I know all about the megahertz myth crap...but i was juss curious about what power can a g4 800- 1 gigahertz compare to a pentium or amd equivalents. I'm talkin about mobile chip equivalents.... Does a 800mhz g4 run about 1.4 pentium or amd moblie chips. Juss wonderin cuz I may soon get a mac and want to compare price and performance.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
A megahertz question
- Thread starter Cheap-Chopstix
- Start date
- Sort by reaction score
Well, it depends.
Id peg the 800MHz G4 ibook to compete with the low end Centrino laptops (1.3GHz and below), though the iBook costs a pretty penny less than a Centrino solution.
As a rule of thumb for me, I always jsut multiplied the Apple rated MHz speed by 2 or 2.25(sometimes even 2.75 to 3) to compare with x86 chips, especially Pentium chips with their unrefined clock speeds that do not accurately read performance. For AMD chips, I drop my multiplier to 1.5 to 1.75 (sometimes even 2) because of the efficient clock cycles AMD chips are known for.
Newer chips however, like the Athlon 64 series and the G5, are on a whole new level and are hard to gauge.
Id peg the 800MHz G4 ibook to compete with the low end Centrino laptops (1.3GHz and below), though the iBook costs a pretty penny less than a Centrino solution.
As a rule of thumb for me, I always jsut multiplied the Apple rated MHz speed by 2 or 2.25(sometimes even 2.75 to 3) to compare with x86 chips, especially Pentium chips with their unrefined clock speeds that do not accurately read performance. For AMD chips, I drop my multiplier to 1.5 to 1.75 (sometimes even 2) because of the efficient clock cycles AMD chips are known for.
Newer chips however, like the Athlon 64 series and the G5, are on a whole new level and are hard to gauge.
The simple answer is yes. But things are never that simple.
When just comparing raw clock speeds (which is at most half the power equation), the G4 is going to do quite well against a Pentium (particularly the poorly designed P4). A G4 will beat out any Pentium (or AMD for that matter) of the same clock speed, and will beat out Pentiums (particularly the P4) for up to another 1 GHz (the G4 doesn't compare quite as well with P3s or AMDs).
Of course, there are plenty of other things to consider, too. Most importantly there's RAM and cache. The new iBooks are intentionally lacking in both departments (the Powerbooks are far more powerful there).
A 1.4 GHz Pentium will probably give an 800 MHz G4 a good run for its money, but a 1 GHz G4 has got Intel beat.
My advice would be to avoid the new iBook, and opt for a Powerbook (12 or 15 inch) instead (unless you're really on a budget, in which case the iBook has a whole lot of bang for the buck). The boosted cache and higher max RAM really makes a difference in overall power.
When just comparing raw clock speeds (which is at most half the power equation), the G4 is going to do quite well against a Pentium (particularly the poorly designed P4). A G4 will beat out any Pentium (or AMD for that matter) of the same clock speed, and will beat out Pentiums (particularly the P4) for up to another 1 GHz (the G4 doesn't compare quite as well with P3s or AMDs).
Of course, there are plenty of other things to consider, too. Most importantly there's RAM and cache. The new iBooks are intentionally lacking in both departments (the Powerbooks are far more powerful there).
A 1.4 GHz Pentium will probably give an 800 MHz G4 a good run for its money, but a 1 GHz G4 has got Intel beat.
My advice would be to avoid the new iBook, and opt for a Powerbook (12 or 15 inch) instead (unless you're really on a budget, in which case the iBook has a whole lot of bang for the buck). The boosted cache and higher max RAM really makes a difference in overall power.
I would say maybe 1.5 for most things, 3 is WAY to much, compare any photoshop test with a 3GHZ p4 vs a DUAL 1.25GHZ mac, the g4 gets STOMPED every time
First off, the Centrino is an enhanced Pentium III, a chip that pretty well matched the G4 back in the day. Therefore, do not expect an 800MHz G4 to come close to a 1300MHz Centrino (it fares well, but it's not really close).
Against the Pentium IV, the G4 kicks ass. However, the G4 has only recently taken on a smaller manufacturing process that Intel is already improving upon (130nm to 90nm). Quite simply, this means that the G4 will likely never be a competitor with the P4 (that's what the G5 is for).
The Athlon has been a solidly built chip over the years, and has performed very well clock for clock versus the G4. I might give the G4 a slight edge in clock for clock. However (and again), the G4 clocks not even remotely close to the Athlon's top clock speed at present. AMD will also be switching to a 90nm process, which will mean a significant speed boost across the line. Motorola has just begun 130nm.
Against the Pentium IV, the G4 kicks ass. However, the G4 has only recently taken on a smaller manufacturing process that Intel is already improving upon (130nm to 90nm). Quite simply, this means that the G4 will likely never be a competitor with the P4 (that's what the G5 is for).
The Athlon has been a solidly built chip over the years, and has performed very well clock for clock versus the G4. I might give the G4 a slight edge in clock for clock. However (and again), the G4 clocks not even remotely close to the Athlon's top clock speed at present. AMD will also be switching to a 90nm process, which will mean a significant speed boost across the line. Motorola has just begun 130nm.
Intel chips are horible when it comes to power per clock cycle. the new amd chips compare a 2.2 athlon xp (non 64) to a 3.2 ghz p4. THis is just the desparity with in x86 chips. I would say when comparing g4 to p4 multiply the g4 by 2. when comparing a g4 to an amd chip multiply by 1.5. This is just my estimate. the whole thing is just an estimate because one has to acount for different cores. different cache sizes and amounts of memory.
when it comes to the g5 and the athlon 64 the specs are about the(the ghz are about the same. dual g5=2.2 athlon fx(roughly)
when it comes to the g5 and the athlon 64 the specs are about the(the ghz are about the same. dual g5=2.2 athlon fx(roughly)
I deal with Multi-GHz PCs at work all day and let me tell you, they ain't fast. The P3 wasn't bad, and AMD makes some pretty good chips (got a Duron myself 
on my cheapy PC). The Centrino is also surprisingly speedy, and not that bad on battery life. But after playing with my Step-Mom's Dell Laptop for awhile, you couldn't pay me to use one. Ugh. Hot, heavy, battery life was nowhere near my Mom's old PowerBook (or new iBook).
Benchmarks mean nothing, specs mean nothing. Go to a CompUSA or Best Buy, play with some Laptops. Buy whatever works best for what you do.
I like the new G4 iBook.
on my cheapy PC). The Centrino is also surprisingly speedy, and not that bad on battery life. But after playing with my Step-Mom's Dell Laptop for awhile, you couldn't pay me to use one. Ugh. Hot, heavy, battery life was nowhere near my Mom's old PowerBook (or new iBook).Benchmarks mean nothing, specs mean nothing. Go to a CompUSA or Best Buy, play with some Laptops. Buy whatever works best for what you do.
I like the new G4 iBook.
It is very difficult to say how much faster one chip is then another. Some processors are great at integer but mediocre at floating point, while some are great at floating point and fail on the integer side. Some tasks need strong integer; some need strong floating point. In the real world, you could use both very heavily. Then you have some processors can handle some tasks really well but are just adequate in others. No one makes a processor that can do it all better then anyone else all of the time. So if you want to know how much faster one chip is compared to another, you need to define faster in what sense, in this case, doing what.
Very true (just look at the recent Itanium 2 benchmarks. Damn does that thing kick ass at floating point). I'd also like to second what Rice_Web said. The Pentium-M (Centrino) is a very nice little chip, and I think the G4 would really need to be on a smaller manufacturing process to compete well with it (Moto's .13 micron process is being pretty damn funky from all indications). A .09 micron G5 with better power management (the hypothetical 980) would probably do the trick quite nicely.
The multiplier isn't really important, and unless you'll be doing a lot of video work, neither is the Megehertz. I'm running a 700 Mhz eMac, and it does almost everything I need. The truth is, a new computer made by anyone is fast enough for daily computing tasks. Dualie towers and powerbooks are for pro users, not regular Joe's. Besides, we all use mac's because of the OS, not because they are super fast. On-paper-speed isn't everything. I wouldn't sweat the processor, and focus more on system stats like RAM, cache, and videocard. That's where a computer really gets its power anyhow.
FYI: My 700 Mhz eMac compiles SETI data 6 times faster than my AMD 533 Mhz.
FYI: My 700 Mhz eMac compiles SETI data 6 times faster than my AMD 533 Mhz.
This is something we have discussed in threat "THANKS APPLE..." somewhere in pages 4-6 I think....
This is what I wrote there:
******************************************
http://www.pcmag.com/article2/0,4149,1274138,00.asp again.
I just actually realized that Dual G4 1.42 won DUAL XEON 3.06Ghz in three tests out of six. This should have something to tell when it comes to the question of performance.
According to that test which I trust very much the performance of the new Apple laptops cannot really stay far away from the PC laptops. The real question is rather that are the programs optimized for the processor. In this case it seems to me that Sorenson & Avid have done a good job because as you can see even G4 leaves XEON far behind !!!
(if you scroll down on the PC Magazine's page there is a link at the end of the 2nd chapter called: "see performance table")
******************************************
What is interesting is that with the video editing tasks Windows machines were left far behind even by G4 while even G5 had difficulties to beat Windows machines when it came something like working with stills...
Interesting - isn't it ?!?
This is what I wrote there:
******************************************
http://www.pcmag.com/article2/0,4149,1274138,00.asp again.
I just actually realized that Dual G4 1.42 won DUAL XEON 3.06Ghz in three tests out of six. This should have something to tell when it comes to the question of performance.
According to that test which I trust very much the performance of the new Apple laptops cannot really stay far away from the PC laptops. The real question is rather that are the programs optimized for the processor. In this case it seems to me that Sorenson & Avid have done a good job because as you can see even G4 leaves XEON far behind !!!
(if you scroll down on the PC Magazine's page there is a link at the end of the 2nd chapter called: "see performance table")
******************************************
What is interesting is that with the video editing tasks Windows machines were left far behind even by G4 while even G5 had difficulties to beat Windows machines when it came something like working with stills...
Interesting - isn't it ?!?
I don't understand why the Mac community keeps ragging on the Pentium 4's performace. I myself am an AMD fan but I must admit that the P4 was well designed. Did it ever occur to anyone that Intel purposefully made the P4 weak per clock cycle so that it could clock much higher overall? It's actually a lot like the RISC philosophy, where chips are supposed to be kept simple. Apple added the SPEC rate results on the G5 performace page as a marketing gimmick. The results in no way mean that a chip is slower or faster than another. The P4 did, after all, beat any Mac and AMD offerings until the G5 and Athlon 64 came out (although its being slower than the G5 is still under dispute).
And I don't understand where you all got your blind opinions of the G4 and the "multiply it by this to see how fast it is" crap. Motorola's chip division was bought out for a reason. Apple did the smart thing by going back to IBM for the G5. The only problem is, you'll probably never see a G5 in a laptop considering the Power Mac needs 9 fans (granted, they're low speed fans). You can't compare how fast two different chips are by comparing how fast they are at the same clock speed, but instead by comparing the fastest chips currently available from each line. The G4, per clock, is around as fast as a P3 (Centrino) or an Athlon/XP, except that the Centrino is available at 1.7 GHz and the latest Athlon XP runs at 2.2 GHz while the G4 is stuck at 1.25 GHz in laptops and 1.4 GHz in the last generation of Power Macs (although there's usually two of them to compensate for their slowness).
Jonathan,
What makes the P4 so well designed? Intel lost the MHz battle with the PIII and they used the P4 to reclaim it. They are using MHz to get work done, AMD follows suit to a lesser degree though. The higher the MHz the more power is required and more heat is generated. While you can have a chip running at less than half the MHz, consumer half the power and get the same amount of work done. Intel could have easily taken the existing Pentium line, improve it and had processors that accomplished more work for every tick. The P4 is better at FP at the same clock rating, but much worse at integer, by about 20%. The sole purpose of the P4 was to win the MHz battle. Most consumers are sheep and think that MHz is the end all. That is true of processors from the same manufacturer and family and that is it.
You could even say that Intel has their own marketing gimmick, it's called MHz.
First, who says the G5 needs nine fans? Are all nine fans for the processor(s)? No they are not, it has different cooling zones.
What makes the P4 so well designed? Intel lost the MHz battle with the PIII and they used the P4 to reclaim it. They are using MHz to get work done, AMD follows suit to a lesser degree though. The higher the MHz the more power is required and more heat is generated. While you can have a chip running at less than half the MHz, consumer half the power and get the same amount of work done. Intel could have easily taken the existing Pentium line, improve it and had processors that accomplished more work for every tick. The P4 is better at FP at the same clock rating, but much worse at integer, by about 20%. The sole purpose of the P4 was to win the MHz battle. Most consumers are sheep and think that MHz is the end all. That is true of processors from the same manufacturer and family and that is it.
You could even say that Intel has their own marketing gimmick, it's called MHz.
First, who says the G5 needs nine fans? Are all nine fans for the processor(s)? No they are not, it has different cooling zones.
Originally posted by alphaone
Hey Vlade, what's up with your 'tar? Are you saying that windows is good and macs are bad???
I'm laughing right now, and every time someone makes a comment about it. Take a guess what Operating system I really use if I have 500 posts on a mac forum
The Pentium 4 is poorly designed compared to the P3, as well as Apple and AMD's chip offerings. I'm not saying Intel didn't realize this, though. Although it definitely has nothing to do with being "simple". They intentionally threw together a processor that's poorly designed, but capable of higher clock speeds. They did it so they could hammer this whole "clock speeds are everything" concept into PC users' minds.
Of course, clock speed isn't everything. A Pentium 3 could easily beat out a P4 of the same, or even moderately faster, clock speed. Things like cache, RAM, and overall design are actually more important than clock speed. The P4 simply isn't a well designed processor. It's inefficeint, it gets very hot, they can't even put two of them into one tower without going to a whole lot of work.
Intel took a calculated risk in pushing the P4 over the P3, and it worked at smashing AMD and Apple at first. Now, the P4 is reaching its limit and AMD and Apple are catching up (mostly thanks to IBM on the Apple side). Intel is going to have a hard time moving its desktop systems ahead without doubling back on their "clock speed is everything" message.
Okay. Then I guess the Pentium 4 will never be in a laptop. It's too hot.
What?!
It already is in a laptop?!
Why, that would make your point silly and wrong!
Also, I in no way like the idiots at Moto's microprocessor division. I don't have aa "blind opinion" of the G4, and most people in these forums also don't. We realize the G4 is old, and getting slow. The G5 is better, and we're just at the beginning of a transition over to 64-bit computing, and more importantly, over to IBM processors.
Of course, the G4 is still better designed than the P4.
No, they're not all for the processor, nevertheless the G5 doesn't exacltly run cool. And more MHz doesn't mean the processor will be hotter or that it will require more power (the G5 comes with a 700 Watt power supply, and it only has one optical drive and usually only one hard-drive). The P4 is also viable in laptops, whereas the G5 won't be for a very long time. Back in the days of P3/P4 vs the original Athlon, Intel used to even critisize AMD because the chip ran hot and required a lot of power (even though it was faster and was clocked a lot lower than the P4). I don't agree with Intel's use of MHz for marketing but their combination of low performance per trick combined with a whole lot of ticks makes (made) it better than the competition. Intel also couldn't have just continued with the P3 as it doesn't scale very well at higher frequencies. A P3-M Centrino at 1.7 GHz is only as fast as a 2.4 GHz P4, and that's the fastest Intel can make the Centrinos go right now.
Oh, and if you're saying that the P4 is 20% faster at floating point than at integer calculations, then why did the G5 win (according to Apple anyways) against the P4 in FP tests but not integer tests?
A new P4 box will kill a G4 at almost anything. Photoshop included. Just look up some tests.
The G5 on the otherhand is pretty close for most stuff... sometimes ahead, sometimes behind... and really shines on altivec tasks.
The G5 on the otherhand is pretty close for most stuff... sometimes ahead, sometimes behind... and really shines on altivec tasks.
Originally posted by pyrotoaster
The Pentium 4 is poorly designed compared to the P3, as well as Apple and AMD's chip offerings. I'm not saying Intel didn't realize this, though. Although it definitely has nothing to do with being "simple". They intentionally threw together a processor that's poorly designed, but capable of higher clock speeds. They did it so they could hammer this whole "clock speeds are everything" concept into PC users' minds.
Of course, clock speed isn't everything. A Pentium 3 could easily beat out a P4 of the same, or even moderately faster, clock speed. Things like cache, RAM, and overall design are actually more important than clock speed. The P4 simply isn't a well designed processor. It's inefficeint, it gets very hot, they can't even put two of them into one tower without going to a whole lot of work.
Intel took a calculated risk in pushing the P4 over the P3, and it worked at smashing AMD and Apple at first. Now, the P4 is reaching its limit and AMD and Apple are catching up (mostly thanks to IBM on the Apple side). Intel is going to have a hard time moving its desktop systems ahead without doubling back on their "clock speed is everything" message.
Of course, the G4 is still better designed than the P4.
Have you considered that they made the chip what you call "poorly designed" for a reason? There are 2 main factors for a CPUs speed, MHZ and work per cycle. When multiplied, they give you an approximate speed. Intel PURPOSELY designed the P4 to have high clock speeds and a lower work per cycle, BUT when you multiply its 3.2GHZ it makes it a VERY VERY fast chip, way better than any G4, Athlon XP (the athlon 64 and G5 are challenging it now, its unclear what chip is the fastest, they all lead certain tests).
Tell me how a poorly designed chip is 2-4 times faster than a "better" G4 chip. Its ok to like the mac, but don't be a mac zealot
The Pentium 4 is a poorly designed chip. It's just intentionally so.
Intel opted for a compromised design because it allowed for very fast clock speeds. In turn, this gave Intel the ability to make faster processors (yes, I admit that openly). It doesn't change the fact that the P4 is poorly designed.
It was simply a short-sighted (although they might have very well realized that) way to quickly increase clock speeds and overall power. The catch is that the poor chip design would eventually make the P4 run far too hot, and become very difficult to make faster (although that eventually happens to most chip designs).
Just think about it. A few years back the P4 was jumping a GHz a year. Now it only gains a couple hundred MHz, and consumer machines almost never see the speed increase because the fastest P4s run too hot for a cheapo Dell or Gateway.
The G4 is a better designed chip than the P4. Of course, that doesn't mean its going to be faster (though within 500 MHz, the G4 will be faster).
You said it yourself, Intel PURPOSELY made the Pentium 4 poorly designed.
Originally posted by pyrotoaster
You said it yourself, Intel PURPOSELY made the Pentium 4 poorly designed.
I said it with quotes around it, I was just referring to when you said it, I didn't mean it. I meant to say that the P4 is a "inefficient per clock cycle chip", but its MHZ make up for it, therefor it is a good chip. An example is having a cluster of processors at lower clock speeds, such as a big computer cluster and a REALLY fast single chip. Your logic would suggest that the cluster is inefficient because has LOTS of computers at slow speed, and therefore the single fast efficient CPU would be better because it is designed better (despite the fact that it is 2-4 times as slow).
Sometimes it doesn't matter how the CPU maker does it, as long as the end result is good (and I don't care if my CPU runs hot, as long as it doesn't melt its OK with me, its in a box that I don't even touch)