Hacker Newsnew | past | comments | ask | show | jobs | submit | ylk1's commentslogin

Most of the production engines still use tumble flow combustion

For reference: Nissan e-power with recuperation ~50%: https://www.nissan-global.com/EN/INNOVATION/TECHNOLOGY/ARCHI...

Mahle ~45% with pre-chamber: https://www.mobilityengineeringtech.com/component/content/ar...

Toyota dynamic force engines ~40% : https://www.thedrive.com/tech/18919/toyota-develops-worlds-m...

Toyota's newest engines claim even more: https://global.toyota/en/newsroom/corporate/40850156.html

Honda hybrid ~40%: https://hondanews.com/en-US/honda-automobiles/releases/relea...

Startup Carnot engines claim 70% efficiency: https://carnotengines.com/technology/


>Mahle ~45% with pre-chamber

That looks deceptively simple and counterintuitive. Mentioned Honda CVCC injects fresh air/fuel mixture into pre chamber. Mahle passive doesnt at all. How does it accomplish detonation in the first place? How does the fresh air/fuel get into the chamber thru those tiny nozzle openings? there must be something they arent showing like additional channel leading close to intake valve.


What’s the catch with the Carnot engine? If if’s real then surely this would be a global news worthy breakthrough?


> Carnot is targeting the hardest to abate sectors including [long haul] marine [transportation], heavy-duty vehicles and primary off grid power.

Even if great advances were being made, these are not very interesting sectors to the vast majority of people and the industries have massive inertia. Presumably Carnot's therefore unlikely to be grabbing headlines.


But such an improvement in efficiency would surely reduce fuel costs, which matter a lot to consumers, though maybe not in the US where it’s cheap.


The geometric compression limit for F1 is 18:1. They make it sound like they can go farther.


Mercedes crossed 50% thermal efficiency in 2018 with their F1 engine: https://www.roadandtrack.com/motorsports/a15049580/mercedes-... Their recent interviews suggest they are up by 2 to 3% on that

Honda did too but they never marketed the explicit number.


Ferrari's partner MAHLE has pre-chamber ignition productized for OEMs: https://www.mahle-powertrain.com/en/experience/mahle-jet-ign...

I think, Maserati MC20 has this tech.


The only manufacturer to release any details about their powertrain technology.


They've detailed their work on engine/MGU-K/MGU-H/ESS here: https://global.honda/en/tech/motorsports/Formula-1/


UMmmmm, actualy there are no details, it's more like bait, layed out in front of the rabbit hole. What it does outline/hint at is the methodology of there thinking about engine development, within the arbitrary rules of F1. And what they, and everybody else leaves out, is that limitations in material sciences is the reason for the engineering heroics, and that access to say, the single crystal materials used for jet turbine hot section blades, would allow higher combustion temps and pressures, and greater efficiency, in NA engine, that would outperform anything built yet. The reason that F1 cant do that is money, and compared to true airospace development,F1 has lint in there pockets, and have falken back onto 1940's turbocompound aircraft engine technology.


what are you even talking about? A lot of stuff is clearly laid out on how they achieved better efficiency.

You can get more power out of NA with better materials you've stated but not efficiency.

F1 wanted more road relevancy and hence ended up with turbo-charging and Fuel limits


zero mechanical details, nothing verifiable is zero information, plus there is a long venerable tradition of lying like hell whenever you talk about competitive advantages :) efficiency goes with materials, as a higher temperature and shock loading, will permit higher compression ratios, and that is as far as I know the only way to exract energy from a piston engine well unless the piston part is just viewed as a gas generator, but whoa betsy, the rabbit hole is looming before us, and having listened to folks involved in engine development (including F1), I am just a spectator,engines are hard, and these days anyone who could do ICE, is looking at other stuff, or retirement ceramics have come close, but dont have the reiability when produced at the scale needed for mass market


Engines are hard, but Honda's problem always was the gearbox, which is much harder. And it's not simulatable.

But I like Honda. They have the very best company leader.



New accelerator in upcoming chips.


Nice read!


Can you try micro bench marking Apple Cores? Nobody seems to be doing that, yet people claim they are the best in business.


I don't have any Apple cores easily available, but the code [1] is open if anyone wants to try it (I don't know how POSIX-y the iOS compile environment is, though).

One caveat is that just because you don't find a performance difference, doesn't mean the optimization isn't happening. It could simply be the case that write throughput is not the limiter, but rather the latency * occupancy product is the limiter. E.g., if it takes 50 ns to go from L2 to RAM, and there are only 10 buffers available to hold these requests, then the maximum bandwidth is 64 bytes / 50 nanos * 10 buffers = 12.8 GB/s regardless of the maximum possible bandwidth of each component.

An eliminated store may still take this full latency (since it still has to read the value from RAM), so even if all writes are eliminated, the performance may remain at 12.8 GB/s - but you would save power and memory and L3 bandwidth for other cores... but the optimization would be tough to detect by looking at performance alone. You'd need to look at performance counters (does those exist for iOS devices?).

[1] https://github.com/travisdowns/zero-fill-bench


He was an actual Microarchitect in his early career. This guy clearly has good knowledge on every aspect of microprocessor design: transistor, interconnect, process complexity, core layout & utilization, ease of design, circuit complexity, time to market, micro architectural behaviors, ISA & software patterns. From what I've seen he is really good at setting targets for the teams, especially given that he has been at multiple companies, he clearly knows what is good and what is lacking. Setting aggressive targets and motivating the team towards them. Something the previous management failed at.


Great article as usual! Nice analysis. btw, just wondering, were you a processor architect in your previous roles?


No, my training and background is on the software side and so the hardware stuff is only a hobby.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: