Skip to main content

What is a sound card?

Do you still need a sound card for your PC?

A sound card allows computers to have sound. Pretty simple, right? But let’s dig deeper. Here’s a closer look at the tech that defines a sound card, and what to know if you want to buy one.

A brief history of sound cards

Image used with permission by copyright holder

Today’s sound cards are hardware rectangles that plug into motherboards via PCI (or are completely integrated, a.k.a. onboard), then connect to speakers and mics, managing the sound capabilities of the computer.

The first computers did not have sound cards—they weren’t considered necessary for the basic tasks that computers were designed to perform. Instead, early devices had basic internal speakers that could produce square wave audio — those “beeps” and “boops” that everyone associates with clunky, first-wave computers.

As computers grew more complex and started entering the consumer market in the 1980s, manufacturers quickly realized that they needed better ways of creating sound, especially for advanced applications and general entertainment purposes. IBM and other manufacturers turned toward manufacturers like Adlib and Creative Labs, which just happened to be working on new sound card technology to move beyond the blips and instead replicate music, voices, and more.

By the late 1980s, computers started hitting the market with built-in sound cards. At the beginning, these sound cards focused on very specific applications. They were created for music composition, or speech synthesis, or (increasingly) specific computer games. Over time, sound cards gained more versatility, and were soon working across many different kinds of software.

Basic functions

Image used with permission by copyright holder

Audio files on a computer are, like everything else, stored as code. That digital information can easily store a lot of sound waveforms, but it can’t create sound—those “analog” waves that need to spread through the air and impact our ears. The sound card translates audio from digital code to the sound waves as needed.

To do this, the card uses a DAC, or digital to analog converter. The converter’s job is to translate the audio file code into electrical impulses, which travel via the sound card’s connections to speakers. The speaker’s drivers turn the electrical impulse into physical sound waves, and the rest is up to our ears. All speakers, internal or external, must be connected to the sound card to work properly.

However, sound cards also have another very important function. They have to do the same thing in reverse. If your computer has a mic (and nearly all do these days), then it too is routed through the sound card. Here, cards use an ADC, or analog to digital converter, that translates sound waves created by your voice into code that becomes an audio file.

Sound cards can also have additional functions, such as serving as a MIDI interface for those who want to create a little electronica. Today’s sound cards are usually streamlined and highly integrated to cut down on costs (with software drivers managing extra features), but some versions still have such built-in capabilities.

Voices and channels

Image used with permission by copyright holder

Sound cards have both voices and channels, and this can be a little confusing. Let’s carefully define both these features.

Voices: Voices refer to how many independent sounds from different sources a sound card can manage at the same time. When your computer is playing an login melody but also dings when a new email comes in, that’s using two voices. Manufacturers have always been able to give sound cards plenty of voices to use from both hardware and software sources, primarily because one of the first purposes of sound cards was to help create electronic music. So early sound cards typically had either 9 or 18 voices. The numbers quickly grew until the average sound card often had 64 voices, 32 available from software and 32 available from hardware. Modern sound cards pay less attention to hardware sources and focus on software that can produce as many voices as needed, so it’s less common to rate a sound card by voices these days.

Channels: Sometimes people will use the term “channels” to mean the same thing as “voices.” Technically, channels should be used in the traditional sense, which is how many audio outputs the sound card can handle. Now we are on more familiar ground. Stereo sound has two channels, 2.1 stereo allows for a subwoofer, 5.1 channels include surround sound, and 7.1 channels provide the best surround sound. Importantly, you’ll need your sound card to support at least as many channels as the audio system you pair it with.

Upgrading sound cards

A sound card can typically be replaced with a different card, which is nice for repairs and especially handy if you want to upgrade your sound card to a better version. Sound hardware integrated into a motherboard can’t be replaced, but it can be disabled, letting you switch to a better PCI sound card.

Advanced sound cards can improve audio, help add more clarity to digital sound, or use processing power of their own to help lighten the CPU load. There are also outboard sound cards that can be connected via a USB port for more active sound management. You can purchase both internal sound cards and peripheral versions, as long as you are sure they are compatible with your computer.

While sound cards can be expensive, basic models are sufficient for most people. Popular options include the Asus Xonar and Creative Sound Blaster X-Fi series. It’s even possible to enhance the sound capabilities of a laptop with a USB sound card like the Creative Go! sand Creative Play! series.

Editors' Recommendations

Tyler Lacoma
Former Digital Trends Contributor
If it can be streamed, voice-activated, made better with an app, or beaten by mashing buttons, Tyler's into it. When he's not…
These are the best graphics cards to use in an external GPU enclosure
The Razer Core X Chroma external graphics card on a desk next a laptop and a monitor.

Using an external graphics card is a great way to boost your laptop's graphics power for gaming or professional 3D work, but buying the best eGPU enclosure is only half the battle. You need to buy the graphics card, too, and it's not as straightforward as just buying the best graphics card you can afford. As powerful as Thunderbolt 3 and 4 are, they're not fast enough for the fastest of cards.

But that doesn't mean you can't have a great experience using external graphics. Here are the best graphics cards to use as eGPUs.

Read more
Intel surprise launched a new graphics card, but it doesn’t make any sense
The Intel logo on the Arc A770 graphics card.

Intel's Arc A580 graphics card is finally here, but we're not sure if it's ready to compete against some of the best GPUs out right now. In fact, based on the reviews, it is ill-prepared to face the current GPU market. What went wrong? It's not even about the performance, even though it leaves a lot to be desired. It's mostly that it seems this GPU makes little sense to buy compared to the competition, including Intel's own offerings.

It's been a bumpy road for the Intel Arc A580. The rest of the Arc Alchemist lineup, including the flagship Arc A770, launched a year ago after multiple delays. The Arc A580 was announced alongside them, but then, Intel went quiet. Many people thought the card was quietly canceled, but then it resurfaced a few days ago -- and now, it's officially out, with a $180 price tag. We also know the full specs of the GPU.

Read more
There’s only one use for an RTX 4080 Ti, and it’s not what you think
Three RTX 4080 cards sitting on a pink background.

According to the latest round of leaks, Nvidia may be readying an RTX 4080 Ti, set to come out in the first few months of 2024. While such a GPU would definitely rank high among the best graphics cards, it's not exactly necessary right now -- there are plenty of high-end GPUs floating around. There is one reason I would like to see an RTX 4080 Ti, though; to push down the price of the RTX 4080.

MEGAsizeGPU, who is a frequent speculator in the graphics card space, shared over the weekend that an RTX 4080 Super, or maybe Ti, will come out in just a few months. Given Nvidia's current naming convention in this generation, a Ti card seems more likely. As per MEGAsizeGPU, the card will be based on Nvidia's flagship AD102 chip and will have a power consumption below 450 watts. The other specs are a mystery, but I wouldn't be surprised if Nvidia added some more VRAM to the card -- 20GB seems likely. The bump from AD103 to AD102 should also deliver a decent upgrade in CUDA core counts -- dare I say in the 12,000 to 13,000 range?

Read more