The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Big frequency dip in treated room - what are my options? Studio Monitors
Old 1 week ago
  #1
Gear Nut
 

Thread Starter
Big frequency dip in treated room - what are my options?

Hi all. Sorry in advance for the long post, but here goes...

I have a small home studio (non-professional), which I’ve been slowly improving using the advice shared on this forum and elsewhere. Despite various acoustic treatments, I'm unable to fix an especially large 'Grand Canyon' of a dip between around 60Hz - 100Hz. I understand that this dip is very typical of small studios and usually caused by SBIR cancellation.

My question is: what can I do about it?

Some background information on the room:

The room measures 5.3m (l) x 2.1m (w) x 2.3m (h). Walls are plasterboard mounted on brick. There is a window in front of the desk (between the front wall bass traps), which is covered by a canvas roller blind.

The room is treated with bass traps built by Blue Frog Audio using Ecose mineral wool:
  • Front corner bass traps: 16" x 16" quoted effective down to 40Hz with a ~10cm air gap between traps and wall
  • Side panels and ceiling cloud: 5" thick effective down to 80Hz
  • Rear left corner bass trap: 16" x 16" effective down to 40Hz
  • Other rear wall bass traps: 7" thick effective down to 65Hz
Monitors are a pair of Unity Audio The Rock MKII, frequency range 37Hz-25kHz +/- 3dB.

Notes on room measurements:

I've attached some hopefully useful diagrams to this post. All measurements were taken from the listening position (approx 38% room length) using a calibrated omni mic (Sonarworks XREF 20) via Fuzzmeasure. 10 second sweeps. Measurements taken were: left speaker (shown red), right speaker (shown blue) and both speakers (shown cyan).

NB: I’m very much a noob when it comes to acoustic measurement, but have tried to follow the advice in the stickies as best I can! Don't hesitate to let me know if I've made a complete hash of it

What I'd specifically like to know is...

1) Can I realistically tackle such a big frequency dip between 60Hz-100Hz by adding more traps to reduce reflections/cancellation? If so, what types of traps would you recommend and where should they be placed? (noting the current layout of the room shown in the Sketchup images - see next post below).

2) Should I instead try to fix the big dip by installing a subwoofer to help even out the low frequency range? (I've read with interest this article on SoundOnSound and this corresponding thread on Gearslutz). If so, should I opt for buying one great quality sub (e.g. the Unity Audio Avalanche) or two good quality subs (e.g. Dynaudio 9S)?

Very grateful for any advice. Thanks!
Attached Thumbnails
Big frequency dip in treated room - what are my options?-etc_l-r.jpg   Big frequency dip in treated room - what are my options?-freq_full-range_3.jpg   Big frequency dip in treated room - what are my options?-freq_full-range_24.jpg   Big frequency dip in treated room - what are my options?-freq_low-end-l-r_48.jpg   Big frequency dip in treated room - what are my options?-room-modes-plot.png  

Big frequency dip in treated room - what are my options?-waterfall_full-range_24.jpg   Big frequency dip in treated room - what are my options?-waterfall_low-end-l-r_48.jpg  
Old 1 week ago
  #2
Gear Nut
 

Thread Starter
Here are some Google Sketchup images of the room itself...
Attached Thumbnails
Big frequency dip in treated room - what are my options?-room_perspective1.jpg   Big frequency dip in treated room - what are my options?-room_perspective2.jpg   Big frequency dip in treated room - what are my options?-room_side-1-wall-removed.jpg   Big frequency dip in treated room - what are my options?-room_side-2-wall-removed.jpg   Big frequency dip in treated room - what are my options?-room_top.jpg  

Old 1 week ago
  #3
Gear Nut
Get a decent subwoofer. That allows you to optimize the placement purely with regard to bass response and without any of the constraints from normal stereo image etc. (since subwoofer frequencies are very hard to localize). I had the same issue and getting a sub solved it almost perfectly.
Old 1 week ago
  #4
Gear Maniac
 
ReDRuMx's Avatar
Narrow room, sidewalls very close, those panels are not going to stop SBIR nulls from the speakers.

Subwoofer, good advice.

Have you ever tried placing the speakers in the corners of the room? At each side of the window. You could try it if it's not too much work (would require moving the top half of the corner trap I guess).
Old 1 week ago
  #5
Gear Nut
 

Thread Starter
Quote:
Originally Posted by Antti H View Post
Get a decent subwoofer.
Quote:
Originally Posted by ReDRuMx View Post
Narrow room, sidewalls very close, those panels are not going to stop SBIR nulls from the speakers. Subwoofer, good advice.
Thanks guys. That confirms my thinking that a subwoofer is probably the best way to try tackling the 60-100Hz dip.

Do you think one sub will be sufficient or should I be aiming for two? Discussion elsewhere seems to suggest that using two subwoofers to fix SBIR nulls is a markedly better solution that using one by its own. But I’d much rather use just one if that’s viable (for reasons of both space and budget!).

Quote:
Originally Posted by ReDRuMx View Post
Have you ever tried placing the speakers in the corners of the room? At each side of the window. You could try it if it's not too much work (would require moving the top half of the corner trap I guess).
I haven’t tried this. It’s tight for space behind my desk and so to place the speakers in the corners I’d need to either:

1) Remove the corner bass traps entirely;
2) Remove the top corner traps and mount the speakers on top of the lower traps or on wall brackets; or
3) Try some kind of ‘DIY soffit’ mounting solution, trying to place the speakers inside the top corner traps… something I’m almost guarantee to make a mess of

So adjusting the horizontal positioning of the speakers is a little difficult, but not impossible.

EDIT to say that I did also consider pushing the speakers back closer to the front wall (to try and avoid the SBIR cancellation notch), but they're already quite close at approximately 50cm distance.
Old 1 week ago
  #6
Gear Maniac
 

Get 2 subs and place them left and right of the speakers
Old 1 week ago
  #7
Gear Nut
 

Thread Starter
Quote:
Originally Posted by arctic audio View Post
Get 2 subs and place them left and right of the speakers
Thanks. So 2 subs are definitely better than 1 for the application?

A quick general question for all...

As an aside, I noticed when taking measurements that the left speaker response consistently shows a stronger dip in the 60-100Hz than the right-hand speaker (as seen in the more detailed freq response graph at 1/24 octave smoothing).

I have a small rack unit on the left-hand side of my listening position which obscures some of the bass trapping - is this likely to be the cause of the increased dip on that side (i.e. additional surface causing additional reflections?). The left and right-hand sides of the room are otherwise fairly symmetrical at that point.
Old 1 week ago
  #8
Gear Nut
Quote:
Originally Posted by Skol303 View Post
Thanks. So 2 subs are definitely better than 1 for the application?
Yes, although it's somewhat controversial whether it's better to run the subs as stereo or with the same mono signal. The latter allows you to optimize the combined placement to provide even smoother frequency response. In any case, even a single mono sub is better than no sub when you have bass issues.
Old 1 week ago
  #9
Gear Guru
 
DanDan's Avatar
Options

A GS here ended up with that often occurring LF dip, despite extensive treatment. He ended up using three subs, with a very good result, including noticeably controlled low modes.

Aside from that, and which would most likely benefit with or without subs, there are simple options.
Place your speakers almost touching the Front Wall, to avail of that free clean LF boost and extension. If it is a bit floppy, block the ports in your speakers and or use Eq or DRC. A client of mine had a similar space. Basically a corridor at the side of a live room. Plse look carefully at the Pic. Speakers are almost touching the Front Wall. There is about 30cm of fibre trap at the side of each speaker. Overhead is a massive fibre trap. Underneath both speakers will also be fibre filled. These deep corner traps, including overhead, extend gradually narrowing along the desired Reflection Free Zone.

DD
Attached Thumbnails
Big frequency dip in treated room - what are my options?-corridor.jpg  
Old 1 week ago
  #10
Gear Nut
 

Thread Starter
Quote:
Originally Posted by Antti H View Post
Yes, although it's somewhat controversial whether it's better to run the subs as stereo or with the same mono signal. The latter allows you to optimize the combined placement to provide even smoother frequency response. In any case, even a single mono sub is better than no sub when you have bass issues.
Thanks, that's good to know. I think I’m going to audition a single sub from my supplier, see how that works on its own, and then audition further from there as necessary.

Quote:
Originally Posted by DanDan View Post
Place your speakers almost touching the Front Wall, to avail of that free clean LF boost and extension. If it is a bit floppy, block the ports in your speakers and or use Eq or DRC.
Nice… that’s going to be my first step

It’s difficult for me to install traps directly behind the monitors due there being a window and not a wall; but I have an idea to suspend some 4-5” thick mineral wool panels from the ceiling so they hang directly behind the monitors, once I’ve moved them closer to the front wall (or window as it happens to be).

I’ll move the monitors first anyway and post some updated measurements here for reference.

PS: my monitors are sealed and not ported. The manufacturer doesn’t specify a recommended minimum speaker/boundary distance (I’ve contacted them to confirm), but I assume that I could move the speakers within say 20cm quite comfortably without a rear port to worry about - ?
Old 1 week ago
  #11
Gear Guru
 
DanDan's Avatar
Behind the Monitors

This is confusing because there are conflicting recommendations.
My take. Flush Mounting is the ultimate. As close as possible is the next best thing.
There is no advantage to adding absorption behind the speakers unless it has a huge surface area, full wall ideally. Even then, the loss of free LF boost.....
A window behind the speakers IS an LF absorber. No need for any more.
Back to next best thing..... I would get the speakers as close to or into that wall. If possible on the window ledge.
Whatever the manufacturer says, all speakers can be placed as close to the Front Wall, or in it, and all experience the same well known acoustic phenomena. Increased LF, but driven from the boundary, so strong modal stimulation.

Most of us like the bass.

Due to the repeated myth that speakers need some distance from the wall it is hard to shake. 20cm? Try 2mm, but on the window ledge is more to the point.
If you still have weak LF, consider blocking that window.


DD
Old 1 week ago
  #12
Gear Nut
 

Thread Starter
^ Thanks again DanDan, it’s great to have some clarity on this.

Unfortunately I can’t block up the window (it’s required as a fire escape from the room according to UK planning law); but I was just today thinking about fitting a freestanding bass trap in the window space and then moving the speakers right up against it within a few mm.

From what you’re saying, I assume there’s no need to fit any additional bass trapping in the window? I should just move the monitors as close to the window as possible - or better still, inside the window space itself. I'm going to experiment with that next chance I get.

If I still have a weak LF after doing so, I might consider getting a sub. I'm going to audition one anyway to see how my room responds.

Thanks again to you and the other contributors here - really appreciate the advice
Old 1 week ago
  #13
Gear Nut
Quote:
Originally Posted by Skol303 View Post
I should just move the monitors as close to the window as possible - or better still, inside the window space itself.
The reasoning behing moving the speakers close to the front wall (iow, close to the reflecting boundary) is to move the first notch to as high frequency as possible where either absorption behind the speaker is more effective or the speaker is directional enough that little energy is radiated towards the boundary.

Unfortunately when you do the math it's non-trivial to make it work like the theory says. Assume your speaker is 30 cm deep (or 25 cm + 5 cm room for bass reflex port). Then the highest you can get the first notch is 343 / 0.3 / 4 = 286 Hz. At that frequency the speaker is almost omnidirectional so it will radiate sound backwards which will reflect and cause a notch at that frequency.

Another option is to put absorbing panel right between the speaker and the wall. Assume 10 cm panel & same speaker size as before -> 343 / (0.3+0.1) / 4 = 214 Hz. Problem is, a 10 cm thick panel with no air gap may not attenuate so well at that frequency, so you will still get some reflection. A 15 cm panel is better but is of course quite large. You also need to cover significant area behind both speakers - a single 60x60cm panel likely won't do much.
Old 1 week ago
  #14
Gear Guru
 
DanDan's Avatar
Nada

+1 Antti. A small area of trap will not do anything at LF.
The other reason to get close to the Front Wall is the free LF boost, and extension of the lower freq cut off downwards.
I have often seen the response get worse with 100x60x1200 traps inserted.
In this case the boundary behind the speaker is half window, so no chance.
Inside the window frame, flush to the walls can get a big of free LF and raise that dip frequency. Or just close, but more on wall than on window.

DD
Old 1 week ago
  #15
Gear Nut
 

Thread Starter
Thanks guys, I'm getting a great education here! Appreciated.

My monitors have a depth of 29cm; rounded up to ~30cm for argument's sake. No port on the back as they're sealed.

So by Antti's calculation, my best option is to forget sacrificing any more centimetres with bass trapping to the rear and just place the monitors as close to the front wall/window as possible (mindful that the window is already a kind of a bass trap, as mentioned by DD above). Or better still place the monitors inside the window space itself - not sure if that's viable, but I'll try it.

And by doing so this might push the SBIR notch upwards to a maximum of around 280Hz, which for me is a much more 'workable' frequency at which to have a notch than 60-100Hz. And I might get a little LF boost thrown in for good measure.

Totally appreciate that it might not work so easily in practice, but this at least gives me a very sensible solution to try out.

Thanks for the clear explanations guys! You've done a great job in demystifying acoustic physics for this noobie

I'll let you know the outcome.

EDIT: sorry, one final clarification... is it definitely not worth me adding bass trapping to the window space? I've read elsewhere that this can help to reduce colouration due to reflections of higher frequencies by the glass. The windowsill itself is around 23cm wide, so if I'm unable to mount the speakers in the window space, I could potentially fit a fairly chunky bass trap here (8" thick) which might also help with LF reflection - ?
Old 1 week ago
  #16
Gear Maniac
1 sub

i'd strongly recommend just 1 sub: way easier/less critical to position
Old 1 week ago
  #17
Gear Nut
Quote:
Originally Posted by deedeeyeah View Post
i'd strongly recommend just 1 sub: way easier/less critical to position
You mean feeding mono signal. Having two mono subs is actually easier for positioning as long as you have a measurement mic.
Old 1 week ago
  #18
Gear Maniac
1 sub cont.

there's no benefit from using two subs in a rather small room - start with one!
Old 1 week ago
  #19
Gear Guru
 
DanDan's Avatar
Clarifications

Quote:
EDIT: sorry, one final clarification... is it definitely not worth me adding Bass trapping to the window space? I've read elsewhere that this can help to reduce colouration due to reflections of higher frequencies by the glass. The windowsill itself is around 23cm wide, so if I'm unable to mount the speakers in the window space, I could potentially fit a fairly chunky bass trap here (8" thick) which might also help with LF reflection - ?
Nothing is definite in small rooms with many unknown construction and other details. Elsewhere be Dragons...... There is no HF radiation to the rear of a loudspeaker so no possible HF colouration of the sound emanating from that speaker. Dragons are not great acoustically. But if you have flutter echo from back wall to front wall, sure fibre treat the window, although a curtain would be enough. And since your BW will be extensively treated this won't happen in any case. 8" inches is impressive sir......;-) By all means try it and test it, but bass trapping is way more efficient in corners, and your window is already floppy to LF.

DD
Old 1 week ago
  #20
Gear Nut
 

In the hi-fi world they typically recommend moving the speakers away from the wall. Pumping the room at it's boundaries promotes strong bumps and dips due to standing waves being generated at 1/2 wavelength intervals of the room size. In your case, 5.3 meters works out to 1 wavelength of 64Hz, almost exactly where your dip is. If you move the mic from the front of the room to the back you should see it peak and dip as you move along the 5.3m dimension.

So perhaps cutting the main monitors off well above 60 Hz and using subs that are not placed at that same boundary might help the situation. The ideal is typically to put the speakers at about 1/3 of the dimension of interest.
Old 1 week ago
  #21
Gear Guru
 
DanDan's Avatar
Dragons

Jeez..... Dragons are coming to Shore....
Two subs allow way more options and are way more powerful than one. Using multiple subwoofers to improve bass: The Welti / Devantier and Geddes approaches - Acoustic Frontiers
The Hi Fi World....... no comment necessary. But it is nonsense to move speakers arbitrarily away from the wall. At or in the wall, the LF boost is gradual and smooth and easily eq'd down, if necessary, which it rarely is.
On the other hand the common Hi Fi distances remove particularly bass notes from the LF response. Very far from the Front Wall or Very Close are the best options. Genelec used to recommend Less than 20cm or More than 2.2M.
REW Results Comparing Layout Options

DD
Old 1 week ago
  #22
Gear Nut
 

Quote:
Jeez..... Dragons are coming to Shore....
What do you mean by "dragons"?

My understanding has always been that flush mounting(or mounting speakers in the wall) fixes the half-space/full-space radiation problem, making mids(assuming cabinet dimensions are in the midrange wavelength area) sound better, but aggravates the bass standing wave problem.
Old 1 week ago
  #23
Gear Maniac
1 sub cont. II

have you ever been measuring/listening how two spaced subs compare to one sub in the center?!?

you'll be in trouble in this room with two (or more) subs unless you use a second one in the rear to cancel out unwanted bass (check for klein+hummel, now neumann kh; they called it 'aram' if i remember right)

of course the one and only sub you need needs to be powerfull enough/on par with the mains.

also, translation to lcr+sub or to 5.1 (if ever needed) is far easier.

(no dragons here but thousands of hours on the road and in studios, re-/designing/installing and recording/mixing)
Old 1 week ago
  #24
Gear Guru
 
DanDan's Avatar
Research

There are many many papers and other writings on the benefits of multiple subs.
Quote:
One subwoofer at each wall midpoint is the best in terms of Std, Max-ave and Max-min but does not support low frequencies particularly well. Two subwoofers, at opposing wall midpoints, performs very nearly as well as four at the midpoints and gives a much better LF factor. One subwoofer in each corner also has good low frequency support, but does not perform quite as well as one subwoofer at each wall midpoint, in terms of Std, Max-ave and Max-min. If cost and aesthetics are considered, subwoofers at 2 wall midpoints is preferred.
Certainly a speaker flush mounted will drive the axial mode maximally. Both in terms of location and LF boost due to the now much larger baffle.
If the boost is too much, simple Eq can reverse it. The extra modal activity due to location at or near the boundary is minimal at the lowest modes where the wavelengths are vast in comparison to the distances from the Front Wall.
Missing Bass due to SBIR is IMO a way worse problem than Modes. Ideally we use treatment that addresses both. e.g. Flush Mount speakers and a deep BW treatment. Do room modes even matter? - MusicPlayer Forums

Puff....

DD
Old 1 week ago
  #25
Gear Maniac
besides the costs

Old 1 week ago
  #26
Gear Guru
 
DanDan's Avatar
Evidence

Quote:
there are at least as many papers why NOT to use more than one sub.
OK, a link to one or two please.

DD
Old 1 week ago
  #27
Lives for gear
 
Jens Eklund's Avatar
Old 1 week ago
  #28
Gear Nut
 

Thread Starter
Thanks again guys, really appreciate your insights on this.

Antti H & deedeeyeah… different views on the optimum number of subwoofers. Must admit I’ve been reading up on this and opinion is certainly very polarised. I did find this useful paper that sheds some light on the topic - here’s a link for reference:

Subjective preference of modal control in listening rooms (PDF Download Available)

“The configurations associated with the highest quality are based on active removal of energy for the room... These are also the most expensive to implement since they require, at least, two subwoofers and a digital signal processing unit.”

That said I may order two subs to audition and try for myself; that way I can experiment with both single and dual options.

Quote:
Originally Posted by DanDan View Post
There is no HF radiation to the rear of a loudspeaker so no possible HF colouration of the sound emanating from that speaker… But if you have flutter echo from back wall to front wall, sure fibre treat the window… By all means try it and test it, but bass trapping is way more efficient in corners, and your window is already floppy to LF.
Thanks DanDan, makes sense. I’m going to forego treating the window space for the time being and focused my efforts elsewhere.

Quote:
Originally Posted by u87allen View Post
Pumping the room at it's boundaries promotes strong bumps and dips due to standing waves being generated at 1/2 wavelength intervals of the room size. In your case, 5.3 meters works out to 1 wavelength of 64Hz, almost exactly where your dip is. If you move the mic from the front of the room to the back you should see it peak and dip as you move along the 5.3m dimension. So perhaps cutting the main monitors off well above 60 Hz and using subs that are not placed at that same boundary might help the situation.
Interesting! That chimes a little with the recent measurements I’ve taken.

----------

So, my recent measurements… I’ve now moved the monitors closer to the front wall so that the speaker cones are approximately 40cm from the wall (I could possibly move them closer to around 30cm max; any further would mean mounting the speakers in the window space itself and that’s going to be tricky - but not impossible).

I’ve taken some new measurements and the key points seem to be:

1) The closer wall proximity has given a noticeable LF boost and also helped to attenuate some of the 60-100Hz dip (by around 10db approx), as predicted by DanDan. So that’s helped.

2) However, the null remains and at the same frequency - so I’m assuming this means I also have a modal problem between 60-100Hz as suspected by u87allen. So it’s looking like a double-whammy of SBIR and room modes. This appears to be corroborated by the measurements taken when I move the listening position L/R/front/back.

See graphs attached:

Red shows original listening position
Blue shows new listening position with monitors closer to front wall

1 Full frequency range, L speaker only
2 Low end frequency range, L&R speakers combined
3 Low end waterfall, L&R speakers combined

Orange shows adjustments in relation to new listening position (blue). I’ve also included Sketchup pictures showing the approximate placement of measurement mic for each.

4 LP moved forwards towards front wall
5 LP moved backwards away from front wall
6 LP moved left
7 LP moved right

Interestingly moving the mic/LP to the right almost totally negates the null… but would be a crappy position from which to mix! Dammit.

So… back again to asking what are my options?

So far I’m considering possibly all of following:

1) Mount the monitors inside the window space somehow.
2) Bass trapping the bejesus out of the room! Possibly including some hefty gobo-style bass traps that I can position at the back wall when mixing.
3) Experimenting with a subwoofer(s).

Any other comments or ideas very welcome and valued.

Thanks everyone
Attached Thumbnails
Big frequency dip in treated room - what are my options?-1-freq_full-range_24.png.jpg   Big frequency dip in treated room - what are my options?-2-freq_low-end-l-r.jpg   Big frequency dip in treated room - what are my options?-3-waterfall_low-end-l-r_48.jpg   Big frequency dip in treated room - what are my options?-4-freq_lp-forwards.jpg   Big frequency dip in treated room - what are my options?-5-freq_lp-backwards.jpg  

Big frequency dip in treated room - what are my options?-6-freq_lp-left.jpg   Big frequency dip in treated room - what are my options?-7-freq_lp-right.jpg   Big frequency dip in treated room - what are my options?-listening-position-lp.jpg   Big frequency dip in treated room - what are my options?-lp-back.jpg   Big frequency dip in treated room - what are my options?-lp-forward.jpg  

Big frequency dip in treated room - what are my options?-lp-left.jpg   Big frequency dip in treated room - what are my options?-lp-right.jpg   Big frequency dip in treated room - what are my options?-photo-lp.jpg  
Old 1 week ago
  #29
Gear Nut
 

Thread Starter
Quick note to add that when I play a low frequency test tone (10Hz increments), I get a big null at 60Hz - the signal drops out significantly - so I think there’s definitely a monster of a room mode at work here!

...and SBIR to boot. Ah the joys of a small room
Old 1 week ago
  #30
Gear Nut
Quote:
Originally Posted by Skol303 View Post
Quick note to add that when I play a low frequency test tone (10Hz increments), I get a big null at 60Hz - the signal drops out significantly - so I think there’s definitely a monster of a room mode at work here!

...and SBIR to boot. Ah the joys of a small room
Simple way to check if it's SBIR from the front wall is to move the speaker a bit and see if the notch frequency changes. For side / back wall / floor / ceiling SBIR you can move the mic / listener and check the same thing. If the notch frequency (but not necessarily amplitude) stays the same, it's modal.

E: For the reference, I consider SBIR to be a horrible term that mostly just confuses people. It's just a reflection (or multiple reflections), nothing more.
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+  Submit Thread to Reddit Reddit 
 
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump