Intel Core i5-2500K
Intel Core i5-4670K
|
3DMark06 CPU score benchmark
|
Estimates processor performance based on how fast it runs such
CPU-intensive code as game logic, physics and pathfinding AI. The
test requires DirectX 9.0.
|
| | |
| | |
Apache web server - dynamic pages benchmark
|
Shows an average number of dynamic HTML pages served per second by
local instance of Apache 1.3.41 web server. The test has much greater
error margin than other tests.
|
| | |
| | |
Apache web server - static pages benchmark
|
Shows an average number of very simple static HTML pages served per
second by local Apache 1.3.41 web server. The test has much greater
error margin than other tests.
|
| | |
| | |
Blowfish encryption benchmark
|
Reports encryption speed of pre-defined 32KB text block. The data is
encrypted using Blowfish algorithm.
|
| | |
| | |
CINEBENCH R10 CPU score benchmark
|
Measures performance of multi-core rendering of a photo-realistic 3D
image using CINEMA 4D software engine.
|
| | |
| | |
Intel Core i5-2500K
Intel Core i5-4670K
|
CINEBENCH R10 render time (seconds) benchmark
|
Measures time taken to render a photo-realistic 3D image using CINEMA
4D software engine.
|
| | |
| | |
CrystalMark ALU benchmark
|
Runs Fibonacci, Napierian, Eratosthenes and QuickSort integer
performance tests on all CPU cores, and reports the sum of results.
|
| | |
| | |
CrystalMark FPU benchmark
|
Executes MikoFPU, RandMeanSS, FFT and Mandelbrot FPU-dependent tests
using all CPU cores, and reports the sum of results.
|
| | |
| | |
Euler3D benchmark
|
Solves specific computational fluid dynamics problem, and
reports a CFD cycle frequency in Hertz.
|
| | |
| | |
FLAC encoding - multiple processes benchmark
|
Reports the number of songs, converted from WAV to FLAC
format by all CPU cores using the best possible compression.
|
| | |
| | |
Intel Core i5-2500K
Intel Core i5-4670K
|
Fritz Chess Benchmark benchmark
|
Measures CPU speed by evaluating chess board positions using
Deep Fritz 12 engine.
|
| | |
| | |
H.264 video encoding - multiple processes benchmark
|
Measures the number of frames, converted by all logical processors
from MPEG2 format to H.264 format using single-pass method and high quality settings.
|
| | |
| | |
H.264 video encoding - multithreaded benchmark
|
Measures the number of frames, converted by all logical processors
from MPEG2 format to H.264 format using multi-threading system calls,
dual-pass method and high quality settings.
|
| | |
| | |
LAME MP3 encoding - multiple processes benchmark
|
Reports the number of MP3 songs, encoded by all CPU cores per one minute.
|
| | |
| | |
PCMark05 CPU score benchmark
|
Measures CPU performance by running data compression, decompression,
encryption and decryption tests as single-threaded tasks and in
parallel.
|
| | |
| | |
Intel Core i5-2500K
Intel Core i5-4670K
|
PCMark05 score benchmark
|
Measures overall system performance by running 16 general, CPU,
Memory, Graphics and HDD tests, and uses results to calculate a
weighted score, expressed in PCMarks.
|
| | |
| | |
Sandra Dhrystone (MIPS) benchmark
|
Runs Dhrystone test on all cores, and reports estimated integer
performance in MIPS (Millions of Instructions Per Second).
|
| | |
| | |
Sandra MultiMedia Floating Point (it/s) benchmark
|
| | |
| | |
Sandra MultiMedia Integer (it/s) benchmark
|
| | |
| | |
Sandra Whetstone (MFLOPS) benchmark
|
Runs Whetstone test on all cores, and reports estimated floating-point
performance in MFLOPS (Millions of Floating-Point Operations Per Second).
|
| | |
| | |
Intel Core i5-2500K
Intel Core i5-4670K
|
Sandra Whetstone SSE2 (MFLOPS) benchmark
|
Runs SSE2-enabled Whetstone test on all cores, and reports estimated
floating-point performance in MFLOPS (Millions of Floating-Point
Operations Per Second).
|
| | |
| | |
XVid video encoding - multiple processes benchmark
|
Reports how many frames per second can be transcoded by all CPU cores
from MPEG2 to XVid (MPEG4) format, utilizing single-pass conversion
method.
|
| | |
| | |
wPrime v1.55 (32M) benchmark
|
wPrime benchmark measures time taken by all logical processors to
calculate square roots of numbers from 1 to 33554431 (32M).
|
| | |
| | |
Both OC@4.4GHz easily
Very easy to do and almost 100%
With 4.5GHz and up, chance of success decrease.
negatively surprised
So you are saying that after 2 generations i5-4670K is only
13% to 17% faster than i5-2500K?. Doesn't sound very promosing for the future.
Why bother testing these without overclocking?
Why buy either of these CPUs if you don't plan to overclock? So, why test them without overclocking? Pointless.
I've run my 2500K @ 4.8GHz using a Noctua NH-D14 air cooler for 40 hours a week for the last couple of years with no stability issues.
From what I read, only 30% of 4670K CPUs can even reach 4.6GHz, 20% 4.7GHz, 10% 4.8GHz. So, you could probably only count on 4.5GHz.
A 2500K @ 4.8GHz would compete reasonably well with a 4670K @ 4.5GHz.
The real point of comparing these two CPUs is to help existing 2500K owners figure out if an upgrade is worthwhile. The answer is no.
I agree with Michael from above
When i looked at this review, side-by-side, at first look i said "wow, should've waited for Haswell because it's the last generation in 4 cores family", and started a little to regret buying a sandy.
After reading Michael's post, i totally agree with him. This article is useless. You're just comparing base products which are unlocked, rather than comparing them full boosted (BECAUSE THAT's WHY THEY'RE UNLOCKED). I was reading it just to see if it's worth it to upgrade. But thanks to Michael, i know it doesn't!
Cheers,
L.
I5 25 VS I5 46
Indeed like they say above, no need to upgrade, unless when by OC there is a big difference.
Power Consumption
Maybe the performance is not much better but efficiency became fantastic.
I guess when you overclock a i5-2500k to 4,8 you will have idle power of nearly 100W. Or tell me what do you have when I'm wrong.