Several years ago, I created a list of ESXi versions with matching VM BIOS identifiers. The list is now complete up to vSphere 7.
Your Linux runs on a VMware VM, but on which ESXi version? Even without access to the host nor vCenter, you can see for yourself: run "dmidecode" and look at lines 10, 11 and 12. The BIOS release date, the address and the runtime size are unique for each ESXi version. Starting from 7.0 U3, no address and size is reported, and the "Phoenix Technologies LTD" name changed to "VMware, Inc.". Look up your result in the following table:
NB These DMI properties are set at boot time. Even if your VM gets live-migrated to a host running a different vSphere version, your VM will keep the values it got from the host it booted on. What you see is the vSphere version of the host your VM booted on. It is the VM power-on that matters, so a guest OS reboot will not regenerate the DMI properties. A guest OS shut down on the other hand will also power off the VM, and the next power-on will regenerate the DMI properties.
Your Linux runs on a VMware VM, but on which ESXi version? Even without access to the host nor vCenter, you can see for yourself: run "dmidecode" and look at lines 10, 11 and 12. The BIOS release date, the address and the runtime size are unique for each ESXi version. Starting from 7.0 U3, no address and size is reported, and the "Phoenix Technologies LTD" name changed to "VMware, Inc.". Look up your result in the following table:
ESXi version | BIOS release date | Address | Size |
---|---|---|---|
ESX 2.5 | 04/21/2004 | 0xE8480 | 97152 bytes |
ESX 3.0 | 04/17/2006 | 0xE7C70 | 99216 bytes |
ESX 3.5 | 01/30/2008 | 0xE7910 | 100080 bytes |
ESX 4 | 08/15/2008 | 0xEA6C0 | 88384 bytes |
ESX 4U1 | 09/22/2009 | 0xEA550 | 88752 bytes |
ESX 4.1 | 10/13/2009 | 0xEA2E0 | 89376 bytes |
ESXi 5 | 01/07/2011 | 0xE72C0 | 101696 bytes |
ESXi 5.1 | 06/22/2012 | 0xEA0C0 | 89920 bytes |
ESXi 5.5 | 07/30/2013 | 0xEA050 | 90032 bytes |
ESXi 6 | 09/30/2014 | 0xE9A40 | 91584 bytes |
ESXi 6.5 | 04/05/2016 | 0xEA580 | 88704 bytes |
ESXi 6.7 | 07/03/2018 | 0xEA520 | 88800 bytes |
ESXi 6.7 U2 | 12/12/2018 | 0xEA490 | 88944 bytes |
ESXi 7.0 | 12/09/2019 | 0xEA480 | 88960 bytes |
ESXi 7.0 U3 | 25/06/2021 | VMW71.00V.18227214.B64.2106252220 |
Comments
Nick from Deep Signal Studios suggested I get in touch with you. I have a LOT of sequences created with a GEM Bachmann WS2 workstation (that I no longer have) that I would really like to convert to MIDI. The files are in the GEM ALL format. Nick suggested i "beg" you to possibly build an app that will convery ws2 .all to .mid.
ANy help would be greatly appreciated!
Thank you so much for your time,
Gary Strickland
Nick must have found or remembered my blog post from 2014: http://virtwo.blogspot.com/2014/07/gem-ws2-midi-system-exclusive-structure.html . I discuss the process of converting the floppy .ALL format to the WS2 MIDI dump format. This was and is useful to people who have a WS2 without a working floppy drive. I later published my code on https://github.com/bertdb/gem-ws2-floppy-to-midi-dump/ so others can use it too. Unfortunately, the data structure of the 60KB memory dumps was never fully documented by GEM/Bachmann, nor successfully reverse-engineered by hobbyists, apart from the header information that contains the filename. A complete understanding of how voice, global, style and sequencer data were stored in memory would be necessary to convert the sequencer data to generic MIDI files. With a WS2 and a lot of free time, one could experiment with binary comparisons of dumps where e.g. only one voice parameter changed. This would reveal where that parameter is stored. However, for the sequencer data this is much, much more complex. It is impossible to generate two recordings with exactly one difference (note number, or velocity, or timing). Personally, I don't have a WS2 anymore either, so further experimentation is not going to happen on my side.
The only feasible way of converting .ALL to MIDI is: buying a second hand WS1/2/400 or WS2 MIDI workstation, loading the file, and playing it back, attached to a MIDI sequencer that can save the recording as .MID.
Kind regards,
Bert de Bruijn.
Bert.