TX16Wx Software Sampler 2.0 Beta Released - Announching TX16Wx Professional (Updated 2012-10-19)

[sinkmusic]

I'm buying it soon
I wish it could read a bit more formats, so i wouldn't need any other sample player (well, i'm not much a smaple guy, except for the ones i create myself, but i have a couple of multilayered electric pianos and stuff, and i wish i don't need several sample players to read them).
SoundFont is great, but it seems that Gig is also rather popular, isn't it ?
In fact, i wouldn't expect tx16w to read 1:1 any sample format, but if at least it was able to read more formats with proper sample mapping (keymapping, velocity mapping) and proper ADSR, this would be way enough for my needs

[elcallio]

Thank you.
I am very glad to see such a fast rate of updates with a careful listening to user requests

On question abvout sampling : how do you select the sampling rate ? Will it record the audio stream according to the soundcard's rate, the host's rate ?
And if/when you will add FALC support, will it be possible to record in FLAC ?

Sampling is done in host rate, a plugin cannot change this. I will eventually add post-resampling support so you can downsample files as well.
FLAC support will include save capability as well, so yes.

I have a question regarding this... Let's say that someone creates an instrument that is sampled in 24-bit 96khz, but I am running my host at 16-bit 44.1khz. Will this be a problem in the current versions? I ask, because you say that you will 'eventually' adding support for post-resampling and downsampling. How are instruments currently going to be kept compatible?

--Sean

Sorry, maybe I was unclear. Those thing are really two different things. Samples that you load are typically in a variety of formats, resolutions and sample rates. When playing these, the engine will "compensate" for the difference in host resolution and sample rate, pretty much the same way that you play a sample in different pitch. Thus loading 96khz, 24bit, 32bit, 8bit etc samples makes no real difference when playing though quality will of course always be constrained by the least of original and host resolution/sample rate.
So, to sum up, libraries and source material is never a problem (except maybe footprint).

What I meant was the ability to reduce the quality of loaded or sampled material to either "lo-fi" it, or simply reduce memory footprint. I.e a bass sample probably plays back fine in 22Khz, while higher pitched material require the full 44.1khz to reproduce well.

Cheers

[elcallio]

I'm buying it soon
I wish it could read a bit more formats, so i wouldn't need any other sample player (well, i'm not much a smaple guy, except for the ones i create myself, but i have a couple of multilayered electric pianos and stuff, and i wish i don't need several sample players to read them).
SoundFont is great, but it seems that Gig is also rather popular, isn't it ?
In fact, i wouldn't expect tx16w to read 1:1 any sample format, but if at least it was able to read more formats with proper sample mapping (keymapping, velocity mapping) and proper ADSR, this would be way enough for my needs

I'm intending to add more formats in the pro version, though I didn't really aim for .gig (because of DFD). Akai and E-mu are perhaps my two highest prios.

[lalo]

Akai and E-mu are perhaps my two highest prios.

!

[makerprofaze]

While I may be asking for it, here's a pleasant little tnar for ya.

TX should be viewed in it's proper class, as SAMPLER in company with Maschine, MPC-R, Morgana, Geist (though it's keyboard based, which aside from Morgana is unique).

Morgana is lofi by design - very stupid gimick in my opinion. Short Circuit is great in a bunch of areas, but lacked sampling ability. Battery is good in some areas, but being a drum sampler, lacking in others. Directwave can sample vSTs, and that seemed like it's only shining factor. Geist can run standalone and does sample, plus has a great sequencing system, but for me, TX is a supeior SAMPLER due to the fact that you don't have to save every sampled wave before mapping. That stage made me dislike Geist- but it is great for a drum sampler. TX makes book keeping totally tranparent if you want (save with project).

Reason can sample, but it costs more and only uses a few proprietary formats. Not much exporting/converting out of Reason, except for audio as far as I know (up to v5 anyway).Kotakt is focused on commercial libraries, and doesn't sample. I didn't have a heap of fun loading my own waves into it as it is quite complex - I never achieved a streamlined method of doing things with all those busy windows. It certainly rules as industry standard for sellers of sounds, but that's irrelevant for someone who wants to add SAMPLING to their regular creative workflow. Must have just been me, after all everyone seems to love it.

As far as converting sounds, seems that a non musical session of instrument conversion is in order with a product like Awave, ExSC, CDXtract or Chicken Translator. Those are apps that focus on conversion and do quite well. And not one of them can SAMPLE.

Commercial libraries are genius how much expression gets packed in, but it takes a team workin night and day to take all those SAMPLES and create all the layers and keyswitches and scripts and such. When I read that a library took years to make, I think - wow, what kind of budget did they operate on - just to SAMPLE?!!. I wonder how much people really expect to achieve with a streamlined method of SAMPLING like TX - those libraries are built by Sound Library Professionals. To approach those results you'd probably have to transfer focus from making Music to Buiding Instruments. The work is astronomical that goes into those collections. This is the reality of sampling which many are unaware of as commercial libraries are easy to buy, and somehow the term 'sampler' unfortunately refers to 'sample player'. Aside from the emerging hardware/software solutionls only just emerging, apparetnly SAMPLERS are legacy machines from the 80's/90's - with very few software equivalents. Not to say TX can't compete. In fact, I'd like to see native TX instruments pushed to the limits of it's ability. We may all be pleasantly surprised.

What we have here is a brand new vst and instrument format, and as such we need to help to improve it, but we should also be concerned with taking advantage of it to the fullest and make our own sounds and even share a few. This is what will really help to show others and the commercial industry that a viable product has arrived - by the music communiy's reaction and usage. Converting other formats of course is key, but original material is what I really look forward to.

Click the red dot.

PS there's always LinuxSampler for gig format/DFD - and free too!

[geoslake]

Amazing !
Will you implement soundfont patch loading, so that we don't have to load the whole soundfont first ? (here, how it looks in samplelord)

[DarkStar]

THis is getting better and better.

A couple of FRs:

(a) In the File Browser: when I stop auditioning a sample I get a click or abrupt stop - can a fade-out be added?

(b) Saved Program (or Bank or Performance ...) : the sample paths are saved in the xml file as Full Absolute paths; saving with Relative paths would be useful too, as an option

[sinkmusic]

The sampling feature is great, but i guess there is still room for improvements :

- after a while, i notice a slow GUi responsiveness while trying to move start/end points ;

- I wish that in the editor, we could have : +/- 3dB boost, +/- 6dB boost (and not just normalize), and also being able to make fades in the audio editor (i guess that it isnot a priority FR, but still ) ;

- I didn't get how to make the threshold work : even by moving the threshold and hitting record, the recording seems to begin as soon as i hit the record icon, and not when a sound is coming into the sampler... I must be doing something wrong ?

- +1 for being able to call the PDF manual from the GUI, it is quicker !

- i baaaadly wsh for a feature where each recorded sound would be automagically mapped just after the previous recorded file. This way, i wouldn't have to break my workflow to make a new keymapping each time i record a file ; This mode could be called "drumsampling" or something like that (1 note = 1 sample) ?

- Is there any way to merge 2 groups (i accidentally created 2 groups, and now i don't know how to put the samples into a single group) ?

[sinkmusic]

and another FR for the adio editor : being able to swap Left and RIght channels for the sample...

[dexterbella]

On the subject of FRs - how about a few more envelope options? At the moment we have 5 LFOs available but only 3 envelopes - one is tied to amplitude and the other two are ADRs. Perhaps the step LFOs could get an envelope mode or a dedicated filter ADSR could be added?

[audiojunkie]

Thank you.
I am very glad to see such a fast rate of updates with a careful listening to user requests

On question abvout sampling : how do you select the sampling rate ? Will it record the audio stream according to the soundcard's rate, the host's rate ?
And if/when you will add FALC support, will it be possible to record in FLAC ?

Sampling is done in host rate, a plugin cannot change this. I will eventually add post-resampling support so you can downsample files as well.
FLAC support will include save capability as well, so yes.

I have a question regarding this... Let's say that someone creates an instrument that is sampled in 24-bit 96khz, but I am running my host at 16-bit 44.1khz. Will this be a problem in the current versions? I ask, because you say that you will 'eventually' adding support for post-resampling and downsampling. How are instruments currently going to be kept compatible?

--Sean

Sorry, maybe I was unclear. Those thing are really two different things. Samples that you load are typically in a variety of formats, resolutions and sample rates. When playing these, the engine will "compensate" for the difference in host resolution and sample rate, pretty much the same way that you play a sample in different pitch. Thus loading 96khz, 24bit, 32bit, 8bit etc samples makes no real difference when playing though quality will of course always be constrained by the least of original and host resolution/sample rate.
So, to sum up, libraries and source material is never a problem (except maybe footprint).

What I meant was the ability to reduce the quality of loaded or sampled material to either "lo-fi" it, or simply reduce memory footprint. I.e a bass sample probably plays back fine in 22Khz, while higher pitched material require the full 44.1khz to reproduce well.

Cheers

Thanks! That really clears things up!

--Sean

[Crackbaby]

The random slice player thing for pro version, do you have any idea when it will be available? I can slice up and make programs for Poise but it's a lot more work than what doing it with TX16Wx would be ... so if i can hold on for a little it would save me a lot of work!
Any rough estimation would help .. a week, a month ... a year

[elcallio]

The sampling feature is great, but i guess there is still room for improvements :

- after a while, i notice a slow GUi responsiveness while trying to move start/end points ;

- I wish that in the editor, we could have : +/- 3dB boost, +/- 6dB boost (and not just normalize), and also being able to make fades in the audio editor (i guess that it isnot a priority FR, but still ) ;

- I didn't get how to make the threshold work : even by moving the threshold and hitting record, the recording seems to begin as soon as i hit the record icon, and not when a sound is coming into the sampler... I must be doing something wrong ?

For threshold: (sorry, I didn't add tooltips yet), third button from the top in leftmost column is the "threshold trigger" (below "L" (listen), above "T" (transport)). When active, any signal exceeding threshold will trigger sampling

- +1 for being able to call the PDF manual from the GUI, it is quicker !

- i baaaadly wsh for a feature where each recorded sound would be automagically mapped just after the previous recorded file. This way, i wouldn't have to break my workflow to make a new keymapping each time i record a file ; This mode could be called "drumsampling" or something like that (1 note = 1 sample) ?

You discovered a bug. If you don't activate "auto detect pitch", it should not set a root key -> fixed (drum) key mapping. But right now it does anyway (C4). Will be fixed asap.

- Is there any way to merge 2 groups (i accidentally created 2 groups, and now i don't know how to put the samples into a single group) ?

Key mapper menu, "combine groups" will merge groups.

[elcallio]

The random slice player thing for pro version, do you have any idea when it will be available? I can slice up and make programs for Poise but it's a lot more work than what doing it with TX16Wx would be ... so if i can hold on for a little it would save me a lot of work!
Any rough estimation would help .. a week, a month ... a year

I've previously said I will have it available at the end of this year, and barring it takes too long to get 2.0 vanilla stable (my first prio), and I don't have to do to much "real" work to support myself, I hope to hold myself to it.

My intention is to release the Pro version in parts as well, i.e when a feature is done you get a new TX release (or side-release) with it in. That way I can give my supporters some value for their money incrementally and try to make sure features are fairly stable. Knock on wood.

[elcallio]

On the subject of FRs - how about a few more envelope options? At the moment we have 5 LFOs available but only 3 envelopes - one is tied to amplitude and the other two are ADRs. Perhaps the step LFOs could get an envelope mode or a dedicated filter ADSR could be added?

One not well documented (as in not) feature of the Mod ENV is that the T2/L2 point will be held for as long as MIDI key is held (unless group is in oneshot mode), and will transition to T3/L3 on release. Thus in effect, they are ADSR envelopes (but bi-level).

That being said, I have indeed thought about having the Step LFO:s having a "single cycle" feature. This is however a little more tricky than it sounds, but eventually...
Cheers