• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

"Nehalem" Lead Architect Rejoins Intel to Work on New High-Performance Architecture

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,670 (7.43/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
The original "Nehalem" CPU microarchitecture from 2008 was pivotal to Intel, as it laid the foundation for Intel's mainline server and client x86 processors for the following 12-odd years. Glenn Hinton, the lead architect behind "Nehalem," announced that he is rejoining Intel after 3 years of retirement, to work on a new high-performance CPU project. Hinton states that his decision to rejoin Intel out of his retirement was influenced by Pat Gelsinger joining the company as its new CEO. Jim Keller, a CPU architecture lead behind several commercially-successful architectures, recently left Intel after a brief stint leading an undisclosed CPU core project. Keller later took up the mantle of CEO at hardware start-up Tenstorrent.

Pat Gelsinger leading Intel is expected to have a big impact on its return to technological leadership in its core businesses, as highlighted in Gelsinger's recent comments on the need for Intel to be better than Apple (which he referred to as "that lifestyle company") at making CPUs, in reference to Apple's new M1 chip taking the ultraportable notebook industry by storm. The other front Intel faces stiff competition from, is AMD, which has achieved IPC parity with Intel, and is beating it on energy-efficiency, taking advantage of the 7 nm silicon fabrication process.



View at TechPowerUp Main Site
 
Awesome news.
 
I think they made an offer Glenn H. couldnt refuse, IDK coming out of retirement just to face all that pressure and stress again.

Also, Nehelam spelled backwards is rooster.
 
Last edited:
Please, for the love of humanity design something for mainstream without an iGPU. RKL could've been 10-core without the iGPU.
 
Sounds like exciting times ahead from Intel. They finally got up and wide awake from their long slumber.

Please, for the love of humanity design something for mainstream without an iGPU. RKL could've been 10-core without the iGPU.
There is no one size fits all solution to this problem. You may not want it, but there will be others who would prefer to have the iGPU since they are not running anything graphically intensive. RKL may be die size limited, but I feel it is more power and thermal limited, thus, Intel had to make do with 8 cores instead of 10. We can wait and see when it is officially launched and reviewed by independent sources.
 
Nehalem was a hell of a CPU. the one that started the core i trend. and it was a major performance leap compared to previous prescott.
Still one a core i7 950 inside my laptop, and it shows its age compared to what we have now....

If One its leading architects comes out of retirement to "work with intel" again for a "high performance" architekture, you realized the shiet theyve been pushing for the past decade.
Which is what ive been bambling all along, Minimam generation performance increments, and once AMD flexed their muscle tired of their shiet(intels shiet), intel entered panik mode.
Intel: "we need to get our shiet together guys"
Intel2 : "faq, what the faq do we do, we cant build cpus!"
intel3: "i know what, we hire nehalem arhitect to build as a cpu"
intel2 "but hes retired"
intel3: "we gief him a shittonn of money - we sucked more than enough from plebs and enterprises during the "milk-era", he wont have a choice but to come out of retirement"
intel1: "good point people, good jub, make it so"

Then we hear this news on TPU
 
Had to go back to my nearly 12 year old Nehalem system after my just over 5 year old Skylake's mb died, and now I'm waiting for Ryzen 5000 and Rocket Lake to release.
 
Please, for the love of humanity design something for mainstream without an iGPU. RKL could've been 10-core without the iGPU.

Has nothing to do with the iGPU...

Getting rid of the GPU doesn't magically free up power budget or space for more cores.

The 10900 and 10900K literally stretched the ringbus to its limits and the core to core latency was already showing the strain.
 
Last edited:
Oh this is good news for Intel it sounds like. Nahalem and X58 platform is for sure something special I think. I mean, I been on this for almost 12 years now with first a I7 920 and now i7 980X.

So the same person behind nehalem, might be what Intel needs. To move on to something game changing again. Cause let's be honest. Intels releases after nehalem and Sandy bridge has not been what I will call impressive.

Honestly for now, I am more impressed what AMD has come from and to with there zen chips.
 
Effectively, Intel is still using Nehalem. All they did was made some teeny tiny IPC improvements, increased clock speeds and core counts, updated the memory and PCI-E controllers and added new instructions (usable only in very specific applications).

They have not even touched the cache subsystem at all (just increased L3 size for higher core counts). They are finally doing that with Rocket Lake, but it is still just an increase in L1 and L2 sizes.

If you match a Comet Lake CPU core count, clock speed and memory speed to a Nehalem CPU, you will get marginally better performance after 12 years. Crazy.

Right now they are stuck not only on the same architecture, but also the same manufacturing process, which is why there has been zero progress made since 2015 Skylake.

Hopefully all these personnel changes will help. They really need some big changes. Thank God for AMD, they finally made Intel wake up.
 
go buy the hedt platform and quit crying.
Ok, thx. AMD has mainstream iGPUless CPU, APU and HEDT and look at how they're doing with that. Wonder what I would buy if I didn't need a discrete card and didn't mind RAM speed limited infinity fabric CPUs. Very few reasons to really have an iGPU all up and down the product stack. It shouldn't even be in a desktop CPU especially if they're making their own discrete solutions now they could sell along side of it. Ultra thin laptops and space limited products is about the only place they belong.
Has nothing to do with the iGPU...

Getting rid of the GPU doesn't magically free up power budget or space for more cores.

The 10900 and 10900K literally stretched the ringbus to its limits and the core to core latency was already showing the strain.
it doesn't HAVE to be 5GHz and I didn't really know about the core latency on Intel. How does it compare to a 5900x/5950x using Infinity fabric?
 
The article didn't make it overly clear, has Hinton been at Intel the whole time up until 3 years ago when he retired because as impressive as Nehalem is he didn't seem to do much after that, certainly after Sandy Bridge in 2010.

It is good news in general as it shows Pat is already reforming Intel and bringing good people back into the company but personally Keller would have been my target to finish what he started.
 
Please, for the love of humanity design something for mainstream without an iGPU. RKL could've been 10-core without the iGPU.
The chip was designed to work with a defective iGPU and be sold as F/KF. So if 20% of chips have a manufacturing defect, and half of those have a defect only in the iGPU because the iGPU is so big, it's a 90% yield for Intel and not 80%.
 
I see much lauding for Nehalem when it was just a proper improvement on 1366, which was an expesive socket/platform that only at triple-channel showed that gargantuan leap.
Meanwhile, i5 700s and i7 800s were stalemating Core 2 Duos and Quads, respectively (nevermind higher TDP), the latter on bog-priced 775s with DDR3 already because FSB+DMI could still punch DMI 2.0 in the face, but "OMG we have Hyperthreading again", except half the world was still on WinXP so whatever, and it was only when Sandy came around (and Windows 7) that that changed (then the 2500K and 2600K became the staple).

If this guy is going to bring anything worthwhile, is to shove all links to older uArchs to the trash and start from scratch, as AMD had to do with Summit Ridge.
 
X58 Platform was the f..king BEST!
 
Nehalem was a hell of a CPU. the one that started the core i trend. and it was a major performance leap compared to previous prescott.
Still one a core i7 950 inside my laptop, and it shows its age compared to what we have now....

If One its leading architects comes out of retirement to "work with intel" again for a "high performance" architekture, you realized the shiet theyve been pushing for the past decade.
Which is what ive been bambling all along, Minimam generation performance increments, and once AMD flexed their muscle tired of their shiet(intels shiet), intel entered panik mode.
Intel: "we need to get our shiet together guys"
Intel2 : "faq, what the faq do we do, we cant build cpus!"
intel3: "i know what, we hire nehalem arhitect to build as a cpu"
intel2 "but hes retired"
intel3: "we gief him a shittonn of money - we sucked more than enough from plebs and enterprises during the "milk-era", he wont have a choice but to come out of retirement"
intel1: "good point people, good jub, make it so"

Then we hear this news on TPU
was it? I thought it was rather meh until sandy bridge showed up
 
Nehalem was a hell of a CPU. the one that started the core i trend. and it was a major performance leap compared to previous prescott.
Still one a core i7 950 inside my laptop, and it shows its age compared to what we have now....

If One its leading architects comes out of retirement to "work with intel" again for a "high performance" architekture, you realized the shiet theyve been pushing for the past decade.
Which is what ive been bambling all along, Minimam generation performance increments, and once AMD flexed their muscle tired of their shiet(intels shiet), intel entered panik mode.
Intel: "we need to get our shiet together guys"
Intel2 : "faq, what the faq do we do, we cant build cpus!"
intel3: "i know what, we hire nehalem arhitect to build as a cpu"
intel2 "but hes retired"
intel3: "we gief him a shittonn of money - we sucked more than enough from plebs and enterprises during the "milk-era", he wont have a choice but to come out of retirement"
intel1: "good point people, good jub, make it so"

Then we hear this news on TPU
it's also possible government involved bringing him back, Microprocessor is strategic asset for US, and Intel is the only Semiconductor company that have in House Fabs in US


there simply so much force that won't allow intel failing
 
Looks like they are putting the dream team back together! Genuine excitement here. I love my X58. She is nice and safe in her box.

I am an Intel fan for sure, but.. AMD is on to something too..

Looking forward to see what these guys can do!

I have a feeling I will only be on AM4 for a year.
 
I see much lauding for Nehalem when it was just a proper improvement on 1366, which was an expesive socket/platform that only at triple-channel showed that gargantuan leap.
Meanwhile, i5 700s and i7 800s were stalemating Core 2 Duos and Quads, respectively (nevermind higher TDP), the latter on bog-priced 775s with DDR3 already because FSB+DMI could still punch DMI 2.0 in the face, but "OMG we have Hyperthreading again", except half the world was still on WinXP so whatever, and it was only when Sandy came around (and Windows 7) that that changed (then the 2500K and 2600K became the staple).
Expensive? Look at the reviews from the past.


Hi-End 1366 mobo for 350$ + 950 for 330, or "cost effective" 1366 for 200 + 920/930 for 240 (300 on release). And just a year before 775 mainstream mobo for 140$ or high-end for 250 + 8500/9550 for 300$. 1156/1155 was about the same as mentioned 775 platform.
And everyone who invested in x58 could use today with some upgrades, and its still capable. unless you want 300+FPS 4K HDR RTX Ultra detail AAA gaming experience of lootboxes and poor netcode.
 
Please, for the love of humanity design something for mainstream without an iGPU. RKL could've been 10-core without the iGPU.

That'll never happen in the mainstream... especially with what they've invested in Xe..
 
I see much lauding for Nehalem when it was just a proper improvement on 1366, which was an expesive socket/platform that only at triple-channel showed that gargantuan leap.
Meanwhile, i5 700s and i7 800s were stalemating Core 2 Duos and Quads, respectively (nevermind higher TDP), the latter on bog-priced 775s with DDR3 already because FSB+DMI could still punch DMI 2.0 in the face, but "OMG we have Hyperthreading again", except half the world was still on WinXP so whatever, and it was only when Sandy came around (and Windows 7) that that changed (then the 2500K and 2600K became the staple).

If this guy is going to bring anything worthwhile, is to shove all links to older uArchs to the trash and start from scratch, as AMD had to do with Summit Ridge.

The i5-750 was a sub-200 $ CPU that could be easily overclocked with BCLK to 3.6+ GHz on air and any chipset. I loved that CPU, it was when quad cores started to make sense.

Sandy Bridge is when they started locking everything away under K SKUs and P/Z chipsets. The only great thing about this architecture was high clock speeds. It just went downhill from there.
 
Back
Top