Choosing the right studio PC / What to look for in an audio system for recording

Where with other types of audio work we tend to focus on the overall available performance, when recording and processing live audio we have the additional latency concerns in mind. Common advice from many software firms has always been to focus on single core performance, over generally slower higher core count setups and sound recording is the scenario where this rings the truest.

When recording through the audio interface the interface driver captures the incoming audio within the ASIO buffer time frame that you set and then passes it to the CPU for the processing to be carried out at regular intervals. You can raise this setting to increase the overall performance, but also raise the monitoring latency you hear or more crucially to our needs you can also lower it to achieve less latency but this will put an increased load on the processor overall.

When a track is armed for recording within a sequencer, it will work to keep entire channel including its processing chain on the same processing thread. Working this way is more efficient than splitting the load and then recombining it again, so you will tend to find that a core with a faster clock speed will allow you to do more at the very lowest ASIO settings before any overload dropouts start to occur.

However, not all ASIO buffers are equal and a 64 buffer on one interface may offer you better latency than a 64 or even a 128 buffer on another model. Where the processer comes in here, is that depending on the driver efficiency, some interfaces may handle X number of plugins on a 64 buffer and another may manage Y amount at the same buffer setting.

A faster single core will allow more work to be done at low latencies within a single channel, which is great for recording. In an ideal world, we would look for the fastest single core performance, but ideally with as many cores reaching that speed on the chip as possible. CPU’s will often have a single core turbo rating, as well as a lower “all core” target speed and it’s this slightly lower figure that guides us to how well the chip will handle our workloads once we put the system to use.

With sound being captured from outside of the computer we find ourselves largely working with audio tracks when mixing. Whilst it is common for extra effects and soft-synth instrumentation to be added to the track later in the recording process, handling just the recorded audio itself whilst mixing is relatively undemanding when compared to trying to generate a full orchestras worth of sounds within the system.

This means that for many recording focused setups, the midrange CPU’s that concentrate on lower core counts and higher clock speeds can often prove to be a better fit than many of the higher-end setups with more cores and lower base clocks. Even the more modest midrange CPU's like the Intel 13700K or AMD 5900X are able to handle a sizable number of plug-ins, whilst offering fast responsive solutions for solo projects or smaller group recordings.

For a purely recording setup where audio is being captured and streamed from the internal storage, a good fast drive and even the basic 16GB of memory is often fully capable. Once you start to introduce plugin software to the mix, be it your effects or more keenly the addition of audio libraries the type of physical memory becomes more important. Our low-end systems often feature DDR4 which is still a great value solution for dealing with a large number of audio channels with processing. Where the DDR5 memory really comes into its own on the larger Intel and AMD systems is the extra bandwidth can help to offer substantial gains in overall polyphony when dealing with big sample-based libraries like Contact. For recording projects adding a sizable number of elements in the box DDR5 and a minimum of 32GB is recommend.

Our solutions feature a small M.2. drive for the Windows OS and applications. These super-fast drives allow for a snappy and responsive system with quick load-times. Systems will often offer a second M.2. option, which we tend to recommend for handling large sound libraries or doing video editing in multi-media projects. For general sound recording and editing, you may favour having more space overall for capturing your projects and large SSD’s can often prove more cost efficient in comparison to a large M.2. drive, whilst still offering enough performance to capture or stream hundreds of audio tracks.

We also offer larger mechanical drives in most of our systems for backup and archival, although this is optional as we understand that many users will choose to migrate older drives over to any new machine. To allow for this we always aim to ensure plenty of expansion capabilities where possible, although do contact us with details should you have any specific requirements for your build.

Our systems focus heavily around the low noise Fractal and be quiet! case models, with the respective Fractal Define 7 and be quiet! 500 being ever popular choices. Both ranges have great acoustic design and their own strengths, where the Fractal cases often feature heavier, denser panels with thick sound treatment to minimise any background noise, with quiet although average in performance stock fans. The be quiet! cases tend to have less physical treatment, instead trading off to include the more efficient and near silent be quiet! fan range as default. Our recommendation to anyone looking for the very best low-noise, high-performance combination is to match the Fractal case with the be quiet! fan upgrade in order to attain the very best of both worlds.

Many of our specs also feature be quiet! coolers and PSU’s as well, where they use the same great low noise fans across the range of supporting products, helping us to ensure the minimum of interruptions in your studio.

Both Intel’s Z790 platform as well as AMD's Ryzen chips feature-built graphics that perform superbly for audio applications. With many boards offering two display outputs natively, many users will find that this is perfectly adequate for running their studio screen arrangement. Upgrade options are available and we aim to include "Hybrid" cooling options which can run passively under lighter workloads. This helps to ensure your DAW sessions remain uninterrupted, whilst offering you performance when it's needed.

All of our systems feature wired Ethernet connections as standard, with many boards offering either pre-tested Wi-Fi included on-board or as an optional add on. Wi-Fi has a history of being disruptive in audio systems and as such is included in our testing when validating machines. Whilst there are a number of add-on options, each price point offers newer and faster supported Wi-Fi standards. If you are unsure of the best card to pick, we advise checking to see the maximum speed your router supports and aiming to match or exceed it with your choice of Wi-Fi adaptor.

When looking for a suitable audio recording interface there are often a host of features to consider, although two of the key focal points tend to be does it have enough connections and how well can it handle low latency recording? Internal sound cards offer the fastest processing route for your audio and often some of the best performance results overall. The limit of the form factor often makes the connection requirements of an average recording session tough to meet, at least not without a number of additional expanders also being introduced to the setup. This sometimes can also be a strength however as cards like the RME MADI FX can offer best in class performance along with the ability of up to 390 channels via hardware expanders

USB as a standard remains the most wide spread and its notable strength is the sheer variety of available interfaces and features found at across the various prices points. USB tends to be the highest-latency solution although high-end models like the Babyface Pro by RME which offers stand out USB performance, often measuring up to the best Thunderbolt interfaces, all the way down to models designed for those first starting out. Firms such as Focusrite, Audient and Steinberg who focus more on offering the full package with an eye on value, which aim to strike a balance of studio ready performance along with features and cost.

Thunderbolt continues to grow in popularity with PC users and some of its earliest adopters were studios looking to leverage its great low latency performance, which comes close to using an internal PCIe card. At this time we find that the market for Thunderbolt interfaces has largely moved to focus on higher I/O count devices, often making them attractive to users doing live recording.

This plays to the strengths of the standard as its high-speed connection to the rest of the system when matched with well-written drivers, can achieve superb latency whilst offering plenty of bandwidth. With interfaces often featuring a high number of I/O connections, we’ve seen some highly featured solutions appear for anyone wishing to record larger groups and bands. High performance RME models and the additional processing capabilities offered by the UAD Apollo range continue to remain popular. These have also been joined by more value-orientated packages that still manage to offer a tremendous amount of features and performance such as the also popular Presonus Quantum models.

As the technology progresses and a wider selection of interfaces continue to become available, the appeal of Thunderbolt will continue to grow for anyone recording music. With this in mind mainboards need to have Thunderbolt support included within the BIOS as well as at a physical level using a dedicated header for any future expansion. Where possible we aim to include this support with our systems, although this may not apply to every model and please do contact us for more information where required.