Demo FIR filters

mops911

Registered
Thread Starter
Joined
Jul 27, 2025
Posts
3
Hello

I am contemplating to jump into DRC. I use HQP and wonder how much more CPU the convolver requires.

Do filters typically come with 65k taps? And does HQP upsample it for 352k music files? Also for DSD, I understand that HQP converts to PCM, convolves and then converts back to DSD. Correct?

Could anybdy share some "neutral" or "null" filters so I can check out the additional load of HQP?

Much appreciated
T
 
You can create dummy filters with the Audiolense demo. The dummies attenuate the sound. That’s all they do.
 
Hello

I am contemplating to jump into DRC. I use HQP and wonder how much more CPU the convolver requires.

Do filters typically come with 65k taps? And does HQP upsample it for 352k music files? Also for DSD, I understand that HQP converts to PCM, convolves and then converts back to DSD. Correct?

Could anybdy share some "neutral" or "null" filters so I can check out the additional load of HQP?

Much appreciated
T
Try asking over at audiophilestyle.com. Guys over there have lots of experience with this stuff.
 
HQP scales the filters to DSD and convolves in DSD, as Jussi from HQP has explained to me. Yes, you can upscale 352K PCM material to DSD if you like. I have done all of my Audiolense filters at 352K and HQP takes it from there. Very CPU/GPU intensive particularly for 5.1 MCH and 5.1.2 Atmos material. I run an i9-10900K CPU and an EVGA GeForce RTX2080 Ti XC GPU and I can’t scale DSD64 5.1 MCH material to DSD128 with HQP heavy modulators and Audiolense filters. JCR
 
Further to the preceding message, I'm wondering if there's a way to upscale the filters for PCM and save them so they can be used with players other than HQPlayer.

Would any upscaler that could save files work for that purpose or is there a special method for doing this?

I'll be grateful for any direction anyone has to offer.

Joel
 
Further to the preceding message, I'm wondering if there's a way to upscale the filters for PCM and save them so they can be used with players other than HQPlayer.

Would any upscaler that could save files work for that purpose or is there a special method for doing this?

I'll be grateful for any direction anyone has to offer.

Joel

AL can save the filters at various sample rates when you create/save them. Is that what you are asking about?
 
HQP scales the filters to DSD and convolves in DSD, as Jussi from HQP has explained to me. Yes, you can upscale 352K PCM material to DSD if you like. I have done all of my Audiolense filters at 352K and HQP takes it from there. Very CPU/GPU intensive particularly for 5.1 MCH and 5.1.2 Atmos material. I run an i9-10900K CPU and an EVGA GeForce RTX2080 Ti XC GPU and I can’t scale DSD64 5.1 MCH material to DSD128 with HQP heavy modulators and Audiolense filters. JCR
Are you sure it convolves in DSD? I was under the assumption that nothing can be done in DSD. And all DSD goes to PCM, colvolves there in back to DSD ...
 
AL can save the filters at various sample rates when you create/save them. Is that what you are asking about?
Ideally, I dont want to configure anything during playback, meaning not changing the filters. But that HQP handles all rates automatically. My output is fixed to DSD256 and no matter the input, it should be all done automatically. Possible?
 
AL can save the filters at various sample rates when you create/save them. Is that what you are asking about?
I know it can, but I'm not sure whether it upsamples the existing filters or only saves the filters at their original bit and sample rates.

I thought it only saved the filters as they are.

Thanks for your quick reply.

Joel
 
Last edited:
ChatGPT indicates that the sample rate of the filter should not be converted by an external upsampler to maintain phase and whatever else might be affected.

Still, it would be good to know what other options might be available to upsample the filter.

Joel
 
ChatGPT indicates that the sample rate of the filter should not be converted by an external upsampler to maintain phase and whatever else might be affected.

Still, it would be good to know what other options might be available to upsample the filter.

Joel
Pretty sure the whole point of adding the re-sampler functionality was to provide the means for AL to upsample while maintaining correction as best as possible. Will phase change? I suppose it could. Will you hear it? Doubtful.

If you want another option for resampling filters, if you haven't already seen it, take a look at https://accuratesound.ca/hang-loose-convolver-hlc/
 
Further to the preceding message, I'm wondering if there's a way to upscale the filters for PCM and save them so they can be used with players other than HQPlayer.

Would any upscaler that could save files work for that purpose or is there a special method for doing this?

I'll be grateful for any direction anyone has to offer.

Joel
You can do that in Audiolense
 
To follow-up on my earlier inquiry, what about increasing the number of taps employed by the filter beyond 65K?

Some may feel this is unnecessary also placing an added burden on the CPU.

Still, if I want to try this, is it possible and, if so, how would I do it?

Thanks in advance.

Joel
 
To follow-up on my earlier inquiry, what about increasing the number of taps employed by the filter beyond 65K?

Some may feel this is unnecessary also placing an added burden on the CPU.

Still, if I want to try this, is it possible and, if so, how would I do it?

Thanks in advance.

Joel
You can increase the number of taps in the filter design tool up to 132K.
 
Thanks Ofer.

I should have said 131K in my post.

I'm talking about creating a filter with well beyond that number, again, understanding that increasing the size places a greater load on the CPU.

Joel
 
Thanks Ofer.

I should have said 131K in my post.

I'm talking about creating a filter with well beyond that number, again, understanding that increasing the size places a greater load on the CPU.

Joel
Think you can manually enter anything you like but there might be limitations. A filter can only be so big.
 
This might be a fun read for you all... http://archimago.blogspot.com/2021/07/mega-taps-upsampling-remasteros-pan.html

I use Mitches Hang Loose Convolver for Digital Room Correction, using FIR filters, as a VST in Audirvana... Its really nice and flexible with an upscaler for the filters, or it will use multiple filters if your happen to make the filters in all your favorite frequencies...
 
Your welcome @joelha, have played with upscaling a little bit, tired PGGB, have million+ taps mods to SOX, use a Chord Quest with 49,152 taps on board and can override with millions of taps embedded at 768kHz... Fun to play with and not much small signal difference other than bloated FLAC files... Lol...

I do like that the HLC allows 6 filter banks, that can switch in virtually realtime, to A/B your various DRC filters...
I also use various mastering quality, old skool equalizers, to add a little mojo to taste, beyond the B&H room curve I use for the DRC stage of my signal chain...
 
Last edited:
Thanks again for your post, ddude003.

So you think adding more taps hasn't been worth the trouble or it's worth the trouble but expectations should be managed about the degree of improvement?

Joel
 
Again, your welcome @joelha... I think there is a point of diminishing returns... And, you might find that for yourself and your kit... Do as you like and find your own way... Every ear and brain is different per individuals as is your kit... Vive la difference...
 
Back
Top