Page 1 of 1

HW 6 upgrade questions

PostPosted: Sun Nov 29, 2020 6:50 pm
by johnstump_organist
Two questions...
1. So if I upgrade to 6 and find I don't have enough computer power to take advantage of the new 96khz options, either together or singularly, will 6, running at 48khz, still operate on the system that 5 is operating on without major glitches?
2. If I can't use the 96khz options until I get a new computer, is there any noticeable improvement in sound quality over 5, in running 6 at 48 khz?
Thanks in advance,
John

Re: HW 6 upgrade questions

PostPosted: Sun Nov 29, 2020 10:14 pm
by JulianMoney-Kyrle
John,

I believe the sound engine has been rewritten, so I think probably there is an improvement in sound quality even without using the 96KHz option. The other option has something to do with re-pitching the samples (which apparently is necessary almost all the time, not just for retuning or using the modelled tremulant) and is separate from processing at 96 KHz, though my understanding is that each one has its own overhead in terms of polyphony, which is halved by using either and quartered by using both. Martin used to maintain that polyphony was the single biggest factor in determining the realism of a HW installation, but how polyphony rates against the imporvements in audio processing I really don't know.

Certainly the upgrade from 4.2 to 5 was noticeable and took me by surprise as I wasn't expecting it. The upgrade from 5 to 6 is also noticeable, particularly when playing full organ when I can hear a greater clarity and less distortion in the upperwork.

Re: HW 6 upgrade questions

PostPosted: Mon Nov 30, 2020 5:36 am
by mdyde
Hello John,

Thanks very much for the interest in v6.

There are two separate options in v6 that affect audio quality:

- The 'higher definition pitch shifting' organ preference.
- The option to run the audio engine at 96 kHz.

Enabling either one very roughly halves the polyphony that the computer is likely to be able to manage (x25% in total if both are enabled). Both options are beneficial, but the 'higher definition pitch shifting' option is the one that makes the biggest difference in most cases. (See also: viewtopic.php?f=1&t=19209#p144310 .)

johnstump_organist wrote:1. So if I upgrade to 6 and find I don't have enough computer power to take advantage of the new 96khz options, either together or singularly, will 6, running at 48khz, still operate on the system that 5 is operating on without major glitches?


If you enable neither of the two options then there will be no difference in performance (polyphony/CPU requirements), compared to v5. Hence there is no performance disadvantage by upgrading to v6.

johnstump_organist wrote:2. If I can't use the 96khz options until I get a new computer, is there any noticeable improvement in sound quality over 5, in running 6 at 48 khz?


The above two options give the main improvements to audio quality (improved perceived clarity, due to considerably lower distortions). Aside from those there are some minor sonic and fidelity improvements, but they're likely to be much less noticeable than those from the two options above.

JulianMoney-Kyrle wrote:I believe the sound engine has been rewritten, so I think probably there is an improvement in sound quality even without using the 96KHz option.


Just for the record: the audio engine was rewritten for v5, and only changed for v6 to support the two new options above (and some other minor sonic tweaks).

JulianMoney-Kyrle wrote: Martin used to maintain that polyphony was the single biggest factor in determining the realism of a HW installation


I've certainly said previously that a major benefit of using wet samples for organs (recorded from constant microphone positions) is that they can reproduce the organ/building's natural acoustics very effectively, giving a realistic perceived spatial impression, and resulting perceived clarity. Wet sample sets (especially large, and/or very wet, ones) often need a lot of polyphony. Hence being able to achieve high polyphony is often a necessary part of achieving realism with wet sample sets (unless one sticks to using smaller and/or less-wet sample sets, and/or smaller registrations, of course).