Advice On Tracking With USB Interface With Guitar Plugins

[Adam Christopher]

Hi All,

I recently got a 2.0 USB interface to replace my old firewire. The USB interface sounds better with the pres/converters versus the firewire I was using before. But of course, the drawback of USB 2.0 is latency can be an issue.

I have had good luck with recording using mics with the Audio-Monitoring Switching Protocol Off but if I am looking to do direct stuff (no mics) with plugins like guitar plugin vsts, I get the phasey latency sound when I have the Tape Style Playback/Input Switching or Tape Style/Input Always On. Of course when I playback the recording, it sounds much better but never seems to be as clean as I had recorded with Audio-Monitoring Switching Protocol Off. I've had best luck with adjusting the Out buffer size to 128 or 256.

Does it sound cleaner because I am playing differently to what I am hearing (no delay when actually recording with Audio-Monitoring Switching Protocol is off) or is latency still affecting the recorded sound when Tape Style/Input Always on or Tape Style Playback/Input Switching is engaged?

The thing is that I need to be able to hear the vst plugins while recording. I seem to remember having few issues when I was using a Mackie firewire but this seems to be problematic for me with the Steinberg UR28m that I have grown to love because of the nice pres/conversion. I was also using the MR816 (firewire) before but that thing broke on me and I went for a cheaper replacement.

Before any suggests the downloading the ASIO4all drivers, I tried that and it is no different. Also, my computer is an I5 so its not that either.

Maybe recording with vst plugins like Guitar Rig or whatever is useless with USB interfaces. I know a lot of people on this board are advocates of PCI Express or even more PCI but for my modest needs I just want to be able to record with vsts (with distortion sounds) without it sounding like sludge every time. Its not the same experience to record just a clean signal and tweak the plugin at the mixing stage.

I tried that Force Real Time Priority Class option and am not sure how much of a difference it makes.

Maybe I'm looking in the wrong menu and should be looking to utilize other options other than the Audio-Monitor switching protocol?

Would love to hear your feedback and how you typically handle the vst guitar plugins if anybody here actually dares use a USB interface for that stuff.
What do you guys do with the guitar plugins?

Maybe I just need to ditch USB altogether. I'm not recording a whole band or anything its just me and its only an issue if I am try to hear plugins (any of them) when I record.

Thanks in advance.

AC

[tomasino]

Hi All,

I have had good luck with recording using mics with the Audio-Monitoring Switching Protocol Off but if I am looking to do direct stuff (no mics) with plugins like guitar plugin vsts, I get the phasey latency sound when I have the Tape Style Playback/Input Switching or Tape Style/Input Always On. Of course when I playback the recording, it sounds much better but never seems to be as clean as I had recorded with Audio-Monitoring Switching Protocol Off. I've had best luck with adjusting the Out buffer size to 128 or 256.

AC

It depends. (I know, doesn't help much.)
It's unique for every system, setup and mix project.

There's many ways to address this problem n' you're on one of the right tracks.
See the Input Monitoring section of the Manual. Describes details on how to adjust input/output buffers for optimal monitoring performance.


I hit this just last night. After adding a nice compliment of plugs to a mix..the Levelizer in a couple of places and the Sonoris Pitch/Time on two Returns,
decided to re-track the acoustic guitar (with the new CloudLifter doodad :cool:) n' behold.. monitoring latency.

Just have to be careful with Latency Inducing and/or Buffer Changing plugs.
e.g.


the Levelizer (will auto adjust buffers)
Sonoris Pitch/Time
Sonoris MultiBand Compressor (seems to require a little larger buffer size.. in conflict with high performance monitoring.)
many VST's

OR
you could use the "BuildMix to Hot Track" feature and then disable all other channels while recording new tracks.

OR
I think you could use Mix Templates to toggle between a latency avoiding setup for recording/tracking and straight ahead mixing setup.

I'm sure there's more..

[Jeff Scott]

I generally track with buffer sizes 2x64 or 1x64. When mixing and starting to add a bunch of plugins....I increase it to 4x64. This helps with plugs like multiple instances of the Levelizer.

At 2x64 I don't have discernable latency. I use POD Farm Platinum primarily as my Guitar plug...and I'm running an instance of Superior Drummer playing Midi Drums with no problems

1x64 is when I'm tracking Edrums.

2x128 ...and I hear latency.

Hope that helps...

[Adam Christopher]

It depends. (I know, doesn't help much.)
It's unique for every system, setup and mix project.

There's many ways to address this problem n' you're on one of the right tracks.
See the Input Monitoring section of the Manual. Describes details on how to adjust input/output buffers for optimal monitoring performance.


Thanks very much, Tomasino. I already tried various buffer settings and couldn't get any desirable results. The Force Real Time Monitoring makes some difference but it doesn't solve the latency problem.

It's not so much latency caused by plugins. If I am Tape Monitoring mode, I get latency no matter what signal I am passing through without plugins. So I am starting to think that this particular interface I am using is ill-suited for this application. My only option might be to record a clean signal with the Tape Monitoring off and then adding the plugins at the mixing stage as if reamping in the digital realm.

I'm considering now that I may need to get a PCI Express card but I like the convenience of my little box and don't want to deal with messy breakout cables in my small apt. If I can get the latency to a usable level I would much prefer to stick with my modest setup.

[Adam Christopher]

I generally track with buffer sizes 2x64 or 1x64. When mixing and starting to add a bunch of plugins....I increase it to 4x64. This helps with plugs like multiple instances of the Levelizer.

At 2x64 I don't have discernable latency. I use POD Farm Platinum primarily as my Guitar plug...and I'm running an instance of Superior Drummer playing Midi Drums with no problems

1x64 is when I'm tracking Edrums.

2x128 ...and I hear latency.

Hope that helps...

Thanks Jeff. That's helpful to know what settings work for you. I have my doubts though that this will work because there's a huge difference between the RME in your sig and my USB interface or any USB for latency I would presume. I'm thinking I might have to settle for recording a clean signal and "reamping" or getting a PCI express sound card.

[Angie]

Adam,
Do you have SAC? There are a few of us who use SAC as the front end in the studio situation. It offers better latency handling.

[Microstudio]

You just need to test your settings to find one that offers the best latency. I have no problem with latency when playing Guitar, Drum, Synth..ect VI's

Just know the 11R is usb and it's made for guitars and uses ASIO drivers so it's not the USB it's all in your setting.

[Adam Christopher]

Thanks Angie and Matt. No I don't have SAC. I'll have to look into other settings then to see if I get it to work but there's no probably if I have Tape Monitoring turned off. No latency there. The problem is when I am trying to record with plugins. That is encouraging to know that it works well with the 11rack USB. I can record fine with my Yamaha THR, which is also USB. But in those cases the sound is coming from the hardware not the plugins. I guess I'll have to play around with the settings. Thanks.

[IraSeigel]

Adam,
I think you're comparing apples and oranges. The USB device sounds better because you're using better preamps than you had with the Mackie/Firewire device. Your issues with the USB are probably not solvable. You'd be better off going back to Firewire and using some better preamps.

As you mentioned, you went for the cheaper replacement, but you really get what you pay for.

Ira

[Adam Christopher]

Adam,
I think you're comparing apples and oranges. The USB device sounds better because you're using better preamps than you had with the Mackie/Firewire device. Your issues with the USB are probably not solvable. You'd be better off going back to Firewire and using some better preamps.

As you mentioned, you went for the cheaper replacement, but you really get what you pay for.

Ira

Thanks Ira. I know the USB device sounds better because both the preamps and converters are better. That's why I got it in the first place. I'm sure you are right that going back to firewire (or going PCI Express) is the solution. In that case, it probably makes sense to see if my broken MR816 can be salvaged. That guy is way better all around than Mackie (when it works).

I don't agree with you, however, on I went for the cheaper replacement and got what I paid for. In the Steinberg family, their flagship is the UR824. It has slightly better conversion than that unit I have but it also has the exact same latency as their flagship model, UR824 (also a USB unit). The UR824 was the follow up to the MR816 (firewire) that I have and is not working for me. That's more of a matter of investing in the wrong technology not because I went for a "cheaper replacement". It looks like they banked a lot on users wanting Cubase integration for direct monitoring (and convenience of USB) but outside of Cubase it's not very good latency-wise.

Aside from the preamp/converter quality, part of the appeal for the UR28 is simply I like the small footprint. I am pleased with the sound aside from the latency issue when using plugins. If it's microphones or any other source, there are no issues. The preamps on the Steinberg units are fantastic and so is the conversion even on the lower end units.

When I have time, I'll have to see what they can offer me for repair. Thanks again for the response.

[IraSeigel]

I, too, am drawn to the smaller footprint of the USB devices, specifically the RME UCX (or UFX). At least for recording. I'm not sure I could get the latencies low enough for SAC use.

Unfortunately, I'm not yet willing to pay the going rate for either of those units:)

[Adam Christopher]

I, too, am drawn to the smaller footprint of the USB devices, specifically the RME UCX (or UFX). At least for recording. I'm not sure I could get the latencies low enough for SAC use.

Unfortunately, I'm not yet willing to pay the going rate for either of those units:)

I know! it's tough to have your cake and eat it too. Whatever that means :)

[sebastiandybing]

I have been using rme firefaceUC for years and the babyface for more than
a year. I have been using both with sac and saw in live
situations and with film sound rec. where sac are doing
a live mix for cameras.

I have for years used a HP6410 with a 2.1ghz
core2 first gen. for film sound rec. at 1/64 buffer,
it have never failed.
I have now updated to a Lenovo T520 where I have
2 hd, one with XP and one with win8,
On xp it run the usb devices like sports cars,
on win8 I do have problems with low latencis, eventhough
sometimes it seems running fine and suddenly there
are dropouts or small clicks in the sound.

Sebastian

[Adam Christopher]

I have been using rme firefaceUC for years and the babyface for more than
a year. I have been using both with sac and saw in live
situations and with film sound rec. where sac are doing
a live mix for cameras.

I have for years used a HP6410 with a 2.1ghz
core2 first gen. for film sound rec. at 1/64 buffer,
it have never failed.
I have now updated to a Lenovo T520 where I have
2 hd, one with XP and one with win8,
On xp it run the usb devices like sports cars,
on win8 I do have problems with low latencis, eventhough
sometimes it seems running fine and suddenly there
are dropouts or small clicks in the sound.

Sebastian

Thanks Sebastian. Are you using guitar plugins or any kind of tape monitoring so plugins are engaged at the same time? I haven't had any problems with the default settings but with tape monitoring engaged, that's when its not working.

[Adam Christopher]

Hi All,

After experimenting with a bunch of different setups, I was able to identify partially what some of my issues were but I still do have some latency issues that remain. First, the extreme latency it seems was specific to one of the inputs on my hardware interface. I had a stereo preamp that was connected via the headphone output. It is supposed to be a stereo signal with trs cables but I made it mono. When this unit is connected, I was getting the extreme latency when tracking as a mono track. However, I don't really need to use tape monitoring mode since I get the sound from this preamp anyways. With tape monitoring off, latency is no issue at all. But I was able to at least figure out that the extreme latency in that particular setup was a specific case.

I tried one of the other inputs with just a guitar going into the hi z input with tape monitoring and my favorite guitar amp sim Overloud Th2. The latency was much, much better. However, I was only able to get desirable results at 1 buffer at 64 + Real Time Performance Mode, which the most extreme setting. At that setting, I get lots of dropouts and even then I still notice a bit of latency but I can actually hear what the plugin is doing without too much distraction. I tried a few other DAWS and some of them seemed to handle the 64 ok while others didn't. But it was at least encouraging that I could get in the ballpark of something usable.

First off, all your posts were very helpful. I tried out Jeff's settings and that definitely pointed me in the right direction. Also, Angie's post about using SAC as a front end really intrigues me. Since I was able to get usable performance out one of my DAWs at buffer settings of 64 (before the dropouts hit), I'm thinking that SAC could possibly make the difference. I read some of the old posts and I got the impression that performance at low buffer settings isn't one of Saw's main strengths as things are set up in a way to allow steady speed and stability and overall performance while SAC as a live application is optimized for low buffer settings. But of course, I wouldn't know other than what I gathered from reading. I couldn't get the sense whether it made a small marginal difference or a huge difference.

Can anybody point me in the right direction of how I can demo SAC within Saw and what's the easiest way to figure it out? A cliff notes version would be preferable lol. The awesome microstudio videos helped me get up to speed in Saw but I would be 100% clueless with trying to route things with SAC as a front end. Not something I can easily conceptualize. Keep in mind that I don't have the same background as a lot of you with live setups and routing so I'm hoping that I won't need too much hand holding with something that would probably be relatively easy for you to figure out. Thanks again everyone.

AC

[Shawn]

If I'm understanding this correctly you are trying to use the tape monitoring input switching when trying to play live through vst plugins right?

Doesn't the tape monitoring/input switching add the dry input signal to the monitor signal? and if so, that would not be how I would want to monitor while playing guitar through a plugin because the dry signal would be straight through with no processing, while being mixed with the more latent effected guitar sound coming from the plugin, it probably would not sound good at all.

I would turn off the tape style input monitoring, then lower the buffer size to something at or under 128 samples if it will run stable and play/record that way.

:)

[Adam Christopher]

If I'm understanding this correctly you are trying to use the tape monitoring input switching when trying to play live through vst plugins right?

Doesn't the tape monitoring/input switching add the dry input signal to the monitor signal? and if so, that would not be how I would want to monitor while playing guitar through a plugin because the dry signal would be straight through with no processing, while being mixed with the more latent effected guitar sound coming from the plugin, it probably would not sound good at all.

I would turn off the tape style input monitoring, then lower the buffer size to something at or under 128 samples if it will run stable and play/record that way.

:)

Thanks Shawn. That makes sense and considered that might be the best option. Record a dry signal and then apply the vst plugin at the mixing stage. But then its not very fun to be able to play distorted rock guitar sims with a completely dry clean signal and not hear what its going to sound like when you are playing. There's got to be a way to play and hear the plugin at the same time and you can only hear with the tape monitoring engaged. I'm not using high latency plugins but I can hear the latency before plugins are already engaged in tape monitoring mode.

[Ian Alexander]

Maybe this has been mentioned, but some interfaces come with a mixer applet that will let you set up a direct monitor of your input signal mixed with playback. In my case with a Lynx card, the input signal still goes through A/D and D/A, but the delay is about 1 millisecond. No, you won't have reverb or whatever, but you'll hear it almost analog live. And you can leave your SawStudio buffer settings where you get no dropouts or hiccups. HTH.

[Adam Christopher]

Maybe this has been mentioned, but some interfaces come with a mixer applet that will let you set up a direct monitor of your input signal mixed with playback. In my case with a Lynx card, the input signal still goes through A/D and D/A, but the delay is about 1 millisecond. No, you won't have reverb or whatever, but you'll hear it almost analog live. And you can leave your SawStudio buffer settings where you get no dropouts or hiccups. HTH.

Thanks Ian. Yeah, the UR28 has that option for monitoring but it only works with Cubase and I'm definitely not going to be using that. I am considering going Lynx but if I got something like the AE16E, I think I'd have to buy a separate D/A converter (since that has no D/A) on top of a new preamp.

[Shawn]

Adam, How low can you set the buffer and get stable playback?

How are you routing your input?

Try this,

With SAW closed, open your audio device's control panel and set it's buffer size to 128 samples

Open SAW and in the options menu, turn OFF the auto audio monitor switching, then in the same menu turn on the auto record/srp latch, also in the same menu go to audio device setup and make sure that you are using 2 or 1 out preload buffers

In the mixer menu click on pre-fx patch signal flow, once the dialog opens, click on input 1, this moves that channel's pre fader fx patchpoint to be after the attenuator section of the channelstrip and before the EQ/DYN section, doing this also moves the record tap point with it, the record tap point is just after the pre fader fx patchpoint.

Insert your guitar amp plugin into the pre fader fx patchpoint on input channel 1

Arm track 2, when the record meter pops up, click on the input selection window, set it to take it's audio input from mixer input channel 1

Now click on the record ready button on the recording transport control panel

SAW's mixer should go "live" and you should hear only the effected guitar signal.

Hit the record button to record.

[Adam Christopher]

Adam, How low can you set the buffer and get stable playback?

How are you routing your input?

Try this,

With SAW closed, open your audio device's control panel and set it's buffer size to 128 samples

Open SAW and in the options menu, turn OFF the auto audio monitor switching, then in the same menu turn on the auto record/srp latch, also in the same menu go to audio device setup and make sure that you are using 2 or 1 out preload buffers

In the mixer menu click on pre-fx patch signal flow, once the dialog opens, click on input 1, this moves that channel's pre fader fx patchpoint to be after the attenuator section of the channelstrip and before the EQ/DYN section, doing this also moves the record tap point with it, the record tap point is just after the pre fader fx patchpoint.

Insert your guitar amp plugin into the pre fader fx patchpoint on input channel 1

Arm track 2, when the record meter pops up, click on the input selection window, set it to take it's audio input from mixer input channel 1

Now click on the record ready button on the recording transport control panel

SAW's mixer should go "live" and you should hear only the effected guitar signal.

Hit the record button to record.

Thanks Shawn. I never thought of this or tried any of this. I was using 128 with 1 buffer. I had little to non dropouts at 128 but frequent dropouts at 64. The latency was very noticeable in my setup at 128 but not at 64.


The Process I was using was setting up the guitar prefader patch on channel 1. Then I'd arm track 1 to a mono input set to Analog Input (on the Steinberg USB) 1 left channel.

I'll have to give this a shot. I never thought to use this auto record/srp latch option. I'll have to read up about what it actually does. I'm not sure I follow this part


Arm track 2, when the record meter pops up, click on the input selection window, set it to take it's audio input from mixer input channel 1

but I'll give it a shot tonight to see what it does and report back. Thanks again. whether or not this solves my issues, I have a feeling that I will get a different result.

[Ian Alexander]

Thanks Ian. Yeah, the UR28 has that option for monitoring but it only works with Cubase and I'm definitely not going to be using that. I am considering going Lynx but if I got something like the AE16E, I think I'd have to buy a separate D/A converter (since that has no D/A) on top of a new preamp.
You may be right, but check this out:
"Latency-free DSP-powered monitoring with one REV-X reverb and four Channel Strips with any DAW by using the latest dspMixFx technology"
That's from this page:
http://www.steinberg.net/en/products/audio_interfaces/ur_series/ur28m.html
It appears that you have to start the dspMixFx software from the Windows Start menu. I would not expect to be able to access it from within SS, but I can't access the Lynx mixer applet from within SS, either.

[Shawn]

Adam, I forgot to mention in my post above that you have to set input channel 1 to use your hardware audio input, do that at the top of the channelstrip, also remember to set the mono switch to use the correct side of your input device, left or right only.

[Shawn]

The srp/record latch just engages both functions at the same time.

[sebastiandybing]

What OS, cpu, and if you now it what motherboard type
are you using.
If you are running XP you should not have all these problems,
if you use win7 or 8 you really have to do all Bobs win7 tweeks,
if you use a very old comp. the newest usb cards dont support them
proberly, btw. RME has a guide with mb types there are supported.
Ofcause all new one are, but please tell us what hardware you are using.

Are you tacking with open or closed headphones ?,

Sebastian

[Adam Christopher]

What OS, cpu, and if you now it what motherboard type
are you using.
If you are running XP you should not have all these problems,
if you use win7 or 8 you really have to do all Bobs win7 tweeks,
if you use a very old comp. the newest usb cards dont support them
proberly, btw. RME has a guide with mb types there are supported.
Ofcause all new one are, but please tell us what hardware you are using.

Are you tacking with open or closed headphones ?,

Sebastian

WIN8 Sandy Bridge I5-2350, Acer motherboard

It's definitely a modern computer. But I haven't done any of the Windows 7 tweaks. I woudln't have that that it would make that much of a difference for recording latency. I hadn't found any slowness with my system but didn't know that this I'm using closed headphones. Apparently, I'm reading that the latency sucks on the Steinberg USB units according to what I'm reading, which are basically their most current offering. All of their newer products are USB not firewire.

[Adam Christopher]

Arm track 2, when the record meter pops up, click on the input selection window, set it to take it's audio input from mixer input channel 1

I was following everything up until this part. I have no problem arming channel one to the analog device. But every time I set up the input on channel 2 from mixer input channel 1 I get the following message:

Source Buffer UnderRun...Source Data Not Retrieved Fast Enough!

I tried changing the buffer settings to 512 with 4 buffers just to see if it wouldn't setup because the settings were too low and still didn't work. But regardless of that, whatever I am doing,
I can't actually hear the sound of the effects yet.

Edit:

I just tried a couple of other DAWS and I was able to get only 6 ms latency with Samplitude at 64 with no dropouts whatsoever and barely audible latency with Tracktion at the same settings. I'm thinking I must be setting up something wrong in SAW. Maybe I need to try this SAC but I don't even know how to set up SAC to hear the effects since there is no tape monitoring mode.




When recording from console channels assigned to device inputs, rather than from the device input directly, the record signal is taken from the just after the Pre-Patch point.

If you re-arrange the Pre Patch point to be in front of the Dyn and EQ section of the channel strip (do this in the Mixer Menu in the Pre-FX Patch Signal Flow option), then you can record flat and still use all the eq and dynamics and pst patches for monitoring.

This allows you to create a complete mix for the control room without affecting the recorded signal. Solos and mutes can also be used without affecting the recording.

Bob L

I just pulled this guy out of the archives. It sounds then like this is what we are trying to achieve. Now I think I understand Shawn why you were suggesting this process. If that's the case then I just need to figure out how to do it lol. It sounds like then it would be possible to record a dry signal without latency but still hear the effects processing it in real time (so to the player it sounds just like they are playing through the effect in real time). That's a neat concept if I understand the logic correct. Now I just need to figure out where I misstepped if that is the case.

[sebastiandybing]

There is one big difference between Saw
and many other DAWs and that is Saw dont
accept any kind of drop outs in the audio stream.
This is actually a good thing.

I am also using Reason witch now is 64bit and updated
to win8, in win8 Reason, Sac and Saw will run in to
low latency troubles sooner or later doing recording using
usb, I have not tryed firewire, to me firewire is dead meat.

Sac's engine will keep going even if there is problems with
the audio stream from the soundcard, thats a good thing
doing a live show, and you may think everything is ok, eventhough
your machinery is struggeling in basement.

Your comp. hardware is perfect for audio,
I am bit concern with the Steinberg sound card,
is it Yamaha who have made it,
Can you see what kind of asio version it support,
I think steinberg has made a new version3.
Is it possible for you to lent a rme babyface or a another
rme usb audiocard to test.

One thing that actually help me on a Win7
machine I had (and sold again) was to setting the bios
on the motherboard to 1 cpu.
Do you know how to enter the bios.

In the power settings make sure to set all
parameters to fuld speed, no usb management,
play with the grafic card settings, there is uselly
3 modes full, mid and energy mode, try all three.

At my day job I am working at a big theater where I also do
video work, we use Dataton watchout video servers.
Anyway what I want to tell is that the win7,8 tweeks paper
is even bigger than the Bob tweeks.
You may download the watchout manual, the tweeks
is in the back off the manual, they are a bit more
step by step than Bobs tweek paper is.

And remember to send all crasch reports to microsoft,
I bomb them with reports, and I do think its getting
better and better from each update.
Lets see maybe win8 will be really good for audio.

[Adam Christopher]

There is one big difference between Saw
and many other DAWs and that is Saw dont
accept any kind of drop outs in the audio stream.
This is actually a good thing.

I am also using Reason witch now is 64bit and updated
to win8, in win8 Reason, Sac and Saw will run in to
low latency troubles sooner or later doing recording using
usb, I have not tryed firewire, to me firewire is dead meat.

Sac's engine will keep going even if there is problems with
the audio stream from the soundcard, thats a good thing
doing a live show, and you may think everything is ok, eventhough
your machinery is struggeling in basement.

Your comp. hardware is perfect for audio,
I am bit concern with the Steinberg sound card,
is it Yamaha who have made it,
Can you see what kind of asio version it support,
I think steinberg has made a new version3.
Is it possible for you to lent a rme babyface or a another
rme usb audiocard to test.

One thing that actually help me on a Win7
machine I had (and sold again) was to setting the bios
on the motherboard to 1 cpu.
Do you know how to enter the bios.

In the power settings make sure to set all
parameters to fuld speed, no usb management,
play with the grafic card settings, there is uselly
3 modes full, mid and energy mode, try all three.

At my day job I am working at a big theater where I also do
video work, we use Dataton watchout video servers.
Anyway what I want to tell is that the win7,8 tweeks paper
is even bigger than the Bob tweeks.
You may download the watchout manual, the tweeks
is in the back off the manual, they are a bit more
step by step than Bobs tweek paper is.

And remember to send all crasch reports to microsoft,
I bomb them with reports, and I do think its getting
better and better from each update.
Lets see maybe win8 will be really good for audio.

Thanks Sebastian. Very helpful. I'll have to check out some tweaks but I don't really want to mess with the bios. I wonder if SAC would solve this for me. If so, I would probably bite on it even though I'm not a live sound guy but I would need to test it first I haven't figured out how to monitor the effects sound in SAC. By the way, last night coincidentally Steinberg made available their latest drivers on the site. They are slightly better but the latency lingers even a buffer setting of 64 with 1 or 2 buffers.

[sebastiandybing]

Well. Sac is a wonderful software, but
Bob has design it for low latency and to
do so it will not accept plugin cousing even
the smallest latency, so you could be unlucky
and your guitar plugs wont work in sac.
I my self has been using amplitube 2 and 3
with good luck in sac.

God luck.

[Adam Christopher]

Well. Sac is a wonderful software, but
Bob has design it for low latency and to
do so it will not accept plugin cousing even
the smallest latency, so you could be unlucky
and your guitar plugs wont work in sac.
I my self has been using amplitube 2 and 3
with good luck in sac.

God luck.

Checking this out more tonight. Thanks again Sebastian.

[Shawn]

Adam, check your pm's, give me a call, I'll walk you thru it.

[Adam Christopher]

Adam, check your pm's, give me a call, I'll walk you thru it.

Shawn...... Thanks so much!!!!! Bob actually diagnosed what my issue was. I'm actually going to address this as a separate post. Thanks again. I can't tell you how much I appreciate it!

[Adam Christopher]

Hi Guys,

I want to say thank you all so much for your suggestions. Bob figured out what my problem was. I gave him a quick call this evening to get some advice and he asked me to explain my situation. He immediately caught my words when I described the latency as a phasey sound. Then he pointed out that a combfiltering effect could happen by hearing the direct monitoring signal and the plugin sound at the same time.

Here is what happened. With the Steinberg unit, it was set up default for direct monitoring (at low levels). I needed to install the mixer software to mute the analog inputs and then I was able to hear just the sound coming into SAW! Problem solved.

I am on a Windows 8 system and now everything is running smooth as butter even without tweaks. I can get a very playable guitar sound anywhere from buffer settings of 64 to 256, although I haven't really found the optimal setting yet.

I feel kind of embarrassed that I made such a bonehead mistake. I probably would have caught that sooner if it wasn't set up that way by default with a low level direct sound in the background. It looks like the Steinberg unit was meant to be used in conjunction with the mixer software.

I want to say thank you all for the time and support. You folks are awesome. This was frustrating me to no end when I couldn't get things to work right and am glad I asked. I learned a lot from everyone in the process.

Thanks again Shawn, Sebastian, Matt, Angie, Jeff, Ira, Ian, and Tomasino.

[sebastiandybing]

Haha, how do you think we feel, I almost hope you
will run into a problem with win8 at some point.
Next time I will say call Bob ofcause.

Sebastian

[Carl G.]

Thanks Angie and Matt. No I don't have SAC. I'll have to look into other settings then to see if I get it to work but there's no probably if I have Tape Monitoring turned off. No latency there. The problem is when I am trying to record with plugins. That is encouraging to know that it works well with the 11rack USB. I can record fine with my Yamaha THR, which is also USB. But in those cases the sound is coming from the hardware not the plugins. I guess I'll have to play around with the settings. Thanks.

AC I'm after the same (as probably a lot of us) 'low latency' monitoring... but I believe with SawStudio in "Live" mode (to be able to use the VST's while recording/monitoring) you'd actually be far better off, like Angie says, by using SAC as your 'front end console' to SAWstudio. However, if you like just using SawStudio (which is OK too with the machine you have) you'll have to be very selective about the plugins used - because depending on those... you may need to relax your buffers (***increase your latency***) in order to record properly...(whereas with SAC.. you might get by with much lower buffer settings using the same plugs).

Why not just try the demo of SAC and use it as the front end to SAC and see how it works for you. (works better for me... but on my current machine ... latency is too high for my liking ... so I'm working on that change now). So.....I'd be interested in following your final best solutions!

[Adam Christopher]

AC I'm after the same (as probably a lot of us) 'low latency' monitoring... but I believe with SawStudio in "Live" mode (to be able to use the VST's while recording/monitoring) you'd actually be far better off, like Angie says, by using SAC as your 'front end console' to SAWstudio. However, if you like just using SawStudio (which is OK too with the machine you have) you'll have to be very selective about the plugins used - because depending on those... you may need to relax your buffers (***increase your latency***) in order to record properly...(whereas with SAC.. you might get by with much lower buffer settings using the same plugs).

Why not just try the demo of SAC and use it as the front end to SAC and see how it works for you. (works better for me... but on my current machine ... latency is too high for my liking ... so I'm working on that change now). So.....I'd be interested in following your final best solutions!

Thanks Carl for the helpful suggestions. I'm actually pleased enough with what I am getting from my guitar plugins now. I know this turned out to be a long thread with troubleshooting but I finally figured out my main issue was that I was getting phasing because I was silly enough not to notice that there was a low level direct monitoring signal coming from the interface while SAW was playing. Bob suggested that this may be the case and he ended up being right. I never installed the Steinberg mixing software beforehand. But when I downloaded it, I saw that I could mute the signal coming from the interface and did so and after hearing just the DAW signal in SAW, the phasing issue was gone.

I actually tried demoing SAC to see this difference. But somehow couldn't get sound out of it. I went over this with Bob and he noted that in all his years he hadn't seen any kind of weird issues with getting sound from SAC. I know it must have something to do with my interface set up. There was a monitor signal going into the input but nothing coming out of Ouput 1 (the Master Output). Must have something to do with muting the signal from the direct monitoring mode must not be good enough. But I ended up moving on from not getting SAC to work because I was pleased enough with what I was getting out of my SAW setup.

I probably have more modest needs than a lot of the pros here on the SAW forums so it isn't exactly real-time monitoring that I was after. It was mostly that phasey sound that I was mistaking for unusable latency.

Both Sebastian (beforehand and thanks to Sebastian again for the help) and Bob suggested I do the Windows 7 tweaks as well in Windows 8 but I am pleased enough with how SAW is running for me.

I'm still playing around with the buffers and I can get ok latency at 96 to 128. 64 is not yet stable for me. But I am find a number of combinations now usable whereas they weren't before.

If I do move on though to another or more real-time solution that works, I will be sure to update.

[Carl G.]

Hi Guys,

Bob .... pointed out that a combfiltering effect could happen by hearing the direct monitoring signal and the plugin sound at the same time.
.....was set up default for direct monitoring (at low levels). I needed to install the mixer software to mute the analog inputs and then I was able to hear just the sound coming into SAW! Problem solved.

I feel kind of embarrassed that I made such a bonehead mistake.

Great catch! "Bonehead" mistake... FUNNY play on words! Because now with your corrections... any 'phase' latency you hear now (mostly noticeable when Voice Tracking SRP in Live Mode) will be just that - "BoneHead" (all through the bones in your head!.... inner-ear bone conduction phasing difference from what you hear coming in from headphones against what you are hearing from your own vocal cords in to your inner ear).

My goal is to get that to a minimum by using buffer settings of 32x1.
Not there yet... so still processing externally with external hardware console monitoring.

But thanks to your thread... some great advice is brought forward! (Dataton Watchout manual Win8 tweaks for instance)

[Adam Christopher]

Great catch! "Bonehead" mistake... FUNNY play on words! Because now with your corrections... any 'phase' latency you hear now (mostly noticeable when Voice Tracking SRP in Live Mode) will be just that - "BoneHead" (all through the bones in your head!.... inner-ear bone conduction phasing difference from what you hear coming in from headphones against what you are hearing from your own vocal cords in to your inner ear).

My goal is to get that to a minimum by using buffer settings of 32x1.
Not there yet... so still processing externally with external hardware console monitoring.

But thanks to your thread... some great advice is brought forward! (Dataton Watchout manual Win8 tweaks for instance)

Haha, that made me laugh Carl. Thanks for that! But I think you're giving me a little too much credit. I usually leave the funny stuff to guys with the witty comments like yourself or a Davey Labrecque.

The whole bone thing reminds me of freshman year in college in bio class where my prof mentioned that women on average have better hearing because of physics - more optimized bone density for hearing.

I'm very glad that more than just myself got something out of this what turned out to be longer thread. I can't tell you how frustrating it was hearing that phasey sound and how relieved I was once I got everything up and running just the way I wanted.

For anybody that has read this far, I found my optimal setting for me. Buffer size of 96 with 1 buffer. 64 is a bit unstable and I can hear some latency at 128 but 96 for me is the ticket. I'm not looking for perfection but usuable recording latency for guitar and 96 works for me. Obviously 32 or better for Carl and co might be a must for real professional work.

Oh and completely OT for Carl. I asked around from more of my techie buddies and any research and he told me for SSDs Intel, Samsung, and Crucial are the best for reliability. But Samsungs latest generation is more like neck and neck with Intel for reliability/compatibility and Crucial is a bit behind the other two but still very reliable. Intel will certainly work just fine but you'll pay a premium for a little bit less hd space. My buddy said it's good to have some extra hd space though for headroom because on any brand there is always the possibility of write errors and the more headroom you have, the less prone you will be to have write errors eating into performance over time. I read similar claims on all those topics on the net but anyways I'm only going off what I heard/read. Firsthand experience from someone you know (your super smart son for instance) is always the best reference. Good luck!

[AcousticGlue]

It was just recently that I pointed out for like the 4th time to Echoaudio about Layla24 that I wanted to NOT hear the guitar and the plugin. They had replied to mute the input on the console. This fixed it for me. Now that being said I can do 128/2 in SAW and not hear latency. Used to be I thought I was hearing latency. The In and Out has to be set same. 128 is the defacto standard for plugins. It was explained that ASIO uses hardware to determine the latency set. So yes I can drop to 64/2 and it is fine 98% time. If I set to 64 in SAW or Reaper it works fine with Amplitube3. Latency wise. It was explained on web that 128 on most PCs equates to 5ms latency. Drops to only 4ms at 64. How long does it take to blink?

Note: :Later Explorer versions that 6 in SAW have some kind of issues that has been being experimented with in SAW. So if you update to say version 8 there are some things that start to happen. Plugins with presets that go from folder to folder will get buffer underruns at times. Clicking on it and continuing fixes the issue.

What I would like to see in SAW and SAC is a config section to point VST to so it pulls from directories used rather than editing a file.

[Carl G.]

Firsthand experience from someone you know (your super smart son for instance) is always the best reference.
The reason I'm on the forum is because of the synergistic intelligence gathering and wisdom found here. :)
(even though 7 super smart sons & Brilliant daughter)

I'll have to look at that Samsung. Thanks for the info. I assume SSD's don't except a write error.. and then rewrite to a portion of mem that is good... (a question I forgot to ask).

[Adam Christopher]

The reason I'm on the forum is because of the synergistic intelligence gathering and wisdom found here. :)
(even though 7 super smart sons & Brilliant daughter)

I'll have to look at that Samsung. Thanks for the info. I assume SSD's don't except a write error.. and then rewrite to a portion of mem that is good... (a question I forgot to ask).

Yes, that is what I was told that it rewrites to a portion of the memory that is good, which is why having a bit of headroom can be good.

Also just came across this thread now that may be helpful. I think the samsung runs a little faster than the intel and has more memory for less but depending on the model, the intel can have a three-year or five-year warranty, which is a big draw for some. It's nice that they don't have a virtual monopoly on ssds though like they do with processors.

http://www.overclock.net/t/1295493/which-ssd-intel-vs-crucial-vs-samsung

Very cool.... I see you are blessed with a very intelligent and big family. The more the merrier as they say;

[Adam Christopher]

It was just recently that I pointed out for like the 4th time to Echoaudio about Layla24 that I wanted to NOT hear the guitar and the plugin. They had replied to mute the input on the console. This fixed it for me. Now that being said I can do 128/2 in SAW and not hear latency. Used to be I thought I was hearing latency. The In and Out has to be set same. 128 is the defacto standard for plugins. It was explained that ASIO uses hardware to determine the latency set. So yes I can drop to 64/2 and it is fine 98% time. If I set to 64 in SAW or Reaper it works fine with Amplitube3. Latency wise. It was explained on web that 128 on most PCs equates to 5ms latency. Drops to only 4ms at 64. How long does it take to blink?

Note: :Later Explorer versions that 6 in SAW have some kind of issues that has been being experimented with in SAW. So if you update to say version 8 there are some things that start to happen. Plugins with presets that go from folder to folder will get buffer underruns at times. Clicking on it and continuing fixes the issue.

What I would like to see in SAW and SAC is a config section to point VST to so it pulls from directories used rather than editing a file.

Yes, that was the fix for me. Mute the inputs! I still can't get exactly what I like out of 128/1, but 96/1 does it for me I know that like you said 128 is supposed to be 5ms on most pcs but I think those things are bound to vary depending on the computers and drivers.