Search results

I used to use a TEAC UD-501 which has three filter options: Slow, Sharp and Off. I tried to hear a difference between the modes and couldn't, even comparing with the "Off" filter. It was somewhat frustrating that my golden ears weren't telling me anything. On the other hand, I haven't given...

It’s an interesting idea, seems to be showing the controlled directivity down to lower frequencies. Not sure if this is imaging, but it is a nest measurement.

Please just return it. Nobody should spend so much money on a DAC and headphone amp unless they just want to stimulate the economy in these trying times.

In sighted listening sessions I've always enjoyed what I heard. I suspect they tend to be setup well in good rooms, and they definitely look impressive. Without knowing how state of the art speakers work, they look as though their form follows function (though those in the know realize it does...

Not only is there nothing wrong with “sounds good to me”, the only alternative is “sounds good to others.” There is no theoretical equivalent of a straight to the brain as audio is not recorded as a sound field, it is recorded as individual channels.

The most important part of a system is the room. So it’s also expected that there are many lousy rooms with well-measuring components that still don’t sound very good. The same is true of the converse. Most electronics that are considered hifi will sound good in a good room.

Not sure, but I suspect not.

Maybe you weren’t measuring the right things

A USB interface with analog output such as MOTU M2 is the most practical advice. There's no evidence I'm aware of that the ADC stage in equipment like KH 750 is anything but audibly transparent, so you can save some money and simplify using an analog connection. Going up in price from there is...

It does not. Installing the Motu driver is required, and it is not compatible with Windows on ARM processors at this time.

Alternatively it is doing what it is supposed to do, which is retrieving information and synthesis a reply without judgment. The judgment is still to make in your own hands. I would prefer this over a model that censors information and judges for me.

Happy to help:

I find it's very important to check the attribution of information. Here's an excerpt from Copilot as an example. The initial reply does not include any sources, but if you ask as a follow-up they are provided. I would in general be very skeptical of text-only replies without a source as these...

There's no such thing as a "neutral" AI model. No matter what there is an editorial decision made by humans of what training data to feed the system. If you were making an audio system agent, should you train it on Audio Science Review? Audiogon? The collective works of Floyd Toole? Amazon.com...

For these systems there are pre-filters, post-filters, and also RAG. Pre-filters halt queries that are known to produce an undesired response, related to copyright content, or things along those lines. Post-filters act upon the generated response. For example they operate content safety...

ChatGPT isn’t one model, it’s an agglomeration of many models, which also exist in different versions. So a query asked of ChatGPT will give a different reply depending on which version is being used. Between the ChatGPT service and Microsoft’s, they are likely also tuning for more or less...

I’m much more interested in blind speaker testing, but I know that’s a lot harder to pull off.

Maybe you could use GRADE to do the measurement if you already have GLM. https://www.genelec.com/glm-grade

Probably a few ways to look at this. By experience, high frequency causes less excursion than low frequency. The force and therefore the acceleration of a driver is proportional to current. For a similar peak voltage (and current) polarity reverses sooner with higher frequency, therefore...

Cynically, because people are very willing to pay for it Practically, because it can be achieved with good engineering at a nominal cost Philosophically, there is no reason