How sure are you that the abundance of EQ filters does not change sound?
Apologies for getting a little technical (and I do not want to flex my knowledge), but do you know about the order of digital filters used in Roon or any other software? How are the collection of filters connected? Is it cascade? Is it parallel? All of these design choices determine how far you can go in terms of adding EQ filters.
I am being careful to not upset Amir here but as a general rule of thumb, you need to stick to very few filters.
For folks who want to know more about this, you can take a look at the technical article below:
If you use large and very sharp peak filters I've read that it can cause ringing, but you shouldn't really use such filters in headphone EQ's anyway due to the uncertainty of absolute precise locations of dips when you have them on your own head vs the GRAS device, so it doesn't pay to use such filters anyway regardless of the ringing variable.....so for all intents & purposes EQ'ing headphones doesn't add a degredation of sound if done properly.
About filter order, it doesn't matter which order they're listed in your software, so you can have them in any order you like.
Regarding deciding on number of filters to use, it doesn't matter as long as you're not using stupidly sharp filters (which I talked about earlier in this message). On a practical level, it's true that you might not want to correct absolutely exactly to the Target Curve when using measurements published on the internet, because there's no guarantee that your unit of headphone will have all the minute peaks & dips of the measured unit, yet instead it's more likely to have the larger distinct features - so you might make the argument that it's only logical to correct for the larger features. However, if someone like Oratory has measured many many units of a headphone, see HD600 and HD650 for example, then all the peaks & troughs he includes in his published measurement is valid, as the averaging process will only allow peaks & dips to remain that are generally consistent across most measured units.....so in those cases you could argue that it's more valid to correct exactly to the Target Curve (hence probably using more filters). Furthermore, one other case, if you've had your particular unit of headphone measured on a GRAS (eg by Oratory), ie you sent him your headphone, then I think it's quite valid to be quite aggressive and accurate with the EQ in terms of hitting the target curve, because unit to unit variation variable is removed - indeed for the headphones I've sent Oratory I've found that aggressive EQ works positively on those, although the success of such an approach might vary from person to person depending on how far their anatomy is removed from that which the GRAS device is attempting to mimic, so that's a consideration if being aggressive with EQ....but you'd test with music listening to see if your particular aggressive filters were a positive or negative impact.
Going back to our original point we were discussing, the fact that the Stealth doesn't really need to be EQ'd for a good experience doesn't mean it's automatically sonically superior to another model of headphone that has been EQ'd to a Target Curve, certainly not using the initial argument you made that "EQ degrades sound".....at the worst EQ reduces your volume output overhead due to having to use a negative preamp to cover the boosts which also theoretically lowers you SINAD too because the DAC is operating at a lesser output for most of the frequency range, but the SINAD argument I also don't think is audible and especially not if you've chosen a good DAC & amp.......and likewise a good DAC & amp will give you plenty of volume output overhead for allowing for a negative preamp.