Well, because some people may not get it..yes things are said that are out of line, but that's what friends do when they argue...soEvilDragon wrote:spacekid wrote:I'm really not trying to be a jerk
Hey, me neither
Peavey ReValver 4 - ACT Combo and ACT Rack released!
- KVRian
- 1007 posts since 12 Sep, 2005
-
- KVRer
- 26 posts since 16 Nov, 2013 from UK
I've been wondering how the ACT inputs work, so I've done a bit of investigation
First of all, it asks you to play as many notes as possible on your guitar to "build a profile". I'm assuming that what happens here is there's a FFT running constantly, and it's collecting all of the frequencies you hit. There's probably an array representing frequencies from 0 Hz to 40 KHz and whenever your guitar makes one of those frequencies (from the note you're playing and its harmonics) it gets ticked off in the array. When it's finished (probably after ticking x amount of boxes) it has a rough idea of the frequencies your guitar pickup or mic can reproduce.
I imagine the target ACT profile (of a Telecaster, say) is compared to the input profile (your Strat, say) and they figure out what they need to do to make your input profile either A) more accurately match the target profile or B) flatten your profile out so there's essentially a flat frequency response to work with when they apply the impulse response.
There's definitely an impulse response at work - first of all they have ACT profiles called Takemine Mic A and similar, so it stands to reason IRs would be involved. Second of all - I recorded an input profile using the Reaper JS white noise generator rather than a guitar, so the frequency response of that input profile should be reasonably flat. Second of all I got a 0.1s clip of white noise and fed it into revalver running 3 different ACT profiles, and there's definitely a little bit of a tail created when it's run through revalver:

Out of interest, using the white noise input profile, I played white noise into a few different ACT profiles to get a sense of the curves:

Tele style neck pickup is pretty much what you'd expect, frequencies dropping off pretty hard >2k. Takemine Mic'ed is pretty flat - louder in the bass region. Resonator is almost flat but has little peaks which I'm assuming are from the resonator's resonances
First of all, it asks you to play as many notes as possible on your guitar to "build a profile". I'm assuming that what happens here is there's a FFT running constantly, and it's collecting all of the frequencies you hit. There's probably an array representing frequencies from 0 Hz to 40 KHz and whenever your guitar makes one of those frequencies (from the note you're playing and its harmonics) it gets ticked off in the array. When it's finished (probably after ticking x amount of boxes) it has a rough idea of the frequencies your guitar pickup or mic can reproduce.
I imagine the target ACT profile (of a Telecaster, say) is compared to the input profile (your Strat, say) and they figure out what they need to do to make your input profile either A) more accurately match the target profile or B) flatten your profile out so there's essentially a flat frequency response to work with when they apply the impulse response.
There's definitely an impulse response at work - first of all they have ACT profiles called Takemine Mic A and similar, so it stands to reason IRs would be involved. Second of all - I recorded an input profile using the Reaper JS white noise generator rather than a guitar, so the frequency response of that input profile should be reasonably flat. Second of all I got a 0.1s clip of white noise and fed it into revalver running 3 different ACT profiles, and there's definitely a little bit of a tail created when it's run through revalver:

Out of interest, using the white noise input profile, I played white noise into a few different ACT profiles to get a sense of the curves:

Tele style neck pickup is pretty much what you'd expect, frequencies dropping off pretty hard >2k. Takemine Mic'ed is pretty flat - louder in the bass region. Resonator is almost flat but has little peaks which I'm assuming are from the resonator's resonances
- KVRian
- 1007 posts since 12 Sep, 2005
Good work. Yes, it seems very picky, so after trial and error, I found that staying with the same source amp + mic position + cabinet used is important when adding ACT Rack profiles(use the same original source profile that you made, and apply a target that is derived from that same source profile).If not, the results get muddied/phased.kohitsu wrote:I've been wondering how the ACT inputs work, so I've done a bit of investigation
First of all, it asks you to play as many notes as possible on your guitar to "build a profile". I'm assuming that what happens here is there's a FFT running constantly, and it's collecting all of the frequencies you hit. There's probably an array representing frequencies from 0 Hz to 40 KHz and whenever your guitar makes one of those frequencies (from the note you're playing and its harmonics) it gets ticked off in the array. When it's finished (probably after ticking x amount of boxes) it has a rough idea of the frequencies your guitar pickup or mic can reproduce.
I imagine the target ACT profile (of a Telecaster, say) is compared to the input profile (your Strat, say) and they figure out what they need to do to make your input profile either A) more accurately match the target profile or B) flatten your profile out so there's essentially a flat frequency response to work with when they apply the impulse response.
There's definitely an impulse response at work - first of all they have ACT profiles called Takemine Mic A and similar, so it stands to reason IRs would be involved. Second of all - I recorded an input profile using the Reaper JS white noise generator rather than a guitar, so the frequency response of that input profile should be reasonably flat. Second of all I got a 0.1s clip of white noise and fed it into revalver running 3 different ACT profiles, and there's definitely a little bit of a tail created when it's run through revalver:
Out of interest, using the white noise input profile, I played white noise into a few different ACT profiles to get a sense of the curves:
Tele style neck pickup is pretty much what you'd expect, frequencies dropping off pretty hard >2k. Takemine Mic'ed is pretty flat - louder in the bass region. Resonator is almost flat but has little peaks which I'm assuming are from the resonator's resonances
-
- KVRAF
- 1655 posts since 3 Mar, 2009 from Colorado Springs
I apologize if this were already discussed, am I understanding correctly that sound demos were not intended to be that representative of what you're hearing because you're using your monitors clipped as part of your tone in your recording setup?
Also, I always really liked that ReValver had a bunch of analysis tools built in - interesting to see the results of hosting some plugs and checking out their impulse responses, although I personally tended to fall off the deep end with it and start naval gazing after a while and not really trying to make music... I think I might have a little too obsessive of a mindset to be satisfied with so many exposed parameters, to be honest. Always an urge to tweak things more. However that's my own focus problem, not at all knocking the software for being easily tuned.
Also, I always really liked that ReValver had a bunch of analysis tools built in - interesting to see the results of hosting some plugs and checking out their impulse responses, although I personally tended to fall off the deep end with it and start naval gazing after a while and not really trying to make music... I think I might have a little too obsessive of a mindset to be satisfied with so many exposed parameters, to be honest. Always an urge to tweak things more. However that's my own focus problem, not at all knocking the software for being easily tuned.
- KVRian
- 1007 posts since 12 Sep, 2005
It's only coming through as what I'm hearing on my own speakers or anyone who has my model of speakers(Event's). You really have to try out the modules, mix and match them, to your own amp'd up speakers. I'm shopping for an extra nearfield(or other)to use as a makeshift rig. I'm pretty sure I can load another audio driver from a second USB audio interface on one computer and switch between my 2 interfaces..but I haven't tried that yet. I'm still just having fun with what this can do..and it can do a lot now. Warning though: The Randy(412) presets I put up are slightly off, as I nudged them away from the original mic position..found and fixing that nowAgreed wrote:I apologize if this were already discussed, am I understanding correctly that sound demos were not intended to be that representative of what you're hearing because you're using your monitors clipped as part of your tone in your recording setup?
- KVRian
- 1007 posts since 12 Sep, 2005
It's easy to fall off the deep end, more so now. For my test above, I tried to keep out too much push. That sounds easy, but because it's something that other software can't do, you tend to do it. That's when you need to either back off from it, or decide to somehow capture what is coming out of your system(because it WILL be unpredictable). As for backing off of the push, I tried using el cheapo headphones to make sure it at least came through ok without too many problem areas..let me know how it worked out in the example above. I'm 100% positive it'll come through better than the others. As far as software parameters, yeah it's deep..I found that you can 'install' a horn to any cab..it has 3 parameters to check to engage it internally, on any cab.Agreed wrote:
Also, I always really liked that ReValver had a bunch of analysis tools built in - interesting to see the results of hosting some plugs and checking out their impulse responses, although I personally tended to fall off the deep end with it and start naval gazing after a while and not really trying to make music... I think I might have a little too obsessive of a mindset to be satisfied with so many exposed parameters, to be honest. Always an urge to tweak things more. However that's my own focus problem, not at all knocking the software for being easily tuned.
- KVRian
- 1007 posts since 12 Sep, 2005
Ok, so using headphones to cross-check is a must. I recreated an old amp of mine, from memory. A Fender Blues Jr. This shows a bit more accurately what can be done in Revalver. All apologies for that stuff earlier, it only sounded right to me. https://soundcloud.com/spacekid1-1/blues-jrish
-
- KVRist
- 41 posts since 2 Nov, 2019
spacekid wrote: Sat Aug 20, 2016 4:12 pmThat's exactly what happens there.kohitsu wrote:I've been wondering how the ACT inputs work, so I've done a bit of investigation
I'm assuming that what happens here is there's a FFT running constantly
Revalver uses files with 'actprofile' extension, which consist of small header and 8192 points of FFT in single float format (4 bytes). Frequency step depends on bitrate though. So, for 192k - for instance - step is about 0.5*192k/8192 ≈ 11.7 Hz, max frequency obviously is samplerate/2 (192k/2 = 96k).
User input profiles stored in separate profile files, while prepared ACT profiles stored in files which in order stored in other files (with 'sa' extension). That 'sa' files are just containers, which consist of ACT profiles, images, samples and some other stuff.
Btw you can create your own prepared profile since that 'sa' format is quite easy. There are some data and files which meaning I not clear but you can use a ready profile, copy it, and change it... though it requires a bit of programming, or manual hex-data correction.
-
- KVRist
- 41 posts since 2 Nov, 2019
If someone is interested I've written a simple extractor that can extract files from Revalver .SA container files. It creates the directory with the same name as SA file +adds 'DIR' at the end of name, and it puts all files from .SA container in that directory
(By default all interesting stuff is stored in 'All Users\Peavey Electronics\ReValver 4' or 'username\Peavey Electronics\ReValver 4' if you didn't choose 'Install for all users' option when installing Revalver)
As a bonus it can load .actprofile files and draw their frequency response.

Extractor:
https://wdho.ru/4X7m
(By default all interesting stuff is stored in 'All Users\Peavey Electronics\ReValver 4' or 'username\Peavey Electronics\ReValver 4' if you didn't choose 'Install for all users' option when installing Revalver)
As a bonus it can load .actprofile files and draw their frequency response.

Extractor:
https://wdho.ru/4X7m
-
- KVRian
- 1194 posts since 27 May, 2008
If I wanted my guitar sound like something else id sell it
-
- KVRer
- 3 posts since 7 Feb, 2025
AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.
Hello, do you still have this software? Is it extracting any sort of IR file from these profiles or do I have a chance to convert the EQ responses into an IR ?ASTN wrote: Thu Dec 26, 2019 9:13 pm If someone is interested I've written a simple extractor that can extract files from Revalver .SA container files. It creates the directory with the same name as SA file +adds 'DIR' at the end of name, and it puts all files from .SA container in that directory
(By default all interesting stuff is stored in 'All Users\Peavey Electronics\ReValver 4' or 'username\Peavey Electronics\ReValver 4' if you didn't choose 'Install for all users' option when installing Revalver)
As a bonus it can load .actprofile files and draw their frequency response.
Extractor:
https://wdho.ru/4X7m (https://wdho.ru/4X7m)
-
- KVRist
- 41 posts since 2 Nov, 2019
I have some version of it. I don't even remember already what it does and why I had made it...
Anyway, I tried to open a file 'composite ir\cabs\430A 4x12.sa' and it made a folder with .ogg files and picture. My guess - those OGG-files are IR of the cab. You can use any software to convert them in WAV if needed (or you can convert it online)
![Image]()
I can try to upload the app somewhere but it's up to you to learn how to work with that crap )
Anyway, I tried to open a file 'composite ir\cabs\430A 4x12.sa' and it made a folder with .ogg files and picture. My guess - those OGG-files are IR of the cab. You can use any software to convert them in WAV if needed (or you can convert it online)
I can try to upload the app somewhere but it's up to you to learn how to work with that crap )
You do not have the required permissions to view the files attached to this post.
-
- KVRer
- 3 posts since 7 Feb, 2025
My actual intention is to extract the act input profiles actually, like guitar profiles, not the cabs, i have lots of them.ASTN wrote: Fri Feb 07, 2025 5:13 pm I have some version of it. I don't even remember already what it does and why I had made it...
Anyway, I tried to open a file 'composite ir\cabs\430A 4x12.sa' and it made a folder with .ogg files and picture. My guess - those OGG-files are IR of the cab. You can use any software to convert them in WAV if needed (or you can convert it online)
I can try to upload the app somewhere but it's up to you to learn how to work with that crap )
-
- KVRist
- 41 posts since 2 Nov, 2019
aha.
Ok, I tried to extract 'Acoustic Folk.sa' and I got bunch of files. However '*.actprofile' files crash my app...
It's midnight over here, tomorrow I'll try to open the project and see what I can do. Technically making a WAV impulse response from a full FFT is a trivial task, but I need to refresh my memories on what's under the hood of those '*actprofile' files.
I'll keep you updated.
Ok, I tried to extract 'Acoustic Folk.sa' and I got bunch of files. However '*.actprofile' files crash my app...
It's midnight over here, tomorrow I'll try to open the project and see what I can do. Technically making a WAV impulse response from a full FFT is a trivial task, but I need to refresh my memories on what's under the hood of those '*actprofile' files.
I'll keep you updated.
-
- KVRist
- 41 posts since 2 Nov, 2019
Ok.
I've checked it. Apparently 'actprofile' files DON'T have a full FFT. It's just the amplitude response without the phase one.
Technically I could make an impulse response using constant-phase approach or minimum-phase approach (which is more tricky but more realistic from physics perspective). However, it requires some time.
I've checked it. Apparently 'actprofile' files DON'T have a full FFT. It's just the amplitude response without the phase one.
Technically I could make an impulse response using constant-phase approach or minimum-phase approach (which is more tricky but more realistic from physics perspective). However, it requires some time.
