Understanding REW

[Iapetus]

I just got my UMIK-1 and opened up REW, loaded the calibration file, ran the sweep and displayed the graph at 1/6

Does this mean there's nothing to be done to improve? I was about to add a subwoofer to my setup but after looking at this graph, wouldn't it be pointless for music at least?

[GaryT]

That's pretty good looking line, you can tweak your slope if you prefer a brighter presentation or to follow a target curve, but apart from that I wouldn't do much to that. 

 

There is more data you can get if you look at your waterfall graphs for the room decay times. That will let you know where you need to focus any room treatment changes. 

 

A sub might make flattening the bass more a bit easier

Edited December 3, 2024 by GaryT

[Satanica]

1 hour ago, Iapetus said:

I was about to add a subwoofer to my setup but after looking at this graph, wouldn't it be pointless for music at least?

 

Not necessarily, you might enjoy a bass boost. But then again, your speakers might be capable of providing that (with EQ)

[mbz]

Drawing a "line of best fit" at about 75dB gives 4kHz-10kHz down by 5dB and 12-14kHz is close to -10dB. Think you need to expand the Y/SPL scale. Yeah probably wait on the sub woofer.

Haven't used REW (yet) but I think there is scope for some tweaking, happy to be corrected by others...

[Iapetus]

This is what Wiim Pro Room EQ tries to do... needless to say it destroys the music.

[Keith_W]

Try changing the vertical scale on that graph. It is rather zoomed out. I can spot a few problems but the vertical scale will make it quite a bit clearer. 

Would you benefit from a subwoofer? This is where I depart from most other "measurement oriented" people. My subwoofers have been set up to take advantage of Griesinger's "stereo bass" theory - that subwoofers set up on either side of the listener provides more spaciousness. He calls it "bassiousness". So I have a nice little demo. I created two DSP profiles - the first has the speakers run full range down to 20Hz. The second has the speakers high passed at 50Hz with everything below sent to the subs. Both were equalised to the same target curve, and volume matched. I can switch between filters nearly instantaneously by pushing a button on the convolver. 

 

So they should sound the same, right? They don't. The version with the subwoofer sounds so much bigger and more enveloping with weightiness and authority over the version without. I don't have a satisfactory explanation for it. 

[adam0bmx0]

On 03/12/2024 at 8:00 PM, Iapetus said:

This is what Wiim Pro Room EQ tries to do... needless to say it destroys the music.

 

Looking at the measured sweep in this very basic plot by Wiim, i'm assuming in your REW plot you're using some sort of EQ?

If you are using EQ and the Wiim measured plot is accurate for the raw non-EQ sweep, then you're adding a huge amount of boosted EQ below (roughly) 70Hz to achieve that nice line, as your speakers drop like a rock below 70Hz but rise a bit again around 38Hz.

[Iapetus]

22 minutes ago, adam0bmx0 said:

 

Looking at the measured sweep in this very basic plot by Wiim, i'm assuming in your REW plot you're using some sort of EQ?

If you are using EQ and the Wiim measured plot is accurate for the raw non-EQ sweep, then you're adding a huge amount of boosted EQ below (roughly) 70Hz to achieve that nice line, as your speakers drop like a rock below 70Hz but rise a bit again around 38Hz.

Both have no EQ applied, which makes me believe that the iPhone microphone just sucks.

[adam0bmx0]

Ah yes, it probably has a narrow measurable bandwidth, as at the other end, after 2.2KHz the sweep shoots off into space!  

[andyr]

On 03/12/2024 at 4:59 PM, Iapetus said:

I just got my UMIK-1 and opened up REW, loaded the calibration file, ran the sweep and displayed the graph at 1/6

Does this mean there's nothing to be done to improve? I was about to add a subwoofer to my setup but after looking at this graph, wouldn't it be pointless for music at least?

 

 

Pretty damn good, if you ask me.  

 

A sub or, preferably, a pair might improve your low-end SQ if you rolled off your mains at, say, 40 or 50Hz ... to a. reduce distortion on the mains and b. lift the frequencies below 40Hz.

 

Plus, if you have DSP available ...removing the dip @ 14kHz might be a good thing.

 

[Iapetus]

Ok so hear me out. I was an idiot. The original graph did have WiiM Room Correction applied, and yes, when zooming in it no longer looked amazing.

 

But I did get a sub! An SVS SB-3000.

 

Green: No Sub, No EQ

Yellow: Yes Sub, No EQ

Red: Yes Sub, Yes EQ

 

I used all 13 available EQ bands (3 from sub, 10 from WiiM)

 

I'm sure I can do better. This was the first try.

I'll be listening to any suggestions

 

Overall thoughts. Bass is no longer boomy, I can easily control it now by changing the volume in the sub, I hear parts of songs that were previously hidden by either that 46hz null, the 80hz boom, those dips under 115hz... tldr: 10/10 would do it again.

 

 

Edited December 24, 2024 by Iapetus

[Iapetus]

[Keith_W]

When posting graphs, please do not chop off the horizontal and vertical scales. It makes it very hard to interpret.