2.EXE to use only one CPU. And we had a long talk about multi CPU/CORE computing and how it is done. And we had a long talk about computers in General.
But still a company like SIMS2's releasing a one CPU game. I can't even start to describe what I feel. Shame on you SIMS2.
Free Life
Boaz
NUMA isn't even part of the discussion. Using die space for a redundant memory controller is pretty stupid. Think beyond 4-way.
https://www.accountkiller.com/removal-requested