How to compare the sound of 2 different speakers ? (HS80M and HS8)

[baniza]

Hey everyone,

I`m studying audio engineering and want to compare 2 speakers (yamaha HS80M and HS8) as a research project.

One part is to compare both on the sound quality/ differences.

I`ll involve 20-30 other audio engineering students in my test. I thought about placing both speakers horizontally and on top of each other (with foam inbetween and the upper ones tilted downwards, so that both speakers emit the sound in a straight line to the ears).
All that in a mixing studio with proper acoustics and on te speakers on monitor stands fbehind the console (for comb filter optimization..)

I want to know about the difference in frequency (or general sound) and impulse response. It will be an A/B comparison.
Now my question: What signals should I use ?

Maybe a Bass and an acoustic-guitar file, both good for analyzing differences in tone and impulse response I think.
Maybe different files, that cover different frequency parts (Bass for lower Hz, Guitar for middle Hz, Hihat for upper HZ,.. or else)

Do you have any suggestions regarding general survey setup and which sound/ files to use ?

Thanks folks,

Ferdinand

 

[Adrian-MarQ]

I think a variety of recordings that your participants likely haven't heard would be useful in getting an unbiased opinion. I've been involved in a similar process (as a participant). I usually have to listen to recordings that have multiple instruments throughout a track.

Also, make sure that your participants are blind to the model that they are listening to.

 

[baniza]

Could you tell me more about the survey/ etst you were part of ?

Thanks for the reply

I think a variety of recordings that your participants likely haven't heard would be useful in getting an unbiased opinion. I've been involved in a similar process (as a participant). I usually have to listen to recordings that have multiple instruments throughout a track.

Also, make sure that your participants are blind to the model that they are listening to.

Yes, I`ll have them wear blindfold. But how should they answer the questionsheet ? Did you wear blindfolds when participating ? If so, did you just tell the person making the survey what you thought and he wrote it your answers down ?

Was it a speaker-comparison you participated in ? If so, were the speakers placed horizontally, on top of each other ? (That`s what I`m gonna do, but I`ve heard that there could be phase issues involved due to the horizontal placement. I think however that if the speakers are placed properly (speakers placed horizontally and vertically slightly tilted so that the tweeter emits the sound to straight to the ear), and the participants sit exactly in the sweetspot without moving around (that`s where the phase issues kick in I guess) then it should be fine.. ?

It would be cool if you could give me a detailed explanation of the survey itself. Also about specific questions they asked you regarding speaker comparison (if they did).

Ok, then I`ll look for tracks with different instruments (covering different frequency ranges).
Do you listen through the entire track on 1 pair of speakers and then again on the other one, or is it just short audio files (5 seks) on one speaker pair and right afterwards on the other one ?

I thought about setting up a protools session with 2 different stereo tracks and 2 different outputs (because of 2 speaker pairs) and make an arrangement, so that it switches automatically between the 2 (with a 1 second break inbetween or so), without having to switch manually (via monitor selection tool). On the other hand I think it might be easier for the participant to tell the difference if he himself can switch speakers throughout the track (via monitor selection tool, like big knob or so).. Do you have any thoughts on that ?

Thanks man, I appreciate it !

Ferdi

 

[Adrian-MarQ]

Well, I didn't do a comparison of speakers. I listened to and compared audio files that were processed in some way (ex: listened to a WAV vs. a mp3). I personally would place the speakers so that they form an equilateral triangle with the position of the listener.

As far as questions go, I would have questions like "Which audio example was louder?" and have a checklist with the answers "Example 1", "Example 2", or "Same". I'd have a similar style of questions regarding frequency and clarity. It might be a good idea to have some open ended questions like, "Describe how audio example one sounds compared to audio example two". Also, I'd play the same audio files for both speakers. I'd probably use short audio files, (5-20 seconds) just because everyone might not be able to keep a 1 minute long audio clip in their memory.

For the idea of the participant switching between monitors, I don't see that as being a bad idea. Just make sure that they are unaware of the model information of monitors.