GForce Sequential Prophet-5

[Funkybot's Evil Twin]

So I did a comparison of my Rev 4 hardware to the Rev 4 software looking at tuning, particularly in relation to the Vintage knob.

I think my first post in this thread was me saying something like, 'the Vintage knob on the software isn't as extreme as on my hardware' and I've wanted to test that out. The answer was "sort of, but not really" (I think I had the software set to 5 voices when I drew that conclusion - more on that below).

Methodology (in case a hardware owner wants to validate/dispute the results):

1. Created an INIT patch on both. On the hardware, hold RECORD and press PRESET. On the software, select an INIT patch from the browser.

2. Set the Voice mode to Round Robin on both. On the hardware, hold BANK SELECT then press GLOBALS, then Program Select button #2. On the software, go to the settings icon.

3. Set the software to 10 voices (the INIT patch is only 5). Or keep it at 5 if you don't have a 10-voice Prophet Rev 4.

4. Added "ReaTune" to a track in Reaper, played the same middle C repeatedly and noted the setting within roughly 0.5 cents. I wasn't getting more granular than that. Repeated that 10 times.

5. Dialed in knob positions by eye on the hardware and used the tool tips on the software. Wasn't trying to get it 100% exact. Close enough is good enough.

Things I observed during this test:

1. The differences are smaller than I thought.

2. The voices get progressively more out of tune. So if you want the tightest tuning, using 5 voices can help.

3. As you can see though, some voices on both the software and the hardware can get really out of tune with the vintage knob maxed out. We're talking 30 cents off for the highest voices.

Assumptions:

1. I'm assuming the most "in tune voice" was voice 1.

2. I've similarly assumed that the most out-of-tune voice was voice 10. Limiting it to 5 voices helps confirm that.

3. I'm otherwise kind of guessing/assuming on the voice numbers. I don't think there's a way to know for certain, but I think those assumptions hold true. Either way, this is close enough.

Results:

The numbers represent the average number of cents off each voice was from being a perfectly in tune middle C.

Important Edit: I modified this to use the Absolute deviation so -14 and +14 average out to 14, not to zero. Interestingly, doing so changed the results to show the software is actually a little more out of tune than the hardware.

Code: Select all

Rev 1
Hardware	14.75
Software	18

Rev 2
Hardware	8.2
Software	9.2

Rev 3
Hardware	2.05
Software	4.95

Rev 4
Hardware	0.6
Software	3.45

And here are the details...

Code: Select all

Voice	Tuning 	Abs	Vntg	Hardware/Software
1	1	1	Rev 1	Hardware
2	7	7	Rev 1	Hardware
3	-4.5	4.5	Rev 1	Hardware
4	12.5	12.5	Rev 1	Hardware
5	-11.5	11.5	Rev 1	Hardware
6	18.5	18.5	Rev 1	Hardware
7	-17	17	Rev 1	Hardware
8	24	24	Rev 1	Hardware
9	-23	23	Rev 1	Hardware
10	28.5	28.5	Rev 1	Hardware
1	1	1	Rev 2	Hardware
2	4	4	Rev 2	Hardware
3	-2	2	Rev 2	Hardware
4	7	7	Rev 2	Hardware
5	-6.5	6.5	Rev 2	Hardware
6	10.5	10.5	Rev 2	Hardware
7	-9.5	9.5	Rev 2	Hardware
8	13.5	13.5	Rev 2	Hardware
9	-12.5	12.5	Rev 2	Hardware
10	15.5	15.5	Rev 2	Hardware
1	1	1	Rev 3	Hardware
2	1.5	1.5	Rev 3	Hardware
3	0.5	0.5	Rev 3	Hardware
4	2	2	Rev 3	Hardware
5	-1.5	1.5	Rev 3	Hardware
6	3	3	Rev 3	Hardware
7	-2	2	Rev 3	Hardware
8	3.5	3.5	Rev 3	Hardware
9	-2.5	2.5	Rev 3	Hardware
10	3	3	Rev 3	Hardware
1	1	1	Rev 4	Hardware
2	1	1	Rev 4	Hardware
3	1	1	Rev 4	Hardware
4	0.5	0.5	Rev 4	Hardware
5	0	0	Rev 4	Hardware
6	1	1	Rev 4	Hardware
7	0	0	Rev 4	Hardware
8	0.5	0.5	Rev 4	Hardware
9	0.5	0.5	Rev 4	Hardware
10	-0.5	0.5	Rev 4	Hardware
1	3	3	Rev 1	Software
2	3	3	Rev 1	Software
3	-10	10	Rev 1	Software
4	16.5	16.5	Rev 1	Software
5	-14	14	Rev 1	Software
6	22	22	Rev 1	Software
7	-23	23	Rev 1	Software
8	27	27	Rev 1	Software
9	-28.5	28.5	Rev 1	Software
10	33	33	Rev 1	Software
1	3	3	Rev 2	Software
2	0	0	Rev 2	Software
3	-6.5	6.5	Rev 2	Software
4	9.5	9.5	Rev 2	Software
5	-7	7	Rev 2	Software
6	11	11	Rev 2	Software
7	-12.5	12.5	Rev 2	Software
8	13	13	Rev 2	Software
9	-14.5	14.5	Rev 2	Software
10	15	15	Rev 2	Software
1	3	3	Rev 3	Software
2	-2	2	Rev 3	Software
3	-4.5	4.5	Rev 3	Software
4	6	6	Rev 3	Software
5	-3.5	3.5	Rev 3	Software
6	5.5	5.5	Rev 3	Software
7	-6.5	6.5	Rev 3	Software
8	5.5	5.5	Rev 3	Software
9	-7	7	Rev 3	Software
10	6	6	Rev 3	Software
1	3	3	Rev 4	Software
2	-2.5	2.5	Rev 4	Software
3	4	4	Rev 4	Software
4	4.5	4.5	Rev 4	Software
5	-2	2	Rev 4	Software
6	-3.5	3.5	Rev 4	Software
7	-5	5	Rev 4	Software
8	3	3	Rev 4	Software
9	-4.5	4.5	Rev 4	Software
10	2.5	2.5	Rev 4	Software

[JerGoertz]

I compared the Multi/Poly with the GForce plugin today. I find that for the most part, the Multi/Poly (starting with the Pro(n) template) can be very close (Yes, even the filter FM) and uses slightly less CPU. You need to set the filter cutoff to 121 and remove the filter frequency drift to match the harmonics with the Prophet plugin. The only thing that sounds significantly different is combining the xmod with oscillator sync (which is cleaner than the Prophet's but close enough). I find the combination to be musically and sonically more interesting at times in M/P than Prophet-5. The way the filter resonance responds to sync/PWM is a bit different in M/P. Compared to the GForce plugin, the multi/poly's Pro filter has its resonance vibrating more intensely with oscillator sync sweeps, but not as much for PWM. And if you want the self-oscillation to match the oscillator levels in M/P, bring the filter resonance knob to around 70.

The Multi/Poly is a beautiful synth. I liken it to DMG Audio's TrackComp: It doesn't try to look analog, but it's extremely flexible and the sound is absolutely there.

[abernathy]

I have another bug in the Audio Unit version. OSC 2 Fine setting reverts to 0 cents when opening saved Logic Pro project. I had a project with several Prophet 5 factory presets that had various OSC 2 Fine settings. They all are set to 0 when reopening the project. I first thought "wow, this IS authentic. It hasn't warmed up yet... yikes". I just noticed that when I first open an instance of P5, the It's A Prophet 6 init preset has in incorrect Fine setting of 0 cents. If I switch to another preset and back to it then it is correct.

And GForce did acknowledge my effects lock not working issue and they're working on it.

So the MIDI reset issue is NOT a GForce bug, just something weird in my Logic Pro. My SFC-mini V4 controller uses MIDI CC 6 for an Oscillator fine tune setting, and my Logic Pro resets that when a plugin opens. Spent half the day trying to solve it. The only thing I learned is that CC 6 really shouldn't be used for controller mapping.

[djanthonyw]

I compared the Multi/Poly with the GForce plugin today. I find that for the most part, the Multi/Poly (starting with the Pro(n) template) can be very close (Yes, even the filter FM) and uses slightly less CPU. You need to set the filter cutoff to 121 and remove the filter frequency drift to match the harmonics with the Prophet plugin. The only thing that sounds significantly different is combining the xmod with oscillator sync (which is cleaner than the Prophet's but close enough). I find the combination to be musically and sonically more interesting at times in M/P than Prophet-5. The way the filter resonance responds to sync/PWM is a bit different in M/P. Compared to the GForce plugin, the multi/poly's Pro filter has its resonance vibrating more intensely with oscillator sync sweeps, but not as much for PWM. And if you want the self-oscillation to match the oscillator levels in M/P, bring the filter resonance knob to around 70.

Agreed. Multi/Poly is actually what I am using as my Prophet-5 emulation instead of a dedicated plugin. This Gforce emulation is really good though maybe hitting that extra 2% for authenticity.

[234north]

The rest of the post was right about building up the tuning tables however and potentially needing to delete them for troubleshooting.

Yeah but saying "there is no autotune button as such" when there is a button right on the front of the panel clearly visible that says "tune" makes me question anything after that. It's like IvyBirds making up that the VCA envelope controls are right under the VCF envelope controls and therefore easy to make a mistake while performing by grabbing the wrong one... That the calibration process has been modernized along with the electronics shouldn't be surprising and it doesn't make the automatic tuning process any less than on the older gear. The benefit is that it should be less necessary than on the older gear.

AI hallucinations? Faulty human memory? Misspeaking/typing? My next post on this thread will be on tuning. Are you a Rev 4 owner too? Might be interesting to compare notes.

No offense but none of those are excuses. Only need to look at their website. We're not talking ancient history here, they're currently on sale But funny enough on the Internet you can find just about anything, including the rev 2 user manual.

Yeah I own a rev 4.

[234north]

AI hallucinations? Faulty human memory? Misspeaking/typing? My next post on this thread will be on tuning. Are you a Rev 4 owner too? Might be interesting to compare notes.

Mistakes are not allowed when arguing about synthesizers. This is serious business. Lives are at stake.

Mistakes shouldn't be allowed when you're talking about something so easy as which/where controls are on a synth panel when it can be looked up. This fancy thing called the Internet and all.