Tal J-8

[RexBanner]

So is EvilDragon the guy in his avatar? I'm always confused why a guy that looks like a guitarist from a technical death metal band is here.

Avatar looks like guitarist in a metal band. But bloke behind the posts is more of a synthesist and intellectual.

It is me, 10 or so years ago while I still had my hair.

Nowadays I look more like here.


(But I still play guitar and \m/)

I see some seven string guitars.

[EvilDragon]

Only one, but it's tuned down to C standard (with low G)

[kodnin]

In Live 10, Mojave 10.14.6, the back/fwd preset arrowv are erratic and most often jump 2 or 3 presets ahead of the next one. Also, some arps don't synchronise with the DAW tempo.

Yes, can confirm the preset browsing issue. It's odd. If I remember correctly there was also a similar issue on the TAL-Mod back in the day, but I might be wrong.

BassLine 101 has the same issue. I’d expect previous and next to go through the subcategory first, instead of (seemingly) random through all subcategories.

[EvilDragon]

It's actually an issue with how Live sets some host automation events. It happens in some other plugins too...

[himalaya]

Since I was part of the conversation about the oscillators, I wanted to respond to some further comments and thoughts.

Huge thanks to Patrick on the latest updates!

To some it's ED, and to a lesser extent me, being pedantic, but I think it's ultimately assisted in resulting in a better, more realistic emulation.

How do you know that this change made the J-8 a more realistic emulation? Did you compare it to the original hardware used for this emulation? Such comments are very presumptuous.

Patrick is on record saying that J-8 is behaving like his real hardware Jupiter-8, but here we have people who have no means to compare the plugin to the hardware stating that it didn't behave like the original hardware used for this emulation, and now after the reworked oscillator tuning options, you state that it sounds like the hardware...People are imagining how it should sound and push that opinion as fact. If there was another user with a 14bit Jupiter 8 in this thread confirming that J-8 indeed fails at emulating the oscillator drift/phase/etc, then we would have a valid counter argument.

In my experiments, I made my vintage Roland SH-2 behave like J-8 with its fine tune set to zero. Arturia's Jup-8V also displays the same phase 'locked' sound like J-8 when fine tune is set to zero and all effects by-passed and the strong pitch randomisation set to zero. This points to the fact that what Patrick had coded in the first version was correct. Other synths also display this behaviour.

However, the new osc tuning options are of course welcomed. Anything that allows us to tweak the sound to our liking is a great addition. But it is very presumptuous for anyone to say that the original wasn't a good emulation if you have no hardware as reference to make direct comparisons. Its as if you were saying that Patrick needed the advice of people who don't have the real, hardware 14bit Jupiter-8 for direct comparison, to help him code a better emulation. That's just crazy.

Anyway.
Here is Arturia's JUP-8V

www.electric-himalaya.com/stuff/TAL/Art ... -test1.wav

TAL J-8
www.electric-himalaya.com/stuff/TAL/TAL-J8-test1.wav

The same phase locked sound although captured at different phase cycles in each example.

The original version of J-8 makes lush analogue sounds, and the new version, with new osc tuning options, does so too. Let's post some audio of sounds that people make with J-8. It sounds absolutely gorgeous, even at version 1.1 without the new tuning options.

[perpetual3]

So, did the SC changes completely eliminate the original J-8 behavior?

Since I was part of the conversation about the oscillators, I wanted to respond to some further comments and thoughts.

Huge thanks to Patrick on the latest updates!

To some it's ED, and to a lesser extent me, being pedantic, but I think it's ultimately assisted in resulting in a better, more realistic emulation.

How do you know that this change made the J-8 a more realistic emulation? Did you compare it to the original hardware used for this emulation? Such comments are very presumptuous.

Patrick is on record saying that J-8 is behaving like his real hardware Jupiter-8, but here we have people who have no means to compare the plugin to the hardware stating that it didn't behave like the original hardware used for this emulation, and now after the reworked oscillator tuning options, you state that it sounds like the hardware...People are imagining how it should sound and push that opinion as fact. If there was another user with a 14bit Jupiter 8 in this thread confirming that J-8 indeed fails at emulating the oscillator drift/phase/etc, then we would have a valid counter argument.

In my experiments, I made my vintage Roland SH-2 behave like J-8 with its fine tune set to zero. Arturia's Jup-8V also displays the same phase 'locked' sound like J-8 when fine tune is set to zero and all effects by-passed and the strong pitch randomisation set to zero. This points to the fact that what Patrick had coded in the first version was correct. Other synths also display this behaviour.

However, the new osc tuning options are of course welcomed. Anything that allows us to tweak the sound to our liking is a great addition. But it is very presumptuous for anyone to say that the original wasn't a good emulation if you have no hardware as reference to make direct comparisons. Its as if you were saying that Patrick needed the advice of people who don't have the real, hardware 14bit Jupiter-8 for direct comparison, to help him code a better emulation. That's just crazy.

Anyway.
Here is Arturia's JUP-8V

www.electric-himalaya.com/stuff/TAL/Art ... -test1.wav

TAL J-8
www.electric-himalaya.com/stuff/TAL/TAL-J8-test1.wav

The same phase locked sound although captured at different phase cycles in each example.

The original version of J-8 makes lush analogue sounds, and the new version, with new osc tuning options, does so too. Let's post some audio of sounds that people make with J-8. It sounds absolutely gorgeous, even at version 1.1 without the new tuning options.

[zvenx]

No it didn't. With the sc vco tuning at zero i think it behaves like j8 1.1.1-1.1.5.
Rsp

[Funkybot's Evil Twin]

Since I was part of the conversation about the oscillators, I wanted to respond to some further comments and thoughts.

Huge thanks to Patrick on the latest updates!

To some it's ED, and to a lesser extent me, being pedantic, but I think it's ultimately assisted in resulting in a better, more realistic emulation.

How do you know that this change made the J-8 a more realistic emulation? Did you compare it to the original hardware used for this emulation? Such comments are very presumptuous.

Himalaya, I have nothing but respect for you but Patrick came into this very thread and said he also couldn't get the tuning as exact on his personal Jupiter-8 hardware as the plugin was behaving. Then he said he was going to run more tests against his hardware. Then what did he do? He changed the behavior of the plugin a day or two later.

Just to remind everyone, here's the exact quote (emphasis mine):

"I tried this with my Jupiter 8 and it is the same. But i wasn't able to set the detune knob exactly to zero. So it starts to drift away when removing the sync. I will make some more tests..."

Do you think Patrick changed the behavior to make it behave LESS like his Jupiter-8? Do you believe Patrick was lying when he said he couldn't get his Jupiter-8 dialed in tune as much as the the software? What did he do after conducting more tests? Added the option to add slight offsets to the per-voice oscillator tuning.

Isn't the most plausible answer here that Patrick took a step back, did the additional analysis he said he was going to do, and made the adjustments he felt were warranted? And if you still want it tighter than hardware can ever be, you still have that option.

Why is this even coming up again? It's settled. Everyone won. You can still get the tuning tighter than hardware, or get some tuning slop if you want it.

[mholloway]

Everybody won

The J-8 is Beastly. The Unison is very smooth. This is also the first time I've actually made use of keyboard splits -- hey, turns out they are fun! Who knew...

I've been working on material with it all week now, and J-8 does not disappoint "in the mix". I say this because, well, you know how some synths are very exciting when you load them up individually and play with them, but somehow never actually make it into an *actual song* of yours? J-8 has passed this all-important test. It's gonna be a workhorse for me (I love Patrick's take on the Roland filter that much, indeed).

-M

[zvenx]

@FET, that is not why he added the stuff in sc. The tests didnt change his opinion about his modeling. It was a compromise so that as you said everyone wins. But i do agree, no need to relitigate it.
Rsp

[Funkybot's Evil Twin]

@FET, that is not why he added the stuff in sc. It was a compromise so that as you said everyone wins. But i do agree, no need to relitigate it.
Rsp

But he did say he couldn’t get the tuning that tight on his hardware? And he did change the default preset trimmer values so the default now has a tiny tuning offset? And that tiny offset does change the behavior with fine tune at zero and when sync comes off? Just want to be sure. But no need to religitate anything.

[zvenx]

I assume you mean he added a preset trimmer not changed.
Rsp

[Funkybot's Evil Twin]

Changelog (emphasis mine):

Added VCO 1 and VCO2 to the SC (Service Control) dialog. This gives you the option to detune them on a voice basis. We added a very small value per default to remove the unwanted static phase lock between the two VCO's. This way the phase slightly drifts per default.

...So the default behavior has changed. Haven’t compared against existing presets so not sure if it’s a global change or just the default preset. But there was a change indeed.

[zvenx]

I am not sure what his new default trimmer values were cause I messed with them and then took them back to zero (double clicking) and it is behaving like it was for me before the added trims. Maybe his default is not having them at zero.
But I do agree the default behaviour has changed, I just disagree with your theory why it was changed.

It's actually I believe the second note that he changed because of listening further/testing further:

"Removed notch filter effect when syncing a saw with almost exact the same frequency. "

rsp

[Funkybot's Evil Twin]

I am not sure what his new default trimmer values were cause I messed with them and then took them back to zero (double clicking) and it is behaving like it was for me before the added trims. Maybe his default is not having them at zero.
But I do agree the default behaviour has changed, I just disagree with your theory why it was changed.

It's actually I believe the second note that he changed because of listening further/testing further:

"Removed notch filter effect when syncing a saw with almost exact the same frequency. "

rsp

I don’t understand what my theory is. The change log itself describes the phase locked behavior as unwanted. This was the exact behavior that evil dragon reported and I later did. I’m not sure that’s a theory anymore. It just seems like it’s an established fact. Patrick confirmed his own hardware could not be tuned so tightly. Then he changed the behavior by adding trimmers and setting their default value to not zero as indicated in the change log. To eliminate the unwanted phase lock behavior as indicated in the change log. This shouldn’t be controversial. It’s an improvement we all get to benefit from. This was long dropped as an issue after the update was posted earlier today but then I got called out for saying unwanted phase lock behavior that was removed didn’t make the emulation better and now you’re saying it changed for other reasons?

Here is my only theory, some users on a forum pointed out something weird, the developer looked into it, listened to his own hardware, and ultimately changed the behavior to improve it.