Intel's Future Chips: News, Rumours & Reviews

[goldstone77]

Intel Announces Discrete GPU Prototype
Intel's GPU core consists of a vector core called "EU (Execution Unit)". One EU has two 4-way vector units. FP32 (32-bit floating point arithmetic) unit is 4-way, 128-bit long vector unit. The EU includes two vector units and 7 multithreaded general purpose registers (GRF), instruction fetchers and branch units.

Spoiler

In the prototype chip, six EUs are bound and constitute "Sub-Slice (SS: sub slice)". In addition to the EU, samplers (texture units), L1 / L2 caches, thread dispatchers, data ports, etc. are attached to the sub slices.

In this chip, it seems that three sub slices constitute one "Slice". Three EUs from SS 0 to SS 2 can be seen in the die picture. Of the three EU sub-slices, the new power control mechanism of this time is implemented in two. SS1 and SS2 are newly designed EU.

SS0 is designed as usual for comparison. A slice composed of three SSs is a full-function GPU core, and includes a graphics fixed function (FF: Fixed Function), a command streamer, an L3 cache, and the like.

The test chip has 1 slice configuration, and it carries 3 sub slices in total, 18 EUs. The product summation unit of the FP32 (32-bit floating point operation) in the vector unit is a calculation of 144 units in total. The low-end GT 1 of Intel graphics for PC is 96 FP 32 in standard implementation.

Because the test chip is 144 units with low frequency LP implementation, roughly speaking it can say that it is a graphics level performance for low end PC. Also, a normal GPU core comes with a multimedia core including a moving picture codec etc. However, it is not specified whether it is contained in this prototype chip or not.

The GPU chip announced at ISSCC includes not only the GPU core but also a system agent (SA: system agent) including a control system and around I / O. The memory interface is the host side. For that reason, the SA also has a large paging cache of 4 MB. It is connected to the host PC via the FPGA bridge. As far as the configuration is concerned, it is a prototype for demonstration experiments.

For more information and more graphs you can look here :https://translate.google.com/translate?depth=1&nv=1&rurl=translate.google.com&sp=nmt4&tl=en&u=https://pc.watch.impress.co.jp/docs/column/kaigai/1107078.html&xid=17259,15700021,15700105,15700122,15700124,15700149,15700168,15700173,15700201

However, the prototype chip of this time is not a level that can be put on the market as a product. It is a GPU chip for technical verification to the last. It is based on existing Intel 's built - in GPU architecture and examines power and performance control technology using integrated voltage regulator (IVR).

They are just using this for testing power, it's not a new design or anything special it is just being used to test voltages.

Edit: It's also mentioned here:

Update: It turns out Intel presented a paper describing a new power management technique for exiting GPU technology rather than a new discrete graphics solution, contrary to earlier reports. According to an Intel spokesperson:

Last week at ISSCC, Intel Labs presented a research paper exploring new circuit techniques optimized for power management. The team used an existing Intel integrated GPU architecture (Gen 9 GPU) as a proof of concept for these circuit techniques. This is a test vehicle only, not a future product. While we intend to compete in graphics products in the future, this research paper is unrelated. Our goal with this research is to explore possible, future circuit techniques that may improve the power and performance of Intel products.

https://liliputing.com/2018/02/intel-unveils-discrete-gpu-prototype.html

 

[juanrga]

There is not any update. SRAM densities have been known since past year and it has been reported that Intel 10nm has slightly larger SRAM cell densities than competing 7nm nodes. On the other hand Intel 10nm has a density lead in logic over competitors 7nm

Bolded part is wrong, Intel 10 nm dosn't have larger SRAM cell densities than 7nm. i is the other way around. 7nm has denser SRAM than intel 10 nm. To be more precise, GloFo 7nm HD cell is 15 % denser than intel 10nm HD cell and 7nm HP cell is 25% denser than intel 10 nm HP cell.

You are right.

 

[-Fran-]

https://techreport.com/news/33306/rumor-mobile-coffee-lake-chip-benches-pop-up-in-geekbench-database

Beep, bop, beep!

Cheers!

 

[YoAndy]

https://techreport.com/news/33306/rumor-mobile-coffee-lake-chip-benches-pop-up-in-geekbench-database

Beep, bop, beep!

Cheers!

Those are great scores for a laptop

 

[goldstone77]

Those a great score for a laptop or desktop!

 

[YoAndy]

Those a great score for a laptop or desktop!

Yes but Laptops are usually weaker than desktops, meaning that those scores(4980 single-thread) are greater for a laptop than for a desktop.
For comparison the desktop i7-8700K scores 5941-single-thread
The best Ryzen is the AMD Ryzen Threadripper 1950X 3.4 GHz (16 cores) and scores 4389-single

https://browser.geekbench.com/v4/cpu/compare/6157346?baseline=7136186

 

[-Fran-]

Those a great score for a laptop or desktop!

Yes but Laptops are usually weaker than desktops, meaning that those scores(4980 single-thread) are greater for a laptop than for a desktop.
For comparison the desktop i7-8700K scores 5941-single-thread
The best Ryzen is the AMD Ryzen Threadripper 1950X 3.4 GHz (16 cores) and scores 4389-single

https://browser.geekbench.com/v4/cpu/compare/6157346?baseline=7136186

Not to be a party pooper, but keep in mind you don't know under what *cooling* conditions those tests were ran. The Turbo is really impressive for a laptop CPU, no arguments against that, but now the question is for how long that is going to be sustained and with what type of sacrifice.

Wish for the best, expect the worst, right? 😛

Cheers!

 

[juanrga]

Those a great score for a laptop or desktop!

Yes but Laptops are usually weaker than desktops, meaning that those scores(4980 single-thread) are greater for a laptop than for a desktop.
For comparison the desktop i7-8700K scores 5941-single-thread
The best Ryzen is the AMD Ryzen Threadripper 1950X 3.4 GHz (16 cores) and scores 4389-single

https://browser.geekbench.com/v4/cpu/compare/6157346?baseline=7136186

Not to be a party pooper, but keep in mind you don't know under what *cooling* conditions those tests were ran. The Turbo is really impressive for a laptop CPU, no arguments against that, but now the question is for how long that is going to be sustained and with what type of sacrifice.

Wish for the best, expect the worst, right? 😛

It is a 45W chip. This is not a CoffeeLake U model for uiltrabooks. But still impressive that a 45W model ST performance is only 80% that of top 95W models and ahead 180W models from the competence.

 

[-Fran-]

Those a great score for a laptop or desktop!

Yes but Laptops are usually weaker than desktops, meaning that those scores(4980 single-thread) are greater for a laptop than for a desktop.
For comparison the desktop i7-8700K scores 5941-single-thread
The best Ryzen is the AMD Ryzen Threadripper 1950X 3.4 GHz (16 cores) and scores 4389-single

https://browser.geekbench.com/v4/cpu/compare/6157346?baseline=7136186

Not to be a party pooper, but keep in mind you don't know under what *cooling* conditions those tests were ran. The Turbo is really impressive for a laptop CPU, no arguments against that, but now the question is for how long that is going to be sustained and with what type of sacrifice.

Wish for the best, expect the worst, right? 😛

It is a 45W chip. This is not a CoffeeLake U model for uiltrabooks. But still impressive that a 45W model ST performance is only 80% that of top 95W models and ahead 180W models from the competence.

So what if it's a 15W part or not? You can just volt it up and make it suck up 150W if you want to get higher scores. I'm just putting the salt on the table, you're free to use it if you want, but don't spill it.

EDIT: Typos.

 

[-Fran-]

https://techreport.com/news/33317/report-first-benchmarks-of-intel-hades-canyon-nuc-surface

In case anyone is interested in a few first bench numbers for the Intel-AMD combi.

Cheers! 😛

 

[YoAndy]

https://techreport.com/news/33317/report-first-benchmarks-of-intel-hades-canyon-nuc-surface

In case anyone is interested in a few first bench numbers for the Intel-AMD combi.

Cheers! 😛

And what will be their ?

 

[YoAndy]

https://techreport.com/news/33317/report-first-benchmarks-of-intel-hades-canyon-nuc-surface

In case anyone is interested in a few first bench numbers for the Intel-AMD combi.

Cheers! 😛

And what will be their price?

 

[-Fran-]

https://techreport.com/news/33317/report-first-benchmarks-of-intel-hades-canyon-nuc-surface

In case anyone is interested in a few first bench numbers for the Intel-AMD combi.

Cheers! 😛

And what will be their price?

Eleventy million dawgllars.

No, I really don't know, but you can imagine premium-laptop range, I'd say. So, ~1K?

Cheers!

 

[YoAndy]

https://techreport.com/news/33317/report-first-benchmarks-of-intel-hades-canyon-nuc-surface

In case anyone is interested in a few first bench numbers for the Intel-AMD combi.

Cheers! 😛

And what will be their price?

Eleventy million dawgllars.

No, I really don't know, but you can imagine premium-laptop range, I'd say. So, ~1K?

Cheers!

1k for a box that can't play 4k, Have heard about this $400+ other box named XBOX ONE X that can play games at 4k and 1440p.:lol:

 

[-Fran-]

1k for a box that can't play 4k, Have heard about this $400+ other box named XBOX ONE X that can play games at 4k and 1440p.:lol:

Well, it is smaller than an XBox and PS4 (from what I could see) and it also consumes way less? Around ~220W?

I can't argue with the "not 4K for 1K is weird" idea, but this is a full fledged PC anyway. Gaming laptops in the 2K range can do 4K, just barely. My own new laptop, in theory, can do some 4K with an i7-7700HQ and a 1070, but it's way over 1K.

So, in context/perspective, 1K sounds possible. Not necessarily reasonable. It's Intel at the end of the day putting this product
out there 😛

EDIT2: https://www.amazon.com/Intel-NUC-mini-NUC6i7KYK-Core/dp/B01DJ9XS52

If you add RAM and M2, it is ~1K, so this one might be even above that.

Cheers!

EDIT: Typos.

 

[YoAndy]

1k for a box that can't play 4k, Have heard about this $400+ other box named XBOX ONE X that can play games at 4k and 1440p.:lol:

Well, it is smaller than an XBox and PS4 (from what I could see) and it also consumes way less? Around ~220W?

I can't argue with the "not 4K for 1K is weird" idea, but this is a full fledged PC anyway. Gaming laptops in the 2K range can do 4K, just barely. My own new laptop, in theory, can do some 4K with an i7-7700HQ and a 1070, but it's way over 1K.

So, in context/perspective, 1K sounds possible. Not necessarily reasonable. It's Intel at the end of the day putting this product
out there 😛

EDIT2: https://www.amazon.com/Intel-NUC-mini-NUC6i7KYK-Core/dp/B01DJ9XS52

If you add RAM and M2, it is ~1K, so this one might be even above that.

Cheers!

EDIT: Typos.

And that one that is $566 comes with Intel Iris Pro Graphics 580, witch is worse way worse than a GTX 960M. I'm just saying that for the price and just for gaming, I'll rather buy the xbox one x..

 

[-Fran-]

1k for a box that can't play 4k, Have heard about this $400+ other box named XBOX ONE X that can play games at 4k and 1440p.:lol:

Well, it is smaller than an XBox and PS4 (from what I could see) and it also consumes way less? Around ~220W?

I can't argue with the "not 4K for 1K is weird" idea, but this is a full fledged PC anyway. Gaming laptops in the 2K range can do 4K, just barely. My own new laptop, in theory, can do some 4K with an i7-7700HQ and a 1070, but it's way over 1K.

So, in context/perspective, 1K sounds possible. Not necessarily reasonable. It's Intel at the end of the day putting this product
out there 😛

EDIT2: https://www.amazon.com/Intel-NUC-mini-NUC6i7KYK-Core/dp/B01DJ9XS52

If you add RAM and M2, it is ~1K, so this one might be even above that.

Cheers!

EDIT: Typos.

And that one that is $566 comes with Intel Iris Pro Graphics 580, witch is worse way worse than a GTX 960M. I'm just saying that for the price and just for gaming, I'll rather buy the xbox one x..

Good for you! 😀

 

[goldstone77]

Intel Cascade Lake SP: Up to 245 watts TDP (too) through 3D XPoint DIMMs note
9.3.2018 12:35 clock Volker Rißka

Intel's upcoming server-processor generation codenamed Cascade Lake-SP will be able to build on a maximum of 28 cores as the successor to the Skylake SP , however, the TDP of the new Xeon CPUs increases again significantly, from 205 watts today up to 245 watts in the future.

In addition to adjusting the clock rates, the Cascade Lake SP are the first processors, which not only offer DDR4-2933 in the server environment but also support the new memory latches Apache Pass . According to rumors from Asia, the DIMMs based on 3D XPoint should be powerful but also extremely power hungry . They are said to have approximately three times the power consumption of conventional DIMMs, with 15 to 18 watts being discussed there. According to previously known plans, the new DIMMs should be usable only on selected CPU models, in half of the available memory banks.

https://www.computerbase.de/2018-03/intel-cascade-lake-sp-245-watt/
Is it just me or is that Purley 4S configuration look familiar?

 

[juanrga]

Intel Cascade Lake SP: Up to 245 watts TDP (too) through 3D XPoint DIMMs note
9.3.2018 12:35 clock Volker Rißka

Intel's upcoming server-processor generation codenamed Cascade Lake-SP will be able to build on a maximum of 28 cores as the successor to the Skylake SP , however, the TDP of the new Xeon CPUs increases again significantly, from 205 watts today up to 245 watts in the future.

In addition to adjusting the clock rates, the Cascade Lake SP are the first processors, which not only offer DDR4-2933 in the server environment but also support the new memory latches Apache Pass . According to rumors from Asia, the DIMMs based on 3D XPoint should be powerful but also extremely power hungry . They are said to have approximately three times the power consumption of conventional DIMMs, with 15 to 18 watts being discussed there. According to previously known plans, the new DIMMs should be usable only on selected CPU models, in half of the available memory banks.

https://www.computerbase.de/2018-03/intel-cascade-lake-sp-245-watt/
Is it just me or is that Purley 4S configuration look familiar?

3D XPoint DIMMS have to consume more power than ordinary DIMMS, because 3D Xpoint is persistent memory. The adequate comparison is with other kind of persistent memory technologies such as SSDs. It cost more power to read the info from a SSD and move it to ordinary DIMMS, that having the data directly stored on 3D Xpoint DIMMs

That Purley 4S configuration looks familiar, because Intel has been using it for ages. This is 4S configuration for ancient Haswell Xeons

 

[-Fran-]

Oh, this is going to be interesting.

Why odes XPoint being "persistent" affects power consumption like that? Haha, I'm already looking forward to your answer, Juan.

Cheers! xD

 

[goldstone77]

3D XPoint was one of the biggest over hyped let downs for me that Intel ever came up with, and it easy to see why Micron and Intel split away from this combined venture!

 

[goldstone77]

Intel’s six-Core i9-8950HK, i7-8850H, i7-8750H Cinebench score leaked
Published: 15th Mar 2018, 14:52 GMT
https://videocardz.com/75326/intels-six-core-i9-8950hk-i7-8850h-i7-8750h-cinebench-score-leaked
Desktop like scores coming from laptop chips! And the leak for the Core i9-8950HK@4.8GHz I found was true.

http://www.tomshardware.com/forum/id-1581001/intel-future-chips-news-rumours-reviews/page-58.html#20656062

 

[YoAndy]

That's really good for Laptop users.

 

[-Fran-]

https://techreport.com/news/33404/rumor-eight-core-coffee-lake-s-cpu-shows-up-in-3dmark-database

Plan B might get 8 cores!

Cheers!

 

[goldstone77]

Damn, the "Core Wars" are going to be great for the consumer!