Jump to content

Active, DSP, REW, Xilica 101


rplace

Recommended Posts

While @Chris A as helped so many of us ( @Coytee, @Rudy81, @babadono, @Khornukopia ) just to name a few with getting up and running. I was hoping to get a more basic thread going where we might be able to answer the more questions we all have....or maybe I'm the only one. Rather than just a yes/no or "do it this way" if you could provide some why that would be great. I know all of our systems are different so often the answer will be listen and decide for yourself. Still it seems there has to be quite a bit of common ground.

 

Here are some of my questions and some comments. Hopefully you all will put in your $0.02 and make this constructive for all.

 

  • If you had a hypothetical 2-channel only system in a symmetrical room and got one speaker dialed in with perfect SPL/Phase/Delay/etc. would you:
    • Copy those settings over to the other channel and be done
    • Copy those setting to the other channel and start measuring/tweaking/adjusting and arrive with two "different" but perfect speakers
    • Start over from scratch on the other because they are two independent items

 

I would think you would copy them over and be done. For a few reasons. Those of you whom just take Roy's Jubilee settings and do nothing with them (or Chris' file exchange method). Or for that matter if you buy a pair of speakers with passive networks the two speakers are the same. Conversely, I could see the argument that you need to dial each speaker in individually. Because the driver might be slightly different, the placement a bit off, your listening position, room construction that you can't readily see, treatments, etc. But I also think if you go down that path, the argument could be made that, at least in a theoretical sense, your perfect right speaker could be a Bessel, 6dB, 3ms delay with 5 PEQs while your left speaker could be Butterworth, 24db, 5ms delay with 8PEQs. Kind of a wild scenario I know and not realistic but I think it illustrates the point that they could be quite a bit different. For now I am making my two identical by copying my measured to the other. What do you all do?

 

  • If your speakers are dialed in to your liking and swap out a piece of equipment (specifically amps but could apply to anything if you are OCD). Do you?
    • Not worry about it because the speakers are dialed in
    • Start with a new preset for your new equipment and dial it in all over again
    • Something in between

 

I go back and forth in my mind with this. The speakers are what they are and are dialed in, but they are part of a system as a whole. That dialing in happed with that specific equipment. If amps & topology/preamps/source/DAC/Tubes/Cables, wire, power (ducking) can all impact the sound it would seem like every tiny change requires a soup to nuts reboot of your Active DSP configuration. The flip side is you don't really do that with a pair of speakers with passive networks. I'm exaggerating a bit here to make my point, but honestly I could be swayed by either side of the argument, if there even is an argument to be made. I'd think there is a balance to be struck somewhere like say Tube consumer, single ended amp to Solid state balanced pro amp needs to be looked at (the gain at a minimum). But we seem to like finding rabbit holes for the sake of finding them. Does every change require a new configuration?

 

  • How important is exact repeatability on mic position and measurements?

 

Sure you want to be as close and repeatable as possible but at what point does the OCD kick in? I'd swear sometimes I forgotten to mute the left side when taking right site measurements and they have been the same and when I realized and corrected that. Other times it seems like a leave everything turned on, all pieces in the same exact position, step out of the room for some reason and come back 5 minutes later and he measurements show different SPLs and different amounts of Phase Wrap. Sometimes I'll do measurements back to back in less than a minute and get slightly different results. Really how can any of those be? Are there some limits to the relatively cheap mini-DSP USB mic? Is it something I'm doing like sitting differently in the chair, my heating or A/C coming on? I thought the mic was very directional and did not pick up too much from around it. If "things" around it are picked up how would moving it all outside help in any way beyond room influences. Seems like you would pick up all sorts of outside noise if the mic was not very directional. I mark the mic stand's position on the floor with tape, keep the same measured distance, try not to move the mic's position relative to the stand. What else?

 

 

  • Like 2
Link to comment
Share on other sites

Well....

I have been just dialing one speaker in and copying to the other side.  I guess a perfectionist might dial each one individually, but I spend enough time on this stuff as it is.

 

As far as the differences in readings from one reading to another, it does happen to me as well. Granted, the differences are not large, just a slight change here and there. I use a lab calibrated ECM8000 which should be a reasonable mic for what we need.  Even my Audyssey Pro kit will show different readings from one set of takes to another. 

 

As you know, I too have been spending a lot of time working on this since I learned about @Chris A's technique.  I always learn a lot, but also need to be careful not to lose sight of the end game.....listening pleasure. For now I am taking a break from tweaking until I have some more time on a rainy weekend. This is a lot of fun and I sure wish I understood this complex subject more clearly.

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, Rudy81 said:

I always learn a lot, but also need to be careful not to lose sight of the end game.....listening pleasure.

 

Agreed, I've gotten some good ear time in recently and very happy with the direction I'm heading/feeling. Yes feeling, I think I am approaching and understanding some of what Chris and other have been discussing in the "Subconscious Auditory Effects " thread.

 

I've recently stumbled upon what I believe is a game changer for me. Too long for this thread; when I have the time/desire I post something in its own location to hopefully help others perhaps. Long story short I believe for the past year I've had a bad delivery mechanism (computer and sound card) on my input side. I repurposed an old laptop and have gotten more consistent measurements (part of my questions above) and better results. I know what Rudy and Chris have for measurements I'd be curious to see others so we can compare and contrast.

 

For a long time I was getting measurements with several wraps of phase typically 6 to 10 depending on XO point, PEQs, Delay, etc. but it seemed like very slow and incremental progress and often lateral moves or steps backwards in sound. Mind you it never sounded bad but we always want more. With my new (old) computer in place I've gone from Phase plots that typically looked like this:

5XPhaseNo2.jpg.d513cb59fd6c564d3733a5e801525b3e.jpg

 

To more like this:

Phase.jpg.d4fe585c62f507ddaad6d2ec42b45b0b.jpg

 

 

and Finally with a bit of tweaking this: Which looks and much more importantly sounds to me a LOT better. Not just different but substantially better. I'd be happy to post GD, SGram, Impulse for any interested but they have all improved quite a bit as well.

Nov25Phase.jpg.f432ef58681ed1eee0131ee42b31c608.jpg

 

 

Thoughts as to why the top plot does not have the rapid zig-zag of phase wraps but the next two do? The rough spectrum form 6K to 10K is similar in dips and humps for all 3.

 

  • Like 2
Link to comment
Share on other sites

1 hour ago, rplace said:

If you had a hypothetical 2-channel only system in a symmetrical room and got one speaker dialed in with perfect SPL/Phase/Delay/etc. would you:

  • Copy those settings over to the other channel and be done
  • Copy those setting to the other channel and start measuring/tweaking/adjusting and arrive with two "different" but perfect speakers
  • Start over from scratch on the other because they are two independent items

 

I measure just one and then copy, because of the time involved, but thinking about it now, the auto-EQ systems measure each individual speaker and the results are different for each one.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

26 minutes ago, rplace said:

Thoughts as to why the top plot does not have the rapid zig-zag of phase wraps but the next two do?

Were you using the same smoothing? Did you notice if "frequency-dependent windowing" is now turned on? Because the REW default is that it usually is on (IIRC), and you have to watch this when re-loading REW.

 

The reason for the rapid phase wrap at 6 kHz and higher is that there is a dual-peak of acoustic energy (SPL) above 6 kHz which you can see in the spectrogram, and the phase calculation within REW can't decide if the main SPL is earlier or later.  This isn't really phase wrap but a dual pulse of energy. For the casual observer, the following can occur at higher frequencies:

  1. re-radiation coming from one end of the driver that is farthest away from the microphone (like a bounce of acoustic energy)
  2. re-radiation from the mouth of the horn (i.e., termination bounce)
  3. two drivers playing is parallel above the crossover frequency (like the K-55 and the K77 drivers do in the older Klipsch Heritage loudspeakers having no low pass filter on the midrange driver circuit, and the K-55 is always lagging the K-77 tweeter)
  4. something around the microphone or within 1 foot (30 cm) of the mouth of the horn when taking measurements re-radiating acoustic energy

Etc.

 

The obvious thing to do is use more acoustic absorption over all objects near the loudspeaker under measurement or near the microphone.  You can even place absorptive material around the mouth of the horn itself and take more measurements to see if this is contributing.

 

Chris

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

13 minutes ago, Chris A said:

Were you using the same smoothing? Did you notice if "frequency-dependent windowing" is now turned on? Because the REW default is that it usually is on (IIRC), and you have to watch this when re-loading REW.

 

For sure 100% same smoothing. I only uses Psycho. I can't answer the Frequency Dependent Windowing question as I don't know where to find that setting. I'll do some digging.

 

13 minutes ago, Chris A said:

which you can see in the spectrogram

 

This is exactly the education I'm looking for in my DSP 101 class or Dummies tutorial. I still don't understand much about the spectrogram. I understand nothing about the colors and shapes that look like flames. All I really know is to look at the dashed line. This is what I think I know

  • vertical line top to bottom is good
  • gentle tailing off to the right at the bottom is good
  • bumps/hump rapidly to the right of XO gives you clues as to how much delay is needed to "pull" it back to the left
    • Put cursor at dotted line near XO (left axis) and read ms on bottom axis

Can you please show me or tell me in the SGram below where I can "see" what you describe? Also my basic understanding of that hump to the right suggests to me I need another 700um of delay, but when I put that in a get a wrap of phase right at the XO point. When I iterate very small numbers around my current 1.5ms range plus or minus say .05 to .1 ms that "extra" wrap of phase goes away and the spectrogram look pretty much the same no matter that number. I think I have delay close, do I? This particular SGram has 1.552ms of delay on the HF, if I up it to ~2.2 (1.5ms + 700um) phase and Sgram are worse.

 

1844278511_275XoSGramQuestions.jpg.678c9dba5fb0847bbe55e7ebd388a354.jpg

 

I think I understand the Group Delay better. Here is the GD for same Spectrogram above with the cursor at intersection of XO and Excess GD. It suggests to me I need another 364ms of delay, is that correct? If so adding that to my existing 1.55 makes both the GD and SGram worse. 1.4-1.6 seems about the sweet spot for Delay in the current set up. And it currently sounds stunning. Maybe I am just not sensitive to those "problems" above 6K?

 

GDNov25.jpg.7e7c7bbc388369745ee992ec5960016b.jpg

Link to comment
Share on other sites

2 hours ago, rplace said:

Can you please show me or tell me in the SGram below where I can "see" what you describe?

Let's use an example of an air-motion transformer (AMT-1) made by ESS in California:

61LoDkiLlQL._SL1000_.jpg

 

The full-range spectrogram is shown below:

 

1371839838_ESSAMT-1Spectrogram.thumb.jpg.e68161eb1f5271b526ebf26569f7f7f8.jpg

 

Note that the spectrogram "spike" is slender and almost vertical.  This implies that the driver has very good internal damping or has very low moving mass (in this case, it has both).  This is one of the reasons why the sound of the AMTs are "crisp" and transparent (i.e., they have a lot of clarity).  Impulses in the music tracks come and go, and the driver responds almost perfectly, that is until we get below 1 kHz, where the driver looks to be "smearing out" to the right at ~800 Hz.  This looks like it's getting close to resonance.  In fact, it is--but not the type of resonance that you'd expect, i.e., moving mass of the diaphragm with its suspension restoring it back to the rest position.  Rather, it's more likely that it's the moving mass of air in the horizontal "horn" that surrounds the driver, thus acting like a real horn.  In this case, only in the horizontal direction--not the vertical direction (perhaps more discussion on that later). 

 

Also note that there is lesser vertical spike (yellow color) at about 1.8 ms behind the original spike centered at 0 ms.  This is the backwave of the dipole AMT-1 reflecting off the wall behind it at ~12 inches, then arriving at the microphone after the original spike of acoustic energy (two-way travel of the wave = ~24 inches round trip distance).  If the rear wall were farther away, the resulting sound would subjectively sound more spacious due to the added delay of the reflected pulse being processed by the human hearing system to indicate a larger internal space (longer delays = larger dimensions of the room). 

 

In other drivers, you see these dual vertical spikes where the rearward-traveling acoustic impulse reflects off something inside the driver itself (like the enclosure of the driver's backwave, the back end of a complex-shaped moving diaphragm that has significant depth, or the mouth of a horn reflecting sound back into the horn itself, then returning at a later time).  This results in a smearing of impulses (as far as the human hearing system is concerned).  This is why lighter mass moving diaphragms sound more airy and light. 

 

In the case of your spectrogram...

 

275Xo SGram Questions.jpg

 

...you see a secondary pulse of acoustic energy at 0.7 ms from 2 to 4 kHz (and where the cursor is located in time), indicating that a reflection or some farther removed portion of the diaphragm from the microphone is radiating the same frequencies--but farther back than the first radiating portion of the diaphragm.  This isn't what you're looking for, and it results in a less transparent sound.  Additionally, the spectrogram pulse is getting quite wide, indicating that moving mass is smearing out the impulse (relative to the AMT-1 spectrogram above).  This general widening of the spectrogram pulse indicates a less "airy" sounding driver.  Moving cones tend to have spectrograms that look wider and wider as the frequency decreases because of their moving mass.  Compression drivers don't have to move very far (in fact, they only have to move above 1/5th the distance as a cone to create the same SPL on-axis), so they sound very airy and light in comparison.

 

Now, all the above gets more complicated due to the minimum response of the human hearing system vs. frequency, but in general the comment above accurately describes what is occurring. 

 

I'll stop there and let you ask questions.

 

Chris

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Very interesting discussion Chris.  I am wondering if the Oris back chamber is causing that secondary pulse.  For those not familiar with the Oris horn, the chamber that encases the full range driver has two items at the back. One is a thick 1" absorbent material and the other is a hard MDF type back that seals the chamber.  Mine has a small hole about 1/2" in diameter. (At least, that is how my Oris are set up. I do know the geometry and materials for the chamber back have changed over the years.) Many folks run the back chamber open with no backing and claim that it sounds more 'airy' and transparent.  I have taken sweeps with and without the back chamber cover and there is a SPL difference in the lower frequencies. 

 

Rich, you might try one sweep with and one without the back chamber cover. 

 

I do use my back chamber cover and this is what I got during my last runs.

 

Edit: I'm beginning to wonder if it would be better to run our Oris chambers open, thus removing any rear chamber reflections which could affect driver performance.  In my our case, we don't rely on the full range driver for bass performance.  Hmmmm....? For years I ran with the chamber open after my experiments with open baffles.  I loved the open baffle sound. Last time I changed drivers I re-installed the back chamber covers and forgot about the option to leave the covers off.

 

Ctr spectrogram final.jpg

Link to comment
Share on other sites

30 minutes ago, Chris A said:

 

I'll stop there and let you ask questions.

 

Very informative. I can see on both SGrams the reflectivity of the AMT and mine. Seeing and reading the desired behavior (narrow red/yellow bands vs my widening one and "extra stuff" to the right at same frequency) is very, very helpful. Thanks! Now I know what the goal is.

 

I've always just thought the purpose was to used the dashed line to determine delay at XO point but suspected there had to be more to the colors themselves and widening/flaring of the colors. @Rudy81 You are a mind reader I was thinking the same thing about backs of Oris horn before I read your post. I originally like them better off, but can't remember why I eventually started running them on. New measurements to follow.

 

Q1.

Is my delay correct based on what you can see/read in the above SGram and GD plots?

 

Q2.

What can I do if anything to get rid of my "hump to the right" between ~200-600 knowing that for now the XO is at about 275? I'd guess I might be able to move the XO point back up around 600-800 and perhaps make the HF driver not work so hard, if that is a thing. My LF bins have pretty good individual measurements (Phase and flatness) up to about 1500-2000. However I like the sound of the Oris 150/Audio Nirvana neodymium drivers better than my LF open baffle and previous drivers I've tried.

Link to comment
Share on other sites

1) That jump to the right at your crossover frequency of the dotted curve is the indicator that I look at, and your excess group delay curve.  Presently the excess group delay curve says that you're there, but the spectrogram looks to be saying you need 600 µs.  I'd go with the excess group delay plot here.

2) Hump to the right: not sure--it could be related to dipole bass bins (as I assume you have).  Move the crossover frequency up to 800 Hz.

 

Chris

  • Thanks 1
Link to comment
Share on other sites

On 11/25/2019 at 10:21 AM, rplace said:

If your speakers are dialed in to your liking and swap out a piece of equipment (specifically amps but could apply to anything if you are OCD). Do you?

  • Not worry about it because the speakers are dialed in
  • Start with a new preset for your new equipment and dial it in all over again
  • Something in between

 

So nobody taking a swing at this?

Link to comment
Share on other sites

You probably know my feelings on this subject:  If I can hear a significant difference with some amplifier swap or I think that one or more amplifier gain knobs have been twiddled, I'd be inclined to measure SPL and phase response differences to help visualize what's occurring. 

 

I always keep the microphone/stand nearby to plug in, and the Xilica stays plugged into a USB port on the computer, so the only real action is pulling out the foam rubber pad from the nearest closet and placing it on the floor between the loudspeaker and the microphone. The whole thing takes 5 minutes or less, including bringing up REW on the computer and setting preamp channel muting/gain levels. 

 

I don't usually start from scratch zeroing all PEQs, unless I've got new drivers that I'm putting into the setup or I'm testing some driver/horn combination for someone else.  The only exception is when I tried out the Danley style PEQ--only crossover approach (i.e., using no "named" crossover filters).  That required zeroing all channels to get the raw response of each driver.  From there, I can see where the crossing frequency should be.  I then begin running REW's EQ facility to find more optimal flattening PEQs, layering them on in waves using updated measurements until I get the response as flat as I can while accounting for microphone/room boundary and room mode cancellations (identified by non-flat areas of the excess group delay curve).

 

But to dispel fears of being OCD about it: I usually go months or a year between measurements and DSP crossover PEQ updates if everything sounds fine.  There is no drift in response over time like there is with passive crossovers due to aging of components.  The drivers themselves are stable after their initial break in period. 

 

Chris

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

I see no difference in Ethernet vs. USB except that you have to set the port address on Ethernet.  Once that's done, it's equivalent either way. 

 

Chris

Link to comment
Share on other sites

Just an FYI, if you don't already know this it is a pretty good time saver. I was always changing to the All SPL section to set the smoothing in REW....every reading, All SPL, smoothing drop down list, Psycho Acoustic, repeat.

 

If you go to Preferences|Analysis you can set a default smoothing and it will be applied to every reading. You can still apply none or any other type for a given measurement if you need a one-off smoothing on the All SPL screen.

 

REWPref.JPG.ae5c062994a78c26aa5831e5c2061f68.JPG

  • Like 1
Link to comment
Share on other sites

On 11/25/2019 at 1:17 PM, Chris A said:

Let's use an example of an air-motion transformer (AMT-1) made by ESS in California:

61LoDkiLlQL._SL1000_.jpg

 

The full-range spectrogram is shown below:

 

1371839838_ESSAMT-1Spectrogram.thumb.jpg.e68161eb1f5271b526ebf26569f7f7f8.jpg

 

Note that the spectrogram "spike" is slender and almost vertical.  This implies that the driver has very good internal damping or has very low moving mass (in this case, it has both).  This is one of the reasons why the sound of the AMTs are "crisp" and transparent (i.e., they have a lot of clarity).  Impulses in the music tracks come and go, and the driver responds almost perfectly, that is until we get below 1 kHz, where the driver looks to be "smearing out" to the right at ~800 Hz.  This looks like it's getting close to resonance.  In fact, it is--but not the type of resonance that you'd expect, i.e., moving mass of the diaphragm with its suspension restoring it back to the rest position.  Rather, it's more likely that it's the moving mass of air in the horizontal "horn" that surrounds the driver, thus acting like a real horn.  In this case, only in the horizontal direction--not the vertical direction (perhaps more discussion on that later). 

 

Also note that there is lesser vertical spike (yellow color) at about 1.8 ms behind the original spike centered at 0 ms.  This is the backwave of the dipole AMT-1 reflecting off the wall behind it at ~12 inches, then arriving at the microphone after the original spike of acoustic energy (two-way travel of the wave = ~24 inches round trip distance).  If the rear wall were farther away, the resulting sound would subjectively sound more spacious due to the added delay of the reflected pulse being processed by the human hearing system to indicate a larger internal space (longer delays = larger dimensions of the room). 

 

In other drivers, you see these dual vertical spikes where the rearward-traveling acoustic impulse reflects off something inside the driver itself (like the enclosure of the driver's backwave, the back end of a complex-shaped moving diaphragm that has significant depth, or the mouth of a horn reflecting sound back into the horn itself, then returning at a later time).  This results in a smearing of impulses (as far as the human hearing system is concerned).  This is why lighter mass moving diaphragms sound more airy and light. 

 

In the case of your spectrogram...

 

275Xo SGram Questions.jpg

 

...you see a secondary pulse of acoustic energy at 0.7 ms from 2 to 4 kHz (and where the cursor is located in time), indicating that a reflection or some farther removed portion of the diaphragm from the microphone is radiating the same frequencies--but farther back than the first radiating portion of the diaphragm.  This isn't what you're looking for, and it results in a less transparent sound.  Additionally, the spectrogram pulse is getting quite wide, indicating that moving mass is smearing out the impulse (relative to the AMT-1 spectrogram above).  This general widening of the spectrogram pulse indicates a less "airy" sounding driver.  Moving cones tend to have spectrograms that look wider and wider as the frequency decreases because of their moving mass.  Compression drivers don't have to move very far (in fact, they only have to move above 1/5th the distance as a cone to create the same SPL on-axis), so they sound very airy and light in comparison.

 

Now, all the above gets more complicated due to the minimum response of the human hearing system vs. frequency, but in general the comment above accurately describes what is occurring. 

 

I'll stop there and let you ask questions.

 

Chris

 

I had not heard of the AMT until you posted.  Interesting concept for a speaker that can go from 800hz-20khz. 

Here goes the stupid question of the day.  Could the AMT be placed behind a front loaded horn?

Link to comment
Share on other sites

".....

I had not heard of the AMT until you posted.  Interesting concept for a speaker that can go from 800hz-20khz. 

Here goes the stupid question of the day.  Could the AMT be placed behind a front loaded horn?

...."

 

This suggestion has been tossed around over at DIY audio in the past. The short answer is "probably not a good idea".  Although one guy "Badman" apparently had some success.

 

That style of tweeter has certain strengths and weaknesses. The system's design (goals etc) needs to be thought out first, before choosing and integrating a tweeter

 

Good Luck,

-Tom

Link to comment
Share on other sites

Not stupid at all, in fact...

See: https://www.diyaudio.com/forums/planars-and-exotics/171441-hornloading-heil-amt1.html

 

...and from Speaker Builder Magazine, March, 1982 (look particularly at the third image for frequency response plots with/without horn):

 

200986d1292393653-hornloading-heil-amt1-

201040d1292435762-hornloading-heil-amt1-

 

201041d1292435762-hornloading-heil-amt1-

201045d1292436009-hornloading-heil-amt1-

/201046d1292436009-hornloading-heil-amt1-

201049d1292437302-hornloading-heil-amt1-

201050d1292437302-hornloading-heil-amt1-

201051d1292437495-hornloading-heil-amt1-

201053d1292437553-hornloading-heil-amt1-

201055d1292437807-hornloading-heil-amt1-

201056d1292437807-hornloading-heil-amt1-

201057d1292437840-hornloading-heil-amt1-

 

Chris

Link to comment
Share on other sites

1 minute ago, PrestonTom said:

".....

I had not heard of the AMT until you posted.  Interesting concept for a speaker that can go from 800hz-20khz. 

Here goes the stupid question of the day.  Could the AMT be placed behind a front loaded horn?

...."

 

This suggestion has been tossed around over at DIY audio in the past. The short answer is "probably not a good idea". 

That style of tweeter has certain strengths and weaknesses. The system's design (goals etc) needs to be thought out first, before choosing and integrating a tweeter

 

Good Luck,

-Tom

 

I am finding out that it has been an idea for a while.  Been spending time on search engines checking it out. 

Link to comment
Share on other sites

Chris, I can't look for the specific threads over at DIYaudio, but there were some conceptual arguments and empirical data that were at odds with this configuration.

 

Perhaps a potential designer would want to look at those before investing the money. As I recall, those tweeters were not inexpensive.

 

In the intended application, is the idea to 1) use the horn for controlling dispersion, since they are dipoles, the dispersion already has some control. or is to 2) get horn loading (the throat smaller than the "tweeter")?

 

Good Luck,

-Tom

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.

×
×
  • Create New...