I have never tried but why not.
I don't use AVERLAB for acoustic measurement, only for electric measurement as it is designed for THD+N, SNR, Crosstalk and Level measurements.
But considering the generator output to amplifier and input from a microphone you can make all measurements I have already mentioned.
Now to say that AVERLAB is the best device for this function, I recommend you to contact directly AVERMETRIC or may be Paul will answer you with more accurate details and capabilities.
Regards
Arrg! I just spent 20 minutes writing a long response to this thread and the system had already logged me out! I'll try again...
For some reason I usually don't get email notifications of posts on this site, but sometimes I do. Ahh, the mysteries of the online world.
We've been feverishly working away on the software and firmware for AverLAB, which is my excuse for being radio silent for a while. That's my story and I'm sticking to it, anyway.
This new version, AverLAB 3, is in beta, and has changes major and minor all through the program. It has a new UI, with a single main window containing resizable panes for Project, Test, Sequence, and Results. These panes can remain docked in the main window or "pulled out" and positioned anywhere on the screen or on other desktops and monitors, as desired. Other improvements include faster measurements, better repeatability, A-wtg, C-wtg, HP, LP filters available on all measurements (except FFT, of course). In addition, the software now supports "Virtual Devices" where you can connect multiple AverLAB units and combine them into a single multi-channel device to get as many channels as you need in a single system.
There are a lot of changes under the hood, and now that it's almost done this new architecture allows us to add features to the program much more easily, so you'll start to see new features on a regular basis. As always, these updates are no charge.
If you have an AverLAB and are not already in the beta program, send me a message at beta@avermetrics.com and I'll give you access. That way you can test some of the new features and provide feedback on what works and doesn't, and how you'd like to see AverLAB work in the future. The more feedback we get, the better the program will be.
Best regards to all,
Paul
For some reason I usually don't get email notifications of posts on this site, but sometimes I do. Ahh, the mysteries of the online world.
We've been feverishly working away on the software and firmware for AverLAB, which is my excuse for being radio silent for a while. That's my story and I'm sticking to it, anyway.
This new version, AverLAB 3, is in beta, and has changes major and minor all through the program. It has a new UI, with a single main window containing resizable panes for Project, Test, Sequence, and Results. These panes can remain docked in the main window or "pulled out" and positioned anywhere on the screen or on other desktops and monitors, as desired. Other improvements include faster measurements, better repeatability, A-wtg, C-wtg, HP, LP filters available on all measurements (except FFT, of course). In addition, the software now supports "Virtual Devices" where you can connect multiple AverLAB units and combine them into a single multi-channel device to get as many channels as you need in a single system.
There are a lot of changes under the hood, and now that it's almost done this new architecture allows us to add features to the program much more easily, so you'll start to see new features on a regular basis. As always, these updates are no charge.
If you have an AverLAB and are not already in the beta program, send me a message at beta@avermetrics.com and I'll give you access. That way you can test some of the new features and provide feedback on what works and doesn't, and how you'd like to see AverLAB work in the future. The more feedback we get, the better the program will be.
Best regards to all,
Paul
We haven't added features specific to acoustic measurements yet, but you can indeed do basic FR and THDN measurements for acoustics using AverLAB. You may need an output amplifier, and perhaps a mic preamp for best results.
Single-ended, AverLAB's output impedance is 25 ohms. The output driver is not designed to supply power, and combined with the relatively high output Z you may need an outboard amp to drive low impedance transducers. The inputs have up to 54 dB of gain, but you may find you need more for best results. With an output power amp and input preamp AverLAB will work well for these basic measurements.
Paul
Dear Paul,
I brought the Averlab along time and waiting the digital test for I2S and jitter test.
When will finish this function?
Also I want to know can the Averlab use as CRO to view the wave?
Hi Schmike,
We're working as fast as we can to add all the features to the AverLAB software that the hardware is capable of supporting, but unfortunately things always take more time than you think the should.
I've taken over the software development myself since we lost our developer. Understanding someone else's code is a slow process. It's well-written under the hood, but it's taken me months of work to understand it--crawling into someone else's brainspace is a scary thing to do! But we're getting close to a new release later this month, and this version gives us a platform to add all of these features, and lots more.
I really appreciate your patience, and I realize it's been a while waiting for these things. I hope you're able to stick with us, and know that we're adding features as fast as possible.
P
We're working as fast as we can to add all the features to the AverLAB software that the hardware is capable of supporting, but unfortunately things always take more time than you think the should.
I've taken over the software development myself since we lost our developer. Understanding someone else's code is a slow process. It's well-written under the hood, but it's taken me months of work to understand it--crawling into someone else's brainspace is a scary thing to do! But we're getting close to a new release later this month, and this version gives us a platform to add all of these features, and lots more.
I really appreciate your patience, and I realize it's been a while waiting for these things. I hope you're able to stick with us, and know that we're adding features as fast as possible.
P
Dear Paul
I hope you can do that.
Now the features still less than the keysight DSOX-1102g.
And keysight only cost usd1100.
DSOX1102G can read the waveform, frequency response with phase shift. Also can read the I2S.
all features is same as Averlab and additional can read the waveform because it is CRO.
I own the averlab 18 months and still waiting the updates.
Now a day the C/P value of the Averlab still very low.
Hope I can wait the full features of it before I forget this mechines.
You can’t compare an oscilloscope with an Audio analyzer. They are different tools designed for different works. Yes of course you can find some common function but be careful with the performance versus what you need.
If you want to compare AVERLAB, take examples like DSCOPE, AP or other Audio analyzer around the same price.
The case of the DSOX-1102g is a modern, basic and expensive oscilloscope with poor performance: THD -40db, 8bits resolution, poor generator BW…. Nothing impressing for that price. You can find much better oscilloscope!
Regards
Dear pdouillard,
I am the APX515 owner and find the apx515 cannot test the I2S.
I find the Averlab can test the i2s.
But I own the Averlab 18 months and the function still stay in the basic level.
At the same time I find the scope can get the same basic function with Averlab which can have sweep gen for frequency, FFT and waveform.
So I disappointed on it which I read some news said Averlab is the AP killer.
Hi Schmike,
I understand your frustration. FWIW, I'm frustrated too that it's taken us so long to make progress. But I feel like we're finally getting somewhere, and I'm looking forward to getting AverLAB 3.0 out so we can start adding some of these features.
pdouillard is right that DSO are useful for some tasks, and not very good tools for others. I had a quick look at the specs for the scope you mentioned, and they are typical for these LCD-based scopes. You can find better features in a DSO at 1/2 the cost with some of the Hantek and Rigol systems, but none of them are useful for audio, beyond "signal present" uses. They all have 8-bit ADCs, so audio measurement performance will be very poor. I'm not sure which scope you are looking at, but the Keysight one can decode I2C (as an upgrade), but not I2S. But maybe I missed something in the datasheet...
I hope we'll be able to deliver the features you want quickly. We're certainly working hard to try and do that.
P
I understand your frustration. FWIW, I'm frustrated too that it's taken us so long to make progress. But I feel like we're finally getting somewhere, and I'm looking forward to getting AverLAB 3.0 out so we can start adding some of these features.
pdouillard is right that DSO are useful for some tasks, and not very good tools for others. I had a quick look at the specs for the scope you mentioned, and they are typical for these LCD-based scopes. You can find better features in a DSO at 1/2 the cost with some of the Hantek and Rigol systems, but none of them are useful for audio, beyond "signal present" uses. They all have 8-bit ADCs, so audio measurement performance will be very poor. I'm not sure which scope you are looking at, but the Keysight one can decode I2C (as an upgrade), but not I2S. But maybe I missed something in the datasheet...
I hope we'll be able to deliver the features you want quickly. We're certainly working hard to try and do that.
P
Dear Paul,
It just for some basic feature as Averlab compare with Averlab.
Now the Averlab can test Frequency Response, FFT, Distortion and crosstalk only.
Just same as the Scope is not useful with audio.
As the same time I am using the APX515 so I get very disappointed on the Averlab which is the AP killer.
I am looking forward to the growth of the Averlab because I have paid for it.
Hope I can see the useful Audio Analyzer on 2020.
Impulse response, phase is the importance test on audio.
Hope I can use on 2020.
Hi Schmike,
I'm sorry you are disappointed. I notice that you unsubscribed from the beta email list after the announcement for beta version CL1494, so you may not have seen any of the work we've been doing over the last 6 months. I'd be glad to resubscribe you to the beta. Please contact me directly at paul@avermetrics.com and I'll get you setup again.
We've made a huge number of improvements--the program is very different from AverLAB 2.2, with a lot of new features. Please have a look at the download page for the beta version to see the list of changes. I've been releasing a new version every week or two.
If there are specific features you are looking for, please contact me directly so I can make sure they are on the upcoming feature list.
Paul
I'm sorry you are disappointed. I notice that you unsubscribed from the beta email list after the announcement for beta version CL1494, so you may not have seen any of the work we've been doing over the last 6 months. I'd be glad to resubscribe you to the beta. Please contact me directly at paul@avermetrics.com and I'll get you setup again.
We've made a huge number of improvements--the program is very different from AverLAB 2.2, with a lot of new features. Please have a look at the download page for the beta version to see the list of changes. I've been releasing a new version every week or two.
If there are specific features you are looking for, please contact me directly so I can make sure they are on the upcoming feature list.
Paul
Dear Paul,
I think I still in the list who can download the beta version.
I have checked the beta version before the last message I post.
I can download the CL1643.
The UI is different but the feature is still too basic too.
I think jitter test is most importance which cab beat the APX515.
Hope I cab see the better feature on Averlab.
I think I still in the list who can download the beta version.
I have checked the beta version before the last message I post.
I can download the CL1643.
The UI is different but the feature is still too basic too.
I think jitter test is most importance which cab beat the APX515.
Hope I cab see the better feature on Averlab.
Last edited by a moderator:
We continue to look at the best way to measure jitter to see if we can implement a useful jitter measurement in AverLAB. The problem with indirect jitter measurements such as those used by AP is that the residual jitter is very poor. The AP's jitter measurement has a residual on the order of 500ps RMS or more. This is of very little value for modern digital systems. Even the lowest cost modern audio interface available now has jitter better than 200ps RMS. Unless you can measure well below the jitter of your DUT, having the feature is not useful.
To measure our clock recovery PLL, I use an HP8568B and some software. This system measures RMS jitter by measuring the phase noise of the clock, then calculating the RMS integral of the phase noise over the desired bandwidth. I use a custom GPS disciplined oscillator as the timebase to achieve a residual jitter of <10ps. Our PLL typically measures around 16ps RMS jitter from 100Hz to 40Khz bandwidth. There are dedicated jitter measurement systems that can measure jitter directly, but they are very expensive (on the order of $90,000USD to buy, $5000/week to rent.) AverLAB was not designed to measure jitter. We get very few requests for this feature. What we've found in the market is that modern crystal oscillators, clock recovery PLLs, DIR's, and codecs have low enough jitter for most products, using the reference designs for these parts. So many designers no longer feel the pressure to measure jitter like they did years ago. That said, state of the art clock generators still need to be measured in order to achieve the best performance, so we're very interested in the feature.
Paul
To measure our clock recovery PLL, I use an HP8568B and some software. This system measures RMS jitter by measuring the phase noise of the clock, then calculating the RMS integral of the phase noise over the desired bandwidth. I use a custom GPS disciplined oscillator as the timebase to achieve a residual jitter of <10ps. Our PLL typically measures around 16ps RMS jitter from 100Hz to 40Khz bandwidth. There are dedicated jitter measurement systems that can measure jitter directly, but they are very expensive (on the order of $90,000USD to buy, $5000/week to rent.) AverLAB was not designed to measure jitter. We get very few requests for this feature. What we've found in the market is that modern crystal oscillators, clock recovery PLLs, DIR's, and codecs have low enough jitter for most products, using the reference designs for these parts. So many designers no longer feel the pressure to measure jitter like they did years ago. That said, state of the art clock generators still need to be measured in order to achieve the best performance, so we're very interested in the feature.
Paul
I have found little use for direct jitter measurements. What is important are the jitter artifacts in the audio output and those can be much easier to measure. (If you want I do have an HP 5370 that can measure cycle to cycle jitter to 20 pS.) The Meitner analyzer was a good model for a jitter analyzer using FFT and looking at sidebands of a carrier and identifying the sidebands helps to identify the aggressors causing jitter.
Another seperate hardware solution uses a PLL (like you have) and a phase detector to catch the phase modulation on the clocks. I have a design for one from ultra-analog which could be adapted.
Still jitter issues today are either a defective part or a real design failing. FWIW the standard SPDIF receiver from AKM is good for less than 20 pS just following the reference design.
Another seperate hardware solution uses a PLL (like you have) and a phase detector to catch the phase modulation on the clocks. I have a design for one from ultra-analog which could be adapted.
Still jitter issues today are either a defective part or a real design failing. FWIW the standard SPDIF receiver from AKM is good for less than 20 pS just following the reference design.
Hi Demian,
You're exactly right. The FFT sideband method is a standard indirect way of measuring jitter. It's very good at measuring relatively higher amounts of jitter, but starts to fall apart with lower level jitter, since in that case the sidebands may be at or below the system noise floor. The PLL version has much to recommend it--it's effectively a synchronous detector with lots of processing gain. It's an analog to the downconverter/filter/detector/integrator method, which is what measuring phase noise with an SA boils down to. Using the PLL for the measurement might be how we would implement this in a future product--but the AverLAB hardware is not setup for this.
Your comment pretty much sums up the state of jitter in the modern age: "FWIW the standard SPDIF receiver from AKM is good for less than 20 pS just following the reference design." For many applications the issue is inherently solved. Now we have opamps that are nearly perfect, PLLs that are nearly perfect, codecs with specs well beyond both the upstream and downstream source material and audio chain performance, and so on. But we still need to verify and validate designs, so there is some need still for measurements like jitter. But it didn't make sense to add the cost for this to the AverLAB product, at least in its current implementation.
Best,
Paul
You're exactly right. The FFT sideband method is a standard indirect way of measuring jitter. It's very good at measuring relatively higher amounts of jitter, but starts to fall apart with lower level jitter, since in that case the sidebands may be at or below the system noise floor. The PLL version has much to recommend it--it's effectively a synchronous detector with lots of processing gain. It's an analog to the downconverter/filter/detector/integrator method, which is what measuring phase noise with an SA boils down to. Using the PLL for the measurement might be how we would implement this in a future product--but the AverLAB hardware is not setup for this.
Your comment pretty much sums up the state of jitter in the modern age: "FWIW the standard SPDIF receiver from AKM is good for less than 20 pS just following the reference design." For many applications the issue is inherently solved. Now we have opamps that are nearly perfect, PLLs that are nearly perfect, codecs with specs well beyond both the upstream and downstream source material and audio chain performance, and so on. But we still need to verify and validate designs, so there is some need still for measurements like jitter. But it didn't make sense to add the cost for this to the AverLAB product, at least in its current implementation.
Best,
Paul
A separate box using a PLL and maybe a VCXO could provide a source for jitter analysis. The frequencies needed are limited and all you need is a known analog out to feed into the analyzer. The analyzer could process the demodulated baseband for useful results. I even persuaded a Yamaha tuner to do this and it worked pretty well, even with a Wenzel to test its floor. I shared the tricks but those seem to have vanished some DIYaudio upgrades ago. Unless your going after ultra low phase noise stuff it would mostly be generic parts.
For SOTA work there is a reason a timepod is so expensive.
For SOTA work there is a reason a timepod is so expensive.
I'm sorry to read this - and am surprised that there aren't more comments. I guess that the market for Aver's product simply isn't large enough to pay for the kind of R+D expense that's needed to keep up with the lower-priced AP offerings?
As well as having a 2722, I have UPVs and a Dscope. I don't think it's a coincidence that R+S haven't updated the design functionally in over 15 years (you can change to Win-7 and there are a few utitlity upgrades, but that's about it to my knowledge). The newer Prism offering doesn't appear to have a vast amount of features that do not exist on the Dscope 3, which has been around for even longer than the UPV.
From where I'm sitting, it would appear that AP have the market pretty much sewn up in all sectors. I'll be surprised if R+S release another flagship (and we should also remember that the analogue performance (B1 oscillator fitted) - is no better than previous R+S models from the 90s).
As well as having a 2722, I have UPVs and a Dscope. I don't think it's a coincidence that R+S haven't updated the design functionally in over 15 years (you can change to Win-7 and there are a few utitlity upgrades, but that's about it to my knowledge). The newer Prism offering doesn't appear to have a vast amount of features that do not exist on the Dscope 3, which has been around for even longer than the UPV.
From where I'm sitting, it would appear that AP have the market pretty much sewn up in all sectors. I'll be surprised if R+S release another flagship (and we should also remember that the analogue performance (B1 oscillator fitted) - is no better than previous R+S models from the 90s).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.