Frequency Response Graphs For Hd500 Eqs

[pfsmith0]

In the process of trying to figure out what a 4 Band Shift EQ is, I put my HD500 on the test bench and measured the frequency response of ALL the EQs in the HD500. I thought you'd find it useful so I'm sharing the results here. For example, what does a 100% shift mean? What does 0% Q mean? And just where is Freq = 38% in the Parametric EQ? Did you know when you set 2.2kHz of the Graphic EQ to -12dB you'll get a +12dB boost (!) at 1kHz before it dips down at 2.2kHz? No wonder it didn't work very well to eliminate a pesky feedback problem, but caused another one instead. Using these can give you insight into which EQ you really want to use.

 

I've attached the most pertinent pictures here (zipped together to make it easy). The only potential problem is the 4 Band Shift Hi Freq which has mismatched L/R channels (~3dB at 1kHz)! Every other EQ and every other setting has perfect L/R balance (less than .1dB mismatch) but this one does not. I've contacted Line 6 support before submitting this here but they see nothing out of the ordinary (i.e., it sounds fine, which may be true, but there's still a mismatch, big enough to be called a bug in my opinion). I've labeled those two graphs channel 1 & 2 because that's what my test equipment calls them. Unfortunately I didn't have the forsight to find out which was left and which was right before I tore down the setup. But the fact of a mismatch remains.

 

Some things you need to know to properly interpret these:

1. An Audio Precision APx525 was used to generate the sound source and measure frequency response.
2. Output = S/PDIF
3. Input 1 = Mic. Input 2 = same. Mic level on back adjusted to give -20dBFS output with the Graphic EQ
4. Unless otherwise indicated, the gain of all the other EQs were set to nominally give -20dBFS. The gains were recorded so you can see which EQs have built-in gain or loss compared to the Graphic EQ.
5. There were no other FX or amps in the signal chain.
6. Sample rate = 44.1kHz (factory default)

I have more complete files that contain the actual data (not just pictures) as well as more settings than shown in these few pictures attached here. Until I can find a better place to put it, you can download the entire 160M set of Excel files from: https://dl.dropboxusercontent.com/u/56306778/HD500%20Measurements.zip. I am open to ideas of where to put this on a more permanent basis.

 

Enjoy! I hope you find these as useful as I have.

----------------------------

I took the HD500 back into the lab and verified that it's the Left Channel that has the odd frequency response in the Hi Freq portion of the 4 Band Shift EQ. I re-uploaded the Excel files to indicate this and re-uploaded the new 4 Band Shift EQ figures here.

----------------------------

I measured the frequency response of the Q Filter (per meambobbo's post) and attached it here. I also updated the giant zip of Excel files. I think this could emulate a pretty Morley wah if you make toe = hi Freq/hi Q/hi Gain and heel = low Freq/low Q/ low Gain. Also, mix attempts to add the filtered signal to the straight signal, which it does. But the filtered signal has some processing delay so the mix includes some notches in the 4k-8kHz region you may not be expecting. You can see this in the attached graphs. For the signal levels I was using (-20dBFS thru the Graphic Equalizer) you can also see some gain compression with the Q filter. So if you want it clean, stick to the lower Gain levels.

-----------------------------

Added a PDF file that lists the frequencies (in Hz) represented by the Parametric Frequency (in %).

Updated links to ALL my measurement files:

Amp Frequency Response Pre and Full (Excel)

Cab DEP Parameters (Excel)

Cabs with 121 Ribbon (Excel)

Gains (Excel)

Mic Frequency Response (Excel)

Amp Tone Controls (pdf)

Amp-Pre Frequency Response (pdf)

Cab Frequency Response (pdf)

FX Gain (pdf)

Model Pak Frequency Response (pdf)

Global EQ (zip)

Huge 185M EQ frequency response file (zip)

 

This last one was the original subject matter of this thread.

 

 

 

 

 

 

 

 

 

[silverhead]

Really useful stuff, pfsmith0. Thanks for sharing this.

[meambobbo]

this is MOST excellent!  PLEASE let me add them to my site: foobazaar.com/podhd/toneGuide/

 

I have some data posted in the EQ section, as well as the link to Matt Mayfield's excellent video demonstrating the EQ's frequency responses, but these really help dial it home.

[geppert]

Absolutely wonderful work and superb documentation. I downloaded the 160 meg Zip.

 

Thanks very much for doing this.

[pfsmith0]

Certainly! Use this on your site as you see fit. If you have a more permanent place to store the files then I am interested in that as well!

 

this is MOST excellent!  PLEASE let me add them to my site: foobazaar.com/podhd/toneGuide/

 

I have some data posted in the EQ section, as well as the link to Matt Mayfield's excellent video demonstrating the EQ's frequency responses, but these really help dial it home.

[meambobbo]

Certainly! Use this on your site as you see fit. If you have a more permanent place to store the files then I am interested in that as well!

 

Thanks so much! Trust me your work will get lots of visibility this way, and the files will be free to access over the web wherever you are whenever you like.

[meambobbo]

all right - pictures are up!  i linked to this thread and gave you props.  let me know if you want me to include anything else.

 

http://foobazaar.com/podhd/toneGuide/eq#effects

[meambobbo]

absolutely bizarre about the shift EQ output - I'm going to experiment with that myself.  I want to post Ch. 1 as well, but I want to figure out which one is L and R.

[pfsmith0]

Nice write-up meambobbo. I did not realize I maybe should have included the Q-filter. If/when I get the unit back into the lab I'll measure that as well - and determine which channel (left/right) has the messed up 4-band shift EQ high freq response. Also, please note that the Excel data file has the parametric peak frequencies for 1% adjustments. The graphs only show 5% so you may want to download the whole zip file from the link in the first post and include a 1% table in your write-up. I can't attach it here due to size constraints.

[Astaroth_CY]

Excellent work. Line 6 should be paying you and the other people doing research on their machine.

[snhirsch]

Thanks for all the good work.  It's a very odd marketing decision for Line-6 not to a) label effects parameters in terms of frequency and b)not make a family of graphs part of the technical documentation.  I don't think they're giving their customer base enough credit.  There may be a few folks who don't care about the details, but many of us do.

[snhirsch]

I did a  PDF compilation of the graphs in a 2-up arrangement suitable for printing.  Unfortunately, it's too big to upload as an attachment.  If anyone wants to host it, let me know where to upload?

 

 

[Astaroth_CY]

You can put it up on Dropbox and share it. You get 2GB of cloud space free.

[pfsmith0]

I don't disagree with you, but I do see an advantage for using % instead of absolute numbers. With % they can always update or fine tune the frequencies to make them more useful thru future software updates. That being said, I don't find % very useful, although they're better than nothing. The whole reason I started this work was because I DIDN'T fine % useful. I needed the absolute frequencies so I could 1) understand what they do and 2) use them in a more direct manner instead of experimenting around (too time-consuming). But, I'm glad you're finding these graphs as useful as I do.

 

 

Thanks for all the good work.  It's a very odd marketing decision for Line-6 not to a) label effects parameters in terms of frequency and b)not make a family of graphs part of the technical documentation.  I don't think they're giving their customer base enough credit.  There may be a few folks who don't care about the details, but many of us do.

[meambobbo]

Nice write-up meambobbo. I did not realize I maybe should have included the Q-filter. If/when I get the unit back into the lab I'll measure that as well - and determine which channel (left/right) has the messed up 4-band shift EQ high freq response. Also, please note that the Excel data file has the parametric peak frequencies for 1% adjustments. The graphs only show 5% so you may want to download the whole zip file from the link in the first post and include a 1% table in your write-up. I can't attach it here due to size constraints.

 

I took one look at one of those spreadsheets and said, WHOA.  I simply don't have time to parse all that data.  I think the 5% increments are good enough for most people.  I've never heard a complaint for more precision - it just kind of helps to give you a better frame of reference than x %.  Keep in mind I made the tables a ways back.  They seem to agree with your data from a quick spot check, but there may be some divergence.  But I think everything is good enough to give people a close idea of what to expect, which is the main point.  I missed some of the finer details that you discovered with your more detailed analysis, like the boosts before dips with the Parametric high/low params, and the 2.2kHZ Graphic EQ.  Those can be bewildering if you don't know about them.

 

Q Filter is a strange beast.  The parameters interact with each other in unpredictable ways, particularly mix and gain.  If you ever do an analysis on them, you'll see exactly what a mean.  It's difficult to explain in words or to even predict.  But for the purpose of pre-EQ I use it for, it's not that pertinent.

 

Anyway, let me know if you want me to add anything or change anything.  Again, great work, and everyone appreciates it.

[pfsmith0]

Please note that I attached frequency response graphs for the Q Filter in the first post of this thread. Also, I identified the Left channel as having the odd frequency response in the Hi Freq portion of the 4 Band Shift EQ. The giant ZIP file containing all the data has been updated with this new information and can still be found in the Dropbox link shown in the first post. Enjoy!

[Ball_ed]

Great work, thanks for sharing bro...

[Warfare]

Excellent work, thank you, I will take a deep look !

[pfsmith0]

I attached a PDF file to the original post the lists the Frequency (in Hz) represented by Parametric EQ Freq (in %).

[toneman2121]

so let me try to understand these graphs. you use the to set the output level of the frequencies so they are all at the same level, mainly for recording. if this is even close to correct, please just answer with a simple yes or no because i couldn't comprehend more than that. :wacko: 

[pfsmith0]

Hi toneman, I would answer with a simple yes/no but I don't understand your statement "you use the to set the output..." I think a very important word is left out between "the" and "to". Would you please repeat the question?

[toneman2121]

these

[pfsmith0]

Yes, that is one of the uses for this information.

[Vettacon]

The info presented, which is excellent, begs the question: Did Line 6 do this just to mess with users? Why reinvent the wheel with basic eq principles at the expense of confusing/frustrating users more? it appears little narcissistic on their part.

 

Thanks again to those putting the time in to de- code.

[DeanDinosaur]

 it appears little narcissistic on their part.

 

Thanks again to those putting the time in to de- code.

 I have to agree, i  find very hard to believe that they couldn't provide that in the PC/Mac Editor at least and not on the hardware itselt. It would have probably taken them minutes. Considering all the routing options in POD FARM/XT/X3 that seemed extremely intelligent and well though out, as much as I like the HD modeling, the image of Line 6 intelligent design was completely shattered when the EQ is still in Percent after three years of release!

[snhirsch]

 the image of Line 6 intelligent design was completely shattered when the EQ is still in Percent after three years of release!

 

That, and the utter lack of overload indication in the signal chain. 

[unperfectcircle]

Thanks a lot for sharing your efforts!

I try to avoid internal EQs because I don't really like the way they color the tone, but that's some great info here!

[malexand20]

I recently purchased an HD-500 and I'm still in the learning stages. In trying to reproduce some of my favorite X3 live tones I need to use EQ. Last night I became very frustrated with the lack of Line 6 documentation for EQ parameters and responses; the "trial and error" method was really trying my patience. Needless to say, your documentation plus MeAmBobbo's website (which is great!) posting is like a breath of fresh air! I agree with Astroth_CY's comment - Line 6 should be paying you for doing their work. 

 

Line 6: Great products MINUS adequate documentation = mediocre service! Invest in the time to better help your customer base, most of us don't hold degrees in sound engineering. P.S. Still like your products. 

[meambobbo]

the use for the EQ graphs is to be able to anticipate the effect it will have on your tone.  the narrow frequency bands that move opposite to how one would anticipate are very useful to be aware of.  Trying to use the EQ's and getting less-than-satisfactory results may be due to some of these quirks.  Maybe there's an alternative EQ, or maybe there's a better way to go about it.  For example, rather than using the high and low parameters of the Parametric EQ to cut lows and highs, you're likely to get better results using the Mid-Focus EQ in only one channel, mixing to approximate how much of a midrange boost you want, without completely chopping off highs and lows.

[HDProJohn]

For example, rather than using the high and low parameters of the Parametric EQ to cut lows and highs, you're likely to get better results using the Mid-Focus EQ in only one channel, mixing to approximate how much of a midrange boost you want, without completely chopping off highs and lows.

 

The use of the Mid-Focus EQ after the amp modeling instantly turned my tone's suck knob way down. I forgot if it was you, meambobbo, or someone else who had first brought its use up, but that trick alone saved my sanity and ears. The IRs are extremely fizzy and harsh in the high end.

[lilguitar]

Quality work man, this is so much better than the hunt and peck or trial by error approach that we've be subjected to, allowing  us a peek behind a curtain that should never had existed.

 

And props to you for all the time you invested in this lil research project, but even bigger props for sharing with the community! 

[wasbeerik]

THNX!

[HankNJ]

Because of the cryptic EQs, (percent values rather than frequencies), my Pod HD500X has been sitting in the closet for about a year. I use the EQs heavily on my POD XT Live and I bought the HD500X as a replacement. Needless to say I am deeply discouraged.

 

How can I submit a formal request for change to the software development team for HD500X at Line 6? Based on the responses to this thread alone, it is in their interest to change the displays to frequencies. THANKS!

[sapalma]

GRACIAAAAASSSSSSSSS!!!

[gunpointmetal]

I was told by Line 6 that the EQs in the HD stuff will NEVER be set to actual frequencies. I was told that it is impossible due to the programming, basically, which seems like a big, dumbass oversight. These graphs are helpful!

 

EQ's in % seems like something for Amplifi/Spider level equipment, but no for something you expect people to actually gig with. Everyone says "use your ears"....my ears tell me I need to find 1500 Hz and kill it......so I gotta fiddle around for twenty minutes trying to find it.....

[pfsmith0]

Yep! I agree. But you now know that 1500 Hz is about 62%.

[Tmidiman]

Because of the cryptic EQs, (percent values rather than frequencies), my Pod HD500X has been sitting in the closet for about a year. I use the EQs heavily on my POD XT Live and I bought the HD500X as a replacement. Needless to say I am deeply discouraged.

 

How can I submit a formal request for change to the software development team for HD500X at Line 6? Based on the responses to this thread alone, it is in their interest to change the displays to frequencies. THANKS!

I have to agree. I'm not sure what Line 6 was thinking when they put frequencies as percentages. I can only hope this is corrected in the future.

 

And Mr. Smith, thank you for your hard work. My appreciation is beyond words!

 

Mr. Meambobbo, ditto! Love your website.

[Tmidiman]

I was told by Line 6 that the EQs in the HD stuff will NEVER be set to actual frequencies. I was told that it is impossible due to the programming, basically, which seems like a big, dumbass oversight. These graphs are helpful!

 

EQ's in % seems like something for Amplifi/Spider level equipment, but no for something you expect people to actually gig with. Everyone says "use your ears"....my ears tell me I need to find 1500 Hz and kill it......so I gotta fiddle around for twenty minutes trying to find it.....

I think it was an oversite, plan and simple. It may also be very difficult to correct in the programming. My hope it that in the next generation of Pod gear this is corrected.

 

Every piece of gear I have uses frequencies for adjusting detailed EQ parameters. It makes things easy when sharing settings between them. Not having them drops their flags ship product below other products.

[Dshow]

This is great stuff - thank you!

[Contrasseur]

Oh my god I'm so bummed out about this. I was about to buy one of these too, but EQ is probably THE MOST important effect. You can control feedback, make guitars and basses sound like totally different models, alter your distortion, protect your speakers, and fundamentally control your tone. I can't believe how badly Line 6 fell short here! That 2.2kHz band on the GEQ... what do they expect me to do with that?