spinorama.org

[tifune]

that's a bug, now fixed. The closest is the JBL M2 (and very very close).

Thanks! It is mostly fixed, although the last 2 comparisons arent loading. Only the first 8 most similar (which is probably enough)

 

[amarsicola]

...

In the stat page, you now have a new value chart which should be what you want. Filters will come soon in the custom tab.
Custom should allow people to plot whatever they want over time. Currently it is a bit buggy but can improve quickly.
...

done (and other parameters that were missing).

Thanks

Thank you very much ! The "value" graph looks fantastic!
On the problem of price, in any case i assume that the best option is some sort of genAI Agent (gemini, chatgpt, copilot etc) to avoid the problem of unit/pair price, it could also be implemented via an agent that updates a .csv file that is somehow uploaded to spinorama at a given time interval

 

[phoenixdogfan]

Is there a way to include THD and Multitone distortion charts and possibly comparing across speakers? The comparison part may not be feasible because the two principals who have done most of these measures appear to use different formats for displaying them (and Amir doesn't do Multitone at all), but perhaps something could be done?

 

[pierre]

Is there a way to include THD and Multitone distortion charts and possibly comparing across speakers? The comparison part may not be feasible because the two principals who have done most of these measures appear to use different formats for displaying them (and Amir doesn't do Multitone at all), but perhaps something could be done?

I am trying. I am building a ML pipeline that learn to parse the image automatically and generate the data. It is not yet at the quality i want. It does well on spinorama files for now so i could index automatically Erins website for ex. If i can get it to work I plan to index all the graphs from Amir automatically.

You can also try to convince Amir to release more data but I have asked every 2 years or so and the answer has been no so far. For Erin, he send the data to me from time to time also without the other plots. Nobody wants to share the Klippel recordings which would be ideal.

I am slowly building my own scanner with 3 motors, the software is mostly working (extracting the room from the measurement), the robotic part is still not working well.

I am focusing on releasing SotF now. It is a large software and without focus it will never go out. For the curious SotF is a music player with plugins, recording capability, autoeq for spinorama and headphones, roomeq for your speakers all bundle together in a ideally easy to use way. You should be able to play with it soon.

 

[sigbergaudio]

I am slowly building my own scanner with 3 motors, the software is mostly working (extracting the room from the measurement), the robotic part is still not working well.

Are you building your own NFS? Where is the thread sharing this interesting endeavour?

 

[a4eaudio]

If you are going to load the site up with Preference Scores, target zones, distortion, etc that are not part of CTA-2034 (aka Spinorama) are you going to rename the site?

 

[Matias]

Are you building your own NFS? Where is the thread sharing this interesting endeavour?

There was this thread for the adventurous.

DIY 3D Speaker Scanner - the Mathematics and Everything Else

Let me get this out of the way first. Yes, my goal is to figure out how to build a 3D speaker scanner similar to Klippel's. Of course the first step is to figure out the theories. In my post back in August, I wrote that I knew next to nothing about the mathematics of acoustics. I also wrote...

www.audiosciencereview.com

 

[pierre]

Are you building your own NFS? Where is the thread sharing this interesting endeavour?

I have various open source projects that already takes much of my time. I have 2 private projects for now, 1 being an automatic scanner. I have not decided yet if this one will be open source or not. BOM is around 5k (large high quality motors for with low tolerances, 1 pc, some ikea furnitures, microphone+interface etc). Software is tbd.

It works differently from the Klippel.
- i don't have enough space to have the microphones run around the speaker. I choose to rotate both (the speaker and the microphone to compensate)
- i tried various models to remove the room from the measurement. i have simplified a bit by assuming the room will be a parallelepiped rectangle. I am using a GPU for speeding computations but a mac mini m4 is good enough. I have tried various things: expansion like Klippel but also more brute force with FEM/BEM and some control theory. A lot of things can be computed now on modern HW. For a simple speaker (like genelec 8330 which is the tested one) concordance is good ~1%.

I think audio society and small speaker company would benefit. I am not ready to take "real customers" yet.
I have been thinking of 2 ways forward:
- pure open source, no guaranteed support, no feature guaranty
- bounty model: when someone want a feature they pay for it. it would incentivise developers to contribute. I am not money constraint I am time bounded.

With Claude, things changed and I may be able to do on this one what I am doing on SotF: build a large software, portable and complicated all by myself with 5% of my energy.

I also have too many constraints at the moment, the key one is that i don't have the physical space to deploy and test. I need to empty the basement to do so and that means summer only (CH is very rainy most of the time).

 

[pierre]

If you are going to load the site up with Preference Scores, target zones, distortion, etc that are not part of CTA-2034 (aka Spinorama) are you going to rename the site?

Possibly. I bought the domain name for 10 years and pure inertia means that it is likely to stay the same. For a small website on a focused topic it is already picking up steam (avg around 10 req/s without the bots). SotF is also deployed on the same domain. I am open to better branding.

 

[Tom Kamphuys]

There was this thread for the adventurous.

DIY 3D Speaker Scanner - the Mathematics and Everything Else

Let me get this out of the way first. Yes, my goal is to figure out how to build a 3D speaker scanner similar to Klippel's. Of course the first step is to figure out the theories. In my post back in August, I wrote that I knew next to nothing about the mathematics of acoustics. I also wrote...

www.audiosciencereview.com

Keep an eye on that thread. There haven't been a lot of posts recently, but a lot of work has been done behind the scenes.

We're finishing stuff and will post an update soon!

 

[mglobe]

I’ve been seeing the shading on the estimated in-room response comparison hiding the curves for a little while now.

 

[mglobe]

I’ve been seeing the shading on the estimated in-room response comparison hiding the curves for a little while now.View attachment 518611

I’ve subsequently realized that it’s possible to turn the shading off by tapping on them in the legend of the chart, but I’d suggest that it would be better to leave the heavier (“Band +-1.5dB”) shading defaulted to off.

 

[Matias]

Pierre, any reason why on my phone (Android + Chrome), using the "desktop version", it shows fancier graphs and more data on Rankings; while on my desktop (Win11 + Chrome) still shows the old versions? Are these changes coming to both the regular mobile and desktop versions?

 

[pierre]

Pierre, any reason why on my phone (Android + Chrome), using the "desktop version", it shows fancier graphs and more data on Rankings; white on my desktop (Win11 + Chrome) still shows the old versions? Are these changes coming to both the regular mobile and desktop versions?

no, there only 1 website and no custom per device customisation except graph trying to pack more stuff is less space on mobile. possibly a caching effect. try a hard reload on desktop, possibly ctrl-shift-R?

 

[Matias]

Shift-F5 did the trick on desktop. Duh, sorry for the dumb question.

 

[kharan]

Hi @pierre , just wanted to say that this revamped interface looks very good and polished. Cheers.

 

[Matias]

Indeed the new interface is a huge upgrade, love using it in dark mode too.
Though I wish the trend lines on the On Axis and EIR graphs also had the db/oct notation like the main CEA2034 has.