Love this and your other video with spitfire. Appreciate you examining these libraries in a controlled manner and including technical details like the midi data, audio buffer, etc.
@Robert Lydecker, Thanks for the kind words. In order to make these videos 'scientific' the detailed settings are relevant. Now the tests reproducible, and viewers may come up with suggestions for improvement. Or avoid a few shortcomings.
@SampleModeling Thanks for the positive feedback. For a fair comparison one has to repeat these experiments every time an instrument update is being released. The type of EWI tests depends on the development approach: pure physical modeling vs. the hybrid approach (samples with modeling scripts).
Do you know if the SWAM sounds have been updated at all since this video was made? Just curious because there aren't many good examples online with more "classical" stuff
@elifarmer3815 Indeed, these videos are 3 years old now (Mar 2024), and a lot has changed since then. The current SWAM instruments are v3.7.2, with improved timbre modeling and better MIDI CC realtime control. It is on my to-do list to produce YT videos with more recent EWI examples of classical music playing, and do a comparison between SWAM and other modeled/scripted libraries. Thanks for asking.
@elifarmer3815 In case you are still interested I recently (April 25, 2024) uploaded a video entitled 'SWAM instruments playing Mozart, Schubert, Brahms and Ravel string quartet excerpts' that demonstrates the solo strings (violin, viola and cello) sound of version 3.7.3.
attack and big leap transitions definitely sound too synthetic on AM to me :( For legato and slow passages SM is soo good. Thanks for great review and comparison!
What did you mean by your "general trend" comment at the end? I was unsure what specific trend you were referring to - do you mean that the unrealistic aspect of SWAM legato is pronounced in the higher pitched instruments?
@The Unrestricted Knowing, I had to replay the video to find the 'general trend' statement at 7:38. My observations regarding timbre and 'natural' sound (significantly determined by the attack phase and transitions) when playing 'classical' music fragments become more outspoken for the lower register instruments in the SWAM brass family (the trombone in my view is the weakest, lacking natural sound when performing with EWI; trumpet is better when compared to the horn sound). On the other hand, be aware that this video from 2020 is based on the first SWAM Solo Brass V3 release, and now (July 2022) we are on V3.5. There have been many improvements in EWI response, and the sound engine has been updated. So I should redo these comparison videos, but I am waiting for the right moment (balanced and stable release, plus my time permitting). Does this answer the question?
@@FransAbsil Yes it certainly does. Thank you very much for taking the time. I am debating a purchase. I have a Leap Motion controller put away in the box. I am fascinated with these because of the potential expressiveness. My Cello v.1. seems pretty buggy (although it is still useful - one model is very good at doing staccato, which offers precise control beyond what other string libraries offer. Do you think the Audio Modeling Strings have improved, and do you think there are some instruments still that Sample Modeling excels at better than Audio Modeling?
@The Unrestricted Knowing I'll try and answer your questions. Yes all SWAM instruments have improved; the physical modeling sound is better in the recent V 3.5.1 (rel. Jul 14, 2022), controller configurations and response are more flexible for various controller types (I've watched demo videos with the Leap Motion controller). I am not aware of alternative modeled woodwinds; the weakest there I find the clarinet sound (and thus I will also use sample libraries with articulations, and keyboard/mouse input). Sample Modeling brass sound is better, but SWAM is catching up. I will not use SWAM brass in symphonic mockups in exposed orchestration, but certainly in other music types (I've used solo trumpet, trombone sections with sustained chords, lead horn and tuba lines). Most TH-cam SWAM brass demos focus on the jazz / pop sound, while overdoing glides and shakes, unnaturally excess vibrato (more clowning around with these instruments, than convincing classical music examples). The strings: well, Sample Modeling have very, very good (ensemble and solo) strings, but SWAM is getting close soundwise and outperforms with realtime controllers (the Kontakt scripts, I suspect, are the limiting factor for SM). I use the solo violin frequently (in fact today, Jul 31, 2022, I published a video that features a pair of SWAM violins), also the lower strings (except for the solo double bass, where I still haven't found a convincing classical sound). I have bought all SWAM products, because I believe in and will support their ambitious developments. Long story, but hopefully it makes sense in your selection process.
@@FransAbsil Hey, thank you for being so clear and concise and helping me choose. A lot of $ changes hands but few of us have the time to get as deep as you have gotten. Your thoughts mean a great deal. Have a great day!
The Aaron Venture Infinite Brass collection has been suggested to me before. In this set of videos I did compare the two model-based instrument bundles that I currently own. That might change in the future. Thanks for the feedback.
@alleyway3215 A few things to note: a) these comparison videos are becoming outdated by 2024. SWAM has progressed to release 3.8.2. b) the results are specific for AKAI EWI performance, and therefore negatively biased by the velocity values derived from the pressure measurement. These days I still find the SM Horn sound still better than SWAM (the latter requires most careful tonguing and legato control), but the SWAM realtime control wins. Thanks for the comment.
Neither of these is fully convincing to me yet, given a complex project to complete and lack of time to tweak. The SM library seems okay with slower material, and yet I feel compelled to wait for Audio Modeling to make some break through because I love the possibilities with controllers with a fully modeled approach - and I've been waiting awhile.
@cachelesssociety5187 Be aware that this comparison was made almost two years ago, and that your judgement may be affected by my selection of examples and my playing. It has relevance when using the Akai EWI controller. Audio Modeling SWAM is now (August 2023) at V3.7.2; the response to this specific wind controller has been adapted significantly. The modeling has improved although the brass instruments sound still has doubtful characteristics. I should redo the tests with the updates. I am always experimenting with updates and tweaking the presets. Besides each SWAM instrument requires a somewhat different playing style on EWI (like the real thing), so practice also is a relevant factor. Thanks for your comment.
@@FransAbsil Thanks for the detailed reply. I'm considering the "All In" - I bought the SM strings bundle - The horn and reeds always fascinated me as a modeling challenge and even though it's not quite there yet I'm intrigued that you've seen improvements.
Thanks for the video.... No comparison to do...audio modeling french horn is totally crap...whereas sample modeling horn is very warm, accurate and totally real...
@darwin3440 Thanks for sharing your observations. As I have responded before, be aware that this video was created 2 years ago, and that Audio Modeling are continuously updating the SWAM instruments. I have reported my EWI playing experience with the company directly shortly after doing these comparisons.
Love this and your other video with spitfire. Appreciate you examining these libraries in a controlled manner and including technical details like the midi data, audio buffer, etc.
@Robert Lydecker, Thanks for the kind words. In order to make these videos 'scientific' the detailed settings are relevant. Now the tests reproducible, and viewers may come up with suggestions for improvement. Or avoid a few shortcomings.
fantastic comparison.
@SampleModeling Thanks for the positive feedback. For a fair comparison one has to repeat these experiments every time an instrument update is being released. The type of EWI tests depends on the development approach: pure physical modeling vs. the hybrid approach (samples with modeling scripts).
Do you know if the SWAM sounds have been updated at all since this video was made? Just curious because there aren't many good examples online with more "classical" stuff
@elifarmer3815 Indeed, these videos are 3 years old now (Mar 2024), and a lot has changed since then. The current SWAM instruments are v3.7.2, with improved timbre modeling and better MIDI CC realtime control. It is on my to-do list to produce YT videos with more recent EWI examples of classical music playing, and do a comparison between SWAM and other modeled/scripted libraries. Thanks for asking.
@@FransAbsil Ok, awesome! Thanks for the reply :)
@elifarmer3815 In case you are still interested I recently (April 25, 2024) uploaded a video entitled 'SWAM instruments playing Mozart, Schubert, Brahms and Ravel string quartet excerpts' that demonstrates the solo strings (violin, viola and cello) sound of version 3.7.3.
attack and big leap transitions definitely sound too synthetic on AM to me :( For legato and slow passages SM is soo good. Thanks for great review and comparison!
You're welcome. I do so hope that the fully model-based virtual instruments will improve timbre-wise with every update.
What did you mean by your "general trend" comment at the end? I was unsure what specific trend you were referring to - do you mean that the unrealistic aspect of SWAM legato is pronounced in the higher pitched instruments?
@The Unrestricted Knowing, I had to replay the video to find the 'general trend' statement at 7:38. My observations regarding timbre and 'natural' sound (significantly determined by the attack phase and transitions) when playing 'classical' music fragments become more outspoken for the lower register instruments in the SWAM brass family (the trombone in my view is the weakest, lacking natural sound when performing with EWI; trumpet is better when compared to the horn sound). On the other hand, be aware that this video from 2020 is based on the first SWAM Solo Brass V3 release, and now (July 2022) we are on V3.5. There have been many improvements in EWI response, and the sound engine has been updated. So I should redo these comparison videos, but I am waiting for the right moment (balanced and stable release, plus my time permitting). Does this answer the question?
@@FransAbsil Yes it certainly does. Thank you very much for taking the time. I am debating a purchase. I have a Leap Motion controller put away in the box. I am fascinated with these because of the potential expressiveness. My Cello v.1. seems pretty buggy (although it is still useful - one model is very good at doing staccato, which offers precise control beyond what other string libraries offer. Do you think the Audio Modeling Strings have improved, and do you think there are some instruments still that Sample Modeling excels at better than Audio Modeling?
@The Unrestricted Knowing I'll try and answer your questions. Yes all SWAM instruments have improved; the physical modeling sound is better in the recent V 3.5.1 (rel. Jul 14, 2022), controller configurations and response are more flexible for various controller types (I've watched demo videos with the Leap Motion controller). I am not aware of alternative modeled woodwinds; the weakest there I find the clarinet sound (and thus I will also use sample libraries with articulations, and keyboard/mouse input). Sample Modeling brass sound is better, but SWAM is catching up. I will not use SWAM brass in symphonic mockups in exposed orchestration, but certainly in other music types (I've used solo trumpet, trombone sections with sustained chords, lead horn and tuba lines). Most TH-cam SWAM brass demos focus on the jazz / pop sound, while overdoing glides and shakes, unnaturally excess vibrato (more clowning around with these instruments, than convincing classical music examples). The strings: well, Sample Modeling have very, very good (ensemble and solo) strings, but SWAM is getting close soundwise and outperforms with realtime controllers (the Kontakt scripts, I suspect, are the limiting factor for SM). I use the solo violin frequently (in fact today, Jul 31, 2022, I published a video that features a pair of SWAM violins), also the lower strings (except for the solo double bass, where I still haven't found a convincing classical sound). I have bought all SWAM products, because I believe in and will support their ambitious developments. Long story, but hopefully it makes sense in your selection process.
@@FransAbsil Hey, thank you for being so clear and concise and helping me choose. A lot of $ changes hands but few of us have the time to get as deep as you have gotten. Your thoughts mean a great deal. Have a great day!
Thanks for this, any chance you could do the same with Aaron venture’s infinite brass?
The Aaron Venture Infinite Brass collection has been suggested to me before. In this set of videos I did compare the two model-based instrument bundles that I currently own. That might change in the future. Thanks for the feedback.
As a brass player, Sample Modeling gets the prize.
@alleyway3215 A few things to note: a) these comparison videos are becoming outdated by 2024. SWAM has progressed to release 3.8.2. b) the results are specific for AKAI EWI performance, and therefore negatively biased by the velocity values derived from the pressure measurement. These days I still find the SM Horn sound still better than SWAM (the latter requires most careful tonguing and legato control), but the SWAM realtime control wins. Thanks for the comment.
Neither of these is fully convincing to me yet, given a complex project to complete and lack of time to tweak. The SM library seems okay with slower material, and yet I feel compelled to wait for Audio Modeling to make some break through because I love the possibilities with controllers with a fully modeled approach - and I've been waiting awhile.
@cachelesssociety5187 Be aware that this comparison was made almost two years ago, and that your judgement may be affected by my selection of examples and my playing. It has relevance when using the Akai EWI controller. Audio Modeling SWAM is now (August 2023) at V3.7.2; the response to this specific wind controller has been adapted significantly. The modeling has improved although the brass instruments sound still has doubtful characteristics. I should redo the tests with the updates. I am always experimenting with updates and tweaking the presets. Besides each SWAM instrument requires a somewhat different playing style on EWI (like the real thing), so practice also is a relevant factor. Thanks for your comment.
@@FransAbsil Thanks for the detailed reply. I'm considering the "All In" - I bought the SM strings bundle - The horn and reeds always fascinated me as a modeling challenge and even though it's not quite there yet I'm intrigued that you've seen improvements.
Thanks for the video.... No comparison to do...audio modeling french horn is totally crap...whereas sample modeling horn is very warm, accurate and totally real...
@darwin3440 Thanks for sharing your observations. As I have responded before, be aware that this video was created 2 years ago, and that Audio Modeling are continuously updating the SWAM instruments. I have reported my EWI playing experience with the company directly shortly after doing these comparisons.