MS-plus: are ZD2 from MS-plus compatible with G1Four/etc, and vice-versa

[mungewell]

The immediate answer is NO, it looks like Zoom have jeri-rigged the ID and Groups to be different between the series

$ python3 ../decode_effect.py -s B1_BOMBER.ZD2
0x07000010 : B1_BOMBER.ZD2, Bomber (v1.50 8.45%)

$ python3 ../decode_effect.py -s MS_BOMBER.ZD2
0x09000040 : MS_BOMBER.ZD2, Bomber (v1.10 8.45%)

The symbols listed in the ELF Code segment are very similar, and there are relatively small difference in the hexdump of the Code

$ diff  B1_BOMBER.ZD2.code.sym MS_BOMBER.ZD2.code.sym
28c28
< 0909210
---
> 0348410
39c39
< 0908410
---
> 0536810
70c70
< 0334410
---
> 2108010
105c105
< 0331610
---
> 1780410

These symbol name differences might just be as the ZD2 has been re-compiled. Looking at a 2nd effect (ZNR) it has completely different number symbols too...

[mungewell]

There are also some differences in the GroupName, although I am not too sure how these affect the actual operation. Maybe just used for ZT2 and grouping of ZIC in the library (MS-plus).

B1 FOUR/unzipped$ find . -name "*.ZD2.dump" -exec grep groupname {} \; | sort | uniq 
    groupname = u'AMP' (total 3)
    groupname = u'CABINET' (total 7)
    groupname = u'DELAY' (total 5)
    groupname = u'DRIVE' (total 5)
    groupname = u'DYNAMICS' (total 8)
    groupname = u'FILTER' (total 6)
    groupname = u'MODULATION' (total 10)
    groupname = u'REVERB' (total 6)
    groupname = u'SFX' (total 3)

ms-60b/decoded$ find . -name "*.ZD2.txt" -exec grep groupname {} \; | sort | uniq    groupname = 'BASS AMP' (total 8)
    groupname = 'DELAY' (total 5)
    groupname = 'DRIVE' (total 5)
    groupname = 'DYNAMICS' (total 8)
    groupname = 'FILTER' (total 6)
    groupname = 'MODULATION' (total 10)
    groupname = 'PITCH SHIFT' (total 11)
    groupname = 'PREAMP' (total 6)
    groupname = 'REVERB' (total 6)
    groupname = 'SFX' (total 3)
    groupname = 'SYNTH' (total 5)

[mungewell]

I just added some options which might help us...

$ python3 decode_effect.py --help
...
  --force-id FORCE_ID   Force the ID to a particular value (in Hex)
  --force-name FORCE_NAME
                        Force the Name to a particular string (max 11 chars)
  --force-gname FORCE_GNAME
                        Force the GroupName to a particular string (max 11 chars)
  -o OUTPUT, --output OUTPUT
                        output combined result to FILE

Something like :-)

$ python3 ../decode_effect.py --force-id 0x07000010 -o MOD_BOMBER.ZD2 MS_BOMBER.ZD2
Checksum Recalculated: 0x9966b592
$ python3 ../decode_effect.py -s MOD_BOMBER.ZD2
0x07000010 : MOD_BOMBER.ZD2, Bomber (v1.10 8.45%)

[mungewell]

Attempting to install MS effect on my (broken screen) G1Four...

$ python3 ../zoomzt2.py -U BOMBER.ZD2
Uninstalling effect: BOMBER.ZD2

$ cp MOD_BOMBER.ZD2 BOMBER.ZD2

$ python3 ../zoomzt2.py -I BOMBER.ZD2
Installing effect: BOMBER.ZD2
uploading...

$ python3 ../zoomzt2.py -R -w -s FLST_SEQ.ZT2
Group 1 : DYNAMICS
0x01000010 : COMP.ZD2, (v1.40 1)
...
Group 7 : SFX
0x07000020 : AUTOPAN.ZD2, (v1.40 1)
0x07000041 : LPROLL3S.ZD2, (v1.10 1)
0x07000050 : HOTSPICE.ZD2, (v1.30 1)
0x07000010 : BOMBER.ZD2, (v1.10 1)
...

The real question is whether I can actual load it into a patch, which is a little more difficult with the broken screen.

Current patch is empty

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
$ python3 ../decode_screens.py temp.bin
---
Effect: Bypass (Off)
 : 
 : 
 : 
 : 
---
Effect: Bypass (Off)
...

Convert ID to 7-bit hex

>>> a = 0x07000010
>>> print(hex(a & 0x7f))
0x10
>>> print(hex(a >> 7 & 0x7f))
0x0
>>> print(hex(a >> 14 & 0x7f))
0x0
>>> print(hex(a >> 21 & 0x7f))
0x38
>>> print(hex(a >> 28 & 0x7f))
0x0

and install to current patch (slot 0)

$ amidi -p hw:2,0,0 -S 'f0 52 00 6e 64 03 00 00 01 10 00 00 38 00 f7'

Decode screens changes, but does NOT display the name

$ python3 ../decode_screens.py temp.bin | head
---
Effect:  (Off)
 : 
 : 
 : 
 : 
---
Effect: Bypass (Off)
...

Downloading current patch does however

$ python3 ../zoomzt2.py -c temp.zptc
$ python3 ../decode_preset.py -s temp.zptc
Name: Empty     
Patch Volume: 100
Effect 1: 0x07000010
   Enabled: True
   Param 1: 0
   Param 2: 0
   Param 3: 0
   Param 4: 0
   Param 5: 0
   Param 6: 0
   Param 7: 0
   Param 8: 0

[mungewell]

If i revert the effect to the B1Four version, it does show on 'screens'.

$ python3 ../zoomzt2.py -U BOMBER.ZD2
Uninstalling effect: BOMBER.ZD2
$ cp B1_BOMBER.ZD2 BOMBER.ZD2 
$ python3 ../zoomzt2.py -I BOMBER.ZD2
Installing effect: BOMBER.ZD2
uploading...

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
$ python3 ../decode_screens.py temp.bin | head
---
Effect: Bomber (Off)
Decay : 50
Tone : 4
Mix : 30
ON/OFF : TRGGR
---
Effect: Bypass (Off)
 : 
 : 

This at least means the pedal is not 'broken' by uploading the wrong/MS-plus version.

Note: when forcing the ID, we only changes the number in the header of the ZD2, nothing within the Code segment as altered.

[mungewell]

Alternate approach, take the Code from the MS effect and jam it into B1 effect file. This means that all the other segments are 'as-was'...

$ python3 ../decode_effect.py -D MS_BOMBER.ZD2 -C -o MOD_BOMBER.ZD2 B1_BOMBER.ZD2
Checksum Recalculated: 0x438fb019
$ python3 ../decode_effect.py -s MOD_BOMBER.ZD2 
0x07000010 : MOD_BOMBER.ZD2, Bomber (v1.50 8.45%)

$ python3 ../zoomzt2.py -U BOMBER.ZD2 
Uninstalling effect: BOMBER.ZD2
$ cp MOD_BOMBER.ZD2 BOMBER.ZD2
$ python3 ../zoomzt2.py -I BOMBER.ZD2 
Installing effect: BOMBER.ZD2
uploading...

$ python3 ../zoomzt2.py -R -w -s FLST_SEQ.ZT2 | grep -A 5 SFX
Group 7 : SFX
0x07000020 : AUTOPAN.ZD2, (v1.40 1)
0x07000041 : LPROLL3S.ZD2, (v1.10 1)
0x07000050 : HOTSPICE.ZD2, (v1.30 1)
0x07000010 : BOMBER.ZD2, (v1.50 1)
Group 8 : DELAY

'Screens' works, and it even makes sound when triggered...

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
simon@the-void:~/zoom-zt2-sdw-github/MS_vs_B1Four$ python3 ../decode_screens.py temp.bin | head
---
Effect: Bomber (Off)
Decay : 50
Tone : 4
Mix : 30
ON/OFF : TRIGGER
---
Effect: Bypass (Off)
 : 
 : 

Now I am doubting myself, did I screw up process or trick myself?

$ md5sum *BOMBER.ZD2
ee15a464cc57161677ca42387c6085f4  B1_BOMBER.ZD2
5c6196b5e155841a44c193f291ac4061  BOMBER.ZD2
5c6196b5e155841a44c193f291ac4061  MOD_BOMBER.ZD2
c0743e090025b1bbd919ad9f3133fdef  MS_BOMBER.ZD2

And a fresh download from pedal

check$ md5sum *.ZD2
5c6196b5e155841a44c193f291ac4061  BOMBER.ZD2

And I even power cycled pedal to confirm that it does/still works after.... it did!

So does this mean that there's something OTHER than the Code segment that's preventing it working?

[mungewell]

There's not too much different; first 'ZDLF' segment and the 'final bytes'... I can quickly rule out the final bytes.

$ python3 ../decode_effect.py -D MS_BOMBER.ZD2 -C -F -o MOD_BOMBER.ZD2 B1_BOMBER.ZD2
Checksum Recalculated: 0x438fb019
$ python3 ../zoomzt2.py -U BOMBER.ZD2 
Uninstalling effect: BOMBER.ZD2
$ cp MOD_BOMBER.ZD2 BOMBER.ZD2
$ python3 ../zoomzt2.py -I BOMBER.ZD2 
Installing effect: BOMBER.ZD2
uploading...

$ python3 ../zoomzt2.py -R -w -s FLST_SEQ.ZT2 | grep -A 5 SFX
Group 7 : SFX
0x07000020 : AUTOPAN.ZD2, (v1.40 1)
0x07000041 : LPROLL3S.ZD2, (v1.10 1)
0x07000050 : HOTSPICE.ZD2, (v1.30 1)
0x07000010 : BOMBER.ZD2, (v1.50 1)
Group 8 : DELAY

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
$ python3 ../decode_screens.py temp.bin | head
---
Effect: Bomber (Off)
Decay : 50
Tone : 4
Mix : 30
ON/OFF : TRIGGER
---
Effect: Bypass (Off)
 : 
 : 

So it seems that's its the first segment, which isn't fully parsed by my scripts:

[mungewell]

So it looks like the Byte (or ULint32) after the checksum is a 'target pedal'... changing 0xA0 to 0x07 lets the original effect (with altered ID) run on the G1Four.

$ python3 ../decode_effect.py --force-id 0x07000010 -o MOD_BOMBER.ZD2 MS_BOMBER.ZD2
Checksum Recalculated: 0x9966b592
$ hexedit MOD_BOMBER.ZD2


$ python3 ../zoomzt2.py -U BOMBER.ZD2 
Uninstalling effect: BOMBER.ZD2
$ cp MOD_BOMBER.ZD2 BOMBER.ZD2
$ python3 ../zoomzt2.py -I BOMBER.ZD2 
Installing effect: BOMBER.ZD2
uploading...

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
$ python3 ../decode_screens.py temp.bin | head
---
Effect: Bomber (Off)
Decay : 50
Tone : 4
Mix : 30
ON/OFF : TRIGGER
---
Effect: Bypass (Off)
 : 
 :

check$ hexdump -C BOMBER.ZD2 | head
00000000  5a 44 4c 46 78 00 00 00  92 b5 66 99 07 00 00 00  |ZDLFx.....f.....|
00000010  00 00 00 00 01 00 00 00  00 00 00 00 00 00 00 00  |................|
00000020  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00000030  00 00 00 00 00 00 00 00  80 00 00 00 00 00 00 00  |................|
00000040  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00000050  00 00 00 00 00 00 00 00  01 31 2e 31 30 00 00 07  |.........1.10...|
00000060  10 00 00 07 42 6f 6d 62  65 72 00 00 39 b6 11 53  |....Bomber..9..S|
00000070  46 58 00 30 00 00 00 f1  b1 59 ff e5 00 00 00 00  |FX.0.....Y......|
00000080  49 43 4f 4e b6 00 00 00  42 4d b6 00 00 00 00 00  |ICON....BM......|
00000090  00 00 3e 00 00 00 28 00  00 00 18 00 00 00 1e 00  |..>...(.........|

[mungewell]

Scripted

$ python3 ../decode_effect.py --force-target 0x07 --force-id 0x07000010 -o MOD_BOMBER.ZD2 MS_BOMBER.ZD2

Whole process repeated for MS-plus' ZNR effect

$ python3 ../decode_effect.py --force-target 0x07 --force-id 0x01000040 -o MOD_ZNR.ZD2 MS_ZNR.ZD2
Checksum Recalculated: 0x37a89eee
$ python3 ../decode_effect.py -7 -s MOD_ZNR.ZD2
0x01000040 : MOD_ZNR.ZD2, ZNR (v1.10 4.35%)
40 00 00 08 00 

$ python3 ../zoomzt2.py -U ZNR.ZD2 
Uninstalling effect: ZNR.ZD2
$ cp MOD_ZNR.ZD2 ZNR.ZD2
$ python3 ../zoomzt2.py -I ZNR.ZD2 
Installing effect: ZNR.ZD2
uploading...

$ amidi -p hw:2,0,0 -S 'f0 52 00 6e 64 03 00 01 01 40 00 00 08 00 f7'

$ amidi -p hw:2,0,0 -S 'F0 52 00 6e 64 02 00 09 00 F7' -r temp.bin -t 2

1023 bytes read
$ python3 ../decode_screens.py temp.bin | head -n 15
---
Effect: Bomber (Off)
Decay : 50
Tone : 4
Mix : 30
ON/OFF : TRIGGER
---
Effect: ZNR (On)
Detect : EFXIN
Depth : 100
THRSHLD : 30
Decay : 0
---
Effect: Bypass (Off)
 :

Freaking Awesome!

[thammer]

Did you get the MS+ ZNR effect to run on your G1Four? :-o

[mungewell]

Yes, after adaption both MS-plus BOMBER and ZNR ran on my G1Four - it's screen is broken so a little awkward to control...

@thammer I'd be interested what value the MS-75CRD+ and MS-50+ use for the 'target' byte

./OPTCOMP.ZD2
00000000  5a 44 4c 46 78 00 00 00  90 97 bf 62 a0 00 00 00  |ZDLFx......b....|
                                               ^^
./MINIMARK.ZD2
00000000  5a 44 4c 46 78 00 00 00  c0 7e 45 47 80 00 00 00  |ZDLFx....~EG....|

I believe that this is a bit field, and the effects for a single pedal seem to have overlapping values. ie the true value for MS-60B+ is 0x80 and 0xA0 overlaps this.

[mungewell]

@thammer

Since you have the MS-50G+ and MS-70CDR+, can you please do a summary ('-s -m') print of the effects on each?

Like
https://github.com/mungewell/zoom-zt2/blob/master/zoom_fx_ms-plus/ms-60b/effects_sort.txt

[thammer]

Here are the summaries for MS-50G+ and MS-70CDR+.
ms50gp_effects_sort.txt
ms70cdrp_effects_sort.txt

[thammer]

Yes, after adaption both MS-plus BOMBER and ZNR ran on my G1Four - it's screen is broken so a little awkward to control...

@thammer I'd be interested what value the MS-75CRD+ and MS-50+ use for the 'target' byte

Here are the first 16 bytes from OPTCOMP.ZD2 from the MS-70CDR+:

5A 44 4C 46 78 00 00 00 81 32 AF 3D 90 00 00 00

[thammer]

... and this is the start of VIBRATO.ZD2 from the MS-50G+:

5A 44 4C 46 78 00 00 00 EF DF 14 57 90 00 00 00

[mungewell]

Thanks for the sorted lists, I've uploaded. Looks like IDs between MS-50G+ and MS-70CDR+ roughly match, but there is major 'collision' with MS-60B+ ID's... maybe this is deliberate to make sharing effects (as pedals/patches would be confused with duplicate IDs).

Here are all the ZD2 effects the GL7 knows about. I think that if this 'target' is a bit-field, it may be of required capability (more memory, double screen, expression pedal) rather than particular pedal model. None of these bits overlaps the MS-plus values...

$ find "A1 FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00
03 00 00 00  
06 00 00 00  
07 00 00 00  
$ find "A1X FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
03 00 00 00  
06 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "B1 FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
03 00 00 00  
06 00 00 00  
07 00 00 00  
$ find "B1X FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
03 00 00 00  
06 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "B3n" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
01 00 00 00  
03 00 00 00  
05 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "B6" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
08 00 00 00  
0f 00 00 00  
$ find "G11" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
04 00 00 00  
05 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "G1 FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
06 00 00 00  
07 00 00 00  
$ find "G1X FOUR" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
06 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "G3n" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
01 00 00 00  
05 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "G3Xn" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
01 00 00 00  
05 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "G5n" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
01 00 00 00  
05 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "G6" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
04 00 00 00  
05 00 00 00  
06 00 00 00  
07 00 00 00  
0f 00 00 00  
$ find "H8" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
03 00 00 00  
06 00 00 00  
07 00 00 00  
$ find "R20" -name "*.ZD2" -exec bash -c "hexdump -C '{}' | head -n1 | cut -c 48-60" \; | sort | uniq
02 00 00 00  
03 00 00 00  
06 00 00 00  
07 00 00 00

For example does '0x08' (in 0x0f) mean that the effect requires an expression pedal?