It is currently Fri Mar 29, 2024 6:59 am


DIES IRAE DIES ILLA

A discussion forum for anything even marginally Hauptwerk-related.
  • Author
  • Message
Offline

DomD

Member

  • Posts: 31
  • Joined: Sat Oct 24, 2020 11:15 am

DIES IRAE DIES ILLA

PostWed Feb 02, 2022 1:10 pm

Please, wait, Hauptwerk backs up....
Please wait, Hauptwerk ............
Please wait.................
Please.............I don't want to wait a so long time !!!!!!!!!!!!!!!!!

Could you, please, only propose something more simple, less time-consuming, less unbearable, .......recaching all samples (either you have 10 or 100) is getting always longer and making caches bigger.
I must say with regret that it's certainly the last time I accept to upgrade my software in such conditions if you don't solve this aspect of your upgrade system. . Enough is enough.

Of course, I 've to admit willingly that this software is only the best one..... when all is installed...... but it's also the champion to keep us waiting....

Please, don't discourage HW fans !
Thanks !

.
Offline
User avatar

mdyde

Moderator

  • Posts: 15446
  • Joined: Fri Mar 14, 2003 1:19 pm
  • Location: UK

Re: DIES IRAE DIES ILLA

PostWed Feb 02, 2022 1:53 pm

Hello Dom,

Thanks very much for upgrading, and sorry that regenerating caches does time some time. New versions only cause caches to be regenerated if there is a genuine need for it (because the cache format needed to change, which we try hard to avoid, given that it's time-consuming), but v7 does fall into that category. The caches don't actually change significantly in size between versions (although of course I appreciate that you might have more sample sets than you did last time you upgraded). Implementing a new mechanism to improve cache generation speed is logged as a high priority enhancement.

You can cancel a backup being made if you like, but I wouldn't recommend it unless you'd already made a recent backup.
Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.
Offline
User avatar

orgelton

Member

  • Posts: 28
  • Joined: Sun Aug 26, 2012 7:13 am
  • Location: Rhineland-Palatinate, Germany

Re: DIES IRAE DIES ILLA

PostWed Feb 02, 2022 2:34 pm

Dear Martin,

I too found the time for the recaching very annoying. It took me two full days just to wait from sample set to sample set to load the next sample and press the <Enter> key twice two answer the two questions. That was my weekend.

I asked quite some time ago to provide batch functionality so that we could automate the creation of the cache file. Perhaps you could think about a more elegant approach: you could set up a small editor page that reads in the names of all the existing sample sets and allows you to tick the ones you want to recache. I wouldn't mind running the system for a day and a night to do this task, as long as it didn't take up my time in such a tedious way.
Offline
User avatar

mdyde

Moderator

  • Posts: 15446
  • Joined: Fri Mar 14, 2003 1:19 pm
  • Location: UK

Re: DIES IRAE DIES ILLA

PostWed Feb 02, 2022 2:41 pm

Thanks, orgelton.

Yes -- your suggestion regarding a batch re-caching function is still logged as an enhancement request, and sorry we haven't implemented that one yet.
Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.
Offline

DomD

Member

  • Posts: 31
  • Joined: Sat Oct 24, 2020 11:15 am

Re: DIES IRAE DIES ILLA

PostWed Feb 02, 2022 5:15 pm

mdyde wrote:Hello Dom,

Thanks very much for upgrading, and sorry that regenerating caches does time some time. New versions only cause caches to be regenerated if there is a genuine need for it (because the cache format needed to change, which we try hard to avoid, given that it's time-consuming), but v7 does fall into that category. The caches don't actually change significantly in size between versions (although of course I appreciate that you might have more sample sets than you did last time you upgraded). Implementing a new mechanism to improve cache generation speed is logged as a high priority enhancement.

You can cancel a backup being made if you like, but I wouldn't recommend it unless you'd already made a recent backup.


Thanks Martin for your answer. I know that's not easy....but it could be a serious advantage. I noticed also that recached S-sets load slower than with HW6.
Offline

Theorbe

Member

  • Posts: 97
  • Joined: Sat Jan 11, 2020 7:10 am

Re: DIES IRAE DIES ILLA

PostWed Feb 02, 2022 7:50 pm

Deleted by author.
Last edited by Theorbe on Sat Jan 13, 2024 6:59 pm, edited 1 time in total.
Offline
User avatar

mdyde

Moderator

  • Posts: 15446
  • Joined: Fri Mar 14, 2003 1:19 pm
  • Location: UK

Re: DIES IRAE DIES ILLA

PostThu Feb 03, 2022 4:38 am

Thanks, Dom.

Theorbe wrote:Hi Martin / other users

Would it be worth my while adding a feature to my random detuning utility to automate the cache rebuild? The same automation and list extraction / processing techniques I'm already using for the detuning would seem to apply in this case.


Thanks very much, Andy. If you have the time an inclination, I expect some people would appreciate that.

Note that the Organ Configuration Wizard will pop up the first time that any given organ is loaded in v6. If the organ was last used in v5+ it will just prompt to confirm the audio quality settings, whereas if last loaded in v4 it would also show the wizard pane giving the option to reset or migrate audio routing, etc. Hence if sending keystrokes to automate it you'd potentially need to handle those cases, so as to send the appropriate number of 'next' wizard button keypresses (Return/Enter on Windows, although that keyboard shortcut doesn't seem to work in Qt wizards on macOS). You could perhaps read the OrganID [assuming the organ definition isn't encrypted], then read the version it was last loaded in via its organ-configuration XML file, then send the appropriate number of key-strokes to tab through the wizard screens when the wizard finally appears after loading an organ. However, that might not help much, since it wouldn't work for encrypted organs. Perhaps it would be possible and better to get the utility simply to monitor the windows that appear and then send the appropriate number of 'next' screen presses accordingly. It would be rather fiddly. Still, I suppose you would need to make the mechanism have some means of detecting when an organ has finished loading anyway, in order to know when it would be appropriate to initiate loading the next one.

Edit: P.S. You could conceivably try to find the OrganIDs for encrypted organs by searching Hauptwerk's logs for the INF:2157 messages with the appropriate organ definition filenames (assuming that the user had loaded the relevant organ in the not-too-distant past, since the logs get rotated and old events will eventually be purged from them, so that they don't grow indefinitely).
Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.
Offline

Theorbe

Member

  • Posts: 97
  • Joined: Sat Jan 11, 2020 7:10 am

Re: DIES IRAE DIES ILLA

PostThu Feb 03, 2022 1:50 pm

Deleted by author.
Last edited by Theorbe on Sat Jan 13, 2024 6:59 pm, edited 1 time in total.
Offline
User avatar

mdyde

Moderator

  • Posts: 15446
  • Joined: Fri Mar 14, 2003 1:19 pm
  • Location: UK

Re: DIES IRAE DIES ILLA

PostThu Feb 03, 2022 2:01 pm

Thanks, Andy.

Theorbe wrote:Regarding the organ config xml, could you confirm that it’s the value of “Control_CurrentHauptwerkVersion” that determines if the extra wizard pane appears (if <5)?


Yes -- that's the one.

Theorbe wrote:Might it be best to read the value of “lastodf” to associate the organ name with ID (which itself is contained in the organ config filename)?


Yes -- that's an excellent idea, that should work in all cases [provided that the organ was loaded in a version >=4.0.0, which was when that attribute was introduced, but I very doubt you need to worry about anybody having used v3 recently].
Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.

Return to General discussion

Who is online

Users browsing this forum: No registered users and 12 guests