Separate names with a comma.
Discussion in 'Mac Basics and Help' started by Sawtooth811, Dec 4, 2007.
Which is faster?
G4/1.6GHz Vs. Pentium 4/2.8GHz?
At what? Neither processor can be compared as they have no equivalent usage. You can't run OSX on a Pentium 4 and you can't run windows on a PPC processor.
Yes they can be compared! Of course, they'll be compared at benchmark apps! You know stuff like CINEBENCH and GeekBench (Which is inaccurate).
Just remember the megahertz myth.
It's all about calculations per cycle, not cycles per second.
id prolly put my money on the pentium
I think the G4 compares to a 2.2 Pentium 4 if anything. I think a 2.8 ghz would have a slight edge, then again, I'm kinda guessing.
Don;t forget the "Megaherz Myth"!
It's hard to run a test,
I tried comparing my sisters 1.33 ghz 12 inch PowerBook to my dads 2.2 ghz pentium 4 pc using Photoshop trial on both machines.
After 2-3 tests, the PC won each time with the Retouch Artists test by a couple of minutes.
Thing is, the PowerBook has a 4200 rpm hard drive, a slower hard drive. The 2.2 ghz Pentium has a 7200 desktop drive inside.
OS/x does run on the p4 .. been there done that, runs fantastic actually... id say the 2.8ghz p4 ( if it has a lotta cache) can maybe barely outrun a 1.6ghz g4
Right there, the HD speed, that's a big factor, because:
(1. The G4 is a laptop
(2. Like you said, the hard drive is slower
(3. The G4 has a terrible graphics MEMORY (Not Card) compared to the P4's VRAM.
Tell me the full specs of both Computers.
How come the 867MHz Solo G4 can be up to 80% faster than a 1.8GHz Pentium 4?
The Pentium 4 2.80Ghz did have HyperThreading, so it could run 2 threads if programmed correctly.
It would most likely walk all over the G4 at 1.6Ghz unless the application were AltiVec optimized.
I doubt it, because I'm talking about a Sonnet G4 7447, and I believe it's the second fastest G4. The G4 867 Apple was using that was up to 80% faster than a P4 1.8GHz. But then again, the later P4's like the 2.8GHz or 3.2GHz has HyperThreading, but I still believe the G4 is more powerful, although the P4 was power-hungry, also. Besides, my 450MHz G4 Sawtooth runs OS X.4 faster than a DELL Optiplex GX280 runs XP. I know why.
I believe that a G4 Dual 1.8GHz, or Single 2.0GHz will crush all P4's.
Why does it matter?
If the Pentium 4 is HT then I would say the Pentium is about 1.2-1.4 times faster than the G4. The G4's aren't all they're cracked up to be, yea, I'm very familiar with the megahertz myth but in all reality the G4 isn't quite as efficient as Apple claims them to be. I had a G4 1.0 Ghz and it wasn't really anywhere close to my Pentium 4 HT 2.8 Ghz (it was about half the speed). I'm basing this off of video encoding and mp3 encoding, and I do realize you have a 1.6 and not a 1.0 but thats gonna pick up that large of a gap I noticed in performance between the two. As for my C2D 2.4 Ghz its crazy fast compared to my old G4 and I'm very happy Apple switched to Intel of instead going with IBM or Motorola again.
Remember, the graphics card, Hard Disk, and RAM may have not been as efficient, good, or as fast as the P4's. Tell me the specs of both computers.
768 Megs ram
60 gig 4200 rpm hard drive
10.5 Leopard running CS3 Trial
HP Desktop Pavilion:
2.2 Ghz Pentium 4 No HTT
512 mbs ram
80 gig 7200 rpm hard drive
Windows Xp running CS3 trial
I think the PC got 15 minutes, the G4 got 18 minutes if I remember correctly.. (Retouch Artists Test)
Graphics card shouldn't really matter in these two situations because I don't believe G4s have ever been dependent on system ram and have always had their own separate video card in either an agp or pci slot, nor should HDD as long as they are both 7200 rpm (I guess one was ide and the other was serial, but again not a HUGE difference) as for memory the P4 had 512mb at the time and the G4 had 768mb so the G4 had the advantage. Oh and this is a G4 PowerMac not a PowerBook.
And another thing, the G4's also always a considerably lower FSB and therefore a slower memory compared to the Intels of the time so no matter how fast your processor upgrade is, its always going to be slightly hindered by this....I believe.
I wasn't mainly on VRAM, I was mainly on HD speed. The PC may have had a faster SATA HD, than the Mac's FAR slower Ultra-ATA HD. Plus, RAM is good for Multitasking speeds, not necessarily flatout speed.
Because the Motorola PPC processor uses only 7 pipeline stages and the P4 uses 20 stages. The pipeline in the PPC has a much shorter path to travel to complete the processing stage. Because the P4 has to travel so much further before processing is complete Intel had to increase the clock frequency to make it run close to or as fast as the PPC G4 that clocked at a much lower frequency.
So when you render a photo in Photoshop using both the P4 and G4 they may render a complex file nearly the same time and in many cases the G4 will beat the P4. Due to cache misses at times the G4 can easily beat the P4 because of the shorter path. The 20 stage pipeline in the P4 has a greater chance of having more cache misses henceforth creating the infamous hourglass.
This is the main reason why you should not buy on clock speed alone unless both CPU's are the same such as 2.2 Ghz Core 2 duo and the 2.8 in the new iMac.
But in terms of AMD where their processors are clocked way different and their cache is much smaller they can appear slower than the Intel when that's not the case.
P4's were a relatively big bust by Intel anyways. They believed they could sell lots of P4's by advertising 2-4GHz processors despite having ridiculous heat and power dissipation issues. Oh, they sold quite a few of them, having a 3.0Ghz model, then a 3.1, 3.2, etc. all the way up.
Then AMD snuck in with K8 and thoroughly trashed P4 by offering cooler, more power efficient solutions with lower clock speeds that could beat higher clocked P4's. At same clocks, AMD easily trounced Intel.
However, Intel came back with a vengeance with the Conroe family and its derivatives which beat AMD at their own game.
But, its a lot more complex than that. There are tons of factors you have to take into account. For one, the actual microarchitecture of the core makes a huge difference. The Core 2 Duo / Quad / Extreme lineups use one of the best designed architectures we've ever seen in CPU history. Extremely efficient and can be overclocked very well because they have lots of room to grow in speed / heat etc. The K8 architecture of the AMD Athlon 64's on the other hand easily beat the NetBurst disaster (the P4).
Also, cache sizes make differences. Increased cache sizes create dramatic performance increases for the Core 2 lineup, whereas cache size on the AMD Athlon 64 doesn't boost as much in performance both due to the fact that its architecture varies as well as the fact that it has an integrated memory controller.
There are tons more factors to put into play, but that's the reason why you can NEVER compare two CPU's next to each other unless they are within the same family (as in a Q6600 vs. a Q6700 vs. QX6800, etc.) then you can say w/ absolute certainty the differene in Mhz and how much in real time speed it matters.
Other than that, you'll have to benchmark using the same platform, only with different CPU's to verify speeds, but that's just not possible with ppc vs. intel.
Actually, if it is about winning I don't see AMD even competing.
These numbers are quite old but they show the crucial difference between Intel and AMD.
It was a very wise decision for Apple to go the Intel route.
you got that a little backwards .. intel introduced the much longer pipeline to increase the clock frequency
and "travel much further" ? if a pipeline is filled each CPU can complete one instruction per clock which means the P4 is going to win because of the higher clockrate
the disadvantage of the net burst architecture with it's long pipeline is the pipeline crashing when a branch is miss-predicted which means intel needed a very sophisticated branch prediction and a lot of fancy other stuff which simply costed too much die space and thus wasn't that energy efficient
that said my AMD Athlon Xp 2000+ with 1.66 GHz is faster than my G4 1.42 GHz in cinebench 10 so i suspect the P4 with 2.8 is faster than a G4 1.6
if we are talking about a P4 with hyperthreading the G4 is getting spanked no matter what