Please ensure Javascript is enabled for purposes of website accessibility Jump to content

HX Effects woes - inability to dial out 8k+ issues (long read)


z3albw1rr
 Share

Recommended Posts

I've been doing some experimenting with why the HX FX, well, sucks, and here's what I've got so far:

 

1. The following image shows the difference between a direct guitar signal and the signal coming out of the HX Effects (two of them in a row in this case) with both OFF and in True Bypass mode. This is done using Logic's "Match EQ" where you pull in a reference file, and then a file to compare it to, and it shows you what you would need to do to make them the same.

 

You can see that there is not very much difference in the signal - could be attributed to the devices, or one additional 18' cable:

 

296747238_DirectvsHXFXOff..thumb.png.bccf8ffe7ba9cd60a862d2016ad9cde0.png

 

2. This next image is of the HXFX ON this time, but bypassed in True Bypass Mode:

 

510741219_Directvs.HXFXTBMode.thumb.png.e9d79f0a2efeaa36509e0b8462776529.png

 

You can see there's not much difference as one might expect since True Bypass with the device or or off should pass the same signal.

 

3. The next image is with the HX FX on, with all the effects in a path bypassed. Here is where a huge problem starts to show itself. Notice that everything above 8k is a problem. Match EQ here is telling us that in order to make the signal going through HXFX be the same as that of the signal not going through the device, we would need to CUT anything above 8k by about 6-7 db (note: everything from about 6k down appears to be boosted, and what this means is the output of the HXFX when on is just slightly below unity):

 

49291893_Directvs.FXbypassed.thumb.png.5ce92d3b379ac4d9ed52e80d9fdf634b.png

 

4. This final image shows where the problem becomes an even bigger issue:

 

93111474_Directvs.6khicut-12db.thumb.png.d4507d071e11425d5da40a1adb9c2b0c.png

 

This is the Hi/Low Cut Block with everything above 6k cut -12db - which is the max.

 

Notice that everything above 8k is basically the same as before.

 

That means THIS EQ IS NOT AFFECTING THOSE FREQUENCIES.

 

And it's not just this EQ. In fact, I set it to 8k first, and there was little difference. I used 6k here to show that it's working - because we can assume 6k is OK based on the previous image and since we've cut that too (and it's a shelf EQ, which has a slope) it's telling us we'd need to BOOST 5k and 6k ish - which makes sense because it's getting cut by the EQ.

 

But everything from 8k up SHOULD BE doing just like 5 and 6 K - it should be ABOVE the line, NOT below it!

 

I've repeated the experiment with a number of different EQs and a number of other effects set to clean or as transparent as they can be.

 

In every case, this same "problem area" is present.

 

Now, I can use a Channel EQ in Logic and correct it there.

 

So what that tells me is that the HX Effects is adding this POST effects - it's doing it on the Output of the device.

 

Because if you dial all the 8k and above out with an EQ, it shouldn't be there, right? But it is - it's a little different as you can see in the 6k example, so that means you're not feeding the signal in there - but SOMETHING is creating that problem area - and since none of the blocks can significantly affect it, it means it's getting after the blocks.

 

Now, since I have 2 HX Effects, I checked this with both of them together, and individually. IOW, it's not that I'm cutting 8k on the first HXFX and then it's getting added back on the 2nd one. It happens alone or in tandem.

 

Basically, there is no way to get rid of this 8k+ 6-7 db boost in frequencies without putting a physical EQ after the devices (which will have to be part of my next set of experiments).

 

Now, to be transparent, Guitar Speakers don't necessarily all produce frequencies above about 5k equally. But many do get up into that 8k range and some in the 10k range. And anyone who knows how the Equal Loudness Curves work knows that a 6db difference between 5 and 8k at an overall volume of 40db is not going to be the same as at 80db (the kind of levels we often play at).

 

[also: note, while these are not going through guitar speakers, they are a direct comparison at the input of an interface into a DAW, so they are 1:1 in that regard - this is "what's coming out of the HX FX"]

 

What that means is that high frequencies that might not be a problem at home volumes suddenly become "hissy" "fizzy" and just "harsh" mess at a higher volume.

 

Anyone who says "well you're not going to hear that anyway" is wrong. I can already hear an audible difference in the tone between images 2 and 3 - which is no effects on (when playing through an amp). It sounds "thinner" and "weaker" (most likely because of the overall volume drop below unity) and has a harsh high end - "harsh" or "tinkly" or whatever kind of words you want to use to describe it - basically, "not as good/full as the unaltered signal".

 

Now my goal thus far has been to "correct for the tone difference".

 

I've seen this many times before on forums and I'm sure anyone still reading has too: "there's this digital harshness you just can't dial out".

 

Well now I see why - you can't actually dial it out. I mistakenly thought, "oh I'll just slap an EQ on there and get rid of those high frequencies I'm hearing".

 

Oh, and BTW, I'm in my 50s, and my ears are shot, and I've been using all kinds of things for 35 years of playing, and am no stranger to Line 6 products either - I was selling AxSys 212s in music stores when they first came out.

 

But I can still hear this difference pretty easily. I can't imagine what it must be like for you poor souls who can still hear high frequencies - you're like dogs cowering in fear!

 

I hate to have to add an EQ pedal behind the FXs, but it looks like that's the solution.

 

I will pass this through some Amp/Cab models in Logic to see what they do, and I'll eat my hat if it gets rid of everything. I'll do it with a live amp and mic too (now that I've narrowed down what to compare). And I'll try an EQ behind it as well just to see what happens - and if my ears agree with my eyes.

 

 

 

 

 

 

 

 

 

 

 

Link to comment
Share on other sites

This got me curious because I've done a lot of spectral analysis with Helix (some would say a disturbing amount, LOL), and never noticed these kinds of anomalies. So I double-checked, just in case. The high shelf and hi/low cut do what they're supposed to be doing - cutting high frequencies when you set them to cut high frequencies.

 

I don't know how Logic's EQ matching works, but here's a wild guess...if there's EQ involved, they're probably not using phase linear filters, so there could be phase shifts that alter the response when summed. Higher frequencies would be more susceptible to phase issues. But let me emphasize this is 100% a guess. All I can vouch for is that pink noise and spectrum analyzers are designed specifically to show frequency response, and what they show is what I would expect to see.

 

  • Like 2
  • Upvote 2
Link to comment
Share on other sites

On 7/22/2022 at 11:42 PM, craiganderton said:

This got me curious because I've done a lot of spectral analysis with Helix (some would say a disturbing amount, LOL),and never noticed these kinds of anomalies. So I double-checked, just in case. The top image is pink noise going through the high shelf at 2.2 kHz, with maximum cut. The lower image is of the hi/low cut at 5 kHz with maximum cut. It looks like they're doing what they're supposed to be doing - cutting high frequencies when you set them to cut high frequencies.

 

 

 

Remember that I'm comparing a direct signal to that going through HXFX. In your experiment above it looks like you're just comparing two things already in Helix - If I were to do this it would show the exact same thing - but I'd be comparing a file from HX to another file from HX, not a file that doesn't include HX in the path at all.

 

 

 

Quote

I don't know how Logic's EQ matching works, but here's a wild guess...if there's EQ involved, they're probably not using phase linear filters,

You'll notice the bottom right of each image has a setting for Phase which is set to Linear.

 

Quote

All I can vouch for is that pink noise

Good point about the Pink Noise and really I should repeat the experiments by sending that through.

 

However right now we're talking apples and oranges - I'd get the exact same result if I ran pink noise through the two EQs you used with those settings.

 

What I'm doing though is comparing pink noise straight into an interface versus pink noise through the HX then into an interface.

 

When the HX is in the path, it adds 8k+ frequencies to the signal as compared to the direct signal.

Link to comment
Share on other sites

On 7/22/2022 at 11:42 PM, craiganderton said:

This got me curious because I've done a lot of spectral analysis with Helix (some would say a disturbing amount, LOL),

 

Since I'm obviously starting to follow in your disturbed footprints, here are a couple more pictures :-)

 

1. This shows the Direct signal (which is either no HX FX in the path, or the HX FX in the path but either turned off, or set to Analog Bypass Mode and bypassed - which are all basically the same as my 1st image in the 1st post) as compared to the Kinky Boost with it set to no boost and no bright switch:

 

1789005423_Directv.Kinky.thumb.png.ffa51d9f23ea31d757cbae4f0a1e99ed.png

 

Now you'll notice here that the Kinky Boost is lessening the frequencies starting about 2k.

 

Now, this is pretty evident when you listen to it - in fact, you can throw a Drive in the path, and put the KB either before it or after it - and when you put it after it you can hear how much of the highs it shaves off. So that's what that pedal is supposed to do.

 

But, notice this "problem area" from 8k up is an issue still.

 

2. This image is the bypassed path with HX FX on, compared to the KB again:

 

1266318219_Bypassvs.Kinky.thumb.png.52053fd3b3bd426ac0aefbdfac325e4f.png

 

See the difference? This is what the KB is doing to the "dry" signal "in the box" - which is what it should do. It's basically cutting things from 2k up.

 

If you put the KB in an empty path, it's a bit louder overall - and that's evident by this saying you'd have to turn it down from 20 up to about 1.5k (i.e. the line above the midline is showing us what we'd have to do to compensate to get it to sound like the dry signal)

 

So this graph agrees with everything you'll hear if you compare a dry signal in HX FX to having the KB in the path - it'll be slightly louder over all, and start cutting off the high freqs around 2k.

 

But image #1 is what concerns me - comparing a Dry signal WITHOUT HX FX on to one with it on.

 

So you turn it on, load up whatever, and it "does what it's supposed to do", but there's this extra boost from 8k+ that is so strong it counteracts and even negates what the block is supposed to do.

 

Everything in the 2nd image that's above 8k that is above the line is like the rest of that image above the line - they're high frequencies that are being CUT by the block. But in the 1st image the block is only able to do that for frequencies below 8k - once it reaches 8k it is now boosting those frequencies.

 

So SOMETHING is happening within the device that is boosting 8k+ frequencies as compared to the direct signal, even if you include any block that should cut those frequencies - it does cut them absolutely within the box as compared to the dry signal in the box, but that dry signal in the box already has that extra boost in those frequencies.

 

Thus when you compare it to a direct signal, those frequencies are STILL boosted no matter what you do with the effects in the box.

 

It strikes me as insane that an effect block like a Low Pass Filter or anything like that shouldn't attenuate the frequencies above 8k.

 

It's like it works as expected below 8k but when you cross that threshold it basically ignores it - in essence it's "adding it back in" despite the blocks cutting it.

 

Now, I'd be fine calling this a potential anomaly in the Match EQ - since it's so consistent. And even if HX FX was adding frequencies above 8k at the output, you'd think feeding it a highly cut signal would bring it up, but not THAT much - I mean the result would look more like the 2nd kinky boost image instead (and BTW, I haven't discounted that the upper stuff on the KB signal could in fact be just that - this 8k+ stuff being added back in and "fighting" with the KB's upper signal - maybe the KB should look more like a typical LPF attenuation of the upper freqs - a smooth decrease from 2k up with a 6 or 12db /octave slope).

 

But the issue here is, I hear it. And I heard it first. And what's happening visually really does represent what I'm hearing. So that leads me to believe this is not some glitch in how Logic's Match EQ is reporting, or how I'm doing the experiment and so on.

 

EDIT: I did just repeat the experiment using Tonal Balance Control which gives you a visual line of the frequency content present. The lines read IDENTICAL between images 1 and 3 in my original post. So that counters this whole theory. I don't have a Spectrum Analyzer I can "freeze" so I'll need to do that with Pink Noise that can remain more consistent (my original clip is a Guitar open E chord). There is a spectrum analyzer in Logic's EQs and it doesn't show much energy in the signal above about 2k and TBC is showing there's energy up to 6k - both well below this 8k "glitch".

 

So either Match EQ is hearing something TBC and the spectrum analyzers don't or can't represent, or the Match EQ is adding something faulty. Again though, the problem is, my ears are hearing non-identical signals, so that they show identical in TBC would lead me to believe it's not as precise as Match EQ...

 

I'm still prepared to eat my hat on all this though...

 

 

Link to comment
Share on other sites

On 7/23/2022 at 12:58 PM, z3albw1rr said:

I'm still prepared to eat my hat on all this though...

 

I've discovered over the years that eating one's hat is best complemented by a good Cabernet Sauvignon :)

 

All I could really do to replicate your setup was to test whether "That means THIS EQ IS NOT AFFECTING THOSE FREQUENCIES." There are a lot of variables in your setup and I don't know what they are. 

 

If I get the time I'll take a pink noise generator, blast it into an interface direct and look at the spectrum, then blast it through a bypassed Helix and see what happens. I suspect the frequency responses will be the same.

  • Like 3
Link to comment
Share on other sites

On 7/23/2022 at 3:19 PM, craiganderton said:

 

I've discovered over the years that eating one's hat is best complemented by a good Cabernet Sauvignon :)

 

All I could really do to replicate your setup was to test whether "That means THIS EQ IS NOT AFFECTING THOSE FREQUENCIES." There are a lot of variables in your setup and I don't know what they are. 

 

If I get the time I'll take a pink noise generator, blast it into an interface direct and look at the spectrum, then blast it through a bypassed Helix and see what happens. I suspect the frequency responses will be the same.

I repeated the experiment today with an SM57 in front of the amp - everything else the same.

 

Some interesting results.

 

First, here's the Bypassed versus "on but all effects bypassed" image:

 

1195900224_WithSM57.thumb.png.7bae5938311349b4662c2e31e757d845.png

 

You can see the "blurble" is now down around 5k, and tapers off earlier.

 

I suspect this is of course due to my putting an SM57, in a certain position on a speaker - both of which are not full range response devices.

 

Now THE REALLY ODD THING this time was, if I put an EQ on it actually would counteract it as you'd expect.

 

I don't have an image of it but when I put an EQ on and cut 8k - and assuming a certain bandwidth around it (not sure what that number is though - the default is 7, and I put it on 1, thinking 1 octave bandwidth) with a cut of -10db.

 

I got an almost flat line - I cut too much and went back and tried -8db and that was even better - there were little bumps here and there in those high frequencies but we're talking maybe half db or 1 db ripple - at those frequencies and amounts you're not going to hear those.

 

And audibly, the bypassed tone was WAY closer to the engaged (but with all blocks bypassed) tone - which makes perfect sense.

 

I could tweak it some more, and will, but I already got fairly close just trying to reverse engineer from the graph.

 

Then I tried drives before the EQ - OMG, SO much better.

 

I just thought the things like the Boss OD and SD were horrible. Super scratchy and hissy/spitty/fizzy.

 

Now pretty much every single thing I ran through sounded like I would have expected it - and the legacy effects now sound more identical to the ones in the M13.

 

I'm not sure why the EQs wouldn't affect that higher frequency 8K+ in the other tests, and why they would here, but still they do reflect visually what I'm hearing audibly (and I suppose what the mic is hearing too).

 

This is probably why I've run across a few people saying they just run a Kinky Boost at the end of their chain and is just makes it all sound better - well yeah, it's cutting all those highs in the non-bright setting.

 

It seems though from my previous experiments a simple Low Pass/ High Cut won't do it. The bottom end seems to need a little oomph too - note the slightly bigger hill down around 35 hz.

 

I'm going to tinker with some EQs and see what I can do to "repair" my tone.

 

Since I use a Volume pedal end of chain, what I might just do is put an EQ in place of the volume pedal, and control it's overall level with the expression pedal to basically do the same thing. Since I'm likely going to have to use an EQ block to "fix" things, might as well use it for the Volume pedal too - don't waste a block on it that way.

 

 

Link to comment
Share on other sites

I can see that you're having a lot of fun with this and not to be a killjoy, but is there some reason why the average HXFX user would (or should) care about all of this?

 

I don't own an HXFX, but all of this should be the same for any of the Helix family, maybe even x2 for the Floor/LT/Rack, and neither my Floor nor my HXS (essentially an HXFX with amps) sucks at all, even when used without Helix amps, as is the use case for the HXFX.

 

Seriously, wouldn't your time be better spent playing your guitar?

 

I'm pretty sure that there's no prize at the bottom of this Cracker Jack box...

 

You'll probably think that I'm being snarky, and maybe I am, just a little. But I'm not trying to hurt your feelings, I'm asking a serious question here...

 

Why should anyone care?

 

  • Upvote 3
Link to comment
Share on other sites

No offence, but this looks like a huge waste of time to me.

 

If you don't like the sound you can get out of the Helix, just move on, sell it and buy something else as digging at that level is going nowhere in terms of possible solutions.

 

It's like when you buy a new guitar, and you don't like something in the core of it, like the frets, or the tone, or even the color....and you start swapping parts and wasting days, weeks...and money, trying to change the nature of the beast, until that day you get bored and you flip it. When there is something that bugs you that level it will never go away because of the grown bias affecting everything you'll be doing with it.

 

100% no offence, just my opinion. :)

 

Then, if you care of my POV on the specific subject; I never heard something that wrong as you are reporting, really. :)

  • Upvote 1
Link to comment
Share on other sites

IMO... if you need to do that much "fixing", something else is seriously wrong.

Have you tried a 2nd EQ/Analyzer beyond "match EQ"... no experiment is complete without 2nd/3rd/4th opinions. 

 

The results Craig Anderton posted look much more like I would expect... and I've trusted Craig's troubleshooting/testing methods since the early 80's. 

As he says in his post....  "There are a lot of variables in your setup and I don't know what they are."

  • Upvote 1
Link to comment
Share on other sites

  • 2 weeks later...
On 7/29/2022 at 1:42 AM, rd2rk said:

Why should anyone care?

 

 

Well, I care because I actually care about tone :-D

 

Seriously though, the problem is I have a number of external drives I love, and always get compliments on my tone when I use them. I also have an M13 that I spent a lot of time dialling in tones on. And I like the sound of my clean amp - all my effects pedals bypassed.

 

So I would like to "transfer over" my analog pedals as I was able to do into the M13 (not exact, but close enough), and transfer over the M13 sounds, and still play clean through my amp.

 

The problem is, the HX FX is noticeably altering my tone. And the graphs demonstrate exactly what I'm hearing. Furthermore, the legacy effects on the HX FX do not sound the same as the M13.

 

I ran into this exact same problem when I moved from the "4" series (DL4, MM4, DM4) to the Pod XT Live - it messed up my bypassed tone and you couldn't correct for it, and the sounds from the 4 series pedals were all drastically different.

 

In this case now, the legacy models are more obviously "supposed to be the same".

 

I'm trying to figure out if there's anything I can do to "fix" this problem.

 

I've done some more experimentation since my last post and simply putting a Hi-Cut, or any of the blocks that cut highs - Kinky Boost, some drives, etc. naturally do it or I can use an EQ.

 

But what triggered the post was the inability to dial out that 8k boost. Using a mic to record it's not such a problem, but still it's hearing something different than what I am in the room - but it's closer and again my ears and the graphs agree - it's adding some highs because it's a buffer.

 

The 4 series and M series were both True Bypass, and the M series had a switch for analog or digital bypass, such that anything you ran with a dry thru (like many of the delays) didn't alter the tone (it did but very insignificantly).

 

So it matters to me because I'm comparing it to great tones I've spent a lot of hours on and would like to have still because I use them regularly.

 

And there seem to be some things going on tonally with this device that is making it difficult (and maybe impossible) to get those same kinds of tones.

 

I see some demos online of how close it sounds to a TubeScreamer and so on, but when I try to mimic the sound of the M13 or the pedal I based that M13 on - which are Tube Screamer type sounds I just can't get the HX FX to sound as close to the real pedal as I could the M13, nor get the HX FX to sound as close to the M13's version.

 

 

Link to comment
Share on other sites

On 7/29/2022 at 10:52 AM, PierM said:

If you don't like the sound you can get out of the Helix, just move on, sell it and buy something else as digging at that level is going nowhere in terms of possible solutions.

 

 

Well that's fair, but consider this: I had an M13 that sounded great. Why shouldn't this "new and improved product" that contains the legacy effects version from the M13 sound just as good?

 

The problem is, all things point to I should be able to get the sounds out of the Helix, but something's going on...

 

But yeah starting to look like I need to sell it and just waste the time going through analog pedals and just doing it old school. I don't have much money, and this was a "once in a decade" purchase (my M13 I bought in 2013) and I don't like messing with selling things at a loss so I'm trying to "make it work" so I didn't lose my butt on this.

 

 

  • Upvote 1
Link to comment
Share on other sites

On 7/29/2022 at 4:35 PM, codamedia said:

IMO... if you need to do that much "fixing", something else is seriously wrong.

Have you tried a 2nd EQ/Analyzer beyond "match EQ"... no experiment is complete without 2nd/3rd/4th opinions. 

 

The results Craig Anderton posted look much more like I would expect... and I've trusted Craig's troubleshooting/testing methods since the early 80's. 

As he says in his post....  "There are a lot of variables in your setup and I don't know what they are."

Holy crap, is that llike the real Craig Anderton? I thought maybe it was just a screen name someone picked as a fan or something! If that's the case, I'm honored - I had one of his home studio books from back in the 80s or early 90s.

 

I haven't tried any other Analyzers because I don't really have any. But Craig's test was comparing to internal sounds. I'm comparing a signal without the HX FX to a signal with it in line.

 

I explained in my other response, my problem is I'm trying to make effects from another Line 6 product contained in a newer Line 6 product sound like the effects from the older Line 6 product! The Classic Distortion in my HX should sound the same as the Classic Distortion in my M13.

 

But it doesn't.

 

Do I need a buffer? Do I need a Radial Dragster? Is it even possible to have the same effects from different devices sound the same? As I said in my other post this happened with the move from the 4 series to the Pod XT live. The M series sounded the same as the 4 series.

 

So the commonalities are, the 4 and M series are true bypass, while the HX and Pod were not. I don't know that it's fair to compare the Pod and HX architecture but one thing is clear: The HX FX changes the tone when it is plugged in, even with all blocks bypassed (or not even in the path).

 

This means it's "doing something" to the tone, and that's resulting in the sounds being different not only from my direct signal, but from my other Line 6 products with "the same" sounds in them.

 

If there is a way to negate that, great. But it seems that a lot of people don't hear it, or don't it doesn't bother them, or aren't trying to compare to other things, and so on. Great for them, sucks for me though.

 

IOW, if I wanted a host of different sounds, I would have bought a Boss or Fractal or Headrush or something. Instead I wanted a newer version of my M13 that had the same sounds in it - which "supposedly" it did because of the "Legacy" moniker and all the names and controls being the same.

 

But they're not. And I'm not even sure that being able to "fix" any of these EQ difference will solve that problem, but if I can figure out how to make the thing pass "direct signal without altering it" by adjusting EQ then those sounds may be much closer to being able to be dialled in.

Link to comment
Share on other sites

On 8/12/2022 at 4:19 AM, z3albw1rr said:

Well that's fair, but consider this: I had an M13 that sounded great. Why shouldn't this "new and improved product" that contains the legacy effects version from the M13 sound just as good?

 

The problem is, all things point to I should be able to get the sounds out of the Helix, but something's going on...

 

But yeah starting to look like I need to sell it and just waste the time going through analog pedals and just doing it old school. I don't have much money, and this was a "once in a decade" purchase (my M13 I bought in 2013) and I don't like messing with selling things at a loss so I'm trying to "make it work" so I didn't lose my butt on this.

 

 

 

It happens sometimes. For example, I still prefer the fender amp sims I do have in my FH 1500, which is considered "old gen" - and never really digged those in the Helix, at a level that I dropped the amp/cab all together and moved to a Iridium for the amp/cab simulation. Never looked back. Still using the HX for everything else I need and that works for me.

 

Happens, especially when you are expecting a specific sound that you already know and that's printed in your memory producing a bias (like your old M13 stuff), or in my case when you know the real deal (as I've those fender amps so I know how they should sound), and I can say they aren't just there in the HX but they were in the Firehawk...so yeah, I moved on to find a smaller footprint platform (than the boat anchor FH1500) that worked for me for that specific tone.


 

Link to comment
Share on other sites

the forums would be pointless if some of us keep saying 'just play/don't care/waste of time'. something in the device may bother someone. please let people review in a good or bad way. and maybe there would be another one suggest a solution or maybe a software error would be noticed and would be fixed.. we wouldn't know unless we write/read and answer. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...