Slashdot Mirror
AMD Moving to a 400MHz Bus?
An anonymous reader writes "According to this tantalizing Infoworld Scoop, AMD soon introduce a 400 Mhz bus. Seems that SiS's big announcement at CEBIT is the SiS748 chipset, which supports both 400 MHz DDR & AGP 8X, and is targeted at the upcoming Athlon 3200+."
Argh... do i wait for athlon64 or opteron, or do I get one of these bad boys?! Decisions, decisions...
My patience is infinite, my time is not.
There have been rumors about AMD going for a 400MHz bus for quite some time now. Some chipsets even have experimental support for it. With the Athlon 64 being delayed until September I would say that is the only way for AMD to try and stay competitive with the Barton core.
Maybe I'm being a little arrogant, but I still feel this isn't really much to be that excited about.
.: Max Romantschuk
AMD better forget these little incremental speed bumps and switch to a whole new architecture this year if they want to remain competetive. The current architecture is like milking a deadhorse and they are already running waay too hot. They need to make something big enough to give it a new name.
Whale
It seems some what proper that so many of the revisions of the athlon have had horse names since they seem to keep beating it till they know its good and dead.
Is a 400MHZ bus really gonna help them all that much? How much more can this chip design take?
Power Corrupts,Absolute Power Corrupts Absolutely, leaving one person(group)in charge is absolutely corrupt.
The 4GHz is only when heavily overclocked. The fastest retail P4 runs at 3.06 GHz, which actually is about the same speed as the 3000+. The 3200+ should give a nice speed boost. Actually, the fastest-overclocked Athlon runs at about the same speed as the fastest-overclocked P4. Not only that, but both processors are at about the same price.
You computer will blow up if the processor speed-steps below 600Mhz...
No, Intel is not at 800MHz FSBs yet. They are about to move to 200MHz quad pumped, whereas AMD are about to move to 200MHz dual pumped buses.
Not only that, but both processors are at about the same price.
.. never thought I would say that!
Not to mention the AMD runs cooler than the Intel chip.
Whoa
I was actaully a little worried that when a Macs switched to the PPC970, memory fast enough for it's initial 450MhzDDR bus would be prohibitively expensive. They might have been forced to increase the bus multiplier to maintain their target price point, or they might have just needed really expensive RAM.
With this 400mhz bus and a bit of upwards evolution, this shouldn't be a problem by the time 970 based macs are released. yay
"The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
Just how much influence does the bus speed have on the system as a whole?
My CPU is running at 266mhz now, what improvment would I see if I upgraded to a 333mhz bus chip with the same clock speed?
Just curious!
The general rule of thumb for upgrading it to put it off for as long as you can, and then buy as close to the top of the line as you can afford.
We who were living are now dying
With a little patience
I've got my Athlon XP 2100+ running on a 400 MHz FSB (of course, that's overclocked)... but it definitely does seem "snappier" than the 266 MHz FSB. Certain apps seem to benefit from the extra bandwidth, but not everything.
I just recently bought an Abit based NForce2 Athlon Motherboard. I have my DDR3200 running at a pretty 200mhz (so 400mhz DDR) and my FSB is at 181mhz (so 362mhz DDR). I have made some changes so I need to try for a 200mhz (400mhz DDR) FSB again. I can tell you that just upping the FSB and your memory bandwidth can have great performance benefits for memory intensive apps (such as gaming). So this will be a great boost for the current XP line. Oh, and in case anyone is wonding, I have an XP2100+ (1.73ghz) running very nicely at 2.2ghz!
Unstable Apps: Our Android Apps Don't Suck
Huh? they're talking about bus speed. Intel is about to reach 200Mhz quad-pumped (800Mhz) and AMD is about to reach 200Mhz double-pumped (400Mhz).
If you don't know what you're talking about prevent yourself from posting.
I usually try to help my friends build their computers. The key things to remember are: the best time to build a computer is as long as you can hold out, because the longer you wait, the faster and cheaper your hardware is going to be. Also, look at pricepoint. You know how Walmart has those labels on their price tags that say "17.8 cent/oz" ? Do that with your hard drives. Right now, 120gb drives are $100 on pricewatch, 160gb drives are $160. 120 has the best price point. Athlon prices increase about $10 each until you get to 2200, where they jump by like $40. Keep in mind that if you build a computer for a reasonable amount of money you can do it more often if you choose to and you'll never finish building one and be unhappy with the speed.
Whale
I can choose a "top of the line" system, or a system that has 75-80% of the performance for half the price.
As a result, it's cheaper to buy a "lower end" system at a lower price and just replace it with a "lower end" system a year later. I'll get two systems, one of which is better than today's "top of the line", for the same price as one "top of the line" machine today.
Make sure you get something upgradable, of course.
Just look at CPU prices: Athlon XP 2500+ CPUs run around 2x the price of a 2000+. 3000+ CPUs are double that again. That's 4x the price for 1.5x the performance. Same for RAM, and to some degree hard drives. (With hard drives, you often get more "bang for the buck" by getting something close to top of the line. 120 gigs or so is currently the sweet spot as far as price per gig, and that's close to top of the line these days. But as soon as you jump to 160 or 200 gigs the price skyrockets. If you go down in size, you're spending not much less and getting significantly les.)
retrorocket.o not found, launch anyway?
Things are very different. Look at the Quake 3 engine. The higher the FSB goes, the faster it renders. Most modern (and not so modern, Q3 is OLD now) engines not only take up more space than the available cache, but they also rely on streaming large amounts of data from main memory. Higher FSBs, as long as memory speeds keep up with them, will accellerate most modern and not so modern gaming engines.
I vote: Porn Stars!
'Sylvia' is the initial version with a 400 MHz DDR bus and 512K of cache.
'Jenna' is the second version, with extended SSE instructions and a bus speed jump to 500 MHz DDR.
'Traci' is the big jump to a 800 Mhz DDR bus and a 1MB L2 cache. It makes its debut with the 'Peter' Northbridge chipset along with a Southbridge that has no name, but a very wide bus...
A deep unwavering belief is a sure sign you're missing something...
Check out the CRC problems on MSI's (VIA) KT400 mbs. If winxp/via drivers happen to drop ide speeds from UDMA to PIO due to CRCs (still looking into it), other KT400 mbs might be experiencing the same problem, except that the users don't know about it - maybe this is the case where ignorance is not bliss. In the case of KT4V, MSI's idiotic solution to CRC problem is to slow down the ide speeds. The latest trend in the AMD world is to blame manufacturers of different components on the motherboard or the "other" guy, when something is wrong. Good examples are AMD and Tyan, Tyan and Crucial.
Anyone remember just a couple years ago when you could actually plan out a simple upgrade to your computer that would make it perform better for a modest price?
Toss in some extra RAM, wow no swapping!
Replace that CPU, doesn't Quake run good now!
The furious pace of bus speed changes have pretty much killed these types of upgrades for home/desktop users. Adding more PC2100 ram to their system when they know they're getting a DDR400 mobo is highly annoying. And forget about popping a new P4 or Athlons into your 1 year old mobo. Gotta buy $300 of new RAM and a $200 new DDR666-PC31337 AsusBitDragonMSI Ultra Deluxe to go with it!
Bleh.
Actually, g++ will peg your CPU. Compiling C++ code is processor intensive enough that reading files from disk isn't the bottleneck. During a big C++ compile, my processor usage will however around 70-80% on a 2GHz P4.
A deep unwavering belief is a sure sign you're missing something...
That is the sensible approach, but it does have a cost: one quarter of your penis length.
Yikes, that's one inch I can't afford to lose.
I recommend that if you need to shave costs, you still buy the best motherboard you can afford, then skimp on easily-upgradeable parts that will come down in price very rapidly after the first spasm of bleeding edge sales, such as CPU and video card.
By starting with a good motherboard, you also maximize the upgradeable lifespan of the system, because it is more likely to support newer components on down the line.
~REZ~ #43301. Who'd fake being me anyway?
Why is this a big deal when Intel is moving to 2x faster (800MHz) bus later this year?
I am well aware that the 400mhz bus on p4s is quad pumped, and is truely running at 100mhz fsb (or a bit more for 533).
:)
:) bad habbits)
I am aware that AMD 400mhz bus is a double pumped 200mhz fsb.
Could someone explain what "double pumped" actually means? if I think back I remember hearing something about how in doulbe pumped.. the cpu grabs data off the bus at the beginning, and the end of a single clock cycle. is there a downside to doing things this way?
Or perhaps.. this is the best way things should be done, and cpu designers should concentrate on LOWERING mhz (for heat/energy reasons), and UPPING the amount of data/instructions it can do in a single clock cycle?
So eventually we could move toward a computer that can run on a single clock cycle, which would be a mhzless computer? I know there is theory somewhere in there
Would it not help voltage/heat greatly if the bus was 33mhz and (12x) pumped?
--Zuchini.
(I keep writing my name, erasing and using an alias instead
I built my first Athlon system after seeing how well a friend made out with his Slot-A Athlon system. Socket-A 1.2GHz was the bleeding edge at the time so I got the 1.0GHz model w/ pc133.
After I'd pulled out half of my hair trying to figure out why it refused to run stable (or as stable as my Celeron 300A w/ Win98), I ditched the mobo for one of the new KT266A boards. It ran sort of OK for a month, then stopped running for more than 1 or 2 minutes at a time. After more hair-pulling I noticed that the power supply was putting out 3.9v on the 3.3v line and the power regulators on the board were too hot to touch. A new power supply, motherboard, and CPU later I was back in business... or so I thought.
It still had weird unexplainable crashes at odd times, so I installed Win2K and that helped a little, but it still wasn't stable. I was determined to have an Athlon system that was stable, so instead of going Intel, I started scouring message boards for anything that would help me get through this.
Many posts pointed at a buggy implementation of ACPI in the first revision of Soundblaster Live's, which I had. Got a Game Theater XP, but same problems. Finally found some info about how VIA wasn't even implementing PCI to spec, so it was time to look for a non-VIA board.
At the time it was mostly all-VIA or an AMD northbridge + VIA 686B southbridge combo. nForce had just come out, so I was hesitant to try an untested chipset, so the only other alternative was AMD's new SMP chipset. SMP was something I'd always wanted to try out and the Asus board could use my old RAM and CPU until I could afford a pair of the pricey Athlon MPs, so thats what I got.
About a year later, I'm still running it with a single XP 1800+ and it has been very stable. The only time it ever crashes is when I let it get too dirty and the geForce3 overheats.
Oh yeah, review sites are absolutely worthless if you're looking for stability. "Its very stable!" really means "It didn't explode while we overclocked the hell out of it for the 3 days we had it."
So anyway, my point is that some of us get burned bad by Athlon systems. Its only because I'm so stubborn (or maybe insane) that I stuck with it until I got it stable instead of switching to an Intel system and preaching the evils of AMD.
I beg to differ, I think the front side bus upgrades are the most important thing they do. You see, FSB is how fast you can get data to and from that super fast processor. You don't want a bottle neck between your memory and your L2 cache, or your Cache and your processor. I like that their upgrading, but I read an press release about it, over the summer, So I think it's retarded that their putting it in slashdot now.
Anonymous Cowards - Oh God, How I hate you
The real question, at least in my mind, is whether they will make AthlonMP's with the 400-MHz bus. While it's not a wrap-up, indications seem to say that they won't, because it would compete with the hammers.
Seeing as how the AthlonMP motherboards have seperate busses for each processer, imagine if Nvidia made an "nforce" chipset with dual-channel memory for dual Athlons - each processer could get full memory bandwidth at the same time. That would be truly impressive, especially for RDBMS servers where you live and die on bandwidth.
But, of course, such a monster would be a direct competitor with the Hammers - and AMD's got too much at stake to let the Hammers fail.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
The problem with such an approach is that it would drastically increase latency between the memory and the processor. While theoretically you could transfer an enormous amount of data in a short amount of time, random access to data would be very slow. The actual bandwidth comes into effect when a large amount of data is being fed to the processor in a steady stream, but when you need only a small bit of data, it would travel at the actual bus speed across the pipe, in this case 200mhz for the Athlon and 133mhz for the Pentium 4. This is why the Pentium 4 excels using Rambus at 32ns while the Athlon prefers 8ns DDR.
Not true! I have my KT266a motherboard here running a barton, it's just got the FSB underclocked, it runs cool and faster than my old tbird. And this system has PC3200 DDR RAM=, it just is running at PC2100 speeds right now. My next purchase wil be a new mobo that can take FULL advantage of the CPU an RAM. Look at the Intel side, they change the PHYSICAL pinout so you CAN'T do this. The athlon has been on one single pinout while intel has done FC-PGA, FC-PGA2, 427(?), 472(?).
DOn't underestimate the power and value you can get from underclocking.
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
All digital data is synchronized to a clock, be it source-synchronous (i.e. clock comes with data), which is the case with DDR, or recovered clock (i.e. clock information is based on rate of change of incoming data). Whatever scheme you get, you will still have a clock inside at some point.
Traditionally, the memory elements or registers on a chip will ignore incoming data until the clock signal undergoes a positive transition, i.e. logic low to logic high. At that point, assuming the data has been stable for a long enough period of time before and after the clock edge, it will be captured. However, since there is only one positive edge per clock cycle, data can only be captured on that edge.
In a double-pumped scheme, what you have is a set of 2:1 multiplexors that go to two different sets of registers. One is sensitive to positive edges, the other is sensitive to negative edges, i.e. logic high to logic low transitions. If you simply wiggle the data out faster, and you have a double-pumped scheme with a small FIFO buffer, you can recover data twice as fast as a single edged scheme. On the interface itself, there are special low skew low insertion delay clock distribution schemes that enable this to happen without too many problems.
In a quad-pumped scheme, you actually have two separate clocks that are 90 degrees out of phase with each other. In effect, you have two positive and then two negative edges to work with internally now. You wiggle data out at 4x the single data rate, and have 4:1 multiplexers to the registers, plus (again) a careful layout of the internal clocks.
The area overhead in such schemes is minimal (~10% for DDR) and really takes advantage of the speed of on-chip devices. It does take some special consideration, but from the perspective of increased die size, it's not a problem. Power, however, is significantly increased for both I/O (SSTL-2 type stuff) and for core devices because of the data rates, and that is also a consideration during design of not only the power distribution, but also the package/module design and the board design.
And, FYI, Rambus uses multiple serial/deserialization (SERDES) that wiggles data between a pair of signals (positive and negative) whose voltage differential is recovered, not for individual levels, which (supposedly but not actually) simplifies matters. Transmitting data via this differential is actually much faster than a single-ended scheme like DDR currently is (single ended meaning all I/O refer to a common ground (and voltage reference)). Then they even IIRC get into exotic schemes like multi-level differential (i.e. steppings between 0 millivolts differential and full swing). I could be wrong about the latter though...
Just take the bathroom fan out and mount it on these bad boys .... ya'll will be good to go @ about 60c.
actually I am happy to see you, however that is in fact a banana in my pocket.
(http://www.tomshardware.com/business/20030314/ce
Also, every time AMD adds more cache or increases the FSB speed, the processor gets a lower clock rate to product number ratio. The 2700+ with 256Kb of L2 Cache is clocked the same as the 3000+ with 512Kb. So, even if they shipped 3200+'s with a 400 MHz FSB, it would probably be clocked about the same as a 3000+ (at like 2166 MHz). All in all this isn't a bad thing, but you wouldn't be getting an extra 200+'s AND the increase in speed from the faster FSB, the FSB performance bump is figured in to the model number.
they are already running waay too hot.
Actually, AMD processors are cooler than the equivalently-performing Pentium 4 chips.
Athlon XP 3000+ max heat: 74.3W
Athlon XP 3000+ typical: 58.4W
Athlon XP 3000+ temperature limit: 85C
Pentium 4 3.06 GHz theoretical max heat: 109.0W
Pentium 4 3.06 GHz thermal design power: 81.8W
Pentium 4 3.06 GHz temperature limit: 69C
What Intel calls "thermal design power" is sort of similar to what AMD calls the "typical" number. It's 75% of the theoretical max temp, so the theoretical max temp for the Pentium 4 would be 109.0W. But the P4's clock throttling would keep it from hitting that theoretical max temp.
My source for all this:
http://users.erols.com/chare/elec.htm
Note also that since your power supply isn't 100% efficient, and since the power supply has to produce one Watt for each Watt your system dissipates, that a complete system with a Pentium 4 will dissipate over twice the difference of just the CPUs. In other words, for our example, the Pentium 4 dissipates about 23W more, so the Pentium 4 complete system will dissipate even more than 46W compared to the Athlon XP system. I'm not sure how efficient a typical power supply is, but if we assume 66% efficiency, the total for the Pentium 4 complete system would be about 58W more than the Athlon XP system.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
Is that 400 MHz in Intel or AMD numbers? Are they going to release it as the 533+ bus operating at 400MHz? You know, so consumers won't get confused...
Best Buy rep: Based on what you described to me, I would recommend this Compaq with an Athlon 2100+ processor.
Average customer: Is that a Pentium? How fast is it?
Best Buy rep: Actually, it's roughly equivelent to the Pentium 4 architecure, and runs at about 1.8GHz.
Average customer: Oh, give me whatever's cheaper.
Best Buy: *sigh* Have you taken a look at our eMachines yet?