Has Clock Speed Outlived Its Usefulness as a Processor Specification?

lcpiper said:
Why?

I've made more then enough HD decisions based solely on the spindle speed.
Click to expand...

It all depends on what you want. If you want high performance and don't care about much else, 10,000RPM is your friend. If you want longevity and solid reliability for the long term, it is not. That's the problem with trying to rate and compare things on a component-level basis. It depends on what you're trying to do with it. Just because System A has a faster processor core doesn't mean it's the best for what you're trying to do. It's not as simple as it was during the Pentium days.
 
rudy said:
He mentioned he didn't think that the company was intentionally deceiving consumers. I beg to differ that is EXACTLY what intel and the OEMs are doing. Ever since we shifted to this mind numbing dumb i3/i5/i7 naming scheme we have seen an explosion in moronic system configurations.
Click to expand...

Man I agree with that. Been building my own machines and OC'ing since a pentium 233, but today if I just look at the processor's names, I have no fucking clue what they are. Quad core, quad core hyperthreading, what their clock speed is...all hidden, and with the variety in models but very similar names, it can get damn confusing. Last time I upgraded my CPU I had to spend a while just figuring out what was Sandy, Ivy, Haswell, hyperthreading, not HT,...jesus.
 
Phoenix333 said:
It all depends on what you want. If you want high performance and don't care about much else, 10,000RPM is your friend. If you want longevity and solid reliability for the long term, it is not. That's the problem with trying to rate and compare things on a component-level basis. It depends on what you're trying to do with it. Just because System A has a faster processor core doesn't mean it's the best for what you're trying to do. It's not as simple as it was during the Pentium days.
Click to expand...

Which is why no one uses 10K drives for Enterprise SANs?
 
TwistedAegis, maybe it just takes some keeping up.

http://en.wikipedia.org/wiki/Intel_Core

One could easily argue that Intel themselves has decided the same thing and that the new naming conventions are an effort to fix the problem. Really, it can't be too hard to research the new offerings and which computers offer them before buying. It took me less then a minute to find this wiki site, only a couple to spot the difference between an I3 and i7, and a moment longer to see the evolution of the difference series of chips. At that point I just have to get some information on which computers offer what I want or print the page to shop on the fly.

It was sort of like researching the different option packages available for the 2015 Challenger before I went looking to buy one.

It is true, the times have changed, the options are many. If there is a complaint it's that not all manufacturers advertise what it is they are selling very well. But I think the good ones still do.
 
Sure, if I take a moment to brush up, I can figure it out. But even remembering the naming conventions...was it the 3570, a 3570S, a 3570K, a 3570M...or was it a 4670?

Those of us on [H] have no issue after doing a little research, but I had to do more than my usual to catch up and figure the differences between Sandy, Ivy and Haswell. For a non-[H] person, I can't imagine the gobbledigook it appears to be.
 
Its all a game.

I could go on explaining companies targeting ignorant consumers or ones that simply don't know how to research properly. They also love the consumers who get involved and research too much looking at products that are not even out to drive the hype machine.

Has the clock speed of a processor outlived its usefulness as a processor specification?
No. Has it outlived its purpose as an advertising specification? Probably, but that is not the question. Take for instance those poor souls who render images with processors tell them to down clock their CPUs! They might give you a tasteful response since their rendering times increase greatly finally pushing them to use things like iRay finally. The CPUs clock speed is really suppose to be the main difference in speed in a given processor series. Do companies screw this up by making an almost impossible naming scheme with letters and numbers? If I said no to that question I deserve to burn!
 
lcpiper said:
TwistedAegis, maybe it just takes some keeping up.

http://en.wikipedia.org/wiki/Intel_Core

One could easily argue that Intel themselves has decided the same thing and that the new naming conventions are an effort to fix the problem. Really, it can't be too hard to research the new offerings and which computers offer them before buying. It took me less then a minute to find this wiki site, only a couple to spot the difference between an I3 and i7, and a moment longer to see the evolution of the difference series of chips. At that point I just have to get some information on which computers offer what I want or print the page to shop on the fly.

It was sort of like researching the different option packages available for the 2015 Challenger before I went looking to buy one.

It is true, the times have changed, the options are many. If there is a complaint it's that not all manufacturers advertise what it is they are selling very well. But I think the good ones still do.
Click to expand...

It's harder when you're comparing processors that, if you were going to use the name as a reference, wouldn't be comparable at all. For example, some Core i3's of the U variant vs. Bay Trail Celerons.
 
schizrade said:
What are you blabbering about?

I said I sure could use a factory 5ghz chip for certain application workloads I have, and now you are going on about how it would not help at all. I have OC'd a 4790k to 4.8 from 4.0 and in the applications in question (OCR) the output just goes up. on a 200,000 page doc, 4ghz vs 5ghz will make a sizeable time difference. Is there a limit to the benefit for faster clocks? Sure. But I see a 25% reduction in time from 4-4.8 ghz in OCR.

So It sure as shit would benefit me.
Click to expand...

It sure is nice how you ignore context just so you can insult someone else for "needing to have something to say."

Clock speed/Speed of Light speed limits and the ability to increase the clock are dependent on die size.

Both for reasons of heat and signal propagation.

After a certain point, shrinking isn't possible which is why they need to go 3D: build UP as the die itself cannot expand outward without running into signal timings.

When dies used to be larger and built on a much bigger process node, 5Ghz was causing problems. The same problems I described earlier in the thread in which you derided me for not knowing what I'm talking about.

For timings to remain constantly in sync, electrical signaling distances decrease the higher the clock speed of the CPU is.

Your response here like you just wanted to shove a stick up your ass and get all offended without considering what is actually being said.
 
VIC-20 said:
That isn't a new issue either as laptops are completely custom one off designs, not carbon copies of a reference design from Intel with nothing changed but a sticker.

Take a look at this round up of Pentium 233 MMX laptops in 1997 and explain to me why the CPU score isn't identical

https://books.google.ca/books?id=oO...w#v=onepage&q=thinkpad 770 benchmarks&f=false

The cooling designs that OEMs and ODMs use are typically atrocious and change almost weekly. But that's just one piece of the puzzle. Location of memory, chipset, I/O, are all different and is bound to have some effect on system performance.

Brand name PCs, even desktops, should always be reviewed as a complete entity. Period. Specs don't matter if the manufacturers can change whatever they like.
Click to expand...

Please tell me you havent been hanging onto that article for way to long just in case this thread happened :)
If not couldn't you seriously find an example of something in the last decade(s)
 
Clock speed hasn't been my performance metric since Tualatin - Willamette performance flop.
 
Laureolus said:
Which is why I said that OEM/Manufacturer should have to list what their device can actually do at, say 23c ambient, instead of the chip's stock speeds.

But that's require some truth in advertising. :(
Click to expand...

Ooooh, I understand. It'd be nice to have a standardized performance metric, but that's not really in the best interests of OEMs to publish so I kinda doubt we're ever gonna get something like that. We'll just have to keep reading total platform reviews on products to get an idea of how things perform. Anandtech has a huge database of somewhat comparable benchmarks that are super useful for that kinda thing.
 
IMHO, clock speed alone was never a particularly good indicator of performance.

I mean, back in the 286 days both intel and AMD manufactured 286's were identical so you could compare those, but you couldn't compare a 286 clock to an 8086 or 8088 clock

Every generation after that, 386, 486, Pentium/586, Pentium II/K6, Pentium III/Athlon, Pentium 4/Athlon XP etc. etc. only got worse as far as the usefulness of comparison of clock speeds even among chips of the same generation, let alone against chips of previous generations.

They were still roughly close enough that one could use a rule of thumb clock speed comparison of the same generation up through the first Athlon's and Pentium III's but once the Pentium 4 came out in 2000, 15 bloody years ago, clock speed alone became an utterly useless way of comparing any two chips not of the same architecture/family.

This Digital Trends site isn't quite up to date with the trends. This article may have made sense if it were published in 2000, but that was a long time ago now, but even then it should have come with huge caveats, as clock speeds were not a perfect comparison even before the P4 launch.

Kids who were 1 year old in 2000 are now driving cars. That fact alone demonstrates how useless this publication is.
 
I think it's a good indication of how processors compare among other like model processors. But, there are so many different models, that it's becoming less of a guide. An i3 vs an i7 at the same speed doesn't mean a whole lot. Even i7's can vary depending on the submodel (mobile, low power, etc.). I don't mind throttling, but some consumers would need to. [H] folks generally won't have their CPU throttled more than likely. We take out time and put effort to make sure it doesn't happen through better cooling, wire management to get the best airflow, etc..

I have a Pentium 4 @ 3.4 that is extremely slow compared with a Core i7 @ 3.4. So, that doesn't work. An i3 vs. i7 at the same clock speed can even be a big difference.

Clock speed is one indicator of performance, but you need to take in the other factors as well.
 
You must log in or register to reply here.
Back
Top