• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

GR Research Speaker Upgrade Review (Sierra-2EX V2)

Rate this speaker "upgrade:"

  • 1. Poor (headless panther)

    Votes: 348 96.1%
  • 2. Not terrible (postman panther)

    Votes: 7 1.9%
  • 3. Fine (happy panther)

    Votes: 3 0.8%
  • 4. Great (golfing panther)

    Votes: 4 1.1%

  • Total voters
    362
I think Amir meant "Smoothed" instead of "Gated" in that comment about resolution, because "Gated measurement" in itself does not mean low resolution and in fact is also how the HF Klippel data is obtained on a faster run.
Not correct. Gating by definition, reduces the number of samples in time domain. Heisenberg's uncertainty principle kicks in here, mandating that reduced number of samples results in lower resolution in frequency domain. That is why his graph starts at 200 Hz as his software software "knows" this. But resolution above 200 Hz is also compromised which some people don't realize and is not indicated on the graph.

As to Klippel, it does indeed rely on gating for high frequency region. But then uses the dual scan phase separation to filter out the reflections below that limit, fully preserving the resolution from start to finish. Danny could do the same by performing ground plane measurements and stitching the two together but he is not doing that.

For an example of someone who knows this, note this remark in many of James Larson's measurements:

"The Philharmonic BMR HT Tower speakers were measured in free-air at a height of 4 feet at a 2-meter distance from the microphone, with the microphone raised to an 8’ elevation that was level with and aimed at the tweeter center. The measurements were gated at 8 milliseconds. In this time window, some resolution is lost below 400 Hz and accuracy is completely lost below 200 Hz. "
 
Not correct. Gating by definition, reduces the number of samples in time domain. Heisenberg's uncertainty principle kicks in here, mandating that reduced number of samples results in lower resolution in frequency domain. That is why his graph starts at 200 Hz as his software software "knows" this. But resolution above 200 Hz is also compromised which some people don't realize and is not indicated on the graph.

As to Klippel, it does indeed rely on gating for high frequency region. But then uses the dual scan phase separation to filter out the reflections below that limit, fully preserving the resolution from start to finish. Danny could do the same by performing ground plane measurements and stitching the two together but he is not doing that.

For an example of someone who knows this, note this remark in many of James Larson's measurements:

"The Philharmonic BMR HT Tower speakers were measured in free-air at a height of 4 feet at a 2-meter distance from the microphone, with the microphone raised to an 8’ elevation that was level with and aimed at the tweeter center. The measurements were gated at 8 milliseconds. In this time window, some resolution is lost below 400 Hz and accuracy is completely lost below 200 Hz. "

Are you planning on making a video for YouTube demonstrating this GR modification? The responses in the comment section would likely be, as they say, priceless ...
 
Well, let's look at what Danny thinks he has done vs. what he has actually done.

1777773653698.png
1777773566411.png


Danny's measurement: his modification boosts the low midrange between 200Hz - 500Hz by about 2dB. Also restores the high treble >10kHz to flatness. This comes at the expense of noticeably more lumpy treble frequencies, notable a dip at 3kHz, which we can see in both measurements.

Amir's measurement: very slight low midrange boost in the same region I highlighted, about 0.5-1dB at most. Slight peak in high treble > 15kHz.

After looking at this, we observe that the 3kHz dip and the treble boost >15kHz can be seen in both measurements. This can potentially be blamed on faulty microphone / improper calibration / improper measurement technique.

But what is curious is why Danny measures a low midrange boost between 200-500Hz, and Amir doesn't. The difference between Danny's "before and after" can not be explained away with an improper cal file or faulty mic - if he used the same mic, the same fault would be observed. What exactly is in that bag of mods that he sent you, and does he state what the intention of those mods are for? Is there anything in that bag that might produce a shelf?
 
Well, let's look at what Danny thinks he has done vs. what he has actually done.

View attachment 529525View attachment 529524

Danny's measurement: his modification boosts the low midrange between 200Hz - 500Hz by about 2dB. Also restores the high treble >10kHz to flatness. This comes at the expense of noticeably more lumpy treble frequencies, notable a dip at 3kHz, which we can see in both measurements.

Amir's measurement: very slight low midrange boost in the same region I highlighted, about 0.5-1dB at most. Slight peak in high treble > 15kHz.

After looking at this, we observe that the 3kHz dip and the treble boost >15kHz can be seen in both measurements. This can potentially be blamed on faulty microphone / improper calibration / improper measurement technique.

But what is curious is why Danny measures a low midrange boost between 200-500Hz, and Amir doesn't. The difference between Danny's "before and after" can not be explained away with an improper cal file or faulty mic - if he used the same mic, the same fault would be observed. What exactly is in that bag of mods that he sent you, and does he state what the intention of those mods are for? Is there anything in that bag that might produce a shelf?
I guess reproducibility could also be an issue here. There are a lot of variables with a spread in component values. If he 'fixes' one sample, its questionable it will work the same in other instances of the same mod.
 
Well, let's look at what Danny thinks he has done vs. what he has actually done.

View attachment 529525View attachment 529524

Danny's measurement: his modification boosts the low midrange between 200Hz - 500Hz by about 2dB. Also restores the high treble >10kHz to flatness. This comes at the expense of noticeably more lumpy treble frequencies, notable a dip at 3kHz, which we can see in both measurements.

Amir's measurement: very slight low midrange boost in the same region I highlighted, about 0.5-1dB at most. Slight peak in high treble > 15kHz.

After looking at this, we observe that the 3kHz dip and the treble boost >15kHz can be seen in both measurements. This can potentially be blamed on faulty microphone / improper calibration / improper measurement technique.

But what is curious is why Danny measures a low midrange boost between 200-500Hz, and Amir doesn't. The difference between Danny's "before and after" can not be explained away with an improper cal file or faulty mic - if he used the same mic, the same fault would be observed. What exactly is in that bag of mods that he sent you, and does he state what the intention of those mods are for? Is there anything in that bag that might produce a shelf?

Its possibly just the circuits Danny is using are reacting with the drivers impedance. One can quite easily end up with bumps in this range, if theyre not implementing a good filter. I sometimes apply a parallel notch to fix these low range bumps if no other option is available.
 
Gating by definition, reduces the number of samples in time domain. Heisenberg's uncertainty principle kicks in here

Gating and smoothing of measurements are the "Shallow Hal" effect for measurements far outweighing the "Beer Googles" effect.
 
I've been telling the guy that his test gear is grossly inadequate for years -- he uses some elderly Clio system from Italy which is to a Klippel NFS what a rusty Fiat 500 is to a Mercedes-Benz S-Class.
The Clio he uses is entirely capable of decent measurement results. It's everything else that is the problem.


But what is curious is why Danny measures a low midrange boost between 200-500Hz, and Amir doesn't. The difference between Danny's "before and after" can not be explained away with an improper cal file or faulty mic - if he used the same mic, the same fault would be observed. What exactly is in that bag of mods that he sent you, and does he state what the intention of those mods are for? Is there anything in that bag that might produce a shelf?
That's an excellent question.
 
Oh my, Danny is getting ripped here. Seems like he stuck his nose into something that did not need fixing. I have the Ascend Acoustics Sierra-2EX V2 and they are excellent, I suspect I will never part with them as well as my Philharmonic BMR's. Those two speakers have one thing in common - the RAAL. To my ears it produces wonderful music, with its limits of having narrow vertical dispersion.
 
What exactly is in that bag of mods that he sent you, and does he state what the intention of those mods are for? Is there anything in that bag that might produce a shelf?
Of course he didn't send me anything. Someone bought it from him, or was the original person who sent him the speaker, and Dave from Ascend built the crossover and installed it.
 
I could not help but notice the absence of this disclaimer from the text of this review.

No loudspeaker was harmed in the creation of this review.

When the dust eventually (sorta) settles, I hope the rescue of a large whale in the North Sea inspires an ASR member to remove this modification and restore this Ascend Sierra-2EX V2 'sample' with OEM parts. ;)
 
Back
Top Bottom