MARF 248r source code available

Discussing some incredible modules that don't quite fit into the other forum categories.

Moderators: Joe., lisa, luketeaford, Kent

Post Reply
User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Wed Feb 05, 2020 11:59 am

OK, real fix #1 is in. The Infamous Gate1/Gate2 LED reversal is fixed in the software. My hardware has a fix for this, but so did this software so it was back to being reversed again. We probably need some kind of board versioning to tell the software to reverse or not. For now I’ll let the software be correct, Gate1 firing lights Gate1 LED. Gate2 firing lights Gate2 LED. We’ll see what happens.

-s

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1484
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: MARF 248r source code available

Post by cygmu » Thu Feb 06, 2020 5:17 am

stevenb wrote:
Wed Feb 05, 2020 11:59 am
OK, real fix #1 is in. The Infamous Gate1/Gate2 LED reversal is fixed in the software. My hardware has a fix for this, but so did this software so it was back to being reversed again. We probably need some kind of board versioning to tell the software to reverse or not. For now I’ll let the software be correct, Gate1 firing lights Gate1 LED. Gate2 firing lights Gate2 LED. We’ll see what happens.

-s
Sooo many thanks for your work on this. This is real progress and very promising. I am actually going to build up my MARF PCB now :)

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Thu Feb 06, 2020 7:18 am

Awesome news - you ROCK :sb: :sb: :sb:
Buchla format cases & parts - www.samodular.com

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Sat Feb 08, 2020 10:36 pm

Look who showed up today :yay:


maxl0rd
Common Wiggler
Posts: 62
Joined: Mon Aug 05, 2013 1:51 pm

Re: MARF 248r source code available

Post by maxl0rd » Sun Feb 09, 2020 10:18 am

stevenb is doing a bang up job on this so far. We have a good chance at getting the most egregious bugs ironed out for v2 owners.

We’re tracking the outstanding bugs as issues on the public github project.

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Sun Feb 09, 2020 12:36 pm

maxl0rd wrote:
Sun Feb 09, 2020 10:18 am
stevenb is doing a bang up job on this so far. We have a good chance at getting the most egregious bugs ironed out for v2 owners.

We’re tracking the outstanding bugs as issues on the public github project.
And big thanks to maxl0rd for his GIT work and organizational skills. It takes a village :tu:

Disclaimer:
Don't read too much into what I show working. There is still a lot of work to do to make sure that it is working correctly. Think of what I share as "Proof that it CAN work", not proof that it is finished. I'm just a little excited and want to share it with you all...

That said:



Steven

jimfowler
Veteran Wiggler
Posts: 550
Joined: Mon Feb 22, 2016 1:13 pm
Location: New York
Contact:

Re: MARF 248r source code available

Post by jimfowler » Sun Feb 09, 2020 1:07 pm

oh boy oh boy oh boy!!!

Electrostatic
Veteran Wiggler
Posts: 544
Joined: Sat Aug 04, 2012 4:45 pm
Location: UK

Re: MARF 248r source code available

Post by Electrostatic » Mon Feb 10, 2020 7:18 am

Damn. :eek:
My main two problems with the MArF (v2) is the clocking issue (i.e. no gates out on the clocked channel) and the advance issue (i.e. it doesn't work at all). And it looks like both are possible to fix.
This is incredible news, thank you very much for the hard work stevenb.
I'd offer to help with testing but I'm not entirely sure I know how to update the source code. Will read up on it, I'm going to have to anyway to get your finished code installed one day, so would like to help you get to that point if I can.

Cheers. :banana:

User avatar
Siren
Common Wiggler
Posts: 56
Joined: Fri Sep 16, 2011 6:29 am

Re: MARF 248r source code available

Post by Siren » Mon Feb 10, 2020 1:08 pm

This is fantastic! Thank you so much for doing this.

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Mon Feb 10, 2020 5:15 pm

Will implement the STM32 on PCB2 including a ground plane asap. Lets see if that helps getting rid of the jitter.
So no separate STM board on the next revision. PCB2 will come pre-soldered with the STM32.
Buchla format cases & parts - www.samodular.com

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Tue Feb 11, 2020 2:34 pm

weedywhizz wrote:
Mon Feb 10, 2020 5:15 pm
Will implement the STM32 on PCB2 including a ground plane asap. Lets see if that helps getting rid of the jitter.
So no separate STM board on the next revision. PCB2 will come pre-soldered with the STM32.
:yay: :tu:

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Tue Feb 11, 2020 2:50 pm

stevenb wrote:
Tue Feb 11, 2020 2:34 pm
weedywhizz wrote:
Mon Feb 10, 2020 5:15 pm
Will implement the STM32 on PCB2 including a ground plane asap. Lets see if that helps getting rid of the jitter.
So no separate STM board on the next revision. PCB2 will come pre-soldered with the STM32.
:yay: :tu:
I decided to not implement the STM32 to the board yet but changed PCB2 to 4-layers with a ground plane and cleaned up some traces.
Also added a ground plane to the STM32 board. I'd like to investigate the issue and implementing the STM32 doesn't really help. Will the ground plane only on the STM32 do the trick ? If not, I'll try PCB2 with ground plane.
Ground seems flaky. I just added that 0.01uF cap to the CV output on stage 1 by using a nearby ground point. Noise was only reduced a tiny bit. I wired the cap directly to ground on the busboards and connected the other end of the cap to CV out on stage 1 and wow - HF noise was greatly reduced.

New files for PCB2 and STM32 have already been sent to the factory so hopefully I'll have something to test here in about a week.
Buchla format cases & parts - www.samodular.com

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Tue Feb 11, 2020 3:27 pm

weedywhizz wrote:
Tue Feb 11, 2020 2:50 pm
stevenb wrote:
Tue Feb 11, 2020 2:34 pm
weedywhizz wrote:
Mon Feb 10, 2020 5:15 pm
Will implement the STM32 on PCB2 including a ground plane asap. Lets see if that helps getting rid of the jitter.
So no separate STM board on the next revision. PCB2 will come pre-soldered with the STM32.
:yay: :tu:
I decided to not implement the STM32 to the board yet but changed PCB2 to 4-layers with a ground plane and cleaned up some traces.
Also added a ground plane to the STM32 board. I'd like to investigate the issue and implementing the STM32 doesn't really help. Will the ground plane only on the STM32 do the trick ? If not, I'll try PCB2 with ground plane.
Ground seems flaky. I just added that 0.01uF cap to the CV output on stage 1 by using a nearby ground point. Noise was only reduced a tiny bit. I wired the cap directly to ground on the busboards and connected the other end of the cap to CV out on stage 1 and wow - HF noise was greatly reduced.

New files for PCB2 and STM32 have already been sent to the factory so hopefully I'll have something to test here in about a week.
I noticed that too. If I grabbed ground right where it entered PCB2 HF noise was reduced

-s

maxl0rd
Common Wiggler
Posts: 62
Joined: Mon Aug 05, 2013 1:51 pm

Re: MARF 248r source code available

Post by maxl0rd » Tue Feb 11, 2020 4:52 pm

Adding caps to low pass the noise out of the DAC outputs is probably a decent kludge for all the modules out there, but will add some slew, right?

User avatar
Peake
I'm in ur DIY. Filling cases with Buchla
Posts: 6818
Joined: Fri Nov 07, 2008 10:00 pm
Location: Loss Angeles

Re: MARF 248r source code available

Post by Peake » Tue Feb 11, 2020 5:39 pm

-This is fantastic news
-Has it been made private so hostile forces who have teased an interest in cloning Buchla cannot download and monetize it
This is not the place I'd imagined it to be.

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Tue Feb 11, 2020 5:44 pm

Peake wrote:
Tue Feb 11, 2020 5:39 pm
-This is fantastic news
-Has it been made private so hostile forces who have teased an interest in cloning Buchla cannot download and monetize it
You mean Mehringer leechers ? :hihi: This sucker will keep ripping off everything he can find.
Good point ! We definitely have to think about that :mrgreen:

Btw: wasn't his intention to clone Suzanne Ciani ?

I can see a lot of people step in for the Devil Fish guy. I have never seen that before. But still, most customers don't care. Cheap stuff for cheap people.
Getting something build locally for sure eats more cash. Something people won't understand.
Yes I do clone Buchla as well, I'm a one-man-show and I'm trying with guys like Steven and Dave to fix the stuff. Not a cloning empire at all.
Now back to the topic :mrgreen:
Last edited by weedywhizz on Wed Feb 12, 2020 3:10 am, edited 2 times in total.
Buchla format cases & parts - www.samodular.com

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Tue Feb 11, 2020 8:20 pm

maxl0rd wrote:
Tue Feb 11, 2020 4:52 pm
Adding caps to low pass the noise out of the DAC outputs is probably a decent kludge for all the modules out there, but will add some slew, right?
If the planets align and the cap value is chosen wisely you may or may not notice the slew, but there would be some yes.
Then again all things are relative. Is a slight amount of slew worse than an audible warble?

-s

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Tue Feb 11, 2020 8:21 pm

Peake wrote:
Tue Feb 11, 2020 5:39 pm
-This is fantastic news
-Has it been made private so hostile forces who have teased an interest in cloning Buchla cannot download and monetize it
Oh my, here comes reality crashing in :sstorm:

Hardware can be copied by inspection. Can't hide it.
Software is another issue. Should I not expose the changes I'm making in software and just supply an executable? Is the modified software mine to do that? Interesting questions. I'm doing this for the community not some corporation..

-s

architexture
Learning to Wiggle
Posts: 40
Joined: Fri Nov 30, 2018 2:59 pm

Re: MARF 248r source code available

Post by architexture » Tue Feb 11, 2020 8:42 pm

> Should I not expose the changes I'm making in software and just supply an executable?

That's entirely up to you, however a good dev could likely do the same. The original is also open sourced so truly software only get's better when there are more eyes upon it, and when other people can contribute.

By locking things down in a complied source only you limit the ability of others to improve upon the changes you made, which were improvements on changes that others had made, etc etc......

As a software engineer I would say that releasing your branch of the code is a viable asset to many and would strongly encourage it, especially since your version is based off of an entirely open source version.

Again, that's your call but unless you have some proprietary trade secret that you are attempting to patent etc..... open-source is the way to go.

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Tue Feb 11, 2020 10:37 pm

architexture wrote:
Tue Feb 11, 2020 8:42 pm
> Should I not expose the changes I'm making in software and just supply an executable?

That's entirely up to you, however a good dev could likely do the same. The original is also open sourced so truly software only get's better when there are more eyes upon it, and when other people can contribute.

By locking things down in a complied source only you limit the ability of others to improve upon the changes you made, which were improvements on changes that others had made, etc etc......

As a software engineer I would say that releasing your branch of the code is a viable asset to many and would strongly encourage it, especially since your version is based off of an entirely open source version.

Again, that's your call but unless you have some proprietary trade secret that you are attempting to patent etc..... open-source is the way to go.
Thanks for the lovely speech. Seriously, you are 1000% correct. It was my intent all along to give my work back to the community. I was just voicing my concerns. I wish I had a proprietary trade secret, but I do not.

I am fortunate to have the spare time as well as experience to donate to this effort. Not only have I increased my knowledge, but I have met some truly awesome people. I am honored to be part of this great community and it's people who give freely. The experience has been incredible. It would not have happened if somebody didn't let the code fly free... which I will do when the features are working and it's cleaned up.

A special thanks go to Steffen, Kyle, Jim, Max, and Dave to name a few. We're only in the middle of the project with miles to go before I sleep.

I think I better dance now... :bananaguitar: :nana: :wookie:

Steven
Last edited by stevenb on Wed Feb 12, 2020 1:20 pm, edited 1 time in total.

User avatar
Peake
I'm in ur DIY. Filling cases with Buchla
Posts: 6818
Joined: Fri Nov 07, 2008 10:00 pm
Location: Loss Angeles

Re: MARF 248r source code available

Post by Peake » Wed Feb 12, 2020 12:20 am

I truly hope no one interpreted what I said as "stop" because that's absolutely not the case...just don't want to bastards to monetize all this hard and beautiful work, want it to only be in the hands of the community!
This is not the place I'd imagined it to be.

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Wed Feb 12, 2020 3:11 am

Peake wrote:
Wed Feb 12, 2020 12:20 am
I truly hope no one interpreted what I said as "stop" because that's absolutely not the case...just don't want to bastards to monetize all this hard and beautiful work, want it to only be in the hands of the community!
Fully agree.
Buchla format cases & parts - www.samodular.com

User avatar
stevenb
Common Wiggler
Posts: 141
Joined: Wed Dec 18, 2013 11:16 pm
Location: Nirvana Silly, California

Re: MARF 248r source code available

Post by stevenb » Fri Feb 21, 2020 6:07 pm

Well, here we are. The first publicly available alpha build of MARF firmware v2.5a6. It is only for v2 MARF hardware. Here's a list of what was fixed:

1) The advance switch works now. It respects loop settings and Programmed Outputs fire.
2) Pogrammed Outputs did not fire on a stage where there was a STOP programmed when the MARF was externally clocked. This now works properly.
3) Calibration was totally broken. It now appears to work properly.

I need people to test this and make sure those features work properly. Pay particular attention to LED status lamps and stuff like that. I wouldn't use it for mission critical applications, but do what you want. I'm going to attach the firmware file to this post. I hope that's appropriate.

You may still have to do the hokey pokey to flash your MARF with the v2.0 firmware. See Don Brown's MARF page for directions.
https://modularsynthesis.com/roman/buch ... 48marf.htm . Basically it's this:
Finally I found one hint about pulling BOOT0 high. Boot0 is pin 60 and tied low through a 10K resistor. Vdd is pin 64 so I simply jumper the two while programming. Programming made no difference.
This is caused by a conflict with DIP switch #3 which I forced to "On" in the code. This will be addressed at a later time.With this version of firmware you will have no problems with re-flashing.

Feedback is appreciated. Thanks to everyone for the support. There are more fixes to come.

Enjoy.

-s

MARF_v2.5a6.hex.zip
You do not have the required permissions to view the files attached to this post.

User avatar
batchas
Super Deluxe Wiggler
Posts: 4617
Joined: Wed Nov 09, 2011 2:51 pm

Re: MARF 248r source code available

Post by batchas » Sat Feb 22, 2020 2:29 am

stevenb wrote:
Fri Feb 21, 2020 6:07 pm
Well, here we are. The first publicly available alpha build of MARF firmware v2.5a6. It is only for v2 MARF hardware. Here's a list of what was fixed
I realized last week it was for v2 when I did read in detail about the issues mentionned in this thread.
batchas wrote:
Wed Feb 05, 2020 8:11 am
I'll test it too and report back when it's ready. Thanx a lot :tu:
Means I won't test it, as I don't have a weedywhizz version.
-----------------------------------
Batchas website
Bandcamp page

weedywhizz
Super Deluxe Wiggler
Posts: 1141
Joined: Wed Oct 20, 2010 3:51 pm
Contact:

Re: MARF 248r source code available

Post by weedywhizz » Sat Feb 22, 2020 2:35 am

batchas wrote:
Sat Feb 22, 2020 2:29 am
stevenb wrote:
Fri Feb 21, 2020 6:07 pm
Well, here we are. The first publicly available alpha build of MARF firmware v2.5a6. It is only for v2 MARF hardware. Here's a list of what was fixed
I realized last week it was for v2 when I did read in detail about the issues mentionned in this thread.
batchas wrote:
Wed Feb 05, 2020 8:11 am
I'll test it too and report back when it's ready. Thanx a lot :tu:
Means I won't test it, as I don't have a weedywhizz version.
Its for the (through-hole) EMS version 2.1 and my version. I remember that Roman shipped those faulty 2.0 and months later shipped 2.1 to those who got v2.0.
V2.0 was faulty as hell and had missing traces. So actually this is for rev 2.1.
Buchla format cases & parts - www.samodular.com

Post Reply

Return to “Buchla, EMS & Serge”