Quo Vadis rusefi?

Post Reply
stefanst
contributor
contributor
Posts: 703
Joined: Wed Feb 17, 2016 12:24 am
Location: USA 08530

Quo Vadis rusefi?

Post by stefanst »

Or, in 1980s CEO speak: "What is our current mission statement?"
It seems we have moved past the original idea of providing control for experimental and crackpot engines.
One goal that seems reasonable is to see rusefi spread as far as it can. Spreading open-source goodness across the planet is a worthy goal.

I'm asking this loaded question because we all try to help improve the product constantly. Usually we just think of features we would like to see and assume that others may like these features as well. And then our core developers decide if this is something they want to tackle right away, definitely later, maybe later, maybe not at all, but if an independent developer wants to take a shot at it, do it.

Most of us can probably think of so many possible features to keep a 10 person team of soft- and hardware developers each busy for a few years. But that's not realistic and it would likely lead to a super-bloated product that misses the mark.

In order to better evaluate what features are important, we need to figure out where we want to go.

One key question seems to be: What is/are the target market(s)?
a few things we should likely dismiss right out of hand are:
- Current luxury or high performance vehicles with very steep price-tags
- Super-oddball types of engines like hit-and-miss engines https://en.wikipedia.org/wiki/Hit-and-miss_engine, turbines and the likes
- Highly safety-critical applications like helicopter engines/turbines, ocean-going vessels

Some probably more realistic markets to target:
- Conversions of most sizes carburetor engines, be it in a lawnmower, or an old V12 from the 60s.
- Low(ish)-budget race cars
- At least in the US: Pre-OBDII injected car engines that are not subject to emissions
- Fun projects involving engines

What other worthwhile targets are out there? What should be the primary focus,
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

A big part of why I've decided to create my own system was how full of magic and too-black-boxy other things seemed.

I know that at the moment rusEfi is not much more usable than other systems and probably even less user-friendly - but the intention was to have easy setup and easy operation. I know that this requires documentation which we currently have very little - but at least some things like engine sniffer and formulas tabs are direct consequences of that original idea.

I hope that with relatively small effort things could be simplified a lot, I just do not know how exactly :( But I would love to implement ideas in this area.

Also understandable firmware code which users do not care for :)
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
stefanst
contributor
contributor
Posts: 703
Joined: Wed Feb 17, 2016 12:24 am
Location: USA 08530

Re: Quo Vadis rusefi?

Post by stefanst »

russian wrote:[...]I know that this requires documentation which we currently have very little - but at least some things like engine sniffer and formulas tabs are direct consequences of that original idea.[...]
A big problem with documentation on a project like this is that whatever we write about the code and the hardware will be outdated by the time it's released to the wiki. For example, I think a few of us wouldn't mind posting some documentation on Tunerstudio screens, but maintaining that documentation will be a huge pain.

On the documentation:
There is also the issue that many people trying to implement rusefi, especially on an as yet unproven engine, don't have sufficient knowledge of exactly how engines work and how to tune them. So, they need to be taught all this by other forum users. Basically they are getting a training session organized by amateurs. Especially @ and @, but many others as well, spend a lot of time on the forum, educating people on engine function and operation which is not really rusefi specific.

Does anybody know of any decent online resources on how injected engines work? Some sort of compendium? Preferably something not part of the documentation of another aftermarket EFI system?
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: Quo Vadis rusefi?

Post by kb1gtt »

stefanst wrote:- At least in the US: Pre-OBDII injected car engines that are not subject to emissions
I thought I would comment on this. Not all states use OBDII as the official emissions. I understand in CT that every 2 years you have to get a tail pipe sniffing, and if you fail emissions levels you have fix it with a cap on spending of around $500. AKA emissions issues don't put your family into bankruptcy, but you are forced to fix things like failed O2 sensors. So I'm not sure how far from ODBII we should steer. There certainly are really handy diagnostics tools, and logging capabilities that I don't see any problems with doing. However I would steer clear of VIN and emissions OK bits. Basically do not spoof or participate in tom trickery, but OK to do diagnostics.

Also keep in mind that an old V8 70's gas guzzler has less of a carbon footprint than a new Prius or other hybrid. Many states recognize that keeping cars on the road longer is a better way to fight carbon emissions. Our new cars simply consume very large amounts of energy to produce. Many states have no emissions regulations. Also keep in mind companies like bully dog, and others flash OEM ECU's thousands of times, and appear to suffer from no legal issues. I under stand the fear of such things, but also keep in mind that people need to experiment to make things better. I personally feel see a huge feature of open source projects like that as a way to educate people. If people have honest information, they can start to make better future systems. You don't think that our rocket engineers didn't illegally blow something up in there back yards do you?

I just wanted to make quick comment. That's my 2 pent's, we welcome yours.

Any how, trying to find a market.... Are we talking software or hardware? I can see many desired paths to pursue. I would think if we wanted to make this the next great thing, we would want to identify a car like a Kia, and make a PNP system for some kind of car that will have a long life of low cost modifications. Either that or we could make a new approach on how to control an engine. I'm a fan of an optical bus which connects several modules together. Think about minimizing radio emissions issues, and the flexibility of picking your own modules to make what ever configuration you need.

At the end of the day, it really relies on what are the core developers motivated to get done. So perhaps the question should be, how do we get more developers interested and developing the next great thing.
Welcome to the friendlier side of internet crazy :)
Flying
Posts: 13
Joined: Fri Apr 21, 2017 8:35 am

Re: Quo Vadis rusefi?

Post by Flying »

rusEFI appeals to me because:
- I am cheap, but have lots of time to burn
- I require a flexible/adaptable system for my planned projects, this means no black boxes
- I'm not afraid of the rabbit hole

I first looked at Megasquirt, but wondered why a project that was brought up on open source-ness didn't have any way for me to construct the hardware outside of kits. Then I realized its the mature product of a past era.
Between all of the open source ECU projects out there, rusEFI seems to be the one with the most potential. It's has reasonably fast hardware, a handful of working cars, sequential injection!, TunerStudio, and a consistent development history.

I want a way to convert my carbureted motorcycle to FI, I want to tune my Accord, I want a cheap, open-source standalone engine management solution for aging engines, such as 4AGEs and M50s. From what I can see, this is all possible with rusEFI, it just needs a push towards organizing the information.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: Quo Vadis rusefi?

Post by kb1gtt »

I guess this holds true for our documentation :) A clean documentation must be a sign of a dirty project :)
http://www.gearboxmagazine.com/a-clean-garage-is-a-sign-of-a-dirty-mind/

One suggestion for the marketing stand point, cheap and low cost are two different things. I look for low cost which doesn't always mean cheap.
Welcome to the friendlier side of internet crazy :)
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

stefanst wrote:
russian wrote:[...]I know that this requires documentation which we currently have very little - but at least some things like engine sniffer and formulas tabs are direct consequences of that original idea.[...]
A big problem with documentation on a project like this is that whatever we write about the code and the hardware will be outdated by the time it's released to the wiki. For example, I think a few of us wouldn't mind posting some documentation on Tunerstudio screens, but maintaining that documentation will be a huge pain.
I do not agree with this view. I believe that as of right now the rate of change is pretty low, most of the basics are set and would not change much. And even these basics are not documented properly.

I believe now is good time to document at least current state of the affairs. Yes maintaining would be nasty, but mainteining is not the problem right now - lack of navigation, consistency and content is the current problem in my view.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

kb1gtt wrote:I would think if we wanted to make this the next great thing, we would want to identify a car like a Kia, and make a PNP system for some kind of car that will have a long life of low cost modifications...

it really relies on what are the core developers motivated to get done. So perhaps the question should be, how do we get more developers interested and developing the next great thing.
I belive there is a bit of a contradiction here - in order to have more contributores we need wider coverage. Look at the great work @ is doing with firmware flexibility - he has the need for flexibility and he helps with that, he has the need for UART and he helps with that.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

kb1gtt wrote:Any how, trying to find a market.... Are we talking software or hardware?
A combination on both?

But I would like to encourage custom hardware - I believe Frankenso should be a great reference platform while projects like e36 custom board would be the real way to spread the project.

I would also like to remind everyone that RUS in rusEfi stand for "reliable universal simple" - that's pretty much the direction (well, three opposite directions :D) to move in.

Unfortunatelly I only have this many hobby hours each week, and who knows for how long would I enjoy current number of hobby hours. Unfortunatelly the day job is not going away any day soon - I doubt there is a way to monetize rusEfi while keeping it open source. Google AdSense to become a real $ source we would need to increase forum views 100-fold.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Quo Vadis rusefi?

Post by AndreyB »

AndreyB wrote:
Mon May 08, 2017 7:34 pm
A big part of why I've decided to create my own system was how full of magic and too-black-boxy other things seemed.
2020 was the best year rusEFI had so far!

* we had a major documentation push while moving wiki to github https://github.com/rusefi/rusefi/wiki
* we've started making releases https://github.com/rusefi/rusefi/releases
* we now have https://github.com/rusefi/rusefi/blob/master/firmware/CHANGELOG.md
* we have new amazing https://rusefi.com/
* we've made a fuss in v12 community
* we had knock knocking and GDI idling, wbo is kind of WBOing
* I am sure I am forgetting a lot

Between rusEFI LLC and independent assemblers, between microRusEFI and Proteus I believe we had about a 100 rusEFI ECUs sold in 2020.

But.

But very few of these 100 units are driving on a regular basis, either that - or we simply do not know how many are drivers how successfully.

But rusEFI did not get fundamentally "smarter" in 2020 - it's still just an ECU which requires major human expertise to operate.
Both https://github.com/rusefi/rusefi/blob/master/misc/Strategy_2019.md and https://github.com/rusefi/rusefi/blob/master/misc/Strategy_2020.md were talking about getting more developers on board but the number of people contributing code is still extremely low. As of right now, there is still a lot of soldering before any code could be written and tested on a real vehicle. We were also not successful in establishing a stream of user feedback so far.

Personally I have very little interest in just selling 300 units in 2021. Being not satisfied with how things are going I plan to try something different in 2021:

* I would like to shift focus from universal board into PnP boards - after all, a significant percentage of vehicles on this planet share quite a small set of physical connectors, and Hellen-One looks to be the way to design many boards quickly
* more of these boards would have some kind of wifi connectivity on them
* some might have microphones - a log without audio is often not doing it
* many would physically click into OEM connectors but while _looking_ OEMish would be pretty raw in terms of functionality in the beginning. Those would still not be satisfactory end user products - those would still be tinkering platforms.


dev_strategy Slack channel has a pinned item with Feb 13, 2021 12:00 PM Eastern Time (US and Canada) first rusEFI conference call.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
Post Reply