Helix DSP.3 help thread

It has been a part of the software since DSP PC-Tool v2, so pretty much any DSP they have sold has the ability. It is far more refined now, for sure, but they can all do it.
 
It's part if the pc tool software and has been out for a while now. Probably over a year

Sent from my SM-G975U using Tapatalk

Yea, I was thinking I had seen it as an option. Just wanted to make sure I could try it out on the ole DSP.2
 
Regarding the auto-EQ - someone above mentioned that it is only capable of auto-EQ'ing using 31 bands of graphical EQ - is that correct? So it won't adjust Q or freq if/as needed like a true parametric auto-EQ? Not saying that is a huge issue - just trying to understand it's capabilities.

Thanks.
 
You always want to set your gain structure for two things. The first is max output from that device to where it isn't clipping the signal. The second is max input of the next device inline. You set it for whatever requires the lowest setting. If you need 2v to your sub, set your dsp output to max 2v for those channels. If your jl amp can handle the 6v out of the dsp, set your dsp output for those channels to 6v and set the JL amp for the correct setting.

You don't want to send more voltage in to your amp than it can handle, this will overdrive the inputs on that device and could cause a bunch of issues.
Ok, so right now, with my current setup (DSR-1 DSP), I really should be reducing the sub output channels on the DSP since it's capable of 4V output and the sub is only capable of 2V input? Obviously, it's "working" as-is, even without me reducing the sub output channels on the DSP at all, but I guess I run the chance (even if remote) of the sub clipping the input signal at louder volumes. Do I have that right? If I have that right, then that means I would need to lower the output on the DSP and increase the gain on the sub (in this case the sub and sub amp are all in one box since it's an under-seat sub).

Just making sure I understand properly.

Thank you!
 
It has been a part of the software since DSP PC-Tool v2, so pretty much any DSP they have sold has the ability. It is far more refined now, for sure, but they can all do it.

OH boy oh boy oh boy! Sounds like my Dayton will be out when I do the final install!
 
Ok, so right now, with my current setup (DSR-1 DSP), I really should be reducing the sub output channels on the DSP since it's capable of 4V output and the sub is only capable of 2V input? Obviously, it's "working" as-is, even without me reducing the sub output channels on the DSP at all, but I guess I run the chance (even if remote) of the sub clipping the input signal at louder volumes. Do I have that right? If I have that right, then that means I would need to lower the output on the DSP and increase the gain on the sub (in this case the sub and sub amp are all in one box since it's an under-seat sub).

Just making sure I understand properly.

Thank you!

We're talking the maximum in and the maximum out, so only at the absolute peak of a transient would the DSP overdrive the amp, assuming the DSP was set at its very max as well. This is one of the best problems to have, btw. Excessive gain from your source is just confidence that you will not have to bring your gains up much. In a musical situation, the chance of this peaking over is extremely slim.
 
Just finished up a tune on my own car after doing a new install. Decided to try the Auto EQ before I did anything like I said. I was able to set crossovers, inouts/outputs, take, go into the rta and make my own curve, tune left, tune right, tune sub, tune global... all in 15 minutes. I would have zero shame in giving the end result to a customer, where I usually spend 2+ hours tuning. Verified the measurements and adjustments with rew... to a T. I'll get a video up shortly.

Sent from my SM-G975U using Tapatalk

So did the Helix auto-EQ simply make graphic EQ adjustments? That’s what I got when I played with it. Nothing wrong with that, if that’s the case, just curious if I did something incorrectly.

BTW Nick, I sat through the Helix seminar at KnowledgeFest (with Julian & Florian) too. Recognized you there but failed to introduce myself & say hello. You gave me some phone support several months back on an Illusion C8 with shifted motor.

Thanks!
Ross
 
Regarding the auto-EQ - someone above mentioned that it is only capable of auto-EQ'ing using 31 bands of graphical EQ - is that correct? So it won't adjust Q or freq if/as needed like a true parametric auto-EQ? Not saying that is a huge issue - just trying to understand it's capabilities.

Thanks.

That was me. I’m still waiting for confirmation that the Auto-EQ is limited to graphic EQ changes only, or if there’s a setting to enable parametric changes in the Auto-EQ. When I attempted using it several weeks back, it gave good results sound-wise, but but all changes were at the standard 31 preset center frequencies and no Q adjustments were made.
 
When I ran my helix dsp i used the auto eq all the time and it really does a damn good job i absolutely loved it


Sent from my iPad using Tapatalk
 
It's part if the pc tool software and has been out for a while now. Probably over a year

Sent from my SM-G975U using Tapatalk

Its definately been out over a year maybe even two because I used the auto eq feature probably two yrs ago at least because thats when i had helix rta mic


Sent from my iPad using Tapatalk
 
Is there any need for a FiX86 to be used when using the dsp.3 and my oem HU? It sounds like the helix input signal analyzer would take place of the FiX? Does the dsp.2 have this feature?
 
Is there any need for a FiX86 to be used when using the dsp.3 and my oem HU? It sounds like the helix input signal analyzer would take place of the FiX? Does the dsp.2 have this feature?
You would not need it if you are familiar with using an rta and how they/an audio signal works. The dsp.2 does not have this feature. Only dsp's with the ACO platform

Sent from my SM-G975U using Tapatalk
 
You would not need it if you are familiar with using an rta and how they/an audio signal works. The dsp.2 does not have this feature. Only dsp's with the ACO platform

Sent from my SM-G975U using Tapatalk

I’m sort of familiar I guess but by no means an expert. Never played with RTA but that was my next step. I’d rather have everything and be able to work with it than not.
 
This thread is turning into a great Helix DSP.3 help thread - I really appreciate all of the assistance so far!

With that said, I have a few more questions for you guys....

1. How can I make "global" EQ changes that will affect either all or only a subset of channels when the PEQ filters are completely different between the channels? Is there any sort of "global EQ" function in Helix? I thought that maybe I could use input EQ for this, but it doesn't look like input EQ is available for optical inputs?

2. Is there anyway to import multiple REW EQ filter export files without overwriting previously imported REW filters? Unfortunately, since REW only supports 20 filters (hate that!), I sometimes have to export multiple sets of filters from REW (20 the first time and then up to 11 the next time). It seems like the Helix "R" import function just overwrites any previously-defined filters every time, so there is no way to import more than 20 REW filters "directly"... It would be nice if it would just start the import from the filter that you currently have selected - so I could highlight the 21st filter in the Helix app and then press "R" and it would import starting at the filter I currently have selected. Otherwise, I guess I can manually combine the two REW filter export files...

3. Physical connectivity question for the DSP.3 power/ground wires. Can I just get the DSP.3 power/ground directly from my power/ground connections at my amp (put the DSP.3 power/ground wires into the amp power/ground "ports" along with the amps existing power/ground wires. Or should I only do that with the power wire and run a separate ground wire? Obviously, I'd put a separate 5A fuse on the power wire as well...

Thank you! I'm going away for work next week, so I'm going to wait until I get back to do the install (don't want to risk having any unexpected audio issues during my trip!), so I'm just trying to figure out a few remaining things... :-)
 
With that said, I have a few more questions for you guys....

I can answer a couple of those:

1) I believe you simply Link which ever channels you wish to EQ globally together with each other then make the EQ changes.

2) Not really sure on importing REW filters, as I always hand-input. Sorry:hmm:

3) Yes, you can simply piggyback off the amp's power & ground connections, using there input terminals as sort of a junction, b/c the DSP draws virtually no current.
 
I can answer a couple of those:

1) I believe you simply Link which ever channels you wish to EQ globally together with each other then make the EQ changes.

2) Not really sure on importing REW filters, as I always hand-input. Sorry:hmm:

3) Yes, you can simply piggyback off the amp's power & ground connections, using there input terminals as sort of a junction, b/c the DSP draws virtually no current.
Thank you.

For #1, I do realize that I can link channels and changes will be made to both channels in a "relative" way, but lets say, for example, that after all of my individual channels are EQ'd with PEQ filters (and I don't have any/enough free on all channels to add new PEQ filters for this function) and I wanted to cut 2500hz-5000hz by a few DB on all channels (let's say front and rear speakers in this case - for "tonality" changes) - is there any way that I can globally alter that frequency range across those channels without interrupting the per-channel PEQ filters (which may or may not be defined for the range of frequencies that I want to alter)?

On my DSR-1, each channel has a 31-band PEQ, but there there is also a 7-band "global" EQ that affects all channels. I'm looking for that type of functionality with the Helix. If I were using Analog inputs, I think I could do that with the input EQ (which is separate from the rest of the EQ), but it looks like a digital input doesn't have the input EQ available.

Thanks again!
 
This thread is turning into a great Helix DSP.3 help thread - I really appreciate all of the assistance so far!

With that said, I have a few more questions for you guys....

1. How can I make "global" EQ changes that will affect either all or only a subset of channels when the PEQ filters are completely different between the channels? Is there any sort of "global EQ" function in Helix? I thought that maybe I could use input EQ for this, but it doesn't look like input EQ is available for optical inputs?

You really shouldnt need to be using any more than 10 or so bands of parametric eq per channel if you are 2 way and eq'ing efficiently and not "beating the response into submission". What I do when tuning is use the first bands of eq minus the first 2 (40- whatever) to eq the individual driver. Then start at the opposite end to do eq for a whole side. Then the bands in the middle for doing global. OR, you can see my new post about the helix rta and Auto EQ and use the graphic bands when it's done and not have to worry about them being different.


2. Is there anyway to import multiple REW EQ filter export files without overwriting previously imported REW filters? Unfortunately, since REW only supports 20 filters (hate that!), I sometimes have to export multiple sets of filters from REW (20 the first time and then up to 11 the next time). It seems like the Helix "R" import function just overwrites any previously-defined filters every time, so there is no way to import more than 20 REW filters "directly"... It would be nice if it would just start the import from the filter that you currently have selected - so I could highlight the 21st filter in the Helix app and then press "R" and it would import starting at the filter I currently have selected. Otherwise, I guess I can manually combine the two REW filter export files...
Not that I know of. You also should not need to use all 31 bands of parametric eq for a single driver. Not even close


3. Physical connectivity question for the DSP.3 power/ground wires. Can I just get the DSP.3 power/ground directly from my power/ground connections at my amp (put the DSP.3 power/ground wires into the amp power/ground "ports" along with the amps existing power/ground wires. Or should I only do that with the power wire and run a separate ground wire? Obviously, I'd put a separate 5A fuse on the power wire as well...

Thank you! I'm going away for work next week, so I'm going to wait until I get back to do the install (don't want to risk having any unexpected audio issues during my trip!), so I'm just trying to figure out a few remaining things... :-)

Yes, you can

Sent from my SM-G975U using Tapatalk
 
You also should not need to use all 31 bands of parametric eq for a single driver. Not even close

I always see people say that you should never use more than 10 bands of EQ... Then why the hell is it such a good feature to have 31-bands of parametric EQ per channel?!?! I mean if you have the bands available, why *wouldn't* you want to get your response to match the target curve as closely as possible? I thought that was the whole point of a target curve? In my case, my 3.5" dash speakers play 500hz - 20khz - that's a pretty big range to flatten out, so I frequently find that 10 bands isn't enough, especially if I'm using REW to EQ down to a flatness target of 1 - which again, why wouldn't you? I guess I'm not understanding something here.

Same for my rear channels - I EQ those as well (again, why not?) - and I have them playing a significant frequency range (I'm using 75hz - 4khz) - so again, sometimes, more than 10 bands are needed to really flatten the curve.

In both of these cases, REW will use all 20 bands more of the time, especially with a flatness of 1 using a smoothing of 1/12th - and still have issues to correct even after 20 bands.

Now for a sub or even a midbass, I can completely understand maybe only needing 10 bands - but even those, you can easily use more to match the curve more closely.

So you guys just don't use more than 10 bands - ever? For my dash and rear deck speakers, I can actually use 31 bands to meet my target curve and still not have things "perfectly" matching my target curve (using REW auto-EQ). :-)

But I'm new at this - please help me understand - is it actually "bad" to match a target curve as close as possible?

That is unfortunate that Helix doesn't have any kind of "separate" global EQ - that will actually be a step backwards from the DSR-1 in terms of functionality. Not the end of the world, but....

Thank you!

EDIT: By the way, I am not trying to "shoot the messenger" - just trying to understand what is considered "good" EQ and what is "too much". I've used up to all 31 bands for my 3.5" coaxial speakers and they sound great - just can't imagine that having a more "jagged" response would make it sound better - but maybe it will? Should I only use REW to tune to a flatness of 2? or 3?
 
I always see people say that you should never use more than 10 bands of EQ... Then why the hell is it such a good feature to have 31-bands of parametric EQ per channel?!?! I mean if you have the bands available, why *wouldn't* you want to get your response to match the target curve as closely as possible? I thought that was the whole point of a target curve? In my case, my 3.5" dash speakers play 500hz - 20khz - that's a pretty big range to flatten out, so I frequently find that 10 bands isn't enough, especially if I'm using REW to EQ down to a flatness target of 1 - which again, why wouldn't you? I guess I'm not understanding something here.

10 is just an arbitrary number. Its not that you SHOULDNT be using X amount of bands. But needing a lot hints that something is wrong with the install and EQ'ing is not the answer. That said, your dash speakers do need to cover a very wide passband, so its obviously going to need more than say a typical midrange. That said, the car will also dictate that but i highly doubt itll need any more than 15. If it genuinely needs more, thats fine. But needing a ton of EQ hints at problems elsewhere. The reason why the helix's 30 parametric bands are so nice is not because of the number itself, but because what the helix software can do with them that no other dsp does... relative linking and adjustment, which is kinda what my last post was talking about. Just an FYI, you dont need to match your target curve to a T. You also dont need 1/12 octave smoothing. Not saying you are doing this but I know a lot of people do this and its just not necessary. Again, maybe save this tune, wipe the eq, and try the auto tune. check my recent thread about that.



Same for my rear channels - I EQ those as well (again, why not?) - and I have them playing a significant frequency range (I'm using 75hz - 4khz) - so again, sometimes, more than 10 bands are needed to really flatten the curve.

In both of these cases, REW will use all 20 bands more of the time, especially with a flatness of 1 using a smoothing of 1/12th - and still have issues to correct even after 20 bands.

wow, looks like i predicted right lol. I didnt even read this far when writing the above lol. try smoothing to 1/3, and mathcing to 2db. or again, running the helix auto eq. also, you shouldnt be running your rears down to 75. run them from 250-4k, 12db slopes with a differential signal.



Now for a sub or even a midbass, I can completely understand maybe only needing 10 bands - but even those, you can easily use more to match the curve more closely.

stop worrying so much about the curve. worry more about getting left to equal right. Relying on the curve is fine for beginners. But push to get better, unless you are content and dont care, but you seem like you want to get better at this.



So you guys just don't use more than 10 bands - ever? For my dash and rear deck speakers, I can actually use 31 bands to meet my target curve and still not have things "perfectly" matching my target curve (using REW auto-EQ). :-)

But I'm new at this - please help me understand - is it actually "bad" to match a target curve as close as possible?

again, dont put so much focus into matching the curve. Its not necessary. focus more on making things match from left to right, knocking down peaks, and making things linear.


That is unfortunate that Helix doesn't have any kind of "separate" global EQ - that will actually be a step backwards from the DSR-1 in terms of functionality. Not the end of the world, but....

It doesnt, and its something ive wanted for a bit now, but tbh, its not that big of a deal, unless you are relying purely on rew's auto eq with a not-so-smoothed-out measurement (like 1/12 octave). But see my other posts. You can use the bands to your advantage and set them up in a way so that you have global eq at the end.

Thank you!

EDIT: By the way, I am not trying to "shoot the messenger" - just trying to understand what is considered "good" EQ and what is "too much". I've used up to all 31 bands for my 3.5" coaxial speakers and they sound great - just can't imagine that having a more "jagged" response would make it sound better - but maybe it will? Should I only use REW to tune to a flatness of 2? or 3?

just stop doing auto eq with 1/12 octave. Thats absurd, especially at high frequencies. Last time i tuned my car with rew (before i changed my install), i had a total of something like 8 bands of eq on my whole system. The reason i had so few, and i realized i forgot to mention this above.. using levels and crossovers to get the response as close to a "target" as possible, then using 1/2 octave smoothing to knock down the peaks. Thats it. Best, and most consistent that setup ever sounded. So yeah, you should be using levels and crossovers to get the acoustic response as close to the target before doing any eq. that will save you time and eq bands.

But again, check this out. it will use graphic eq and you can link everything and adjust relatively after the fact.

https://www.caraudiojunkies.com/sho...Car-Audio-Junkies-Exclusive&p=46563#post46563
 
Back
Top