-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Coffeelake and Donutlake
- Thread starter batboy
- Start date
- Joined
- Sep 19, 2015
- Location
- Corsica
Systems used are on 16GB/3200CL14 memory as well.
And ame goes for all benchmarks : ipc increase, 5% max per clock.
Regarding SP32m, it a whole 10% lower for the 8700K (marginal on SP1M, confirming what you say above).
- Joined
- Sep 19, 2015
- Location
- Corsica
- Joined
- Jul 31, 2004
Do you run an operating system? If so, there will be other processes to use those N available threads. Windows itself runs 40+ processes. Would you rather have context switching on 4 fast and hot cores, or everything running just fine on 16 slow and cool cores that are still more than enough for the tasks assigned to each?
- Joined
- Sep 19, 2015
- Location
- Corsica
That makes a lot of sense!
- Joined
- Mar 7, 2008
For general consumer level tasks, I'd argue 4 not-slow cores are generally plenty for a good experience. It would be likely that 16 slow cores will give a worse general experience and most of those extra cores will sit around mostly idle while fewer cores are maxed out for longer. All this complicated by variable active core turbo... I have a 14 core Xeon system, and it definitely doesn't feel that as great on a desktop as consumer quads. The OS background tasks are generally minimal impact, outside of a Windows Update which cripples anything no matter how fast.
Let's take the ideal scenario. If you had perfect core scaling, the total throughput is proportional to cores * clock (* architecture adjustment * HT/SMT benefit). When we buy more cores, we aren't just buying more cores. We're buying a bigger number of core * clock, otherwise more cores wont be worth it. I've done a lot of testing in my specific compute areas of interest (comparable workloads to Prime95, linpack etc.). Multi-threading helps but the cases it helps are not universal. It can shift the bottleneck to other areas like disk, cache or ram, or simply showing up weaknesses in the software. For a given total potential, fewer faster cores will generally give more real world throughput than an equivalent case of more slower cores, all else being equal. Yes, faster cores tend to be less power efficient. But more cores = more silicon area = more cost too. It is a tough balance and we can only buy what is right for us at a given time.
- Joined
- Jul 31, 2004
That's what "turbo" is for already
But really, x86 needs to somehow implement the "big.little" thing ARM has. I'd love to have 4 fast cores for games and 12 low-power cores to deal with all the background processes. We already have pretty decent power management allowing different cores to run at different speeds, just not enough cores to really take advantage of that, and AFAIK no software signalling to say "this process is not at all urgent. DO NOT throttle the core up to 4 GHz just to get it done quicker, FFS."
- Joined
- Feb 16, 2008
- Location
- East Coast
If intel could deliver a small improvement across gaming/single core/multi core many would happily buy, including myself. This isnt work computers updated upon need, this is home/hobby computers updated like iphones.. shiny new toy to explore and play with....but new toys arent fun with zero improvement or a loss in some areas. If cpus stay on silicon and the gains continue to flatline, im going to need a new hobby. Hope the 8 core comes out in late 2018 on 10nm and with some improvement across board over my broadwell e...instead of another sidegrade.
- Joined
- Mar 7, 2008
That's what "turbo" is for already
For laptops or other mobile devices that could be interesting, although I'm not convinced there is enough need for that many low power cores. The optimisation is for minimum power consumption over time. There are cases where short high power bursts combined with longer periods of deep power save are more efficient than running a lower power burst for longer, or more frequent shifting in and out of power save state. I also wonder how much of a modern PC power consumption when substantially idle is the CPU, as opposed to other parts?
In terms of background processes getting in the way, I think in part that is an OS issue. Windows just was never set up to do multi-tasking well, and things can interfere with others too easily.
At this point, I wouldn't buy a 4c/4t processor new if I gamed...there are already several titles where more than 4 threads improve performance.
Would I upgrade? That is up to the person doing the upgrading... I'm certain you know what questions you need to ask yourself to determine if its 'worth' the upgrade.
Would I upgrade? That is up to the person doing the upgrading... I'm certain you know what questions you need to ask yourself to determine if its 'worth' the upgrade.
- Joined
- Mar 7, 2008
Speaking of 4c4t, I have an OS running on the 8350k now.
https://valid.x86.fr/cxp9qh
Gonna do initial tests first before OC and more.
https://valid.x86.fr/cxp9qh
Gonna do initial tests first before OC and more.
- Joined
- Aug 14, 2014
Intel Will No Longer Provide Per-Core Turbo Frequencies, Making Motherboard Tuning Impossible - https://www.extremetech.com/computi...quencies-making-motherboard-tuning-impossible
- Joined
- Sep 19, 2015
- Location
- Corsica
I think I've never checked the Per Core Turbo frequency, lol!
WGAF.
Still trying to figure out why i should care... as in, what bearing does this really have?
Who cares? Overclocking disables Turbo Boost anyway so it's irrelevant. Further, it's not exactly hard to figure out the max turbo boost. Just open Prime95, hit go and use CPU-Z or Windows Resource Monitor to see what the max clock speed is. Not exactly rocket science. Intel trying to hide it's max turbo frequencies is about as effective as Ford trying to keep secret what brand of tires they ship their cars with...
Point is that they did barely anything. Most performance gains are from faster internal connections and cache. Their marketing is working great and it's not hard to convince most users that new is better. If you were on any Intel event then you know how much BS they can say in one hour
As it was already said, larger cache = slower cache. They have to improve it in other way to make it at the same time faster and larger but it's not so easy. It's also one of the reasons why AMD FX was so slow ... while Intel had 2-4MB L2 then AMD was pushing 8MB. Also difference in popular software like games is visible because of cache. Many people are complaining because of Skylake-X slow L3 cache.
Most changes are in chipsets, pcie lanes, additional features/controllers, lower power usage, more cores etc. Performance on single cores is not much higher than couple of years ago but there is higher turbo clock, faster memory controllers and some other things.
Finally 8th gen of mobile Intel CPUs will have more cores. Since first Core gen almost all i3 and i5 and most i7 were 2 cores. Now i5 are 2/4 or 4+HT so something is moving.
In desktops there are less changes than in other series. Now you can get 16 core Atom CPU which has 32W TDP but has full functionality of higher series. Couple of years ago at this TDP were 2-4 cores. For home/office users it doesn't matter but in business in global scale it's huge difference. In laptops in the same TDP you can get 2x more cores what makes possible to build ultrabooks with more than 2 cores. Even though performance per core isn't much better then overall it's a big difference.
I agree more cores is great, however like mackerel said a lot of programs can't benefit with more cores.
The reviews I looked at there are some games now that don't know what to do with 6 cores 12 threads and that hindered performance a little compared to the i7 7700k. I would go with the 8700k for future proofing for years to come.
Last edited:
- Joined
- Aug 30, 2004
- Location
- Santa Barbara, CA
I've come to learn this (parallel execution) recently with FPGA programming (designing a feedback control system).
![]-[itman](/forums/data/avatars/s/9/9309.jpg?1650325391){kind=avatar}
