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

Editorial Jim Keller on Moore's Law, Microprocessors, and Designing Chips from Scratch

Raevenlord

News Editor
Joined
Aug 12, 2016
Messages
3,755 (1.23/day)
Location
Portugal
System Name The Ryzening
Processor AMD Ryzen 9 5900X
Motherboard MSI X570 MAG TOMAHAWK
Cooling Lian Li Galahad 360mm AIO
Memory 32 GB G.Skill Trident Z F4-3733 (4x 8 GB)
Video Card(s) Gigabyte RTX 3070 Ti
Storage Boot: Transcend MTE220S 2TB, Kintson A2000 1TB, Seagate Firewolf Pro 14 TB
Display(s) Acer Nitro VG270UP (1440p 144 Hz IPS)
Case Lian Li O11DX Dynamic White
Audio Device(s) iFi Audio Zen DAC
Power Supply Seasonic Focus+ 750 W
Mouse Cooler Master Masterkeys Lite L
Keyboard Cooler Master Masterkeys Lite L
Software Windows 10 x64
Jim Keller on Lex Fridman's AI Podcast shed some light on his thoughts on the microprocessor design fundamentals as he sees them. In a hour-and-a-half-long interview, he approaches Moore's Law and its much lauded - and ubiquitously repeated - death, as well as the need for both iterative and zero-point microprocessor design requirements.

Mr. Keller approaches the usual microprocessor design loop, where a company develops a new design from scratch and then looks at the most fundamental way of adding performance. Usually, he says, easy 10% performance increments can be found by simply looking at a design and increasing execution units - increase a buffer here, increase a cache over there, put in another add processor on this part of the pipeline. However, he also speaks of how this process in itself is limiting, inasmuch as doing this often will eventually guide processor designs towards a bottleneck and the diminishing returns problem, where any more additions made to the design don't seem to increase performance - mostly just adding complexity, area and power requirements, and generally convoluting a given design.





Mr. Keller says - and this should go without saying - that that is the point where teams should see that the design in itself is already limiting - and that that's when you do something from scratch. And for Mr. Keller, fast advances in performance in the world of microprocessors depend on 3-5 year cycles of building a new processing design from scratch. His work in the microprocessor world usually comes in at this point - Mr. Keller has established himself as one of the foremost leaders of microprocessor design teams when looking at a clean-slate design, with his stints at AMD (collaborating on their K7, K8, and K12 designs alongside the HyperTransport bus), Apple (for their A4 and A5 microprocessors, which are now in the iterative phase), Tesla (with their HW 3.0 designs for autonomous driving) and AMD's Zen.



Of course, then comes Intel, which Keller himself describes as having a microprocessor development mindset that's closer to a 10-year sustained designed rather than the 3-5 year development schedule for a new architecture he favors. Interestingly, in the podcast, Jim Keller approaches this microarchitecture mindset on Intel from a short-term and long-term disaster perspective. According to him, repeating and refining a recipe (like Intel did many years with their Core architecture [author's side-note]) is the most efficient way to go about it: incrementally improving a design, saving money and taking a low-risk approach to processor development, albeit threatened by the diminishing returns equation we mentioned earlier.



The problem, according to Mr. Keller, is that managing quarter to quarter means that there is fear in hitting a short-term disaster with a rewrite from scratch; companies thus look to "milk" every ounce of profit from a previous design by incrementally improving it. However, this primes companies to hit a long-term disaster, much like we see today with Intel (it's not a disaster for a multi-billion dollar company like Intel, but you get the point): its architecture, which didn't go through a from-scratch design phase for years, was superseded by AMD's new Zen design and its iterations.

According to Mr. Keller, managers that don't look only quarter to quarter but have a longer outlook know that a short-term disaster - a new architecture design not panning out with all the costs that entails - is better in the long-term outlook than the other way around. This is what AMD looked to do after its Bulldozer architecture and derivatives - a company that was in financial dire straits saw that the only way forward was to start another design from scratch, instead of keeping on shoveling itself towards the ground with more years funneled into a dying microprocessor design that couldn't compete with Intel's previous or current ones. As to this, it's likely we'll be seeing the fruits of Mr. Keller's stint at Intel in the near future, though for now, it's still somewhat nebulous as to what designs will actually surface from his more than two-year tenure at the blue giant.



Closing up this article, which doesn't aim to be a summation of the AI Podcast, but aims to highlight some interesting tidbits present there, are Mr. Keller's thoughts on team management for a technological project. According to him, he sees the existence of abstraction layers in microprocessor design teams. Where a team of 10 humans works well together, and a team of up to 100 people may be able to function properly under a single supervision, any more than that and teams have to be divided, with organizational boundaries having to be set - and here too appears the diminishing returns equation. Jim Keller says some very interesting things regarding this, as in, that humans in general aren't getting smarter - so there is a fundamental limit to how much "processing power" you can have in a team working on a set project, considering team and size and communication capacity caps that derive from the fact that we are, well, humans.

And of course, you can have the most brilliant people in their given field working in a single design - but sometimes those people won't work well together. Adding to that, microprocessor development can sometimes lead toward instances where functions and transistor parts from where performance increments are being etched into a given product design change, and some people in the team aren't needed anymore, because their particular field of expertise has reached a diminishing returns plateau that is no longer pursuable.

TL;DR: Jim Keller is clearly an extremely accomplished microprocessor designer, but also a project leader, and has very clear ideas regarding the industry and his field of work. You should read the entire article and then move onto the podcast.

View at TechPowerUp Main Site
 
Joined
May 30, 2015
Messages
1,942 (0.56/day)
Location
Seattle, WA
He's also said before that ground-up redesigns are mostly a thing of the past. A 'ground-up' clean slate design today is comprised mostly of existing logic blocks rearranged to support or be supported by new logic. I.E. Bulldozer's FlexFPU in Zen.
 
Joined
Feb 18, 2005
Messages
5,847 (0.81/day)
Location
Ikenai borderline!
System Name Firelance.
Processor Threadripper 3960X
Motherboard ROG Strix TRX40-E Gaming
Cooling IceGem 360 + 6x Arctic Cooling P12
Memory 8x 16GB Patriot Viper DDR4-3200 CL16
Video Card(s) MSI GeForce RTX 4060 Ti Ventus 2X OC
Storage 2TB WD SN850X (boot), 4TB Crucial P3 (data)
Display(s) 3x AOC Q32E2N (32" 2560x1440 75Hz)
Case Enthoo Pro II Server Edition (Closed Panel) + 6 fans
Power Supply Fractal Design Ion+ 2 Platinum 760W
Mouse Logitech G602
Keyboard Razer Pro Type Ultra
Software Windows 10 Professional x64
While Keller is correct that Intel has spent too long beating a dead horse, it's also true that the 10nm disaster has not helped them in that regard. In essence, the 10nm disaster has caused a CPU disaster.
 
Joined
Nov 4, 2017
Messages
14 (0.01/day)
I'd give my whole PC, my PS4, my smartphone, smartwatch, half of my bank account if that guy would agree to have two or three beers with me. What a legend! All the best to this guy, been watching he's moves since forever, not a single mistake.
 
Joined
Aug 30, 2006
Messages
7,223 (1.08/day)
System Name ICE-QUAD // ICE-CRUNCH
Processor Q6600 // 2x Xeon 5472
Memory 2GB DDR // 8GB FB-DIMM
Video Card(s) HD3850-AGP // FireGL 3400
Display(s) 2 x Samsung 204Ts = 3200x1200
Audio Device(s) Audigy 2
Software Windows Server 2003 R2 as a Workstation now migrated to W10 with regrets.
I'd give my whole PC, my PS4, my smartphone, smartwatch, half of my bank account if that guy would agree to have two or three beers with me. What a legend! All the best to this guy, been watching he's moves since forever, not a single mistake.
That’s quite an accolade!
 
Joined
Nov 4, 2005
Messages
12,013 (1.72/day)
System Name Compy 386
Processor 7800X3D
Motherboard Asus
Cooling Air for now.....
Memory 64 GB DDR5 6400Mhz
Video Card(s) 7900XTX 310 Merc
Storage Samsung 990 2TB, 2 SP 2TB SSDs, 24TB Enterprise drives
Display(s) 55" Samsung 4K HDR
Audio Device(s) ATI HDMI
Mouse Logitech MX518
Keyboard Razer
Software A lot.
Benchmark Scores Its fast. Enough.
When will AI be smart enough to develop its own chip that performs better than a human designed chip?
 
Joined
Jun 3, 2010
Messages
2,540 (0.48/day)
I would enjoy it, had Jim Keller conceptualized microarchitecture design using the key word "Amdahl's Law" within its framework. He is the best candidate to tell it to integer based brains such as myself.
 
Last edited:

bug

Joined
May 22, 2015
Messages
13,843 (3.95/day)
Processor Intel i5-12600k
Motherboard Asus H670 TUF
Cooling Arctic Freezer 34
Memory 2x16GB DDR4 3600 G.Skill Ripjaws V
Video Card(s) EVGA GTX 1060 SC
Storage 500GB Samsung 970 EVO, 500GB Samsung 850 EVO, 1TB Crucial MX300 and 2TB Crucial MX500
Display(s) Dell U3219Q + HP ZR24w
Case Raijintek Thetis
Audio Device(s) Audioquest Dragonfly Red :D
Power Supply Seasonic 620W M12
Mouse Logitech G502 Proteus Core
Keyboard G.Skill KM780R
Software Arch Linux + Win10
He's also said before that ground-up redesigns are mostly a thing of the past. A 'ground-up' clean slate design today is comprised mostly of existing logic blocks rearranged to support or be supported by new logic. I.E. Bulldozer's FlexFPU in Zen.
Well yes, x86 has seen how many implementations to date? Further designs are bound to take a lesson or two (or even basic blocks) from the previous ones, so in that sense, you can't really call them "clean slate". You have to read that term in the context.
 
Joined
Dec 26, 2006
Messages
3,862 (0.59/day)
Location
Northern Ontario Canada
Processor Ryzen 5700x
Motherboard Gigabyte X570S Aero G R1.1 BiosF5g
Cooling Noctua NH-C12P SE14 w/ NF-A15 HS-PWM Fan 1500rpm
Memory Micron DDR4-3200 2x32GB D.S. D.R. (CT2K32G4DFD832A)
Video Card(s) AMD RX 6800 - Asus Tuf
Storage Kingston KC3000 1TB & 2TB & 4TB Corsair MP600 Pro LPX
Display(s) LG 27UL550-W (27" 4k)
Case Be Quiet Pure Base 600 (no window)
Audio Device(s) Realtek ALC1220-VB
Power Supply SuperFlower Leadex V Gold Pro 850W ATX Ver2.52
Mouse Mionix Naos Pro
Keyboard Corsair Strafe with browns
Software W10 22H2 Pro x64
A little constructive feedback, perhaps make the source link a bit bigger, bolder with a bit more contrast.

Took me a bit to find it.

Cheers
 
Joined
Oct 22, 2014
Messages
14,170 (3.82/day)
Location
Sunshine Coast
System Name H7 Flow 2024
Processor AMD 5800X3D
Motherboard Asus X570 Tough Gaming
Cooling Custom liquid
Memory 32 GB DDR4
Video Card(s) Intel ARC A750
Storage Crucial P5 Plus 2TB.
Display(s) AOC 24" Freesync 1m.s. 75Hz
Mouse Lenovo
Keyboard Eweadn Mechanical
Software W11 Pro 64 bit
Joined
Mar 10, 2010
Messages
11,878 (2.20/day)
Location
Manchester uk
System Name RyzenGtEvo/ Asus strix scar II
Processor Amd R5 5900X/ Intel 8750H
Motherboard Crosshair hero8 impact/Asus
Cooling 360EK extreme rad+ 360$EK slim all push, cpu ek suprim Gpu full cover all EK
Memory Corsair Vengeance Rgb pro 3600cas14 16Gb in four sticks./16Gb/16GB
Video Card(s) Powercolour RX7900XT Reference/Rtx 2060
Storage Silicon power 2TB nvme/8Tb external/1Tb samsung Evo nvme 2Tb sata ssd/1Tb nvme
Display(s) Samsung UAE28"850R 4k freesync.dell shiter
Case Lianli 011 dynamic/strix scar2
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup, corsair void pro headset
Power Supply corsair 1200Hxi/Asus stock
Mouse Roccat Kova/ Logitech G wireless
Keyboard Roccat Aimo 120
VR HMD Oculus rift
Software Win 10 Pro
Benchmark Scores 8726 vega 3dmark timespy/ laptop Timespy 6506
Am I the only one who just got half way through the interview on YouTube then realised.

I've seen this before, a good while ago.
Worth a second listen tbf.
 
Joined
Feb 23, 2008
Messages
1,064 (0.17/day)
Location
Montreal
System Name Aryzen / Sairikiki / Tesseract
Processor 5800x / i7 920@3.73 / 5800x
Motherboard Steel Legend B450M / GB EX58-UDP4 / Steel Legend B550M
Cooling Mugen 5 / Pure Rock / Glacier One 240
Memory Corsair Something 16 / Corsair Something 12 / G.Skill 32
Video Card(s) AMD 6800XT / AMD 6750XT / Sapphire 7800XT
Storage Way too many drives...
Display(s) LG 332GP850-B / Sony w800b / Sony X90J
Case EVOLV X / Carbide 540 / Carbide 280x
Audio Device(s) SB ZxR + GSP 500 / board / Denon X1700h + ELAC Uni-Fi 2 + Senn 6XX
Power Supply Seasonic PRIME GX-750 / Corsair HX750 / Seasonic Focus PX-650
Mouse G700 / none / G602
Keyboard G910
Software w11 64
Benchmark Scores I don't play benchmarks...
Joined
Nov 18, 2010
Messages
7,595 (1.48/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 41
I posted this vid somewhere here months ago.
 
Top