Trying To Understand REW and Mini DSP

[MR.MAGOO]

What Will Mini DSP Do That Audyssey or YPAO Does not?

So mini DSP will be a permanent addition to my AVR, plugged in between the subwoofers and AVR? Then if I change subwoofer locations or get different subs I re-calibrate the mini DSP?

Just wondering if it makes a noticeable difference (besides to my wallet).

 

[lovinthehd]

YPAO and Audyssey all you need do is move the provided mic to various positions around your primary seat and those programs will set delays, levels and eq automatically for the subs and speakers.

In your proposed addition for further management of your subs, using a measurement mic (like a miniDSP Umik-1 or a Dayton UMM-6) you can take measurements with the help of the REW software, but is a bit more involved on your part; the REW software can also generate downloadable (to the miniDSP 2x4 model of your choice, for example) adjustments for delay, level and eq of your choice. Might check out the REW website and download the software and look around. It has a feature to help plan sub placement in given room dimensions that you can use without a measurement mic.

Also good information at the miniDSP website

 

[PENG]

What Will Mini DSP Do That Audyssey or YPAO Does not?

So mini DSP will be a permanent addition to my AVR, plugged in between the subwoofers and AVR? Then if I change subwoofer locations or get different subs I re-calibrate the mini DSP?

Just wondering if it makes a noticeable difference (besides to my wallet).

If you have Audyssey XT32 SubEQHT already I doubt you need the mindsp EQ system unless you want to invest a lot of time learning how to use to it. Even then, I am not sure if it can do a better job, and if you don't know what you are doing you could do more harm than good. If you just want to use REW to plot some graphs then you only have to invest USD75 on a MIC such as THIS

Pogre has recently acquired both the minidsp 2X4 HD and the Umik-1 mic so hopefully he will chime in soon and offer his opinion/experience.

 

[Pogre]

I've found that Audyssey XT32 did a really good job in my room. Enough so that I've yet to open the box my MiniDSP came in. I've taken many REW sweeps before and after rc. I don't think it's worth the time and effort to incorporate a mini. At least for me.

If you want, just get the mic and take some measurements. if you think you can do better with a mini... I might have one for sale...

 

[Pogre]

 

[KEW]

I know @ATLAudio has been using a miniDSP along with XT32 for a few months now. He has found some pretty good utility out of it. I'm pretty sure he considers it a critical component of his system at this point.
All I can say is he has shown me some impressive in-room response graphs for bass.
He is using the miniDSP to tune his subs, but I forget the sequence (ie before Audyssey or after).
Hopefully he will share some of his experience!

 

[lovinthehd]

I know @ATLAudio has been using a miniDSP along with XT32 for a few months now. He has found some pretty good utility out of it. I'm pretty sure he considers it a critical component of his system at this point.
All I can say is he has shown me some impressive in-room response graphs for bass.
He is using the miniDSP to tune his subs, but I forget the sequence (ie before Audyssey or after).
Hopefully he will share some of his experience!

Usually you would run the minidsp setup prior to Audyssey but am curious if he indeed did it the other way around....

 

[ATLAudio]

I'm a firm believer that Audyssey XT32 goes by the Jerry McGuire line, "help me, help you." The more challenging the correction it has to perform, the less you're going to get out of it. I have a (massive) 60hz hump were two (massive) room modes are within 4hz of one another. MiniDSP will alleviate this hump completely at the MLP. I then run Audyssey and get a far, far better correction.
Granted, the Audyssey correction is an improvement from without doing anything, but using miniDSP certainly has it's place.

I don't have anything to post, like before and after, but I've done it a few times and always get the same results. Without removing this hump, Audyssey rolls the bass off at 30 hz, and does other changes which aren't optimal. If I remove this hump prior to Audyssey, I do a few more tweeks, and I have very flat response down below 10hz with my two subs.

If you're tech savy enough, and it's not terribly hard, REW with a Umik1 and MiniDSP is a $200 investment and very effective.

 

[KEW]

I'm a firm believer that Audyssey XT32 goes by the Jerry McGuire line, "help me, help you." The more challenging the correction it has to perform, the less you're going to get out of it. I have a (massive) 60hz hump were two (massive) room modes are within 4hz of one another. MiniDSP will alleviate this hump completely at the MLP. I then run Audyssey and get a far, far better correction.
Granted, the Audyssey correction is an improvement from without doing anything, but using miniDSP certainly has it's place.

I don't have anything to post, like before and after, but I've done it a few times and always get the same results. Without removing this hump, Audyssey rolls the bass off at 30 hz, and does other changes which aren't optimal. If I remove this hump prior to Audyssey, I do a few more tweeks, and I have very flat response down below 10hz with my two subs.

If you're tech savy enough, and it's not terribly hard, REW with a Umik1 and MiniDSP is a $200 investment and very effective.

Interesting!

Does Audyssey successfully address the 60Hz hump on its own, or does it only partially handle it?

Let me make sure I understand what you are saying about 30Hz vs 10Hz rolloff. Are you saying that Audyssey actually treats it differently (presumably) based on the the situation at 60Hz? IOW, you are not using the miniDSP to get the Flat to 10Hz response, but Audyssey does this if it is not busy having to deal with the 60Hz EQ?
This makes me realize how little I know about Audyssey. I know it has lots of filters, but can it assign them to whatever frequencies it needs to? I've been thinking in terms of a traditional PEQ which has evenly distributed bands of EQ. It is just now occurring to me that as software Audyssey might be "shooting it's wad" tamping down your massive humps and not have resources left for the subsonic frequencies (which should not be first priority).
I'm not sure how much detail Audyssey has given about these details but does that fit your (more mature than my) conceptual model of what it does?

 

[ATLAudio]

@KEW

Kurt,
Audyssey will turn down the sub gain far more if I don't eliminate the 60hz hump first. Due to this gain adjustment I simply don't have enough steam to make it very low. Again, help me help you. The more you do things right on your end, the more Audyssey will provide. No, I'm not using DSP to achieve the <10hz.

 

[PENG]

Very interesting indeed, the fact is, Aydyssey have no problem flattening the 20 to 100 hz for me and Pogre, we both posted our graphs. I also sent mine to SVS and they (Ed, I think), also thought it was very good. I may actually invest in the mini, either the $205 2xHD or the more expensive one that has Dirac build in for one of 2 ch system. If and when that happens, I will share the results.

 

[KEW]

Yeah, his results for the subs were pretty impressive.!
I love my PSA XS15se's as clean musical subs but don't consider them the final word in low frequencies, by any means!
Hopefully he can correct me, but IIRC, he was getting flat down to 10Hz at something over 100dB (the subs are corner loaded). I'm not sure that is something I want or need, but it is quite impressive!

 

[PENG]

Yeah, his results for the subs were pretty impressive.!
I love my PSA XS15se's as clean musical subs but don't consider them the final word in low frequencies, by any means!
Hopefully he can correct me, but IIRC, he was getting flat down to 10Hz at something over 100dB (the subs are corner loaded). I'm not sure that is something I want or need, but it is quite impressive!

I would love to see that "flat" curve he has. Is it truly flat? I thought getting the subs to do +/-2 to 3dB from 15 to 100 Hz using 1/12 smoothing is excellent because that's all I could achieve with XT32 SubEQHT and SVS sort of told me so. My subs are not identical twins (PC12U and PB13U) and probably is a bigger challenge for Audyssey. Actually I do have one dip at around 51 Hz that was about a touch outside the +/1 3dB range. There is no way I could fit another PB13U in where the PC12U is sitting so I have to settle for less. In theory (as I understood it) I can't see how REW based mini can work better with Audyssey especially in that sequence, but I must have misunderstood something.

 

[ATLAudio]

I would love to see that "flat" curve he has. Is it truly flat? I thought getting the subs to do +/-2 to 3dB from 15 to 100 Hz using 1/12 smoothing is excellent because that's all I could achieve with XT32 SubEQHT and SVS sort of told me so. My subs are not identical twins (PC12U and PB13U) and probably is a bigger challenge for Audyssey. Actually, I do have one dip at around 51 Hz that was about a touch outside the +/1 3dB range. There is no way I could fit another PB13U in where the PC12U is sitting so I have to settle for less. In theory (as I understood it) I can't see how REW based mini can work better with Audyssey especially in that sequence, but I must have misunderstood something.

A miniDSP 2x4 (non-HD, which is the one I use) is a $100 investment, you won't be disapointed.

I'm +/- 3dB down to 9.5HZ, and +/- 2dB down to 12HZ with 1/12 smoothing.

Like I said, Audyssey almost always helps, but only as much as it's helped. I have a square shoebox of a room with 4 exterior walls. The only way it can deal with the massive 60 hz bump I have is to start by turning the gain down. A lot. Audyssey has to make a lot of assumptions for it to be as easy as it is to use. It can only make guesses at my room size, and specific dimensions. If it simply applied -16dB of PEQ to that 60hz dip, as I do, it might work in my room, but totally be crazy in another. Simply put, Audyssey will turn down gain to get the system into a range it can work with. Thus, turning down the entire gain range also limits my max SPL extension.

 

[Winkleswizard]

While I agree minidsp is a great deal, it is likely to be a major investment in your time.

Moving the microphone slightly can reveal very different frequency responses. Measuring from many locations to equalize room response is a major investment in time and manually attempting to adjust is a very iterative process as there can be significant dependencies. More subwoofers and other speakers usually helps, but adds complexity.

Accurate location of each microphone position for every measurement cycle requires discipline. Am not saying there is no value, but is not for everyone.

For comparable money, you are likely to get better return by adding some room treatment. In any case, you should get audible differences, but it is easier if you are trying to optimize one sweet spot vs. many.

Am familar with Audyssey, do you know if YPAO does anything special for multiple subwoofers?

One final note, better can be subjective in home theater applications. Reducing standing waves should mean better sound, but if the listener likes the impact produced with them, reducing the standing waves may not be considered an "improvement". I have separate systems for music and home theater and tune them for the application.

Ww

 

[ATLAudio]

@Winkleswizard

You make some good points, but let me clarify some things.

"While I agree minidsp is a great deal, it is likely to be a major investment in your time."

It will require a learning curve with REW, but it need not be a major investment of time.

"Moving the microphone slightly can reveal very different frequency responses. Measuring from many locations to equalize room response is a major investment in time and manually attempting to adjust is a very iterative process as there can be significant dependencies. More subwoofers and other speakers usually helps, but adds complexity."

The idea that's worked best for me is to give Audyseey/ YAPO etc the best to work with by using MiniDSP to eliminate major bumps and dips. That way any notch adjustments done post RC will not effect much in other seats; and generally it doesn't. However, I check them too, and they measure very well too. My suggestions with MiniDSP are subwoofer exclusive, and more subs is always a good thing, but for most rooms two subs strikes the best balance, all things considered, four is optimum, but not always possible for a variety of domesticated restrictions.

"Accurate location of each microphone position for every measurement cycle requires discipline. Am not saying there is no value, but is not for everyone."

This isn't hard.

"For comparable money, you are likely to get better return by adding some room treatment. In any case, you should get audible differences, but it is easier if you are trying to optimize one sweet spot vs. many."

I agree that room treatment has benefit for sound above 250 cycles, but not below. Applicable sound treatment for below 80hz costs much more, and takes up A LOT of space, and YMMV big time.

"Am familar with Audyssey, do you know if YPAO does anything special for multiple subwoofers?"

Not sure

"One final note, better can be subjective in home theater applications. Reducing standing waves should mean better sound, but if the listener likes the impact produced with them, reducing the standing waves may not be considered an "improvement". I have separate systems for music and home theater and tune them for the application."

How much we reduce them, and or how much we increase them more, and or other areas is deep deep down the road of preference. MiniDSP is the king of allowing one to curve response to their own preference. That said, I set my MiniDSP to my preference with movies (generally one standard), and since I use my Laptop via iTunes for music, I use the on-board graphic EQ to adjust to the many standards seen in music recording.

 

[Winkleswizard]

Please forgive any newb forum etiquette mistakes on my part, but my intent was to ask Magoo about his Yamaha's YPAO. Also would help to know about his room size...

Also, what you can do with with active bass management is dependent on speakers. What can be done with
PSA XS15s is unlikely to be accomplished with Prosub 800s. Whereas the PSA's design can handle a lot, not sure about the Prosubs. Equalizing some designs risks potentially damaging overexcusions. Also, what may seem fine at calibrated measurement levels, may be very problematic when driven to more continuously or to home theater listening levels.

Agree that low frequency room treatment can be expensive, but should clarify that where a minidsp 2x4 is very reasonable, their room eq solutions are much more expensive.

If you are technically inclined, suggest starting with a minimal investment in a UMIK-1 and try taking some room measurements. Before you go buy a minidsp, check your processing setup. You may be able to tweak equalization without buying any additional equipment.

Minidsp products offer some great capability, but it is a technology that can create unintended effects too!

Ww

 

[KEW]

Minidsp ... is a technology that can create unintended effects too!

Ww

Can you elaborate/give examples on this?

Thanks!

 

[ATLAudio]

@Winkleswizard

"Also, what you can do with with active bass management is dependent on speakers. What can be done with a PSA XS15s is unlikely to be accomplished with Prosub 800s. Whereas the PSA's design can handle a lot, not sure about the Prosubs."

And what one can do with a PSA V3601 might not be likely with a PSA XS15se. Perhaps you're not clear on how much you can do with MiniDSP. You can cut a large room bump, such as the one I had, with ANY sub using MiniDSP allowing Audyssey and or Yapo to achieve a better outcome.

"Equalizing some designs risks potentially damaging overexcusions."

Speaking for most if not all ID sub makers, don't take it as some sort of challenge, but they have enough onboard DSP, and digital and analogue limiters to be incredibly resilient. At worst they'll sound like garbage before they are damaged. That said, improvement can be made with any subwoofer and smart use of a MiniDSP, you might not be able to achieve all possible goals as a more expensive sub, but improvement is virtually a certainty.

"Also, what may seem fine at calibrated measurement levels, may be very problematic when driven to more continuously or to home theater listening levels."

Who said to only test at one level? This is why you'll want to do compression sweeps to understand how you'll perform. I actually find distortion characteristics quite audible during MiniDSP REW testing, and yet don't hear them during actual material. In essence, you've got it backwards.

"Agree that low frequency room treatment can be expensive, but should clarify that where a minidsp 2x4 is very reasonable, their room eq solutions are much more expensive."

Sure, I could be wrong, but I'm pretty sure everyone here was referring to the MiniDSP 2x4.

"If you are technically inclined, suggest starting with a minimal investment in a UMIK-1 and try taking some room measurements. Before you go buy a minidsp, check your processing setup. You may be able to tweak equalization without buying any additional equipment."

I'd buy both together for less than $200, and if you don't need to miniDSP, send it back. Once you find out how your set up measures, you'll want to tweek it.

"Minidsp products offer some great capability, but it is a technology that can create unintended effects too!"

Like what? How?

 

[Winkleswizard]

Can you elaborate/give examples on this?

Thanks!

Sure...

1. To be more specific about my earlier caveat, if you eq a woofer to below its box resonant frequency, the woofer unloads and can go into damaging over-excursions. This is more likely in bass reflex and passive radiators designs, but if using minidsp, can add subsonic filter to reduce. I have experienced this accidentally and it was not pretty.
2. If your minidsp config uses advanced biquad programming (the basic UI handles the biquad programming for you), minidsp has a disclaimer "...first confirm the overall transfer function to confirm that your biquad computations are correct. miniDSP cannot be liable of incorrect/unstable plug-in biquad settings being loaded into the platform." I do not use this feature, but memory gets corrupted in almost every microprocessor-based system...
3. I have used minidsp 2x4 in 2 active speakers designs. It has not happened often, but now and then, have had them "crash". The first time it happened, it took a while to troubleshoot and had me worried I had an amp fail. As well-built as a minidsp seems, it still a software-based product and it seem reasonable that it might hiccup eventually.
4. Clean power is important (surprise ;-). You can power a minidsp via your computer USB port, but I have experienced noise. I have had varying results with different AC adapters and have had better results with beefier ones with in-line ferrite cores.

Despite some "challenges", I really like the minidsp and it has reduced iterative prototyping in speaker design for me. So, if I seemed alarmist, did not intend to be. Was just trying to encourage caution for first timers and share my experiences.

While I have spent some time trying to apply PEQ with minidsp to reduce room uglies seemingly successfully, am not convinced it sounded better (even though it measured better). Seems Pogre had a comparable experience. If others have had different experiences, I can respect that as well.

Ww