Jump to content

Latency during playback in Cubase 5


jack259

Recommended Posts

I'd not heard of that before! That's very cool. It might help a bit yeh, i'm not sure how it works. Normally you'd need a 64bit cpu to handle any more than 3.5Gb of RAM. If you have a 64bit cpu and OS why not get some more internal RAM put in? Either way I think your project is just too intensive for your current cpu.

 

If you're on Windows, plug in your memory stick, right click < properties, ReadyBoost.

 

I don't think you can use it for file storage whilst it's being used for ReadyBoost, so it's best if you've got a spare. Luckily I have a 4GB one doing nothing.

 

Edit:

I'm blessed with 3GB of RAM, and my OS is 32bit (Windows 7 Pro). Which is good considering the laptop is over 2 years old and cost around £400. Obviously it came with Vista, I upgraded for £40~ around 14 months ago.

Link to comment
Share on other sites

It will definitely be your soundcard, if it's stock it will be useless at recording.

If you're recording at school on a desktop the chances are it's got a decent card in it.

 

Things like freezing tracks (which you definitely can do in Cubase 5 (it's the little snowflake icon on the track box) and having good virtual memory will help during playback but it's all about the ASIO device.

Also, do NOT have any inserts or effects running during recording. Only add them during mixing.

 

If you've said I haven't noticed, what latency time are you getting on the device menu?

 

Also, what bit rate, sample rate and latency sample number are you using? (Your soundcard may not let you adjust the latter)

However you can definitely set the bit and sample rates. If you're having trouble set them as low as possible without compromising quality 44,100Hz, 16 bit. Increasing them will sap your power and raise your latency exponentially.

Having them lower than that will help again, but sound increasingly poor.

 

As for fixing the problem with an external USB device, it's all a question of how much you're willing to spend. Also check out the make first, because some are real pigs when it comes to Windows 7.

Link to comment
Share on other sites

It will definitely be your soundcard, if it's stock it will be useless at recording.

If you're recording at school on a desktop the chances are it's got a decent card in it.

 

Things like freezing tracks (which you definitely can do in Cubase 5 (it's the little snowflake icon on the track box) and having good virtual memory will help during playback but it's all about the ASIO device.

Also, do NOT have any inserts or effects running during recording. Only add them during mixing.

 

If you've said I haven't noticed, what latency time are you getting on the device menu?

 

Also, what bit rate, sample rate and latency sample number are you using? (Your soundcard may not let you adjust the latter)

However you can definitely set the bit and sample rates. If you're having trouble set them as low as possible without compromising quality 44,100Hz, 16 bit. Increasing them will sap your power and raise your latency exponentially.

Having them lower than that will help again, but sound increasingly poor.

 

As for fixing the problem with an external USB device, it's all a question of how much you're willing to spend. Also check out the make first, because some are real pigs when it comes to Windows 7.

 

Yeah I know it wouldn't be much kop at recording. I only use my laptop for mixing/VST projects. The studio computer has an excellent sound card.

 

Bit Rate 24 bit (not sure how to change this)

Sample Rate: 44.100kHz

I think the latency sample number is 2048.

Link to comment
Share on other sites

Yeah I know it wouldn't be much kop at recording. I only use my laptop for mixing/VST projects. The studio computer has an excellent sound card.

 

Bit Rate 24 bit (not sure how to change this)

Sample Rate: 44.100kHz

I think the latency sample number is 2048.

 

To lower the bit rate, go to the drop down project menu and select project setup, then you can change the bit-rate to 16. Cubase will ask if you want to downsample the parts you've already recorded (this is fine, and you can keep the 24bit work you've already done as well).

You'll probably know this, but the quality drop isn't too bad as 44.1kHz 16bit is standard CD quality.

 

Just remember that if you ever took the project back to school, while you can upsample back to 24bit, the quality will not be as high as it originally was.

 

The latency number is interesting because it should be giving you the best performance - in terms of playback.

For example, I normally try to record at 48 samples and look at the latency difference between that and 2048

sample.jpg

 

So yeah, without upgrading the soundcard, freezing tracks is probably your best option. VST instruments, for example, can hog beastly amounts of power, so if you're happy with the sound you can export the individual track as a .wav file, then import it again. That way you get the sound you want, but without your computer having to process the instrument in realtime.

 

*edit - regarding dual core/quad core computers, I could be wrong but I'm pretty sure Cubase 5 isn't optimised to use multiple cores (unlike pro tools I believe) So, dual core is good because one can handle Windows and another core, Cubase, but quad core would just mean two go to waste.

Link to comment
Share on other sites

You'll probably know this, but the quality drop isn't too bad as 44.1kHz 16bit is standard CD quality.

 

The difference is the noise floor of 16 bit is around -90dB and 24 bit around -120dB, unless you listen to music at dangerously high volumes, there's not a major difference in basic terms.

 

The difference comes when you start mixing, adding dynamics processing and so on, as that brings the noise floor up, so having more dynamic range in the first place does make a bit of a difference.

 

 

 

Edit - You only need to worry about latency if you're actually playing along to something or recording a live musician. If you're just mixing, whack the latency right up and ease the load on the cpu.

Link to comment
Share on other sites

To lower the bit rate, go to the drop down project menu and select project setup, then you can change the bit-rate to 16. Cubase will ask if you want to downsample the parts you've already recorded (this is fine, and you can keep the 24bit work you've already done as well).

You'll probably know this, but the quality drop isn't too bad as 44.1kHz 16bit is standard CD quality.

 

Just remember that if you ever took the project back to school, while you can upsample back to 24bit, the quality will not be as high as it originally was.

 

The latency number is interesting because it should be giving you the best performance - in terms of playback.

For example, I normally try to record at 48 samples and look at the latency difference between that and 2048

sample.jpg

 

So yeah, without upgrading the soundcard, freezing tracks is probably your best option. VST instruments, for example, can hog beastly amounts of power, so if you're happy with the sound you can export the individual track as a .wav file, then import it again. That way you get the sound you want, but without your computer having to process the instrument in realtime.

 

*edit - regarding dual core/quad core computers, I could be wrong but I'm pretty sure Cubase 5 isn't optimised to use multiple cores (unlike pro tools I believe) So, dual core is good because one can handle Windows and another core, Cubase, but quad core would just mean two go to waste.

 

I tried that - still some latency in there.

 

I think I'll just deal with it and stay late after school to remix the song. I've done several mixes/masters but I'm never happy.

 

I think it may just be coincidence then that the studio computer is quad core then

Link to comment
Share on other sites

It will definitely be your soundcard, if it's stock it will be useless at recording.

 

He's not recording. In my experience, a shit soundcard is fine for mixing the most intensive projects, so long as you have sufficient CPU power.

 

*edit - regarding dual core/quad core computers, I could be wrong but I'm pretty sure Cubase 5 isn't optimised to use multiple cores (unlike pro tools I believe) So, dual core is good because one can handle Windows and another core, Cubase, but quad core would just mean two go to waste.

 

Where did you get that impression? I'm interested as i'd like to optimize my next machine for cubase.

 

 

The difference is the noise floor of 16 bit is around -90dB and 24 bit around -120dB, unless you listen to music at dangerously high volumes, there's not a major difference in basic terms.

 

The difference comes when you start mixing, adding dynamics processing and so on, as that brings the noise floor up, so having more dynamic range in the first place does make a bit of a difference.

 

Edit - You only need to worry about latency if you're actually playing along to something or recording a live musician. If you're just mixing, whack the latency right up and ease the load on the cpu.

 

All good advice IMO. You shouldn't record audio at less than 24bit. I wouldn't really know about the set up you need for VST instruments etc? And yeh, whack the latency right up!!

Link to comment
Share on other sites

He's not recording. In my experience, a shit soundcard is fine for mixing the most intensive projects, so long as you have sufficient CPU power.

 

 

Where did you get that impression? I'm interested as i'd like to optimize my next machine for cubase.

 

 

All good advice IMO. You shouldn't record audio at less than 24bit. I wouldn't really know about the set up you need for VST instruments etc? And yeh, whack the latency right up!!

 

He's not recording, but he said he's getting popping/distortion etc. That's usually an overburdened system or a conflict (one of my m-audio cards refuses to get on with Windows 7). Since we can assume it's power in this case and upgrading a laptop is a pain, an external soundcard would help reduce the processing load on the system.

 

I just know that a lot of programs were slow in adapting to make the best use of multi-core systems, many not using more than two. However, Cubase 5 does have a multi processing option and I think Steinberg have been releasing patches so it takes full advantage of quad or octo core systems, so Cubase 6 and up definitely will (even free Reaper does). So If you're building from scratch, get the best processor you can.

I just didn't think he should exclusively blame his lack of quad core, since audio processing would be better helped by a soundcard, in my opinion.

 

As for recording quality, of course I agree that ideally you'd record in the best available but you also have to be realistic in regards to the equipment you're using and I would definitely say 16 bit is an acceptable minimum in these circumstances.

Link to comment
Share on other sites

He's not recording, but he said he's getting popping/distortion etc. That's usually an overburdened system or a conflict (one of my m-audio cards refuses to get on with Windows 7). Since we can assume it's power in this case and upgrading a laptop is a pain, an external soundcard would help reduce the processing load on the system.

 

so much much is the soundcard involved in? I thought it'd only really help if it had some sort of on board DSP which handles certain VSTis or VST plugins.

Link to comment
Share on other sites

All good advice IMO. You shouldn't record audio at less than 24bit. I wouldn't really know about the set up you need for VST instruments etc? And yeh, whack the latency right up!!

 

is it important to have your ("theoretical") noisefloor that low when analog stuff (preamps, instruments, mics etc.) is noisier than that anyways?

Link to comment
Share on other sites

so much much is the soundcard involved in? I thought it'd only really help if it had some sort of on board DSP which handles certain VSTis or VST plugins.

 

Yeah, the majority of processing is dumped on the system, his seems capable enough if it's running windows 7 with 3gb of ram. It's difficult to know what he's running without seeing his project. A decent dedicated soundcard should help run VSTs and reduce the drain on the system though, especially compared to a laptop's onboard one.

Link to comment
Share on other sites

DPTP - yeh. I like to record with plenty of headroom to avoid digital clipping at all costs. Recording at 24bit enables that.

 

Noodles - in what way would a sound card help with VSTs? Most soundcards simply provide DA AD, inputs and outputs, and MIDI. To take processing load off the CPU you generally have to use a seperate card (can be external) which is just designed for plug-ins IE Universal Audio AUD PCI cards etc.

 

But if you know of a soundcard that handles VST processing...?

Link to comment
Share on other sites

I'm confused, you ask how a sound card can help with processing, then say you can get additional cards which can help. I think we have crossed wires, because these are the ones I'm talking about. Brands like EMU, SM audio, Manifold Labs and UA make hardware that can help process VSTs, I'm pretty sure UA even do a card for laptops.

 

I'm just trying to approach the problem from his individual point of view (ideal set-ups and settings being irrelevant), eliminating all the other options which is why I mentioned inserts, bit rates, freezing and exporting first. I've never had to fix this problem on a laptop before.

Link to comment
Share on other sites

soundcards have nothing to do with VSTs, thy don't have onboard CPUs or anything, the only platform which helps with running VSTs is the UAD stuff (but it works only for their own plugins) i think and maybe protools have something but i know nothing bout protools, there are no general VST running devices except CPUs.

 

(when you see a rise of performance from changing your soundcard it is because the driver, genereal windows drivers should handle a millions of shit soundcards with orgasmic3d technologies and other bullshit where better soundcards have proper drivers which can focus only on one thing. except creative audigy stuff where the drivers are basically just there for trolling the poor customers)

 

also RAM doesn't count until it is not full, harddrive speed counts a really little bit (a few percents of the overall performance), the most important thing is CPU power.

Link to comment
Share on other sites

is it important to have your ("theoretical") noisefloor that low when analog stuff (preamps' date=' instruments, mics etc.) is noisier than that anyways?[/quote']

 

Not really in all honesty, it's far from essential to record in 24 over 16 bit as a lot of types of music rarely have the dynamic range to take advantage of it.

 

The main reason to do it is just to have as high a quality recording as you can, as you can always lower the quality later, but impossible to upgrade it.

Link to comment
Share on other sites

I disagree. I think recording in 16 bit is the reason most kids are taught to record as hot a signal as possible, which often leads to clipping. I was taught at uni that the "best" signal is as high gain as possible, in the context of digital recording. I realise that's total bollocks now.

 

I'm not tsaying there's anything wrong with recording to 16bit, but if it's possible, always record in 24bit.

Link to comment
Share on other sites

I disagree. I think recording in 16 bit is the reason most kids are taught to record as hot a signal as possible, which often leads to clipping. I was taught at uni that the "best" signal is as high gain as possible, in the context of digital recording. I realise that's total bollocks now.

 

I'm not tsaying there's anything wrong with recording to 16bit, but if it's possible, always record in 24bit.

 

The as hot as possible thing is a hangover from using tape, nothing to do with 16 bit at all. As I pointed out earlier, 16bit/44.1kHz has like 90dB of dynamic range, which is plenty 99.9% of the time!

 

But yeah, always best to record as high quality as you can, so if you have the options of using 24bit, then definitely do it, no harm will come of it.

Link to comment
Share on other sites

Jack I don't a soundcard is going to solve your problem. You have the facility to "freeze" tracks - just use that.

 

Haze - Yeh, you're probably right that it's a hangover from tape. I'm probably just bullshitting to justify my opinions. I wonder if it's because I use lower quality gear that I get more noise when recording at 16bit? My soundcards are M-Audio Deltas which I thought were pretty good. I do have a lot of shitty home made cables but surely they'd be just as noisey for 24bit?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...