marlin-config
marlin-config copied to clipboard
akaJes
mct vs marlin-conf
Hello,
I've done these experiments many times with the Tevo Tornado 1.1 firmware ; it's a Marlin 1.1.5.
Firest of all, cloned 4 Marlin repos.
repo1 : opened in marlin-conf, set to 1.1.5 repo2 : opened in marlin-conf, set to 1.1.9 repo3 : opened in marlin-conf, set to 1.1.9 repo4 : cloned using mct clone (so it's 1.1.9 at this time)
I copied (overwrited) config files in repo #1 with Tevo Tornado config files. then opened 2x marlin-conf instances, side by side. 1st with repo #1 (1.1.5+Tornado files), 2nd with repo #2 (vanila marlin 1.1.9)
And then, duplicated modifications from #1 to #2 by hand.
Doing this, repo #2 was turned into a "Tornado 1.1.9".
Then, I drag/drop'd original Tornado config files into marlin-conf opened with repo #3 (1.1.9) ; doing this, config files were converted to 1.1.9 version.
Then, I conpared results from these 2 upgrading methods (by hand vs marlin-conf, repo #2 vs repo #3). Guess what ? They are identical. Great ! (I also copared results from direct code editing : perfect !)
At this point, repo #2 == repo #3
Then, I tested mct... I copied config files into repo #4 (mct clone), and did "mct git 1.1.5 json" (remember, Tornado files are from Marlin 1.1.5), and then "mct git 1.1.9 h". As I understand mct, the first command generates a "diff file" that describes the work to do on newest versions.
I got 1.1.9 config files, but they are very different from the amrlin-conf generated ones. Some parameters get commented, some others commented out, some values are not copied, and so on. It is havoc ! It looks like marlin-conf and mct don't use the same routines. marlin-conf ones appear to be just perfect. Not mct ones.
I did this many times.
But I could be misusing mct !!! On one hand, uou give very few information about mct, and on the other hand, it is probably the most powerfull tool.
I find it disturbing no one has responded to your issue....But having said that your post is somewhat confusing. You seem to be saying Marlin Conf is perfect and MCT is not. If that is the case why are you posting the issue here? Isn't MCT the same? In which case should your question why are the results different? Not that it matters it seems that the developer is not inclined to participate in these forums very often.
It's alright I am kinda new to programming and I am trying to learn on my own but may have to take online classes
On Sat, Mar 30, 2019, 11:50 PM cinealfa [email protected] wrote:
I find it disturbing no one has responded to your issue....
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/akaJes/marlin-config/issues/80#issuecomment-478311905, or mute the thread https://github.com/notifications/unsubscribe-auth/AuLjB4cKqX8GR3qREI9d7glugikZJzZcks5vcD6qgaJpZM4XLdPs .
@cinealfa :
mct and marlin-conf are 2 different applications developped by akajes, and mct usage is described on the marlin-config main page : see "mct help" (end of page). mct is installed separatly, and needs node.js
In the OP, I just describe all the steps needed to compare firmware upgrading : manual vs marlin-conf vs mct (optional), from a random untouched OEM firmware. Needs 4 repos : one for original, remaining untouched, one for manual upgrading, one for marlin-conf upgrading, and one for mct upgrading. Then winmerge is used to compare all 4 versions.
marlin-conf is not perfect, but does 99% of the job. See issue #81 for example. Wrting a parser is not easy, and you can imagine the problems when nearly every new version of Marlin comes with different headers versions, #define and so on. There is no support from Thinkyhead, as he explained in a interview with Thomas Sanladerer.
I suspect marlin configurator to be an abandonware...
II think you are right Joe, It is ashamed because it definitely makes configuring easier. Perhaps it could be folded into the new 2.1? It makes sense to have a fast easy way to do updates.
