How fast is a mac compared to a pc?
- Thread starter Convert
- Start date
Viro
Registered
Look at it this way, CPU clock speed (Ghz/Mhz) is a very poor indicator of performance. It's almost like saying that my car is faster than your car because mine revs up to 8000 RPM while yours only revs up to 5000 RPM. A lot more factors are involved in determining the car's performance. Same things with CPUs.
Now on to the P4. Intel designed this processor to scale up to very high clock speeds and basically have a very big GHz number. In order to do this, Intel had to do lots of hacks to the CPU and the result is a processor that doesn't do as much per Hz than other processors. That will explain why an Athlon 64 3200+ (which is actually a 2GHz processor) performs more or less equal to a P4 3.2 GHz.
Architecture wise, the G5 is very similar to the Athlon64/Opteron. You can expect the performance to be very similar, when all things are equal. The G4 and G3 processors are roughly similar to the P3 so you'll get more or less the same performance with an equally clocked P3.
But as with all things in life, not everything is equal. Most software is written to take advantage of the x86 (i.e. P4, Athlon, etc) architecture and a straight port to the PPC (G3, G4, G5, etc) architecture doesn't always result in superb performance. The quality of compilers for the PPC architecture isn't as good as those for x86. This is quickly changing with the work Apple and IBM are doing so it may not be a factor for much longer.
There are somethings that G4 and G5 chips can do very very well. One of these thigns is encryption. On some algorithms, a 800 MHz G4 can perform better than P4 5 GHz (if one exists). RC5 is an example of such an algorithm. Take a look at http://www.distributed.net for some benchmark numbers. Anything that can be optimized with Altivec such as digital signal processing (media encoding, image processing, sound processing, etc) will show a very significant boost in performance.
To sum things up, its hard to compare PCs to Macs. Just be happy with the Macs and know that you aren't running Windows, an OS that should be forced by law to carry a warning that it can be hazardous to use.
Now on to the P4. Intel designed this processor to scale up to very high clock speeds and basically have a very big GHz number. In order to do this, Intel had to do lots of hacks to the CPU and the result is a processor that doesn't do as much per Hz than other processors. That will explain why an Athlon 64 3200+ (which is actually a 2GHz processor) performs more or less equal to a P4 3.2 GHz.
Architecture wise, the G5 is very similar to the Athlon64/Opteron. You can expect the performance to be very similar, when all things are equal. The G4 and G3 processors are roughly similar to the P3 so you'll get more or less the same performance with an equally clocked P3.
But as with all things in life, not everything is equal. Most software is written to take advantage of the x86 (i.e. P4, Athlon, etc) architecture and a straight port to the PPC (G3, G4, G5, etc) architecture doesn't always result in superb performance. The quality of compilers for the PPC architecture isn't as good as those for x86. This is quickly changing with the work Apple and IBM are doing so it may not be a factor for much longer.
There are somethings that G4 and G5 chips can do very very well. One of these thigns is encryption. On some algorithms, a 800 MHz G4 can perform better than P4 5 GHz (if one exists). RC5 is an example of such an algorithm. Take a look at http://www.distributed.net for some benchmark numbers. Anything that can be optimized with Altivec such as digital signal processing (media encoding, image processing, sound processing, etc) will show a very significant boost in performance.
To sum things up, its hard to compare PCs to Macs. Just be happy with the Macs and know that you aren't running Windows, an OS that should be forced by law to carry a warning that it can be hazardous to use.
blue&whiteman
PowerPC Fanatic
macs use risc (reduced instruction) cpus vs wintel machines that use cisc (complex instruction) cpus. in virtually every case a cisc will lose to a risc of the same mhz because risc cpus do more with less so to speak.
the fastest mac cpu at the moment is 2.5GHz vs 3.4GHz (or higher?) for pc.
the fastest pc's right now are a bit faster at things like gaming over the fastest macs. remember though that macs are made for digital arts in large part so the hardware will always keep things like photoshop or 3D animation running faster. pixar uses macs.
both platforms have their place but just remember that the whole windows experience is pretty much just a bad copy of the mac gui. gates has been biting apples tech for years.
the fastest mac cpu at the moment is 2.5GHz vs 3.4GHz (or higher?) for pc.
the fastest pc's right now are a bit faster at things like gaming over the fastest macs. remember though that macs are made for digital arts in large part so the hardware will always keep things like photoshop or 3D animation running faster. pixar uses macs.
both platforms have their place but just remember that the whole windows experience is pretty much just a bad copy of the mac gui. gates has been biting apples tech for years.
ElDiabloConCaca
U.S.D.A. Prime
Yes. Macs are faster than PCs. You can argue all you want about clock speeds and front side busses and what have you, but when it comes down to it, it's all about the experience, right?
When you're working on a fast computer, it's fun -- it's neat to see tedious processes finish in 12 seconds. It's cool to have a computer that can process a word document in 4 seconds. It's cool to say, "I pull 12.4 gigaflops with my computer!" But imagine if you actually liked using the computer. Sometimes I hop on my Mac just to hop on my Mac. I don't wanna surf the internet, I don't want to read PDF files, I don't want to do anything with my newsgroups. I don't wanna be available on iChat. I just wanna have a little Mac experience for a minute.
Macs may be faster or slower than PCs, but not much in either direction. They're about equivalent. The Mac will spank the PC at certain things, and the PC will spank the Mac at (less important) certain things, but in the end, they're about the same. But would you rather be using Mac OS X or Microsoft Windows? If you answered "Microsoft Windows," then you haven't used Mac OS X yet.
When you actually enjoy working on your computer, then that means that your computer is the fastest. It doesn't matter if your PhotoShop filter takes 2.3 seconds on a PC, but 2.5 seconds on a Mac... which would you rather be using? Which would make the day go by quicker? Which would you look forward to using? Which would make work seem like play, and which would make work seem like hell?
Macs are faster. Don't believe the hype.
When you're working on a fast computer, it's fun -- it's neat to see tedious processes finish in 12 seconds. It's cool to have a computer that can process a word document in 4 seconds. It's cool to say, "I pull 12.4 gigaflops with my computer!" But imagine if you actually liked using the computer. Sometimes I hop on my Mac just to hop on my Mac. I don't wanna surf the internet, I don't want to read PDF files, I don't want to do anything with my newsgroups. I don't wanna be available on iChat. I just wanna have a little Mac experience for a minute.
Macs may be faster or slower than PCs, but not much in either direction. They're about equivalent. The Mac will spank the PC at certain things, and the PC will spank the Mac at (less important) certain things, but in the end, they're about the same. But would you rather be using Mac OS X or Microsoft Windows? If you answered "Microsoft Windows," then you haven't used Mac OS X yet.
When you actually enjoy working on your computer, then that means that your computer is the fastest. It doesn't matter if your PhotoShop filter takes 2.3 seconds on a PC, but 2.5 seconds on a Mac... which would you rather be using? Which would make the day go by quicker? Which would you look forward to using? Which would make work seem like play, and which would make work seem like hell?
Macs are faster. Don't believe the hype.
Viro
Registered
blue&whiteman said:
A generalization that isn't strictly true these days. Most PC processors are RISC behind the scenes. The P4 has 120 rename-registers that allow it to convert complex x86 instructions into RISC-like microcode. That is why they can perform that well.
The CISC vs RISC war was decided long ago. RISC has its short comings. CISC has its short comings. That's why most processors these days use a hybrid approach to computing. The G4 and G5 aren't strictly RISC processors either especially since they have Altivec.
Viro
Registered
ElDiabloConCaca said:
I agree that it's all about the experience. That is the only reason I'm using Macs instead of PCs. I like Macs much more than PCs. On my Mac, Things Just Work (tm). I end up spending less time fiddling with settings and more time getting work done.
However, I disagree that Macs are faster than PCs. That may hold true with media apps, but it falls flat in other areas. Games don't run as fast on Macs as they do on PCs. Head over to http://www.barefeats.com and look at the benchmarks there.
PCs run mathematics software faster than Macs do. Looking at MATLAB bench, a single P4 3.4 GHz runs almost twice as fast as a dual G5. I'm including the result of the benchmark here.
While Macs may not be as fast as PCs (no use in denying that), they are fast enough for me and I've found them much more usable than PCs. Your mileage might vary, of course
Attachments
Zammy-Sam
Desertchild
The reason I actually moved to macs was not because of the design or macosx, but only because of the performance of blast-alignment tools such as BLAST and CLUSTALW.
Even in those days it was known that PowerPC processors perform a lot better in this field than other systems including the x86 processors. Apple published these results:
and I totally agree. My tibooks performs a lot better in such alignments than our Dell 3.06Ghz HT box.
As actually everyone pointed out here: it all depands on the area and field you need the performance. And when it comes to non-professional use, every current mac is fast enough - if not faster - for that purpose and macosx is the much better companion.
Even in those days it was known that PowerPC processors perform a lot better in this field than other systems including the x86 processors. Apple published these results:
and I totally agree. My tibooks performs a lot better in such alignments than our Dell 3.06Ghz HT box.
As actually everyone pointed out here: it all depands on the area and field you need the performance. And when it comes to non-professional use, every current mac is fast enough - if not faster - for that purpose and macosx is the much better companion.
It depends on what you are doing with the machine. For instance, SETI@home on my Sawtooth G4 upgraded to 1.4GHz, 100 MHz system bus is about the same speed as my P4 1.5 GHz(1.4 overclocked to 1.5) 133MHz system bus at work.
This is without any processor-specific optimizations.
For Photoshop and anything that uses Altivec, the G4 & G5s are going to be substantially faster.
Also, the G5 is very fast for processing huge chunks of parallel data such as video, protein folding, etc. due to the ultra fast system bus and memory bandwidth.
This is without any processor-specific optimizations.
For Photoshop and anything that uses Altivec, the G4 & G5s are going to be substantially faster.
Also, the G5 is very fast for processing huge chunks of parallel data such as video, protein folding, etc. due to the ultra fast system bus and memory bandwidth.
If you monitor the memory activity of the protein folding (I'm guessing you're referring to Stanford's F@H project) it doesn't do frequent reads/writes. Whereas the SETI client does do a lot of memory read/write so it would benefit more from highspeed system bus.Captain Code said:
pds
Registered
Now, I just did the definitive test between a G4 1.25 and a Dull P4 3.2 and found them to be exactly equal.
Both had a gig of RAM and 120gig HD. The peecee had a GeForce with 64 megs of vram and the powermac had the standard issue ATI. They both had 21 inch Sony monitors and were set to run a special crossplatform benchmark - ICNIVAD. I set them up on the sill and launched them at the exact same time and they reached the exact same speed by the time they hit the ground from the 45th floor.
So a 1.25 G4 is just as fast as a 3.2 P4
Both had a gig of RAM and 120gig HD. The peecee had a GeForce with 64 megs of vram and the powermac had the standard issue ATI. They both had 21 inch Sony monitors and were set to run a special crossplatform benchmark - ICNIVAD. I set them up on the sill and launched them at the exact same time and they reached the exact same speed by the time they hit the ground from the 45th floor.
So a 1.25 G4 is just as fast as a 3.2 P4
ElDiabloConCaca
U.S.D.A. Prime
Yes! I will now refer to this thread every time someone asks me which is faster... hehe... now, pds, can you perform this same test on the moon and tell me if both are still as fast as they are on earth?
pds
Registered
Well, we are talking relative speeds here, are we not? There will have to be some adjustments to the ICNIVAD benchmark, but if we can call terminal velocity an absolute, relative to gravitational mass it could be that they are absolutely as fast. But if were to go to the moon, where would I find the 45 storey building?