Thank you for participating as a listener in this hydrogenaudio public listening test. Please carefully read the following guideline on how to perform the individual test sessions, even if you are an experienced listener. The general test terminology and graphical user interface should already be familiar to you. If they aren't, please obtain this important information from the test coordinator before continuing.
The recommended test procedure is divided into four steps: identification of anchors, preliminary ranking, looping, and grading. Each step is described below. You might want to print this page so you can consult it during the test. Even better: practice the procedure in advance!
Identification of anchors
Having started the session, an easy way to get a quick impression of the quality range to be expected is as follows: start the playback, then while playing switch between the stimuli and identify the low anchor. In our case, the low anchor sounds like the reference played back through a telephone. Grade the low anchor with a relatively low score. Then continue by trying to identify the mid anchor. In our case, the mid anchor sounds similar to the reference but notably washed out, blurry, warbly, or different in its stereo image. Give the mid anchor a grade in the middle of the scale based on its overall quality when compared to the reference.
Do not try to find the hidden reference with this procedure! In most tests this is too difficult, can lead to mistakes, and will be done later with a different approach.Preliminary ranking of stimuli
Leave the sound on. Now switch between the remaining stimuli (the hidden reference and the codecs under test) once every few seconds and try to find one or two which sound clearly worse than the others, i.e. with artifacts similar to those of the anchors. You should do this long enough so you can listen to every passage of the item once with each stimulus. Then give the worse-sounding one or two stimuli a grade somewhere between the mid anchor and the highest possible grade.
Do not try to find a worse-sounding stimulus if there is none! If all stimuli sound almost the same and none is clearly worse than the others, go to the next step.Looping through item
Now comes the detailed ranking and grading of the stimuli. Start at the beginning of the item and define loops which are between one and four seconds long. If you want you can align the loop points to the rhythm of the item. Listen to a looped stimulus twice, then immediately switch to a different stimulus and listen to the same loop twice as well. Then reduce the grade of the worse-sounding stimulus. Do this for every possible pair of looped stimuli. Then continue with a new loop.
Do not trust your first impression or only selected loops! Perform each looped-stimulus comparison at least twice, and make sure your loops cover the entire item.Grading of stimuli
Once you reached the end of the item with your loops, you should have assigned a grade to each stimulus which is absolute (compared to the low anchor and reference) as well as relative (compared to all other stimuli). This grade should reflect your impression of the overall quality of the entire stimulus, i.e. a summary of the qualities within the individual loops. At least one of the stimuli should have the highest possible grade. If there are two or more excellent-sounding stimuli and you can't identify the hidden reference, you can take a guess and give one of them the highest grade and the other(s) a fraction of a grade below that (here 4.9).
Do not turn up the volume if you can't find differences between stimuli! A volume increase only helps for very quiet items, otherwise it just leads to auditory fatigue.
Congratulations! You have completed the current session. You can continue directly with the next session, but don't forget to take breaks between sessions (including taking off your headphones) if you feel tired or your ears feel numb. In general, the longer it took you to complete the session, the longer you should rest your ears.