(10-16-2017, 06:12 AM)Subzero75 Wrote: [ -> ]In my OP 8700K is overkill for like 95% pf the population. I'm not spending a $100 more for the 8700K when the 8600K is will perform in emulators just as good when OC'ed. I'm getting the 8600k and will OC it to 5.0GHz. Also from the latest eurogamer review they say that the 8600k and 8700k both OC'ed to 5.0GHz the 8600k produces much less heat and used less power, also said they feel that the 8600k still had some OC'ing capacity in the tank. I'm not getting a 8700k just for bragging rights, I'm not saying either that people who get them are doing it for bragging either.
http://www.eurogamer.net/articles/digitalfoundry-2017-intel-coffee-lake-s-core-i5-8400-i5-8600k-review
I'm just waiting for them to be back in stock 
The problem with relying on overclocking is it's a risk. Intel only *guarantee* the base frequency. You may be unlucky, lose the "silicon lottery", and buy a lemon.
I kinda dislike reviewers talking about OC results without a decent spread of samples to work with - you don't know if they happened to be lucky (or "golden samples" sent to reviewers)
The eight generation CPU's certainly look interesting. Looking at the i5-8400 here I do wonder how it behaves in Dolphin. Now, I would not upgrade my i5-4430 to an i5-8400 purely because of Dolphin since my i5-4430 is more than capable of running most of what Dolphin can throw at me. Of course there is more than simply Dolphin for a Windows system. Obvious the new i5-8400 is a big improvement for native Windows applications due the two additional cores, higher cache and higher turbo frequency. However the base frequency is a bit lower than the i5-4430. Since emulation does not rely that much on a massive amount of cores, the question that begs is if the i5-8400 still offers a major improvement for emulation. I suppose it couldn't be worse than the i5-4430, but would the i5-8400 be capable of Dolphin throwing even more demands to it? Would demanding emulators such as Citra or CEMU see big improvements from it?
I suppose you could say I am mostly interested in an emulation benchmark between the i5-4430 and i5-8400, just for knowing how much of an improvement it will be.
While you are right that in emulation generally single core performance > core count, you are also not taking into account that single core performance is more then just frequency. Things like architecture improvements can improve efficiency and allow more to be done per clock (IPC), btw this is one reason why older AMD CPU with really high frequencies didn't really have great single core performance.
Intel has generally increased IPC each generation (with the exception of Kaby Lake) and there is more L3 cache on Coffee Lake so even if the frequency is a bit lower performance could be higher in single threaded applications.
(10-25-2017, 01:11 PM)TKSilver Wrote: [ -> ]While you are right that in emulation generally single core performance > core count, you are also not taking into account that single core performance is more then just frequency. Things like architecture improvements can improve efficiency and allow more to be done per clock (IPC), btw this is one reason why older AMD CPU with really high frequencies didn't really have great single core performance.
Intel has generally increased IPC each generation (with the exception of Kaby Lake) and there is more L3 cache on Coffee Lake so even if the frequency is a bit lower performance could be higher in single threaded applications.
I am sure that the 8th generation is an improvement in every way, even for emulation and I am aware that intel increases the IPC with almost each generation, but it still begs the question how much of an improvement the 8th generation is. Would it be slightly better for emulation? A lot? Or somewhere inbetween? That's why a benchmark could allow us to compare results.
Coffee lake uses the same uarch as kaby lake and therefore get the same IPC. The larger L3 cache is just due to having more cores. The only difference is the new manufacturing process which has made their transistors smaller and power efficient. This allowed them to move their chips from 4 cores to 6 and increase the clock rates by 200 MHz. It's pretty amazing that the i7 8700 can hit 4.3GHz under 6 core load and 4.6GHz under a single core load while maintaining 65w TDP. For applications using 4 cores or less the performance should be very slightly better on coffee lake (+5%). But for heavily multithreaded apps it will be significantly better (+60%). The benchmark results are pretty much what we expected for dolphin based on preliminary specs, the same as kaby lake. If you live stream dolphin with cpu encoding though you will notice a big difference due to the extra cores (3 for dolphin, 3 for video encoding).
I'm pretty sure Intel released a statement saying the turbo clocks might well push a chip beyond its TDP and that providing cooling for that TDP is only a guarantee of base clocks. The coffee lake six cores are hot chips as I've heard it.
Obviously more cores means more power and more heat, but I think you are getting it confused with the Kaby Lake Extreme, like the Core i9s. Those go INSANELY hot and power hungry! i7-8700ks on the other hand are very manageable, and don't suffer from thermal limitations with the default tim (unlike say, the i9-7900x).
It's both. The i9s are practically soldering irons, but the hexa-core coffee lake chips still draw more than their rated TDP when turboing.
Well, at true stock it pulls precisely the TDP or even under it. As shown on
anandtech and
Gamers Nexus. But with
Multi-Core Enhancement enabled it can go over, but that's not a stock configuration, and the motherboard is doing things to give the CPU more power. There was a lot of confusion about this since lots of sites weren't aware of this at first and Asus and a few other motherboard brands ship with Multi-Core Enhancement enabled by default, because, outside of power use, it's basically a free performance boost! This isn't new, btw, but with more cores Coffee Lake really put the spotlight on this behavior.
(11-26-2017, 06:30 AM)NaturalViolence Wrote: [ -> ]Coffee lake uses the same uarch as kaby lake and therefore get the same IPC. The larger L3 cache is just due to having more cores. The only difference is the new manufacturing process which has made their transistors smaller and power efficient. This allowed them to move their chips from 4 cores to 6 and increase the clock rates by 200 MHz. It's pretty amazing that the i7 8700 can hit 4.3GHz under 6 core load and 4.6GHz under a single core load while maintaining 65w TDP. For applications using 4 cores or less the performance should be very slightly better on coffee lake (+5%). But for heavily multithreaded apps it will be significantly better (+60%). The benchmark results are pretty much what we expected for dolphin based on preliminary specs, the same as kaby lake. If you live stream dolphin with cpu encoding though you will notice a big difference due to the extra cores (3 for dolphin, 3 for video encoding).
I see you are out of the loop on PC hardware. The jump 14nm+ to 14nm++ is simply for efficiency, the size didn't change. Also, you said it was pretty amazing that it can remain 65W for 6 cores 12 threads load. Have you every heard of Ryzen? The R7 1700 does 8 cores 16 threads 65W full load. If you have been following hardware, you'd know that Coffee Lake was paper launched very early, and released half a year before intel wanted to release it. This was because Ryzen was too powerful and taking all of the market share for 2017 CPU's. Coffee Lake and Core i9 was a
DIRECT response to Ryzen and Threadripper.
I appreciate you trying your best to explain everything, but please don't be an intel shill, do a little research next time. AMD is what made this year great for CPU's. I know you will say "but AMD sucked for the past 10 years" and that's because was being monopolistic and didn't allow AMD's better Opteron CPU's from being bought by Dell and HP. You can watch the AdoredTV video on YouTube called "A history of a Monopoly" with intel in the thumbnail, you'll see what I mean. AMD was losing so much money, they had to sell all of their fabs, and even their headquarters. They used their GPU profits to fund both CPU and GPU divisions, and that money was less than a tenth of intel's. All analysts say that it was amazing how AMD survived, but they did, because AMD is amazing.