persingc wrote:I was always curious as to how much if any source code lsdroid and pcmhammer have in common.
Pcmhammer got a lot of code from my ls1-flash free which took me years to figure out. viewtopic.php?f=42&t=3111
I managed to get the write routines working with a little help from Dimented24x7 in my private codebase, but my code wasn't properly multi-threaded and it'd crash if you held the window up with the mouse. I gave that codebase (windows code, and working asm kernels one for read, one for write) to NSFW after he stuck firm to wanting the code for pcmhammer to be open source (I was worried it'd be stolen, but came around to his line of thinking) and with his better windows development skills and my working code base we re-implemented it as a proper GPL OO windows app. He also wanted the kernel to C so that got re-written as one kernel too.
The result of this, is that I totally understand where Tazzi is coming from. Hes not a big company, not going to make serious money, and hes keeping it affordable for the kids in the backyard with their junkyard cars on newer pcms. Thats a huge win, I dont think many understand the skills and the time it takes to do original work. Unfortunately I think we are past the era of 1 person can write an app in their spare time for free. This seems to be a workable solution going forward to the more complex generation of newer cars. I for one am interested to see how it goes. And it absolutely does not mean other people cant write their own things, open source or not, free or not. The more people doing things the better.
persingc wrote:I was always curious as to how much if any source code lsdroid and pcmhammer have in common.
Pcmhammer got a lot of code from my ls1-flash free which took me years to figure out. viewtopic.php?f=42&t=3111
I managed to get the write routines working with a little help from Dimented24x7 in my private codebase, but my code wasn't properly multi-threaded and it'd crash if you held the window up with the mouse. I gave that codebase (windows code, and working asm kernels one for read, one for write) to NSFW after he stuck firm to wanting the code for pcmhammer to be open source (I was worried it'd be stolen, but came around to his line of thinking) and with his better windows development skills and my working code base we re-implemented it as a proper GPL OO windows app. He also wanted the kernel to C so that got re-written as one kernel too.
Ok. I thought it all advanced from your original app. I didn't realize that only pcmhammer came from it.
Nope different source, different languages, different kernels, different features, different people. Though we do speak to each other and help each other when one asks, but we all agree that our things can co-exist and its an interesting journey seeing how other people do it. And we'll help anyone else who asks a specific question to help their development and doesn't just want a code handout they can rip off.
All Parameters in the Testing xdf have been validated. There could still be errors and if you find any please let me know. I'm giving it some time before I merge everything.
Please don't PM me with technical questions - start a thread instead, and send me a link to it. That way I can answer in public, and help other people who have the same question. Thanks!
NSFW wrote:Holy crap that's a lot of parameters. Nice work!
Thank you. I was really surprised by all the etc stuff. A few errors in the low oil protection area have been pointed out to me which I'll fix probably when I merge everything.
Just to keep track of things I find as I comb through this
"C6302 - MAP Sensor Offset" needs "signed" checked so that it can see negative values. Verified my bin files values were correct after selecting the check box.
"B5945 - Cold Engine Noise Reduction Timing" has the row axis inverted. Column is RPM x 11 and Row is Cyl/Airmass x 15.
"B5946 - Cold engine Noise Reduction Multipler" scaler is wrong as well as LSB does not need to be checked. Multiplier tables are generally X/4096
"B0104 - Cylinder Volume" scaler is X/32.767 rather than 32.768. Such a small value it probably won't make much difference. But VE% is calculated from this so having it as accurate as possible would be good.
