Guru3D.com
  • HOME
  • NEWS
    • Channels
    • Archive
  • DOWNLOADS
    • New Downloads
    • Categories
    • Archive
  • GAME REVIEWS
  • ARTICLES
    • Rig of the Month
    • Join ROTM
    • PC Buyers Guide
    • Guru3D VGA Charts
    • Editorials
    • Dated content
  • HARDWARE REVIEWS
    • Videocards
    • Processors
    • Audio
    • Motherboards
    • Memory and Flash
    • SSD Storage
    • Chassis
    • Media Players
    • Power Supply
    • Laptop and Mobile
    • Smartphone
    • Networking
    • Keyboard Mouse
    • Cooling
    • Search articles
    • Knowledgebase
    • More Categories
  • FORUMS
  • NEWSLETTER
  • CONTACT

New Reviews
Corsair H170i Elite Capellix XT review
Forspoken: PC performance graphics benchmarks
ASRock Z790 Taichi review
The Callisto Protocol: PC graphics benchmarks
G.Skill TridentZ 5 RGB 6800 MHz CL34 DDR5 review
Be Quiet! Dark Power 13 - 1000W PSU Review
Palit GeForce RTX 4080 GamingPRO OC review
Core i9 13900K DDR5 7200 MHz (+memory scaling) review
Seasonic Prime Titanium TX-1300 (1300W PSU) review
F1 2022: PC graphics performance benchmark review

New Downloads
FurMark Download v1.33.0.0
Intel ARC graphics Driver Download Version: 31.0.101.4091
Corsair Utility Engine Download (iCUE) Download v4.33.138
CPU-Z download v2.04
AMD Radeon Software Adrenalin 23.1.2 (RX 7900) download
GeForce 528.24 WHQL driver download
Display Driver Uninstaller Download version 18.0.6.0
Download Intel network driver package 27.8
ReShade download v5.6.0
Media Player Classic - Home Cinema v2.0.0 Download


New Forum Topics
Grab for free: Dishonored: Death of the Outider at Epic Games Store Forspoken Benchmark Test & Performance Analysis Review AMD Confirms Strategy of Restraining Chip Supply to Maintain High CPU and GPU Prices Corsair Introduces a Bevy of New Component Products for PC Builders Razer Introduces the Viper Mini Signature Edition Magnesium Alloy Gaming Mouse Amernime Zone AMD Software: Adrenalin / Pro Driver - Release Discovery 22.12.2 WHQL AMD Announces Pricing and Availability for Ryzen 7000X3D Series Processors RTX 4090 Owner's thread RTX 4070 Ti Owner's thread DirectStorage testing reveals that PCIe 3 SSDs are as fast as PCIe 5 SSDs, PCIe 4 SSDs almost similar




Guru3D.com » News » Tesla HW 4.0 Chips baked with TSMC at 7nm have a Monolithic design

Tesla HW 4.0 Chips baked with TSMC at 7nm have a Monolithic design

by Hilbert Hagedoorn on: 08/20/2020 08:24 AM | source: chinatimes | 16 comment(s)
Tesla HW 4.0 Chips baked with TSMC at 7nm have a Monolithic design

If you own or desire a Tesla, then you know these cars have autopilot and are getting closer and closer to fully autonomous driving. To support that latest feature, a new hardware update is pending. The new chips will be significant, massive, and classified as HPC chips.

Tesla will only be able to recover 25 chips per 12-inch wafer from its HW 4.0 design, which does not include yield corrections. The new chip designed in collaboration with Broadcom would be fabbed at 7nm. So that means these chips are bulky and even monolothic in size (which is prone to affect yields bigtime if you ask me). To paint a more clear picture, current HW3 3.0 chips are 20x13mm and can fit 217 on one 14nm wafer. So on 7nm HW3.0 would fit 434 chips compared top the 25 for HW 4.0. The new HPC chips would steer Tesla into autonomous driving. Due to the size of these chips, we also have some doubts about power consumption. Current HW3.0 takes roughly 200 Watts of power while driving. 

Broadcom and TSMC would have targeted the first production by the end of this year. Two thousand wafers, which are suitable for 50,000 chips. Then it would be scaled up in next year with mass production in late 2021.

 



Tesla HW 4.0 Chips baked with TSMC at 7nm have a Monolithic design




« LIAN LI Launches New Fan Interlocking System with the UNI FAN SL120 · Tesla HW 4.0 Chips baked with TSMC at 7nm have a Monolithic design · NVIDIA Announces Financial Results for Second Quarter Fiscal 2021 »

4 pages 1 2 3 4


Maddness
Senior Member



Posts: 2195
Joined: 2014-10-24

#5820010 Posted on: 08/20/2020 08:28 AM
Whilst I'd personally love a Tesla, not really comfortable with an autopilot. I realize it is the future, but I get simple pleasure from just driving my own automobile.

Andy Watson
Senior Member



Posts: 300
Joined: 2017-09-22

#5820039 Posted on: 08/20/2020 11:08 AM
I assume that all current Teslas will have to have these retrofitted to get full self driving for people who paid for that functionality? That's quite a few cars as Tesla is selling 100k per quarter and hopes to expand.

wavetrex
Senior Member



Posts: 1696
Joined: 2008-07-16

#5820058 Posted on: 08/20/2020 01:09 PM
I assume that all current Teslas will have to have these retrofitted
My guess is it will be quite simple to swap in the new card, just like upgrading a GPU.
A few screws, a few wires, done.

Not everyone will do it, like @Maddness , most probably won't care and are okay with just having the safety features (lane keeping, automatic braking, collision avoidance).

The overall cost for Tesla for the swap will be tiny, compared to the massive profits that a fully autonomous army of Robo-Taxis can bring, assuming they manage do achieve Level 5 autonomy.

D1stRU3T0R
Senior Member



Posts: 643
Joined: 2017-08-16

#5820081 Posted on: 08/20/2020 02:22 PM
Is it really worth the struggle to make a new chip? Why don't just create a Custom AMD cheap, since it would be a big win win for both. Tesla would make it they own Epyc/Threadripper which is much more OP than this, like the chinese did, and AMD would ....win.....

Denial
Senior Member



Posts: 14010
Joined: 2004-05-16

#5820082 Posted on: 08/20/2020 02:25 PM
Is it really worth the struggle to make a new chip? Why don't just create a Custom AMD cheap, since it would be a big win win for both. Tesla would make it they own Epyc/Threadripper which is much more OP than this, like the chinese did, and AMD would ....win.....


Because an ASIC designed for Tesla's home grown self driving stack will easily outperform any GPGPU/CPU AMD could make at a fraction of the cost per unit.

There is a reason why ever large company doing AI work is going vertical.

4 pages 1 2 3 4


Post New Comment
Click here to post a comment for this news story on the message forum.


Guru3D.com © 2023