2.00 Firmware Beta Testing

[admin]

   

If you are unfamiliar with the 2.00 Firmware announcement, please read "Version 2.00 Beta Firmware Update Ready"

 

Finished!  We're ready to have this sucker tested by you few and brave souls that throw all caution and sensibility to the wind.   

Just kidding     It is safe and preliminary testing has been done.  What we're looking for are any minor bugs that may be present as well as general feedback from a variety of situations, conditions, and uses to confirm that we haven't made anything worse with our efforts.

Some brief notes regarding the Beta Testing process:

• There are only three units available for beta testing.  Out of fairness, we're going with the first three people to indicate that they were interested in beta and these people are:
  • mvaldez (indicated August 09, 2009, 11:38:57 AM)
  • wdhitch (August 27, 2009, 07:43:57 PM)
  • wizard of ahs (September 08, 2009, 05:42:22 AM)
  • first alternate is intocoasters (October 14, 2009, 08:58:37 AM) should any of the above decline.
• If those people can please confirm they are ready, willing and able to voluntarily beta test the new code it would be appreciated.  Indicate whether or not you have a ScanGauge II (2 or 3 should have it ideally).  Thorough beta testing will require dedicated attention tests as well as normal driving situation tests.  If you can't commit to several hours of testing over the next few weeks then please decline and allow someone else to take your spot.
• In addition to "playing" with and ensuring normal operation in your daily commutes, I will also be devising specific tests that each person will be asked to perform.  It will be that person's responsibility to ensure they have a safe quiet stretch of road where if the worst should happen (which is the pedal being unexpectedly dead) it won't cause an accident!  Remember that no matter what happens, your brake works and will immediately disengage the cruise.
• For the beta testers, there is no need to send your existing unit to us to swap or reprogram.  We want you to have your existing unit both in case of a problem and you need to put it in but also so you can compare similar situations with both units.  We do expect that you will send one of the two units back when beta testing is finished of course.
• Beta testing participation is entirely voluntary.  There is no compensation of any kind implied or offered.  You do, obviously, have everyone's thanks!

If you're one of the mentioned beta testers and have any questions or comments or wish to accept or decline the invitation, please do so here.
 
Others, please feel free to post here if you have any specific suggestions, questions, or comments regarding the actual beta test efforts.  For general discussion related to the 2.00 firmware please begin here to find the appropriate thread.  Thank you.
 
Art.

[wdhitch]

Ready to go!!! And I have a scan guage 2.

[wizard of ahs]

Art,

Would LOVE the opportunity, but work takes me away from home for a while.  I do have the scan guage but must decline at this time.

Please send to Intercoasters 

Good luck to all you beta testers  !!! 

[mvaldes]

Sounds great, I'm ready to go and do have a Scangauge II!

[intocoasters]

Thank you Wizard.... I am sorry that you will not be able to test.

I accept the post of replacement tester.

[admin]

Perfect.  I'll be sending messages to each of you shortly.  Thank you.

[wdhitch]

My little package arrived too!!!!! 

[mvaldes]

Still patiently waiting for mine, NOT!

[mvaldes]

Murphy's Law mostly won out. I had today off and hoped the mail would have my new unit. It did, but the mail carrier was a substitute so it didn't arrive unit after 4 pm, bummer. Anyway, here's what I found in the short time I had to play:

Cruise: Pretty much the same on level ground at local speeds. AWFUL at speeds under 20 mph with any sort of uphill. The steeper the hill the more you feel like you're on a bucking bronco! I tried swapping back to my original unit and repeated the test on the same hill. No surging at all. Put the new unit back and had the same bucking bronco ride.

Cruise speed recall: works just as described.

Pedal map: works just fine. With our little engine, I really didn't see much unitl I got up to "5 and above." The econo mode definitely follows the graph.  A short trial around my neighborhood leads me to believe that I don't have the patience to use it. I'll have to try it in "stop and go" traffic and see if it makes sense there.

This was just an initial run, roughly 1 1/2 hours of playing around doing all the "programming" with the Scangauge to save time. (I will try the switches, just knew I wouldn't get much done in the time available if I had to go to the back and flip switches.)

So far, unless you've got some "tweaking" ideas for the surging at low speeds, I way prefer my original unit.

[admin]

I moved some of the posts to this topic.  This is where we should be discussing the beta testing (the other thread should be specifically for programming via SG discussions).

mvaldes... did you change any of the settings at all prior to your slow speed tests?  I'm just a little surprised to read that because I did quite a bit of testing on a relatively steep hill right outside our location doing about 20 mph.  I thought that I got it very close to how it was before and even better on flats at low speeds.  I'd also like to know if you were in D or manual, and if the "bucking bronco" is due to accelerator pedal action or if it is due to shifting up and down.  If you aren't sure and you were in D, then try again in manual mode in one gear.

Regardless... obviously, 20 mph on steep hills is - or probably should be - a relatively rare situation that won't affect many people at all.  So please continue to test with more normal cruising situations before committing to an opinion such as "I way prefer my original unit". 

[mvaldes]

Art,
I plugged in the new unit and headed out my driveway in "D". Cruised at about 37 for a few miles and then tried some of the hills in the neighborhood.I started up the hill, still in "D" and hit "on" at about 18 mph. I was at about 4400 rpms and the car held the speed but lurched and bucked it's way up the hill. There was no change of gear until I crested the hill and it upshifted. I don't think I've ever gone up that hill using cruise before, so I didn't know how the original unit would have fared. So, in the interest of testing and giving you some data to compare "apples with apples,"  I pulled over to the side of the road and shut down the car. I removed the new unit, put in the original and drove around the block to repeat the experiment on the smae hill (still in "D.") The original unit, on the same hill held 18 mph smoothly in one gear and upshifted after cresting the hill. I stopped, shut down and removed the original unit and re-installed the new one. After trying all the pedal maps, I tried the hill experiment on another steep, but less steep hill. (The pedal map this time was at "5".) I went up and down a bunch of times and had the bucking whenever the speed was set between 18 and 20 mph. At, or above 20, the ride was smooth. Could your testing have been done at 20 or above?
(I tested it below that because there's one particular school zone where I drive every couple of weeks where the speed limit is 15 and I use the cc there at 15 - 18 mph. Come to think of it, the school is on a hill, but it's nowhere near as steep as the one I was testing on today. BTW, that stretch constitutes about 10 minutes of my driving every 2 weeks so it's hardly a frequent use of the unit, but it did prompt me to test at the low end of the speed range.)

Perhaps my choice of words was poor and, if you thought I was in any way belittling your efforts on the new software, I apologize. Remember that this is the first time I've used the unit and I'm trying to actually beta test it by trying it in different settings, some of which may have little application to daily driving conditions. It did fine on the "main" local road at 35- 40 mph with flats and some fairly gentle hills. I still have a lot more different conditions to try and will use it on the interstates tomorrow. What I was trying to say in my original note is that is that I really prefer the way the old software handled low speeds on the steep hills.

If you prefer, I can confine my testing efforts to just my regular commute and "about town" driving which would give a more realistic picture of the conditions encountered in daily driving. I seem to recall you saying something in an earlier post about correcting tiny "overshoots" that most people would never notice except perhaps under extreme conditions so I may have misunderstood my role and I'm trying it out under conditions that are inappropriate to the data you would like to gather from this testing.

Please advise,
Michael

[admin]

There's no inappropriate tests.  Some are going to hold more weight than others, but its all good data.  I certainly took no offence.  Only asking that you (and readers) keep it in perspective and not make sweeping judgements based on individual test results.

You're right in that I doubt I ever tried it below 20.  At 20, but not below.  I'll see if I can duplicate what you're experiencing.  While I don't see that particular situation being of much interest to most people, I am intrigued that the original unit handled that situation better.  When in D though, if the car is picking a very low gear (most notably "1" or "2") then it would be very difficult for the cruise to hold the pedal steady enough to not "buck".  So perhaps the original algorithm is putting the pedal in a range where the tranny is choosing 2nd or 3rd gear and the new algorithm is pressing harder/faster and coaxing it to stay in 1st or 2nd.  If you find you have time, these are some things to try:

1) repeat the test several times - say 3 times - with each cruise control.  Just to rule out that you didn't engage it slightly different.  This is only because I can't see how the old one handled it any differently really.  If you were to engage the cruise JUST as the transmission was shifting, for example, it might put it into a instable loop.
2) repeat the test but in manual gear instead of D.


Hills are tricky.  Hills at low speeds/gears are super tricky!  But we'll see if we can't get it.

Art

[SmartieParts]

I was able to reproduce the moderate "bucking bronco" effect at slow speeds on a somewhat steep hill.  I say "moderate" because I was privy to the early testing where we had some REAL surging happen before the thing was properly tuned!  From repeated tests at 25 to 30kph (~15-18mph), this is what I learned:

• It is moderate when in 1st gear.  I confirmed this by being in manual shift mode.
• It is very slight in 2nd gear. You'd have to be pretty picky to notice I think, but it is there.
• I tried the test multiple times in Drive as well and sometimes it was slight, sometimes moderate and sometimes severe.  I used a program I have to hook my laptop to the car to see what gear it is in and it was consistent:
  • slight when it chose to be in 2nd
  • moderate when it chose to be in 1st
  • severe when it was shifting between the two gears
• The interesting thing was that before I could visually confirm what gear it was in, I wasn't sure that it was shifting.  That's what prompted me to hook it up to the laptop.
• I then repeated the test twice with the old cruise and my results do not agree with Michael's.  In my case, it surged when I took it up in 1st gear and did not when I took it up in 2nd.  However, I didn't think to do any tests in Drive to see if the cruise is somehow prompting it to be in 2nd all the time (is that possible?).

So on my immediate to-do list:

• Run the test again a few times in D with the old cruise and see if it is any better for some reason.
• I think lowering the value of MAX_DELTA_O might improve things.  I had to come back in to get my cheat-sheet, so I thought I'd write this interim report now

 - Steven

[mvaldes]

Didn't get to do the highway speed runs today. You'd think that after the wettest summer on record, people in NJ would have finally learned to drive in the rain; but you would be wrong. The interstates were a parking lot. Best I could do was some of the larger back roads at 50 and 55mph. All worked well.
Here's an odd observation, but I'll throw it out there. When it's a rainy day, like today, I lazily leave the wipers in "auto" mode when I shut off the car. (Have done so for over 17 months and 16,000 miles.) I did the same today but found that when I put the car in "reverse," the rear wiper didn't come on and do its wipey thing. If I turned off the wipers and then put it in "auto" mode and then in "reverse" it worked as it always has, it did its wipey thing. I only mention this because of the earlier wiper/door open issue. It's not that I think it's necessarily cruise control related, I just found it bizarre and thought I'd toss it out there.
When I got home a little while ago, I made a slight detour and tested the new cc in manual mode on the same hill. I only did it in 1st gear and it did the bucking thing to the point of disengaging the cc 'cause the traction control kicked in - wet leaves and such. Tried it twice with the same results. I have not tried it in 2nd, and it's rainy, nasty, etc and I've had a long day so that test will have to wait. If it's not raining tomorrow, I can give it a go in the morning and will have time to switch back and forth between units.
Thank you Steve for proving that I'm not actually crazier than I already think I am. BTW, the hill I was using was steep. (Hint - I live on South Mountain - not anywhere near the height of "real" mountains, but pretty steep going up to the top.)

[intocoasters]

Ok... forgive my inability to have thoughts flow in a comfortable manner for reading....
I will also try to keep my opinion out of the descriptions until the end. I apologize if they slip in from time to time.

The first test I did today was the version retrieval.  Sending 609900 did not give me the 60000200 I expected... I received 6000020100005500 instead.  I didn't think too much of this until some other responses did not seem to match what was expected.  I sent 60FF00 just to reset in case this unit had some previous settings - I did not receive C1EA00, 600CEA0000005500 showed up instead.  Another example: 600001 returned 6000010120007900. 

My area is rather flat, the only hills to speak of are overpasses on the highway.

Now on to the driving, no pedal maps, no changes after the reset.  An hour drive to a friends house in the country - that gave me about 45 minutes at 65 MPH in each direction, and then plenty of time at different speeds on those country-back roads - total driving time today: about 4 hours.

Performed the door open, parking brake, shift to neutral, wiper tests and received expected results.

I learned that I had no idea how the version 1 cruise truly performed in any given situation, I just took it for granted, I guess.  So, after all the tests today with the new unit, I swapped to the original unit for the drive home.

I rarely set the cruise below 30 MPH so these results are difficult to rate.  Ver.2  I could feel and hear the engine accel and decel (less throttle) on flat roads. Ver.1 was similar but seemed slower to respond - what I mean is that version 2 would(seem to) quickly apply throttle or quickly take it away.  The effect was that I could feel the change in throttle more with version 2. 

On the highway:
VER2
no radio so I could listen to the engine.  I could hear and feel the car slightly surging just as in the <30MPH tests.  Just to make sure I wasn't imagining things, I set the Scanguage to monitor TPS and RPM.  The RPM did not help mush at all, the refresh rate is too slow to make any connection to the feel/sound and the numbers listed.  The TPS, however, would fluctuate about 4 points on average.  The resume was nicely aggressive.  It could get up to speed much quicker than version 1.  The 'press and hold to accel' was also noticeably more aggressive with one glitch that I will cover later.  If I were to press the button to disengage the cruise, decelerate naturally (without brake) and then press to engage at a lower speed (say 5MPH slower) the engine would let out a little growl as it surged forward to quickly get back up to that new speed.

VER1 (for the drive home - another hour of driving)
I could hear the engine change pitch slightly, but not as noticeable. Again, the RPM did not refresh quickly enough to associate numbers with sound/feel. The TPS did seem to fluctuate by about 2 points with the Ver.1.  Resume was slower to reach the set speed and the 'press and hold to accel'  was also less aggressive.  If I disengaged the cruise and then set it again while decelerating, it would miss the new speed by about 1MPH and then continue to hold from there (about 1MPH lower than where I think I pressed the button).

Glitch?:
When I first tried the 'press and hold to accel' it was to change from 60MPH to 65MPH. The car did not seem to respond.  I let go of the button and was still at 60 MPH... I pressed and held again - this time I looked at the light. It was blinking slowly, just as it should.  The car SLOWLY accelerated to 65MPH in this case.  I tried again from 65 to 70... much much more quicker.  I slowed to 55MPH and tried again. Quickly accelerates to 60 and then, while still holding the button, it acts as if I released the button at 60MPH.  I did this more than 5 times at highway speeds with the same results.  There seems to be a 'notch' in the 'press and hold to accel' feature at the 60 to 63MPH range.
While on the empty country road I could set the cruise at 20 MPH and then press and hold the button all the way thru the 65MPH range.  Again, it accelerated just fine thru the entire range until the 60MPH mark where it feels like I let off the button and then it slowly reached 63/64 and then accelerated normally again.

Weird, huh?

On to the fun stuff!

the pedal maps are AWESOME! 
I had a BLAST trying out the different maps... even the econo map was fun to play with... It feels like a totally different car.  The B.O.O.H. curve was certainly exciting on the curvy roads out near my friend's house.  I can see why other people 'feel' like the boosters are effective.  The maps really tricked me into feeling like the car is much more powerful - the plus to this method is that I can tweak the feel by changing the maps at will. I am not locked into one profile.

I made no changed to any of the other settings. I still want to try the 'stock' Ver.2 for several days to get a feel for it.


My opinion...
I think the Ver.2 is too quick to make slight changes in order to keep the speed as exact as possible.
I feel like the Ver.1 is smoother, its almost as if it lets the speed drift a little up or down while it takes it's time to correct for variations.
The speed changes feel more abrupt in the Ver.2 right now when compared to the Ver.1.

The pedal maps are more fun and much more effective than I was expecting.

I hope that I was not too critical... this was my first time out with Ver.2.  I am sure that I would not have been able to pick out these differences if I did not have the Ver.1 to do a back to back comparison.

The highway speed control was fine with the Ver.2 .... I just think the Ver.1 was a tiny bit better.