My P4 3.6 Faster Than C4Q In Running Non-Multi-Threads???
Last response: in CPUs
I've waded through several benchmarks, etc. for the C2Q and the Kentsfield, but I'm still unable to answer this question.
Most of the software I use is single-threaded, boring office-type stuff, with the exception of Photoshop. Currently I'm running a P4 3.6. But if I run the same software in a C4Q it's going to run at 2.66, right? So it will be noticeably slower. Or am I missing something? I understand that the CPU functions would be taken over by Core 1 while Core 2 processes my software, but is that going to give me a speed upgrade or downgrade? :?
Most of the software I use is single-threaded, boring office-type stuff, with the exception of Photoshop. Currently I'm running a P4 3.6. But if I run the same software in a C4Q it's going to run at 2.66, right? So it will be noticeably slower. Or am I missing something? I understand that the CPU functions would be taken over by Core 1 while Core 2 processes my software, but is that going to give me a speed upgrade or downgrade? :?
More about : faster c4q running multi threads
That's great to know! Thanks! In seeing the lower clock speeds of the C2s vs. P4s I was getting worried.
BTW, gOJDO, I like your new avatar. Have to admit that I was never too crazy about your old one. But that's because I'm a rightwinger from way back! Dobra vecer gospodin!
P.S. Wusy, I've seen 93C under load in Speedfan and 111C in the BIOS. The Preskillet is not a heater, it's a steel forge! :roll:
BTW, gOJDO, I like your new avatar. Have to admit that I was never too crazy about your old one. But that's because I'm a rightwinger from way back! Dobra vecer gospodin!
P.S. Wusy, I've seen 93C under load in Speedfan and 111C in the BIOS. The Preskillet is not a heater, it's a steel forge! :roll:
Related ressources
- Poll: R U still pre-C2D or up with the times? Discuss - Forum
- Would a P4 Socket 775 3.4 be faster than C2D E6700/X6800 in any way? - Forum
- pi score on P4 3.6 Ghz vs P-D 3.2Ghz - Forum
- Are quad core, duo core run faster than single core CPU? - Forum
- Is Turion 64 X2 TL56 much faster than a Pentium 4 @ 3GHz ? - Forum
Quote:
Currently I'm running a P4 3.6. But if I run the same software in a C4Q it's going to run at 2.66, right? So it will be noticeably slower. Or am I missing something? Obviously Intels brainwashing of Mhz good Ghz better is more insiduous that I thought.
Ok I fell for intels 'its more efficient' speil about C2D, but it is true, and yes more Ghz were faster, but the efficiency of new processors when it comes to executing instructions is such that you do not need vast numbers of Hz to be effective.
(I wonder what you call a collection of Hz's is it a melt? I think that a melt of Hertz sounds ok, any other suggestions.)
Obviously within a processor family more GHz is better, I wouldn't bother with Quad (yet) looking at toms reviews, but you could use it to get cheaper duo.
Would it really be too much to ask to extend what AMD seemed to be implementing but then fell out of with the FX series? A coherent numerical numbering scheme so regular Joes can figure out what is faster than what?
It seemed that Intel was going that route too with the 6300, 6400, etc. but there are huge, non-linear differences between the chips that don't adhere to the 100-jump numbering. And now Kentsfield is supposed to be a 6800??? Huh?
When I buy a car I can easily find out the engine displacement, mpg city, mpg highway, quarter mile time and top speed. Those are the numbers I need to know to understand the performance of what I'm going into debt for the next 5 years to finance.
Wouldn't it be nice if there was a weighted benchmark average, like the DJ Industrials. You take, say, the top 20 most common software that people use. Weigh the usage of each to come up with an algorithm. Factor in the benchmarks. Come up with a number. That's it. Yes, it would not necessarily apply precisely to the guy who 3D draws his life away, or the hardcore gamer, or the incessant pr0n downloader, but the vast majority of people who use PCs are not enthusiasts who are up on the latest SiSandra. They just wanna know how much bang for the buck they're getting in their new PC without having to qualify to write a doctorate thesis on comparative benchmarking.
C4Q is supposed to be 11,000 x faster than Pentium Pro. Ok, show me a number! Call it the C4Q 11000! Then when something comes out that's 10% faster than that (in our DJ Ind.-type benchmark avg.) then call it the C4Q 12100. Is that so damn difficult? Or just pick your baseline, pick your algorithm and stick to it!
But NOOOOOOOOOOOOOOOOOOoooooooooooooooooooooo. We have to have a bunch of code numbers that don't mean squat to most of the millions of people who shell out thousands of dollars for their systems and don't know whether inside the case there's silicon chips or a really smart hamster with a killer fast abacus inside.
There! Rant over!
It seemed that Intel was going that route too with the 6300, 6400, etc. but there are huge, non-linear differences between the chips that don't adhere to the 100-jump numbering. And now Kentsfield is supposed to be a 6800??? Huh?
When I buy a car I can easily find out the engine displacement, mpg city, mpg highway, quarter mile time and top speed. Those are the numbers I need to know to understand the performance of what I'm going into debt for the next 5 years to finance.
Wouldn't it be nice if there was a weighted benchmark average, like the DJ Industrials. You take, say, the top 20 most common software that people use. Weigh the usage of each to come up with an algorithm. Factor in the benchmarks. Come up with a number. That's it. Yes, it would not necessarily apply precisely to the guy who 3D draws his life away, or the hardcore gamer, or the incessant pr0n downloader, but the vast majority of people who use PCs are not enthusiasts who are up on the latest SiSandra. They just wanna know how much bang for the buck they're getting in their new PC without having to qualify to write a doctorate thesis on comparative benchmarking.
C4Q is supposed to be 11,000 x faster than Pentium Pro. Ok, show me a number! Call it the C4Q 11000! Then when something comes out that's 10% faster than that (in our DJ Ind.-type benchmark avg.) then call it the C4Q 12100. Is that so damn difficult? Or just pick your baseline, pick your algorithm and stick to it!
But NOOOOOOOOOOOOOOOOOOoooooooooooooooooooooo. We have to have a bunch of code numbers that don't mean squat to most of the millions of people who shell out thousands of dollars for their systems and don't know whether inside the case there's silicon chips or a really smart hamster with a killer fast abacus inside.
There! Rant over!
Agreed... It would be a lot easier to have consistent numbers, I remember the old PR ratings. The previous set of Intel numbers were designed to make it easier for the consumer I believe, obviously they have never actually spoken to a consumer and asked the question how many part numbers do you want for CPU, how many years of study do you want before you can order what you want...
At least we don't have E6310 Vs E6350 with some minor, but in 6 months time critical security features enabled or disabled. Heck it keeps the reviewers in a job...
At least we don't have E6310 Vs E6350 with some minor, but in 6 months time critical security features enabled or disabled. Heck it keeps the reviewers in a job...
Ugh, it hurts me when I go to stores (Fry's), and I see customers asking the sales technicians, so whats the true difference between this dual core and single core.
Customer: Which ones faster?
Sales people:The single core with a faster clock speed is faster that other single cores, but the dual core is faster.
Customer: Why, and how much faster. Does it justify the added premium?
Sales people: Uh... dual core is faster.
Customer: Whats the difference between 64 bit and 32 bit?
Sales people: 64 bit is better.
I'm standing there like... why do we even bother hiring these people. How about you fire all the sales rep, and cut prices. That would make me buy there so much more.
But then again, if it was that easy to understand computers, I wouldn't feel as elitist as I do when I tell my friends to wait for a new comp, and that prices will drop soon. Shrug
Customer: Which ones faster?
Sales people:The single core with a faster clock speed is faster that other single cores, but the dual core is faster.
Customer: Why, and how much faster. Does it justify the added premium?
Sales people: Uh... dual core is faster.
Customer: Whats the difference between 64 bit and 32 bit?
Sales people: 64 bit is better.
I'm standing there like... why do we even bother hiring these people. How about you fire all the sales rep, and cut prices. That would make me buy there so much more.
But then again, if it was that easy to understand computers, I wouldn't feel as elitist as I do when I tell my friends to wait for a new comp, and that prices will drop soon. Shrug
[quote="13thmonkey"]obviously they have never actually spoken to a consumer and asked the question how many part numbers do you want for CPU, how many years of study do you want before you can order what you want...
Yes, it would be too much to ask for a multi-billion dollar mega-corporation like Intel or AMD to actually spend 1500 bucks on running a focus group to find out that their average customer doesn't have the slightest idea what the hell they're selling!
@Doughbuy: Fry's, Best Buy, Circuit City, Future Shop, etc. hire amoebas. Incoming employees are given an IQ test. All those that fail are hired!
Quote:
Yes, it would be too much to ask for a multi-billion dollar mega-corporation like Intel or AMD to actually spend 1500 bucks on running a focus group to find out that their average customer doesn't have the slightest idea what the hell they're selling!
@Doughbuy: Fry's, Best Buy, Circuit City, Future Shop, etc. hire amoebas. Incoming employees are given an IQ test. All those that fail are hired!
Excuse the double-post but when something as momentous, earth-shaking and historically-turning-point as this concept is dawns on humanity, it must be brought to light without delay.
It was about time that I put my money where my mouth is, therefore... Presenting the Official Orsino Benchmark:
We take THG's excellent benchmark system as a basis as shown in the Kentsfield test.
The three Synthetics (SiSoftware Sandra 2007, PCMark05 Pro and 3D Mark 06) account for 6% each.
The two Multitaskings are 6% each.
The eight Applications (AVG, Winrar, Autodesk 3D, both Photoshops, Acrobat, Word and Powerpoint) have two more added. Repaginate a 500 page doc in Word and recalc a 50,000 cell sheet in Excel. Each one of these is 3% each.
The eight Videos are fine as they are. 2% each.
The four Audios are also fine as they are. 2% each.
The four 3D Games (as a gift to the avid gamers on this board) are 4% each.
Now let's pick a nice round number and a keystone processor to base it on. My vote would have to be to make the baseline CPU the C2D X6800 as the numbers can go up and down from this and it makes no sense to set your baseline anywhere but the fastest CPU today. It will only go up from here.
Therefore, the C2D has an Orsino # of A1000. Set up your algorithm, calculate the hell out of it and figure out all the other processors.
What happens when some of these apps no longer are applicable as they've been superceded? No prob! Change the benchmark constituents every couple of years and then change the first letter of the Orsino # to B! That should keep you happy for the next 50 years or so until CPUs don't exist and we do all our computing through our Borg Implants. Simple, huh?
It was about time that I put my money where my mouth is, therefore... Presenting the Official Orsino Benchmark:
We take THG's excellent benchmark system as a basis as shown in the Kentsfield test.
The three Synthetics (SiSoftware Sandra 2007, PCMark05 Pro and 3D Mark 06) account for 6% each.
The two Multitaskings are 6% each.
The eight Applications (AVG, Winrar, Autodesk 3D, both Photoshops, Acrobat, Word and Powerpoint) have two more added. Repaginate a 500 page doc in Word and recalc a 50,000 cell sheet in Excel. Each one of these is 3% each.
The eight Videos are fine as they are. 2% each.
The four Audios are also fine as they are. 2% each.
The four 3D Games (as a gift to the avid gamers on this board) are 4% each.
Now let's pick a nice round number and a keystone processor to base it on. My vote would have to be to make the baseline CPU the C2D X6800 as the numbers can go up and down from this and it makes no sense to set your baseline anywhere but the fastest CPU today. It will only go up from here.
Therefore, the C2D has an Orsino # of A1000. Set up your algorithm, calculate the hell out of it and figure out all the other processors.
What happens when some of these apps no longer are applicable as they've been superceded? No prob! Change the benchmark constituents every couple of years and then change the first letter of the Orsino # to B! That should keep you happy for the next 50 years or so until CPUs don't exist and we do all our computing through our Borg Implants. Simple, huh?
Quote:
I'm standing there like... why do we even bother hiring these people. How about you fire all the sales rep, and cut prices. That would make me buy there so much more.
Hey, while I don't work at Fry's not all salespeople are that stupid. We actually have a number of people who can explain the architectural differences between a PentiumD, AMD 64 X2 and a C2D.
Quote:
@Doughbuy: Fry's, Best Buy, Circuit City, Future Shop, etc. hire amoebas. Incoming employees are given an IQ test. All those that fail are hired!
As someone who works at one of those places I take offense to that. I do my job damn well and know what I am talking about. Most of the people in the computer department can easily tell you the difference between processors and most have built their own computers. Heck, most of the time if you want to know the motherboard in the tower we can tell you. Its that pervasive attitude that leads to constant arguing with customers who ask me a question then tell me that I'm wrong. (and my reply "If you weren't going to believe the answer I gave you why did you bother asking me?")
No offense meant, Exterous. It is impossible to generalize. In any group of professionals there is a wide range of expertise. I too have met the occasional big box employee who was extremely well informed. But you do have to admit that there are also a fair number who were laying sod the day before and got a new job to get out of the sun!
P.S. To prove how intelligent I am, I named this thread C4Q instead of C2Q. So follow my example!
P.S. To prove how intelligent I am, I named this thread C4Q instead of C2Q. So follow my example!
Quote:
My 3month holiday part-time job during the long, end of year semester break(providing there's no farm work I need to go).I work in a computer shop as a technician along with other long-term technicians. I can either blast them DEAD with my exotic computer knowledge or the stuff I learn in vet school if they don't want me to blabber on about computers.
Is that great???
I thought was the only farmer that came on here lol, so now we can talk about cpu architecture and things like spraying slurry and dosing cows
OK, then here's the deal, Wusy. I fly down to Christchurch, you show me around the Rings locations, then we fly to Taipei, pick out all the goodies, fly to Auckland, get royally pi$$ed at the best bars, build the system in the hotel room, I write you a five figure check and then we go our separate ways. Deal?
JumpingJack, thought you might like to see this.
84C Prescott @ 62% CPU
Exactly 5 minutes into running McAfee Virus Scan, Ambient Temp 21C. Zalman and 305mm Katrina fan on Max settings.
And I've seen 93C on Speedfan.
BIOS always runs about 17-19C hotter. So at this temp, BIOS would read about 102C.
When I trade up to my C2Q I'll keep this chip as a backyard steel forge.
Speedfan has readings that I can never figure out. Wot are the two Vcores and why is the second one so high?
84C Prescott @ 62% CPU
Exactly 5 minutes into running McAfee Virus Scan, Ambient Temp 21C. Zalman and 305mm Katrina fan on Max settings.
And I've seen 93C on Speedfan.
BIOS always runs about 17-19C hotter. So at this temp, BIOS would read about 102C.
When I trade up to my C2Q I'll keep this chip as a backyard steel forge.
Speedfan has readings that I can never figure out. Wot are the two Vcores and why is the second one so high?
Related ressources:
- ForumAMD states K8L aka Barcelona faster than all Intel cores
- ForumP4 2.8 c running very hot (asus p4c800 e deluxe)
- Forumoverclock e6600 at 3.6
- ForumHelp identify old P4 .
- Forumwhat the heck?!? Fear runs faster on MAX...
- ForumWhats better an older 3.6 ghz cpu vs. a newer 2.8 ghz cpu
- ForumOpteron 150 vs. Xeon 3.6 Nocona
- Forum2.4D Cel OC to 3.6 GHz
- ForumCan I run 64bit XP Pro or Server 2003 on P4 2.6Ghz CPU?
- ForumPrime 95... one core faster ?
- ForumCPUs don't make sense !
- ForumDoes Overclocking kill your CPU FASTER ?
- ForumPCIe 2.0 using only 1 video card, is it faster ?
- ForumFX-6200 vs FX-6300?
- ForumWhat's a normal C2D temp. when it's running hard?
- More resources
!
np.