Hi Luca, Sorry if my comments came across as argumentative or as if I was trying to defend the old driver. That was not my intention.
I was only trying to give my honest feedback as an integrator and explain why, from my particular use case, the benefits of hm2_modbus were not immediately obvious to me. I completely understand now why the old driver needs to go, and I have no objection to that. I am not a native English speaker, so I sometimes don't get the tone quite right, especially in a technical discussion. I certainly didn't mean to start a heated argument. Thanks for explaining the reasons behind hm2_modbus and for considering the feedback about the absolute path. That was actually useful for me to understand the bigger picture. Regards, Zdenek -- S pozdravem Zdeněk Zdražil ---------- Původní e‑mail ---------- Od: Luca Toniolo <[email protected]> Komu: [email protected] Datum: 13. 8. 2026 14:23:33 Předmět: Re: [Emc-users] Deprecating the old mesa_modbus / modcompile driver "Let's not get into heated arguments over this.... A user not seeing benefit is not a dis at the new tool or to the author (aka Bertho). Just a genuine question about why the new tool was needed in the first place, which is fair coming from someone for whom the old tool did the job fine. My fault for not explaining why the older one had to go. In kinder terms, the new tool includes the older features and more, it fits better in the code base, and pleases the devs, which we want to keep happy, I was fishing for anything missing, and the fixed path thing was good feedback. The older tool although usable does not fit with the shape of the codebase anymore, it's stinky code no one wants to touch, therefore it had to be rewritten, and will be retired, in due time, with proper warnings in place so no one is left behind. I'm sure more tutorials will come up with time for the new tool, and the old left to history. Let's keep it civil boys, please... Luca On 13-Aug-26 7:41 PM, Bertho Stultiens wrote: > On 8/13/26 1:22 PM, Luca Toniolo wrote: >> I can send PR to make `hm2_modbus` more portable > > There is a PR in the pipeline that does away with the absolute path > message. > > The old mesa_modbus driver is very rigid and completely inflexible. It > cannot have multiple devices on the same bus. It does not support > endian swaps or different sized types, nor can it perform modbus > device initialization or have different speed devices on the same bus. > > All these problems, and a lot more, have been solved with the > hm2_modbus driver. The old driver has been superseded for (more than) > 12 months now. > > >>> However, from my point of view I don't really see a benefit of >>> `hm2_modbus` >>> over the old solution yet. It works for me, but I don't see a feature >>> that would make me prefer it over the old driver. > Does that mean we need to support the old driver ad infinitum because > your use case has no advantage? > > For the developers, the old driver is presenting structural problems > in code organization and exposure. Removing it is the only real > option. We have a proper replacement, hm2_modbus, that can do it all > and better. > > >>> There is also one thing I don't like about the new version: the >>> `mbccbs` parameter requires an absolute path to the MBCCB file. The >>> current documentation explicitly says: > Will be fixed shortly. > _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users " _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
