CPU usage and 96khz Questions about your system stability

[fishmonkey]

Yes it will be completely different between system thats why Im asking. I was just curious about how high people could go since it tells me something about the quality of your audio drivers and system and how it compares to my system, no?

It seem people mistake me for a noob looking at my post count and want to set me straight. I know I can higher the buffer setting but then you get a higher latency. Im used to hardware synths connected to analog mixer so this latency is very annoying to me thats why Im trying to work at lowest buffer as possible.

Thanks for the first answer to my question. I thought most sound card could test at 64 buffer maybe thats another reason I don't get any answers.

Here is my system. Since I said someone with a less system and and old Apollo could go much higher with the cpu usage simulator thats why I was curious. Still I can use much more plugins than him but he can use much more of his cpu before crackles at this low buffer than I could. The cpu usage simulator tells you something right or why is it there? An actual setup at 80% would be hard to compare.

building a Windows box from off-the-shelf parts and overclocking the CPU does not make you a computer expert. i know you think you understand how it all works, but most of what you have written in this thread indicates that you lack a firm grasp of the fundamentals of how computers and digital audio interfaces work and interact with each other. nor do you understand what the Ableton Live CPU meter and CPU usage simulator actually do, and what information they can and can't provide you.

your statements about acoustic latency are off the mark too. for starters, a concert grand piano is about 3 metres long.

[hearteh]

Unbelievable really. I don't expect to get any answers at this point. I've moved on so this thread can be closed.
I am done talking to hurt egos that think they know much more about things than they actually do.

Just for the record before I overclocked my system was much worse and Ive build computers for the last 25 years and Windows is super optimized for audio production. I've also got pro help from the uad that have confirmed there's nothing wrong with my system. The crackles are not in the recording only in monitoring.
The problem I've suspected all along is the uad drivers and thunderbolt ex3. But I didn't ask for help did I? I was hoping for fast answers to simple question about your system it was not a scientific test. I took for granted people knew about certain things.

I've already stated why I asked at 64 buffer.

[fishmonkey]

Unbelievable really. I don't expect to get any answers at this point. I've moved on so this thread can be closed.
I am done talking to hurt egos that think they know much more about things than they actually do.

Just for the record before I overclocked my system was much worse and Ive build computers for the last 25 years and Windows is super optimized for audio production.

I've already stated why I asked at 64 buffer. I didn't ask to get help I asked to get answers about your system.

https://www.youtube.com/watch?v=GiPe1OiKQuk

[hearteh]

I understand all this very clearly it's you that don't understand this maybe you will some day. Now at least you found the cpu usage simulator.

Yes it will be completely different between system thats why Im asking. I was just curious about how high people could go since it tells me something about the quality of your audio drivers and system and how it compares to my system, no?

It seem people mistake me for a noob looking at my post count and want to set me straight. I know I can higher the buffer setting but then you get a higher latency. Im used to hardware synths connected to analog mixer so this latency is very annoying to me thats why Im trying to work at lowest buffer as possible.

Thanks for the first answer to my question. I thought most sound card could test at 64 buffer maybe thats another reason I don't get any answers.

Here is my system. Since I said someone with a less system and and old Apollo could go much higher with the cpu usage simulator thats why I was curious. Still I can use much more plugins than him but he can use much more of his cpu before crackles at this low buffer than I could. The cpu usage simulator tells you something right or why is it there? An actual setup at 80% would be hard to compare.

building a Windows box from off-the-shelf parts and overclocking the CPU does not make you a computer expert. i know you think you understand how it all works, but most of what you have written in this thread indicates that you lack a firm grasp of the fundamentals of how computers and digital audio interfaces work and interact with each other. nor do you understand what the Ableton Live CPU meter and CPU usage simulator actually do, and what information they can and can't provide you.

your statements about acoustic latency are off the mark too. for starters, a concert grand piano is about 3 metres long.

[hearteh]

Do you understand that using hardware will add even more latency going to the input ad? I use both hardware and plugins, I can take advantage of the uad console for the output latency though. But Im not here asking help for setting things up.

Again look at my original question if you can't answer that just leave this thread.

[fishmonkey]

Again look at my original question if you can't answer that just leave this thread.

your original question is poorly conceived, and based on a fundamental lack of understanding about what you are measuring and how you are trying to measure it.

do you know exactly how the CPU usage simulator works? i sure as hell don't, since Ableton hasn't published a description of it anywhere (not one that i can find anyway). what i can see is that it generates a synthetic CPU load on a single CPU core, in a way that matches up with what the Live "CPU meter" measures. the thing is, the Live "CPU meter" is not strictly speaking CPU meter at all — it is a rough measure of Live's ability to process a whole audio buffer in the time available before a glitch occurs. it is far from a precise tool, and you will find many questions on the interwebs from people who are confused about why the Live "CPU meter" doesn't match up with their system CPU meters, and also querying the unpredictable relationship between the Live "CPU meter" and audio glitches.

also, statements like "have a super fast PC", and "Windows is super optimized for audio production" are so vague that they are meaningless in this context. you have jumped to the conclusion that it is the UAD drivers that are the problem, but what is this assumption based on? people regularly complain about their "super fast PC" under-performing because they confuse raw power with rock solid audio processing. maybe it is the UAD drivers, but it might well also be one or more issues with your computer system. for starters, have you run DPC latency tests on your computer? if your machine appears to be glitching out too readily then that is an important troubleshooting step.

btw, your rudeness and personal attacks are unwarranted and reflect poorly upon you as a person.

[hearteh]

Show me where I made personal attacks? You are the rude one, keeps telling my lack of understanding.

Yes Ive run a dpc meter test and I have no problem running audio. Max out at 200us after long time running.
The crackles is not in the recording only in the monitoring. What did I just write about staying on topic? I don't have time for this. The UAD have already told me theres nothing wrong with my pc but could be inherent to windows processing on a PCI bus device and may go away over time.

Of course an optimized windows system will matter in these situations.

Ive already stated it wasn't a scientific test. In real word I can use less cpu than what the simulator shows but its still pretty close. The Ableton cpu meter is of course not the same as the actual cpu. My actual cpu usage is nowhere near the Ableton cpu meter but again I asked how high people could with cpu usage simulator not the actual cpu.

Again look at my original question if you can't answer that just leave this thread.

your original question is poorly conceived, and based on a fundamental lack of understanding about what you are measuring and how you are trying to measure it.

do you know exactly how the CPU usage simulator works? i sure as hell don't, since Ableton hasn't published a description of it anywhere (not one that i can find anyway). what i can see is that it generates a synthetic CPU load on a single CPU core, in a way that matches up with what the Live "CPU meter" measures. the thing is, the Live "CPU meter" is not strictly speaking CPU meter at all — it is a rough measure of Live's ability to process a whole audio buffer in the time available before a glitch occurs. it is far from a precise tool, and you will find many questions on the interwebs from people who are confused about why the Live "CPU meter" doesn't match up with their system CPU meters, and also querying the unpredictable relationship between the Live "CPU meter" and audio glitches.

also, statements like "have a super fast PC", and "Windows is super optimized for audio production" are so vague that they are meaningless in this context. you have jumped to the conclusion that it is the UAD drivers that are the problem, but what is this assumption based on? people regularly complain about their "super fast PC" under-performing because they confuse raw power with rock solid audio processing. maybe it is the UAD drivers, but it might well also be one or more issues with your computer system. for starters, have you run DPC latency tests on your computer? if your machine appears to be glitching out too readily then that is an important troubleshooting step.

btw, your rudeness and personal attacks are unwarranted and reflect poorly upon you as a person.

[fishmonkey]

Show me where I made personal attacks? You are the rude one, keeps telling my lack of understanding.

oh, right, am i meant to pretend that you do understand?

Yes Ive run a dpc meter test and I have no problem running audio. Max out at 200us after long time running.

huh? this whole thread is about you having a problem running audio.

one 64 sample audio buffer @ 96 kHz is only 0.67 ms. note that multiple driver latencies can sometimes stack together, causing an audio processing gltich.

The crackles is not in the recording only in the monitoring. What did I just write about staying on topic? I don't have time for this. The UAD have already told me theres nothing wrong with my pc but could be inherent to windows processing on a PCI bus device and may go away over time.

UAD tech support told you that the problem "may go away over time"? that doesn't make sense.

what is the lowest latency that UAD says the Apollo x6 can achieve on a Windows machine?

[hearteh]

Show me where I made personal attacks? You are the rude one, keeps telling my lack of understanding.

oh, right, am i meant to pretend that you do understand?

Yes Ive run a dpc meter test and I have no problem running audio. Max out at 200us after long time running.

huh? this whole thread is about you having a problem running audio.

one 64 sample audio buffer @ 96 kHz is only 0.67 ms. note that multiple driver latencies can sometimes stack together, causing an audio processing gltich.

The crackles is not in the recording only in the monitoring. What did I just write about staying on topic? I don't have time for this. The UAD have already told me theres nothing wrong with my pc but could be inherent to windows processing on a PCI bus device and may go away over time.

UAD tech support told you that the problem "may go away over time"? that doesn't make sense.

what is the lowest latency that UAD says the Apollo x6 can achieve on a Windows machine?

This thread was actually more about my curiosity how much better MAC is than PC in this department.
Actually that didn't make sense to me either. They don't state this.
This was my max dpc Im usually under 100us.

[fishmonkey]

if the duration of one sample buffer is only 0.67 ms, with the CPU usage simulator pegged at 80% the processing leeway is only about 130 microseconds. so your system would need to have very stable, very low DPC latencies to not exhibit glitches.

don't forget that the DPC latencies can stack. for example, two consecutive DPC stalls of 100 microseconds will effectively create a 200 microsecond stall.

[[jur]]

Hey there,
please take a deep breath and be cool to each other.
Thanks!!!

[hearteh]

if the duration of one sample buffer is only 0.67 ms, with the CPU usage simulator pegged at 80% the processing leeway is only about 130 microseconds. so your system would need to have very stable, very low DPC latencies to not exhibit glitches.

don't forget that the DPC latencies can stack. for example, two consecutive DPC stalls of 100 microseconds will effectively create a 200 microsecond stall.

My average dpc is 67ms running 15 minutes. I just tried running 15 minutes after using my on board graphic instead of nivida. Still no difference I get crackles all the time at 57% CPU. In real situations more like 50% CPU which is around 5% of my real CPU.

[fishmonkey]

if the duration of one sample buffer is only 0.67 ms, with the CPU usage simulator pegged at 80% the processing leeway is only about 130 microseconds. so your system would need to have very stable, very low DPC latencies to not exhibit glitches.

don't forget that the DPC latencies can stack. for example, two consecutive DPC stalls of 100 microseconds will effectively create a 200 microsecond stall.

My average dpc is 67ms running 15 minutes. I just tried running 15 minutes after using my on board graphic instead of nivida. Still no difference I get crackles all the time at 57% CPU. In real situations more like 50% CPU which is around 5% of my real CPU.

are you saying that you've positively ruled out DPC spikes as causing the glitches? i.e. you've had a DPC latency tester running while the audio has been glitching?

[hearteh]

I think its strange I get crackles all the time otherwise no? I was under 5us for some time and still I get crackles. Yes I run at the same time.
I haven't ruled out it fully maybe something I'm missing.

if the duration of one sample buffer is only 0.67 ms, with the CPU usage simulator pegged at 80% the processing leeway is only about 130 microseconds. so your system would need to have very stable, very low DPC latencies to not exhibit glitches.

don't forget that the DPC latencies can stack. for example, two consecutive DPC stalls of 100 microseconds will effectively create a 200 microsecond stall.

My average dpc is 67ms running 15 minutes. I just tried running 15 minutes after using my on board graphic instead of nivida. Still no difference I get crackles all the time at 57% CPU. In real situations more like 50% CPU which is around 5% of my real CPU.

are you saying that you've positively ruled out DPC spikes as causing the glitches? i.e. you've had a DPC latency tester running while the audio has been glitching?

[wearemindflux]

Moderation edit:
Useless comment removed. The mood is calmed down, no need to fire the shit up again