2025: Does somebody needs Hypex etc...?

[Acapella]

Hello together,

I know the title sounds a bit cynical, but I'm wondering this myself.
If there is already a similar thread, please move mine there.

There are now some very good Class-D amplifiers based on the TI TPA3255 chip.
The best (according to Amir's measurements) currently seems to be the 3E Audio A7 (mono or stereo).

Suitable for 2-16 ohms, about 250 watts per channel, XLR and RCA, exchangeable OP amps, what more could you want?

I'm actually wondering now (also in view of the very reasonable price of the TPA3255 amplifiers) whether Hypex or Purifi amplifiers still make sense, provided the power in watts is sufficient for you personally.

The 3E Audio A7 is also specified up to 16 ohms, no idea whether this is a problem with Purifi or Hypex or whether it is just not specified.
Although the 3E Audio A7 does not have a 12V trigger input, this is not really a problem thanks to smart sockets (controllable via Home Assistant) or the Auto-On-Off function.

What reasons (apart from higher wattages) will there still be in 2025 to buy a Hypex or Purifi amplifier instead of a 3E Audio A7 (or a comparable TI TPA3255 amplifier)?
Maybe some users can write their experiences or opinions here, as I am very interested in this topic.

Thanks in advance.

 

[Mike123]

What reasons (apart from higher wattages) will there still be in 2025 to buy a Hypex or Purifi amplifier instead of a 3E Audio A7 (or a comparable TI TPA3255 amplifier)?

Better measurements. I can be calm with purifi that it is completely transparent and does not affect the music in any way.

 

[Mike_Trueman]

Yeah i love those small compact Infineon's GaN FET amplifiers.

 

[Matias]

On the high end stereo or multichannel Hypex and Purifi are very relevant. Only mid level and entry level is where they are not competitive price wise with the chip based class D amps, IMO.

 

[MCH]

Not many have been capable of ironing that load dependency, and the 3E audio aren't that cheap. Seeing the chip in real, one wonders (in the good sense) how can all that heat, and power, can get out of that tiny pad. These amps are a real wonder and very good news for the audio hobby, but I understand why many people will stick to hyper & co.

 

[mocenigo]

Better measurements. I can be calm with purifi that it is completely transparent and does not affect the music in any way.

Also, even in a home environment, we should not forget the crest of transients. With less than ideally sensitive speakers, having (undistorted) 500W or 1000W can be quite useful, even if it is for just a small fraction of a second.

 

[mocenigo]

On the high end stereo or multichannel Hypex and Purifi are very relevant. Only mid level and entry level is where they are not competitive price wise with the chip based class D amps, IMO.

ICEPower does offer competitive products for for 125-250W range.

 

[TurtlePaul]

I always this of this as "what is the price of overkill". The audio world has proven that there is a huge market for overkill in DACs and amps. Why are we still testing DACs at this point? Why do people even bother with that chart if everything in the blue or green is audibly perfect?

I think my system is pretty close to ideal on a price to performance ratio. I have Revel F206 speakers, which retailed for $2,800/pair (but now seems are being phased out, and I paid much less used). I pair them with a MiniDSP 2x4 HD ($225) and a Fosi v3 ($110). I think this is about the ideal performance matching, as the speakers still have by far the worst distortion in this setup.

However, I totally get that people with $3k ,$5k or $10k speakers don't want to have electronics which cost an order of magnitude less than their speakers. Honestly, this is the hardest thing for people to get - how can it be ideal to pair $8k/pair speakers, perhaps F228BE, with an amp that costs merely $100-300 dollars.

Regarding the frequency dependency, I have DSP and can adjust this, but honestly I can't hear 15 khz anyways. Regarding power for transients, the v3 already puts out 150+watts, I live in an apartment building and my math says that for a Hypex to even hit its power level on transients means the cops are being called.

 

[Ognian]

One good reason is that huge power brick which must be hidden / managed. Time will tell if they are reliable as well (I'm not trying to be negative, just cautious). Amps are the only piece in the chain that might pose an electrical hazard, or could destroy a nice pair of speakers if something goes wrong.

From my understanding hypex and purifi when decently implemented (like not painting your ground surface) are very safe in the long term so far.

 

[TurtlePaul]

Time will tell if they are reliable as well (I'm not trying to be negative, just cautious). Amps are the only piece in the chain that might pose an electrical hazard, or could destroy a nice pair of speakers if something goes wrong.

From my understanding hypex and purifi when decently implemented (like not painting your ground surface) are very safe in the long term so far.

This is just fearmongering. Texas Instruments knows how to implement a chip and the TPA3255 has all of the relevant protections. It has over temperature, under supply voltage, over current, over power and DC protections implemented in the chip. Therefore, if one of these amps fails, you can just replace it for cheap and it is unlikely to take the speakers with it.

 

[Ognian]

This is just fearmongering. Texas Instruments knows how to implement a chip and the TPA3255 has all of the relevant protections. It has over temperature, under supply voltage, over current, over power and DC protections implemented in the chip. Therefore, if one of these amps fails, you can just replace it for cheap and it is unlikely to take the speakers with it.

Fair enough on the chip part. What are your thoughts on the power brick? I've seen my fair share of swollen or dead laptop bricks.

Am I wrong in my understanding that hypex and purifi have strict power delivery guidelines? Is it the same case here?

 

[Ropeburn]

Regarding power for transients, the v3 already puts out 150+watts, I live in an apartment building and my math says that for a Hypex to even hit its power level on transients means the cops are being called.

150 or 250W isn't that much of a difference anyway. Generally, every once in a while you might be glad to have the power, even if you never use it otherwise.

Last new year's was when I fully cranked my NC252MP for the first time - we got drunk, and the DAC volume somehow showed 99 at a point for a minute or two. That's when I knew it was indeed a good idea to get this version instead of the 125W version. Speakers did it without complaint too and, believe me, killed this appartment building. Haha! All without obvious clipping or distortion, as far as a bunch of drunkards can tell. There must've been plenty compression. Definitely a good idea to reserve this for the three loudest hours of the whole year in the interest of good neighbourhood, but I doubt we'd have had as much fun with a weaker amp. Probably would've run into problems or something breaking at a point. But no!

Better have reserves and almost never need them, than not having them and needing them occasionally.

 

[TurtlePaul]

Fair enough on the chip part. What are your thoughts on the power brick? I've seen my fair share of swollen or dead laptop bricks.

Am I wrong in my understanding that hypex and purifi have strict power delivery guidelines? Is it the same case here?

The power bricks with the affordable amps work but yes they are cheaper.

There are threads around here regarding dead hypex amps because Hypex uses cheap power supply caps.

 

[Yevhen]

What reasons (apart from higher wattages) will there still be in 2025 to buy a Hypex or Purifi amplifier instead of a 3E Audio A7 (or a comparable TI TPA3255 amplifier)?

I would check the THD at the frequencies >5kHz. The class D chips with the local feedback, like TPA, usually suffer from the high THD in the post-feedback inductor.

I'm sure that with the relatively good speaker, you'll hear the difference between those amps. To me, the cymbals sound crisp enough only with the THD like -100dB or higher.

Purify (Bruno) has some nice articles about the global/local feedback loops.

 

[DVDdoug]

I would check the THD at the frequencies >5kHz. The class D chips with the local feedback, like TPA, usually suffer from the high THD in the post-feedback inductor.

How bad is it? I assume it's still less than 1% and inaudible... And probably not as bad as any speaker...

Personally, I've NEVER heard distortion from ANYTHNG unless it was broken or over-driven. It's not a spec that I worry about.

Purify (Bruno) has some nice articles about the global/local feedback loops.

I do appreciate the potential benefits of negative (corrective) feedback... It's a wonderful thing! But here's a quote from Amir that I saved. It's about speakers & headphones but it can apply to anything:

I don't know why people are so fascinated with how some speaker/headphone is made. What matters are the results.

 

[TurtlePaul]

I would check the THD at the frequencies >5kHz. The class D chips with the local feedback, like TPA, usually suffer from the high THD in the post-feedback inductor.

I'm sure that with the relatively good speaker, you'll hear the difference between those amps. To me, the cymbals sound crisp enough only with the THD like -100dB or higher.

Purify (Bruno) has some nice articles about the global/local feedback loops.

TPA amps also allow post-filter feedback. Bruno's patent expired 8 years ago and now a number of Texas Instrument based amps use this topology (e.g. Fosi v3 mono, 3e Audio A5/A7, Topping PA5 II, Aiyama A70).

TI TPA PFFB Application Report

Also, good luck hearing distortion which is in tones above 10 khz, our ears just aren't that good at those frequencies.

 

[Yevhen]

Personally, I've NEVER heard distortion from ANYTHNG unless it was broken or over-driven. It's not a spec that I worry about.

I don't hear distortion either, but the intermodulation at high frequency make the sound very unpleasant to me. You won't notice it if someone plays solo, but in a complex song with drums, cymbals, electric guitar, distorted or synth bass playing at same time, it gets very pronounced.

 

[Yevhen]

It's a wonderful thing! But here's a quote from Amir that I saved. It's about speakers & headphones but it can apply to anything:

In this case the technology defines a sound signature, and based on my experience all amps with a deep/fast feedback allowing THD -100dB and beyond (at all frequencies, not only at 1kHz) sound very clear

 

[Mike123]

TPA amps also allow post-filter feedback.

All the TPA3255 amps with PFFB that I've see have too much noise at high frequencies on multitone. I really don't like this behavior. It's too much for 2025. Something is wrong with this chip's high frequencies. The Purifi are much cleaner in this test up to 20 kHz.

 

[Yevhen]

TPA amps also allow post-filter feedback. Bruno's patent expired 8 years ago and now a number of Texas Instrument based amps use this topology (e.g. Fosi v3 mono, 3e Audio A5/A7

Maybe that 3e Audio has a global feedback, but it's clearly not working well at high frequencies. That 15kHz curve stays at -65dB