EcuTek vs BRZ-Edit

[zc06_kisstherain]

I kinda miss Hondata's K-Pro capability.
if someone can develop a product like K-Pro, i will jump in it.

[Fast_Freddy]

Originally Posted by mad_sb

Tuner does not need the pro version. You can buy personal and take your laptop to the tuner and have them tune you car.

Self tuning is not for everyone, but for those of us it is for, BRZedit is the only option aside from a stand alone.

Is there a BRZedit tuning forum where users can share knowledge, base tunes, etc.? AEM has a pretty active forum for their electronics and I found it very helpful when learning to tune the FIC. I'm very interested in anything other than ECUTek and their Apple/Microsoft approach. I don't mind paying someone for a good tune but it really bothers me that after buying the tune I can't see it, tweak it or even know the extent of the changes. While for some it's enough to hear that their AFR, ignition and cam timing have been "optimized" and that they won't get any CELs because certain unspecified DTCs have been disabled but no way am I flashing my ECU with that until I know a lot more about what they changed.

[ft_sjo]

EcuTek exists to protect the tuner's IP, which is fair enough. You don't own a tuner's calibration, you're just licensed to use it.

If you don't like that then you have to tune it yourself.

[mad_sb]

Originally Posted by Fast_Freddy

Is there a BRZedit tuning forum where users can share knowledge, base tunes, etc.? .

Right here on this forum. So far i am the only one posting much info though as i am one of the early adopters....


DBW Info

Cam Tuning Info

Fuel Trims & Closed to Openloop transition

E85 HowTo

I have posted enough info that anyone who has tuned a car before with just about any platform should be able to jump right in and get going. I have been taking the time to do this to show this community tuning is not something you need to scared of and not something only an elite few can accomplish in the hopes it would help get the grass roots tuners moving. I think once we have open source support the end user tuning community will take off. I'm really looking forward to EcuFlash etc because it really pulls in the assembly geeks that love to write patches and define maps for fun

[regal]

Originally Posted by mad_sb

Right here on this forum. So far i am the only one posting much info though as i am one of the early adopters....


DBW Info

Cam Tuning Info

Fuel Trims & Closed to Openloop transition

E85 HowTo

I have posted enough info that anyone who has tuned a car before with just about any platform should be able to jump right in and get going. I have been taking the time to do this to show this community tuning is not something you need to scared of and not something only an elite few can accomplish in the hopes it would help get the grass roots tuners moving. I think once we have open source support the end user tuning community will take off. I'm really looking forward to EcuFlash etc because it really pulls in the assembly geeks that love to write patches and define maps for fun

Correct a guy who grew up tuning carburators, but doesn't all this require expensive dyno time? Or could one temporarily install a wideband o2 monitor and develop a tune out on back country roads ?

[arghx7]

Originally Posted by Dynotronics1

only if you read line by line in the code

It has always been, and always will be, that someone will be able to see, on some level, that the code has changed. Question is, can it be done at the dealer, or even the factory level, and is it worth the cost and time to do so?

Short answer is, no.

Having used the commercial-grade OEM tools (Caldesk, INCA, etc) and certain in-house tools I can say that I can EASILY see if anyone has changed something from the production calibration.

First, there can be a software build version into the code that has been flashed onto the ECU. So for example, if the manufacturer changes the spark table from a 10x10 to 16x16, a new software version will be released. That's what moves around a bunch of the memory addresses so that some reverse engineering would need to be re-done on newer model years. Then there are checksum addresses that can be matched to the library of production calibrations. No match = tampering. Now maybe ECUtek basically leaves those in such a state that it hides the change. I'm not an expert on that.

All that's academic and has very little bearing on ECUtek vs BRZedit. But I can tell you right now that Toyota and Subaru have a handful of guys outside of Japan who have access to the tools and information needed to determine tampering. I can't speak for every dealer tool though. From my experience with using dealer tools, they are designed to be idiot proof. You plug into the customer car and it will just update it to whatever the library in the tool specifies. Those libraries can change all the time because the dealer tools require regular updates from the internet. I presume Techstream is like that.

The way it works in at the OEM's is that the actual software code (Matlab Simulink models etc) is kept under very tight security. By that I mean, very few people know exactly how everything is calculated, and often they are not the ones doing the tuning hands on. The tools needed to access the ECU, the memory address files (ASAP2 .A2L format usually), and the actual ROM binary have a little wider availability. It's typically on a need-to-know basis though.

[wparsons]

Originally Posted by jamesm

Still, it beats ecutek's bend-over licensing scheme by a mile. The idea that I can pay a tuner and hourly rate, to tune my specific vehicle on a dyno in person (a bespoke tune, not canned and resold) then somehow not own the product of that work is absurd. If I pay for labor, I own the fruit of it. If I have two BRZ's with the same mods, I should be able to flash MY map to both for the cost of producing the single map. But I can't, because with ecutek you own nothing, they're just letting you use it under their terms.

That's not 100% true... what you pay EcuTek for is the cable and license for the ECU (one per car/ecu). Your tuner could happily generate another ROM file for you to run on a second/third/whatever car. EcuTek doesn't control the ROM file at all, that's up to the tuner. So yes, you need another license but you wouldn't have to pay full price for a second tune.

That said, if you're getting a custom tune, you *should* have the second car custom tuned as well since no two cars are going to behave identically.

Originally Posted by ft_sjo

EcuTek exists to protect the tuner's IP, which is fair enough. You don't own a tuner's calibration, you're just licensed to use it.

If you don't like that then you have to tune it yourself.

Kinda this, I'd say you own the calibration, but it's an encrypted version of it so you can't modify it in any way. You can flash between it and other tunes freely though if you have a cable.

Originally Posted by arghx7

Having used the commercial-grade OEM tools (Caldesk, INCA, etc) and certain in-house tools I can say that I can EASILY see if anyone has changed something from the production calibration.

First, there can be a software build version into the code that has been flashed onto the ECU. So for example, if the manufacturer changes the spark table from a 10x10 to 16x16, a new software version will be released. That's what moves around a bunch of the memory addresses so that some reverse engineering would need to be re-done on newer model years. Then there are checksum addresses that can be matched to the library of production calibrations. No match = tampering. Now maybe ECUtek basically leaves those in such a state that it hides the change. I'm not an expert on that.

All that's academic and has very little bearing on ECUtek vs BRZedit. But I can tell you right now that Toyota and Subaru have a handful of guys outside of Japan who have access to the tools and information needed to determine tampering. I can't speak for every dealer tool though. From my experience with using dealer tools, they are designed to be idiot proof. You plug into the customer car and it will just update it to whatever the library in the tool specifies. Those libraries can change all the time because the dealer tools require regular updates from the internet. I presume Techstream is like that.

The way it works in at the OEM's is that the actual software code (Matlab Simulink models etc) is kept under very tight security. By that I mean, very few people know exactly how everything is calculated, and often they are not the ones doing the tuning hands on. The tools needed to access the ECU, the memory address files (ASAP2 .A2L format usually), and the actual ROM binary have a little wider availability. It's typically on a need-to-know basis though.

What if you flash back to the OEM ROM, can anything be detected as tampering then? I remember reading about the flash count not being updated by the EcuTek flashing process so you can flash between tunes and stock as much as you want but the flash count wouldn't be effected. I've also read that some tuners will update the flash count as needed in case the ecu was flashed by a dealership.

[arghx7]

Originally Posted by wparsons

What if you flash back to the OEM ROM, can anything be detected as tampering then? I remember reading about the flash count not being updated by the EcuTek flashing process so you can flash between tunes and stock as much as you want but the flash count wouldn't be effected. I've also read that some tuners will update the flash count as needed in case the ecu was flashed by a dealership.

I haven't used every development tool or dealership tool for every manufacturer, not by any means. What I can say is this.

If there's an accessible "flash count" on an ECU, I've never seen one. I'm not saying they don't exist, but it's not something you'd ever look at unless you were interested in the hardware more than anything. It's not part of the day-to-day process of using commercial tuning tools on production ECU's. If I want to see whether someone has changed values in the ECU, I can use the built-in comparison function for the tuning tool. It's a pretty simple affair: compare the calibration in the vehicle to the production calibration that's supposed to be in it.

If somebody wants to claim warranty on a dead ECU, and the ECU goes back to the ECU manufacturer, and the manufacturer sees that the ECU has been flashed 500 times, then maybe somebody would mention that in a report or something. Most likely it would end up in the warranty claim guy's Excel spreadsheet used for tracking such things.

Now if there's already a TSB saying x customers are doing y modifications and breaking things, well that might be a different story. The TSB could conceivably involve a procedure for checking to see whether the ECU has been reprogrammed. Think about it this way. If the customer is modifying the programming in the ECU, then breaking things as a direct result of that, and then trying to stick the OEM with the bill, well the OEM has a vested interest in preventing that. Of course determining fault and root causes isn't always straightforward.

[regal]

Originally Posted by wparsons

What if you flash back to the OEM ROM, can anything be detected as tampering then? I remember reading about the flash count not being updated by the EcuTek flashing process so you can flash between tunes and stock as much as you want but the flash count wouldn't be effected. I've also read that some tuners will update the flash count as needed in case the ecu was flashed by a dealership.

Based on the info posted in this thread unless your dealer has an IT forensics team as long as your flash count is less than your dealer visits it would be very unlikely to have a warranty denied. I think a good tune should prevent warranty work.

Really tuning a modern car like this isn't so much about gaining HP on WOT pulls but fine tuning driveability, doing as much to eliminate knock and pre-ignition as possible, custom individual (non-mass market) MAF scaling. Basically things that make the engine run better rather than meet minute emission requirements.

For example I know this car uses the cam timing to suck exhaust back into the combustion chamber (modern EGR), basically there are sacrifices the OEM makes for regulations and for mass production, these are things the consumer should be able to fine tune.


With Scion USA burying their heads in the sand and not releasing the latest OEM flash for the transient ignition retard issues which may help with the DI issues, it really makes custom tuning something I want to do sooner rather than later.

I read a few scary posts with graphs that showed how much knock the stock tune gives (not just after shifting either), the OEM tune relies way too much on the knock sensor to pull timing after the fact, its sloppy. A good tune doesn't rely on a knock sensor, that's just there for as insurance. But what do the OEM's care? They only need the engine to last to 60k miles.

[arghx7]

Originally Posted by regal

Based on the info posted in this thread unless your dealer has an IT forensics team as long as your flash count is less than your dealer visits it would be very unlikely to have a warranty denied. I think a good tune should prevent warranty work.

Again, I'd hesitate to call myself an expert but I've never seen a "flash count" on a dealership tool or OEM tuning tool. Does somebody have any proof that this even exists in anything other than the ECU manufacturer's internal tools? Got a screenshot of a "flash count" in Techstream? I'm trying to understand where this idea is coming from.

For example I know this car uses the cam timing to suck exhaust back into the combustion chamber (modern EGR), basically there are sacrifices the OEM makes for regulations and for mass production, these are things the consumer should be able to fine tune.

It's called dethrottling. I know this sounds strange, but it works on the same basic principle as cylinder deactivation: open the throttle more so that your pumping work decreases. In cylinder deactivation, it's accomplished by shutting off half the cylinders. With charge dilution like EGR, you are adding in extra air to account for the additional unrburnable mixture. This requires the throttle to be opened further, reducing pumping work. Any engine with variable cam phasing does it. Do an AVCS sweep at a given low-load point, and you will see brake specific fuel consumption go down with more phasing.

With Scion USA burying their heads in the sand and not releasing the latest OEM flash for the transient ignition retard issues which may help with the DI issues, it really makes custom tuning something I want to do sooner rather than later.

It's probably sitting in some dude's Outlook Inbox right now, or making its way through three departments' tracking spreadsheets. Maybe it's on the agenda for the next insufferable meeting between two comparable departments in Toyota and Subaru. The guy in charge of that could be on sick leave. The reason behind it is probably more mundane than someone proverbially "burying their heads in the sand."

I read a few scary posts with graphs that showed how much knock the stock tune gives (not just after shifting either), the OEM tune relies way too much on the knock sensor to pull timing after the fact, its sloppy.

For a completely stock car people say they want the best fuel economy possible. That means running as much timing advance as possible and trusting the knock control system. There's nothing unusual about that.

A good tune doesn't rely on a knock sensor, that's just there for as insurance. But what do the OEM's care? They only need the engine to last to 60k miles.

If the ECU pulls too much timing due to poor tuning of the knock control, the cat will overheat. Then the vehicle will fail the long term in-use emissions testing and the OEM could get fined. Durability requirements are getting stricter and stricter due to CARB regulations.

[regal]

Thanks Argh for the detail. Still think working on these things will help squeeze out more long term reliability out of this engine rather than relying on the OEM flashes. I'll gladly give up 1-2 mpg to put off pulling the engine to do any work on it. The way the boxer is crammed in so low and tight to the side rails, its going to be a tough one to work on.

Its funny you mention that if the engine pulls too much timing due to knock correction it will destroy the cats. We have several documented cases that this happens more than we would like with the stock tune, especially with high oil temps. Add that to the heavy "dethrottling" and I wouldn't be surprised if there are already folks with ceramic particles making their way into the engine. I know Toyota had the same issue with their last sports car and got away with it, quite a few MR-S owners lost compression shortly after the warranty ended with cat ceramic breaking down. MsgMike's tear down showed broken up cats.

Things like this are why I want a solid custom tune, Toyota only owes me 60k miles, I want 300k miles.

[mad_sb]

Originally Posted by regal

Correct a guy who grew up tuning carburators, but doesn't all this require expensive dyno time? Or could one temporarily install a wideband o2 monitor and develop a tune out on back country roads ?

You don't need to install a wide band the factory front 02 sensor is a wide range sensor. It is accurate from 12.1:1 up to 20:1... for NA tuning you don't need richer than 12.5:1.

So, all you need is a tuning solution and a virtual dyno app and a safe (closed) road. A dyno is easier and safe and gives quicker feedback but road tuning is still good to make sure the tune is dialed in for loading differences between the street and dyno.

[jamesm]

Originally Posted by ft_sjo

EcuTek exists to protect the tuner's IP, which is fair enough. You don't own a tuner's calibration, you're just licensed to use it.

If you don't like that then you have to tune it yourself.

the problem is that they don't give you that option. so, in 5-10 years when my annointed tuner is no longer around and i need to adjust something, i get to pay all over again or start from scratch with an open source map. you can't just pull and edit the file, because you don't own it. this is one of the only markets i can think of where you can pay for labor and not own the fruit of it, and it has broad practical implications.

if software development terms, you are left with a critical external dependency on this map, with no maintenance guarantee. this is ok if the provider is microsoft and has the money to support customers of a failed effort for years. these guys aren't microsoft. this is exactly why linux won the server war, people naturally fear locked-down external dependencies, for good reason.

now our cars run on one lol.

[crook]

Originally Posted by regal

The only way I could see BRZ-edit becoming a success if a major FI kit came with a tune, that would save $500 off a packaged EcuTek tune. So far none have surfaced. Imagine if the "innovate" super charger had been developed with BRZ edit, it would have retailed for $3300!

Rev_works are doing/have done this, Jamesm could prob chime in here

And I am running BRZedit on my FI kit only because my tuner/the only tuner I trust doesn't have the 'legalities' to tune ecutek (which my FI kit came with) and suggested I go with BRZedit