dbbarron

What would be the typical workflow for reviewing a HPDE session using a SCHD and SoloDL (or other AIM logger). I am comparing this workflow to a video synch'd workflow (like VBOX).

Specifically, I want to focus on actions immediately after data has all been downloaded.

Does one first review the fastest lap on video looking for obvious areas of improvement (e.g., line, using full track, etc.) and then separately review data around these areas? Does one first look for issues in the data first (e.g., Gsum) to identify areas of improvement, and then move over to the video to inspect the attitude of the car at these improvement points, etc.

A bit of a chicken or egg problem?

ProCoach

For the typical SDL/SCHD at-track workflow, I would recommend keeping it simple. Much better to look at a few KPI's repeatedly and identify trends as opposed to heading down a rabbit hole, IMO.

1) Pull the SD card, find and replay the quickest lap.

If you don't already have the information on the video, put in the SCHD config fast lap number/fast lap time along with the normal current lap number/current lap time info. Since RSA will open the accompanying data file to the quickest lap by default, you can see any instances WHY the data might not be optimal FIRST, rather than chasing your tail.

A great deal of your evaluation of the video will parallel your review of the data, but pay attention to transitions in real-time. "Soft" transitions where you release throttle early and cover distance before more rapid (and proper) decel, multiple steering angle inputs before settling on a sustainable arc, and delays in throttle application from the time of brake release to max cornering loading.

If you have an all-in-one camera, and making sure it is calibrated at rest and that the car is on a level surface, make quick comparisons between max brake g loadings between brake zones and comparisons of max sustained cornering loadings between corners. Make sure there are no apparent deficits from proven and achieved high values.

2) Download and open your SDL file.

Hopefully, you've spent some time beforehand creating at least ONE profile (arrangement of informational windows that each file opens by default) that includes GPS Speed, GPS or actual longitudinal g and a GPS track map. You can add a GPS gSum strip chart, if you really want to find out more, but often loading a gSum channel into the colored map is more effective than a strip chart.

Study the cleanliness of your transitions. Pick one brake zone and/or corner complex that shows the most opportunity for improvement for the next session. ONE, at a time. By improving the quality of your basic skill executions, it's going to help everywhere.

If you're having trouble figuring out where to start improving, overlay the three quickest laps and identify the areas with the greatest variation. Prioritize from there.

So, just two things to start. Obviously, it gets much better if you can look at the two (TP/BP) or three (TP/BP/SA) driver control individually, but this will get you started.

The next item might be additional channel colored maps to identify WHERE and for how LONG you are slowest in speed in each corner (better, more easily digestible and more easily actionable than strip charts, IMO), then segment time analysis.

You want to make sure you NEVER make a judgment on a deficit on the basis of one or two measures. You ALWAYS want to look at the video FIRST. Saves a lot of time.

There's a ton more, but that's what I suggest first off.

dbbarron

As I compare systems, not having configurable color gradient maps for specific variables is a big weakness of CT2 relative to say RS2; however, there is that video sync thing going the other way....

Matt Romanowski

I basically do Peter's steps in reverse.

I look at the data first. I have a number of profiles set up for each area that I want to focus on. There are photos of the profiles below. I do this for the fast lap and a few laps together that are close in lap times. I feel that this look at data is the most productive use of time to quickly ID the areas for improvement and not have to watch multiple laps of data.

First, I always go through a quick channels report that show the car health and some basic performance parameters. I'll check to make sure there were no over revs, good oil pressure, no high temps too high, etc.

Next, I'll look at a general speeds profile. I look for sharp (saw tooth) transitions from throttle to brake. throttle application, overall speeds, and good corner min speeds (sharp V shape in slower corners and U shape in faster).

Then, I look at a braking profile to see how the brakes are applied, how well they are trailed off (different for every corner), shape of the traction circle (same info as gsum).

After that, I'll look at cornering info like corner radius (shows if you are early, middle, or late apexing), gsum, under/over steer, and a few other measures.

With that, I have a good overview of the entire picture of what is going on and I'm able to do it in a few minutes. With that, I'll pick no more than 3 things to work on the next session. Depending on what it is, it may be only 1 thing! My track debrief maps/reviews have a spot to write these things down, as well as a spot for 3 things that were done well (you always have include something positive too!).

Once I've done that, I will usually go to the video (sometimes during the above if I can't see why something in the data happened) and use the video to start filling in the story of what happened and get some info on the why. The video also helps to fill in the plan for improving. If you have selected something like trail brake into turn 1 more, then you can use the video to see where you are getting out of the brakes now, where you want to be doing it, and visualize the whole thing.

Last thing for an overly long post already - making good use of your profiles is critical to quick analysis! With the profiles set up, you can make 1 click and see the next set of info without having to configure anything!

For folks with a question like yours, I highly recommend taking John Block's webinars at www.auto-ware.com. He will show you how to set up the profiles, give you his set - which may or may not work for you - and most importantly, teach you the methodology to do all of this yourself.

Hope this helps some and gives you some good ideas on the use of profiles.

ProCoach

That's right. Every system has it's pluses and minuses.

Matt's profiles, while attractive, comprehensive and informative, are mostly overkill for a 40 minute analysis and turnaround. Many of these measures are not available (brake temp/shock displacement) from most cars I've seen. So pick and choose. You can still achieve a great deal in a short time.

Matt Romanowski

I think you misread. There is no shock data in those profiles and the brake temp is displayed but I never mentioned using it for what the OP was asking about. I use the profiles I have for my car and the brake temps are there, but most folks I work with don't have that as you said. Even a regular Solo will let you do the speed analysis, braking with Long G instead of pressures, friction circle, G Sum, corner radius, and speed and/or long g GPS maps. That is still certainly a great analysis set.

I can do the workflow I posted above in 5-10 minutes going through the data. That is why I use profiles and preconfigured and auto applied math channels. It automates the manipulation and display of the data so all I have to do is analyze. That is one of the great features of AiM and MoTeC (and the other higher end systems like Cosworth, Bosch, etc).

In my experience and opinion, I can break down data much faster than watching a number of video laps. I use the data to ID areas and then use the video for further study. Maybe it comes down to how comfortable you are with numbers or the visuals?

As to overkill, I guess it depends on what you are trying to get out of the data/video and what the driver's abilities and performance level are. Some people's low hanging fruit is picked from the ground and some requires extension ladders.

ProCoach

Hah! True.

I'd say most folks we are addressing need a step ladder, which is what this equipment is for many.

Biggest variance in our workflows sounds like the difference in our reliance on video.

In my experience, drivers of even significant skill (pro level, IMSA and WEC) can identify opportunities for improvement by forming a workflow based on the differences between what they remember doing and what they did do, discovered when they review the video. MUCH more so for most drivers here on Rennlist...

Only then can the examination of "extension ladder" data measures like throttle speed and brake aggression, math channels that use raw sensor information to yield an objective measure for what has been (before data) a subjective observation. These are the areas most more advanced drivers can improve, IMO.

The ability to validate, confirm or augment observations in the video using data measures is what accelerates my analysis, and makes it better than either one, on their own... when studying how people learn and assimilate the information needed to progress to the next level, the power of the visual aids, particularly channel color maps and "intelligent" (data infused/overlaid) video has been most useful.

To the OP, keep it simple and add speed slowly, just like on the track.

Happy New Year!

Matt Romanowski

I think we do basically the same things, but in reverse order. One of the best things with motorsport is there are many ways to do the same thing, all which can have equal effectiveness for the user.

I like the data approach because I can look at many laps and compare tendencies across multiple laps, not just their fast lap video. Then I can develop a learning plan and approach that fits their overall abilities and styles, not just their one lap.

I would not agree with part I added italics to. In my extensive training on how to educate and tutor people that started nearly 20 years ago, I've learned and applied that everyone learns differently. One of the worst things that instructors do is to assume that everyone learns the same. One of the most important jobs an instructor (in any field) has is to do is quickly figure out how their student(s) learn and tailor their teaching to that.

This circles back around to the OPs original topic. You have to develop a workflow that is good for you. There is no ONE correct way. Some people will start with video, some will start with data. It's all what produces the best results for you!

ProCoach

The beauty of the latest crop of simple, powerful data and video acquisition tools is that finally, after decades of being the province of full time engineers and professional race teams, they are now capable of providing information in such a way that track day, DE, club level and entry level pro drivers can coach themselves.

Most folks that do this are scientists, engineers, professionals and business people. They understand the value of metrics and more importantly, how identifying the trends of those metrics can direct self improvement. Safely and effectively...

The hardest part about collecting and digesting these informational metrics is deciding on what to do next with them. That's where my work is focused. Not on "teaching them," but helping them develop strategies and workflows to guide their own self coached progress. To improve in a lasting way, THEY have to OWN IT!

The challenges drivers at every level other than rank novice are remarkably similar and consistent. After learning the "best execution of fundamental skills," it's a constant cycle of breaking down those individual skill executions and improving EVERY ingredient of them.

While different drivers may assimilate complex concepts in a variety of different ways, the challenges of improving key performance indicators are remarkably similar. And these tools are a must to do just that.

We can certainly agree on that!