Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Geek Culture / Processors!

Author
Message
Indicium
16
Years of Service
User Offline
Joined: 26th May 2008
Location:
Posted: 7th Dec 2009 23:00
Hey, im quite annoyed by the fact i can't really find anything on this.

Say i had two computers with the exact same specs, except one was using a intel pentium 4 2.4ghz processor, and one was using an intel celerone 2.4ghz processor, what would be the difference :S?
Phaelax
DBPro Master
22
Years of Service
User Offline
Joined: 16th Apr 2003
Location: Metropia
Posted: 7th Dec 2009 23:13
The biggest difference between the two is the amount of cache. P4 will have more cache, thus making them perform better.


> SELECT * FROM users WHERE clue > 0
> 0 rows returned
NeX the Fairly Fast Ferret
20
Years of Service
User Offline
Joined: 10th Apr 2005
Location: The Fifth Plane of Oblivion
Posted: 7th Dec 2009 23:25
I always thought that when Intel made processors not all were equal - microinconsistencies that make some more efficient than others. Coupled with extra cache and there's your high-end CPU.

Athlon64 2.7gHz->OC 3.9gHz, 31C, MSi 9500GT->OC 1gHz core/2gHz memory, 48C, 4Gb DDR2 667, 500Gb Seagate + 80Gb Maxtor + 40Gb Maxtor = 620Gb, XP Home
Air cooled, total cost £160
Phaelax
DBPro Master
22
Years of Service
User Offline
Joined: 16th Apr 2003
Location: Metropia
Posted: 8th Dec 2009 01:02
Depending on which P4 he has, it could also be using less power than the celeron. But as far as I'm aware, they have the same instruction sets (MMX/SSE)


> SELECT * FROM users WHERE clue > 0
> 0 rows returned
Toasty Fresh
17
Years of Service
User Offline
Joined: 10th Jun 2007
Location: In my office, making poly-eating models.
Posted: 8th Dec 2009 01:11
While on teh topic of processors, here's a question I can't seem to get right: Say you had a quad core 2.2ghz processor, because it is quad core does that mean it is effectively 8.8ghz?

NeX the Fairly Fast Ferret
20
Years of Service
User Offline
Joined: 10th Apr 2005
Location: The Fifth Plane of Oblivion
Posted: 8th Dec 2009 01:24
In a manner of speaking, but there's no guarantee that every tick on every core will be used and I can imagine that the more spread out a program is the more work needs to be done to keep a program in sync and thus the efficiency maybe begins to drop off as the core numbers reach the hundreds?

Though that is a long way off and I am getting far ahead of myself.

Athlon64 2.7gHz->OC 3.9gHz, 31C, MSi 9500GT->OC 1gHz core/2gHz memory, 48C, 4Gb DDR2 667, 500Gb Seagate + 80Gb Maxtor + 40Gb Maxtor = 620Gb, XP Home
Air cooled, total cost £160
Benjamin
22
Years of Service
User Offline
Joined: 24th Nov 2002
Location: France
Posted: 8th Dec 2009 01:25
I'd like to point out that comparison of clock rates only makes sense with processors of the same family, as much newer processors have a more efficient design and even with a lower clock rate can outperform older processors.
Phaelax
DBPro Master
22
Years of Service
User Offline
Joined: 16th Apr 2003
Location: Metropia
Posted: 8th Dec 2009 01:45
Especially when comparing 32 to 64 bit.

Most applications are only ever using 1 core, so most of the time quad cores are wasted until more companies start writing code to utilize them. It's kinda like having a dual processor board, not a whole lot applications would use both simultaneously. Studio Max would, and from what I've heard it was one of the most efficient programs at utilizing dual CPUs.


> SELECT * FROM users WHERE clue > 0
> 0 rows returned
Zotoaster
20
Years of Service
User Offline
Joined: 20th Dec 2004
Location: Scotland
Posted: 8th Dec 2009 02:09
Keep in mind that in Intel machines such as Pentium, it can take up to 800,000 transistors just to increase the PC register by 1... they could be far more optimized

"everyone forgets a semi-colon sometimes." - Phaelax
Dazzag
22
Years of Service
User Offline
Joined: 26th Aug 2002
Location: Cyprus
Posted: 10th Dec 2009 09:32 Edited at: 10th Dec 2009 09:36
Quote: "Say you had a quad core 2.2ghz processor, because it is quad core does that mean it is effectively 8.8ghz?"
No definitely not. The better something is programmed to take advantage of multi-cores then the closer it will come to your 8.8ghz target, but somethings just cannot be done.

On one end of the scale consider an algorithm that needs to calculate a million items, but each is dependent on the outcome of the last item. You cannot then split the task between the cores. So you are then effectively running at the slowest speed for that process. ie. a 2.4 Ghz single core (assuming architecture is identical to your quad core, and the rest of the PC is identical, with both software and hardware) should take about the same time to complete this task (any OS automatic clever logic for multi-core shannanigans may even make the quad slower as it attempts to automatically assign multiple cores instead of the single core that just gets on with things). Although running another app would probably then drag much more on the single core than the quad while the process runs

On the other hand say you are converting a video file. Assuming that each part of the file can be converted seperately then you could hand over each bit to a seperate core. If you have four cores then you should technically get close to four times the speed. I have a converter that will process a video on each core (not as good as my example granted) so if you have more than one video then it is a fair bit faster (disk access etc not taken into account there).

Although the other day I read something about Intel producing a demo 48 core processor. Nice. Would be great one day to have so many cores that each major process is handled by it's own core(s).

Cheers

I am 99% probably lying in bed right now... so don't blame me for crappy typing
Current fave quote : "She was like a candle in the wind.... unreliable...."
El Goorf
18
Years of Service
User Offline
Joined: 17th Sep 2006
Location: Uni: Manchester, Home: Dunstable
Posted: 10th Dec 2009 16:42
rule of thumb is go with the more expensive, unless you're looking at an inter-brand comparison

http://notmybase.com
All my base are not belong to anyone.
NeX the Fairly Fast Ferret
20
Years of Service
User Offline
Joined: 10th Apr 2005
Location: The Fifth Plane of Oblivion
Posted: 10th Dec 2009 19:26
Quote: "I have a converter that will process a video on each core (not as good as my example granted) so if you have more than one video then it is a fair bit faster (disk access etc not taken into account there)."


Rather than doing the video on the GPU and each audio channel on a different core?

Athlon64 2.7gHz->OC 3.9gHz, 31C, MSi 9500GT->OC 1gHz core/2gHz memory, 48C, 4Gb DDR2 667, 500Gb Seagate + 80Gb Maxtor + 40Gb Maxtor = 620Gb, XP Home
Air cooled, total cost £160
Dazzag
22
Years of Service
User Offline
Joined: 26th Aug 2002
Location: Cyprus
Posted: 11th Dec 2009 11:34
Quote: "Rather than doing the video on the GPU"
Last time I looked my video card was too old for the Nvidia GPU encoding (I believe it's the 8000 series that started it and mine is the 7000 series). Not that I'm that bothered as I do that sort of thing once in a blue moon. Although that still goes with the efficient idea of using multiple processors. As long as a task can be easily and efficiently split between processors and each task can run in parallel then you will get massively increased performance compared to a single processor. But not all tasks can be as easily split up as say video encoding (each processor handles a different part of the video), and even then not all programmes are programmed efficiently. eg. I have another video encoder that just uses one core on one file at a time. No advantage to multicores whatsoever really apart from using other apps at the same time are much slicker than on a single core.

Cheers

I am 99% probably lying in bed right now... so don't blame me for crappy typing
Current fave quote : "She was like a candle in the wind.... unreliable...."

Login to post a reply

Server time is: 2025-05-25 14:59:27
Your offset time is: 2025-05-25 14:59:27