Topic: Hypercard Comp Blowout (Gefell m210, JWmod AKG 463, Beyer mc950, Charter Oak)

[Todd R]

Well, I've been collecting and trading hypercard mics like baseball cards, and along the way did some comps.  Unfortunately, I don't think I got any comps done with the AKG 393's or MBHO 603 hypers when I had them.  But I've done a bunch of comps of some of my hypers, all set up against my Gefell m210 hypers.

In the list are:
Gefell m210
Beyerdynamic mc950
Jim Williams modded AKG 460B with ck63 caps
Charter Oak m900 with hypercard caps

Basics on the comps:
I've selected portions of 1 to 3 songs from various shows I've done some comps of.  In each case, I have two sources head-to-head (never had enough motivation or cables to do 3 at once).  The song snippets have no processing other than some post gain.  For each track, each pair of sources had gain added such that their average RMS power levels for the whole source were equal to each other.  (I chose to have the average sound levels be equal for each, not to have each source boosted to 0dbFS).

Mics were set up on the same stand, set up with the same pattern (DIN), within no more than an inch or two between each other.  For all sources, the mics were all on the same t-bar, with each mic pair interleaved with one another, and thus were very close to each other (care was taken to not block any mic with another and to leave some amount of space to not cause significant reflections). Windscreens were not used (all indoor shows).

For the comps, I've assigned randomly what source is listed as source 1 and which was source 2 (using a coin flip).  This was done per track, not per set of music.  So for instance, for a set of 3 Yonder songs, source 1 may sometimes be Gefells and sometimes be AKG's.  All sources were originally recorded in 24 bits, some at 48k, some at 44.1k.  I've dithered all sources down and saved as 16 bit flacs.  But I haven't resampled, so if you listen by burning to CDR, you'll need to re-sample the 48k files to 44.1k.

Comparisons from the YMSB show and G Love set are pretty good comps as all mics were just straight into a Tascam DR680.  The Phish comp used Gefell m210> EAA PSP2 > 680 and jwmod AKG 463s> Naiant littlebox (output transformers engaged) > 680.  The WSP comp used Beyer mc950 > Lunatec V3 > 680 and Gefell m210 > Naiant littlebox (transformers disengaged) > 680.  So these comps aren't the best.  Still though, I've done head-to-head comps of the littlebox to the V3 and littlebox to the PSP2, and in both cases the results between the littlebox and the PSP2/V3 were pretty close, so hopefully these hyper comps at least give some sense of the differences between the mics.

The comps are:

Gefell m210 vs JWilliams mod 463 -- YMSB show and Phish show
Gefell m210 vs Beyerdynamic mc950 -- G Love set and WSP show
Gefell m210 vs Charter Oak m900(hc) -- YMSB show

(BTW, the Charter Oak mics look almost exactly like the Peluso CEMC6's other than color.  Supposedly use the same capsule, and have upgraded electronics like the Pelusos.  They fared very well in a large -- 40 mic?? -- shootout reported in Tape Op, they were found in blind tests to be one of the top selected mics.  Along with Milab vm44's  )

Ok, so that's the long spiel.  I've posted all the head-to-head comps on megaupload.  I'll put all that info and the megaupload links in another message....

[page]

yay!!
will listen later.

[Todd R]

Here are the comps, with the megaupload links:

Source Comparisons:

Gefell m210 - JWilliams modded AKG 463 comp:

ymsb2010-08-26:
Gefell m210 > Tascam 680
JWilliams modded AKG 463 > Tascam 680

http://www.megaupload.com/?d=LJ80ZO8N

Track 1 (Bolton Stretch)
Track 2 (Ten)
Track 3 (2 Hits)


ph2010-10-10
Gefell m210 > EAA PSP2 > 680
JWilliams modded AKG 463 > Naiant littlebox (tranfo) > 680

http://www.megaupload.com/?d=VFUAT6EU

Track 1 ( Loving Cup)


Gefell m210 - Beyer mc950 comp:

G Love 2010-12-30:
Gefell m210 > Tascam 680
Beyer mc950 > Tascam 680

http://www.megaupload.com/?d=V93NDDUB

Track 1 (The Road…)
Track 2 (50 Ways)

WSP 2010-12-30:
Gefell m210 > Naiant littlebox (no tranfo) > Tascam 680
Beyer mc950 > Lunatec V3 (dig) > Tascam 680

http://www.megaupload.com/?d=LFDIO9TM

Track 1 (Goin' Out West)
Track 2 (Surprise Valley)


Gefell m210 - Charter Oak m900(hc) comp:

ymsb2010-08-26:
Gefell m210 > Tascam 680
Charter Oak m900(hc) > Tascam 680

http://www.megaupload.com/?d=1HNIS8TC

Track 1 (Steep Grades)
Track 2 (Spanish Harlem Incident)
Track 3 (Kentucky Mandolin)

[page]

Gefell m210 - Beyer mc950 comp:

G Love 2010-12-30:
Gefell m210 > Tascam 680
Beyer mc950 > Tascam 680

I didn't even have to listen, I'm was rather sure I knew which was which by looking at the wave forms. I don't know if it's all of the 950s, but there is at least one or two sets that the mics are opposite polarity from each other. I remember first seeing it on the set that acidjack had.

edit to add:
Once I inverted said channel, the stereo image cleared up and it was interesting to compare the two. I ended up having to add about 1db to the right channel of both, may just be equipment/environment on my end. Loading both up in audacity, since they were clocked together allowed me to set the first pair on mute and listen to the second pair, then hit solo on the first pair and instantly switch without pausing.

1 question: Were the gefells on top or bottom that night? Not sure it would have made a big difference, just curious.

Neat, comp, thanks.

[it-goes-to-eleven]

Thanks for the effort, and sharing, Todd!

[Chuck]

Thanks Todd.

[newplanet7]

Thanks Todd.

[Todd R]

1 question: Were the gefells on top or bottom that night? Not sure it would have made a big difference, just curious.

Neat, comp, thanks.

Damn, bummer on that polarity problem -- I didn't know about that.  The mc950 pair I have were formerly acidjack's, so it looks like I need to get a repair done.

Neither pair of mics were on top or bottom -- all mics were interleaved on the same t-bar.  So left gefell, left beyer, right gefell, right beyer.  1" risers were used since otherwise the backs of the mics would hit each other in the same plane.  So all mics were within about 1" of each other in the horizontal plane, with each pair having one mic up compared to its mate.

[Todd R]

I ended up having to add about 1db to the right channel of both, may just be equipment/environment on my end.

One more note on your comments:

As I mentioned, I matched each source against the other so that they had equal average RMS power levels, as I think that is how comps should be done (others may prefer comps be done differently).  And to try to again keep things fair, each channel of a source were normalized so that each channel had the same average RMS power levels. Sources were not normalized to 0dbFS (or -0.1 dbFS). 

So sources may (and do) top out at -1 or -2db, either in one channel or in both.  From my standpoint, we hear based on average RMS, not peak value.  If sources were normalized to 0dbFS, then the source that is more "dynamic" (word choice?) and had sharper peaks, once normalized to 0dbFS would actually have a lower average RMS power level compared to a less dynamic source.  Hence, normalized to 0dbFS, the less dynamic source would have higher average RMS power, and people would probably tend to prefer it better (much the reason why we have "loudness wars").

Bottom line, if people want to change the gain on these comp sources, they clearly can.  But I think the way they are normalized is more "fair", so the sources will not have equal peak values (either comparing the left channel to the right channel or comparing one source to the other).  All channels and all sources though should at this point have equal average RMS power levels.

[page]

I ended up having to add about 1db to the right channel of both, may just be equipment/environment on my end.

One more note on your comments:

As I mentioned, I matched each source against the other so that they had equal average RMS power levels, as I think that is how comps should be done (others may prefer comps be done differently).  And to try to again keep things fair, each channel of a source were normalized so that each channel had the same average RMS power levels. Sources were not normalized to 0dbFS (or -0.1 dbFS). 

So sources may (and do) top out at -1 or -2db, either in one channel or in both.  From my standpoint, we hear based on average RMS, not peak value.  If sources were normalized to 0dbFS, then the source that is more "dynamic" (word choice?) and had sharper peaks, once normalized to 0dbFS would actually have a lower average RMS power level compared to a less dynamic source.  Hence, normalized to 0dbFS, the less dynamic source would have higher average RMS power, and people would probably tend to prefer it better (much the reason why we have "loudness wars").

Bottom line, if people want to change the gain on these comp sources, they clearly can.  But I think the way they are normalized is more "fair", so the sources will not have equal peak values (either comparing the left channel to the right channel or comparing one source to the other).  All channels and all sources though should at this point have equal average RMS power levels.

oh yeah, no doubt. I definately agree with your approach on using the RMS values and matching those. Its quite possible that my hearing, or my equipment will cause me to want to push or pull a channel somewhere. I tinkered with them after the first listen (and preference selection) only as an attempt to balance the stereo image while listening for further differences between the two (e.g. AKGs have a sizzle, gefells have a certain color, etc).

Thanks!

[johnmuge]

Thanks for the comp Todd.  Nice job !!

[Gutbucket]

Looking forward to listening. Thanks for putting the effort into this Todd.

[Gutbucket]

I've always been partial to the Gefell sound.  In that light, I simply accepted my own bias and assumed I would prefer the Gef's to the other mics, so for me one challenge here was to pick which samples were the Gef's and I feel I nailed that relatively easily on headphones.  I recorded my guesses, but will withhold posting them so I don't influence other's listening. (in that light, what follows may tangentially influence others as well so if you prefer to listen uninfluenced, feel free to stop reading at this point..)

I will say that there are of course various sonic differences between samples, but what I listened for primarily is what I hear as the tell-tale Gefell traits of clarity (apparent both in the direct sound and also especially the ambient off-axis sound, discernable separately) and a certain sense of 3-dimentional depth.  The depth aspect seems to be the most difficult for the other manufacturer's to match- some samples that sound good or similar in isolation sound rather dimensionally flat in comparison, with all sounds seeming to emanate from about the same depth in space.  I also find that what I believe to be the Gefell samples are generally more musically engaging and exciting to listen to over time, but I realize that is an even more subjective measure- it's real and a useful differentiator for me, but hard to put a finger on or describe meaningfully.

Beyond the simple sonic ID'ing challenge, this was an interesting exercise for me in exploring quality judgments surrounding why I prefer one sound over another and what I personally listen for.  Thanks for the mental exercise.  It helps to keep things in perspective.

I'm somewhat surprised by the Charter Oaks.  Especially on T2 (if I remember right) where I went back and forth a few times.  The samples differed primarily in the 3-d and long term musicality aspects for me and less in clarity, assuming I ID' them correctly of course. 

[OOK]

I was only really interested in the Charter oak mics....I really can't tell much between the gefells and the Charter Oaks all that much, there certainly is not a 1000$ difference....The charter Oaks are very surprising.....  Considering when you spend 1k for a pair with cards, omni's and hypers.  This is only beat by the great Bussman package....Nice Comp


OOK

[Jimmy NOMAD]

Any chance someone can re upload these samples somewhere?  I'd love to give em a listen!