Intel Core i5-7640X 4.0 GHz

[cadaveca]

If anything I actually see this chip as a negative to Intel in the big picture, since it degrades their HEDT Intel brand. Most here will do a deeper dive, but consumers generally should be able to buy a HEDT chip and assume the bells and whistles that are absent in this chip are there. Or call this one a new name (something other than HEDT or with X in the name), that makes it clear that it's a non-HEDT cpu that just happens to work on the HEDT platform for those limited use-cases.

If HEDT is used by business, it makes more sense to have a single platform with many core choice dependent on user, as this is easier to manage by IT people (which is why many business have all their PC's from one single OEM). It's actually quite the feat to have a CPU that does not have internal voltage regulator and one that does, working in the same socket.

For the enthusiast/gaming market, sure it doesn't make a lot of sense when 7600K is available, unless you want to buy a platform with some upgrade life, which 7600K does not have, which is also a selling point for business. They can buy one board now, upgrade CPU later to more cores, easily, rather than having to buy two boards. There is saving in that decision that makes a lot of sense to business, given how consumer technology is progressing with VR and such.

 

[agello24]

as long as my ryzen 5 1600 beats this and some i7 quads in multicore multitasking, im fine. their is not reason you should choose a quad core proc over a higher core proc for multitasking. WE ALL do it. YET the intel FANS feel their quad core is superior.

 

[_Flare]

great test w1zzard,

one more time visualized how efficient the 7700K is in Gaming. 249W for the whore rig, incl. the 166W of the GTX1080, thats 83W for the rest incl. CPU RAM etc.
im curious how coffee-lake will better that ... not "if" ... "how"

 

[jboydgolfer]

i feel like this ISNT a HEDT CPU. based on the following reasoning.

HEDT cpu's (In my mind) support Quad channel memory ~~~ this chip doesnt
HEDT cpu's tend to have Hyper threading ~~~~~ this chip Doesnt
HEDT cpu's tend to be 6+ cores ~~~~~ this one isnt
then theres the other issues that most everyday users , and even enthusiasts wont run into, like PCI-e lane quantity, etc... my conclusion, Intel in an attempt to redefine HEDT, or pull a fast one has advertised this chip as a "budget" HEDT CPU, which IMO, it isnt.

I love intel, but not the idea that this 4690k with a new black box is supposed to be a HEDT-CPU. sorry Intel, this one was a flop, a good chip likely, but not a HEDT chip, you'd have better luck pooping in a bad and telling me it was Burger king takeout.

 

[cadaveca]

I love intel, but not the idea that this 4690k with a new black box is supposed to be a HEDT-CPU. sorry Intel, this one was a flop, a good chip likely, but not a HEDT chip, you'd have better luck pooping in a bad and telling me it was Burger king takeout.

I personally do not think it was ever meant to truly be a HEDT CPU, but merely a CPU that works on a HEDT platform for non-HEDT users, for, like I have posted before, IT managers that have hundreds of boxes to manage. Not all users need HEDT, but some do, and X299 is the platform that can meet the needs of all users, HEDT, or not...

Which makes sense to me, but wouldn't make sense to an enthusiast, other than having an option to get a board that can OC it to the limit and back... ain't so easy to get 4600 MHZ memory on Z270 compared to X299....

 

[Warrgarbl]

Intel does seem quite desperate to throw anything against Ryzen, huh?

 

[Aquinus]

Intel does seem quite desperate to throw anything against Ryzen, huh?

That's not what their stock is telling me. I suspect Intel has been making bank when it comes to the profit margins on these CPUs because only until now AMD really didn't have anything good to counter with. All this is telling me is that Intel has a way to go before they really would start to panic. Intel is a behemoth and we should remember that.

 

[bug]

That's not what their stock is telling me. I suspect Intel has been making bank when it comes to the profit margins on these CPUs because only until now AMD really didn't have anything good to counter with. All this is telling me is that Intel has a way to go before they really would start to panic. Intel is a behemoth and we should remember that.

I said it from day 1 that during 2017, Intel's answer to Ryzen will be mostly smoke and mirrors, in preparation for the real answer (because you can't build a new chip within a few months time frame). I wouldn't be surprised if the apparent mess of lineup Intel has today is merely them testing waters about how to better pile up cores.
But all this matters little, what matters to customers is what we can buy. Coffee Lake is not that exciting (I'll stick to my 6600k tyvm), but maybe, just maybe Ryzen pushed Intel into coming up with something more radical after that. At the same time, as you have said, Intel is a juggernaut and juggernauts are anything but nimble...

 

[Aquinus]

I said it from day 1 that during 2017, Intel's answer to Ryzen will be mostly smoke and mirrors, in preparation for the real answer (because you can't build a new chip within a few months time frame). I wouldn't be surprised if the apparent mess of lineup Intel has today is merely them testing waters about how to better pile up cores.
But all this matters little, what matters to customers is what we can buy. Coffee Lake is not that exciting (I'll stick to my 6600k tyvm), but maybe, just maybe Ryzen pushed Intel into coming up with something more radical after that. At the same time, as you have said, Intel is a juggernaut and juggernauts are anything but nimble...

Edit: Preface; this is 100% speculation.

Exactly and Intel knows this. The real money is in the server and business market because these entities have enormous financial resources compared to the consumer market and when AMD comes out and says, "Hey look, we have a scaleable and cost effective process for producing CPUs with more cores than what's currently available within reasonable power figures and with decent performance", the market starts salivating. Intel is reacting only because it perceives AMD to be a threat. That new i5 on the HEDT platform is Intel's way of trying to make a platform that scales from the very small to the very big, which is what AMD is doing by "gluing" clusters of cores together. As you said, Intel can't react instantly, so what we're seeing is a shift towards the same line of reasoning with something they already have and they've known this and have been working on it for a while, there is older material that describes this but, I wouldn't be surprised if within 1 years time if Intel starts adopting the same strategy of using multiple smaller dies to house clusters of cores. It's an excellent strategy to maintain decent clocks and good yields and Intel would be stupid not to capitalize on the idea but, it's going to take them time to develop it. AMD has Intel on the defensive and while AMD has no real means to substantially harm Intel, it is enough to be considered a threat. Intel has the most to lose and the real question is "how much?"

 

[StrayKAT]

Useless CPU is useless.

None of these are all that appealing to me... I think I'll be happy with Kaby Lake. I'll just jump to i9 when I go to x299.

 

[BiggieShady]

What do you mean pointless ... finally i5 owners can call their PCs workstations

 

[R0H1T]

Edit: Preface; this is 100% speculation.

Exactly and Intel knows this. The real money is in the server and business market because these entities have enormous financial resources compared to the consumer market and when AMD comes out and says, "Hey look, we have a scaleable and cost effective process for producing CPUs with more cores than what's currently available within reasonable power figures and with decent performance", the market starts salivating. Intel is reacting only because it perceives AMD to be a threat. That new i5 on the HEDT platform is Intel's way of trying to make a platform that scales from the very small to the very big, which is what AMD is doing by "gluing" clusters of cores together. As you said, Intel can't react instantly, so what we're seeing is a shift towards the same line of reasoning with something they already have and they've known this and have been working on it for a while, there is older material that describes this but, I wouldn't be surprised if within 1 years time if Intel starts adopting the same strategy of using multiple smaller dies to house clusters of cores. It's an excellent strategy to maintain decent clocks and good yields and Intel would be stupid not to capitalize on the idea but, it's going to take them time to develop it. AMD has Intel on the defensive and while AMD has no real means to substantially harm Intel, it is enough to be considered a threat. Intel has the most to lose and the real question is "how much?"

My guess is too much, if AMD deliver or exceed their targets wrt Zen+ & Zen2 then you can expect a substantial loss of potential revenues for Intel. Guesstimating that it'd be in the low double digits i.e. in billions of dollars, if Intel continues to fumble on 10nm or smaller nodes then that figure could double or quadruple very quickly over the next 5 years or so.

 

[StrayKAT]

Intel does seem quite desperate to throw anything against Ryzen, huh?

I find it impressive though.. how quickly they can release all of these new product lines. Same way I'm impressed that Nvidia is going for the "almost 1080" segment with 1070i (although some here say it's for mining instead. Maybe it is).

 

[EarthDog]

Intel does seem quite desperate to throw anything against Ryzen, huh?

I find it impressive though.. how quickly they can release all of these new product lines. Same way I'm impressed that Nvidia is going for the "almost 1080" segment with 1070i (although some here say it's for mining instead. Maybe it is).

The only thing which is marginally impressive is moving up the timeline on existing products. This isnt an answer, so much as it is what they had to throw at it. It takes years to bring new cpus to market.

 

[StrayKAT]

The only thing which is marginally impressive is moving up the timeline on existing products. This isnt an answer, so much as it is what they had to throw at it. It takes years to bring new cpus to market.

That's still impressive to me.. at least on a marketing level.

That said, I'm not anywhere near the target market. Maybe if I were to build a machine from scratch, I'd do the "buy i5/i7 now, transition to i9 later" thing, but doesn't seem worth it as a recent i7 buyer.

 

[bug]

The only thing which is marginally impressive is moving up the timeline on existing products. This isnt an answer, so much as it is what they had to throw at it. It takes years to bring new cpus to market.

It only takes years if starting from scratch. I'm pretty sure Intel didn't sit on their asses all this time
Just remember how they handled the trainwreck that was Netburst: they went back to Pentium III/M and turned that into Core (plus they bribed their way out).

Though now that I'm looking at dates, it was about two years between Athlon64 and Intel's Core CPUs. And even longer between AthlonXP and Core. Oh well...

 

[Bucket]

I have my 7640x OC per core. #4 is 5.2, #3 5.3, #2 5.4, #1 5.6. Windows reports this as a 5.2 GHz OC, but the clock speed won't exceed 5.1. Regardless, with this OC, I am satisfied that this cpu allowed me to 1. sell my i9 for 3/4 what I paid nearly 2 years ago (moving to x570/3900x soon) and 2. still be productive and even game AAA and record (barely) at the same time. Temps are fine. The rig is stable. Good CPU. My Benchmarks are significantly higher than most everyone reports (this is usually the case, I am an OC enthusiast) and I am doing things that I feel nobody else is with a 4/4 CPU. It definitely spanks the custom build with 7600k in my daughter's rig handily. With the 1200mhz OC, all is good! If you take that OC off BRRRRRRR. The only thing I don't like is that I am losing quad channel support for the 4000mhz corsair kit I purchased with the i9 (7640 lacks required lanes for quad channel support) no big deal, not getting it back with x570, not paying 800+ for another i9, not waiting a year for the zen2 TR HEDT platform. Dual channel it is, at least I can bump back up to 4000mhz.

Useless CPU is useless.

Allowed me to sell my i9 and still use my PC until I jump to x570/zen2 so I am glad this 200$ CPU was available! My OC was much higher than reported here (5.2 aorus x299 gaming 7) my scores thrash a 7700k's in most areas. Without a big OC, these are pretty much useless, with the 1200mhz OC, it runs like a brand new Audi.

as long as my ryzen 5 1600 beats this and some i7 quads in multicore multitasking, im fine. their is not reason you should choose a quad core proc over a higher core proc for multitasking. WE ALL do it. YET the intel FANS feel their quad core is superior.

I'm making the jump to AMD for zen2. As an enthusiast who has a build with a 1600x, as well as a 2600 workstation, and this intel in my home rig, I would take a 2600 or 2600x over this chip any day. I would not take the 1600x over this cpu. The scores and memory support weren't where they should have been for the gen 1 Ryzens.

i feel like this ISNT a HEDT CPU. based on the following reasoning.

HEDT cpu's (In my mind) support Quad channel memory ~~~ this chip doesnt
HEDT cpu's tend to have Hyper threading ~~~~~ this chip Doesnt
HEDT cpu's tend to be 6+ cores ~~~~~ this one isnt
then theres the other issues that most everyday users , and even enthusiasts wont run into, like PCI-e lane quantity, etc... my conclusion, Intel in an attempt to redefine HEDT, or pull a fast one has advertised this chip as a "budget" HEDT CPU, which IMO, it isnt.

I love intel, but not the idea that this 4690k with a new black box is supposed to be a HEDT-CPU. sorry Intel, this one was a flop, a good chip likely, but not a HEDT chip, you'd have better luck pooping in a bad and telling me it was Burger king takeout.

Agreed. Burned by lack of quad channel support. oh well.