I was under the impression the Hz (so 48000 in this case) referred to SNR
Signal-to-noise ratio (dB) is in fact not measured in Hertz (sampling frequency).
1 Like
You know, when I think about it, that makes perfect sense 
1 Like
For speakers on a laptop, I would suggest that they’re not large enough to provide the detail of sound required to provide much, if any, noticeable difference at higher bit rates than 48KHZ.
As far as the headphone jack goes, if anyone has their hands on Framework’s audio expansion card, how does this compare to the DAC currently provided on the Framework 13? A side by side comparison of some specifications by Framework themselves might be useful to have, also.
1 Like
Well, for the actual laptop speakers, I think the FW13 does perfectly fine. They have descent sound profile and their volume is impressive.
My main concern here is the DAC itself. The 3.5 jack doesn’t really matter, it’s just a connector. (Though 3.5 mm jacks suck but I digress on that point) The controller itself is the issue here.
Granted, most audio doesn’t exceed 16bit 48000 HZ - thus, the audio is good enough. But for me, I would like better than ‘good enough’. 24 bit 48000Hz is kind of my minimum. But the higher the better (mainly for nerdy reasons)
Another issue I have with it is the display. 8-bit SDR is not really going to work for video editing/ color grading. Granted, the display looks amazing for a basic 8-bit panel. I would just like to see something that handles color with more finesse.
As for the audio, Realtek is okay. Just okay. Their GUI has the bare minimum of options that I end up never using because there are better programs out there. It would be awesome to see something more creator oriented. Something from Beakn for example (the team the made the GOXLR) working with them could potentially add the ability to route audio to different sources easily, create groups (like one for streaming, one for media, etc) and control those groups independently - basically like a virtual sound board. That would be super cool.
It’s not that the FW doesn’t work, or doesn’t get the job done. It most certainly does everything I need it to - I would just like to see some improvements. I have “audiophile” headphones and 16bit audio is like pairing an RTX 4090 with an intel 7700K
A lot of sound quality on tiny speakers these days comes from equalizing or otherwise processing the audio signal. I do not know how you would do this on Windows, but I’m sure there are tools to achieve things similar to this: https://community.frame.work/t/guide-framework-dsp-better-linux-audio/44645/1
The result is quite impressive for me. I thought the speakers were nice enough, but this makes them sound a good bit better and there is probably more to do if one wants to tinker.
Yeah, that’s not what I was getting at. It comes with realtek installed which has an okay EQ you can play with. And I’m not at all saying the audio quality is bad. Honestly, it’s pretty impressive for what it is. I was just suggesting an option for a sound card that supports higher quality audio.
It’s more a use case thing. Like, I wouldn’t recommend doing video/picture editing on the FW display. Can you? Sure. Will it look descent? Probably. But it’s not ideal for that use case. honestly I don’t even like playing games on 8-bit panels. One might say, “then just go get a gaming monitor and use that” which is a good option. But it would be nice to have an integrated one.
Same deal for the audio. I could go buy an external DAC or just use my phone/ desktop. But it would be NICE to have an integrated option with some cool features that might help content creators or just give people something to play with.
Don’t get me wrong, I’m not dogging on framework here, I love this laptop and wouldn’t get anything else right now. Just would like to see a little more in some areas
The thing is, for general listening, 16 bit, 48KHz is technically enough. This is an oversimplification, but the bit depth primarily affects the dynamic range that can be contained in the file. With audio, that’s generally going to be the difference between the loudest and the quietest sounds that can be played without getting into noise on the low end or clipping on the top end. 16 bit is said to be able to produce approximately 96db of dynamic range. That would allow for a 96db difference between the loudest and quietest parts of a track. That’s plenty for final audio. If you are recording and mastering audio, higher bit depths can be useful to allow more headroom during recording and more freedom when mixing and editing without running into issues. But for final listening, even if you had your music turned up so that the loudest parts were at 116db, the quiets parts of the audio could still be as low as 20db before disappearing into the noise floor. And at that point, the noise floor of whatever amp was pushing 116db of sound would likely be higher than that. So in terms of delivering final audio (vs. recording and mastering), 16 bit should be enough.
A rule of thumb for sample rate (the 48KHz number) is that it takes a sample rate at roughly twice the frequency of the sound you are trying to reproduce in order to properly reproduce that sound. So a 48KHz sample rate should be able to accurately reproduce any sound 24KHz and lower. Since an adult generally cannot hear anything above 22KHz or so (Many say the average is much less than this), the “rule of thumb” is that 44KHz is enough for any sounds within the range of human hearing. Again, mixing and mastering are a bit different. If you are trying to adjust the tone or pitch of a sound, a higher sample rate is beneficial so you don’t get artifacts. But for playback, it shouldn’t be an issue.
This is why the standard for CD audio was 16bit, 44.1KHz. The standard for movies is 16bit, 48KHz.
If you were going to record audio for mixing and mastering, a good starting point is 24bit, 96KHz. As long as you have an audio interface capable of this and a program capable of working with that audio, you’re all set. Being limited to 16bit, 48000 Hz from your speakers and headphones, really isn’t a big deal.
Granted, there are plenty of audiophiles who would dispute these claims. And maybe some of them have a point. But in blind, A-B comparisons, I doubt the majority of people would be able to tell a difference when listening back to audio at 16bit, 48KHz, vs. 24bit, 96KHz, or higher.
When it comes to content creators, the display I can see being a potential limitation, if someone wants 10 bit color for editing, though I don’t see it as a huge limitation myself. But for audio, unless you are recording via the 3.5mm mic input jack (or with the webcam speaker), then the built-in audio card/codec isn’t a limitation. The computer itself is more than capable of working with 32bit float, 196 KHz files. Assuming you have a way to actually record them and a software program that can work with them.
6 Likes
Yeah, got that from your previous posts. Just wanted to share that there is some potential left on the table.
Sorry, I wasn’t meaning to imply otherwise.
The Realtek drivers are okay, and Dolby access is a good option to beef it up a little more too
This is true. But I’m not arguing that the 16 bit isn’t sufficient. Basically 16bit 48khz meets the requirements for good audio. However, as someone that does to video/audio editing and enjoys listening to music in the best quality possible, I would like to see a 24-bit option. Especially since they use a dedicated sound card, not a chip soldered onto the MB. Using my desktop for an example again (which is the primary rig for recording, editing, and all the stuffs) it does 24 bit 192khz. I notice a slight difference between 24 bit 48khz and 96khz. Maybe it’s just placebo but I swear the audio just sounds cleaner. Definitely notice it from 16 bit to 25 bit.
I’ll reiterate, I’m not complaining and saying the FW has bad audio. It would just be nice if there was a hi-fi/ audiophile option
If you’re talking about the headphone jack built into the FW13, the audio DAC chip for that is on the motherboard. The fact that the headphone (+ lid sensor) board is called an “Audio Board Kit” makes it sound like the DAC would be there, but the mainboard schematic shows it located on the mainboard, rather than on the daughterboard.
The Audio expansion card is available for better range and SNR. It uses a Conexant CX31993. I read it’s a well regarded DAC. And I’m seeing it listed as 32 bit / 384kHz PCM, with 24 bit / 96kHz recording.
1 Like
A couple options:
The audio chip is unfortunately on the mainboard directly. The separate board called the audio daughter board actually only has the headphone jack and a sensor for sensing when the lid is closed. That means that to upgrade that chip you’d need a whole new mainboard, which would be expensive. From your comment history it looks like you already have the AMD mainboard, which is already the better audio chip. However a future mainboard could potentially have better audio.
The audio expansion card has a better chip internally, the Conexant CX31993, so you could get one of those and see if it is as good as you want. Otherwise the only option remaining is an external DAC.
1 Like
Is the audio expansion card better than most USBC headsets?
The issue is the audio controller. If you get the expansion card (I might be wrong, as I have not tested it myself) the audio would still be precessing through the laptops internal chip. It’s possible it bypasses it when you use a 3.5mm headset (like a USB DAC?) but if you use BT, USB, etc, it still goes through the laptops chip. So you won’t get any better audio quality. To be fair the 13s audio sounds good as is, I just think there is room for improvement in the future
Actually, the audio expansion card is a USB DAC. So the audio “processing” happens in the audio expansion card, not in the internal audio chip. The audio expansion card is reported to produce higher quality audio (better signal to noise ratio) than the built-in DAC. I don’t know if it’s significantly better, but it is claimed to be an improvement.
Ah cool, that answers that question. It still wouldn’t work for Bluetooth or USB, right? As they wouldn’t be connected to the expansion card? Or does the expansion card bypass the internal chip entirely?
Correct, it wouldn’t affect anything connected to USB or bluetooth. It’s just a USB DAC for whatever is plugged into the headphone jack on the card.
Gotcha. I’m a little disappointed the controller isn’t part of the module in the laptop. If it was, then they could just make a higher quality option in the future and it would be a simple upgrade. It should be possible but maybe it just takes too much on the cost and r&d side to separate it from the MB?
Well, Bluetooth quality isn’t really down to any audio chip or anything in the computer. It’s possible to have software signal processing applied to the digital signal before it is transmitted over Bluetooth, but other than that, the audio quality is going to come down to the codecs and transmission quality of the digital signal via the Bluetooth radio, and the quality of the radio and DAC in the headphones.
The same goes for anything plugged into USB. Any audio sent to any device that is plugged into a USB port is sent as-is, in digital form. From there, it’s going to be down to whatever hardware and software is in whatever device is connected.
The built-in audio chip/controller on the laptop only applies to the built-in speakers and the built-in headphone jack (if there is one). When it comes to USB audio and Bluetooth audio, the source files are just sent, digitally, to whatever device is on the other end, and the audio conversion and processing is all done at the other end.
That was extremely informative. I thought it had to be processed through the DAC no matter what connection type is used. It really won’t matter in my use case then because I prefer to use BT with my laptop anyway. All of my video editing and what not gets done on the desktop so as long as I can enjoy media with the laptop, I’m good. I would still like to see better audio support in the future but that’s just me being a nerd about it
2 Likes