Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Mac vs PC?

That's good to know - have you compared wasapi to the asio4all driver?

You talkin' to me? You talkin' to me?
I've not used ASIO4All since I got the Presonus box around ten years ago. The latter came with its own ASIO driver, so there was no longer a need to use ASIO4All.

External interfaces are not just manufactured to just have better latency. There are many other advantages. better ADC/DAC, better amplifier for both headphones and speakers, MIDI I/O better mic pre amps.

You can't have decent amp and ADC/DAC in a laptop that have direct influence on the latency!

I've been using my vpc1 via usb... and didn't even think about using the midi connections for my UMC204hd. I'm still pretty green around music technology - but if I used the vpc1 with the midi instead of usb (having to power the vpc1 with an external ps right?? or can I leave usb connected for power?) would I just use the vpc1 -> midi out -> behringer -> usb to pc/mac? I'm not quite sure if there would be any reason to use the vpc1's midi input, is there (ie, connecting behringer midi out -> vpc1 midi in)

WASAPI driver latency is usually so high, because general purpose WASAPI drivers use Shared Mode. However if AI maker offers proprietary WASAPI Exclusive or WDM-KS (Kernel Streaming) driver - these can deliver latency similar to ASIO.
By the way, ASIO4ALL is wrapper around WDM-KS driver - it can't be used without WDM, this is why it never worked with Win95 or NT.

It is not the driver which decides if the WASAPI is shared or exclusive. It is the application. With Reaper, we have the choice. But we can have applications which don’t propose the exclusive mode.

Multi-tasking and asynchronous I/O were well understood by the 1970's. If you look at the typical features of a state-of-the-art operating system in the 1970's as implemented in Unix systems and DEC proprietary operating systems, I think NT4 was when Windows caught up to that, and Windows 2000 was the first with widespread usage.

I would add that the design center of Unix was software development, not control of telephone switches.

UNIX is just a set of API the OS must conform to, implementations can be very different.
UNIX-RT and UNIX-RTR both were true real time OSes with UNIX layer.

That may be true today, but was not true in the 1970's.

Yes, IEEE POSIX standardization began in 80th. Before that all versions of UNIX were based on the source code licensed by AT&T to other vendors, and all major version updates were from AT&T.
The first major split happened only when free BSD NET/1 was released, and this was 1989, right after the first IEEE standard was published.

Also about the same time, the unix-based OS for NeXT was released (which became the foundation for Mac OS X), based on Mach that originated in 1985.

I remember those NeXT computers - square black desktop cubes. Almost like the original mac.

That's not historically correct. The first major split was when UC Berkeley implemented what became BSD (Berkeley Standard Distribution) in the late 1970's. There was Bell Unix and Berkeley Unix. Posix started as a result of that split, as an attempt to provide a uniform platform for software portability across Unix platforms.

Nope, intially BSD was just PDP/VAX software distribution package (Berkeley Software Distribution - Pascal compiler, ex, vi, c shell, tty, etc) built on top of AT&T Unix code when Thompson worked in Berkeley. BSD added new utilities, its own networking stack, and still used but slowly replaced AT&T code until Net/2 which was first BSD competely free from AT&T code.

You are right though that POSIX was needed because System V and BSD became technically completely independent, both were widely used by big guns (System V - IBM, HP; BSD - DEC, Sun) and there was risk of application incompatibility nobody was interested in.

I believe the original BSD implementation included a denand page virtual memory system that was lacking in Bell Unix at the tine. By the time of 4.1BSD and Bell System III around 1980, Berkeley and Bell Unix were not compatible platforms for running software. Bell System V just made it worse.

Is there any particular reason why some computer geek cannot develop a replacement for ASIO4ALL (which is ancient and single-client)? As in a multi-client application with a modern interface? I don't need an external box interface for piano gigs as not recording saxophone. Having to not bring a box would be great.

AnotherScott, are you achieving good results with your Surface Pro? I'd like to bring VSTs to a gig without bringing my 17" day gig work laptop. Thanks...

I was happy with the results in my limited experimentation, but never actually gigged with it or stressed it much. I added an Alesis Control Hub to get the latency down, which worked well since it was a small, inexpensive box that gave me audio out AND the MIDI in jacks (which is important because the SP only has one USB port). I ended up adding a small USB hub anyway because I had to connect an iLok, but that's not always necessary, there are lots of VSTs that don't need iLok, and even those that do can now often work without the physical dongle.

BTW, I had previously briefly experimented with the first Surface Pro, but found its screen was too small (it's 10.6" instead of the SP4's 12.3"). Too many Windows apps just don't scale down well, and that was below my threshold of usability. If you need something super small to enhance your rig, I'd give up on proper VSTs and stick with an iPad (even the iPad Mini's 7.9" works).

Robin Vincent designs computers for professional audio studios in the UK.

He also has a series of videos on the SurfacePro for homegamers, with recommended tweaks. He recommends certain SurfacePro models below.

http://surfaceproaudio.com/