Slashdot Mirror
PowerPC 970 Running at 2.5 GHz
kuwan writes "IBM has just released a press release that indicates they have the new PowerPC 970 running at 1.8 to 2.5 GHz making it 'the fastest PowerPC so far.' IBM's original estimates were to have the chip running at 1.4 to 1.8 GHz at introduction, so this is very good news for those of us hoping Apple will use this as their next-generation chip."
Here you can find a more technical details than just press release.
Here is the actual spec about the PowerPC 970.
Ars Technica articles. Apparently, PPC 970 just last year's news. The real news is just the cranked-up speed...
--
Error 500: Internal sig error
This is the same 970 as before. No lengthened pipeline, although the 970 has a relatively long pipeline to begin with. And they probably hit 2.5ghz by selective testing... I haven't seen suggestions they can manufacture these chips in quantity yet. Keep in mind that Intel demos ~5GHz chips every few months or so. Even so, it's promising that the design seems to scale up that far without issues and without needing a process change.
I've had this sig for three days.
"First of all, what is the processor that Apple using now? Isn't it some sort of PowerPC already? I see this one supports Altivec and I know that G3 and G4 Apple computers have the same instruction sets. Is this just another implementation, or is G3 and G4 relatives of this new processor?"
Apple does currently use a PowerPC processor in their computers. They have for the past eight years or so. Currently they're using the "750" edition, a'la G3 and G4, which are supplied by both IBM and Motorola.
"Second: what operating system does the IBM PowerPC run?"
The IBM machines with these series of microprocessors are things like the later generation AS/400s and RS/6000's. There are also some workstation machines (both badged as such and badged differently) with IBM PowerPCs in them. AS/400s use OS/400. RS/6000s can run many different OSes, including Linux and AIX.
"I suspect that the article is just confusing and processor itself is not made by IBM. Right??"
Wrong, at least on who makes the microprocessor. Motorola hasn't been doing so well lately, and even early on they had to deal with IBM to meet quota. IBM's hand in the PowerPC line is visible in Macintosh 5200's, which were common schoolroom computers that are starting to be end-of-lifed. They're dating back to August 1996 or so.
IBM had PL/1, with syntax worse than JOSS,
And everywhere the language went, it was a total loss...
The 970 has the same instruction set (99%) as the G4, but it also has a very, very different internal architecture that should make it quite a bit faster than the G4 at the same clock rate. It's actually a scaled-down version of the Power4 chip, the CPU in a lot of IBM's much larger systems. The Power family is the root of the PowerPC chip, which was actually created by IBM/Apple/Motorola to simply use the same instruction set.
The IBM Power4 runs many of IBM's OS's.
What's more interesting is that the frontside bus of the 970 was designed to scale with processor speed. So the 1.8GHz was supposed to have a 900MHz (well, presumably 225MHz quad-pumped) FSB, using a multiplier of 2. The 2.5GHz, then, has two options... either drop down a notch to use a multiplier of 3 (getting an 833MHz FSB, which is manageable)... or go full-hog and hit a 1.25GHz FSB. While I suspect that for the 2.5GHz chip the answer is, unfortunately, the former, the question is a bit hazier in the case of a 2GHz part... 1GHz is manageable but impressive, whereaz 666MHz simply isn't enough. Of course, they can allow non-simple multipliers and solve the issue, but I do recall that they were planning on supporting only integral multipliers.
I've had this sig for three days.
First of all, what is the processor that Apple using now? Isn't it some sort of PowerPC already? I see this one supports Altivec and I know that G3 and G4 Apple computers have the same instruction sets. Is this just another implementation, or is G3 and G4 relatives of this new processor?
Apple currently uses the G4 and G3 family. The G4 has AltiVec, G3 does not. G4/G3 are product names, whereas 970 are more like model numbers. There all related in that they implement the PowerPC ISA (Instruction Set Archetecture).
Second: what operating system does the IBM PowerPC run?
Depends on who is selling the machine the chip is in. Apple sells OS9 and OSX. IBM has AIX. And of course there's Linux and BSD. These are the most common.
I suspect that the article is just confusing and processor itself is not made by IBM. Right??
Nope, IBM does manufacture the 970. IBM also makes G3's. AFAIK Motorola is the only one making G4's right now (could be wrong here, could be that IBM is cranking some G4's as well). Also note that both Motorola and IBM sell other variations of the PowerPC (most well known is the PPC that powers the Nintendo GameCube).
PowerPC is an open architecture; several companies make different CPUs based on the design. IBM's historically made them for servers (the 970 was originally intended to be a server chip) while Motorola made them for desktops (Apple). Only problem is, Motorola sucks -- and their growth in the wireless business has gotten them to the point where they don't need Apple's business any more, so they have no real reason to improve their CPU line.
The G3 and G4 are also PowerPC chips -- they just are specific models made by Motorola. It's half new implementation, half relative.
Finally, a CPU doesn't run any specific OS -- OSes just have to be written for that CPU (and more generally, for the system architecture that CPU uses). Linux has supported the PPC for a long time; there's a distro called Yellow Dog that specifically targets Macs, and does a good job of it. Mac OS X's kernel, Darwin, has been backported to Intel IA-32. Windows used to be available for Alpha processors. It's just a matter of coding and hardware knowledge.
From reading the specs it says:
9 Fetch, Decode Stages
5-13 OoO Execute Stages
2-3 Dispatch, Commit
So at total of 16-25 pipelined stages. I also notice that the longest(25) is for the Alti-Vec engine. This is very comparable to Pentium 4 which has 26 pipelined stages, although Pentium 4 does not have a vector engine.
Where the Music Matters
here is some info i found.. might help:c hdocs/A1387A29AC1C2AE087256C5200611780
SPECint2000
- 937 @ 1.8 GHz
SPECfp2000
- 1051 @ 1.8 GHz
Dhrystone MIPS
- 5220 @ 1.8 GHz
- 2.9 DMIPS / MHz
Additional Performance
- Peak scalar GFLOPS = 7.2
- Peak SIMD GFLOPS = 14.4
- RC5 : 18M keys/sec
Unfortunately at the very bottom it says that some of this are estimates.. here is the link where I got the info: http://www-3.ibm.com/chips/techlib/techlib.nsf/te
First of all, what is the processor that Apple using now? Isn't it some sort of PowerPC already?
G3 and G4 are Apple marketing terms for current PPC chips, made by IBM or Motorola (the G3s in the iBooks are made by IBM). The only real difference between the two is that the ones with a G4 sticker on it supports AltiVec and SMP (I'm simplifying here for the sake of brevity, before I get flamed). Both are 32-bit chips.
The 970 will probably be called a G5 by Apple (although they may drop the G_ naming convention and call it a PPC64 or something) and is a 64-bit PPC chip based on IBM's Power4 series, with AlriVec tacked on. Power4 is a PPC-derived architecture, specifically designed to run in high-end UNIX servers, where x86 just doesn't cut it. With the 970 IBM are trying to move this technology to the desktop.
Second: what operating system does the IBM PowerPC run?
It will run any OS that runs on current PPC chips (PPC Linux and OS X, for example), although it will probably require OS modifications to take advantage of the 64-bit features of this chip.
I suspect that the article is just confusing and processor itself is not made by IBM. Right??
The chip indeed is made by IBM, as are the G3s in the current iBook range (as I recall Motorola G3s top out at <600MHz, while IBM make them up to 1GHz). Apple is expected to be one of the largest customers for these chips, hence their mention.
I am TheRaven on Soylent News
For comparison's sake, the P4 Xeon @ 1.8ghz pulls 703/717 (int/fp) on SPEC CPU2000.
Assuming a linear scaling in SPEC performance, we can look forward to a 2.5ghz 970 scoring about 1294/1460, which is pretty respectable. Not a world beater (especially for 2H03), but a far cry from the abominable performance of the current G4.
'jfb
To spur "enterprise Linux," Big Bang, the distributed two-phase commit.
Because we're all too lazy to copy/paste
OK, Everyone who wants to understand which processor is fastest should really take a course on processors. Here's the (condensed) deal with the MHz myth:
All other things being equal, faster clock frequency = faster processor. The trick is in the magic words "all other things being equal". If I have a 1 GHz G4 and overclock it to 1.8GHz it will be faster. That's because the processor is using the exact same process but all the steps in the process suddenly take less time.
The problem is that no two processor designs are the same. RISC vs CISC isn't even the only consideration. There are cache sizes/locations, number of pipeline stages, number of pipelines, processor component layout, all kinds of crap. And thats just IN the processor. Motherboard designs don't even enter into my discussion.
PPC and x86 are very different, as well you know if you are a nerd (if you aren't then what are you doing here anyway?). But even processors that run the same instruction set are different enough that clock frequency doesn't necessarily dictate relative processing speed. This is why if you went to tom's hardware when the P4's first came out and looked at the benchmarks, initial P4's were rated as slower than P3's which were running at a SLOWER clock frequency. And I don't think I have to tell you about AMD vs. Intel processors at equal clock speeds.
The point is that clock frequency is a number that represents something that is actually going on inside your processor. It doesn't always accurately represent speeds relative to other processors, but its a pretty good heuristic when used wisely. If you're comparing the speed of different P4's you wouldn't be in error if you said "I want a 2.6GHz P4 because its faster than a 2.2GHz P4". However, you probably would be in error if you said "I want a 2.6GHz P4 because its faster than a 2.5GHz Power5".
The Statue of Liberty is America's lawn jockey.
Assuming the same bus speed (which is impossible, so take these numbers to be within, say, one hundred points of reality) and linear performance progression, the 2.5GHz chip should have:
;)
SPECint2000 =
937 / 1.8 = 520.5 points/GHz * 2.5
Estimated Score ~= 1300
Average P4@3.0GHz score ~= 1080 (the 970 = 20% faster)
SPECfp2000 =
1051 / 1.8 = 583.9 points/GHz * 2.5
Estimated Score ~= 1460
Average P4@3.0GHz score ~= 1100 (the 970 = 33% faster)
RC5 =
18 / 1.8 = 10 * 2.5
Estimated Score ~= 25M keys/sec
Average P4@3.0GHz score ~= 4.3M keys/sec (the 970 = 581% faster)
Take these numbers with a grain of salt, but they're somewhat interesting. I like the RC5 score, especially.
Yeah you're right I didn't account for MMX and SSE.
However there is little comparison.
Alti-Vec
# 32 separate Registers
# 128 bits per register
# No interference with FP registers
# no context or mode switching
# max throughput: 8 Flops / cycle
MMX/SSE
# 8 MMX registers shared with the FPU, 8 for SSE
# 64 bits per mmx register, 128 bits per xmm register
# MMX stalls the FP registers
# context switching required for MMX
# max throughput: 2 Flops / cycle
When you are playing a 3D game do you really want your FPU stalled for vector calculations?
To be fair, you could program your 3D game to do all FPU calculations in SSE. gcc has an option to do this automatically now. And SSE2 is one step ahead of AltiVec in one regard - it supports a few double-precision operations.
But aside from those two nitpicks, I agree completely. I've hand-optimized code for both Pentium/SSE and G4/AltiVec and there's no comparison: SSE provides a small performance boost for a lot of work, while AltiVec provides a large performance boost for a little bit of work. AltiVec has very fancy shift, rotate, and shuffle instructions that are completely lacking in SSE. These are useful for more than just RC5 - they're totally necessary to vectorize many more complicated algorithms without the overhead of putting the data in the right place eating up any potential speed gains.
That's why the 970 in a Mac will easily beat the P4 in a number of tests: Apple has optimized hundreds of system calls to use AltiVec already, so many programs get the speed gain automatically.
While UNIX compatibility in OS X is great, calling it a "true UNIX" is really rather misleading. First of all, the kernel isn't a UNIX kernel, it's a hacked Mach kernel with a BSD compatibility layer. Furthermore, there are very significant differences in userland, including things like a case-insensitive file system, huge changes in system administration, lack of device nodes for things like audio and video, multiple views of the file system (from Carbon/UNIX), etc. Also, the standard UNIX window system, X11, is at best bolted onto OS X.
Now, you may think all these things are improvements to UNIX, and you might be right. However, they make OS X pretty significantly different from UNIX. And while some applications port with no problems to OS X, others require incorporating Cocoa or Carbon code for porting, which can be a lot of effort.
Apple shipped 7484 servers (presumably mainly Xserve) in Q3 2002. In contrast, there were only 3500 Itanium 2 based servers sold in the whole of 2002.
The future looks even better for Apple in the server space, following the recent release of the new Xserve and the Xserve RAID. I can't wait to see an Apple 64 bit PPC 970 blade server to blow the crappy Dell out of the water.
Quoting numbers attributed to Internet World, MacInTouch (Saturday, Jan 12) reports that Apple's share of the server market has more than trebled from 0.2 percent to 0.7 percent (Q3 '01 vs Q3 '02). An equally telling statistic is the fact that approximately 40 percent of growth had taken place by the end of Q2 '02 (ie before Apple's Xserve was released).
In terms of unit sales, Internet World quotes the following for Apple:
? Q3 '01 2,049
? Q2 '02 3,937
? Q3 '02 7,484
MacAddict and others are reporting that the press release has been removed from IBM's site; clicking the link to it in this story now takes one to a listing of IBM's German press releases. The pr on the 2.5 GHz 970 seems to have been completely removed. Might the announcement have been premature?