Im with you on that one. Cali is sooo CPU heavy it´s crazy. None of my other Neural products bother my cpu, but Cali is just crazy.
Hi, I’ve realized lately this is a problem, didn’t notice before, but Cali takes almost 30% of my CPU while Omega Granophyre only takes 5% to 7%. Deactivating the cab section only takes the Cali from 30% to 23%. Really hope there is a fix on the horizon for this.
Same here. The cab section of Cali consumes significantly more CPU than all the other Cab sections of the Neural plugins.
Perhaps it has a more sophisticated/complex IR algorithm behind it? Otherwise I can’t understand the difference…
are there any chances for an update in the future? optimization CPU performance and fixing issues with the cab? why developers do not pay attention to the requests of their users?
I am also wondering that. I’ve started to believe that they are busy with developing new products to make more money, rather than solving issues with the old products. If I didn’t miss it, also NTS VST3 version has some problem in Reaper and they haven’t released any update yet. It is 2 years old product, I think (Not a problem for me, I am happy with the VST2 version but it is still annoying.) I can understand that sometimes it is harder to find and solve bugs than to develop a new product but I want to believe that their team is proficient enough to do that as well.
1 Like
I think we are not worthy of an answer from the Neural DSP
I´ve sent the support an e-mail because of the CPU load-issues and they´ve replied, that they are aware of that issues and that they´re working on an update.
1 Like
They have been aware since a few days after the release date and said that they are working on an update. That’s why people think that their priority is releasing new products rather than fixing bugs.
3 Likes
Just found this thread… Bought the Fortin Cali plug and it’s crackling like crazy. Nolly and Gojira are fine even at low latencies and oversampling with stereo on. The Fortin Cali plug in VST or VST3 barely runs in mono at regular sampling. Sometimes it works fine, sometimes not. Oversampling always sends it into crackling at any latency.
CPU is 6 core and doesn’t get over 40% usage. The stand-alone version works fine, even when played while the DAW is running with dozens of tracks and plugs. I can run dozens of instances of the other plugs like Nolly and Gojira with no issues.
LatencyMon shows bad issues while this plug is crackling but shows that everything is fine without this plug running. It shows that everything is fine with Nolly and Gojira plugs running too.
Since the basic sim must be the same between all their models, how could they make this so bad yet be so slow to fix it?
This thread is going on two years old and zero movement by Neural. We have to face the fact that they are not going to fix it. And Neural has to face the fact I’ll buy from other vendors from now on (and have been). Apparently no support once a product is out. Pity the fools that bought QC…
Can confirm Cali is an absolute CPU hungry monster compared to newer plugs like Gojira. I’ve just been changing tracks over to that in a session I’m working on at the moment or otherwise I’m reaching the point where I can’t get playback anymore even at 2048 buffer size…
Ouch… Thats crazy cpu reports. It def eats more cpu than Granophyre, Nolly and Parallax but not that much more. I can easily make 10 active tracks and a couple of instrument tracks with Superior Drummer, and it runs smoothly on my 2015 24gb ram imac at 32 samples … maybe its because of my soundcard? Rme Ufx via USB?
So it’s now mid September 2021 and Cali is still on 1.0.0
NDSP have been touting new CPU friendly updates since Gojira came out.
But not to worry, they just released a new QC update.
1 Like
I bought this in April 2021 and found the same issue - it was unusable. I wrote to support requesting a refund and based on their response, decided not to persue the refund as they said they were aware of the issue and will be releasing a fix soon. Well, I haven’t been able to use the Cali a single time since purchase - literally like throwing my money away.
Like others, I too have not purchased anything else from them, nor will I in the future.
Is it still too late to refund? Here’s an excerpt from their reply to me:
Thanks for reaching out and sorry to hear you are experiencing that issue.
There have been other users as well that have reported experiencing CPU performance issues only on Windows machines. The dev team has replicated it on their end as well, and has confirmed that they are currently working on an update to be released very soon to clear out any potential bugs that may be contributing to the CPU overconsumption.
We apologize for your inconvenience with this particular plugin’s performance on your machine and are actively working on wrapping up that update for release, it will be out soon.
2 Likes
+1 here…
I bought Cali and it was unusable… I was also promised on this forum an update for Cali. I recently sold my license. Also i have Nolly plugin and still waiting for update since NeuralDSP promised in January 2021 to update the engine for all plugins since Archetype: Gojira released. Today middle of October 2021. I don’t buy their plugins anymore and I feel that I will also sell Nolly soon.
Sorry to hear you’ve been having performance issues. Have you followed any of the optimization steps on our support website? The update to Fortin Cali will be available before the end of November.
November of what year? Serious question.
1 Like
November this year!
Tried out 2.0 and its still a bit of a resource hog compared to Wong but it works fine (tested as a plugin ) and no issues clicking thru presets, whereas before clicking thru presets would provoke issues. Very happy to see that attention was given to this. 
I know this is from a while ago but I figured I’d chime in. I’m using a new-ish i7 equipped lenovo laptop. I own Gojira and Cali and Cali still consumes more resources by far. I know nothing about what is going on under the hood but I’m shocked that the plugin doing real-time pitch shifting outperforms the one that doesn’t.