What Makes Apple's Power Mac G5 Processor So Hot 313
An anonymous reader writes "58 million transistors can drive a lot of power. Apparently, Apple appreciated the choices IBM processor architects made when designing the 970 family. This article provides the 64-bit architecture big picture for the 970 family (A.K.A. the Power Mac G5) and the critical issues in IBM's 64-bit POWER designs, covering 32-bit compatibility, power management, and processor bus design."
Circular Logic (Score:4, Funny)
What Makes Apple's Power Mac G5 Processor So Hot? (Score:5, Funny)
Re:What Makes Apple's Power Mac G5 Processor So Ho (Score:4, Funny)
What makes it so hot (abridged) (Score:3, Insightful)
Re:What makes it so hot (abridged) (Score:2, Flamebait)
For us Apple users - you bet!
Well, except for the fact that the processor's so hot, but you know what I mean
Re:What makes it so hot (abridged) (Score:4, Interesting)
Re:What makes it so hot (abridged) (Score:5, Insightful)
64 bit integers (Score:5, Insightful)
Addressing (Score:5, Informative)
Re:64 bit integers (Score:5, Informative)
Actually, since most modern CPUs are x86 variants, the floating point registers are usually 80 bits wide (and have been since the 1981 introduction of the 8087).
As far as "complex mathematical calculations" go, 64-bit integers aren't really that big a deal. It's pretty rare to need integers bigger than 2^32 but no bigger than 2^64; floating point usually handles big numbers more flexibly.
The big deal with 64-bit CPUs is 64-bit address pointers and operations on them (which usually aren't more complex than adding and shifting).
Re:64 bit integers (Score:5, Informative)
More importantly, it doesn't have to break the 64 bit operations into many successive 32 bit operations. 64 bit operation are not simply 2x32 bit operations, but can be several dozen operations.
An 8-bit microcontroller can perform 64 bit floating point operations correctly. It just takes a long time.
Re:64 bit integers (Score:2)
Re:64 bit integers (Score:3, Informative)
A lot of 32 bit processors already have a 64 bit data bus right now. This is for example the case for the Intel 32 bit processors. So loading an entity bigger than 32 bit is not an issue. The bus width between the processor core and the L1 cache is even wider allowing even bigger chunks to be loaded in one cycle.
As for the floating point unit, it is also designed to do the operations in double precision (64bit) or even more. Once again, full 64 bit processors have no
Re:64 bit integers (Score:3, Informative)
A lot of 32 bit processors already have a 64 bit data bus right now. This is for example the case for the Intel 32 bit processors. So loading an entity bigger than 32 bit is not an issue.
This is not relevant to the instruction stream - you still need two load instructions. Actual bus widths are not visible to executing code - it's simply there to improve bandwidth.
Re:64 bit integers (Score:4, Informative)
You're right that P4 has a lot of rename registers, but those aren't directly accessible. Thus, their usefulness is limited by the CPU's look-ahead. In theory, a compiler can always do better if it has access to the same number of registers because it can look arbitrarily far ahead.
Oh, but you were trolling. You didn't actually expect an answer, did you? Well, you got one.
Nice title... (Score:2)
Re:Nice title... (Score:2, Interesting)
The PPC970FX actually doesn't produce that much heat compared to the current AMD and Intel crop, only dissapating 54W typical at 2.5Ghz (albiet a good deal higher at peak). But due to it's smaller die, it has to dissapate that heat from a smaller area, thus requiring a cooler heat sink to dissapate it into to maintain the same die temperature.
It's basic thermodynamics,
Re:Nice title... (Score:4, Informative)
Not sure if this is the reason, but its not just a matter of amount of heat generated. Its the amount of heat generated in a given area. So if you generate the same, or even less, heat in a smaller area, you may need to resort to something more efficient that air cooling to do the job.
ob Memory (Score:5, Funny)
-GillBates0, 2004.
Re:ob Memory (Score:5, Informative)
This page [plexos.com] makes a fairly convincing argument that 256 bit CPUs should be enough (basically, there would be no way to exhaust the amount of memory a 256 bit CPU could access, because the number of memory locations is about the same as the number of atoms in the universe).
--Mark
Re:atoms in one universe (Score:5, Funny)
Simple. Since we've got a computer capable of simulating the entire universe, we'll just use it to simulate a universe which contains a computer which is capable of simulating more than one universe.
Re: uh oh... (Score:3, Interesting)
Maybe it's possible to disprove this based on the idea that the other universe would need to be physically larger?
Re: uh oh... (Score:3, Funny)
Re:ob Memory (Score:3, Informative)
All i can say is "Whoa".
Teflon underside (Score:5, Funny)
"Turn it over, and you can cook dinner".
Re:Teflon underside (Score:2, Funny)
Re:Teflon underside (Score:3, Informative)
Re:Teflon underside (Score:4, Funny)
Re:Teflon underside (Score:3, Funny)
Re:Teflon underside + Fans = Convection oven (Score:2)
Re:Teflon underside (Score:3, Funny)
Increased Pointer size (Score:5, Interesting)
Re:Increased Pointer size (Score:4, Insightful)
Also, 64-bit pointers allow you to go from a max of 4GB of RAM to 16 billion GB, so the assumption is memory prices will keep dropping and you'll have much more than twice as much RAM on your 64-bit system anyway.
Re:Increased Pointer size (Score:3, Informative)
You may forget that a bitblit, or a bulk memory copy operation can be accomplished in half of the time using the same number of 64 bit registers as 32 bit registers. How do you think common operations like scaling and color transformation will be affected by the increased register size and memory IO path? In my experience (Ultrasparc real world apps like GIMP and OpenSSL) most bulk i
Re:Increased Pointer size (Score:3, Funny)
Re:Increased Pointer size (Score:3, Informative)
But then again, you don't have to use linked lists. The C++ standard template library has all sorts of wonderful containers that may be better suited than simple linked-lists. Java has some neat containers as well.
Re:Increased Pointer size (Score:5, Funny)
Re:Increased Pointer size (Score:3, Funny)
On the other side of the fence, C++'s usage of magic pixie dust and moonbeams is what gives it its flexibility -- and by that I mean the ability to program in the procedural style or object-oriented style or template style, etc. The downside is that magic and moonbeams aren't really...reliable. So yo
Add instruction sets size too (Score:4, Interesting)
The notable exception is the Arm's thumb instruction set [embedded.com] (it's cool).
The sad part "my address bus is bigger than you" is going the "I have more MHz than you" way soon as parallel CPUs (mulit-core or otherwise) become cheaper.. 90% of our tasks are better done parallel than using a single fast chip . Hell , half of the tasks really don't need anything beyond a 300/400 mhz clocks.Re:Add instruction sets size too (Score:2)
Re:Add instruction sets size too (Score:3, Informative)
Re:Add instruction sets size too (Score:2)
If they were 64bits for each individual instruction, I think it would have a huge undesirable impact on execution.
Re:Add instruction sets size too (Score:3, Informative)
Re:Increased Pointer size (Score:5, Interesting)
Re:Increased Pointer size (Score:2)
Re:Increased Pointer size (Score:2, Insightful)
I guess if you plan on shelving out for 64 bit, you should plan on getting more ram w/ it.
Re:Increased Pointer size (Score:2, Informative)
If you build in 32-bit mode [e.g. -m32] you lose the major benefits of the 64 which is namely the extra registers.
Tom
Re:Increased Pointer size (Score:2)
That's assuming each node has a 'next' pointer. That's how they teach it in CS classes, but in the real world it makes more sense to use an index into an allocator. It requires cleverness from the language and library designers, of course, but there isn't necessarily a size increase.
Re: (Score:2, Interesting)
Re:Increased Pointer size (Score:5, Informative)
Sun machines with UltraSPARC processors do this too. They run 64 bit kernels, and applications are 32 bits. Unless you actually need 64 bits, in which case you feed the compiler some differnet options and it makes a 64 bit executable for you.
Both Solaris and Linux do it the same way. When you build a kernel for Linux on an UltraSPARC machine the part about kernel support for different kinds of executables offers you (among other options) 32 bit ELF (which you need), 64 bit ELF (optional), and Solaris emulation (never tried it...).
...laura
Re:Increased Pointer size (Score:2)
If I bought a 64 bit system, simply because it's the "Best", but only got 1GB of RAM, I have less useful memory, because the pointers take up all of my physical RAM. Do the architects of these systems take this into account?
Umm, duh? You increase your footprint by a small amount (say 10%) to get access to a flat 16EB address space. With memory running $200/G for the good stuff, why not?
Re:Increased Pointer size (Score:3, Interesting)
From a developer perspective (at least in software I've worked on) programmers often do not realize that a vast number of projects took dependencies on 4 byte pointers.
Structure alignment, pointer size, etc have plagued all projects I've worked on, even after a cursory perusal at the code indicating "It looks good". Lots of bugs
Re:Increased Pointer size (Score:2)
Address space is one thing. Register size is another: I could be crunching lots of 64-bit numbers, yet stay within 1GB of memory consumption.
Re:Increased Pointer size (Score:2)
WATT figures for G5 vs AMD-64? (Score:5, Interesting)
Re:WATT figures for G5 vs AMD-64? (Score:5, Informative)
2.4 GHz A64- 89 W
3.4 GHz P4(Northwood)- 89 W
3.4 GHz P4(Prescott)- 103 W
Best guess on the 2.5 GHz G5 is around 65 W.
It's the die size that requires additional cooling (Score:3, Interesting)
90nm transiters require less power then their larger counterparts. The problem is, for the same die size they use more power. So you end up with a relatively low power CPU that requires massive cooling.
Apple and IBM had a lot of problems because they expe
Re:WATT figures for G5 vs AMD-64? (Score:2, Interesting)
From Intel's datasheets [intel.com]: P4 90 nm (prescott) 520-550 models 84 W of design power (what Intel recommends the heatsink be able to pull).
550-560 models 115 W of design power.
From AMD's datasheets [amd.com]: design power (measured with max amplitude and nominal voltage) is 89 watts for all power rating
Re:you want heat data (Score:3, Interesting)
Re:WATT figures for G5 vs AMD-64? (Score:3, Informative)
AMD's 90nm 3500+ uses under 67W of power however at 2.2GHz.
Nitpicking... (Score:5, Insightful)
I know the first 2 digits are 1 and 8, but 2^64 bytes is still 'only' 16 exabytes...
Re:Nitpicking... (Score:5, Informative)
When they say 18 exabytes, they're talking base-10, otherwise they would have used the "gibi-" equivalent (exibytes?)
Re:Nitpicking... (Score:3, Insightful)
When they say 18 exabytes, they're talking base-10, otherwise they would have used the "gibi-" equivalent (exibytes?)
If they meant base10, they should have used existing conventions (10^x) rather than trying to shoehorn this kibi stuff into usage. Common usage is that an exabyte is 2^60 bytes. Nobody in there right mind uses exibyte for that. The standard is something that NIST is trying to force on the community without any sort of support - they're exceeding their mandate.
Re:Nitpicking... (Score:2)
So, 2^64 = 1.84467e19 bytes = 1.84467e16 KB = 1.84467e13 MB = 1.84467e10 GB = 18.4467e09 GB = 18 billion GB
PC vs Mac cooling. (Score:2, Funny)
Fans and cooling (Score:5, Interesting)
Your PowerMac G5 has nine fans. Again, not much space left for additional cooling without interfering.
And get this, the PowerMac G5 already uses a liquid cooling setup. The only possible additional mod is to hook the current setup to a resevoir and radiator on the outside of the case, as the inside already has a radiator per CPU and something like a 120mm fan per CPU.
Re:PC vs Mac cooling. (Score:2)
For the record... (Score:2, Redundant)
I doubt that 64 bit computing is that hot for PPC (Score:4, Insightful)
As the PPC instruction set is sane (x86 is not, urgh), beside the extra-instruction needed for 64 bit computing, there are very few difference between a PPC running on 64bit code or a PPC running on 32bit unless of course you have an app which needs more than 4GB of memory or do lots of 64-bit integer calculation..
It used to be... (Score:3, Interesting)
Now that G5's are liquid cooled, it makes me wonder if a 2.5GHz G5 is *really* a 2.5GHz G5, or if it's an overclocked 1.8GHz chip. You know, overclockers really pump things up with cool liquid cooling stuff. What's the fastest a 2.5GHz G5 could run with a traditional cooling system, like a fan and heatsink?
Oh, one more thing before I'm modded as a troll: my G4 PowerBook is my 8th Macintosh. What I'm asking is genuine curiosity.
Re:It used to be... (Score:4, Informative)
I have mod points, but I figured I'd answer your question instead.
When you do a die shrink, you can lower the power required at particular clock rate, or you can run at a higher clock speed with the same power dissipated. So when IBM went from 130nm for the 970 to 90nm for the 970FX, the top clock speed went up from 2GHz to 2.5GHz. Other than the process change, I believe there were very few changes to the chip.
Now, when you go from 130 nm to 90nm, the linear dimension across the chip is ~70% of what is was, and the area of the chip is (70%)^2 or about 50% of the previous chip.
Lets use some numbers, these may not be 100% accurate, but they'll explain the basic concept. The 2GHz 970 had a die size of about 121mm^2 and put out a maximum of 42W. That is about 350mW/mm^2. If we assume that the 2.5GHz 970FX has that same power consumption, but has a die size of 60mm^2, then the 970FX will produce 700mW/mm^2. So you have the same amount of power, but you are trying to suck it out of a smaller piece of silicon. So you need much more efficient cooling to keep the chip temperature the same. Hence, the liquid cooling system in the dual 2.5GHz G5.
Re:It used to be... (Score:3, Interesting)
I guess I should have been a little clearer. I know that the 2.5GHz chip is rated as 2.5GHz. I also know that just the clockspeed alone does not a system make. I realize the same amount of power has to be pumped through smaller pipes and be disapated. I realize it's quiet and I miss whisper-quietness of my graphite iMac (Mac #6).
I guess the question I'm asking -- aside from being a question -- is really meant to be thought provoking. Does the use of a liquid cooling system fundamentally change the defacto
Good followup link from the article (Score:5, Informative)
There's an informative link at the bottom of the article for those requiring a bit more insight into the effect of 64-bit computing.
CPU power defeated by bloated software? (Score:3, Insightful)
The quest for CPU power has been largely defeated by bloated software in applications and operating systems. Some programs I wrote in Basic on an Apple II ran faster than when written in a modern language on a G4 Dual-processor Mac with hardware 1,000 times faster.
Come on. What language are we talking about here? My basement collection includes a II+, a IIe, two IIcs and a Franklin compatible. I challenge anyone to come up with a program in Applesoft that runs faster on one of my museum pieces than on a modern Mac using C++, Java or even Perl. I mod his article -1 for troll.
While software has become bloated and to some extent inefficient, people often forget that we expect a lot more from our computers now than the single-tasking 80 column display days.
Re:CPU power defeated by bloated software? (Score:3, Funny)
_80_ columns?! Why, you young punks had it good! When I was a boy, we only had 40 columns, and we felt grateful, as our fathers had no video display at all. Grandfathers? They used their fingers or an abacus.
You've come a long way, baby.
it's hot. really. (a little perspective) (Score:4, Interesting)
my dual 2.5GHz PowerMac G5 [apple.com] idles at 52C (125F) on CPU A and 50C (122F) on CPU B. the memory controller is actually one of the hotter things, it idles at 62C (143F). however, it's not the hottest thing, of course: at full load (DVD rip+encode or playing 15 videos at once + MP3 + tasks + flicking around Exposé) both CPUs have hit a max of 83C (181F) (the computer is supposed to automatically sleep around 90C or so).
so why so effing hot? i mean, this idles at the max temp my athlon 2500 peaks at! it certainly idles at a hotter temp than it needs to, but i have no problem with that: the system runs the fans dynamically to keep the noise down, so at idle it's not as cool as it could be. the difference in noise in my room when i sleep the athlon is ridiculous - the G5 sounds like a slightly loud external hard drive that's spun up. the system also has a liquid cooling system [overclockers.com.au] to quench the processors. this seems to just keep the processors within their range. the value that i see in it is response to new heat - the CPU temps flick around a lot and are very responsive to load and the loss of load. after ramping up the CPUs to >80C, it take about three or four seconds after the load drops for the CPU temps to drop 15-20C, then maybe a total of ten or twelve seconds to drop to idle temp.
for some real-world perspective... a DVD rip+encode with HandBrake [m0k.org] with using ffmpeg engine, MP3 audio, 2-pass encoding, and gunning for your average 700MB movie time (800-1300kbps?) takes slightly less than the length of the DVD. an hour and a half long movie took about and hour and fifteen minutes to get on to my hard drive. MP3 ripping in iTunes will run up to 28x, but it's not fully loading the processors so i wonder about a drive read bottleneck. the first night i got it, i was at a loss for how to really test the speed on it, so i just decided to open up a shitload of videos. basically i played a DVD (fluff, the GPU does that), opened up something in VLC, opened up about 13 videos in QuickTime of various sizes and formats, played some MP3 music (fluff again, that's ball sweat of a cutting edge proc), and still had enough processing power to comfortably navigate files, chat, browse web pages, and flick around Exposé [apple.com]. around all of these things plus one is when a few of the videos would start stuttering and expose would start dropping frames to keep collapse speed uniform. anything past this would really start robbing time from videos.
all in all? it's fast. it's quiet. it gets hot, but it takes care of itself. coming from a 375MHz G3-upgraded PowerMac 7600 (vintage '98), i'm not doing too shabby. i just decided i'd scramjet at mach 7 to the top of the pack and then sit there for another few years.
Re:anyone else noticed how COOL the AMD-64 chips r (Score:5, Insightful)
I was able to hold my hand on the heatsink and it was barely warm.
It could be because there's inadequate conduction between the CPU core and the heatsink. Check the temperature monitors to make sure it's actually as cool as you hope it to be. It could be that just most of the heat is staying in on the CPU, which would be a bad thing. Hopefully you've already checked this though.
The Athlon64 is very cool (Score:3, Interesting)
I recently upgraded to a 754-pin Athlon64 3000+, and the hottest it's ever been is 51 C, a few degrees more than the room temperature of 43C. On a cooler night, with 100% CPU load for ~2.5 hours [2-pass XviD encoding], it peaked at 47 C. Quite impressive.
Re:The Athlon64 is very cool (Score:4, Funny)
Open a window or buy an air conditioner or something!
Heat wave (Score:2)
Re:The Athlon64 is very cool (Score:2)
Is the 43C really the room temperature or the temperature at some other point in the computer? (which sounds much more likely)
Jeroen
Re:The Athlon64 is very cool (Score:2)
Re:The Athlon64 is very cool (Score:3, Interesting)
Anyways, I've ony ever touched an Athlon64 once. It was a 3000+ and ran 28C at idle. Granted, room temperature was 25C (office aircon). Don't know what it would do in a hotter room.
Room temperature can make an enormous difference. My PC at home is a 2400+ AthlonXP, which ran at about 45C all winter under full load - room temperature is col
Re:The Athlon64 is very cool (Score:3)
Ever been to Africa? How is 32C on an avarage winter's afternoon for you?
Which part? Africa has pretty much every environment present on the planet (possibly excluding tundra). If Ghana is 32C in December, I'd expect it to be about 34C in July.
Re:The Athlon64 is very cool (Score:2, Funny)
Holy Crap! You must be really roductive [slashdot.org] !
Re:The Athlon64 is very cool (Score:4, Funny)
Cheers!
-- CD
Re:The Athlon64 is very cool (Score:2)
Are you aware that a 43C room temperature is incredibly hot? Standard room temperature in an office is 20C. A 47C room is like a scorching Texas day in the middle of summer.
Re:The Athlon64 is very cool (Score:3, Informative)
He might live in Tucson or Phoenix, Arizona; the Sonora-Arizona desert is quite a hot place in summer. Around Hermosillo, Sonora, there's a permanent high-pressure system which pushes off any clouds coming our way, so there's uninterrupted sunlight some
All through July and August, the normal, everyday noon-time temperature is over 45 C, 47 C is not really all that surprising. Once I was out in the street at 50 C.
Being a desert, it cools down qu
Re:anyone else noticed how COOL the AMD-64 chips r (Score:2)
Re:Heat Problem Back Ground (Score:5, Interesting)
That article is some crybaby whining about how expensive the G5s are. "Apple is so dumb. Why would anyone pay that much for a CRAP computer," is what the article sounds like. I think that guy needs to take his superior knowledge elsewhere and try some benchmarks ("512K of cache isn't competitive for $3000")... apparently it is because it's winning benchmarks and people are buying them. Just because you can't afford it doesn't mean it's a bad computer (doesn't mean it's a good computer either). This guy needs to grow up and write an article with facts instead of emotions...
Re:Heat Problem Back Ground (Score:4, Insightful)
Hear hear.
I've priced out equivalent machines to the current Apple offerings, and you do indeed get what you pay for. A dual-CPU 1.8Ghz Powermac is around $2500 sans monitor. Price out a dual-Xeon or dual-Opteron and you end up at around $2000-$2500 for a comparable system.
Where Apple might be missing the boat is in the ultra low end where you can buy a system for $600. (But why should they try to compete down there where margins are razor-thin?)
Re:Heat Problem Back Ground (Score:4, Insightful)
Because poor college students like me, who buy $600 computers is going to buy a PC. And when I get out of college and have the means to drop $2500 on a computer, guess what I'll probably buy.
Actually, the two real reasons why Apple doesn't sell low end machines is that it would undercut sales of their more expensive machines, and totally destroy the second hand Mac market (where used Macs are way overvalued, IMHO). Without a strong second hand market to sell a used Mac for a good price, people will be less likely to buy their high end machines.
Re:Heat Problem Back Ground (Score:4, Insightful)
I don't have a mac here, but opterons and g5s are probably similar. Pick the one you like, they're both expensive
Re:Heat Problem Back Ground (Score:5, Insightful)
Re:Heat Problem Back Ground (Score:3, Interesting)
Re:So sick of it (Score:2, Funny)
Re:So sick of it (Score:5, Funny)
Re:So sick of it (Score:3, Insightful)
But for some reason most people believe they need a 6.8ghz 2.2terabytes of ram and 15,000 rpm hard drives along with a hairdryer add on known as a 3d video card, bringing that machine up to the 400 watts+ power consumption mark.
and where does most of that energy go?? that 's right .
Re:Mac's are doomed! (Score:3, Insightful)
I used to be very pleased that my Macs lacked a CPU fan while x86 users where bolting on everything but the kitchen sink in efforts to cool their systems down.
More important, Macs had just one fan -- the power supply fan -- to cool the entire box while x86 systems were fucking festooned with them.
Nowadays, though, a prudent Mac owner should at least consider his home air-conditioning system's cooling capacity since G4's and G5's will definitely heat up a room.
And I don't even want to g
Re:Mac's are doomed! (Score:3, Insightful)
If you don't need that kind of power, you should stay on the lower end. The G5 iMac or the eMac don't seem to run all that hot.
As for fan count, presuma