'PX7 200K Collection' shrinked to 30K uniques


When I first heard of the original Korg Kronos oriented 'dx7_200k_collection' project I thought that it would be interesting to try and purge it from all the inevitable duplicate Banks and Patches included.

This happened a few weeks before the release of Propellerhead's own version, the PX7-200k-Collection ReFill for their PX7 FM Synthesizer.


After reaching a point where I already had all duplicate SysEx Bank files removed and was going through each bank sets, performing the conversion to .repatch format using the indispensable Propellerhead's DX7-PX7 Patch Converter, I realized that the hardest part of this dups removal project was going to be the surprisingly high amount of cases that I had to decide which bank should be considered the "original" so I could remove all the corresponding duplicates from all the others.

Well, once PH released their 200K ReFill version, I thought that it really wasn't worth my time to revisit the project and go through the very time consuming process of doing it properly even though still thinking that it would be nicer to browse a collection of unique PX7 patches, even if similar sounding,  at least free of exact copy duplicates.

...so I let the project go to sleep, like so many that I have, waiting for better (longer) days :)

...

Recently, a sequence of (soon-to-be-deceased) PH Forum threads (mainly joeyluck) reignited my interest on this matter.

I decided to have a different approach and this time I would search the web for similar or at least sub-sets of this huge 200K collection, trying to identify the original sources of the included banks.

For this, the most useful resource was Bobby Blues DX7 pages which allowed me to better organize some of the more recognizable banks into some sort of historically meaningful folders.

While doing this, the same PH Forums (via Jamesville) referenced another project that achieved a unique Bank count (but not unique patch count), BlackWinny's DX7 collection.

These two allowed me to further break down the amount of unsorted banks that the 200K source provided.

...

Well, after some necessary iterations, here's beta 6 7 of what I like to call Version 1 of this Koshdukai's Slim 'PX7 200K Collection' ReFill project, with 30102 unique patches, obviously still including similar sounding ones:

"PX7 200K Collection (30102 Uniques) Koshdukai's Slim Version 1b6 ReFill"

(beta 7 changelog at the end of this post)

The final Version 1 release will happen once I'm sure there's no unwanted empty folders or anything weird with folder or file naming, etc.


So, Version 1 is basically my attempt to:
  1. remove duplicate Banks from the 200K: Achieved
  2. remove duplicate patches: Achieved
  3. provide an historical context of the Bank sources: Achieved where possible

Am I happy with it? Nope :) ...that's why I'm working on Version 2, keeping goals 1 and 2 but dropping the importance of the 3rd goal because it caused a lot of insignificantly small banks once removed of all the duplicates they had when compared with "higher hierarchy" sources (that's why the folders are numbered). Basically, in Version 2 I'm going to drop the "from the Web" folder completely and feed it into a better organized "Instruments" folder.

There's also a very critical issue that undermined Version 1 and will condition me working on Version 2: The DX7-PX7 Patch Converter has some bugs and limitations regarding particular chars used in the patch names which result in existing patches being skipped from the conversion process not being included in the final converted bank zip.
This is why there's some patches missing from known banks which I would like to see included in Version 2.

I'm going to wait for these issues to be fixed in the converter to start re-building that new project.
Until then, Version 1 will have to do :)


Needless to say, suggestions and comments pin-pointing issues with this beta are very welcome!

...

V1b7 changelog:
  • empty folders removed except the ones at "2 Other Cartridges\Bo Tomlyn\"
  • "7 from the Web\Atari\SYNTHESI\" fixed. (Converter issue)
  • "7 from the Web\Atari's SPECIAL\SPECIALS\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\103\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\138\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\176\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\228\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\245\" fixed. (Converter issue)
  • "7 from the Web\Big Mamma\298\" fixed. (Converter issue)
  • "7 from the Web\from 200K\2003\" fixed. (Converter issue)
  • "7 from the Web\from 200K\DX45\" fixed. (Converter issue)
  • "7 from the Web\from 200K\DX55\" fixed. (Converter issue)
  • "7 from the Web\from 200K\EBANK0\" fixed. (Converter issue)
  • "7 from the Web\from 200K\FULLCRC3\" fixed. (Converter issue)
  • "7 from the Web\Jeff Saxe\BANKP1\16\Fluegel 2" fixed. (Converter issue)
  • "7 from the Web\Jezreel's INCON\INCONI05\" fixed. (Converter issue)
  • "7 from the Web\Tim Conrardy\ORCHESTR\" fixed. (Converter issue)
  • "7 from the Web\Tim Conrardy\PERCUSSI\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-03\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-105B\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-106B\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-10C\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-20C\" fixed. (Converter issue)
  • "7 from the Web\Tim Garrett\TX7-79B\" fixed. (Converter issue)
  • small adjustments to the ReFill splash/icon image.
    •

There are still A LOT of 1 or 2 patch folders in Beta 7 that I'm thinking of simply discard their original Bank folders and put them in the higher folder. Still deciding if I should do that in Version 1.

Comments

  1. UnknownJanuary 22, 2015 at 7:43 AM

    I really needed this.. The original 200k link on the propsshop is dead... This makes life a whole lot easier..thx. :)

    ReplyDelete
    Replies
    1. KoshdukaiJanuary 22, 2015 at 8:02 AM

      oh, so it is! Weird. I guess it's a temporary glitch due to their new site deployment.

      Anyway, I'll keep working on this more compact version a little more, so it'll be available from here if nothing changes :)

      Delete
    2. KoshdukaiJanuary 23, 2015 at 5:21 AM

      Looks like the link to the PX7 200k ReFill is now fixed :)

      Now points to http://cdn.propellerheads.se/ReFills/PX7-200k-Collection.rfl
      at https://shop.propellerheads.se/product/px7-fm-synthesizer/

      Delete
  2. David ReeceFebruary 21, 2016 at 1:17 AM

    Hi Kosh, I just acquired a DX7 and I'm on the hunt for the original DX7 200k sysex patch library, but the only one I've been able to find is the the pX7 converted version - do you have the original?

    ReplyDelete
  3. KoshdukaiFebruary 21, 2016 at 8:34 AM

    Hi :)

    Well, I just did a web search and it looks like it's still in the same place, the Korg Kronos forum at http://korgpatches.com/patches/kronos/dx7_200k_collection
    with the download link http://korgpatches.com/download/a3Jvbm9z/7NTFkeDdfMjAwa19jb2xsZWN0aW9u.zip

    You'll get a lot of duplicate banks and patches within them, but you know that, right? :D

    Good luck with your new DX7! :)

    ReplyDelete

Post a Comment

Popular posts from this blog

Generic MIDI Channel aware Remote Codec+Map

Image
If you have a controller capable of sending MIDI messages on multiple MIDI Channels, like an M-Audio Axiom 's 4 Keyboard zones (so 4 MIDI Channels from the same keyboard) plus the Pads on Channel 10 or an Arturia BeatStep PRO that can send 2 sequences on their own MIDI Channel (1 and 2) and its Drum sequencer on MIDI Channel 10 plus the controller on, say, MIDI Channel 16, etc,  then you certainly felt the need for a MIDI Channel-aware type of Remote Codec like this one. Taking the BSP example, here's how to get its 4 sources of MIDI into Reason with this Remote Codec+Map. Add a Control Surface per MIDI source  with these 3 easy steps: 1)  Add a Control Surface from the " <Other> " Manufacturer and choose the correct " Generic Controller on MIDI Channel __ /< " Model with the MIDI Channel you'll need 2) Give it a proper name, so you know what's the real controller being used with this generic map (that you
Read more

Arturia SparkLE Remote Codec+Map

Image
Hey all :) I've been asked last week to share my Arturia SparkLE Codec so here it is :) (unofficial) Arturia SparkLE Remote Codec (and partial Map) V1.0.2 ReDrum: Added Drum Channel selection support through the Dial encoder Edit Flam with Select+Loop Button Flam Amount  with  Select+Tempo Encoder Shuffle On/Off through Loop Button Actually, this has been sitting on my Remote projects folder for 2 years now (and far from being alone in there, eh), waiting for me to complete the rest of the Remote Map, because my initial need for it was really just to control Kong, Dr.OctoRex and ReDrum, that's why those are really the ones that got some partial love on the included map. Early this year, due to the improvements added to the ReDrum remoteables s ince Reason 8.2 , I ended up making some editing to support the new ReDrum remoteables :) I'm hoping that anyone wanting to extend it or complete it has enough examples already in the partial map I'm
Read more

Random access to a CV values list

Just added this little CV tool in the free download section (on the right): "K's Basics - Random access to a CV values list" It's basically a way to access through CV indexing, a list of 3x32 "cells" of CV values (well, one of the list is within the Note range of CV values, but this example is enough to demo the concept). So, if you're ever in need of generating a specific sequence of CV values controlled by another CV source, this is one way to do it :) It's like having random or sequential access (in whatever direction you want) to a specific custom list of CV or note values, by using a CV controlled "cursor" (just connect the CV "cursor" source to the Rotary 1 CV input and tweak the Combinator modulation programmer Min/Max values to your needs). This device was enough for something I was testing, but I'm sure it can be optimized with some thought and more time. It won't work well in a real-time fast switching situation,
Read more