Overclock.net › Forums › Intel › Intel CPUs › Why is the i5's Performance so much better even though stats are the same?
New Posts  All Forums:Forum Nav:

Why is the i5's Performance so much better even though stats are the same? - Page 5

post #41 of 49
Quote:
Originally Posted by crUk View Post
Clock speed is a marketing plot it does not denote how much performance a chip has.
Its like the megapixel race in cameras specially in point and shoots.
What the marketing departments do not quantify is how efficient a processor is per clock cycle. If the microcoding is equivalent, a faster clock speed yields more performance. But processors are not created equally. For years, RISC processors blew the doors off of CISC processors, even if they were limited in clock frequency. For instance, an 800MHz Power processor would outperform a Pentium III at 1.8GHz - a factor of about 2.3. That stemmed from the fact that for most instructions, a Power processor accomplished in one clock cycle what a Pentium would, on average, require 2.3 clock cycles.

Intel has been forced to adopt more and more RISC like strategies - so their current processors that to a programmer / developer appear to be CISC, because they have a CISC instruction set - are actually RISC-like internally, especially since the vast majority of developers never actually do any low level functions, but rather, do their programming with high or mid-level programming languages, quite often programming through API interfaces rather than with the actual hardware.

The numbers do not mean much for general purpose use. For the most mart, the vast majority of users will never actually see any difference between say, a Phenom II versus a Sandy Bridge. Most users are using light weight tasks that do little computation, tasks that are speed limited not by the processor, but by peripherals. Like, what difference does it make in printing speed, since USB is the speed limitation, or the printer can only digest so much at a time - or for web browsing, where it is network latency that makes all of the difference. The only difference might be with gamers - since they will push the envelope and the processor difference may add a FPS or so. Or with people doing heavy duty rendering for animation or CAD - which is a small segment of the market as well.

It is much like cameras, since megapixels mean little since most people are not creating billboard sized signs out of their pictures. Most people will view their pictures on their computers - displays that are limited to 75DPI. They may print their pictures, in which 300DPI is entirely sufficient because really, higher DPIs are possible but not discernible by the eye. In fact, most people will end up borking their high-megapixel pictures by having bad settings which overcompress their jpegs - rather than going high quality with RAW (or liberal jpeg settings which do not dither the colours into submission).

This has long been a game in the computer industry, pushing out meaningless numbers in the everlasting contest of overkill.

As for something like the i5 - internal microcode that yields more instructions that execute in a single clock cycle, couples with more efficient superscalar operations - make up the performance difference. But an i5 is still borked by having to interface with slow DRAM, slow hard drive interfaces, and the constant need to play nursemaid for all of the other interfaces, like USB that really uses lots of CPU.
post #42 of 49
Quote:
Originally Posted by Derp View Post
How so? The Q9650 beats the phenom II X6 in most games.

@OP, the explanation is that AMD processors have been slower than Intel processors for a frighteningly long time now. For your needs the 2500k should be your only choice, anything else, especially ANYTHING currently sold by AMD would be a waste of time.


Without OC'ing the IMC, yes it is slower in non-multithreaded enviroments (it beats it completely in multithreaded applications anyways.) However, when you do, it beats the Q9650 clock for clock and core for core in both non-multithreaded and multithreaded applications.

Also, games obviously aren't the only applications that people use.
Edited by Nutty - 8/2/11 at 8:09am
    
CPUMotherboardGraphicsRAM
AMD Phenom 9150e @ 1.056v for now. Foxconn Bengel RS780 XFX GT 240 @ 550-1651-3646 Samsung 4GB @ 800mhz 6-6-6-18 
Hard DriveOSMonitorKeyboard
586GB Windows Vista 64bit 19 inch Gateway 1440x900 Gateway keyboard 
PowerCaseMouseMouse Pad
300w Bestec Gateway Case Razer Lachesis Dell mouse pad 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Phenom 9150e @ 1.056v for now. Foxconn Bengel RS780 XFX GT 240 @ 550-1651-3646 Samsung 4GB @ 800mhz 6-6-6-18 
Hard DriveOSMonitorKeyboard
586GB Windows Vista 64bit 19 inch Gateway 1440x900 Gateway keyboard 
PowerCaseMouseMouse Pad
300w Bestec Gateway Case Razer Lachesis Dell mouse pad 
  hide details  
Reply
post #43 of 49
Quote:
Originally Posted by Yugimt View Post

Also, games obviously aren't the only applications that people use.
Quote:
Originally Posted by JackC64 View Post
i'm gonna do like 95% gaming on this thing.
During your defense of AMD's slow CPU's did you bother reading the OP's posts? The post you made calling someone else a troll for comparing the phenom II X6 to the Q9650 in a game was absolutely valid.
post #44 of 49
Quote:
Originally Posted by Yugimt View Post


Without OC'ing the IMC, yes it is slower in non-multithreaded enviroments (it beats it completely in multithreaded applications anyways.) However, when you do, it beats the Q9650 clock for clock and core for core in both non-multithreaded and multithreaded applications.

Also, games obviously aren't the only applications that people use.
AMD Phenom II X6 1090T - 3.2GHz - 3MB L2 - 6MB L3 VS Intel Core 2 Extreme QX9770 - 3.2GHz - 12MB L2
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
post #45 of 49
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
post #46 of 49
Quote:
Originally Posted by $ilent View Post
Are you joking?...Surely you know what HT is?

@OP i5 2500k can be used to fold bigadv, you need native linux to run it. performance/$ wise 2500k is unbeatable at the minute.
I have reason to suspect that user Orion has been out of the loop for a while. Understandable... Look at when he registered, 2005.
Mugendramon
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5 4670k ASrock Z87 Extreme6 R9 290 Crossfire Samsung 8GB DDR3 2133mhz 
Hard DriveHard DriveCoolingCooling
WD Black 1T Samsung Spinpoint F3 1 TB EK Supreme HF Rev 2.0  HW GTX 360 Rad 
CoolingOSMonitorPower
EK Acetal waterblock W7 64 bit Yamakasi Catleap Q270 2560x1440p Corsair AX850 
CaseAudio
Corsair 800D Creative Titanium HD  
  hide details  
Reply
Mugendramon
(18 items)
 
  
CPUMotherboardGraphicsRAM
i5 4670k ASrock Z87 Extreme6 R9 290 Crossfire Samsung 8GB DDR3 2133mhz 
Hard DriveHard DriveCoolingCooling
WD Black 1T Samsung Spinpoint F3 1 TB EK Supreme HF Rev 2.0  HW GTX 360 Rad 
CoolingOSMonitorPower
EK Acetal waterblock W7 64 bit Yamakasi Catleap Q270 2560x1440p Corsair AX850 
CaseAudio
Corsair 800D Creative Titanium HD  
  hide details  
Reply
post #47 of 49
Quote:
Originally Posted by Derp View Post
During your defense of AMD's slow CPU's did you bother reading the OP's posts? The post you made calling someone else a troll for comparing the phenom II X6 to the Q9650 in a game was absolutely valid.
You just dodged the whole top part of my post.

Read my post again.
    
CPUMotherboardGraphicsRAM
AMD Phenom 9150e @ 1.056v for now. Foxconn Bengel RS780 XFX GT 240 @ 550-1651-3646 Samsung 4GB @ 800mhz 6-6-6-18 
Hard DriveOSMonitorKeyboard
586GB Windows Vista 64bit 19 inch Gateway 1440x900 Gateway keyboard 
PowerCaseMouseMouse Pad
300w Bestec Gateway Case Razer Lachesis Dell mouse pad 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
AMD Phenom 9150e @ 1.056v for now. Foxconn Bengel RS780 XFX GT 240 @ 550-1651-3646 Samsung 4GB @ 800mhz 6-6-6-18 
Hard DriveOSMonitorKeyboard
586GB Windows Vista 64bit 19 inch Gateway 1440x900 Gateway keyboard 
PowerCaseMouseMouse Pad
300w Bestec Gateway Case Razer Lachesis Dell mouse pad 
  hide details  
Reply
post #48 of 49
Quote:
Originally Posted by Yugimt View Post
You just dodged the whole top part of my post.



Read my post again.
I read it. I just posted benchmarks link so people could see.
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Core i7 2600K@4.5GHz Asus P8P67 Pro 3.0 (B3) EVGA GTX 1070 SC Black Edition 4x4GB Corsair Vengeance 1600-9-9-9-24-2t-1.5v 
Hard DriveOptical DriveOSMonitor
M500 240GB SSD for OS, 2x256GB M4's Raid0 for ... Generic Windows 10 Pro 64 Bit Asus VK266H 
PowerCase
Corsair RM 750x Cooler Master Storm Scout 
  hide details  
Reply
post #49 of 49
Quote:
Originally Posted by robwadeson View Post
i7 will fold twice as fast as i5 at the same clock speed since folding can take advantage of extra 4 threads
Doesn't work like that at all. 15-30% speed increase over the 2500k for thread-happy applications.
2P Workstation
(13 items)
 
  
CPUCPUMotherboardGraphics
Xeon E5 2699 v3 Xeon E5 2699 v3 ASUS Z10PE-D16 EVGA TITAN X 
RAMHard DriveHard DriveCooling
128GB Crucial DDR4 @ 2133MHz Intel 750 Series 1.2TB 4 x 4TB Constellation RAID5 2 x XSPC Raystorms 
CoolingCoolingMonitorMonitor
RS480 MCP655 Dell U3011 Dell U3011 
Case
Silverstone TJ07 
  hide details  
Reply
2P Workstation
(13 items)
 
  
CPUCPUMotherboardGraphics
Xeon E5 2699 v3 Xeon E5 2699 v3 ASUS Z10PE-D16 EVGA TITAN X 
RAMHard DriveHard DriveCooling
128GB Crucial DDR4 @ 2133MHz Intel 750 Series 1.2TB 4 x 4TB Constellation RAID5 2 x XSPC Raystorms 
CoolingCoolingMonitorMonitor
RS480 MCP655 Dell U3011 Dell U3011 
Case
Silverstone TJ07 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Why is the i5's Performance so much better even though stats are the same?