As far as I know this is not an official price cut from Intel, it's only a store cut and it's Microcenter...which they tend to do so I have the impression that this news is somewhat sensational... I mean of course an official price cut will eventually happen but I don't think this is it. Beside, those price cuts aren't very impressive, Ryzen is still better value.
AMD Ryzen has some strange things on L2 L3 cache level. AMD benching better result than everybody else. Hilbert may interested to check these things. (use Aida64 Zip and similar programs). Even Aida64 author has confirmed these things. I got AMD so don't attack me.
I would like to confirm that several french retailer are also cutting prices on Intel, all i7 and i5 are at least -14% cut on the 2 major retailer
While I give Intel an A for effort by reducing the prices of their CPUs but if they want to be competitive they need to release 6 and 8 core CPUs to the mainstream market at affordable prices.
Good point. Geek sites like this one lose track of what users out there actually use. Same goes with the 4K adoption, etc.
That doesn't matter at all. What matters is that in order to make a console game, your engine needs to be able to scale well with a minimum of 7 cores. All the rest is just PC optimization, which is mostly taking advantage of the fact that PC CPUs are generally much faster per core than the console ones, and compensating this way. But since fast serial performance is never really equal to actual parallelism, you get phenomena like people with Skylake i5s getting stuttering in Battlefield 1. There is no game written for less than eight threads, I assure you. Even the engine made by Obsidian for Pillars of Eternity, and used in the new Torment game, will have to scale properly so that it can be used in a console.
Using Steam hardware survey doesn't provide accurate results. Not every gamer is using Steam. The only reason I have a Steam account, is because it was the only way to get TL2 on Linux. Most of my games are through Origin....
Zen's L3 is a dump cache - it's only used when stuff gets evicted from L2. With Intel the L3 cache always and only includes all the L2/L1 caches. Both methods have different advantages with different workloads - but either way it will lead to differences in benchmark results.
One thing is mostly sure...is that 4 cores 8 threads Ryzen will probably clock at 4.6ghz+ on air, competing directly with 7700k in single thread performance, at lower price. Also even tho I plan to buy a 1700x, I'll probably tweak it to turbo at 4.5ghz+ at 4 cores. Of course if you don't need 8 cores ever, a Ryzen 7 CPU might need be good value, a Ryzen 3 with SMT (4C/8T) will be enough. TBH if you plan to buy a 7700k, I would at least wait for Ryzen 3.