1. Join Now

    AVForums.com uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Tmreq

Discussion in 'TAG McLaren Audio Owners' Forum' started by Neale, Aug 15, 2004.

  1. Neale

    Neale
    Standard Member

    Joined:
    May 22, 2004
    Messages:
    83
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    6
    Location:
    Gerrards cross
    Ratings:
    +0
    Wow I know I am preaching to the converted but have just finished TMREQ my room. I first flattened out the sub a velondyne with its own equalisation and then procedeed to do the rest of the speakers
    I was suprised at how much difference it made to the higher notes the soprano in fith element just sounded so much clearer.
    It was interesting going between pre and post Tmreq with the different zones
    The bass was not so loud at the same volume but much clearer had to readjust the speakers settings
    This was not marketed properly I think IAG have got theselves such a bargain!! (All though I am not sure how much they paid)
     
  2. Mike192

    Mike192
    Active Member

    Joined:
    Jun 10, 2004
    Messages:
    58
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Ratings:
    +5
    Hi Neale,

    I agree with you 100%. I've been using TMREQ for over a year now, since before John's excellent TMREQ wizard was released and it used to take well over an hour per speaker to set up. Its much easier now but even then the results were worth it. I find everything so much clearer, especially in the low frequencies and individual instruments are more distinct. Without the low frequency room resonances you can run at much higher volumes. I certainly wouldn't be without it.

    Mike
     
  3. lenny

    lenny
    Active Member

    Joined:
    Jul 8, 2004
    Messages:
    429
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Paris
    Ratings:
    +6
    I confirm what you affirmed in your postings. I just started playing with TMREQ using just a SPL: the results, even using simply the wizard, and not playing at all with the Q values, are excellent. I was suprised by the fact that it is very easy to use, even if my wife was not happy during the process because of the noise. I fully agree on the remark that Tag had a great product that would have deserved to be marketed with a "Louis Vuitton style" (a French luxury brand).
    However I have a couple of questions:
    1) TMREQ target set point : I fixed my set point to 75dB and than I tuned the subwoofer first (a REL400E) than the main speakers (a pair of old Celestion 5) and finally the surrounds (a pair of little floor standing Audio Reference). In the TMREQ manual I realised that Tag suggested fixing the target set point to 85 dB. If I use that level the sound in my room if far too high. I believe that it is not a big issue, and that was better to optimise the system around the normal level at which I'm going to listen at the movies. Am I wrong? Should I optimize at 85db and than just settle the volume as I want?
    2) Next step, use ETF5 and a sound card: in order to go even further it seems I need to invest in an USB sound card (since the one I have on the company laptop has only a microphone input and an headphone output), the connecting cables and the ETF5 package in order to be able to save and work on the measurements. Am I wrong? The other point is on the ETF5 licence: can I install it on my company laptop and than, as soon I got my "family" computer, install it on the new unit or should I wait the "family" computer since it is not possible to install it more than once?

    Thank you in advance for your opinion.

    Lenny
     
  4. simon40

    simon40
    Active Member

    Joined:
    May 30, 2003
    Messages:
    191
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Dorset
    Ratings:
    +4
    Lenny,

    Not sure as to question 1, but with Q 2 if I remember correctly ETF allow users to register their software on two computers - might be best to check with Doug at ETF.

    Don't forget, if you use IAG/TAGs (John's easy set-up) wizard it is registered to the computer - to use on a second computer you will have to contact the Helpdesk and pay an additional £10.

    Simon
     
  5. UrbanT

    UrbanT
    Distinguished Member

    Joined:
    Jan 6, 2002
    Messages:
    12,876
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Aldershot, Hants
    Ratings:
    +3,036
    I have to say I've seen the light as far as TMREQ is concerned. I've been down the road of trying to use ETF5, setting up the sound card, balancing line in and outs, try to calibrate the sound meter and so it goes on. 13 hours solid of frustration and much swearing gave a fairly decent result.

    Now I have the new, simpler way that has been shown to me, for all those frustrated with this process. How?

    Firstly, its easier with 2 people, grab the missus or a friend. You need a simple text file with all the frequencies from 20 to 200 written on it. It has to be in a certain format which I can email to anyone wanting it. Then, using the 'sine' test tone from the Tag, take an individual reading for each 1hz between 20 and 200. Yep, thats 180 readings. With one person adjusting the hz in steps and calling the readings, anmd the other typing them in, this is a 10-15 minute process.

    Import these to the wizard (ctrl-i). Set the level between 75 and 85db depending on your preference. Next press the 'find peaks' button, then 'add filters', and then 'optimise'. You may want to play with a couple of tweaks if you feel it necessary, but otherwise just 'ctrl-d' to download the settings in to the Tag. Each speaker takes about 20 minutes using this method.

    Apologies to those who do this already, but thought it worth sharing to those still struggling with the massive case study, ETF5 and its hundred options etc. I just wish I had known about this before :)
     
  6. Daneel

    Daneel
    Active Member

    Joined:
    Dec 5, 2002
    Messages:
    2,822
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    68
    Location:
    London
    Ratings:
    +26
    What is the spec of TMREQ, in comparision to say the Pioneer MCACC system which uses 5 or 9 fixed bands for each speaker + / - 6 dB?
     
  7. Stevesky

    Stevesky
    Active Member

    Joined:
    May 13, 2004
    Messages:
    454
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +13
    TMREQ is based around a matrix of 52 ((4*8)+(4*5)) high definition 80bit fully definable parametric filters. You can define a filter at any frequency to 1 Hz resolution with adjustable Q, allowing filters to be created that have a bandwidth of a few hertzs. Ideally you want .5Hz resolution, but TMREQ was a first attempt based on a bit of research done in spare time! :eek:)

    A fixed frequency system with relatively wideband filters is just an auto graphic equaliser that will tweak the overall tonal content of the sound, but isn't much cop for room equalisation purposes. Having examined quite a few room eq systems in receivers recently there does appear to be quite a bit of marketing spin on most of them as most are just auto setup graphic equalisers, but typically do more damage than good.

    For true room eq systems check out people like Meridian, Lexicon, TAG/IAG and TACT.

    I'm sure true room eq will appear in cheaper products, but it does chew up quite alot of DSP resource (= money) and to do auto calibration properly needs a reasonable amount of memory and gets tricky for some speaker configs.
     
  8. lenny

    lenny
    Active Member

    Joined:
    Jul 8, 2004
    Messages:
    429
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Paris
    Ratings:
    +6
    Simon 40 and Urban thank you very much for your help.

    Urban : I think I'll go with your suggestion. Could you please e-mail the text file to my personal address?
    It is going to save me 150USD + sound card + some problems...

    Regards, Lenny
     
  9. UrbanT

    UrbanT
    Distinguished Member

    Joined:
    Jan 6, 2002
    Messages:
    12,876
    Products Owned:
    1
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Aldershot, Hants
    Ratings:
    +3,036
    Email sent Lenny :)
     
  10. Nuno Santos

    Nuno Santos
    Active Member

    Joined:
    Jul 25, 2000
    Messages:
    148
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Portugal
    Ratings:
    +3
    Hi!

    Since I’ve moved to my new house, I have been trying to configure the TMREQ settings on my DP.

    I use a USB sound blaster and after some calibration I started doing my system’s readings, and now have (quite!) a few questions:

    My front and sub readings are in the attachment file (hope it works):

    Q1: Having in mind the speaker’s response, should I XO at 60Hz, or lower?

    Q2: I think I should first correct the Sub and then proceed with the front speakers. After correcting the sub, should I do the readings with bass redirected to the sub (if speakers are configured as bass limited) or should I correct the speakers as stand alone?

    Q3 : Do you accept exactly what the TMREQ wizard configures (after some manual adjustments, if required), or do you use the ETF “calibrated” readings and configure the filters using ETF’s output curves? (therefore using the Wizard for starters only)

    Q4 : Does it make any sense to use a Y connector and read the L&R speakers at the same time? Could there be cancellations between speakers that are not considered when setting up individually?

    Q5 : After the wizard does his thing and you download your filter settings to the unit, did any of you read the speaker’s response again and compared the before & after with the overlay function? Was the outcome similar to the one reported on TAG’s case study? On the waterfalls did found the same outcome?

    Q6 : Do you load the calibration file on the ETF SW when using the RS SPL meter?

    Any extra advice on using the TMREQ is welcome.

    Thank you for your help.

    NS
     

    Attached Files:

  11. Daneel

    Daneel
    Active Member

    Joined:
    Dec 5, 2002
    Messages:
    2,822
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    68
    Location:
    London
    Ratings:
    +26
    Thanks for that Stevesky. Does the system auto-setup or do you have to manually set the EQ points and parameters?
     
  12. Stevesky

    Stevesky
    Active Member

    Joined:
    May 13, 2004
    Messages:
    454
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +13
    If you use the TMREQ wizard then the filter calculation and transfering them to the AV is automatic. You need to do measurements of your room manually with a program called ETF. If you do a search on this forum there was a thread that had copies of the step by step guide that JohnM wrote for the original TMA site.
     
  13. JohnMulcahy

    JohnMulcahy
    Member

    Joined:
    Jun 13, 2004
    Messages:
    181
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Aldershot
    Ratings:
    +20
    Hi Nuno,

    60Hz looks like it would be about right for the crossover on your front speakers, certainly should be no lower - much of the output between there and 50Hz is from the resonance at 57Hz or so, things will look rather different once that is removed.

    After you have done the sub, do the other speakers with bass redirection turned off initially as otherwise the sub's behaviour can mask the behaviour of the main speakers. Then make measurements with bass redirection on and compare the results, some resonances which are present when driving just the main speakers will disappear when the sub is also active, so do not need to be corrected. The same is true when both L and R (for example) run at the same time, you do not need a Y lead to check the behaviour in this case as the TMREQ menus allow you to have the test signal reproduced by more than one speaker.

    Generally the Wizard will do a reasonable job, but there will be resonances which are masked by others when the initial readings are made that will only become apparent after making readings with correction filters in place, so subsequent tweaking can be based on further ETF measurements - this can still be done in the wizard if you have spare filters to use, just load the reading as normal, disable (within the wizard) all the filters that were in use when the reading was taken (so as not to apply them twice, in effect) then use the remaining filters as required to further improve the response.

    On your Q5, it is very sensible to make before and after measurements and compare them. Difficult to confirm the outcome any other way :)

    It is not necessary to load the ETF RS meter cal file for LF measurements as it has no effect at low frequencies, but it can be useful if trying to interpret full range measurements.

    Note that it is also advisable to run an ETF level calibration before making your measurements, so that ETF shows "calibrated" in green. The actual number you calibrate to does not really matter, pick something close to the SPL meter reading when the cal signal is played. The important thing is that by calibrating ETF it will stop auto-scaling the highest peak to 100dB, so you can then make comparisons between left and right speakers, for example, and between LF and full range measurements (note that there is typically a 2.5dB difference in level between ETF LF and full range readings, the wizard automatically adjusts for this when importing the data).

    When making readings after ETF calibration the wizard's target level should be the same for all speakers (assuming they were all calibrated to the same spl level in speaker cal, e.g. all 75dB. If, for example, the surrounds were deliberately calibrated a few dB higher then the target level for them should be higher by the same amount). Check the various speaker responses and pick a target level that gives the best compromise for all.

    Note that although the non-ETF method UrbanT describes above is a bit more laborious for getting the actual measurements, it does give very good results and of course the readings are always true spl values so readings from each speaker can be directly compared and the target level should correspond similarly.

    Regards,
     
  14. lenny

    lenny
    Active Member

    Joined:
    Jul 8, 2004
    Messages:
    429
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Paris
    Ratings:
    +6
    I got it UrbanT. Thanks a lot.

    Regards.
     
  15. Nuno Santos

    Nuno Santos
    Active Member

    Joined:
    Jul 25, 2000
    Messages:
    148
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Portugal
    Ratings:
    +3
    Thanks a lot John for your help.

    So 60 Hz it is. Its a shame that even with the sub (a REL Strata III) there's little response between 56Hz and 33Hz. And then there's that peak at 30 Hz... The valley between 56 and 33 makes it difficult for an even response from the sub.

    So in the TMREQ Wizard, when setting up the filters, I should configure the L&R speakers as full range? Or does the TMREQ SW still uses the XO frequency and treats the speaker as bass limited?

    So I might be eliminating some peaks on the L&R spks that I would not need to and therefore go back and disable them, right?

    So, first do the front speakers, then measure them at the same time to see if there are cancellations creating any abnormal behaviour. If so, rearrange the filter settings on L&R again.

    Do you think it's best to do one filter at a time? I haven't thought much on this, but sometimes I feel that some filters interact making a "global" aproach not as precise comparing to a one at a time aproach.

    Thanks again for your help,

    Best regards.

    NS
     
  16. JohnMulcahy

    JohnMulcahy
    Member

    Joined:
    Jun 13, 2004
    Messages:
    181
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Aldershot
    Ratings:
    +20
    Hi Nuno,

    Is the sub in a corner? If not, I would place it there, or at least against a wall.

    Leave the speakers set up with the XO frequency you have selected. When looking at the results the measurements with bass redirection off should be compared with the corresponding high pass bass management curve the wizard will show, when making a measurement with the bass redirection (and sub) on, the overall target is to have a flat result, of course.

    I wouldn't do one filter at a time, as one of the big advantages of the wizard is that it computes the combined responses of the filters and so allows for their interaction. However, after correcting the major peaks that were visible in the response there may be minor peaks that only then become visible and require correction.

    Regards,
     
  17. Nuno Santos

    Nuno Santos
    Active Member

    Joined:
    Jul 25, 2000
    Messages:
    148
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Portugal
    Ratings:
    +3
    In some experiments I did, after defining the filters with TMREQ, dowloading them to the unit and re-measure the speaker's corrected output, I've found that the ETF result was quite diferent compared to the corrected curve that the Wizard was showing. Is this normal, or am I doing something wrong?

    NS
     
  18. Nuno Santos

    Nuno Santos
    Active Member

    Joined:
    Jul 25, 2000
    Messages:
    148
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Portugal
    Ratings:
    +3
    A few more questions, please… :) :

    When exporting the measured data from ETF, TAG recommended exporting the results for t=160ms. After correcting this curve within the TMREQ wizard and loading the filter settings to the unit, I can see the effect of the filters by measuring the room response again with ETF. For comparison’s sake, shouldn’t I compare again the results for t=160ms? If so how can I do that?

    Thank you for your help,

    NS
     
  19. JohnMulcahy

    JohnMulcahy
    Member

    Joined:
    Jun 13, 2004
    Messages:
    181
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Aldershot
    Ratings:
    +20
    You can either make an overlay in ETF to compare the sets of measurements or load the two sets of exported data into different speakers in the Wizard (e.g. the before into Left and the after into Right) and compare them that way.
     

Share This Page

Loading...