Thursday, June 11th 2020

Jim Keller Resigns from Intel

Intel today announced that systems designer-extraordinaire Jim Keller has departed the company citing personal reasons. Whether or not this is a blow to Intel likely depends on how far Jim Keller brought their Technology, Systems Architecture and Client Group throughout his two-year tenure at the company whilst serving as its Vice President. The semiconductor and chip architecture world isn't being driven by Mr. Keller himself, obviously; there are a number of architects and designers that bring the industry forward through their concerted efforts. However, it's hard to look past Jim Keller's pedigree when it comes to doing his job - if anything, AMD's Zen architecture is a testament to that, and has put Intel in the place we now see it in the CPU world.

To fill in the void, Intel has announced a reshuffling inside their Technology, Systems Architecture and Client Group. Jim Keller will still be serving with Intel for the next six months as a consultant, thus easing the transition. Read the full press-release below.
Today, Intel announced that Jim Keller has resigned effective June 11, 2020, due to personal reasons. Intel appreciates Mr. Keller's work over the past two years helping them continue advancing Intel's product leadership and they wish him and his family all the best for the future. Intel is pleased to announce, however, that Mr. Keller has agreed to serve as a consultant for six months to assist with the transition.

Intel has a vastly experienced team of technical leaders within its Technology, Systems Architecture and Client Group (TSCG) under the leadership of Dr. Venkata (Murthy) Renduchintala, group president of TSCG and chief engineering officer. As part of this transition, the following leadership changes will be made, effective immediately:

Sundari Mitra, the former CEO and founder of NetSpeed Systems and the current leader of Intel's Configurable Intellectual Property and Chassis Group, will lead a newly created IP Engineering Group focused on developing best-in-class IP.
Gene Scuteri, an accomplished engineering leader in the semiconductor industry, will head the Xeon and Networking Engineering Group.
Daaman Hejmadi will return to leading the Client Engineering Group focused on system-on-chip (SoC) execution and designing next-generation client, device and chipset products. Hejmadi has over two decades of experience leading teams delivering advanced SoCs both inside and outside of Intel.
Navid Shahriari, an experienced Intel leader, will continue to lead the Manufacturing and Product Engineering Group, which is focused on delivering comprehensive pre-production test suites and component debug capabilities to enable high-quality, high-volume manufacturing.
Intel congratulates Sundari, Gene, Daaman and Navid as we begin the next phase of our world-class engineering organization and look forward to executing on our exciting roadmap of products.
Source: Intel
Add your own comment

114 Comments on Jim Keller Resigns from Intel

#76
R-T-B
ARF
Is x86 dead with reached IPC wall once and forever?
If this years Ryzen releases have anything to say about it, no.
Imsochobo
Mesh is a hot garbage mess.
Mesh honestly works fine, it's just that it has higher latency, kinda like Ryzen chips. That's not the end all but it does hurt their "best gaming" claims because well, Ryzen already has them beat on every other front.
Posted on Reply
#77
ToxicTaZ
Imsochobo
also interconnect says hi :)
Ringbus doesn't scale.
Mesh is a hot garbage mess.
Good thing Intel 10th series is the last (Sky Lake) and finally done.

Q1 2021
11th gen (Rocket Lake) Willows Cove is the last 14nm++

Intel new 16 Cores big.Little architecture on new H6 LGA 1700 socket.

Now moves too...10nm++
Q4 2021
12th gen (Alder Lake) Golden Cove

Q4 2022
Which brings us to Jim Keller Ocean Cove project on 13th generation (Meteor Lake) 7nm+

Basically Intel Meteor Lake is Intel Zen2 aka version....

13th gen (Meteor Lake) is the coming fix to what everyone is complaining about... Ultra low power, double Cores, 7nm+, Ultra IPC.... More than 80 IPC Gain over 10th generation SkyLake IPC.

I'm really curious how far Jim Keller Ocean cove project is right now...
Posted on Reply
#78
ARF
R-T-B
If this years Ryzen releases have anything to say about it, no.
Zen 3 should be one of the last and final attempts in the development of x86-64 and the PC platform in this shape and form.
Zen 3 started its development at least 3-5 years ago, JK resigns today. There is quite a difference in the time frame.

The Sony PlayStation 5 development and design shows just how primitive the x86 is, where they needed a ton of new specialised functions and dedicated hardware sets in order to replace the missing x86 instructions which they need and the extremely slow execution on the Zen 2 cores if they didn't include these new specialised hardware functions in order to accelerate the gaming experience as they wish.

Try running a Celeron with an HDD on Windows 10.
You will see how painfully slow it is.
A simple update takes dozens of minutes or hours.

My Ryzen system with an SSD took longer than an hour to update from 19 09 to 20 04. This is just unacceptable and I will take the very first opportunity to get rid of x86 system in favour of RISC ARM which is a mignitude of times FASTER.

You will never see, even the slowest Android smartphone to update in hours. :kookoo::rolleyes:
Posted on Reply
#79
Assimilator
ARF
Zen 3 should be one of the last and final attempts in the development of x86-64 and the PC platform in this shape and form.
Zen 3 started its development at least 3-5 years ago, JK resigns today. There is quite a difference in the time frame.

The Sony PlayStation 5 development and design shows just how primitive the x86 is, where they needed a ton of new specialised functions and dedicated hardware sets in order to replace the missing x86 instructions which they need and the extremely slow execution on the Zen 2 cores if they didn't include these new specialised hardware functions in order to accelerate the gaming experience as they wish.

Try running a Celeron with an HDD on Windows 10.
You will see how painfully slow it is.
A simple update takes dozens of minutes or hours.

My Ryzen system with an SSD took longer than an hour to update from 19 09 to 20 04. This is just unacceptable and I will take the very first opportunity to get rid of x86 system in favour of RISC ARM which is a mignitude of times FASTER.

You will never see, even the slowest Android smartphone to update in hours. :kookoo::rolleyes:
LMAO.

Your ignorance is showing again.
Posted on Reply
#80
R0H1T
ARF
My Ryzen system with an SSD took longer than an hour to update from 19 09 to 20 04.
This is just unacceptable and I will take the very first opportunity to get rid of x86 system in favour of RISC ARM which is a mignitude of times FASTER.

You will never see, even the slowest Android smartphone to update in hours. :kookoo::rolleyes:
You know what, get a better SSD. How about the upcoming true gen4 980 pro from Sammy?

Depends on the handset & the size of the update.

I dunno, you seem to fluctuate from anywhere between Tucker Carlson on one end of the spectrum to John Oliver at the other. If you get what I mean :rolleyes:
Posted on Reply
#81
ARF
R0H1T
You know what, get a better SSD.
The problem is these forums are too many anti-objective employees of corporations working with x86, thus shilling and trolling in favour of their job.
Completely ignoring the global users' experiences, and more importantly - the huge list of disadvantages of their own achievements.

My SSD Western Digital Blue 3D NAND M.2 SSD 500GB is already the best SATA3 SSD on the market. Can't get better than it and it's not the problem..
Posted on Reply
#82
EarthDog
ARF
My Ryzen system with an SSD took longer than an hour to update from 19 09 to 20 04. This is just unacceptable and I will take the very first opportunity to get rid of x86 system in favour of RISC ARM which is a mignitude of times FASTER.

You will never see, even the slowest Android smartphone to update in hours
it took me ~5-10 mins including the DL all the way to 'first' boot. Sounds like a personal issue. :)

Funny how you're comparing phone os upgrades to pc's tho...the size of the upgrade packages are a lot smaller. A couple/few GB vs couple hundo MB? Even the much slower arm processors and a phone will chew through it. That said, I've had a few OS updates on my s9+ and the first on the s20+ also took 10 mins...and these are both high-end units with, at the time, the fastest snapdragons(?) on them.
Posted on Reply
#83
ARF
EarthDog
it took me ~5-10 mins including the DL all the way to 'first' boot. Sounds like a personal issue. :)

Funny how you're comparing phone os upgrades to pc's tho...the size of the upgrade packages are a lot smaller. A couple/few GB vs couple hundo MB? Even the much slower arm processors and a phone will chew through it. That said, I've had a few OS updates on my s9+ and the first on the s20+ also took 10 mins...and these are both high-end units with, at the time, the fastest snapdragons(?) on them.
:confused:

Well, the SSD's speeds are 560 MB/s reads , 530 MB/s writes as at shop.westerndigital.com/en-ie/products/internal-drives/wd-blue-3d-nand-sata-ssd#WDS250G2B0A



Now, my tests show:
From HDD1 to SSD1 where the Windows is:



Copy - paste in the same SSD folder:




What can the "personal" issue be ? Platform defect on the Acer Nitro AN515-42 ?
Posted on Reply
#84
EarthDog
Not sure, ARF. Just telling you my experiences with my system (16c/16t cpu, pcie 3.0 nvme m.2) and phones. I'll add to that my son's 2700x system with sata ssd for OS took a bit longer, but nowhere near how long yours did. Same with my 7900x + sata ssd system. No way the latter took 20 mins...60 mins is, seemingly, a problem and not normal.
Posted on Reply
#85
londiste
ARF
Well, the SSD's speeds are 560 MB/s reads , 530 MB/s writes as at shop.westerndigital.com/en-ie/products/internal-drives/wd-blue-3d-nand-sata-ssd#WDS250G2B0A
Now, my tests show:
From HDD1 to SSD1 where the Windows is:
Copy - paste in the same SSD folder:
First one - you are copying from an HDD. The speed is limited by speed of data source read - that is HDD speed.
Second one - small enough files? Small-ish file overhead eats speed more than anything, especially on the same drive.
Posted on Reply
#86
remixedcat
R0H1T
You know what, get a better SSD. How about the upcoming true gen4 980 pro from Sammy?

Depends on the handset & the size of the update.

I dunno, you seem to fluctuate from anywhere between Tucker Carlson on one end of the spectrum to John Oliver at the other. If you get what I mean :rolleyes:
Don't drag poor Tucker into this. LOL.

But "personal reasons" for a resignation in this day and age is prolly the fact he was busted with stuff he shouldn't and he took some plea deal.... I smell shenanigans.

Just like how Bob iger and disney are.... basement bob is still on training wheels and sippy cups for disney stuff.. while iger still has the whips.

Same shit pattern. So where's Intel's basement bob?
Posted on Reply
#87
ARF
EarthDog
Not sure, ARF. Just telling you my experiences with my system (16c/16t cpu, pcie 3.0 nvme m.2) and phones. I'll add to that my son's 2700x system with sata ssd for OS took a bit longer, but nowhere near how long yours did. Same with my 7900x + sata ssd system. No way the latter took 20 mins...60 mins is, seemingly, a problem and not normal.
londiste
First one - you are copying from an HDD. The speed is limited by speed of data source read - that is HDD speed.
Second one - small enough files? Small-ish file overhead eats speed more than anything, especially on the same drive.
Copying ISO Office and Windows files is very fast - an ISO office file of 0.9 GB happens instantly.
Windows ISO file that is larger begins with 1.1 GB/s and ends with around 430 MB/s...

That means that the problem is NOT in the SSD but in Windows itself. Updating is slower than new fresh installation.
Updating is very slow, it's like decompressing by the CPU that eats dozens of minutes, and then copying and pasting very small files.
Posted on Reply
#88
Assimilator
So because Windows updates slowly, x86 is the problem?

Amazing logic there. Absolutely flawless. Impeccable even.
ARF

There are two very important words there that you seem incapable of reading:

UP TO.
Posted on Reply
#89
ARF
The "up to" is wrong. It goes up to all the way up to 1.1 - 1.2 GB/s copying and pasting in the same folder.
Posted on Reply
#90
Lindatje
ARF
The "up to" is wrong. It goes up to all the way up to 1.1 - 1.2 GB/s copying and pasting in the same folder.
No that is not true. :laugh:
It`s read speeds up to 560MB/s and sequential write speeds up to 530MB/s.
Posted on Reply
#91
EarthDog
ARF
That means that the problem is NOT in the SSD but in Windows itself. Updating is slower than new fresh installation
for you, perhaps. Installing windows from a 10 Gbps USB takes around 10-15 mins on nvme drive... the update from DL to boot was, as I said, less than 10 mins.
Posted on Reply
#92
ARF
Lindatje
No that is not true. :laugh:
It`s read speeds up to 560MB/s and sequential write speeds up to 530MB/s.
I have just said that I observe higher speeds and you say it's untrue?

These are just speeds around the saturation of the SATA3 bus (which is by the way 600 MB/s, if you don't know) and no where it's written under what exactly circumstances and conditions a user would observe them.
Posted on Reply
#93
EarthDog
ARF
I have just said that I observe higher speeds and you say it's untrue?

These are just speeds around the saturation of the SATA3 bus (which is by the way 600 MB/s, if you don't know) and no where it's written under what exactly circumstances and conditions a user would observe them.
So... if the limit of the bus is 600 MB/s, and with overhead etc its more realistically 560 MB/s as you see most sata drives top out at, you saw double this speed? I'm not sure what you're trying to defend.

Would that be cache/burst? Either way, I dont care.. its off topic, lol.
Posted on Reply
#94
hat
Enthusiast
Am I missing something? I'm not sure what microprocessor architecture, be it x86, ARM or whatever, has to do with data throughput speeds...
Posted on Reply
#95
ToxicTaZ
Say on topic please!

We are talking about "Jim Keller" and his Ocean Cove project at Intel he leaving.

Talking about load times is a losing battle with technology changing rapidly over the next few years.

Both companies...have new sockets with higher bus speeds huge bandwidth increases with PCIe 5.0 coming. AMD AM5 and Intel H6 LGA 1700 sockets every thing is changing... PCIe 5.0, DDR5, USB4, technology's make all you guys load times irrelevant with the technology that "Jim Keller" was working on...

This is near future stuff....Jim Keller Ocean cove is second generation big.Little...

Intel is moving to 16/32 Cores big.Little architecture and unifying mobile and desktop together under Alder Lake / Meteor Lake over the next few years. This is PCIe 4.0 to PCIe 5.0 and 10nm++ to 7nm+ on 16/32 Cores big.LITTLE

Jim Keller will get more famous once Ocean Cove technology gets in our hands. This is the technology that Jim Keller was working on at Intel.

Power efficiency, IPC increase, Bandwidth, Prices....all fixed with Meteor Lake.
Posted on Reply
#96
remixedcat
hat
Am I missing something? I'm not sure what microprocessor architecture, be it x86, ARM or whatever, has to do with data throughput speeds...
I think those are more related to storage than CPU??
Posted on Reply
#97
R0H1T
ToxicTaZ
Jim Keller will get more famous once Ocean Cove technology gets in our hands.
Or not :twitch:
ToxicTaZ
13th gen (Meteor Lake)
I bet 13th time is not the charm!
Posted on Reply
#98
ARF
remixedcat
I think those are more related to storage than CPU??
No, because the x86 CPUs lack instructions which they need for better storage transfers efficiency and optimisations.
This is why Sony PlayStation 5 has several additional hardware sets which enable those storage speeds of the "825 GB" SSD.

One of the chips is so fast that accelerates as if 9 Zen 2 cores ! :eek:


www.anandtech.com/show/15848/storage-matters-xbox-ps5-new-era-of-gaming/2
Posted on Reply
#99
londiste
ARF
One of the chips is so fast that accelerates as if 9 Zen 2 cores ! :eek:
You do realize this is an ASIC running hardware implementation of whichever compression they are using? It has little to do with the specific processor in the system.

Kraken compression range generally falls in the 2-7x range. basically there is a 3-4GB/s NVMe drive underneath and then there is a hardware compressor/decompressor on the machine side of things. I kind of wonder how long will it take for motherboard manufacturers and/or Intel/AMD to implement something similar if it proves useful to generic use cases. From the descriptions so far, both Sony's and Microsoft solutions seem to be heavily aimed towards texture compression.
Posted on Reply
#100
R0H1T
Sandforce used to do this on their controllers IIRC.

Suffice to say it isn't needed for most consumers, not in the least because file compression is often times limited by the compression algorithm you're using & isn't a one stop shop or solution to every file you throw at it!
Posted on Reply
Add your own comment