[hiflofi]

I wouldn't be surprised about unit variation. I don't really see the controversy. As long as it sounds good to your ears, it's not really too big of a deal right? Or if it isn't your cup of tea, that's OK too.

Is unit variation the crux of this debate? Or is it how good/bad the Solaris is?

 

[toughnut]

The problem here isn't the 4khz gulf on one sample, the problem is the inconsistency between samples. If you want to compare samples, you obviously want to use the same methods and hardware. You're isolating the variables to a single variable: the sample itself.

Exactly. Even if the method or instrument is flawed (assuming) but if the results for each sample is consistent and repeatable, that data still hold valid as comparison between different samples.

Anyway, question now is, I have chance to purchase Solaris with some discount, should I take the jump???? Someone push me off the ledge

 

[Penguino]

If the ported design of the mid range BA was susceptible to having variations of up to 15dbs, not just on 1 spot but several, it would be impossible to match both channels. The variances are matched on each set.
Not trying to dodge your point.
The graphs are telling a story that have more than a few holes.
And you know how I am with graphs anyways.
If this variance is real, more sets will show up eventually. Till then, not much to go on honestly.
Put out the torches. Grab a beer.
Enjoy the music.

Who said anything about torches? I feel like you guys are honestly playing the victim hard here really hard when I don't see much of anything directly offensive towards you guys and it.
As for the channel matching I cannot help but agree. The dynamics involved would be incredibly random but again I am not sure because they may have got a process that worked enough? Its impossible to say and I look forward to seeing the response from @KB if he does make one.

That’s not what I am saying nor implying.

I am just saying measurements are tricky and when something that big comes up this warrants a double check don’t you think?
If I was crinacle that would be my first reflex, get this checked because 15dB seems like a HUGE anomaly anyway that could be heard right away as defective...



I am not in the business not sure a pro would take the time to train me if I myself had the time nor the inclination to learn this or invest in a pro equipment I couldn’t possibly justify paying for...

Measuring iems is not that tricky once you get a rig going. Awhile ago he used this : https://www.amazon.com/Dayton-Audio-iMM-6-Calibrated-Measurement/dp/B00ADR2B84
He got consistent measurements that were plenty worthwhile with it.

As for making sure I'm almost certain someone with so much rep would triple if not quadruple check before posting something so bomb shell.

 

[ddmt]

If the ported design of the mid range BA was susceptible to having variations of up to 15dbs, not just on 1 spot but several, it would be impossible to match both channels. The variances are matched on each set.
Not trying to dodge your point.
The graphs are telling a story that have more than a few holes.
And you know how I am with graphs anyways.
If this variance is real, more sets will show up eventually. Till then, not much to go on honestly.
Put out the torches. Grab a beer.
Enjoy the music.

Do you own or even ever try Solaris or any other CA product? Just asking, you keep dodging answering this question.

Myself, I own Andro S, haven’t had a chance to try Solaris yet, the closest dealer is more than 1000 km from me. Bought Andro S because the high praised from this site, and boy it’s right.

 

[marcussmj]

Exactly. Even if the method or instrument is flawed (assuming) but if the results for each sample is consistent and repeatable, that data still hold valid as comparison between different samples.

Anyway, question now is, I have chance to purchase Solaris with some discount, should I take the jump???? Someone push me off the ledge

You going in blind might be a gamble.

 

[davidmolliere]

If you want to compare samples, you obviously want to use the same methods and hardware. You're isolating the variables to a single variable: the sample itself. Toyota is a reputable brand known for reliable cars yet they had breaking software problems that resulted in major accidents and deaths. Just because someone is reputable doesn't mean they don't make mistakes.

Yes of course you can compare samples with the same gear but you can also compare results between different rigs and here comes the big issue : you admit you can get different results with different rigs (and that does not mean it’s massaging...), which kind of indicates that it’s hard to have an absolutely objective measurement. Who can say that rig X is better than rig Y, except maybe by getting some kind of information on which measurement gear is more accurate.

And reputable brands do make mistakes, but they admit their mistake is the important aspect.

And with that I’ll just stop feeding this debate that is not about the Solaris at all in the end...

 

[Kitechaser]

What Crinacle has measured isn't a huge anomaly. Explain what the anomaly is. A lot of headphones and IEMs have specific peaks and dips for specific tuning purposes. You're suggesting him to use a different rig for what purpose? The only reason someone would want to use a different rig is if they wanted to massage the numbers, like if they want to hide the 6khz peak on the Sennheiser HD800 or Sony EX1000 to make them look better by using a rig

The problem here isn't the 4khz gulf on one sample, the problem is the inconsistency between samples. If you want to compare samples, you obviously want to use the same methods and hardware. You're isolating the variables to a single variable: the sample itself.



Toyota is a reputable brand known for reliable cars yet they had breaking software problems that resulted in major accidents and deaths. Just because someone is reputable doesn't mean they don't make mistakes.

Let me answer this for the last time, lest we go on and on in circles.
The story the measurements are telling is this:
A) campfire audio is knowingly matching, pre assembly, drivers with wild variances of up to 15dbs in some spots. (All the measured sets do not have any imbalances in the L/R Channel) now why they would do that? Only campfire and God knows.

B) the other possibility is that the measurements are taken from different rigs, or ones that are faulty, or have inconsistencies in their methodologies, or manipulated...or whatever ever ever, the list goes on.

This can't be settled by us Westerners sitting in our respective continents.
So how about we wait till more measurements show up. People other than those in Singapore, also have measurement rigs.
Now that this is such a huge firestorm, I am sure people will be rushing to see what the real story is.

 

[toranku]

Let me answer this for the last time, lest we go on and on in circles.
The story the measurements are telling is this:
A) campfire audio is knowingly matching, pre assembly, drivers with wild variances of up to 15dbs in some spots. (All the measured sets do not have any imbalances in the L/R Channel) now why they would do that? Only campfire and God knows.

B) the other possibility is that the measurements are taken from different rigs, or ones that are faulty, or have inconsistencies in their methodologies, or manipulated...or whatever ever ever, the list goes on.

This can't be settled by us Westerners sitting in our respective continents.
So how about we wait till more measurements show up. People other than those in Singapore, also have measurement rigs.
Now that this is such a huge firestorm, I am sure people will be rushing to see what the real story is.

Let's just wait on a response from KB.

 

[marcussmj]

Yes of course you can compare samples with the same gear but you can also compare results between different rigs and here comes the big issue : you admit you can get different results with different rigs (and that does not mean it’s massaging...), which kind of indicates that it’s hard to have an absolutely objective measurement. Who can say that rig X is better than rig Y, except maybe by getting some kind of information on which measurement gear is more accurate.

And reputable brands do make mistakes, but they admit their mistake is the important aspect.

And with that I’ll just stop feeding this debate that is not about the Solaris at all in the end...

You don't get the point, it's not about accuracy it's about consistency. You can argue about rigs but that's irrelevant here.

 

[toughnut]

You going in blind might be a gamble.

No mate, I tried Solaris 3 times. Issue is the fitting. Love the fun sound and high stage.

 

[jeffreyw311]

Yes of course you can compare samples with the same gear but you can also compare results between different rigs and here comes the big issue : you admit you can get different results with different rigs (and that does not mean it’s massaging...), which kind of indicates that it’s hard to have an absolutely objective measurement. Who can say that rig X is better than rig Y, except maybe by getting some kind of information on which measurement gear is more accurate.

Dude, are you not following the thread.

The problem isn't the measurement itself. The problem are the audible differences between samples. You want to use the exact same rig and methods, no matter how theoretically incorrect the frequency response graph produced is. You want to isolate variables to a single variable. In this case, the single variable are the Solaris IEM samples themselves. We know Crinacle is, if anything, consistent with his results so I'm happy to scratch that potential problem off.

If you used different measurement rigs to measure the different Solaris samples, it would not answer the question: are different Solaris actually audibly different or not. The conclusion in that situation would be: are the different measurement rigs at fault or are the Solaris samples actually different?

 

[hiflofi]

Lol why are people questioning the neutrality of @crinacle's approach? Or the consistency? Now, of all times? The guy has measured 100s of IEMs with zero complaints and now he gets called into question?

He loves the Andro and Andro S, which are both Campfire products. In my opinion, he doesn't have an agenda.

What is so surprising about unit variation? That's a known phenomenon. And I'm not seeing channel matching issues in the Solaris measurements anyway lol...

 

[FangJoker]

Out of 11 IEMs on crin's list, I agree with 9 out of 11 for sound signature description and 8 out of 11 on grading order. I trust his opinion for the most part, but there's guys that I PM that have similar source equipment as me and buy similar IEMs and full sized headphones that I get their opinions before buying anything.

If you guys disagree with his list that's fine. I'd like to see some of you make your own threads grading what you heave owned and demoed and to list your source gear and what music was used. I find those types of threads to be helpful when deciding between different things to buy. I used to be the early adopter type of buyer, but I got burned too many times so I now wait for multiple reviews to make a decision.

There are 2 reasons why I will likely pass on the Solaris. The first is that it's too sensitive and I just got the Hugo tt2 for Christmas so I'd expect it to hiss a lot on it the second is that several people have said that speed is not a strong suit. I guess a third reason could be that some have said that it doesn't play well with Asian female singers. Unless I can audition these somewhere or rent them (I wish someone would rent out totl gear to help make buying decisions easier) I am not going to gamble on these playing well with my source.

 

[marcussmj]

No mate, I tried Solaris 3 times. Issue is the fitting. Love the fun sound and high stage.

The specific unit you tried? Maybe you should just pick up that pair at a good price then hahaha.

 

[marcussmj]

Dude, are you not following the thread.

The problem isn't the measurement itself. The problem are the audible differences between samples. You want to use the exact same rig and methods, no matter how theoretically incorrect the frequency response graph produced is. You want to isolate variables to a single variable. In this case, the single variable are the Solaris IEM samples themselves. We know Crinacle is, if anything, consistent with his results so I'm happy to scratch that potential problem off.

If you used different measurement rigs to measure the different Solaris samples, it would not answer the question: are different Solaris actually audibly different or not. The conclusion in that situation would be: are the different measurement rigs at fault or are the Solaris samples actually different?

That's what we have been saying, it's not about accuracy it's consistency but this falls on deaf ears apparently.