Friday, February 10th 2017

8th Gen Core "Cannon Lake" Over 15% Faster Than Kaby Lake: Intel

At an investor meeting in February, Intel touched upon its performance guidance for its 8th generation Core processor family due for later this year. Based on the 14 nm "Cannon Lake" silicon, these processors are expected to have a bigger performance gain over the preceding 7th gen Core "Kaby Lake" micro-architecture, than Kaby Lake had over its predecessor, the 6th gen Core "Skylake."

In a slide titled "advancing Moore's Law on 14 nm," Intel illustrated how Kaby Lake processors are on average 15 percent faster than Skylake parts, in SYSmark. While Kaby Lake has negligible IPC gains over Skylake, the newer chips are clocked significantly higher, making up Intel's performance targets. Unless Cannon Lake is a significantly newer micro-architecture than Kaby Lake, we could expect them to come with even higher clock speeds. Will the Core i7-8700K be a 5 GHz chip?
Source: VideoCardz
Add your own comment

97 Comments on 8th Gen Core "Cannon Lake" Over 15% Faster Than Kaby Lake: Intel

#76
Nosada
dat_boi
Of course it wouldn't make sense to you, you've been a hw enthusiast for thiiiiis long. Ever wondered why others moved on?

Some red pills are really hard to swallow, mainly if they are about things you really like.
Oh this I GOTTA hear!

*eats popcorn*
Posted on Reply
#77
Aquinus
Resident Wat-man
Nosada
Oh this I GOTTA hear!

*eats popcorn*
He says that like people here don't live lives outside of being hardware enthusiasts. :laugh:
Posted on Reply
#78
JunkBear
deu
So NOW their can pull more than 1-4% performance leap!? :D I am a long time intel owner but: "Intel; milking the customer since ; forever...."
Like Apple Do.
Posted on Reply
#79
ratirt
Prima.Vera
If you think AMD's Ryzen will be faster than your (my also) 3770K CPU, you are a very optimistic person. ;)
Optimistic? I think that's already a fact Ryzen's IPC is way better than my I7 3770K's. Since Ryzen matches 6900 and that one is way faster than my 3770k. No doubt about that.

Maybe take a look here.
http://wccftech.com/amd-ryzen-7-1700x-389-8-core-cpu-benchmarks-leaked/

That's what you can get for 400$
Posted on Reply
#80
efikkan
ratirt
Optimistic? I think that's already a fact Ryzen's IPC is way better than my I7 3770K's. Since Ryzen matches 6900 and that one is way faster than my 3770k. No doubt about that.

Maybe take a look here.
http://wccftech.com/amd-ryzen-7-1700x-389-8-core-cpu-benchmarks-leaked/

That's what you can get for 400$
We don't know yet if Ryzen matches i7-6900K overall, time will tell.
But how are you able to determine Ryzen's IPC when we don't even know the clock? These models will feature "XFR" which is an extra boost feature.
Posted on Reply
#81
ratirt
efikkan
We don't know yet if Ryzen matches i7-6900K overall, time will tell.
But how are you able to determine Ryzen's IPC when we don't even know the clock? These models will feature "XFR" which is an extra boost feature.
I'm guessing you didn't read the article. There is confirmation about the frequencies and turbo boost of each model. The benchmark performed for the 1700X was without turbo boost and the frequency was set to 3.4Ghz. Knowing that the performance for this particular CPU will be greater when official release takes place since 1700X will be clocked at 3.4 and boost to 3.8 with XFR it will go 3.8Ghz++. Of course time will tell but you have some information presented here and so far they are very promising especially when 1700X surpasses top Intel products in 5 of 8 benchmarks which gives a relative performance capability. I'm looking forward for the new processors of AMD's.
Posted on Reply
#82
deu
JunkBear
Like Apple Do.
Apple is WAAY worse! :D But the difference is that they are selling a bundle of components with their own stuff to hide away their margin to the users! :D Last time i heard it was about 100-150"% on iphones and about 60%+ on the computer part.

You cant blame Intel for doing it; with NO compatition for years the people setting the price is them so they only have to steir free of monopoly charges :) When it comes to apple people HAVE a choice; its just that alot of people are easily tricked ;)
Posted on Reply
#83
efikkan
ratirt
I'm guessing you didn't read the article. There is confirmation about the frequencies and turbo boost of each model. The benchmark performed for the 1700X was without turbo boost and the frequency was set to 3.4Ghz. Knowing that the performance for this particular CPU will be greater when official release takes place since 1700X will be clocked at 3.4 and boost to 3.8 with XFR it will go 3.8Ghz++. Of course time will tell but you have some information presented here and so far they are very promising especially when 1700X surpasses top Intel products in 5 of 8 benchmarks which gives a relative performance capability. I'm looking forward for the new processors of AMD's.
I'm guessing you didn't either:
"It’s not clear whether the engineering sample was running at 3.4GHz without any turbo functionality during testing or if the application simply failed to read the turbo clock."
BTW, the OS needs new drivers to read turbo frequencies correctly.

We'll need a wide range of benchmarks to determine the IPC, and of course a locked clock frequency. (not that I'm saying anyone should use it that way)
Posted on Reply
#84
ratirt
Well maybe but it's not running 2.0Ghz that's for sure. Besides it was stated in the article that the sample was running 3.4 which was specified on the sample itself. If the turbo wasn't that's ok. It means that it was running stable 3.4 without changes. That is still a great way to determine at least at the beginning the IPC power especially when this CPU holds its own.
We'll need wide range but this is still a benchmark. Whoever will procure benchmarks I'm sure that person will use same suit for benchmarking.
Posted on Reply
#85
efikkan
ratirt
Well maybe but it's not running 2.0Ghz that's for sure. Besides it was stated in the article that the sample was running 3.4 which was specified on the sample itself. If the turbo wasn't that's ok. It means that it was running stable 3.4 without changes. That is still a great way to determine at least at the beginning the IPC power especially when this CPU holds its own.
We'll need wide range but this is still a benchmark. Whoever will procure benchmarks I'm sure that person will use same suit for benchmarking.
The CPUs were specified with turbo frequencies, and you can be pretty sure that turbo is working this close to the release date, especially since AMD wants it to look good.
Posted on Reply
#86
ratirt
efikkan
The CPUs were specified with turbo frequencies, and you can be pretty sure that turbo is working this close to the release date, especially since AMD wants it to look good.
You are missing the point here. What i'm saying and what you have pointed out that the turbo isn't working for AMD but CPU still holds it's own. With the turbo it would be rocking even higher on the charts. That's my point. From what I seen so far with AMD CPU's lineup and performance I'm really amazed by it. Putting to consideration that this is only engineering sample I wouldn't be surprised if the end product be better than the sample one. there's still time for improvement and fixes.
Posted on Reply
#87
efikkan
ratirt
You are missing the point here. What i'm saying and what you have pointed out that the turbo isn't working for AMD but CPU still holds it's own. With the turbo it would be rocking even higher on the charts. That's my point. From what I seen so far with AMD CPU's lineup and performance I'm really amazed by it. Putting to consideration that this is only engineering sample I wouldn't be surprised if the end product be better than the sample one. there's still time for improvement and fixes.
You seriously need to read the previous post again. There is no evidence that turbo is not working, and of course it would be working if the chips are to be released in a couple of weeks.
No, there is not time for improvement and fixes. It takes 4-6 months to incorporate any change in silicon. The only parameters they have the ability to tweak at this point are the binning and the clock speeds/voltage.
Posted on Reply
#88
ratirt
efikkan
You seriously need to read the previous post again. There is no evidence that turbo is not working, and of course it would be working if the chips are to be released in a couple of weeks.
No, there is not time for improvement and fixes. It takes 4-6 months to incorporate any change in silicon. The only parameters they have the ability to tweak at this point are the binning and the clock speeds/voltage.
well it's not like they started fixing it just now since it's been a while from AMD's ZEN announcement. The fixes simply continue and those benchmark leaks are only confirming that AMD's testing the CPU's which may be already tweaked or it is still going on before final release.
There is no confirmation that the turbo boost is working during the benchmark either and in the article author is assuming only that it doesn't. So it depends. What i'm saying is regardless if the turbo is working or not the performance is great and if it really isn't working during the benchmark that would only mean the CPU will performs even better with the boost.
Posted on Reply
#89
TheGuruStud
Intel should be forced to publicly admit they lied, AGAIN. Oh, boy, another 100 mhz bump is 15%? Or they tweaked the iGPU to deliver 15% more perf in one instance? Or they improved decoding 15%?

I mean, they might as well say, "We know you're dumb, so believe it and buy some LOLtel chips."
Posted on Reply
#90
ratirt
Loltel :) man that's a good one :D I've put so much faith in AMD's chips hope they don't disappoint me though it would be hard since i've seen what they can do :)
My pick is 1700x or pro though i'm not sure what pro gives either way i'm buying AMD. even if it aint superior to intel it's worth buying just for the sake of it :D
Posted on Reply
#91
Melvis
I fell asleep reading this thread....
Posted on Reply
#92
redeye
Intel can still get away with saying 10-15% gains, by using the ultralow power cpu's and increasing those cpu's clock speed 10-15%.
A 1Ghz base to 1.15Ghz is still 15% increase...
IMO that is a "cheat" but it is truly a 15% increase on one Cpu, not all...
Posted on Reply
#93
ratirt
redeye
Intel can still get away with saying 10-15% gains, by using the ultralow power cpu's and increasing those cpu's clock speed 10-15%.
A 1Ghz base to 1.15Ghz is still 15% increase...
IMO that is a "cheat" but it is truly a 15% increase on one Cpu, not all...
Disagree with that logic. It is a 15% increase in clock speed not overall IPC. 15% in clock speed is like nothing compare to relative performance. Although it also depends of the CPU and it's capabilities but anyway with CPU's we know it's like nothing.
Posted on Reply
#94
ratirt
Melvis
I fell asleep reading this thread....
Then stop spending a lot of time in front of computer and go to bed and for the record please stay on topic next time. If you wanna share your mood status or discuss sleeping problems create your own thread. I'm sure TPU has a place for that :)
Posted on Reply
#95
Melvis
ratirt
Then stop spending a lot of time in front of computer and go to bed and for the record please stay on topic next time. If you wanna share your mood status or discuss sleeping problems create your own thread. I'm sure TPU has a place for that :)
o_O I dont think you get it, read everyone elses replies and you might understand why I said what I said :slap:
Posted on Reply
#96
ratirt
Melvis
o_O I dont think you get it, read everyone elses replies and you might understand why I said what I said :slap:
Well reading all of it requires time (or is boring for you hard to tell) so if you think that I didn't get it is understandable although wrong and you didn't get even my 1 reply :) which is frustrating. Stay on topic and share stuff of the subject for that particular topic. Melvis, sir.
Hint: I don't care about your sleeping problems ( don't wanna say "we" since others might but that can be discussed in other thread since this one is reserved) :) You could have given at least a comment for what was written above since you had to go through all the replies :)
Posted on Reply
#97
Melvis
ratirt
Well reading all of it requires time (or is boring for you hard to tell) so if you think that I didn't get it is understandable although wrong and you didn't get even my 1 reply :) which is frustrating. Stay on topic and share stuff of the subject for that particular topic. Melvis, sir.
Hint: I don't care about your sleeping problems ( don't wanna say "we" since others might but that can be discussed in other thread since this one is reserved) :) You could have given at least a comment for what was written above since you had to go through all the replies :)
*Facepalm* Do I have to add in everytime the word Sarcasm?
Posted on Reply
Add your own comment