Intel Core i7-9700K 9th Gen CPU Review: Eight Cores And No Hyper-Threading

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

MasterMadBones

Distinguished
Dec 26, 2012
333
32
18,890
47
Good luck getting Ryzen to 5.0GHz or higher on air or water. To reduce the speed to match Ryzen is not a fare comparison. It is like taking a 400 horse power engine [Intel] in a car and restrict the fuel flow so only 200 horse power is available so a 1970 Chevy Biscane with a small block V8 will have a chance against a Corvette.
It's more similar to taking two V8 engines with the same displacement and running them at the same rpm. The performance then becomes dependent on torque (or IPC). For enthusiasts like us, determining IPC is purely a matter of personal interest, because like in internal combustion engines, performance is a function of a number of factors.

From an engineer's point of view, this methodology is extemely valuable when identifying areas of improvent for a product. In the case of Zen+, AMD can now see that most of the performance deficit to Intel is due to clock speed, so a good way to close the gap could be to use a better process node for the next generation.
 
It's more similar to taking two V8 engines with the same displacement and running them at the same rpm. The performance then becomes dependent on torque (or IPC). For enthusiasts like us, determining IPC is purely a matter of personal interest, because like in internal combustion engines, performance is a function of a number of factors.
Yeah and that's the problem right there,just as with the engine measurements you will get a fake number that is only real under those very specific conditions and is irrelevant for daily usage.
Ryzen's best case is cinebench where the CPU can run completely unhindered and it is about 14% ahead in IPC but also about 25% behind in clocks.
The worst case scenario is games where ryzen is about 10% behind while the 9900k might still be hindered by the GPU even at 4Ghz.We can only be sure if someone tests this by lowering clocks even more until FPS starts to drop.

View: https://www.youtube.com/watch?v=RmxkpTtwx1k




In the case of Zen+, AMD can now see that most of the performance deficit to Intel is due to clock speed, so a good way to close the gap could be to use a better process node for the next generation.
It's about a 25% difference in all core max clocks a new node will only mean even lower clocks for AMD,we already see them struggling to stay at 4Ghz.
 

MasterMadBones

Distinguished
Dec 26, 2012
333
32
18,890
47
I never mentioned Zen 2, remember this article was written at a time when Zen 2 didn't exist yet. On top of that, some of your maths doesn't work with the charts you put up.

IPC is not fake, and it's very interesting to some of us as I mentioned. The problems you talk about with measuring IPC are addressed by using a large standardized suite of benchmarks such as SPECViewPerf. These tests are well-documented, which allows us to find the weak points in an architecture.

When talking about process nodes, I've been careful to use the word "better" instead of "newer" or "smaller". For example, Intel 14nm is newer than 22nm, but not better from a performance standpoint. However, 14+ and beyond are better. In Ryzen's case, the transition from 12nm to 7nm has not had the greatest effect on (all-core) clocks, which means that 7nm, although it's newer, is arguably not worse than 12nm.
 

ASK THE COMMUNITY

TRENDING THREADS