Inaccuracy of the pdr, performance data recorder...
#41
Banned Scam/Spammer
So I say you appear distressed by facts and data and your intelligent response is simply copying me? Good job gramps. Again, why would you get so upset with facts and data? That is the only weird behavior around here. It would seem to me, you would want to know what is accurate and what is not. This type of information improves the knowledge of the community so we know what to trust and what is fairy dust.
#42
Banned Scam/Spammer
I was just hoping for a rational response to your negative reaction to facts and data (you see a lot of this in society these days). Anytime I see someone post something so off-base I try to understand why. I guess asking for a halfway intelligent response from you was too much. I will have to lower my expectations in the future.
#43
Banned Scam/Spammer
I was just hoping for a rational response to your negative reaction to facts and data (you see a lot of this in society these days). Anytime I see someone post something so off-base I try to understand why. I guess asking for a halfway intelligent response from you was too much. I will have to lower my expectations in the future.
The following users liked this post:
Z0HS1CK (08-01-2020)
#44
Team Owner
The dragy is as accurate as the gold standard vbox which is used by every magazine publication im aware of. Compared against nhra track timing equipment on hundreds of runs in a dozen or so different cars from awd audis to rwd camaros, to fwd civics ive found it to be accurate myself to .05 in et and .5mph in trap.
The numbers you see in the dragy video overlay are without rollout. Dragy also gives 1 ft rollout. The runs with rollout are...
Supposedly the pdr is not giving rollout numbers. Either way its off and not off by the same amount each run either making it both inaccurate and with poor reliability (statstical repeatability)
The numbers you see in the dragy video overlay are without rollout. Dragy also gives 1 ft rollout. The runs with rollout are...
Supposedly the pdr is not giving rollout numbers. Either way its off and not off by the same amount each run either making it both inaccurate and with poor reliability (statstical repeatability)
Last edited by themonk; 08-01-2020 at 03:31 PM.
#45
#46
Safety Car
Thread Starter
But yes those are the accurate numbers for my car on that surface in that weather at that weight with those miles on the engine. The pdr numbers certainly were not.
That said im not surprised or disappointed by those numbers. This was a very very early first set of baseline data to compare later when the engine is fully worn in. Its also terrible weather conditions so no displeasure on my side
Last edited by bhvrdr; 08-01-2020 at 05:41 PM.
#47
Pro
Wouldn't the PDR system use wheel speed data instead of GPS for speed/timing?
#48
Safety Car
Thread Starter
Thanks for the reply. I initially thought that as some other oem loggers ive messed wth use ecu data such as engine speed, speedometer, wheel speed vs timestamp. After doing some research it looks like at least on the c7 it is reported by GM that they use a 5hz gps module in their pdr. These just arent accurate enough to be used for real comparison data. The minimum for decades has been 10hz loggers.
Last edited by bhvrdr; 08-01-2020 at 05:47 PM.
#49
Le Mans Master
#50
Le Mans Master
#51
Race Director
Member Since: Aug 2019
Location: Central Illinois
Posts: 10,099
Received 11,626 Likes
on
4,651 Posts
Does the non PDR cars also use GPS or an accelerometer to show the 0-60. They also show G loads in all directions and so I don't know if they use the G loads to calculate it or GPS.
#52
#54
Safety Car
Thread Starter
#55
Looks like GM only put a 5hz GPS in the unit. No wonder its off.
Looks like c7 track guys even had a petition setup for gm to fix it as lap times are off upwards of 2 seconds at times. I wasnt aware.
Still plenty happy with the car of course and the performance is great. Still beats my 650hp tvs1900 supercharged camaro ss to 60 mph with ease.
As i said i just like to measure performance on cars and on modifications so i always make sure im doing it accurately.
Looks like c7 track guys even had a petition setup for gm to fix it as lap times are off upwards of 2 seconds at times. I wasnt aware.
Still plenty happy with the car of course and the performance is great. Still beats my 650hp tvs1900 supercharged camaro ss to 60 mph with ease.
As i said i just like to measure performance on cars and on modifications so i always make sure im doing it accurately.
Thanks for the reply. I initially thought that as some other oem loggers ive messed wth use ecu data such as engine speed, speedometer, wheel speed vs timestamp. After doing some research it looks like at least on the c7 it is reported by GM that they use a 5hz gps module in their pdr. These just arent accurate enough to be used for real comparison data. The minimum for decades has been 10hz loggers.
At least on the C7, the PDR uses the speedometer data to display speed which comes from a sensor in the transmission (and another in the diff I think). It does not use the GPS to display speed in the video or in the data (at least without you creating a math channel in Pi Toolbox).
Also, the data overlay on the video does lag the video, so for example if you hop off the gas it will show up in the overlay ~2-3 tenths later than where it actually happened in the video. This is a result of processing time since it is written out as the video is recorded as well as the fact that there is some smoothing of the native logging rate to the frames that are overlaid on the video. I confirmed this with Cosworth engineering last Summer. However, if you open up the MP4 in Pi toolbox and look at a time/distance chart next to a video display in the tool they will be correctly synced up.
Finally, the issue with the PDR lap times being ~1.8s off was something I found and reported to GM and which they later offered a TSB to address. There is an Ask Tadge reply on this but it was a result of some changes made to the GPS satellites to use leap seconds to adjust. It didn't happen all the time and when it did it was very noticeable as the lap time would back up 1.8s during the lap in the video.
I don't know if this helps explain your concern but I wanted to at least provide the additional info.
The following users liked this post:
Z0HS1CK (08-01-2020)
#57
Safety Car
Thread Starter
Interesting data, thanks for posting. I guess its somewhat moot where im concerned but it would be fun to learn more about the pdr snd how it captures acceleration data.
It sounds like it may or may not be different from how it captures lap times. So for laptimes it uses 5hz GPS for the actal lap timing and then uses sensor data for speed overlay which seems to make sense.
Now for 0-60, 1/4 mile time and trap i wonder if they use the GPS or sensor data or a combination. They could use sensor data for 0-60, 60-130, 0-100, 100-0 etc but for 1/4 mile and 1/8 mile they'd have to use GPS which it would make sense to use for all the acceleration timers. I dont know enough about it though to know.
It sounds like it may or may not be different from how it captures lap times. So for laptimes it uses 5hz GPS for the actal lap timing and then uses sensor data for speed overlay which seems to make sense.
Now for 0-60, 1/4 mile time and trap i wonder if they use the GPS or sensor data or a combination. They could use sensor data for 0-60, 60-130, 0-100, 100-0 etc but for 1/4 mile and 1/8 mile they'd have to use GPS which it would make sense to use for all the acceleration timers. I dont know enough about it though to know.
#58
Race Director
Interesting data, thanks for posting. I guess its somewhat moot where im concerned but it would be fun to learn more about the pdr snd how it captures acceleration data.
It sounds like it may or may not be different from how it captures lap times. So for laptimes it uses 5hz GPS for the actal lap timing and then uses sensor data for speed overlay which seems to make sense.
Now for 0-60, 1/4 mile time and trap i wonder if they use the GPS or sensor data or a combination. They could use sensor data for 0-60, 60-130, 0-100, 100-0 etc but for 1/4 mile and 1/8 mile they'd have to use GPS which it would make sense to use for all the acceleration timers. I dont know enough about it though to know.
It sounds like it may or may not be different from how it captures lap times. So for laptimes it uses 5hz GPS for the actal lap timing and then uses sensor data for speed overlay which seems to make sense.
Now for 0-60, 1/4 mile time and trap i wonder if they use the GPS or sensor data or a combination. They could use sensor data for 0-60, 60-130, 0-100, 100-0 etc but for 1/4 mile and 1/8 mile they'd have to use GPS which it would make sense to use for all the acceleration timers. I dont know enough about it though to know.
Anyone wishing to dispute that fact should create an "ask tadge" question and pester that member who controls those questions that are posed to tadge. Good luck getting that past the gatekeeper... getting that one to be answered brings plenty of controversy in this world as shown by the hurt feelings the OP laid on people here to dispel these ridiculous mid 2 second 0 to 60 times NOT THAT THE CAR CANT DO 2.9 WITH A 1FT ROLLOUT. Regardless of how quick or not the C8 is from 0 to whatever, the pdr and dash timer is horrifically inconsistent.
Last edited by 23/C8Z; 08-02-2020 at 09:18 AM.
The following users liked this post:
23/C8Z (08-02-2020)
#60
Le Mans Master
Member Since: Jul 2005
Location: Think BEFORE hitting "Submit Reply"
Posts: 9,415
Received 708 Likes
on
391 Posts
I think as more owners get their cars and start testing, we'll see legit sub 3 second runs.
The car can do it, just how far below is the question.
I'm really looking forward to what the Z06 brings to the table.