Intel claims its 10nm process is a "full generation ahead" of rivals

[HEXUS]

Intel's 10nm transistor density is approximately double that of rivals like Samsung or TSMC.

Read more.

[GinoLatino]

But is not here yet...

[Ozaron]

But is not here yet...

Neither is Vega, mainstream-affordable VR or a genuinely cost-efficient GB/£ storage replacement for spinning rust. All good (and exciting) things in good time.

[Biscuit]

Neither is Vega, mainstream-affordable VR or a genuinely cost-efficient GB/£ storage replacement for spinning rust. All good (and exciting) things in good time.

But in those cases at least we have had tech demonstrations, or existing products to base the future off.
This is just hot air, if you look at the charts, they very cleverly put 2017, then skip 2018 and the next bar is 2019, so its reasonably safe to assume they will have working silicon sometime after 1018 which will take some time to translate into products on shelves. Im guessing consumers wont have anything 10nm in their hands, from intel until after Q2 2019.

"by delivering transistors that are smaller and have lower cost-per-transistor"

Not that the consumer will ever see any of the cost savings...

[Platinum]

Intel has demod chips http://hexus.net/tech/news/cpu/10114...ke-laptop-ces/

[Biscuit]

Intel has demod chips http://hexus.net/tech/news/cpu/10114...ke-laptop-ces/

Ha! Ill just leave then...

[Bagpuss]

Oh, are they now acknowledging AMD as a rival then?

..they must be worried

[bridges009]

Now all they need is a new architecture and a way to gracefully break from their absurd market segmentation model...

[Sunderas]

[QUOTE=Biscuit;23156]

"by delivering transistors that are smaller and have lower cost-per-transistor"

Not that the consumer will ever see any of the cost savings...

Hahahahahaha! Percisely!

[Arbie]

Intel has awesome technology, and the geek in me wants to continue to buy their stuff. 100 million transistors per mm^2!!

However... I've been thinking that without competition, it would actually be bad business for Intel to innovate. Why pay the gigantic R&D costs if you can just coast along, moving all that money to the bottom line? Intel's only PC-world competitor is AMD. If they give up it seems impossible for another credible player to get into this high-stakes game.

Given AMD's valiant effort at recovery with the amazingly competitive Zen line, I am willing to put my money there even if it's not the tech front-runner. Keeping AMD in the game is crucial to advancing the whole industry. If we buy only Intel now, they will soon be the only choice, and they'll offer much less than they would otherwise. We're pretty close to that state of affairs now, as this looks like AMD's last possible effort.

So, think seriously about tomorrow before you choose e.g. Kaby Lake over Ryzen just to get a few extra FPS in some game today.

[devBunny]

But is not here yet...

That does seem to be how it works when you look ahead to see what the future will bring. ;-)

[elites2012]

im going to enjoy this melt down launch!

[Shane74]

"Intel claims its 10nm process is a "full generation ahead" of rivals"

So are it's prices!
Going to have to do better than that Intel, AMD has a winner that puts my 5960X and the price I paid for it to shame.
Next build the cpu is going to be AMD, It's basically half price for the top dog cpu.

[kompukare]

So, think seriously about tomorrow before you choose e.g. Kaby Lake over Ryzen just to get a few extra FPS in some game today.

That's the thing though, humans aren't usually that good at thinking long term even where it is in their (eventual) self-interest.

Competition is the only thing which can keep such big players like Intel (somewhat) honest. While AMD only had Bulldozer design CPUs it was hard to recommend them outside of cheap virtualisation or video encoding workstations, but now that they have a good design - even if Kabylake gets a few more FPS especially in low-resolutiontests - there really is no such excuse. And besides, there are already situations where 4C/8T even with Skylake's IPC advantage and 1GHz more frequency is limited and that is likely to get worse. Plus AMD still offer better value with more cores even at Ryzen 5 1600's lower-than-i5 price and even when games don't advantage of them most applications or background processes will.

[utopic]

Intel has awesome technology, and the geek in me wants to continue to buy their stuff. 100 million transistors per mm^2!!

However... I've been thinking that without competition, it would actually be bad business for Intel to innovate. Why pay the gigantic R&D costs if you can just coast along, moving all that money to the bottom line? Intel's only PC-world competitor is AMD. If they give up it seems impossible for another credible player to get into this high-stakes game.

Given AMD's valiant effort at recovery with the amazingly competitive Zen line, I am willing to put my money there even if it's not the tech front-runner. Keeping AMD in the game is crucial to advancing the whole industry. If we buy only Intel now, they will soon be the only choice, and they'll offer much less than they would otherwise. We're pretty close to that state of affairs now, as this looks like AMD's last possible effort.

So, think seriously about tomorrow before you choose e.g. Kaby Lake over Ryzen just to get a few extra FPS in some game today.

Hear, hear!

[CAT-THE-FIFTH]

That's the thing though, humans aren't usually that good at thinking long term even where it is in their (eventual) self-interest.

Competition is the only thing which can keep such big players like Intel (somewhat) honest. While AMD only had Bulldozer design CPUs it was hard to recommend them outside of cheap virtualisation or video encoding workstations, but now that they have a good design - even if Kabylake gets a few more FPS especially in low-resolutiontests - there really is no such excuse. And besides, there are already situations where 4C/8T even with Skylake's IPC advantage and 1GHz more frequency is limited and that is likely to get worse. Plus AMD still offer better value with more cores even at Ryzen 5 1600's lower-than-i5 price and even when games don't advantage of them most applications or background processes will.

Except in the games I am actually running now,none of the Ryzen CPUs will actually offer me a noticeable upgrade and that will be at 1080p and qHD because the same drops happen in the same places at any resolution,even with slower cards. Its no point worrying about the future if I am not getting an upgrade now - I might as well wait another year. All the games I have problems in,Ryzen won't solve,but KL/SKL will and by extension Coffee Lake,and I think some of you need to understand not everybody is running the latest 8 core aware SP game,or something like BF1. I know ZERO BF1 players,but I know far more people running MMO type games which are based on oldish engines and don't thread well. I know more people running games like PS2,ESO,ES,FO,etc than say AOTS.

The game we play at LANs,is not AOTS but things like Sins of a Solar Empire,etc which is not well threaded. Games like SC2 and WoW are still very popular and have millions of people playing them and probably ONLY those games.

Just because games are old and not the latest gaming fad,it does not means people will abandon them.

If someone asked me to spec them a top end rig for WoW it would be Intel/Nvidia not AMD/AMD. Yes AMD will run it OK,but the Intel/Nvidia combo will still generally be better at least last time I checked. I am not going to lie to them and say in 12 months time an AMD/AMD based rig will destroy Intel/Nvidia in the game. Maybe it will,maybe it won't and its somebodies money I will be wasting.


The problem is that enthusiasts on tech forums will play more modern games but TBH I bought Doom,Deus Ex:MD,etc and I lost interest in them and anyway they ran fine on my 5 year old CPU,but plenty of people who game will be running those poorly optimised titles which hog very few threads which don't run fine. These are the very kind of games, are CPU limited by the performance of one or two cores.

Most of these titles are tied exclusively to the PC and are they poorly optimised with penny pinching devs behind some of the larger ones. Yep,but it is what it is.

So unless the purported patches in time for the Ryzen 5 launch suddenly improve things,its not going to change much.

I also looked at some of the image editing benchmarks with Adobe sw,and again Intel pushed ahead since they processes are not all well threaded:

https://www.pugetsystems.com/pic_dis...800&height=800

I use LR,so again looking at the performance of Ryzen unless again AMD gets a big performance patch,I estimate performance is close to what I have already.

Maybe as I get older,I tend to be far more critical when I upgrade and I tend to get less excited by launches now - maybe if I was a decade younger I might have ditched this i7 when the R5 1600X is released,but unless there is some sudden massive CPU patch which changes performance in even older titles,its not really worth me spending £400.

Its a side-grade and by the time more cores will affect things I run in any meaningful way I might as well have waited to upgraded then and not now.

So I am waiting for Ryzen MK2.

But the problem is by then Intel will also have 6C/12T CPUs - supporting AMD is one thing,but again a CPU is a tool. I will quite happily give up performance to a certain degree and get AMD,but if is too drastic for what I plan to run,then again meh.

There are other situations outside the CPU which also have an effect- especially when even the times I wanted to buy an AMD CPU,Intel had just better quality mini-ITX motherboards and a wider range. Its why I never bought a Llano CPU with my own money - it was better for my purposes than the SB CPU I ended up getting,but I waited yonks for a mini-ITX motherboard and got fed up and the first one released had a tendency to burn out the VRMs.

Its the same reason why I had my Q6600 for so long. I fancied a SFF rig with a Phenom II X6 but Shuttle kind of lost interest in the AMD side of their business and by the time AM3 mini-ITX motherboards appeared I also lost interest.