• The font on the PM GO DJ Media players is much larger than the tiny font on the small Stanton display.
  
• The Stanton Browse list does not retain the directory structure of the media, so you cannot use the media's directory structure to organize your tracks.  The media players in the PM GO DJ do retain the directory structure.  While I generally just have folders for artists, one could also use folders basically as crates, or any other structure you would like (such as a directory for "Halloween Songs").  Of course you can create playlists on the Stanton, but that would be tedious using the Stanton interface itself, or would require using other software on a PC.  Simply using the directory structure for crates or playlists may not be sophisticated, but it is simple and easy.
  
• Quickly moving through large numbers of tracks is actually just as easy or easier than on the Stanton. Quickly browsing the browse list on the Stanton using the jog wheel is not smooth, and suffers from lag.  You either have to scroll slowly enough for it to keep up, or risk significant overshoot.  As mentioned, the Media Players used for the PM GO DJ are fast and direct if you just keep mashing the >> or << buttons.
  
• The combination of the Media Players not stopping at the end of the track, and the Fader not totally silencing the opposite (faded away from) player leads to the next track playing in the background if you forget to stop it.  I think I did this for over half the tracks I played using the PM GO DJ.  I should state that the level of the undesired track is very low, and I don't know if it was noticeable to the audience or not when it happened.
  
• While the PM GO DJ has 2 mic inputs (as opposed the the Stanton's single mic input), they do not pass through, so they can only be used for recording.  A live mic input is really needed, and it would be nice if it had equalization.
  

•  Squelch is poorly implemented. Squelch is another basic function that has been correctly implemented in receivers since the vacuum tube days.  In the Malachite DSP SDR receiver, it seems to cause loud clicks and scratchy sounds if set near the signal level.  There is an NR threshold setting that may be an attempt to mitigate this, but it again, is another setting that other receivers don't have or seem to need, and it also does not actually seem to help.  
  Work is needed on the implementation of squelch.  It should have some hysteresis to eliminate rapidly turning on and off, and the squelch enabling or muting the audio should not cause a noise of it's own.  Even better would be a way to adjust the squelch while looking at the spectrum display, with an indication of the threshold level.  This would allow the user to set the squelch at the desired level above the noise floor.
  
•  Touch screen response time/Settings Panels switching:  The unit feels a little show in touchscren response.  While current firmwares support going from one panel to the other with out exiting the first panel, it takes a second or so to do so.  In the meantime the user does not know if his touch has been "registered" or not. Even going from one item in a given panel to another seems a tiny bit "laggy". I realize that this is not a gaming PC, but I am pretty sure that the STM32H7 is capable of better than it is now.  Just a little software optimization is necessary with respect to displaying the menus/settings panels.
  
•  Encoder Wheels:  I should mention that what follows is from 2 different radios, with different encoders, and different physical layouts of the encoder knobs.  There is not a 1 to 1 correlation between "clicks" or detents, and what is registered on the display.  When tuning, it will often skip one or more steps.  In some cases, it is necessary to position the wheel between detents to tune a desired frequency.  For example, if I am tuned to 118.000MHz, and I would like to set it to 118.200MHz, and the interval is set to 100KHz, it should take exactly 2 detents to get there.  I do understand how these wheels work, and I understand that Encoders with different numbers of detents (and different qualities) are sold.  I think that given the open hardware nature of this project, there needs to be additional software settings to allow matching of the particular encoders to the software, just as there is currently an encoder reversing option.  Essentially a encoder gain or sensitivity setting.
  
•  Excessive encoder clicks to move from one group of memories to the next in the "Band" panel.  It takes many clicks to move to the next, or previous panels.  Ideally, this would be a continuous smooth horizontal scroll, and could also be accomplished by dragging with a finger on screen;  However, I am not asking for this.  As I mentioned I realize that this is not a gaming PC, or a phone, but it should be possible to make going from one group to the next a little easier/quicker.
  
•  Power on, Power off.  There has been quite a bit of discussion on the Russian CQHam forums about turn on and turn off.  IMHO, both the 3 presses to turn on, and the Morse code when turned off are obnoxious. Mobile phone designers were faced with this issue many years ago, and have universally settled on a long press to turn on.  Neither the pre-built Chinese radio I have, nor the one in the case I made, are prone to being accidentally activated.
  
•  Spelling of the Word Standard.  The word "Standard" is spelled wrong in the Low Battery cut-off menu item (is: "Standart", should be: "Standard").
  

•  Ghost Frequencies and Reflections: There seem to be a lot of "Ghost" or phantom frequencies where no actual external signal is present, but a strong or overpowering signal is indicated on the Spectrum Analyzer, and nearby frequencies cannot be tuned.  Signals are often reflected across the range of the spectrum display. The worst seems to be at 60MHz, and all even (at least) harmonics - e.g. 120MHz, 240MHz, etc.  These appear with no antenna connected, appear on both radios, and across different firmwares.  An RF Explorer verifies that there is no actual strong signal at the affected frequencies. This is due to both internal clocks and signals, and due to the RF issues that follow.  Software Defined radios often have a clock shift function to move internal clocks and signals should they interfere with a desired signal.
  
•  Lack of RF AGC.  The unit has settings under the Audio Panel for AGC, but this is an Audio level (AF Gain) AGC only.  It does not prevent strong RF signals from over powering the receiver.  For example, I use mine to listen to aircraft radio (VHF AM). If I set the RF gains high enough to receive the tower, an aircraft flying near me totally saturates the receiver, and is illegible. RF AGC is a pretty basic function, and should be included in such a device.  It should be possible to implement RF AGC in the MSI001 totally in software -- in other words, in a future firmware.
  
•  Too many settings for various RF Gains.  Related to the above, what this unit does have is at least 5 RF Gain settings.  While access to these settings should be available to users wanting to explore the full capabilities of the MSI001, it is too complicated for a user that just wants to tune a frequency and listen, just like they would on a Tecsun or Grunding, etc.  While it is possible to get similar performance to commercial radios, it takes a lot of time messing with these settings to make the Malachite radio perform adequately. At the minimum, they should default to optimal settings for each frequency range and mode.
  
•  Use of the term "Bands" for memories.  "Bands" are defined groups or ranges of frequencies.  While not 100% the same every place, they are generally agreed upon world wide.  The AM broadcast band is from 560KHz to 1.6Mhz.  The short wave "band" is generally considered to be from either the beginning or end of the AM broadcast band up through 30MHz.  Within that range, there are Ham bands, which again may vary in specifics from country to country, but generally use the same terminology (such as 120m band, 40m band, etc.) The FM broadcast band is from 88 to 108MHz FM, and while station assignments and perhaps the exact edges of the band may vary from country to country, something close is generally followed world-wide. The aircraft band is accepted as 108-137MHz world wide.  This radio should not use that term for the Settings panel that allows the user to access memories that have no association with the more common use of the term "band".  "MEM" (Memory), would be a fine label for that button.
  
•  Saving All settings when a frequency is stored to a "Band".  All current settings are saved when a memory position is pressed and held. In general, this is a good idea, but certain settings should not be saved.  For example, Hi-Z or 50 Ohm depends on where the radio is being used, not the frequency tuned.  What if I want to use the radio in either a building (home), or hand-held, but I would like to listen to aircraft frequencies in both cases?  I must now save each frequency twice -- once with the appropriate settings (of which as mentioned there are a lot of) for a (perhaps very high gain) roof mounted antenna, and then again for handheld use with just a whip.  The antenna Hi-Z/50ohm setting, and probably some of the others, should not be saved with the frequency.
  

• It runs from the internal LiPo battery.  I went with the separate boost converter to go from battery voltage to the +5V.  The charger/display board handles charging and battery level only. Battery life seems to be pretty good.  As mentioned, it draws less than 0.25A.  With a 5000mah battery, it will last longer than any party I will ever get invited to.
  
• The issues with the cue switches was due to a short from soldering in very tight spaces.  2 soldering/wiring issues have been corrected.
  
• I have not fixed the issue of the fader not completely silencing the opposite channel. Since the fader (and the headphone cue/live mixer) are just wired as resistors in the line-level audio instead of actual stage gain controls, they are not going to fully cut off the audio.   More engineering is needed.  I am not sure how much more effort I want to put into this project.
  
• I turns out that the fader (linear potentiometer) board was not wired like I though it was.  Basically, both sides are wired in parallel on that board. The result of this was that it was not actually playing in stereo. I had to de-solder and re-work the board to make it 2 electrically independent pots.
  
• If you look closely, you can see that I re-taped one of the membrane keyboards, and that the other is sill coming up near the center.  I'll re-glue this when I have the time and patience to do so.
  
• Labels are not installed.  I have made some labels with Libre Office.  Again, I'll print, cut, and apply them when I have the time and patience to do so.
  

/dev/ttyACM0  /dev/ttyACM1 

sudo addgroup "$USERNAME" dialout 

grig -l 

228  Kenwood	    TS-480		     0.9.5  Untested 

grig -m 228 -r /dev/ttyACM0 -s 19200 

dfu-util -l  

dfu-tool attach