Yeah I agree, it’s like a 303, almost impossible to sound like shite!
But that’s where the ‘ear’ cones in, you have to have great quality control.
When you set the fx and really design what you want, it’s seriously powerful.
Big love the daddys at TT! 
The main principle of it is comping.
I don’t know what is “comping”… 
I see it more as an expanded DJ mixer with FX and the ability to have all tracks, stems, loops play simultaneously…
It’s a dream come true, as way back, I’d load my loops into Live and set everything up to switch between ‘any’ loop every 16th (for example) but it wasn’t particularly intelligent. Legato was on so it all stayed in sync but it was very limiting if you loaded anything with swing or something that wasn’t edited already.
I mean I got results but it was way too much digging, and also the FX side was problematic.
I’m happy with that whole modular type approach where you set something up, see what happens, tweak a bit, see what ‘the system’ does and then really target in and whittle down what your aims are.
I see no difference to plucking a string differently or touching a key, I do something and then change what/how I do it depending on what comes out.
If only human relationships were so easy 
It’s like comping with 8 different loops. It’s a bit like the megabreak on the OT. At any time of the sequencer, you choose which of the 8 loops is audible. And then you can automate sample selection of the sample map and effects like crazy 
And also you can layer more than one at a time! All if you like… And decide whether fx are on a step or stem/lane etc 
With a bit of preparation in your sample map you can get some great results, a bit of silence here and there for instance😉
I can do that with OT without Megabreak. 
And @rotallytuined of course if I go back a bit to DAW and plugins I have to try Ultraloop in priority.
Next year maybe!
No doubt you can 
Tell me how you would do it
I have to say, it’s just incredible convenient in Ultraloop. The interface is genius
Already did, above and in other conversations we had. Find the pieces of the puzzle. 
haha. Ok. I’ll do
I just load up Reaktor and smash through some stuff.
Still working with loops from UL in 2014!
The Make Noise Morphagene has a similar ability but with only one loop/segment per step…
The OT will forever be King!
So here is a try with 4 Amen Break tracks, triggered alternatively by midi tracks arp.
1 generative bar. Not great, but it was fun and surprising. @Unifono, sorry, it’s randomized. 
Here’s your HW tracker eurorack.
https://xor-electronics.com/
And standalone:
https://xor-electronics.com/nerdsynth/
His style is the result of good homeworks i think
Because all of is tracks are single/unique
the quality of these loops lies just by an hard-work of sampling art
In addition to the quality of his melodies He has a real talent for building and mixing loops
As a conductor…It’s just a cooking work
Afta “it 'snice to watch how he goes from a sampler that generates a beat to a TR606 out-of a pedal effect transition go thru atmosphère in same régularity.”
The quality is felt when you work a loop hrdr … what# is on 1/2/3/4 … measures
Creat on computer ,then reused in one type of effect ,then resampled ! and replaying with other sounds in lines …
Take two international artists
The already has facilities for melodies and the other is obliged to sample melodies
They are made very good music!
Thats all !
I will end up with a personal opinion :
Without technology the music would be less deep
For someone who knows his weaknesses …it is easy to work around them
Hmm yeah, probability and other auto-magic processes are no substitute for good programming IMHO, sure they have their place but it would be wrong to consider them even a passable substitute to getting busy with the edits. Leave nothing to chance
Edit - I never seem to enjoy the results of probability, I like to use iterations though as these are by their nature guaranteed to happen when expected, which of course amongst other things can be used to get more milage from a limited sized grid. I do also enjoy using different sized step grids against each-other because again this can be predicted to a certain degree, so maybe my dislike of probability is a result of the brains attraction to looking for patterns, where intricate patterns can be a delight and probability just sounds too random?
Perhaps an idea inspired by the way I proceed,
Keep three identical samples
With a dozen BPM gap between them and pitch effect
Ressampling play with the scenes with definition of filters and eco time
The microprocessor will make a difference with the authors of BPM and will make the particular sample
Then with a micro timing(and scale per track) the loop can evolve in a sense
It’s a way like using the slicemode associated with delay for exemple
I hope this can help you create what you want
Give chance a chance!
At the opposite, it would be hard for me not to use random parameters I use since 20 years!
Ok, I know, I’m totally abusing them! 
I have no interest in machines without random parameters. Octatrack for ever.
I can’t my brain won’t allow it! 
Besides I’m not a super cool French dude with a wild sense of adventure and probably great hair and probably a nice motorcycle too
Joking aside, I should have said I don’t like probability much for trigger/note events but do use it for parametric things where I find it more useful.
So I’ll give P’s a chance
P=Parameters