can you recommend a good cheap low latency board/box

[mhog]

Yes, it seems those brands are for professional studios. I get the impression you are looking for normal quality consumer products instead, though. Steinberg, Zoom, Roland, etc.

My suggestion:

Extremely compact size (around 10cm x 10cm)
24bit 96/192KHz
1 max 2 input line-mic-guitar I/O
USB2.0

Examples:


ICON CUBE MINI


ALESIS CORE1


FOCUSRITE VRM Box


ESI Ugm96


These are WAAAAY more than enough for his needs, all around $80-100 and all good quality devices. I would not spend more, in terms of latency and sound quality they do a very good work. Otherwise it's just wasting money (see the "Ferrari vs Golf" allegory I posted above).

[fluffy_little_something]

Very compact indeed
One thing to consider is whether or not he needs an XLR microphone input with phantom power for quality microphones.
And whether or not the mic preamp supports high-impedance headphones.

[mhog]

Very compact indeed
One thing to consider is whether or not he needs an XLR microphone input with phantom power for quality microphones.
And whether or not the mic preamp supports high-impedance headphones.

No, he claimed he doesn't. He only plays virtual instruments. That is why I am suggesting him to avoid wasting money for nothing. IMO these kinds of interfaces are more than enough for his needs. If I were him and had $500-600 to spend, I would satisfy my "Gear Acquisitin Syndrome" in this way: $100 for an audio interface and the rest in virtual instruments, instead of buying a second hand $1,000 RME just for "hey, look at this professional studio's jewel" sort of showing up... for nothing

[Kr3eM]

IMHO you are just mixing things up. If you import 16 bit within a 24 bit environment, then add some new process end export everything 24 bit, that makes sense. But what you claimed was that you usually export your 24bit audio into "32bit" via a 24bit audio board. This makes no sense at all.

It doesnt make sense because I never claimed to export any 24bit into 32bit and I for sure never mention or talked about a "24bit audio board" by which I not even sure what you actually mean.

I am and was always talking about "rendering your music from DAW into audio file" which you claimed that if you had a 16bit soundcard you could not render it to anything higher than just 16 Bit.

This is as I explained utterly wrong, the soundcards bit depth limitation is only a limitation when it comes to Analog to Digital or Digital to Analog convertion. There is NO such AD-DA conversion involved in the process of rendering a channel like your master to a file. Rendering to file is internal and recording is either internal or external depending on the source.

Your Master Out (if you use Live) is 32 Bit, not 24 Bit. The preference settings is for recording and has nothing to do with the bit depth of Live's "sound engine". Most DAWs are working in 32 Bit and most plugins as far as I know work at 64bit internally so the "environment" as you put it not at all 24 Bit.

Although the Bit setting in live for recording will also decide which bit depth you will get when you consolidate a wav file. In this case there surely will be extra zeros involved depending of the nature of the editing made.
So which setting is to prefered depends on what you are doing.

I personally think there should be seperate settings for internal and external recording as well for the consolidate function because if you are recording one channel with another internally direct during playback and have choosen 24 bit you have recorded 32 bit data to 24 bit. However if you have set the preferences to 32 Bit and record a external source like vocal you will have wasted space on your drive.

Edit: I may be wrong about the consolidation fuction, I just consolidated a pitched 16 bit drumloop to a 24 Bit file but according to Schwa Bitter it generates 32 Bit data (no warp or anything activated) but if I import a 24 Bit drumloop Schwa Bitter says its 24,,, have no idea why or how, just mention it as a correction..)

If you are rendering your track to file from Live I recomend you to choose 32 Bit on export if you are going to master it further or send to a Mastering Engineer. The dithering are designed to be the last process, and if possible be avoided til then.

(since a human ear could not however hear any difference between 16, 24 or 32 bit... only computers can)...

That all depends on the circumstances, but either way I think you may confuse bit depth with frequency range (sample rate)... bit depth is dynamic range.



back on topic
For what is worth I agree that a RME card is overkill for the OP, I wont argue with RME have excellent drivers and products, afterall I have a FireFace UCX myself.

I would consider ESI due to the quality but the drivers can be tricky to install for the average user, well at least it use to be 10 years ago. I had a ESI 192 WamiRack XL, and it served me very well for a long time. A friend of mine eventually went from a Motu to a cheap small ESI model (10-12 years ago) and he was very happy with it.

Yes there are some risks involved with cheap cards and drivers, but if the OP buys from a place like thomann or similair that have a return policy I would take chance... although I would wait with the Beringher cards until they actually have a final release of a driver for windows. As far as I know there's only a beta... although things may have changed, just a heads up.

[Dominus]

Examples:

ICON CUBE MINI

ALESIS CORE1

FOCUSRITE VRM Box

ESI Ugm96


These are WAAAAY more than enough for his needs, all around $80-100 and all good quality devices. I would not spend more, in terms of latency and sound quality they do a very good work. Otherwise it's just wasting money (see the "Ferrari vs Golf" allegory I posted above).

The VRM Box is discontinued and no longer supported. It also gives you 10ms latency. Not quite low latency there. My experience with Focusrite was horrible. The VRM Box was okay, but I then tried a first generation Scarlett. Their driver support was a joke. Thankfully I was able to return the Scarlett before my 30 days was up. (Actually, I may have been a few days over, but the place I got it from noted that LOTS of people had problems with them and they worked with me.)

ESI UGM96 is also discontinued, and support appears to be adding compatibility to OSes every few years. I had a Juli@ years ago, it was okay.

Alesis Core 1 is also discontinued. I don't even see where you can download drivers on Alesis' site.

I'm not finding much on the Icon Cube Mini, but they never released drivers for Mac OSX 10.11 (El Capitan) or Sierra, so I'd expect their support to be lacking as well.

Got any other good suggestions? Your VW Golfs are actually Yugos.

Yes, the RME Fireface I suggested was discontinued long ago. But it's still supported. That's why I suggested RME in the first place.

(Also, it's analogy, not allegory.)

[mhog]

I never claimed to export any 24bit into 32bit and I for sure never mention or talked about a "24bit audio board" by which I not even sure what you actually mean.

Actually, it wasn't me who wrote:

By that logic I should not be able to render my songs to 32 Bit Wav since my soundcard is only 24 Bit, but still I somehow manage to do that. Explain that.

Your Master Out (if you use Live) is 32 Bit, not 24 Bit. The preference settings is for recording and has nothing to do with the bit depth of Live's "sound engine". Most DAWs are working in 32 Bit and most plugins as far as I know work at 64bit internally so the "environment" as you put it not at all 24 Bit.

I guess you are mixing things up again, because the 32 and 64 bit you are referring to are something different from recording/sampling audio resolution (or "audio quality" for ordinary mortals). If a sound was sampled in 24bit/192khz, there is no way to improve this resolution. You can only downsize it (24 to 16, 16 to 12, 12 to 8... etc.). However, if you think re-rendering it in a fake 32bit (within the same 24bit/192khz environment?!) will improve dynamic and quality, ok... do what you like, mate. For me it's just total nonsense, a waste of memory and useless digital noise. And, again, it does not work like this. Otherwise, why introducing higher resolutions? Input/Outuput analog-digital convertion implies "output", too... how can you improve the "printed image" of a sound just adding 8 useless 0000-0000 noise bits remains a mistery for me Unless you mean the remastering process, which is a totally different matter.

[Kr3eM]

I never claimed to export any 24bit into 32bit and I for sure never mention or talked about a "24bit audio board" by which I not even sure what you actually mean.

Actually, it wasn't me who wrote:

By that logic I should not be able to render my songs to 32 Bit Wav since my soundcard is only 24 Bit, but still I somehow manage to do that. Explain that.

Really? language barrier? note the WORD RENDER, we are talking about rendering your master to a file, NOT RECORDING AN EXTERNAL signal thru the soundcard....

Where do I say I route out the master to a 24 Bit audio board ? Seriously ? I just saying that my soundcards bit depth is 24 bit (RME FireFace UCX) and it has NO EFFECT what so ever when you rendering your master channel to a wave file... doens matter if your soundcard is 16 bit, U CAN STILL render your master to 24 bit or 32 bit....

The ONLY thing you are constantly refering to is to record something THRU your soundcards AD...

Your Master Out (if you use Live) is 32 Bit, not 24 Bit. The preference settings is for recording and has nothing to do with the bit depth of Live's "sound engine". Most DAWs are working in 32 Bit and most plugins as far as I know work at 64bit internally so the "environment" as you put it not at all 24 Bit.

I guess you are mixing things up again, because the 32 and 64 bit you are referring to are something different from recording/sampling audio resolution (or "audio quality" for ordinary mortals). If a sample was recorded in 24bit/192khz, and you think re-rendering it in a fake 32bit (within the same 24bit/192khz environment) will improve dynamic and quality, ok... For me it's just total nonsense a waste of memory and useless digital noise. And, again, it does not work like this. Otherwise, why introducing higher resolutions?

No you are not hearing what I am saying, you are just talking about plain recording thru your soundcard into a file. I am talking about rendering a masterchannel with VSTI's and VST plugins an what not....

If a sample was recorded in 24bit/192khz, and you think re-rendering it in a fake 32bit (within the same 24bit/192khz environment)

But I am not re-rendering a 24 bits file I am rendering a master out channel from Live's master out which is the sum of a lot of wav files with plugins and volume adjustments and that environment is not 24 Bit it is 32 Bit...

I took the time to try to see if I didn't just missunderstood your comment, I clearly asked if "Are you saying that if you have a 16 Bit soundcard you can only render your song/track to 16 Bit wav and no higher?" Note that I asked "render" not record...

And no matter what you are trying to say, that statement is wrong.

You could easily download Schwa Bitter, put after Kontakt and load a 16 or 24 bit library and see which bit depth is coming out, the same with the master channel... I don't know about other host but Live will not have a bit depth of 24 on the master channel on regular project.. doesn't matter if your preferences are set to 16,24 or 32 bit recording. It has nothing to do with it...

The " 24bit/192khz environment" is the environment of your soundcards AD-DA converter, not the internal environment of your DAWS soundengine.

[mhog]

Examples:

ICON CUBE MINI

ALESIS CORE1

FOCUSRITE VRM Box

ESI Ugm96


These are WAAAAY more than enough for his needs, all around $80-100 and all good quality devices. I would not spend more, in terms of latency and sound quality they do a very good work. Otherwise it's just wasting money (see the "Ferrari vs Golf" allegory I posted above).

The VRM Box is discontinued and no longer supported. It also gives you 10ms latency. Not quite low latency there. My experience with Focusrite was horrible. The VRM Box was okay, but I then tried a first generation Scarlett. Their driver support was a joke. Thankfully I was able to return the Scarlett before my 30 days was up. (Actually, I may have been a few days over, but the place I got it from noted that LOTS of people had problems with them and they worked with me.)

ESI UGM96 is also discontinued, and support appears to be adding compatibility to OSes every few years. I had a Juli@ years ago, it was okay.

Alesis Core 1 is also discontinued. I don't even see where you can download drivers on Alesis' site.

I'm not finding much on the Icon Cube Mini, but they never released drivers for Mac OSX 10.11 (El Capitan) or Sierra, so I'd expect their support to be lacking as well.

Got any other good suggestions? Your VW Golfs are actually Yugos.

Yes, the RME Fireface I suggested was discontinued long ago. But it's still supported. That's why I suggested RME in the first place.

(Also, it's analogy, not allegory.)

Come on, you are talking as if you were a scientist playing with some atomic bomb or some brain surgeon haha Besides, I own two Focusrite and they both work like a charm, the pre-amps are spectacular and they are ultra-silent, either on my two macs (native) or on my PCs (ASIO). ASIO drivers are universal, and ASIO4ALL is a good driver, too. The tiny audio interfaces might be discontinued, but still on the market, at least here in EU. Otherwise he can find similar "brand new" 2017 products, which do exactly the same job but cost a little more because "2017 products".

Regarding the 10ms latency: it is a very short latency, perfect to play any virtual instrument in a DAW. Everything under 30ms is more than enough. My Roland D50 (hardware synth) has 17ms latency. Almost the same for my Kord Triton (14 ms).

10ms or 5ms... there's no difference at all in terms of human perception of latency (so called "echo"). Besides, he does not even need direct monitoring, not playing external instruments or voice...

[mhog]

But I am not re-rendering a 24 bits file I am rendering a master out channel from Live's master out which is the sum of a lot of wav files with plugins and volume adjustments and that environment is not 24 Bit it is 32 Bit...

The " 24bit/192khz environment" is the environment of your soundcards AD-DA converter, not the internal environment of your DAWS soundengine.

You are wasting your time and energy, IMO. Unless you mean: "I have no audio tracks at all, no voice, no instruments, no samples in my DAW. I only use virtual analog synths and effects, midi tracks only and want them to be recorded ("rendered") in the best resolution available". Still useless IMO, but makes a little sense. Otherwise, I really don't understand this unnecessary exertion of yours.

[Kr3eM]

But I am not re-rendering a 24 bits file I am rendering a master out channel from Live's master out which is the sum of a lot of wav files with plugins and volume adjustments and that environment is not 24 Bit it is 32 Bit...

The " 24bit/192khz environment" is the environment of your soundcards AD-DA converter, not the internal environment of your DAWS soundengine.

You are wasting your time and energy, IMO. Unless you mean: "I have no audio tracks at all, no voice, no instruments, no samples in my DAW. I only use virtual analog synths and effects, midi tracks only and want them to be recorded ("rendered") in the best resolution available". Still useless IMO, but makes a little sense. Otherwise, I really don't understand this useless labor of yours.

Ok, you can't even grasp the idea that you are wrong and or are missunderstanding me, and apparently not even be bothered to investigate it... so yes I will leave you to your ignorance....

https://www.ableton.com/en/manual/audio-fact-sheet/
"31.2.4 Summing at Single Mix Points

.......
Please note that, while 64-bit summing is applied to each single mix point, Live‘s internal processing is still done at 32-bit. Thus, signals that are mixed across multiple summing points may still result in an extremely small amount of signal degradation. This combination of 64-bit summing within a 32-bit architecture strikes an ideal balance between audio quality and CPU/memory consumption.


31.2.5 Recording external signals (bit depth >/= A/D converter)

Recording audio signals into Live is a neutral operation, provided that the bit depth set in Live‘s Preferences window is the same or higher than that of the A/D converters used for the recording. In this context, “neutral“ means “identical to the audio as it was delivered to Live by the A/D converters.“


31.2.6 Recording internal sources at 32 bit

Audio that is recorded via internal routing will be identical to the source audio, provided that the recording was made at 32 bits. To ensure neutral recordings of plug-in instruments and any audio signals that are being processed by effects plug-ins, internal recording at 32 bits is recommended. Please note, however, that if the source audio is already at a lower bit depth, internal recording at that bit depth will also be neutral (assuming that no effects are used); internally recording an unprocessed 16 bit audio file at 32 bits will not increase the sound quality. "

[mhog]

So what? I just tried to put a 2 seconds sample into Ableton and rendered it two times: 16 bit 44.100 and 32 bit 192khz. Results:

wav file 16 bit: 382 kb
wav file 32 bit: 3,25 mb

Almost 10 times bigger for the SAME audio file. Again, what's the point? This is crazy, since the two samples sound EXACTLY the same. Unless you really think 32 bits "sounds better". It doesn't, untill they tell you which is which. I guess the only good reason to waste tons of memory in this crazy way is in professional environments, for instance a classical music orchestra, voice etc. ORIGINALLY RECORDED with that ultra-high resolution and the resulting samples put in audio tracks (Pro-tools). Otherwise it's simply useless. 16 bit 44100 is still (and will remain) the perfect standard in my book.

[mhog]

If you are rendering your track to file from Live I recomend you to choose 32 Bit on export if you are going to master it further or send to a Mastering Engineer. The dithering are designed to be the last process, and if possible be avoided til then.

Actually, if I needed a Mastering Engineer, I wouldn't send him any audio tracks made in my small room, I'd go directly into his thousand dollars studio and record instruments and voices there. Otherwise everyone here is a Mastering Engineer, thanks to the nototious "RTFM" law

[Kr3eM]

You are not reading my post obviously, do even try to understand the context?

So what? I just tried to put a 2 seconds sample into Ableton and rendered it two times: 16 bit 44.100 and 32 bit 192khz. Results:

wav file 16 bit: 382 kb
wav file 32 bit: 3,25 mb

Almost 10 times bigger for the SAME audio file. Again, what's the point? This is crazy, since the two samples sound EXACTLY the same.

from my previous post...

....from https://www.ableton.com/en/manual/audio-fact-sheet (31.2.6)
Please note,
however, that if the source audio is already at a lower bit depth, internal recording at that bit depth will also be neutral (assuming that no effects are used);

internally recording an unprocessed 16 bit audio file at 32 bits will not increase the sound quality. "

Unless you really think 32 bits "sounds better". It doesn't, untill they tell you which is which. I guess the only good reason to waste tons of memory in this crazy way is in professional environments, for instance a classical music orchestra, voice etc. ORIGINALLY RECORDED with that ultra-high resolution and the resulting samples put in audio tracks (Pro-tools). Otherwise it's simply useless. 16 bit 44100 is still (and will remain) the perfect standard in my book.

'


You are so of the context it's scary...I have provided all the facts... there's nothing more to add. You are stuck in the context of one simple process of re-rendering a file. That has never been the subject. You should read up on 32 Bit Float processing in DAWS and how it works and why we use it. As you earlier said, you didn't even know it existed until my post, so do you really think that you have clear understanding of it already? (retorical question)

Facts are presented as well as the context, everything else is reduntant from this point.

[Kr3eM]

If you are rendering your track to file from Live I recomend you to choose 32 Bit on export if you are going to master it further or send to a Mastering Engineer. The dithering are designed to be the last process, and if possible be avoided til then.

Actually, if I needed a Mastering Engineer, I wouldn't send him any audio tracks made in my small room, I'd go directly into his thousand dollars studio and record instruments and voices there. Otherwise everyone here is a Mastering Engineer, thanks to the nototious "RTFM" law

Ohh I so whish you were just trolling me.... you seriously think that a professional Mastering Engineer's studio is the same as a Recording Studio..that's soooo adorable.

[Dominus]

Come on, you are talking as if you were a scientist playing with some atomic bomb or some brain surgeon haha Besides, I own two Focusrite and they both work like a charm, the pre-amps are spectacular and they are ultra-silent, either on my two macs (native) or on my PCs (ASIO). ASIO drivers are universal, and ASIO4ALL is a good driver, too. The tiny audio interfaces might be discontinued, but still on the market, at least here in EU. Otherwise he can find similar "brand new" 2017 products, which do exactly the same job but cost a little more because "2017 products".

Regarding the 10ms latency: it is a very short latency, perfect to play any virtual instrument in a DAW. Everything under 30ms is more than enough. My Roland D50 (hardware synth) has 17ms latency. Almost the same for my Kord Triton (14 ms).

10ms or 5ms... there's no difference at all in terms of human perception of latency (so called "echo"). Besides, he does not even need direct monitoring, not playing external instruments or voice...

Reading your other exchange in this thread, you really have no idea what you're talking about. You're just embarrassing yourself at this point.

ASIO drivers are *not* universal.
ASIO4ALL is not a driver, per se. It still relies on the WDM driver to be installed, so if no driver is available for the hardware, you're SOL.

*You* may not hear a 10ms latency, but some people can.

From the beginning, I mentioned that the RME is expensive, but it is the just about the best that you can get. I asked that other people make suggestions. Your continuance of arguing, along with your poor suggestions, have added nothing to this thread.