It does indeed – thank you!
I think it’s worthy to note that the similarities @Resolve noted between OG HE6 and Tungsten were likely based on the former having Sundara pads on it. With these pads, the two do in fact measure quite similarly!
2 Likes
Yes, for me all HE-6 impressions are with Sundara pads because they are… better.
2 Likes
I would like to hear this setup tbh, HE6 with the stock velour pads is def a bit too intense around 9-11 kHz for me.
Stock pads vs. Sundara pads for anyone curious!
1 Like
I don’t think the normalization is quite right with these. Perceptually the complaint with stock pads was the honk around 1khz. Ear gain is plenty strong with the HE-6 as well, perceptually at least.
Renormalized, though I’m actually not entirely sure the prior normalization was all that problematic, even if it seemed so upon first glance.
Its worth pointing out two things though
-
The midrange being “flat” post-calibration as it is for the HE-6 with Stock pads means it is essentially DF, which I’d agree is indicating likelihood of being shouty in that 1 kHz region (even if its in the bounds). Notice how its rather far off the 10dB slope in this region, even in my first normalization.
-
This ear’s HRTF may be a bit hotter in the pinna-related part of ear gain than your own ears, especially based on those blocked canal measurements you did recently
2 Likes
Yes, I’m just saying perceptually the normalization makes more sense with 1khz being more audibly out of bounds with the stock pads. Not saying that’s how it’d sound for everyone but certainly for me it does.
1 Like
Are there any measurements for the SS version compared to the DS? I hear varied opinions on the differences between the two. Should I save 500$ and just get the SS?
1 Like
Thank you for the reply again, Resolve. From your previous posts, it sounded like you didn’t really want to get into the subject of targets and pref curves here. And I do want to respect that. But for the past week or so I’ve been trying to understand and unravel some of your and listen_r’s remarks here about different heads, rigs and ears, and HpTF variations from rig to rig. And I confess I’m still stumped re how any of this might help to shed some light on Spearthrower_Owl’s original question about the differences with the Harman target. If anything, listen_r’s graphs have only added to my confusion on this, rather than helping to clear things up.
I see some of the variations you mentioned in the two DF compensated curves he posted here, for example…
Whether these variations are due to HpTF, differences in sealing (the flat plate on the GRAS 43AG might seal differently than the KEMAR used for the GRAS DF), or other factors, I couldn’t really say. But I’ll concede that they vary… a bit (though not in a way that’s very meaningful for me).
Listen_r delineates a few of the differences in his previous post above. And I assume both curves were compensated with Ora’s new free-field derived DF curves for the 5128 and GRAS/KEMAR, respectively. In both of these cases though, your pref curve appears to noticeably overshoot the Modhouse Tungsten measurements in the upper mids and low treble. In fact, the overshoot seems to be slightly worse on the new DF compensated GRAS plot than on the 5128 plots!
I don’t have a plot of your new DF compensation curve for the GRAS/KEMAR. But based on the above results, I suspect that if you or listen_r applied it to the Harman target in the same fashion as the Modhouse Tungsten measurements above, you’d find that your pref curve also overshoots the DF compensated plot of the Harman target, in much the same way it does on some of the measurements of neutral-ish headphones.
What am I not seeing here that you two are seeing that explains this apparent discrepancy, and/or makes it ok?
So, yes we should be discussing this stuff somewhere else but for the moment this is fine, I’ll probably move it to a relevant thread eventually.
Regarding the curves, if I understand your question, it’s important to remember that A) the Harman filters are different from a static DFHRTF tilt, which is effectively the comparison you’d be considering, and B) the in-room baseline is still slightly different from the DFHRTF. It’s very close but not the same, specifically right around the ear gain, so there are bound to be some differences.
You can effectively create a similar overall signature to ‘harman’ for the 5128 using the DFHRTF and applying the filters from the research, and comparing that against raw graphs with GRAS would end up showing a more familiar result, with the differences between rigs being the same as what Listener posted above.
I’m not really sure what else there is to consider here. Maybe @listener can elaborate on the point further.
I understand most of the above, Resolve.
Try my suggestion of compensating the Harman target with your new GRAS/KEMAR DF curve. If you really want to be consistent with the Harman research, that might give you a somewhat different and possibly better idea of the slope you should be targeting (or preferring) on your diffuse field compensated graphs. Perhaps for all three rigs.
It won’t be perfect though, because the DF curve you’re using for the GRAS 43AG is apparently based on a mannikin (KEMAR), which may couple a bit differently with over-ear headphones than the 43AG’s flat plate. (There might be some ways to iron some of those differences out as well though, by comparing headphone measurements on the KEMAR and 43AG with the same or similar pinnae and ear couplers.)
Note, since the Harman target is smoothed, you may also need to smooth your new GRAS DF curve a bit before using it to compensate the Harman target, to get a satisfactory result. (I don’t have an easy way to smooth curves in EAPO’s Configuration Editor, but could try converting both to 1/3-octave curves.)
I’ve always believed that the Harman target undershoots a neutral response in the high treble, btw. And that undershoot will probably also be carried forward in the DF compensated plot of Harman (unfortunately). So this approach will be a bit of mixed bag from my perspective.
We’re not compensating with sloped diffuse field though. We’re using flat diffuse field and representing that with bounds based on the Harman filters. I imagine you’re suggesting this to indicate the differences between Harman filters applied to the DFHRTF and Harman OE 2018, but we already know the differences between the DFHRTF and the in-room baseline. So I’m not sure what this seeks to accomplish.
This has to do with response of the ear vs response of the headphones. The response of the ear should be the same, but it is true that flat plate may not couple the same, the key differences are typically that a flat plate presents a more consistent seal and that’s consequential in the bass. But I also do not see this being all that impactful for differences among results using a DFHRTF.
It does, relative to the effects of that ear in a headphone-based use condition (being worn on the head = sound coming from all/no directions). You can see this fairly clearly when comparing the DFHRTF baseline with Harman filters to Harman OE 2018, which is the in-room baseline and those same filters.
My Tungstens order finally shipped after 7 months of waiting.
Right now the wait times are about 8 months for Mimic Orders and 7 months for the Modhouse drop orders.
I just wanted to give people an accurate view of the wait time since it’s not really mentioned in reviews.
6 Likes
Fwiw, I understand you’re not using the sloped version of DF HRTF for your compensated graphs. And that’s why all of your DF compensated headphone plots have a downward slope from F0 in the bass to the treble (generally speaking). This is what I want to see on the Harman target as well. I’d like to see its general downward slope from F0 in the bass to the treble after compensation with the new (flat) higher resolution DF HRTF you’ve developed for the GRAS. And compare that with your current pref curve. Particularly the slope levels in the upper mids and low treble.
Unless you’ve already posted this somewhere else, I don’t believe this is something we’ve seen yet. And I’d do it myself, but I don’t have a plot of the new flat GRAS DF HRTF, or a way to easily apply an appropriate amount of smoothing to it for this type of test with my current EQ software.
The DF compensated plot of the Harman target will probably also have a noticeable peak where it overshoots the concha notch at around 10 kHz in the treble on the GRAS. This is different than the undershoot in the HF. And it’s something I should also have mentioned above… I think we’re all used to ignoring this 10k feature though on Harman graphs. And know that the Harman target probably should’ve had a notch there as well, for better continuity with the GRAS measurements.
Could this conversation be kindly moved to a separate topic? It is not helping with the actual headphone discussion.
4 Likes
I’m ok with it being moved.
Hey guys, new DS Tungsten owner here. Really loving these!
I need some help with a technical question regarding amplification. I have a Ferrum OOR and a Quicksilver here, the OOR has obviously sufficient power to drive the Tungstens no problem but i was surprised when i tried the Quicksilver which sounded exceptional!
Honestly, i prefer the Tungstens on the Quicksilver and it is not even close. What the Quicksilver does for the Tungsten’s soundstage is crazy, this pairing delivers the most organic and realistic sound i have ever experienced. The only thing the OOR does better imo is bass control, everything else goes to the Quicksilver for this pairing.
Ayways, my question is, given the Quicksilver specs, does my findings agree with the math that the Quicksilver is indeed sufficient to power the Tungstens?
The specs are:
Output impedance: 2.2 ohms
Maximum output: 7 volts rms into 30 ohms or higher load
I mostly have the volume pot around 12 oclock but i have read somewhere this doesn’t really mean all that much…
3 Likes
I managed to try a Tungsten at CanJam London, so this is only first impressions, but has anyone else found the imaging to be weird? I got this strange three blob effect that was uncomfortable enough that I had to stop. One friend heard it too but the other friend didn’t
1 Like
Hi @Torq , i was reading some of your replies to amplifier related questions, maybe you can shed some light into this question if you don’t mind? It has been a few days but still no replies to help me understand this thing.
Cheers
The Quicksilver’s maximum output of 7v is enough to get the Tungstens to between 100 and 103 dB/SPL, depending on which version. That’d yield an 80 dB/SPL average listening level, allowing 20 dB for musical peaks.
That assumes, of course, that the unit’s 24 dB of gain is enough for your source to drive the Quicksilver to its limit.
2 Likes