You are touching on listener preference a bit here. Sometimes what initially sounds too quiet is actually the point where the sub becomes more integrated with the mains and less localized. A strong room peak around 40–60 Hz can create a sense of excess weight or impact that people get accustomed to, so reducing it can subjectively sound thinner even if the response is objectively smoother.
I would go with a common recommendation from AVR setups of 30%, no less.
I'm using the same umik-1 w/cal file on both REW (PC) and room fit (iphone).
I use umik-2 and cal file with phone/roomfit but I don't think that makes any difference. Umik-1 would be the same
I spent a lot of time measuring, testing, and reading all I could about the various time delay estimation options in REW. The phase match, phase slope match, and IR match, and others including wavelets. All gave different results. I don't see any way of knowing if there is one best answer. Sometimes it agreed with WiiM, sometimes not. But then it also doesn't agree with itself, so there's something to consider. When there was ambiguity, my conclusion was to discard the big delay numbers (i.e., 30-52ms) and keep the small ones (3-6ms range). Also, fine tuning by ear helps confirm the selection.
I agree it’s nice to have both. My point wasn’t really that WiiM is bad or REW is automatically right. I use both.
I have a longer standing desk setup where I never did the FR+sub alignment. Just recently did it and it wants +8ms for FR. It doesn't make sense to me as FR are very near and line of sight, the sub is further away under the table. To be sure about these recommendations I think it's best to measure with REW and also take a different room position relative to the nearfield setup. If there is a real delay from a component, it will be there constantly. If it's just mic position vs. speaker position, one could chose to ignore that.
In my case it makes sense because my monitors are further away since mounted and sub is closer. I also tested with mains delayed to be certain but it got lesser results
Good point. The total measurable delay has two components: delay due to processing, and delay due to propagation.
hey, how do you send REW test_tones from pc/mac to ultra? it's usb seem to be either file_in or audio_out,That's a good looking result overall.
I had problems with a room mode in a larger space emerging at 38hz, coming back at 76hz and so on.. very wide room, big speakers. Which made me explore how I can make the WiiM Ultra handle that. Initial suggestions were in the range of the default 24db. In REW it's important to tune the equalizer, but also look at the settings for Q - I think "RBJ Q" is default and that is what the WiiM uses.
Not sure if it's wise to do that, but WiiM RoomFit also supports filter stapling. So one can do filter 1 -12db at a frequency and filter 2 the same again - haven't measured that, if it just adds up or has an exponential effect. Nor tried to find proper Q values for that. It cancelled the frequency out more or less completely.
I use the analog output jack on my macbook and connect to the Wiim Ultra's line in. Works great. I also tried saving REW test clips to wav files on a usb stick and inserting this into the Ultra to eliminate any possibility that the macbook's dac could be effecting things, but I got the exact same measured results both ways.
