Ok, thanks, that's interesting.
Maybe Tippach gave Magix a newer version. What we get from asio4all.org is basically from 2015, 2017 came a patch that fixes usb issues in win 10, but that was just a bugfix, nothing more.

 

Asio4all is just a generic driver that was born out of the need for a better driver then what was available for onboard audio devices ( aka Soundblaster16 drivers) if you have an audio interface use the drivers developed for that device. ;)

 

the thread is bout asio4all vs magix not bout fonsis using asio4all while having hardware unit...but thx for input

 

Have a look here:
https://community.osr.com/discussion/181844/how-to-build-a-asio-usb-audio-driver

So I guess that Asio4all makes use of usbaudio.sys, that is the standard Windows audio kernel, while proprietary asio have their own kernel driver, suitable for their boards, and that should improve efficiency.

Indeed, I looked to the device manager, a4a is not listed, while ploytec is.
From ploytec site:
"Ploytec's    Windows USB Audio driver and   Mac OS X USB Audio HAL-plugin driver enable buffersizes down to 32 samples (0.73 ms) and create an ultra highspeed USB audio connection, bypassing the operating system's audio, its mixing and samplerate conversion.

This not only gives you low latencies, but better sound quality also."

Summing up, there could be two ways magix asio could work:
magix product installation checks for a SUPPORTED board (and for sure it works with several, cause I have different board brands on different pcs and it works with both), then:

1) uses usbaudio.sys and WRAPS audio like a4a,
2) OR uses a custom kernel driver for that brand and model of board.​

The last could explain why magix seems to be more efficient than a4a, and why I can use safe 64 samples buffer of magix vs safe 256 of a4a.

 

Last edited: Jul 27, 2019

You are off. Reverb is not included in the chart. It should be chorus and it goes out to 35ms. Above that is a percievable echo or delay.

 

I use win7, so I don't think it's related, see my previous post.

 

Speaking of keyboard latency, I agree with some claim, and only partially with others, I can tell my personal experience about that.

I think this is a serious matter and deserves a separate thread. If you don't start it, I will.

It's a threat!!!

 

Ahem, I already mentioned this @ reply number:
https://audiosex.pro/threads/asio4all-vs-magix-asio.45101/page-2#post-410041
yet we still run on for another 3 pages.
Mama, are we there yet?

Easily answered by looking at Michael Tippach's a4a page and/or the magix driver's settings panel.
latest a4a @ www.asio4all.org is v2.14
latest magix that I've installed is v2.13
(top left of settings panel in both editions)
but I suppose it might have been updated to v2.14 since I last looked.
Maybe somebody with a recent magix install would tell us?

as a performer, acceptable latency is that which one can comfortably play and record at, obviously the shorter the better and obviously without the audio glitching out.
when mixing or other jobs where inputs or (nearly) realtime feedback aren't required, it's perfectly reasonable to extend latency to give your poor old, hard working cpu a bit of a rest.
I guess it's a bit like being nice to someone after a job well done.
everything else is just number crunching geekery that gets in the way of creating music.

good luck everybody

 

Last edited: Jul 27, 2019

That is OUR assumption, based on gui similarities, unless you have a reliable source that ends it once for all.

Not clear, but I guess Plendix compared the hex code and found common snippets or a4a registration, but code can't be exactly the same, and I found that they react differently with same settings, and that simply can't be IF it's the same driver.

If I was sure that a4a and mgx were exactly the same, I would uninstall mgx.

 

Yes, I agree, but I had serious latency issues when playing live vst piano, I'll be more specific about it.

 

I only wish there was a good multi-device ASIO, but i suppose that would require everyone to use a standard like RME and then I could afford nothing at all - LOL
Or they could all have a crappy low standard and no one could do anything.
I suppose the manufacturers would fight it for a variety of reasons, but one I think of is that they don't want you to be able to use the several interfaces you already own due to upgrading. It seems even with the same manufacturer it is either rare or impossible to use two at once.
Well i did "upgrade" my interface from a 2-in PreSonus USB to an 8-in (at once) StudioLive AR8.
I think that is about all the inputs I will ever need as a hobbyist.
I used to use a Mackie 1202 VLZ and I must say ... the interface hardware of the AR8 sucks.
I'm afraid to even turn a knob.