| View previous topic :: View next topic |
| Author |
Message |
Cheer
Guest
|
 Posted: Tue Feb 12, 2019 8:05 pm Post subject: CPU comparison |
 |
|
| I just wanted to ask... Core i7 3930K vs i7 8550u. I know the K would win by a large margin but I wanted to ask: clock-for-clock, core-for-core speed, which one is faster? If you had only 1 core enabled on each one at 3.8GHz, which one would be faster? Only asking because the 8550u is 6 years newers |
|
| Back to top |
|
 |
wren4777
Joined: 13 Dec 2016
Posts: 571
Location: Litija, Slovenia
|
| Posted: Tue Feb 12, 2019 10:27 pm Post subject: Re: CPU comparison |
|
|
| Cheer wrote: | | I just wanted to ask... Core i7 3930K vs i7 8550u. I know the K would win by a large margin but I wanted to ask: clock-for-clock, core-for-core speed, which one is faster? If you had only 1 core enabled on each one at 3.8GHz, which one would be faster? Only asking because the 8550u is 6 years newers |
The 8550U for sure. the 3930K is sandy bridge while the 8550U is kaby lake, and there is a massive IPC improvement between the two.
_________________
My WTB List
My collection
Have a nice day!  |
|
| Back to top |
|
|
Cheer
Guest
|
| Posted: Tue Feb 12, 2019 11:55 pm Post subject: |
|
|
| Ok that helps so far, but one is mobile and the other and desktop. Will that make a difference to the equation or no? |
|
| Back to top |
|
|
wren4777
Joined: 13 Dec 2016
Posts: 571
Location: Litija, Slovenia
|
| Posted: Wed Feb 13, 2019 12:18 am Post subject: |
|
|
| Cheer wrote: | | Ok that helps so far, but one is mobile and the other and desktop. Will that make a difference to the equation or no? |
Sort of, the clock speeds of the 3930K are higher as it is a desktop part and can therefore be cooled better.
_________________
My WTB List
My collection
Have a nice day! |
|
| Back to top |
|
|
Cheer
Guest
|
| Posted: Wed Feb 13, 2019 12:32 am Post subject: |
|
|
| Yes but will the fact that it's a desktop chip as opposed to mobile make a difference to IPC? |
|
| Back to top |
|
|
CPUShack
Joined: 16 Jun 2003
Posts: 34259
Location: State of Jefferson, USA
|
| Posted: Wed Feb 13, 2019 1:13 am Post subject: |
|
|
| Cheer wrote: | | Yes but will the fact that it's a desktop chip as opposed to mobile make a difference to IPC? |
No, the cores are the same between them (for a given architecture) They just adjust things like clock speed, cache, power dissipation etc)
_________________
New for 2025! The CPU Shack has a co-processor!
Visit The CPU Shack of microprocessor history and information. |
|
| Back to top |
|
|
Mixeur
Joined: 06 Jan 2005
Posts: 4038
Location: Sochaux, France
|
|
| Back to top |
|
|
wren4777
Joined: 13 Dec 2016
Posts: 571
Location: Litija, Slovenia
|
| Posted: Wed Feb 13, 2019 2:25 am Post subject: |
|
|
OP asked for a comparison with the same amount of cores at the same clock (aka IPC) - in that regard the new one walks all over the old one.
Also, that's performance in raw terms. In a real life, non content-heavy situation, the new one will again feel better.
_________________
My WTB List
My collection
Have a nice day! |
|
| Back to top |
|
|
crazybubba64
Joined: 03 Jul 2018
Posts: 1371
Location: WI, USA
|
| Posted: Wed Feb 13, 2019 9:18 am Post subject: |
|
|
To be fair, this is comparing a 130W unlocked desktop processor to a low-power laptop chip.
Looking at the passmark page, the single-thread performance isn't *that* far off.
One must also consider the thermal capability of whatever machine the 8550U is in, as thermal throttling can easily hinder a processor's performance.
I wouldn't say there would be much of a difference in everyday use between the two. Although the extra four threads of the 3930k along with double the memory channels might make a big difference in some applications.
_________________
My collection |
|
| Back to top |
|
|
guest
Guest
|
| Posted: Thu Aug 15, 2019 10:00 am Post subject: performance tests |
|
|
hope this adds value to the conversation,
is there a in-house publication which explains the performance bm's given on each cpu profile or comparison charts?
I don't play games very often. I am proned for informatics power CPUs, for installing binaries, compilation, running vm/container, and further network analysis, system tools/reverse as well. |
|
| Back to top |
|
|
misace
Joined: 14 Sep 2012
Posts: 932
Location: Prague, Czechia
|
| Posted: Thu Aug 15, 2019 1:45 pm Post subject: |
|
|
Thereīs a article where they measured IPC (Intsructions per clock) between Sandy Bridge and Kaby Lake. Itīs about 20% for Kaby Lake.
https://www.hardocp.com/article/2017/01/13/kaby_lake_7700k_vs_sandy_bridge_2600k_ipc_review/1
i7-8550U has a configurable TDP, on 25W will be much faster than on 15W.
_________________
Looking For Chips 386 J38605DX, Symphony FPU, Rise Tiger, MP6-II, Winchip 3&4, IDT C6-Early Logo, Nx686, Tesla 8088, L80C86, Bulgarian 8088, Pentium 4 QT69ES. |
|
| Back to top |
|
|
Guest
|
| Posted: Fri Aug 23, 2019 8:49 am Post subject: |
|
|
| misace wrote: | | hardocp review |
great addition, bro.
i shall clarify the performance benchmarks which intrigues me atm are Integer CPU, Floating Point FPU, SIMD and Memory performance level graphs given on CPU profiles, here in cpu-world's detailed reviews.
Relate arguments about specific end proneness among them. |
|
| Back to top |
|
|
|
![[ Hidden ]](https://cdn.cpu-world.com/forum/templates/CPUWorld/images/lang_english/icon_email.gif "[ Hidden ]"){kind=icon}