And that all ready loaded within seconds?
1 Like
Probably between 10-20 seconds.
Closer to 10!
- How long does it take to load your entire set with GP now?
- How long does it take to switch from one patch to the other in GP and how long in Forte?
How large is your Forte file?
Are the plugin configurations included in that file?
About 2 minutes or so…maybe a little longer.
Instantly in both…no discenrable difference.
11 MB
Yes
I’m sure @Angel would also concur with this, but on the same hardware, my Forte files (I can’t even remember the file type anymore) loaded significantly faster than GP. Of course, Forte has been discontinued, and personally I found that it was a bit buggy on Windows 10. It was a different paradigm. It really doesn’t matter if Forte loaded a full setup of songs faster or not. GP is simply different, and in most ways, significantly better. We’re only about two features away from doing everything I really liked about Forte while Forte would have to implement several dozen features to meet my expectations after using GP for three years now.
With Forte, you loaded all instances of all plugins you would use for a full show in a single shot. That would typically lead to a lower number of total plugins used. I also found that the rack itself had significant limitations in terms of total number of plugins loaded on the front of the rack. I hit that limit and it got to a point where there was a UI bug in the last 4 minor releases that made it so I couldn’t select and modify anything on the last plugin in the rack…
X
2 Likes
Agree 100%!
I did not have any bugs with Win10…only with the more recent plugins themselves, which is what prompted me to find a new solution and led me to GP.
1 Like
The question is really what exactly got loaded immediate and what got loaded “later”, perhaps unnoticed.
For example, on my MacPro (3.3Ghz with 12 Core and 96GB ram), if I run Logic Pro X and insert a Kontakt plugin, it takes 5-10 seconds to load the EMPTY plugin, no samples, etc.
Now, a second instance of the same plugin will load in about 1 second but that’s really because the shared library is already loaded so just the instance data has to be created.
If you then go and load an instrument into Kontakt, depending on various configuration settings, it will take several seconds to START loading the instrument and the samples will continue to be loaded in the background.
On the other hand, an empty SampleTank loads in both Logic or Gig Performer pretty much instantly, as do a bunch of the Korg plugins that I’ve tried.
1 Like
Hi Joey,
great to see you here.
In spring 2021 I switched from Brainspawn Forte to GP4, too. It took me months to transfer sound by sound and song by song for a 4 hours show set (excluding breaks) for a best of rock and metal and some Rammstein cover band. Most time consuming was the transfer of individual samples which are not in the factory libraries of NI Kontakt, and rearranging the complex MIDI routings and keyboard layers and splits for my two controller keyboards. But it was worth every minute.
I can confirm the observations of @Joey and @xpansion on loading times.
My old system with Forte:
Built 2014, Dell Precission M4800 i7-M4800MQ 2.7GHz, 16GB RAM 2x 512GB SSD.
Forte rack with 80 scenes = 80 songs loads within 13 seconds.
Forte rack file size is 22MB.
Forte RAM usage when loaded is 1.2GB.
My new system with GP4:
Built 2021, Clevo PB51DDS i7-10875H 2.3GHz, 32GB, 2x 1TB SSD.
Predictive loading is off.
GP4 gig with 70 rackspaces = 70 songs loads within 1 min 23 seconds.
Gig file size is 29MB.
GP RAM usage when loaded is 26GB.
Passmark Performance Test factors of the new system compared to the old:
CPU 3x faster.
RAM 1.3x faster.
SSD 5x faster.
Based from that I would think that Forte would load my whole show within 5 seconds if installed on my new system.
In Forte my rack consisted of:
1x Steinberg Hypersonic jbridged to 64bit
1x Spectrasonics Omnisphere
2x NI Kontakt 5
1x 4Front TruePianos
1x NI B4 jbridged to 64bit
- a few FX.
In GP I have a global rackspace with:
1x NI Kontakt 6, for sample FX and intros/outros I could need on any time during a show, and a low-latency master compressor/limiterfor the sum going to FOH.
I have individual rackspaces for each song because of very different sounds for each song, typical for keyboarders in a cover band, containing typically, but differentyk from rackspace to rackspace:
1x Omnisphere
1x NI Kontakt 6
1x Hypersonic or HalionSonic
1x TruePianos or NI B4.
- a few FX.
What’s the difference? As already mentioned in this thread:
Forte loads only one instance of the plug-ins that make up a fixed rack with a routing that can’t be changed much during the show. And it loads only the sounds and samples for the very first scene aka song during start-up. Switching songs takes 1-2 seconds, in my case, for loading new samples into Kontakt and Omnisphere or any other of the fixed and already loaded plug-ins. There is no predictive loading, and patch persist is impossible. Switching songs too fast could leed to crashs with NI Kontakt, because of trying to load new samples while the previous sample load process hasn’t even finished.
In GP4 I don’t use predictive loading. I like to have everything loaded into RAM as long as it fits into it, to be able to switch instantly between arbitray songs, and to avoid any excessive loading processes during the show. However, with predicitve loading on, load times are much shorter, and I have not observed any issues so far.
With these settings, GP loads individual plug-ins for every rackspace, and loads all samples as requried for them, for all rackspaces or songs. Of course, this takes more time and RAM.
The advantage of this is patch-persist and instanteneously switching between arbitrary songs which is impossible with Forte. Actually this has not been a problem with Forte and my band for 17 years. Changing guitars or tuning to dropped takes much longer, lest we forget ordering some beer or jacky-coke during song breaks …
Honestly, I would still be fine with Forte, if it would still be supported. But Forte is dead, there are no fixes anyymore, there are no new features, and some issues never got resolved.
GP offers all that I need and much more.
For example, for some weird keyswitches triggered by MIDI notes in Forte, I had to use third party programs like Bome MIDI translator and assembler-like codes for programming. In GP I can do much more complicated things with high-level language scripting.
In GP, the rig manager, aliases, and widgets allow to change ccontrollers quite quickly, even if they have very different CC hardware mappings. In Forte that would have been a nightmare.
Latency and most important, latency jitter, in Forte is much worse than in GP, at same buffer sizes of the audio driver.
Missing things in GP are a rig manager including audio devices, and a simple and customizable “scene view” or song/setlist view like in Forte. The GP Selector extension comes close, alternatively a custom OSC surface.
Widgets on panels in GP are not that much important for me, as I don’t controll sounds with widgets and mouse or touch screen, but instead with controllers from the keyboard. So only half of my rackspaces have widgets on a panel at all, only for a few individual MIDI CC that I need for a ever different parameter of a particular sound that I would like to modulate live. And as sounds are very different from song to song, I have to program them within the plugin windows anyway, and not with a few limited widgets.
And finally the gig loading time is of no concern for me. This is done once before sound check. And then stays stable for hours until the last song.
And yes, it took me some time to adapt my workflow on sound design to GP but now I absolutely like it.
2 Likes
I eliminated this by loading all the samples I would need for a show in Kontakt and leaving it “static.” It’s a setting in Forte: “Do not set per scene configuration.” Then have Forte access them through seperate MIDI channels. My Forte switching was instantaneous.
Correct…unles you did it like I did 
AMEN! I am thrilled with the added versatility.
I never experienced this problem in Forte…but it really dosen’t matter now!
I would like to see GP someday be able to automatically recognize what rig you have connected. Forte did that, and I still…even after 3 months…sometimes forget to switch Rigs and am scratching my head asking why I am not getting any sounds when I hit the keys!!, LOL.
As it should!
I am extremely happy with GP and look forward to a great 2022 with it!
Joe
There is no magic here. There are specific situation where you can do it, while other are ambiguous. e.g. I use keyboard controller going through the DIN5 MIDI connector of my audio interface (for some valuable reasons) and have a different rig manager setup for gig and rehearsal. I don’t see how either Forte or GP would manage to automatically switch from gig to rehearsal rig. And there are other similar examples…
After reading your very instructive post, it is clear that Forte doesn’t load 80 songs, but a bit more than 1 single song. The Forte loading time should be rather compared to GP having predictive loading enabled with a loading depth of three.
Well, I personally don’t need such a feature (at the moment…). However I could imagine some rulebased mechanism when starting GP based on a dedicated USB MIDI port and/or audio interface - just like when running GP with a lower channel count audio interface and asking for enabling virtual channels
1 Like
I am pretty sure we would be faced to so many exceptions to the rule that it would end with situations difficult to manage:
- different gig/rehearsal controllers using DIN5 MIDI connectors (audio interface,wireless MIDI transmitters=>no information about the connected controller)
- same GP flight case with audio interface for gig/rehearsal (doesn’t help to recognize the audio interface)
- some lightweight USB controllers used for both gig/rehearsal (are we at gig or rehearsal?)
- some USB controllers not used for rehearsal, but not immediately detected at gig rig startup (e.g. longer booting time of on of my controller than the whole GP gig: if the controller is not there we are at rehearsal… oh, really?)
I am personally face to the above listed situation, but pretty sure the list is not complete.
To clarify what Forte did…upon loading a Forte performance file, it would scan and recognize the audio interface and controllers via their “electronic signature” through USB. If there were multiple controllers connected via traditional Din5 it wouldn’t be effective. Nor could you switch while Forte was active, which GP does remarkably well.
I am quitely chuckling here, because when I started my quest with GP, the credo I heard ad nauseum was “don’t try and compare what and how Forte works to GP…they are totaly different!!” And so, when Forte does something that is quite impressive, then out come the reasons why it is NOT so impressive by trying to compare it to how GP handles things!
The bottom line is (was)…Forte loaded my entire show (74 Scenes) in 10 seconds (I timed it Friday night)…complete with opening around 19 instrument plugins, loading my samples in Kontakt, (and all the IK stuff that uses samples) and was ready to go with the first tune of the set. I could click on any other song (ie: Scene) and it jump to it and I could play it instantaneously without any lag. How it did it, why it did it…dosen’t matter. It did it. Period.
And, for what it’s worth…I can open GP , go to File>Import Song and that one song will still load slower than my entire Forte file!
Agreed, and as I mentioned erarlier…the recognizing of the Rig was dependent on the ID of the unit via USB. So, Forte knew I had a Hammer 88 and a Keylab 61 mated to my Tascam US144 Mk II. At home, it knew my practice/programming Rig combo. It could NOT mix and match. It identified a rig in its entirety…or else it would ask you to choose a previously profiled Rig and reassign what was missing to what was present, and then name it as a new profile.
I haven’t started to compare Forte and GP, but if someone do it, for example concerning the loading time, then one has to do it trying as much as possible to be in the same conditions. @Angel explained perfectly that his 80 songs gig loaded in 13 seconds in Forte and that it took him then in fact, one to two seconds to go from one song to another. With its 80 songs it’s between 1:20 min and 2:40 min more than the initial 13 seconds and it’s about the same loading time as GP. So, it would be interesting if @Angel could tell us how long it takes for GP to download his gig with predictive loading enable with a depth of 3. We still wouldn’t get exactly the same after loading (e.g. with GP switching between two consecutive songs would still be instantaneous), but the comparison would be a bit more fair.
Having said that, I think, the idea of this discussion is not to tarnish Forte’s strengths by sweeping away its assets with some of GP’s more interesting features. At least, that’s not what I’m trying to do. It’s more about understanding what worked well with Forte and seeing if the idea is reusable for GP. It’s not always possible, but IMHO it’s interesting to think about it.
1 Like
Of course! The comparison is not fair.
So here is my loading time in GP on my new system (much faster than my old with Forte) with predictive loading set to three: 25 seconds.
But as said already by Joey and me: Interesting to see and discuss the differences, but in fact the loading time is no issue for me.
1 Like
Correct! But still on a machine many times slower than my actual GP system
Could it be the native multicore support of Forte that enabled a quite efficent loading process?
Again, interesting to discuss but no issue with GP for me.
I disagree.
Many of us have two or three typical setups, maybe with different audio interfaces or different numbers of controllers. I think these configurations, called “hardware profiles” in Forte, can be detected quite well. And if not there always would be an option to adjust by hand in the rig manager, as it is now. Forte did a good job on this.
The main point that is missing in GP at the moment is a rig manager for audio profiles. Doesn’t have to be automatic in frist place, could be manual as the rig manager for MIDI is now.
Right now I’m sitting in my living room. When I start GP (to make some checks for the forum 
) it complains that my RME Babyface Pro is not there, and uses the built-in audio instead. Back in my studio I switch back to the RME Babyface Pro, which has completely lost all its selections for active input and output channels in the audio options! I don’t need any audio input so I disable each of the 16 inputs offered. And I need only two audio outputs so I disable all other 14 outputs.
I have to redo this manually each time after starting GP without the RME.
Going to check if there is a feature request, already …