I agree with this completely. And that’s why I said it was just my personal opinion.
I completely accept my viewpoint is a sample size of 1 and not everyone will agree. I really like being able to make my toolpath in estlcam, export, and upload from my garage pc and I couldn’t do that with AP mode so I’ve no reason to go back.
I’d probably be just as happy with a pendant like the one Steven potter made without a screen, it’s the physical feedback I like. 
It’s by no means a must have. But for the sake of what I think was less than 40 bux it makes me happy.
4 Likes
Since I don’t see the appeal of the pendants I am just trying to go off the info I glean from the posts. I think a vast majority is about jogging. Every so often, it is the ability to just hit start. And sometimes it is about physical buttons. (you I think want all three so a pendant makes the most sense, but for others that just want one we might be able to save a lot of hassle just adding a bluetooth joystick, or Jason’s extension.
I am putting out the optional way to do this things to see how many issues we can take care of without any added expense.
For example I like gcode visualization, but not the tablet mode it takes to get it. Or using a USB cable with a gcode sender. I think Jason had an extension for this as well but I am not seeing it in his repo.
The more people we can make happy without any added cost is best for everything. Less time spent troubleshooting more time spent doing the more fun stuff.
1 Like
Understood, I just really hope that is no longer the case and don’t want people to think you can’t do this. The connection stuff is rock solid these days. A lot of use made up our minds back on 3.7 and to be honest that version was kind of a train wreck that quickly got better.
1 Like
I fall into this one as well. I just got stuck on the jogging part since that was the current conversation 
This would be sweet. Not going to lie. I would probably still use my pendant to jog around, since I already have it. But I would have the webui pulled up on some type of screen just to see it visually
Maybe another way to put this is I am seeing people say things like, “just build a pendant and you will like it better” to new users. That is rough to hear for me. To me that sounds like, “this project is incomplete and you need to spend considerably more money to do it right”. (the cyd cost nearly as much as a jackpot and the m5 version is considerably more).
If we heard, try jason’s extension, or have you tried a $15 bluetooth joypad, more often it wouldn’t be a big deal but if a pendant has so many of you using it I do feel something must be lacking and I want the new users experience to be a great one, not something hurried to another expense.
That speaks more to the outstandingly good value of the jackpot. In reality a lot of users will spend more on the table it sits on, or the filament.
2 Likes
I made changes to Mitch’s tablet extension to support it being in a panel on the dashboard. So, if you added the tablet extension as a panel instead of a page, it just showed a dropdown to select the file and see the preview. The problem is, you can’t run it from there so you’d still have to run that file normally. I don’t use it. Last I checked, there were other changes made that broke folder selection.
There’s a lot of conversation here so I’m not quite sure how to respond. I think pendants are cool but unnecessary. I get that people like physical controls (although I don’t like combining those with touch controls). I get why people like the CYD pendant over the M5 dial version. The fact that the CYD and M5 dial use basically the same code makes the CYD one a bit weird, but I understand not wanting to maintain two completely separate code bases. It’s fair that a pendant make sense if wireless is an issue. This is also challenging because like all interfaces, there are a billion opinions and it’s hard to filter through the noise. Having a lot of options sounds great, but in reality the fewer options the better.
1 Like
Exactly, I am trying to filter it now to make an “official” option, and improve what we have now. That dam “use this network as is” is stopping me from making V3 the clear winner for UI that we could then keep tweaking to bend it to our will.
I just want it to be easier and better, and more clearly defined for our users, preferably without spending more money.
2 Likes
Rereading this topic. Seems to me my ideal option is a pc in my shop for CNC control. I don’t have wifi in my shop unfortunately. Its to far away from the house so STA mode wouldn’t work without a whole new mesh router setup, which I wouldn’t mind having but not today. lol
- Fixes my jog complaints via keyboard, xbox controller or something similar.
- Allows me to upload directly to the sd card and I don’t have to constantly move the sd card around.
- Allows me to install Estlcam and make quick changes right there at the machine.
Not the biggest fan of the webui although it is pretty easy and does work. However, I would be heavily considering wired with a gcode sender if I add a pc in the shop. Don’t see why not other than the cable annoyance.
If the cable breaks connection any time during a job then its a real pain to get started back. Can cause a lot of lost material.
I run my Primo via USB and Lightburn. But the control board doesn’t move on that machine like it does on the LR
That being said I just put one of these by my LR4 (for other reasons than what the pendant covers) and it works great for Estlcam/MM and for the webui.
2 Likes
I get this @CesarH. That is why I like my CYD pendant so much. It has a Hold and E-Stop button on the screen whenever a job is running.
Like @Jonathjon said, I really like using the jog wheel on the pendant to move to the starting position, or to jog around the machine.
1 Like
You know, I’ve largely dismissed gcode senders for a few reasons.
- I don’t have a PC in my shop and I don’t want to keep one out there because it gets hot. Although, I do have a laptop now that I could take out as needed.
- I just don’t want to run another cable. Although, with FluidNC I know that UGS supports wireless connection. And I’m moving cables around. That cable coming disconnected is a concern, but I’m sure that could be mitigated.
- I still consider running off the SD card to be the best option from a performance standpoint since FluidNC is in full control. However, people have been using gcode senders without issues, so perhaps that concern is overblown.
However, from the other side, there are a lot of reasons that gcode senders make a lot of sense. Right now, at least UGS and gSender look like good options.
- Resources - The resources to maintain and develop UIs and pendants on the FluidNC side just isn’t there. So, offloading that from FluidNC helps.
- Features - Since you’re running it on a full computer, it’s a lot easier to add features. Things like gcode visualization is much more practical there than on a microcontroller (unless done via hosted web page - like the tablet UI). Also, this piggy backs on #1 because you have more broad resources to help (FluidNC itself is highly specialized). And the functionality is more generic, so not specific to FluidNC.
- Don’t need Wifi (if using a cable connection)
For me, I have Wifi in my shop but not a PC. The WebUI does everything I need. I think things like gcode visualization is cool, but I haven’t seen anywhere that is actually helpful. The most likely reason is to see the starting position or maybe size, but I know that ahead of time. There’s really very little you actually need out an interface for normal everyday usage. Jogging via joysticks and dials is fun, but you spend very little time doing this. I say all this but I enjoy all these crazy unnecessary features but when we’re talking YBR, we need to be careful not to overcomplicate it.
1 Like
I think you have a big blind spot here, Ryan.
For you and me and a lot of the experienced millenial crowd who have been working on routers since we were teenagers, networking is a fact of life.
But some people don’t want to mess with it. They have past “trauma” around it and they are at best not willing to work on it or at worst experiencing anxiety whenever they have to type anything into a browser.
I would bet you there is a large fraction of your userbase that would give you double to put a wired, physical, no setup pendant in the box and they would love it.
Personally, I would push towards having jackpot v2 have a plug and play port for a pendant and get a pendant as an option in the box. I know you’re trying to get away from options. But I feel strongly that this would make the user experience better for a lot of people that don’t like dealing with networking or wifi or their phones while in their shop. This is the same crowd that is going to want physical buttons and can move a file easily using a thumb drive or sdcard.
If you prefer klipper and octoprint and run pihole at home, or you have a tight budget but willingness to figure it out, then you don’t buy the screen. AP mode is the YBR in this case. If you want to use that GPIO for something else, solder on a header or maybe make an RJ port to wire adapter.
6 Likes
I’m not sure about paying double, but when it came time to replace a faulty SKR board, I had the choice between using another SKR with a (wired) touch screen, or going wireless with a Jackpot and using a networked phone or tablet.
It was a pretty easy decision, I went with the wired choice. Not because I don’t understand networks or technology (my education and employment for the last two decades was in networks and technology). Rather I appreciated the reliability and simplicity of the touch screen, and didn’t want to have to regularly troubleshoot or fiddle with network or connection issues.
So IF I ever went the Jackpot route, I would probably look for a wired controller option, rather than going completely wireless.
But as someone else said, I’m a sample size of one, so maybe I’m a bit of an outlier…
1 Like
That’s exactly how I ended up with an MKS DLC32 and “fantastic” firmware for my laser cutter… I wanted a way to jog the machine without a computer (or other device)… so FluidNC on that board was a no go (and I don’t think I’d come across the other FluidNC ecosystem boards yet).
As it turns out, I never use the touch screen… mainly because it can only jog 10mm at a time, so it takes quite some time to go 6000mm.
Ideally, I’d have a joystick for it.
They’re still legitimate ways to run a machine.
You can even run FluidNC wirelessly this way.
It’s not recommended but it works. I’ve done it.
Very much this. I don’t mind even radically difficult networking setup. But some people want nothing to do with it.
This year at RMRRF there were a pair of participants that were looking first at the Compass and then at LR4. One said to the other about the LR4 “you have to set it up on WiFi and run it with your phone” the other replied “[explitive] that, Nope!” They started to walk away.
I mentioned you could use a sender or a pendant with them and ended up talking for a fair while with them.
Ryan, you want there to be the one true way and there just isn’t.
I can (and do) drive these machines using all of the options, and I see merit in all the use cases.
people’s knowledge and abilities plus experience and biases play into this.
For me, specific circumstances dictate the pro and con for these.
There’s also basic human factors. Someone with large hands or poor eyesight will hate hate hate either a tiny phone screen or the smallest pendant, but might find the larger pendant much more suitable for use.
Frankly all of the pendants feel a bit bush league to me in the enclosure realm. We had a forum member mill an enclosure- now that’s what I’d rather have: at some point I’m going to give that a try.
I hear the talk about how much better the newer FluidNC builds are at networking and I do see improvement.
But that improvement isn’t good to great, it’s more borderline crap to slightly better than average.
There are also much needed improvements in all the UIs web or pendant. More also in documentation about the crappy way some OS will abandon a non-internet network connection.
It sucks there isn’t an ESP-32 with Ethernet that can run FluidNC and that the ESP-32 doesn’t have more available UARTs.
With what we have available we can cover a bunch of use cases that span the majority and several of the larger niche use cases.
This is also true for the firmware itself. FluidNC is a great option for the majority. Klipper, Marlin, RRF, EstlCAM, etc all can run the LR and work for specific users for specific reasons. That’s exactly the same as the Pendant/WebUI discussion.
3 Likes
I usually hit the pause button on time, home the z axis and then move the xy to watever position i want (old firmware) and edit the gcode after if needed, some times i loose the sheet. When that happen wuth stacked sheets it hurts more
Im going to an m5 dial soon. Already bought the sixpack form a v1 user that went thc controller. That machine will get the surface tablet on that cnc cart+wired pendant
I think this is helping a lot. I am starting to understand the bigger picture.
More points of view will still be very helpful.
Since really taking a deeper look at these things I noticed that as well. Mine is not in an enclosure for this very reason…with that said I very much like to design stuff like that. Still need to understand the use cases more to know what it should look like. This is screaming for a 3D milled enclosure for sure.
I even poked around in the pendant UI to take a deeper look.
2 Likes