[dzenno]

Originally Posted by Joshboody

I’m curious about the Cobb boost… the way Cobb “deflected” was a little suspicious and if it’s just a log conversion supposedly then it should be fixed by now. I bet it has to do with load targets and if this is the case that means all load based tables above X rpm are not accurate.

Someone with Cobb 2, log both MAPs, load, and torque.

Josh, why not ask Cobb directly and see what their response is? We surely can't give you an answer here

[Skydive]

As I indicated some weeks ago in another of these "Kill Vishnu" threads, reasonably competent flash tuners can manipulate sensors and the like, just as may be done with piggybacks. In fact, it's fairly common. Of the four leading Audi flash tuners in North America, four of them do it. Rescaled MAF sensor output is equally common in high boost applications as well. (APR, GIAC, Revo, and AMS all did it on Audis, VeeDubs and Porsches in 2001 or so.) It's not a "trick" and it's not *necessarily* sleight of hand, but it is an indicator that a least one of our MSD8x flash tuners have employed the very same kinds of signal and data conditioning that some of you folks are so fond of attributing to Shiv as deception.

The fact is that in order to "fix" the data out issue that comes from the MAP rescaling, (i.e. bias), they will have to calculate the appropriate value for presentation, and I'm willing to bet that it won't be spot on when compared to an outboard sensor of disparate calibration right away either.

As to Dzenno's comment... Why are you getting so fucking defensive? My questions were completely fair, and entirely logical. That's why you "knew" I would ask them. You're just sore because you don't like to be reminded that your car doesn't run the way you like if factory knock sensors are allowed to work the way they do in other cars. (Yeah... Revo and GIAC were doing similar things ten years ago. It's a means to an end, and it's not always a nefarious indicator of anything other than differing philosophy.)

Believe it or not, I give a shit. If you guys are right, I want to know it... but NOTHING I've seen here or in my own data is particularly indicative of the problems you describe in the current Procede implementation with which I have tested. (I have a 2.5, and didn't do any appreciable logging before it was updated, because at the time I was working my own tuning with OLS. It wasn't an issue until I decided to work with the PWM kit.) In the meantime sniveling, half-information attempts to assassinate Shiv based on one dimensional data is frustrating to me. Procede and Cobb are what I have, and the Procede is what I believe to be the right tool for the job *I* want to do. If you want to use a different tool to achieve a different set of goals, that's all well and good. Please refrain from clouding issues that are not related to your own.

Take a deep breath.

[OpenFlash]

Originally Posted by dzenno

Because your respected tuner chooses that instead of shutting us up with factual data, logs...sadly it'll be like this everytime it seems

This is where I get a little fed up Dzenno. You can be a fountain of help to people on this forum. Yet when I'm concerned, you choose to be a fountain of misinformation. At one time, i thought it was simply due to your lack of understanding with respect to technical/tuning matters. But I've seen that your motives aren't nearly as innocent.

So here is your data that you found it necessary to demand. Demand under the guise of caring about refinement. Below are two logs taken earlier today on SIHK335's FBO 6AT on the way to visit Sappersix to pick up the wallet I left in his car this morning SIHK335's car is running the new aggressive maps in his Rev2.5. He makes ~400whp on 91oct and 438whp on 91+meth (dyno results posted a couple weeks ago). To satisfy you, the AFR logs are scaled with a min/max of 10:1 and 20:1.

Map 1 (91 oct, no meth):


Map 4 (91oct+PWM Meth):


And just for reference, here is the Cobb log that boom posted on the previous page that you, interestingly enough, have yet to express ANY concern/comments about:


Is there anything else that you would like? Perhaps you'd like another video of innovate AFR gauge display during a highway pull in his car? Let's just put it all out on the table while we are at it. Because you may hide behind the excuse that I'm a better "typer" than you. When truth be told, I just know a hell of a lot more than you. As do many people on this forum. Which is why we find it possible to express ourselfs with profanity and ill-conceived arguments that tend to morph suddenly into something reminiscent of a magic picture. Stare at it long enough and you can see all kinds of things. All of them fake images. But hey, they captivates kids who tend to blur the lines between reality and fantasy.

Just stop being such a jerk.

Shiv

[jason335777]

Originally Posted by dzenno

Because your respected tuner chooses that instead of shutting us up with factual data, logs...sadly it'll be like this everytime it seems

No offense dude, but I'm sure Shiv know a lot more than you about tuning the 335. I have followed many heated threads (the CPS issue comes to mind) and its clear who the expert is. I dont know what you do for a living, but its probably not tuning 335s all day. I think Shiv sometimes gets a little too hostile, but nonetheless he knows what he's talkng about.

[themyst]

Skydive:

As I previously stated, as long as secondary logging tools verify a flash tune's output, there is no issue. e.g. mechanical boost gauge for boost, external wideband for AFR, etc. Nothing wrong with manipulating sensors to get desired results as long as the logging output is accurate.

It is no secret piggybacks bias the widebands to add fuel. If the latest firmware on rev 2.5/3 corrected the incorrect AFR calculation, awesome. dzenno has other issues with the Procede fuel "control" due to the fact he has a manual, but the AFR bias latency is minimally impacting for an automatic as we do not lift off throttle during a WOT shift.

Our beef with the AFR calculation output on Procede rev.2 is the fact reported AFR was completely different from actual AFR.

edit- since our friendly neighborhood tuner chimed in, here is a log of my car through a 2-3-4 pull. CAN Knock = AFR calculation in this instance. 18 AFRs post-shift going into 3rd? Surely this can't be correct... or is it?

Shiv - Pleast post some logs of customer cars with the same logging parameters through a multiple gear pull like you used to do with StartupJunkie's car. A nice 2-3-4 would do it. Thank you.

[OpenFlash]

Originally Posted by themyst

Skydive:

As I previously stated, as long as secondary logging tools verify a flash tune's output, there is no issue. e.g. mechanical boost gauge for boost, external wideband for AFR, etc. Nothing wrong with manipulating sensors to get desired results as long as the logging output is accurate.

It is no secret piggybacks bias the widebands to add fuel. If the latest firmware on rev 2.5/3 corrected the incorrect AFR calculation, awesome. dzenno has other issues with the Procede fuel "control" due to the fact he has a manual, but the AFR bias latency is minimally impacting for an automatic as we do not lift off throttle during a WOT shift.

Our beef with the AFR calculation output on Procede rev.2 is the fact reported AFR was completely different from actual AFR.

edit- since our friendly neighborhood tuner chimed in, here is a log of my car through a 2-3-4 pull. CAN Knock = AFR calculation in this instance. 18 AFRs post-shift going into 3rd? Surely this can't be correct... or is it?

Shiv - Pleast post some logs of customer cars with the same logging parameters through a multiple gear pull like you used to do with StartupJunkie's car. A nice 2-3-4 would do it. Thank you.

Here are 3 logs from SIHK335's 6AT. Same ones I posted up a couple weeks ago. In the future, please don't mistake me for your waiter. If you want info, try doing a search:

[themyst]

Originally Posted by shiv@vishnu

Here are 3 logs from SIHK335's 6AT. Same ones I posted up a couple weeks ago. In the future, please don't mistake me for your waiter. If you want info, try doing a search:

Hi Shiv, thanks for the data, much appreciated. Are you in a bad mood today? You could have simply linked me to this thread! BTW, have you ever logged this same car on pump gas? I am genuinely curious. Thank you.

[dzenno]

Shiv thanks for the logs. Let's discuss...in your procede log I'm actually surprised it looks worse than before. There's a HUGE lean spike to 15AFR a whopping 1000rpms into the pull, in BOTH logs and that's a pull starting at 4000rpm

And its laughable you're pointing out boom's COBB log as I think you're referring to AFR at 2700rpm before even full boost hit...

I have nothing to sell, my only motive here is and always was to make the community realize that there's something in procede's fuel control which could take refinement, if possible...but your thing remains there's nothing there when all logs including your own show CLEAR leanruns in midrange 1000ROM's into the pull

Skydive, get some logs and we'll talk...until then sayonara

[dzenno]

Shiv, even your second set of procede logs show 14AFR at full boost and high timing post shift, scale them better (10-16 instead of 10-30 and everyone will easily see)...this default AFR scaling really sucks to a point it's unclear...how did I miss to comment on those when they were first posted is beyond me, I failed myself lol

[OpenFlash]

Originally Posted by dzenno

Shiv thanks for the logs. Let's discuss...in your procede log I'm actually surprised it looks worse than before. There's a HUGE lean spike to 15AFR a whopping 1000rpms into the pull, in BOTH logs and that's a pull starting at 4000rpm

And its laughable you're pointing out boom's COBB log as I think you're referring to AFR at 2700rpm before even full boost hit...

Originally Posted by dzenno

Shiv, even your second set of procede logs show 14AFR at full boost and high timing post shift, scale them better (10-16 instead of 10-30 and everyone will easily see)...this default AFR scaling really sucks to a point it's unclear...how did I miss to comment on those when they were first posted is beyond me, I failed myself lol

Keep looking at the logs. It will hit you soon. Also, I suggest trying less to desperately win an argument. And focus instead on learning.

shiv

[dzenno]

Originally Posted by shiv@vishnu

Keep looking at the logs. It will hit you soon

shiv

Lol ok, right on

[themyst]

Originally Posted by shiv@vishnu

Keep looking at the logs. It will hit you soon

shiv

It's like watching paint dry. Trying to read a two point variance on those logs is giving me a migraine. But I do see lean spikes at WOT onset. Nothing catastrophically dangerous assuming the AFR output is accurate, but still can't compare to what a flash tune can do.

[dzenno]

Originally Posted by shiv@vishnu

Keep looking at the logs. It will hit you soon. Also, I suggest trying less to desperately win an argument. And focus instead on learning.

shiv

I'm learning every day, really not hard to look at logs and see spikes where there shouldn't be any lol it's easy when you have nothing to sell or defend...I've never really met anyone who could argue more than you when you yourself post clear examples of the issue in those logs confirming further what I've been harping about since forever...sad it's STILL not resolved and you coose to debate it instead of just accept it and say you've been working on ironing this out for a while now...I have emails from May talking back and forth starting in about February this year when you started working out this issue and were saying you had to find ways to work with short term fuel trims which you at that time added logging for to assist you in understanding WHY this issue happens...did you tell anyone ever why you added those channels or should I just copy the email conversation in here? Man what did you make this harmless thread into...in my first post all I said was:

Great numbers, have Shiv dial in those low/midrange AFR a bit

Wow!

[maui86]

Shiv, is there a delay with the afr logging?

[OpenFlash]

Originally Posted by maui86

Shiv, is there a delay with the afr logging?

There is always a small delay. Around 1/10th of a second due to sampling rate of CAN data. The more important thing to be aware of is that you can't reasonably expect to achieve your AFR target the moment full boost is achieved. This has more to do with the combustion dynamics during sudden increases in engine load. Which is why you often see a brief, but clear, spike to 14-15:1 once full boost is achieved. Only then ramping down to the AFR target. This is illustrated on both the Procede log I posted as well as the Cobb log that boom posted. Unfortunately, there are a couple people in this thread who are selective with what they see

[dzenno]

That delay is same and is there for all the CAN channels procede or Cobb if your question was to attempt to understand if it's possible the AFR log is just delayed while all others are not then that is not the case...just sayin'

[OpenFlash]

Originally Posted by dzenno

That delay is same and is there for all the CAN channels procede or Cobb if your question was to attempt to understand if it's possible the AFR log is just delayed while all others are not then that is not the case...just sayin'

Sorry, but you are incorrect again. The logging delay will depend on the CAN sampling frequency. In the case of the Procede which samples CAN data at 10hz, the delay will be approx 1/10th of second. I don't know exactly what Cobb logs sample at but I suspect they are around 30-40Hz which would make the delay 3-4 times shorter.

Shiv

[dzenno]

Shiv do a procede log starting at 2000rpm and post it that's if you're willing to compare Cobb and procede and compare it to boom's log...

There's nothing there we're missing out on in the Cobb log...AFR is 13 and going richer with no wild leanrun 1000rpms later by the time full boost hits even as low as 2700rpm...

Your procede pulls start at 4k rpm again, deflection at it's finest...congrats

[dzenno]

Originally Posted by shiv@vishnu

Sorry, but you are incorrect again. The logging delay will depend on the CAN sampling frequency. In the case of the Procede which samples CAN data at 10hz, the delay will be approx 1/10th of second. I don't know exactly what Cobb logs sample at but I suspect they are around 30-40Hz which would make the delay 3-4 times shorter.

Shiv

I'm "wrong" because you "suspect" Cobb logs are 3-4 times faster than procede's? Lol no doubt you're right because you're so sure of everything it seems even when you suspect it

I can tell you the frequency of the Cobb AP logs, it logs a sample every 0.04-0.08seconds depending on the number of channels selected for logging...so 12-25 samples/sec so 12-25Hz not 30-40, so guess who's wrong, both of us lol

[OpenFlash]

Originally Posted by dzenno

Shiv do a procede log starting at 2000rpm and post it that's if you're willing to compare Cobb and procede and compare it to boom's log...

There's nothing there we're missing out on in the Cobb log...AFR is 13 and going richer with no wild leanrun 1000rpms later by the time full boost hits even as low as 2700rpm...

Your procede pulls start at 4k rpm again, deflection at it's finest...congrats

Ok... just happened to have exactly what you wanted. Below is a simplified, easy to follow, 4th gear log on the same FBO car. Run started at 2000rpm. Running 17-18psi with meth. The bottom two grid lines represent 13:1 and 10:1. And actual AFR is slightly richer than what is displayed. I'll be happy to demonstrate on an Innovate AFR gauge if you demand it



Shiv

[Brey335i]

Originally Posted by Joshboody

I’m curious about the Cobb boost… the way Cobb “deflected” was a little suspicious and if it’s just a log conversion supposedly then it should be fixed by now. I bet it has to do with load targets and if this is the case that means all load based tables above X rpm are not accurate.

Someone with Cobb 2, log both MAPs, load, and torque.

I have a mechanical boost gauge. I can't tell that anything is wrong.

I'm not perfect, and I still have a lot to learn about tuning, but I should be more than qualified to observe a boost gauge go up and down.

There might be something to what you're saying about load based tables and all that, but as long as my boost levels are within a safe range then I think my motor will be fine. If you have evidence to prove otherwise please share.

Sorry about OT, let's get back to talking about OP's post. Unless you guys want to continue Debating on the Internet. I too have a Ph.D. in this department

[dzenno]

Lol you're scaling AFR 10-40 now?? Hahaha rescale it 10-16 please if you'd like us to see anything useful there