2016 OVTune Tacoma 3.5L Manual Transmission / Engine ECU Reflash

The link says Tremec T65 but that means zero to me cause I don’t know anything about auto mechanics. Sounds cool though. Not sure if it’s an option for 2wd only but since MT requires 4wd and has for some time it seems reasonable to expect if they say Tacoma and MT that it is actually compatible.

 

quick google search shows it as some sort of GM or Ford transmission

 

yeah it’s an older corvette/camaro transmission, so a no go for the 4x4 guys :/

a 4x4 with the T56 would be crazy haha. I wonder if it’s possible to rebuild the current trans and just change the ratios?

 

Hey, Does anyone keep track of their long term fuel trims on their truck? Mine seem to be more negative than I would expect (like -9% and -5% when im just cruising at 2500rpm) and bank 1 is always about ~4% more negative than bank 2. Just curious is anyone has any insight or some data to compare too, haven't been able to find much info elsewhere figured someone here might have some info.

 

So, just wondering as a n00b that if OVT changes these settings, then what does the factory do that is different than what OVT does? Taken to the logical extreme, why can't OVT set the ECU to the "final" best settings and then the only factory changes needed would be for hardware interface changes such as replacing a sensor that has different upper/lower limits, etc...? Obv I don't fully understand what OVT does that is different than the factory TSB changes.

 

Think of it this way -

the ecu / TSB (from Toyota) is the set of algorithms that are used to calculate things

the tune is the table (values) that are used by the algorithms

the ecu is 1000s of lines of code, with many interdependencies. You can’t just copy paste from one to another

updating to the latest TSB, gives you the best set of algorithms with various bug fixes or optimizations. Then you run the tune which gives you the best set of values within the tables.

Basically by going to the latest TS you are getting the best base code from Toyota to then run the tune.

Hopefully that makes sense

 

Well said. I was stumped about how to say anything close to that.


The one thing to add is that it was said the Toyota code is all in machine language. OVT had to back it all out into something usable as a basis.

 

Yeah it’s all in assembly which is low level machine code. It’s all written for the specific processor to tell it what to do.

https://en.wikipedia.org/wiki/Assembly_language

couple other posts for reference

 

So, I have 26 years in software development, mostly back-end (no js/css/ui stuff).
Software is a fickle friend, it can go off the rails for various reasons and edge cases. I probably don't even need to write that, most laypeople understand this to some degree?
You have various options when mucking with someone else's stuff. If you have full source code you can tinker directly but you better know exactly what you're doing and/or test it well. For embedded systems you run the risk of exceeding available memory/storage both with code and data. I assume OVT does not have access to Toyota source on this stuff and does not wish to directly rewrite logic.

Software may also be designed to allow 3rd party implementations to customize/tune/adjust it with configuration data which can be directly manipulated (often this config data lives in a database, sometimes the app itself even has ways of editing the config data). This setup describes the product I work on right now. I assume Toyota does not intentionally design for this, though they likely have set things up reasonably well for their own team to make adjustments and by doing so they have opened the door for OVT.

You can also reverse engineer which means taking machine code and turning it into human readable (to some degree) form. This is easier if debugging information is not stripped out (the symbols for the variable names) so that the code has more meaningful names than it would for a purely reverse-engineered production product without debugging info. Either way, it can be read and understood with significant time investment and some way to simulate the changes.

I believe the primary tinkering done with OVT is to tabular configuration data? I'm not sure on that and I honestly have no idea either. But by manipulation of this configuration information, the code from Toyota is unchanged but the net effect of code + config produces a revised behavior. That's a safer path than trying to rewrite or refine the logic itself because changing the logic carries higher risk.

Now, say Toyota releases an updated version, they might change the name of a variable or more importantly its location or size in the compiled binaries. The compiler and linker are important parts of any software system. Of course I'm assuming these are compiled in some manner, I also have no visibility into that. So you no longer can apply the same old trick as the last version to get your changes in place. Furthermore, if a newer version of base code adds some new configuration capability, you'd likely want to consider taking advantage of that. And most important, if something which used to be possible has changed (split into multiple pieces or just removed entirely) you definitely want to account for that and at a minimum not attempt to overwrite it with changes which worked on older versions.

Bottom line, it's all about minimizing risk. When making changes to someone else's product you want to start with their vanilla setup and then tweak _only_ what you intend to tweak and nothing else. To have such a "patch" or "change" for multiple starting points is quite the challenge. My specific company pays a very hefty price for customizing our solution for our clients vs the ones who retain a vanilla configuration and our license fees reflect that.

NOTE: There are many alternative options including rewriting some or all of the code, provided it's possible to load it. Some of those options carry tremendous risk, but in the absence of legal concerns it might be fun to tinker with a fully homebrewed solution. Sometimes starting with a clean slate is actually the best way forward, many legacy systems are just a total nightmare to work with. Taking this path, though, incurs reinventing some wheels and is not for the meek. There's also a massive investment of time and energy into dev & testing before you can get anything off the ground.

For MT specifically, I'm scratching my head at what is so hard to get right about ETC? I have no experience with this but it seems a linear response to input would work fine. I'm guessing there's a ton of stuff to consider, mostly around safety. Still, MT feels like a different animal than AT in this regard and it sure would be nice to actually have actual control of the throttle without needing software tweaks to match expectations.

 

I think you touched on a couple of self-imposed limitations there. OVT noted that he will not touch parts that affect the TRAC / ABS systems.

There would be nothing that I'd like more than a complete and simple as possible rewrite of all the trash in the Engine (and, where applicable, Transmission) code. I imagine getting all of that to align properly with the other ECUs would be its own nightmare. But with OVTs limitation, I am very much okay with what we get, and now with vF, what we get to mess with.

 

Okay, I think I understand. I'm actually a coder by trade, just shy of four decades experience doing low level assembly and embedded code and reverse engineering of competitor data files to create conversion programs; currently work for a gov't agency interfacing backend databases with frontend web systems, UI/CSS/JS/RSS emphasizing security to prevent hacking, multiple certs, etc. So yes, I get the coding, I simply didn't understand what OVT is doing or how my truck ECU is laid out. I'm assuming it's a custom ASIC with memory-mapped areas for data tables and code execution, limited to a certain size (the aforementioned 4kB). So I understand that if Toyota increases the code and relocates a table then OVT has to change a lot of stuff around. I just didn't know if they were rewriting the actual code or not, but the quoted post from OVT above answers that.

So thanks for the great explanations, I understand much more than I did before! Perhaps sometime I'll share how I was fired from my first real job as a programmer at a local community college at 16yrs old for unexpectedly "hacking" a very large west-coast university on company time and escaped monetary repercussions because when they heard how young I was they didn't want it to get into the news, and the director wrote in my file "laid off due to lack of funds" and told me that he thought I have a promising career as a programmer and he hated to think I wouldn't be able to achieve that goal. Okay, guess I just shared it lol.

 

It's code + memory mapped area of the ROM exactly correct.
I wish it was 4kb. It's 1.5mb of pure machine code.

and we rewrite the logic (code, not tables) in hex. I write out the JARLS, Jmps, Mov, calculate the 80 FF, 07F 00, etc throughout.





QUESTION FOR MT : Regarding anti-stall and 1rst gear DFCO. I need a general consensus on what RPM you want it to trigger back on.
- Also these tables will be included for editing in vF Tuner if you want to go really old school and actually let it stall out.

 

What ever you used for 2nd through 6th has been stellar, IHMO. 650, I think? I haven't had any problem using that for over a year.

So far, Ultimate feels like a tune that doesn't need edits, but you're saying "hold my beer, huh?"

 

I want whatever fuckery happens at ~1500 rpm to happen at 1000 rpm. I think most would agree.

 

1st gear is well below 1,500 already. Sort of splitting the difference between those two.

 

Ok well whatever happens in 2-6 at 1500 rpm when Im coasting to a stop sign and it lets go lol

 

That's weird. The 2.0 MT tunes should all have the DFCO patch in 2nd through 6th with it lowered slightly in 1st.

I'm hardly the best source there because the Test Tune DFCO settings stuck around on my ECU after I flashed back to 1.05 FOB. I was checking DFCO pretty thoroughly in the lower gears yesterday on Ultimate.

 

Any chance of me seeing any of these new fancy developments on my TT tune or am I SOL? Lol

 

I know what you're talking about, you're not crazy lol

Not sure if it happens on my TT tune but for sure on the n/a map

 

Sez the guy with twins hiding under the hood.