Steelshivan

Got to thinking about how far processors have come and decided to play around a little bit with the calculator. I figured it would be neat to see what "effective" clock speeds newer processors might be at. For the AMD processors, I took their clock speed as 160% (or 1.6 times) the regular old Intel P4. For newer Intel stuff, I gave them 20% more speed than the AMD ones shown below. This is a very short list, thought I would share anyway. Please let me know if you think this is fairly far off or what I can do to make it more accurate. Thanks.

True CPU Speeds compared to Intel P4 at 3.0Ghz

Intel P4 3.0Ghz 3.00Ghz
AMD Athlon 3500+ 3.52Ghz
AMD Athlon 3800+ 3.84Ghz
AMD Athlon X2 6000+ 4.80Ghz per core
AMD Athlon X2 6400+ 5.12Ghz per core
Intel E8400 5.76Ghz per core
Intel Q6600 4.61Ghz per core

Edit: Due to overclocking, the above value for the Q6600 (especially compared to the X2 6000+) makes it look pretty damn good if you want a more future-proof processor. Just my opinion. Also, I realize there are minor fluctuations independent of several of the above processors vs the formulas I used, due to the different technologies used in the processors. Figured this was a decent ballpark chart, and it makes fairly good sense for what benchmarks I've seen.

ramodkk

That's a pretty good way to relate effective clock speeds but there's other stuff that make the CPU's different such as L2 Cache and architecture. And to be honest, using your method, the E8400 compared to the P4 at 3.0GHz would be about 10GHz

And remember that having dual cores does not mean that you have X.XXGHz per core. for example on your Athlon X2 6000+ it's 4.8GHz total, not per core.

But your concept is good

__________________
"If at first you don't succeed, you're not Chuck Norris"
E2160 @ 2.7GHz
AC Freezer 7 Pro
Gigabyte P35-DS3L
2GB BallistiX Tracers DDR2-1000
Sapphire HD3870
Rosewill RD500 500W
Rosewill Modded Case
3DMark06: 10,017
Winner of Photo Tourney: What were they thinking? & Grass

Steelshivan

That isn't right...and if it is, to say I missed the boat is an understatement.

The way you say it, a Q6600 has 4 cores, each at 600Mhz (the only way I can possibly get your total-besides, "totals" dont exist in multiple core processors, at least not until they're working 100% together, if that's ever going to be possible). The Q6600 has 4 cores, each at 2.4Ghz. Sure, only 1 or 2 may be in use at any given time, but it still has 4 at that speed.

ramodkk

That is not what I meant, I meant that the processor would be just 4.8GHz, period.

Here's how I see it.

Taking the Q6600 as an example, you don't have 2.4GHz total and neither do you have a 9.6GHz processor. You simply have a 2.4GHz processor based on four cores. Using only one core does not mean that core is going to run at 600MHz.

Take in this qutoe:

__________________
"If at first you don't succeed, you're not Chuck Norris"
E2160 @ 2.7GHz
AC Freezer 7 Pro
Gigabyte P35-DS3L
2GB BallistiX Tracers DDR2-1000
Sapphire HD3870
Rosewill RD500 500W
Rosewill Modded Case
3DMark06: 10,017
Winner of Photo Tourney: What were they thinking? & Grass

Steelshivan

Right, having two cars at 3mph each. Or 4 cores at 2.4Ghz each.

ramodkk

Yep, just remember you don't add them up. Mmany people think that when they hear "dual" or "quad" core.

__________________
"If at first you don't succeed, you're not Chuck Norris"
E2160 @ 2.7GHz
AC Freezer 7 Pro
Gigabyte P35-DS3L
2GB BallistiX Tracers DDR2-1000
Sapphire HD3870
Rosewill RD500 500W
Rosewill Modded Case
3DMark06: 10,017
Winner of Photo Tourney: What were they thinking? & Grass

Kewl Munky

It's easy to think that the cores add up with certain applications running faster. But they are only effective when you have applications that can use all the cores so that the application can have two or four cores loading it instead of one.

hermeslyre

When you consider multi-threaded apps, like kewl munkey mentioned, you'll have to determine performance differently than in simple multitasking terms. When looking at it this way you can add the cores up, in a sort, though not all SMP code is the same, and most SMP code is poor, so the end result tends to be most discordant.

Too bad we can't chart that up, it would require more than calculation, but active testing of specific platforms running specific code.. I regress, the calculation of the c2d chips as 20% more efficient clock for clock seems about right, according to the benchmarks I've seen. I've seen that number grow to as far as 30% as it is highly subjective towards what's being run. The wolfsdale should have a slightly better core architecture as well, elevating it higher than it's older kin.

Steelshivan

Thanks for the input last poster. I'd like to stress that this thread is only about my take on effective core speeds relative to the original P4 for each processor, it has nothing to do with adding up the cores or anything else...just how fast each CORE is. Obvously if I left out the fact that the dual and quad-core processors had more than one core, I would have been "corrected" on that...hence my X2 and X4 afterwards.

If someone wants to start a thread on multi-core processors and try to use benchmarks and tons of variables/stats to determine percentages of certain cores being used, and in what fashion, go for it. I'll probably post there with my input. But this isn't the thread for it.

I'm sorry if I sound harsh, but I'm not a big fan of when misunderstood people put words in my mouth.

ramodkk

Yep, sorry bro, I went off-topic for a moment...