• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Discussion Intel current and future Lakes & Rapids thread

Page 146 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Gideon

Senior member
Nov 27, 2007
980
1,532
136
Yep, prices will have to be adjusted. Action - reaction. Where will it end?
What I love about it the most is that due-to competition from AMD, Intel will finally stop artificially castrating ... *ahem* ... segmenting its mid-range CPUs - selling pretty much the best they can produce at every price point. This should always be the case (and the end-goal for every enthusiast).

The final verdict of course depends on the price and clocks, but overall it's quite impressive what they have been able to squeeze out of the 14nm process. This should hold them over until Zen 3 arrives (probably in the end of Q3 2020), but by then the process limitation is really starting to manifest itself.
 
  • Like
Reactions: pcp7

JoeRambo

Senior member
Jun 13, 2013
849
524
136
The i9 would still be in a really odd place though. It won't be any faster in gaming than the 8 core
When we went from 6 to 8 cores, I think in most games improvement came from larger L3 cache. if 10 cores have 20MB of L3, It could continue scaling gaming performance even if most latency bound games are not using extra cores.
+ Intel has been stuck on with low speed DDR4 for 5 years now, if they can move to JEDEC 3200 speed, that would help them with benchmark numbers on sites like AT that test "stock" that is not relevant for enthusiasts but make certain persons on this forum happy.
 

eddman

Senior member
Dec 28, 2010
239
86
101
It's been rumored in multiple places that the i3 and i5 and i7 will get HT.
Read more closely before replying. 9500 and 9700 are coffee lake and already out there and yet are shown as having HT in the slides.

Does intel make such mistakes in their official material?
 
Last edited:

Shivansps

Diamond Member
Sep 11, 2013
3,019
687
136

EXCLUSIVEHARDWARELEAK
Intel’s Entire 10th Gen Comet Lake Desktop CPU Lineup Leaked – Up To 10 Cores, 20 Thread Core i9-10900, 80W TDP, W480 Chipset With New LGA 1200 Socket.


and now the Intel machine is in high gear,.




No need to shout.


esquared
Anandtech Forum Director
Intel intends to sell 2/2 cpus in 2020 to 2021?... Come on... the I3 is really 4/8 but they cant make the Celeron 2/4 and the more expensive Pentiums 4/4? really?
 

jpiniero

Diamond Member
Oct 1, 2010
8,294
1,356
126
Intel intends to sell 2/2 cpus in 2020 to 2021?... Come on... the I3 is really 4/8 but they cant make the Celeron 2/4 and the more expensive Pentiums 4/4? really?
Core Celeron and Pentium are basically nothing more than throwaway parts that are otherwise defective. So yeah I could see them needing both to be dual core to salvage the parts that have only two working cores.
 

Gideon

Senior member
Nov 27, 2007
980
1,532
136
When we went from 6 to 8 cores, I think in most games improvement came from larger L3 cache. if 10 cores have 20MB of L3, It could continue scaling gaming performance even if most latency bound games are not using extra cores.
Initially I had the very same thought about 9900K, but the fact that 9700K has only 12MB of cache and is performing almost exactly the same lead me to question that theory. It's odd, but it seems that either Skylake doesn't benefit that much from the extra 4MB of cache or that somehow disabling SMT improves performance enough to compensate (has any site done 9900K SMT off vs 9700K benchmarks?)

+ Intel has been stuck on with low speed DDR4 for 5 years now, if they can move to JEDEC 3200 speed, that would help them with benchmark numbers on sites like AT that test "stock" that is not relevant for enthusiasts but make certain persons on this forum happy.
That's true, I hope they change that. Anandtech claims that this is because most of the people don't even visit BIOS to enable XMP (I'm not that convinced in the statistic forCPUs like 9900K)

eddman said:
Read more closely before replying. 9500 and 9700 are coffee lake and already out there and yet are shown as having HT in the slides.

Does intel make such mistakes in their official material?
Yeah, sorry about that. The WCCFTech document seems suspicious for a number of reasons to me (clock-speeds also seem strange). I still think that the rumors of the Comet Lake having SMT accross the stack are almost certainly true.
 

Zucker2k

Golden Member
Feb 15, 2006
1,193
536
136
It's odd, but it seems that either Skylake doesn't benefit that much from the extra 4MB of cache or that somehow disabling SMT improves performance enough to compensate (has any site done 9900K SMT off vs 9700K benchmarks?)
I believe it is SMT off influencing the results. One of the reviews of the 9900K had SMT off results and that was the better result for the 9900K
 

DrMrLordX

Lifer
Apr 27, 2000
16,387
5,289
136
Does anyone know what happened to the Core i7 1068G7 CPU? It's nowhere to be seen months after Intel announced it. It's the only 10nm CPU running at 4.1GHz.
The Acer Swift 5 was supposed to have the 1068G7 as an option. I haven't seen any Swift 5's with that CPU, though.
 

liahos1

Senior member
Aug 28, 2013
573
45
91
Does anyone know what happened to the Core i7 1068G7 CPU? It's nowhere to be seen months after Intel announced it. It's the only 10nm CPU running at 4.1GHz.
i believe its in the dell xps model and the surface pro and surface laptop models
 

D283W

Junior Member
Dec 11, 2017
4
1
51
Does anyone know what happened to the Core i7 1068G7 CPU? It's nowhere to be seen months after Intel announced it. It's the only 10nm CPU running at 4.1GHz.
I've been using a Dell laptop for a little over two months with one of these CPUs in it. Pretty happy with it.
 

yeshua

Member
Aug 7, 2019
125
50
61
itvision.altervista.org
i believe its in the dell xps model and the surface pro and surface laptop models
And you can buy them exactly where? Sorry for being sarcastic but I wouldn't ask if they were indeed laptops being sold with this particular CPU. There's nothing on the market - not even an indication that something is around the corner. On Reddit someone justifiably called this CPU vaporware.
 

D283W

Junior Member
Dec 11, 2017
4
1
51
Can you post a link? I couldn't find any 1068g7 on Dell web site, while there are many with 1065g7.
You're right, I may have missed the difference between the 1065 and the 1068. I think it IS the 1065g7 in the model of the 2 in 1 that I have.
 

Nothingness

Platinum Member
Jul 3, 2013
2,153
397
126
You're right, I may have missed the difference between the 1065 and the 1068. I think it IS the 1065g7 in the model of the 2 in 1 that I have.
Heh, it's not hard to make a mistake given the nomenclature used by Intel and others :)
 

liahos1

Senior member
Aug 28, 2013
573
45
91
ahh my bad. I believe the 1068g7 is the 28w version? Ive seen the 15w 1065g7 sorry for the mistake!
 

IntelUser2000

Elite Member
Oct 14, 2003
7,102
1,644
136
Does anyone know what happened to the Core i7 1068G7 CPU? It's nowhere to be seen months after Intel announced it. It's the only 10nm CPU running at 4.1GHz.
28W CPUs are almost vaporware. They almost never use them. Some Vaios used them but that's about it. They don't even have it on their ark page. Even Apple will likely not use them this generation because they are busy with all the ARM transition rumors. If they do it'll likely be the 1065G7.

Manufacturers are just going to ignore it because the regular 15W U does everything. They can set it to 25W if they want and change it dynamically too. 2-in-1s like the XPS 7390 change its PL1 level depending on whether its in presentation mode, clamshell, or folded tablet. The base frequency is pretty much irrelevant since the system will adjust to whatever the load/thermal/power constraint is.

Officially the 15W U only has two cTDP modes with Base at 15W and cTDPup at 25W, but I've seen much more granular states and some even exceeding the cTDPup figure(such as with previous XPS 2-in-1 using 9W Y mode in some circumstances). Intel makes the chips so if the manufacturer asks for it, they'll allow for any power level between 15W and 25W.

Nearly no one used the previous 28W CPUs either. 3.9 to 4.1GHz is a 5% gain. It's insignificant.
 
  • Like
Reactions: coercitiv

yeshua

Member
Aug 7, 2019
125
50
61
itvision.altervista.org
28W CPUs are almost vaporware. They almost never use them. Some Vaios used them but that's about it. They don't even have it on their ark page. Even Apple will likely not use them this generation because they are busy with all the ARM transition rumors. If they do it'll likely be the 1065G7.

Manufacturers are just going to ignore it because the regular 15W U does everything. They can set it to 25W if they want and change it dynamically too. 2-in-1s like the XPS 7390 change its PL1 level depending on whether its in presentation mode, clamshell, or folded tablet. The base frequency is pretty much irrelevant since the system will adjust to whatever the load/thermal/power constraint is.

Officially the 15W U only has two cTDP modes with Base at 15W and cTDPup at 25W, but I've seen much more granular states and some even exceeding the cTDPup figure(such as with previous XPS 2-in-1 using 9W Y mode in some circumstances). Intel makes the chips so if the manufacturer asks for it, they'll allow for any power level between 15W and 25W.

Nearly no one used the previous 28W CPUs either. 3.9 to 4.1GHz is a 5% gain. It's insignificant.
1065G7 base frequency 1.3GHz
1068G7 base frequency 2.3GHz

We are talking about 77% difference in performance under sustained load. That's not 5% but what do I know anyways. And the conversation is not even about performance per se. It's about Intel's 10nm node being unable to reach 4.1GHz frequency which spells disaster for its theorized desktop Ice Lake CPUs unless someone starts calling NUCs desktop PCs.
 

ondma

Golden Member
Mar 18, 2018
1,497
418
106
1065G7 base frequency 1.3GHz
1068G7 base frequency 2.3GHz

We are talking about 77% difference in performance under sustained load. That's not 5% but what do I know anyways. And the conversation is not even about performance per se. It's about Intel's 10nm node being unable to reach 4.1GHz frequency which spells disaster for its theorized desktop Ice Lake CPUs unless someone starts calling NUCs desktop PCs.
Obviously, if they TDP up the 1065 to 25 watts, there wont in any way close to a 77% difference in sustained performance. You are actually contradicting yourself. First you say the point is the turbo frequency, which as 2000 said is only 5% difference, but them you base your perfomance calculation on the largely irrelevant base frequency.
 
Last edited:

Thunder 57

Golden Member
Aug 19, 2007
1,516
1,393
136
Obviously, if they TDP up the 1065 to 25 watts, there wont in any way close to a 77% difference in sustained performance. You are actually contradicting yourself. First you say the point is the turbo frequency, which as 2000 said is only 5% difference, but them you base your perfomance calculation on the largely irrelevant base frequency.
I don't see how a guaranteed GHz is "largely irrelevant". Yes, the 1.3GHz will certainly turbo as much as it can which is good, but it could fall all the wall back to 1.3GHz vs 2.3GHz. I mean, it's pretty bad when you can't even compare the two chips.
 

ASK THE COMMUNITY

TRENDING THREADS