Slashdot Mirror
AMD and Intel Update CPU Roadmaps
vincecate writes "Recently
AMD updated their processor roadmap. It shows their move to 90 nm and has a range of new processors over the next 1.5 years, including dual-core chips. An
unofficial AMD roadmap shows speeds and performance increasing.
Intel also recently updated their roadmap.
Intel does not show anything faster than
the current 3.6 Ghz in the next 11 months, including the recently delayed 4 Ghz chip, except to say '3.6 Ghz or greater.' Strangely, some of the recent SPEC benchmark results show the 3.6 Ghz chip to be slower than the 3.4 Ghz chip. One possible explanation for this is that the 3.6 Ghz chips will slow down due to 'thermal throttling' if you are not very careful to keep them cool. So it seems like heat may be the reason Intel's roadmap does now show much improvement."
Well, why not just make water cooling mandatory for new CPUs, just like Apple did?
The clock rate of the CPU went up madly through the 90s but the wind appears to have gone out the sails a little. Is the actual speed of the CPU still climbing but they're doing this without adjusting the clock rate?
:P
Don't really keep up on the hardware these days..
Cheers,
Simon.
An unofficial AMD roadmap shows speeds and performance increasing.
And here I was, afraid that they had decided to not increase speeds and performance. That was close.
I just got an MSI K8N Neo Platnium, which is a socket 754 motherboard. Looks like socket 754 is going no where.
The reason the 3.6GHz processor runs slower than the 3.4GHz processor is because they're different processors, not the same processor running at different clockspeeds. Just look at the die photos (www.chiparchitect.com) and you'll see what I mean. The idea is that the new processor will scale to higher clockspeeds which it, uh, already has. (Just look at the "OC records": nobody got an old Pentium 4 beyond 4GHz with standard HSF cooling - nobody. On the other hand, this is more or less straightforward with the new Pentium 4s.
What I don't understand is why more people aren't building Pentium M desktops.
Kind of but not really... There was a time in the nineties where if you waited two years you could get a system at least 2x as fast.
I built my system about two years ago (actually it's a few months short of two years). AMD would have to release the equivalent of 5600+ within a few months to match the speed of the 2800+ they released almost two years ago.
If they were a few months late that would be normal but it looks like it will take far longer.
Hmmm... Pie...
As I wait for the skin to grow back on my eyes from this horrible colour scheme, I can consider the information in the story summary.
We're obviously starting to see a convergence between the industrial processor market and the end-user one. I mean three years ago you would get a dual 3.2GHz (1.6 * 2) system to host a medium sized website, and that kind of horsepower is probably still adequate today. So what kind of apps (I mean, apart from Doom 3) do end users need this kind of grunt for? 3GHz? 3.6GHz? 4Ghz?! If Architects could use AutoCAD 2000 on a 950MHz cpu, without complaint, what has changed? Obviously a speed increase is nice, but three or four times that?
Are we going to see a point where the convergence turns to over taking, and end-user CPU's need to be faster than a lot of corporate stuff?
p.s: I'm aware of shit.slashdot.org, no karma whores please.
--
The last digit of pi is four.
I've noticed this. I got a 1GHz Athlon a few years back and it doesn't seem to be much behind the latest Athlons (especially when I count my athlon's overclocked speed). My previous machine was a 100MHz pentium and that seemed to go out of date really quickly.
Are the new processors really much faster?
Don't get me wrong... I still love my Laptop! :-)
The important concept to keep in mind is that all these computers are powerful enough to do what I need them to do, so merely making CPU clocks tick at a higher rate isn't going to persuade me to run out and upgrade.
Well, it's a matter of opinion but, imo... yeah they are.
But then and again your upgrade cycle seems to be longer then mine. I think it all depends on whether or not you want to run some apps that require a more powerful system or actually runs them a good deal faster.
Hmmm... Pie...
I have always thought AMD is better than Intel (price/performance, no annoying jingle, no annoying "... inside", no "MHz myth"), but now it seems Intel is getting its arse kicked so much I worry AMD might get too complacent.
More people than me seem to have noticed that clock speeds seems to have stalled. I don't necessarily see this as a bad thing - as computers has grown fast enough for me lately. I'm still content with my 1.3Ghz Duron.
.. specialist tasks.
What I personally really, really want to see is cooler CPU's. CPU's that doesn't require a huge fucking fan. CPU's that are content with a heatsink would be nice.
Furthermore, I would love it if Dual configuration became more widespread (and thus cheaper). Personally I would love a multi-CPU machine far more than single-CPU ones.
My personal wishlist:
- 64bit CPUs to become the norm (seems to be happening).
- Cooler CPUs, not requiring fans (seems to be happening, look at the VIA EDEN CPU's)
- Dual/Quad/Multi -CPU configurations becoming the norm in home computers.
I don't care much whether single CPU's grow much faster at the moment, as there doesn't seem to be applications requiering it for regular use. There are of course specialist tasks that require more horsepower, but those are
"Rune Kristian Viken" - http://www.nwo.no - arca
GPUs are not powerful, they are just very efficient at matrix manipulations and calculations related to graphics operations. You can use this to increase performance in some situations, but 99.99% of the time you'd just bottleneck yourself.
...but maybe the cheaper PCs cannot?
Also, a liquid cooler is probably inherent harder for Intel to package with an OEM processor. Affixing a liquid cooler to a processor requires more case aware design than simply clipping a fan to a mainboard socket.
Last week I benchmarked the 2.2Ghz Opteron on 64 bit Linux 2.6 and Java. I got almost three times the performance of a 3Ghz Xeon. For details see http://gregluck.com/blog/space/start/2004-07-29/1# AMD64,_JDK1.5.0_and_Linux_2.6_rock!/
I've got a 2GHz Celeron laptop that is much slower at compile than my Athlon 1.47GHz.
The Celeron is a severly crippled chip, unlike the Duron, which is a respectably performing budget processor. It only has 128KB cache, which is CPU sucide on a P4 core. The P4 needs large amounts of cache to keep its long pipeline filled. People who buy high clock speed Celeron, thinking they're getting a fast CPU are getting massively screwed by Intel. So much so it borders on being an unethical and immoral business practice. The chips are not near as fast as their clockspeed would indicate. One would be much better of with an Athlon XP, Duron, or a slow P4 as a budget processor.
In addition to this, until we start seeing widespread use of PCIe, the downstream AGP bus is still a serious bottleneck as well. Uploading data to the GPU is really fast, downloading maxes out at ~133MB/s.
I haven't had the chance to play with a Pixel Shader 3.0 card yet, so I don't know how useful for generic computation they are. It usually helps if you're trying to process many sets of the same kind of data, rather than evolving one calculation through a long or iterative algorithm.
~phil
Sick, sad, and ultimately moronic
A GPU is still only a CPU that has been optimized for multicore vector operations
eg: a GeForce 6800 is approx 10 programmable pipelines, with some entangling fixed function pipes, such as triangle setup, cache, memory and IO, etc.
6 of those cores are independant vertex processors
4 of them include 4 "pixel pipes" each, which, unfortunately, aren't actually independant, they just include writemasks to prevent unused ones from affecting what they shouldn't. if one pixel pipe slows down, the rest slow with it. Each of these pixel pipes has a 2 stage pipeline, much like the early, early CPUs(im thinking pre-pentium), and 2 execution units, which each consist of a 4 floating point vector processor with inline integer math for addressing, filtering, etc.
so basically it is 10 processors that can handle 20 instructions per clock with an average of 176 data outputs per cycle(22 "pixel/vertex pipes" * 2 execution units * 4 components per vector)
Or 17.6 data outputs per processor per cycle
compare to a prescott
1 processor, 31 stages, 1 pipeline, 4 instructions per cycle, up to 4 components per vector(SSE) = 16 data outputs per cycle, 8x as many cycles per second, 1/2 as many transistors = 128 data outputs requiring half the silicon.
therefore, a dual core prescott could easily do (256/176=) 45% more work at the same price and:
1) have a decent level of precision(64 bit)
2) have integer math
3) be much more flexible and not rely on the AGP bus
4) be a hell of a lot easier to code for
what are you whining about?
consider the concept processor, Niagra, an 8 core, multi-ghz, 3 instruction per clock, 4 way hyperthreading processor, undoubtedly with vector extensions...who needs a GPU for some shitty 32 bit, AGP-bottlenecked results when the CPU is obviously superior?
Where is the roadmap for low-power consumption chips that can operate either fanless, or with low less cooling gear?
I survived just fine on a PII for several years until recently biting the bullet and getting myself a P4 box in a Shutttle Zen XPC case (relatively quiet). I seriously considered getting myself an EPIA box as my main machine, simply because it would be lower power (therefore cheaper to run), silent and enough umph to use mutt, firefox and ssh into the server kit where the real work is done. The only reason I ended up with a P4 is because a friend had a 3GHz one going very cheap.
I want less power, not more. The idea I should overclock, buy liquid cooling systems and should pay a ridiculous amount so I can play some games? I'm sorry, what planet are you all on?
Normal applications are not as likely to drive a processor into thermal throttling as a benchmark is. It sounds like benchmarks are going to need to be rewritten to either be short enough to not cause thermal throttling or to spread the benchmark out so that the CPU has a chance to dissipate heat buildup caused by the artificially intensive benchmark code.
The AMD roadmap says it all: "As market requires". If the market says give me 5 GHz and I'll pay anything you can bet 5 GHz will be on the shelves. Right now you can buy sub $500 supercomputers that sit relatively idle. Word processing, db query, e-mail, web surfing, solitaire - most of the world goes no further.
The next market force is competition. If AMD looks like it will be selling a 4000+ Intel will match that.
Processors capable of this speed are most likely possible. There's no way Intel can sell and support 3.6 GHz without having perhaps seen 2 to 2.5 times as fast in stable operation under extreme cooling. In the lab where they can really reduce the feature sizes and power consumption who knows what is really around the bend?
Has anyone noticed the recent trend in laptop computers? It's all marketing and forcing consumers to buy crap. No floppy drive - so how do you boot if you want to install legacy stuff? Ultra wide screens but I've seen screens at 12" and the laptop has all the interface goodies (CD-RW, parallel port, PCMCIA, USB, speakers, 10/100, modem, headphone/microphone jacks). The widescreen monsters don't have any new interfacing, but are super heavy and don't necessarily have long battery life. The idea is to hook the suckers who will see the error of their ways and buy a new version several years later for the sake of lighter weight and longer battery.
Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
I was talking to my friend about this the other day, and we think that eventually they cannot go that much faster (well, maybe have a SMALL core of the chip that can go faster), and they'll start stacking in parallel instead. Ie, massively hyperthreaded processor cores. So maybe in a few years we'll see 6 GHz chips with 8 or 16 hyperthreaded processors?
We're physicists, though, not engineers, maybe there are some other clever ways to keep pushing the envelope?
make world, not war
Round wafers are easier to manufacture. Keep in mind that the the Si chips are made of is monocrystalline. When growing the wafer it is grown from the center outwards. When it hits the wall of the reactor vessel, it will probably break the crystal structure, and whatever growth that continues after that is not usable as there is a grain boundary. With a circular wafer you hit the edge at the same time.
Also, the wafers used today are what, 300 mm in diameter while the chips are something like 10x10 mm, so there's not much material lost anyway. And the leftovers are simply sent back to the wafer factory to be remanufactured into new wafers, so there's no material lost. Not that it would matter anyway, since Si is among the most abundant materials on earth.
Photons are the mediating particles of electromagnetic force, and it's definitely this force that couples two electrons together, or the electrons to the 'holes' in the doped semiconductors, etc etc. An elementary description of current in a wire is akin to a tube filled with marbles, you push one in, and one comes out at the far end. This interaction between the 'marbles' would be mediated by photons. Of course metals and semiconductors are far more complicated than this picture, but it's a rough start.
It might sound weird to you (it did to me at first), but when you send a 100 MHz signal down a coax cable, you are really sending photons. They're rather low-frequency photons confined to a waveguide, but they're definitely photons.
make world, not war