I'm not sure about the latency since I can't even measure it but I'd go to USB. Being tied to a PCI(e) slot is an awful nuisance these days. Many machines have multiple USB busses, try putting the sound card on a bus without much else if there's a problem. I'm not sure if your application is closer to a language lab or speech recognition. Startech is generally a good company, you can also get these cheaper and faster through Amazon. https://www.startech.com/en-us/search?search_term=usb%20sound I also have a few Chinese clones bought through Aliexpress that seem to work fine for general playback and recording. Preamps aren't just gain but supply operating power to electret microphones. I do amateur nature recordings with home-built 4 transistor preamps and generic electret capsules, maybe $5 parts cost total. On 6/14/21, Maarten Duijndam via Alsa-user <alsa-user@xxxxxxxxxxxxxxxxxxxxx> wrote: > Dear alsa users, > > We are looking for a replacement card for PCI(e) audio that we are using > in our linguistics laboratory. We have been quite happy with the older > ASUS Xonar D2 pci and ASUS Essence STX II PCIe cards. Unfortunately, > these are out of stock in the Netherlands, I don't know whether they are > still being made. We need quite a number of those devices in roughly > 15-25 pc's in our labs. So they were good value for money at roughly 200 > euro or dollars each. The xonars worked out of the box on recent > versions of Ubuntu and this great to have so we do not have to make > small adjustments to the all pc's in the lab. > > Some background on psychology/linguistics research. We are looking for a > sound card that is easy to use, and without many "expert" features. For > quality speech recordings we are looking for devices that has a high > signal to noise ratio (in practice we used an analog mic preamp and put > the preamps output in the line out of the preamp in the line in of the > xonar cards). Also we like to record as naturally as possible and we > don't value Digital signal processors (DSPs) that mimics special preamps. > Additionally, cards using DSPs might have an internal buffer in which > they do there processing and take additional latency. In artificial > grammar learning or other types of experiments in which speech stimuli > are presented to participant in our lab we greatly value the timing of > the presentation of those speech stimuli. That is: we try to schedule a > stimulus (speech or another sound) with 1ms accuracy. So If we know that > ALSA or another backend uses a eg. 20ms playback/software buffer, we > schedule the stimulus 20 ms ahead of time. We would like to keep any > buffering delays as short as reasonably possible. > > So coming back to the original question, the Xonars we were working with > ran fine in for example Ubuntu-18.04 and even some older distributions. > We would like to find a nice replacement that meets the requirements of > a linguistics lab and are relatively easy to maintain. I don't know if > this is the right place to ask, but I hope to get some input for we have > tried quite a number of PCI(e) devices from expensive RME equipment to > cheaper Creative Sound BlasterX AE-5 Plus or very cheap TERRATEC Aureon > devices. Perhaps we should move to USB class 1 and/or 2 compliant devices. > > Hopefully some of you have experience with such equipment and is willing > to give us a nice suggestion. > > Best regards and thank you for your time, > > Maarten > > > > _______________________________________________ > Alsa-user mailing list > Alsa-user@xxxxxxxxxxxxxxxxxxxxx > https://lists.sourceforge.net/lists/listinfo/alsa-user > -- ------------- Education is contagious. _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
