Data file sets by manufacturer now available again

[JohnCoker]

The late, lamented Mark Koelsch used to maintain lists of data files for each manufacturer. I have just added a feature to ThrustCurve.org that does something similar.
If you go to My Outbox and there is nothing in it, you will see links for RASP and RockSim files for each manufacturer.
https://www.thrustcurve.org/outbox/

 

[rharshberger]

Hooray!Thank you John! That will make using new motors so much easier in RS and OR!

 

[Bill S]

Thanks!

 

[Nytrunner]

So by downloading the AT or CTI .eng files, and placing them in my alternate motor directory, I'll have all their available motors including recent additions?

 

[rharshberger]

So by downloading the AT or CTI .eng files, and placing them in my alternate motor directory, I'll have all their available motors including recent additions?

Thats how its supposed to work Nytrunner.

 

[Nytrunner]

Nice, I hate hunting for individual new motors as they come out. Time fora 5yr motor database refresh!

 

[JohnCoker]

With the new site, it's a lot easier to add new features. (I still wish I would have stuck with a SQL DB rather than switched to MongoDB, but that doesn't really prevent anything.)

 

[Banzai88]

Thank you!

 

[DaHabes]

John, a couple of questions. First, what is the source of these files? Are they from the manufacturers? Second, are they ever updated? I pulled them down and looked at them. While the Estes file has the C5, but it's the old C5-5, not the new C5-3. The Aerotech file does not have the recently posted F52C and H13ST, not does it have the 38mm H219T I used last year (2019) for my L1 cert (the H219T was in the Rocksim 10 updated Aerotech RSE file).

At a minimum it would be helpful if there is a datetime stamp in the filename to know the creation date of the original file. Under the assumption that any engines certified after that date are not included; it would be informative as to whether someone should use these to update their existing local files in Rocksim or OR.

Ideally, these would get updated whenever a new engine is certified. I haven't found the manufacturers to be timely in terms of updates to the presumed master files they post on their websites (do they ever)? It would be a value add if this was done and new RSE files posted so that each and every user would not have to update local files individually each and every time there is a change.

Finally, I don't know what to make of the CTI file. I looked at my CTI file from Rocksim 10 and the manufacturer name entry is the same for every engine - Cesaroni Technology Inc. In the file on Thrustcurve, many of the manufacturer name entries are [CTI] Cesaroni Technology Inc. When the names are different they show up in two different lists in Rocksim, which is a PITA. In addition, the Thrustcurve file engine names use the long nomenclature (e.g., 108-G68-WH-13A) whereas the Rocksim10 file use the short nomenclature (G68WH). Hard to know which one is more current.

Would it be advantageous to have a couple of people tag team to updates these files universally over time - one to make the updates for new engines and someone else to validate the updates before the get posted to TRF? Glad to participate if someone else wants to join the fun.

Finally, it would be very helpful if there was one entry in the Propulsion forum for newly certified engines that is a sticky entry and only updated by members that have the authorized information to share regarding these updates. People can start separate threads to discuss usage. Unless it's sticky, it will just roll off the bottom of the screen and most people would never find it. My reason for suggesting this is that there have been several new engine entries posted recently, two with the exact same name but for different manufacturers and engine classes. For something as integral to the hobby as new engines, a more formal forum structure would be very helpful.

 

[JohnCoker]

John, a couple of questions. First, what is the source of these files? Are they from the manufacturers?

The data comes from various sources; there is a field in the ThrustCurve.org database for where the data came from.

Second, are they ever updated? I pulled them down and looked at them. While the Estes file has the C5, but it's the old C5-5, not the new C5-3. The Aerotech file does not have the recently posted F52C and H13ST, not does it have the 38mm H219T I used last year (2019) for my L1 cert (the H219T was in the Rocksim 10 updated Aerotech RSE file).

It's a communal effort. If you see an out of date file, please upload a newer one.

At a minimum it would be helpful if there is a datetime stamp in the filename to know the creation date of the original file. Under the assumption that any engines certified after that date are not included; it would be informative as to whether someone should use these to update their existing local files in Rocksim or OR.

There are created and updated timestamps in the database which you can view on the site. The data formats don't include a field for that (at least RASP doesn't).

Ideally, these would get updated whenever a new engine is certified. I haven't found the manufacturers to be timely in terms of updates to the presumed master files they post on their websites (do they ever)? It would be a value add if this was done and new RSE files posted so that each and every user would not have to update local files individually each and every time there is a change.

The manufacturers have never felt it was their responsibility to provide simulator data. That's why I started ThrustCurve.org. You can read more about the history if you're interested: https://www.thrustcurve.org/info/background.html.

Finally, I don't know what to make of the CTI file. I looked at my CTI file from Rocksim 10 and the manufacturer name entry is the same for every engine - Cesaroni Technology Inc. In the file on Thrustcurve, many of the manufacturer name entries are [CTI] Cesaroni Technology Inc. When the names are different they show up in two different lists in Rocksim, which is a PITA. In addition, the Thrustcurve file engine names use the long nomenclature (e.g., 108-G68-WH-13A) whereas the Rocksim10 file use the short nomenclature (G68WH). Hard to know which one is more current.

ThrustCurve supports various names (the full name, the standard abbreviation and various aliases) which historically have been used. There is an incomplete list of manufacturer names in the NAR combined list.

Would it be advantageous to have a couple of people tag team to updates these files universally over time - one to make the updates for new engines and someone else to validate the updates before the get posted to TRF? Glad to participate if someone else wants to join the fun.

Anyone can submit new entries, and we would all welcome more accurate data. I have a page on how to get started: https://www.thrustcurve.org/info/contribute.html.

https://www.thrustcurve.org/info/contribute.html

Finally, it would be very helpful if there was one entry in the Propulsion forum for newly certified engines that is a sticky entry and only updated by members that have the authorized information to share regarding these updates. People can start separate threads to discuss usage. Unless it's sticky, it will just roll off the bottom of the screen and most people would never find it. My reason for suggesting this is that there have been several new engine entries posted recently, two with the exact same name but for different manufacturers and engine classes. For something as integral to the hobby as new engines, a more formal forum structure would be very helpful.

See the Recent Updated page on ThrustCurve.org: https://www.thrustcurve.org/motors/updates.html
This is live (produced when requested) and requires no manual work to maintain.

 

[DaHabes]

Anyone can submit new entries, and we would all welcome more accurate data. I have a page on how to get started: https://www.thrustcurve.org/info/contribute.html.

[/QUOTE]
HI, John. I have updates Estes and Quest RSE files, datetime stamped, which include the newest engines, to contribute. I am logged in (I can see "My Stuff"). I can't find any link that allows me to upload these files. The "contribute" link for motor data is the only upload link I have found on the website.

 

[JohnCoker]

HI, John. I have updates Estes and Quest RSE files, datetime stamped, which include the newest engines, to contribute. I am logged in (I can see "My Stuff"). I can't find any link that allows me to upload these files. The "contribute" link for motor data is the only upload link I have found on the website.

ThrustCurve.org maintains the data files in a database, per motor. The "sets" are synthesized by putting individual files together. The link you found is the right one, when you create a new data file, just upload it with its corresponding motor.

 

[DaHabes]

ThrustCurve.org maintains the data files in a database, per motor. The "sets" are synthesized by putting individual files together. The link you found is the right one, when you create a new data file, just upload it with its corresponding motor.

"Sets" are fine if you are querying the DB. I've aggregated the new engines into the RSE files that folks can import to Rocksim to update the engine lists in Rocksim (which, unlike OR, doesn't use Thrustcurve as it's source, although it can't be updated from Thrustcurve dynamically, which is a huge flaw). These updated file ought to replace the "Data file sets by manufacturer now available again" that you recently updated, which are indeed out of date. The question is, how?

 

[CarVac]

John, speaking of sets "synthesized by putting individual files together"... is the categorization of motors done programmatically?

Because there's one K motor (out of production Gorilla K695BT) that's somehow lumped in with the L motors.

 

[JohnCoker]

John, speaking of sets "synthesized by putting individual files together"... is the categorization of motors done programmatically?
Because there's one K motor (out of production Gorilla K695BT) that's somehow lumped in with the L motors.

That impulse class is supposed to be picked up from the common name. Not sure what happened there, but I fixed it manually. I'll write a script to go through and check that there are no other bad ones.

 

[JohnCoker]

"Sets" are fine if you are querying the DB. I've aggregated the new engines into the RSE files that folks can import to Rocksim to update the engine lists in Rocksim (which, unlike OR, doesn't use Thrustcurve as it's source, although it can't be updated from Thrustcurve dynamically, which is a huge flaw). These updated file ought to replace the "Data file sets by manufacturer now available again" that you recently updated, which are indeed out of date. The question is, how?

When you download the RockSim set from ThrustCurve, it already aggregates the individual files into a single XML file. Each time you download it, you get the latest entries for every motor for that manufacturer in the database. So for ThrustCurve.org, the database is always the source of truth and all web pages and download files are built from that. There is nothing that can be "out of date," because it's all produced dynamically.

Each time someone creates a new data file and uploads it, it will immediately show up for everyone using the site or downloading data; there are no additional manual steps.

I think I'm still not understanding what you're tasking for.

 

[DaHabes]

When you download the RockSim set from ThrustCurve, it already aggregates the individual files into a single XML file. Each time you download it, you get the latest entries for every motor for that manufacturer in the database. So for ThrustCurve.org, the database is always the source of truth and all web pages and download files are built from that. There is nothing that can be "out of date," because it's all produced dynamically.

Each time someone creates a new data file and uploads it, it will immediately show up for everyone using the site or downloading data; there are no additional manual steps.

I think I'm still not understanding what you're tasking for.

Are you saying that when I download a Rocksim set for a manufacturer (lets say Quest) from this link https://www.thrustcurve.org/outbox/ it aggregates all of the Quest engines before the download? If that is the case, terrific, except that it isn't working correctly. The Quest RSE file only has 8 engines in it and does not include the D20W and C18W, for example. The Quest ENG file has 12 engines and includes the new ones. As for the Estes files, the C5 has a delay of 5 seconds, which is the old OOP C5. The new C5 only has a 3 second delay option. I suspect that is an error in the file. If the intent is to have both available and OOP engines in the files, the entry for the C5 needs to be updated.

Neither of the Aerotech ENG or RSE files have the 38mm H219T in the file, which I used for my L1 last year (and I have another I recently bought). Also, for the Aerotech files, the newly certified K76WN, F52C and H13ST are in the ENG file but none are in the RSE file.

So, all I am saying is the confusion I am experiencing is that the Rocksim files, ENG and RSE, are not consistent and do not have all of the engines for each manufacturer in them. It was not clear that the files were generated dynamically, which is why I was requesting to upload an updated RSE file. Now it seems clear to me that while the files are dynamically created upon download, they are incomplete and do not include all of the engines for each manufacturer in each file, with further inconsistencies between the ENG and RSE versions. Hope this explanation helps.

 

[JohnCoker]

Are you saying that when I download a Rocksim set for a manufacturer (lets say Quest) from this link https://www.thrustcurve.org/outbox/ it aggregates all of the Quest engines before the download? If that is the case, terrific, except that it isn't working correctly.

Yes, that's how it's supposed to work. I will look into the examples you posted to see what's going on there.

Right now, it's only querying in-production motors.

Of course, if there are no data files in the right format for a motor, they won't be included so it depends on what files have been contributed. (It's much harder to make an RSE file than a RASP file, so there are fewer of them.)

 

[JohnCoker]

The Quest RSE file only has 8 engines in it and does not include the D20W and C18W, for example. The Quest ENG file has 12 engines and includes the new ones. As for the Estes files, the C5 has a delay of 5 seconds, which is the old OOP C5. The new C5 only has a 3 second delay option. I suspect that is an error in the file. If the intent is to have both available and OOP engines in the files, the entry for the C5 needs to be updated.

Neither of the Aerotech ENG or RSE files have the 38mm H219T in the file, which I used for my L1 last year (and I have another I recently bought). Also, for the Aerotech files, the newly certified K76WN, F52C and H13ST are in the ENG file but none are in the RSE file.

There have been no RSE files submitted for the Quest D20 and C18.

I updated the info on the Estes C5 motor entry to indicate the only delay is 3s. I also fixed the two latest files (since I submitted them).

I just downloaded the Aerotech.ENG file and it does contain the H219T, but there is no RSE data file for this motor.

Thanks for taking the time to report the issues. If you have created any of these missing RSE files, please upload them which will help everyone.

 

[JohnCoker]

We have managed to collect a lot of data files; thanks to everyone who contributed!

Currently we have 1770 files and info on 1091 motors. Of the 777 non-OOP motors, only 9 are missing data of any kind.

But as Mike Haberer points out, there are many more RASP files than RSE files. I think that's due to the fact that it's easier to create RASP files and they can be used with more programs. So if you are a RockSim user and can't find an RSE file for your favorite motor, you can load RASP files into RockSim as well using Engine Edit. And if you create an RSE file, please upload it to ThrustCurve.org for other to use.

 

[Rob702Martinez]

Did the option to see thrust curves in "pounds" go away? Not very good at reading NMs on the fly.

 

[JohnCoker]

Did the option to see thrust curves in "pounds" go away? Not very good at reading NMs on the fly.

You can choose all units (My Stuff / Preferences) and you can also toggle the units on the thrust curve graphs for the motor and the data file (using the N/lbf buttons on the bottom right).

 

[Rob702Martinez]

Thanks John!

 

[DaHabes]

There have been no RSE files submitted for the Quest D20 and C18.

I updated the info on the Estes C5 motor entry to indicate the only delay is 3s. I also fixed the two latest files (since I submitted them).

I just downloaded the Aerotech.ENG file and it does contain the H219T, but there is no RSE data file for this motor.

Thanks for taking the time to report the issues. If you have created any of these missing RSE files, please upload them which will help everyone.

OK, I get it now. The dynamically created manufacturer RSE files are missing engines because the individual engine RSE files are missing. I was also under the mistaken impression before I got into this thread that ENG files were for single engines and RSE files were aggregates. Those were the missing pieces of info. Can I assume that every engine has a RASP file but not necessarily a RSE file, or are their RASP engine files missing for which there are only RSE files?

 

[JohnCoker]

Can I assume that every engine has a RASP file but not necessarily a RSE file, or are their RASP engine files missing for which there are only RSE files?

Not necessarily; it depends what people submit as new motors are introduced. RASP files are more common because they're easier to make. Sometimes I have the data from the cert. org. when a new motor is certified, in which case it's usually simple to convert it into a RASP file.

 

[JohnCoker]

What would also be possible (I think) is to synthesize each format from the other (plus motor info in ThrustCurve.org). That's something I'll look into at some point.

What would really be cool is if the raw test data could be uploaded and files produced from that.

 

[H. Craig Miller]

JoePfeiffer, an OpenRocket developer, posted this message: "Several of the motor files at thrustcurve.org are failing to download (the vast majority are succeeding). The only diagnostic is a message similar to " [java] Error downloading RockSim for motorID=782". Some hint as to what's wrong is needed..."

Any ideas?

 

[JohnCoker]

JoePfeiffer, an OpenRocket developer, posted this message: "Several of the motor files at thrustcurve.org are failing to download (the vast majority are succeeding). The only diagnostic is a message similar to " [java] Error downloading RockSim for motorID=782". Some hint as to what's wrong is needed..."

That sounds like using the API rather than the manufacturer sets. I'll see if I can reproduce it (given that motor ID).

curl -v 'https://www.thrustcurve.org/servlets/download' -d '<download-request>
  <motor-id>782</motor-id>
</download-request>'

It appears to work for me (using the legacy API through curl), but if I had more details about the exact request being issued, I might be able to say more.