Jim Roseberry

I bought this... thinking it was too good to be true. It is too good to be true. Although the Full Compass website lists/shows Samplitude ProX 4 and SoundForge Pro 12, you actually get Pro X3 and SoundForge Pro 11.

Unbalanced I/O with phone plugs. If you're looking for quality recording, that's a bad start. Also note, with A/D inside the box, about the best noise-floor I've seen is ~-100dB. Most of the better audio interfaces with external converters have noise-floor of -110dB to -118dB (usually more expensive interfaces have lower noise-floor).

Quick speed reference: Conventional HD sustains ~200MB/Sec SATA SSD sustains ~550MB/Sec M.2 Ultra (NVMe) sustains ~3500MB/Sec Where you're going to see the most bang-for-the-buck is with disk-streaming Sample Libraries. They're going to load faster... and you'll be able to pull 17+ times the disk-streaming polyphony vs. conventional HD (6+ times the polyphony of SATA SSD). For most users: We recommend using a SATA SSD for your boot drive (still plenty fast). Use M.2 Ultra SSD (more expensive) "strategically" for that library/libraries that loads slow or where you need massive polyphony. On many motherboards that have two M.2 slots; if you run the second M.2 slot using 4 PCIe lanes (maximum performance), the last two SATA ports will be disabled. You can work around this issue by putting the second M.2 Ultra SSD on a PCIe host card. Note that it has to be installed in a PCIe slot with at least 4 PCIe lanes. If you've got a complex build (multiple M.2 Ultra SSDs, multiple PCIe cards like UAD-2, dedicated video card, etc)... available PCIe lanes can become an issue. Typically, if you're going socket 2066, it's because it's a more complex build. 😉 If you go socket 2066, go all the way and get an i9 CPU with 44 PCIe lanes.

No ads here either... (Win10 desktop)

I spec'd and built every PC Sweetwater sold for the first ~3 years. Note that I was always freelance (not an employee). 😉 We're still on good terms. I (like many of you) buy a lot of gear from Sweetwater. If you go to GearFest 2019, you'll probably see me there looking at guitars.

Hi Bob, When you call... you'll get to talk directly to me. 😉

There's precious little time for me to actually work on music. That time has to be productive. The owner of Above Records used to call me once a week (Rolling Stones, Neville Brothers). Each conversation used to end with, "So... do you have any music I should hear?" Used to almost physically hurt to tell George Daley "no"... because I was just too busy working on the technical side. Life is nothing if not ironic. 😉 At 18 years old, I would never have dreamt I'd know/talk to the guy who signed the Rolling Stones.

Noel would have a real explanation... 😉 I'm guessing it has something to do with the fact that not every process in a DAW can be multi-threaded. Video rendering is much different in that regard.

Yep. Class-Compliant means crap latency performance on PC. You'll want an audio interface with a proper ASIO driver.

It'll get used to produce synth samples for use live on stage.

@abacab Thanks for the run-down! I will go for Max... as that seems to be significantly more complete.

Those who have installed ST4, how are you liking the sounds?

Most major DAW applications have what's called Automatic Plugin Delay Compensation (automatic PDC). If a latent plugin is inserted *anywhere* in the project, all other audio is delayed by that amount to maintain sample-accurate sync. If you have several latent plugins inserted in series, the latency adds up quickly. The workaround is either to avoid latent plugins when tracking... or use the Global PDC Bypass option in the Control Bar. Latent plugins are often Mastering, Linear-Phase, Convolution, or Look-Ahead type processing

If you're like many general-purpose users (Surfing the Internet, Facebook, Email, Office apps), you may get along just fine with 4GB RAM (even though it's running lean). FWIW, I think you'd find the VM Swapfile (even on SSD) is **FAR** slower than DDR4 RAM. 😉 DDR4/2400 has a transfer rate of 19.2 GB/Sec M.2 Ultra (NVMe) SSD sustains 3.4 GB/Sec. SATA SSD sustains 0.54 GB/Sec

Agreed... but not running Win10. If you're left with 1.5GB RAM (or less depending on how well it's optimized), that's running super lean. It wouldn't take much of anything to get the machine hitting the VM Swap-File (and that'll kill performance on any machine).

A well-configured lean install of Win10 will use about 2.5GB RAM when idle. That leaves a whopping 1.5GB for CbB and any plugins. Even if the OP was recording audio/MIDI only (no virtual instruments), that's running too lean and most likely hitting the Virtual Memory Swap-File (which absolutely kills performance).

With 4GB RAM, the machine is likely getting "ram-starved" in which case it'll hit the Virtual-Memory Swapfile (which absolutely kills performance). You need enough RAM for your largest projects. Additional unused RAM doesn't buy any additional performance. Also, check the laptop's DPC Latency using Latency Mon. High DPC Latency can cause audible glitches and transport drop-out.

Mastering, Linear-Phase, Convolution, and "Look-ahead" style plugins are often latent (add additional latency).

@Morten Saether It's good to see you here!! Been a long time... 😉

Latency has but two sources. Audio Interface Latent Plugins Use a dedicated audio interface that has a proper ASIO driver. If you've inserted latent plugin/s anywhere in a project, all other audio is delayed by that amount to maintain sample-accurate sync (Automatic Plugin Delay Compensation). If you're using latent plugins in series, the latency can pile up quickly.

FWIW, Those are the same parameters that are on the Kemper Profiler. 😉 Steven Slate alluded to TH-U being able to load Kemper profiles. What's been released thus far (at least to the general public) doesn't allow directly loading Kemper Profiles. Allowing the user to load Kemper Profiles (assuming they sound identical to the Kemper) would be huge.

The end of the video where he's going thru the Rig Player: He left the tone controls at extreme settings (set when he switched to Cab off and tried using the stock "MESA" Cab)... then discovers that all the profiles sound bad. 😉 With Kemper style profiles, they're a "snap-shot" of an entire Amp/Cab/Mic rig at a specific amp setting. The Kemper (and Rig Player) have a Gain control... but if you get this too far from the original amount, the profile no longer sounds accurate. If you want an accurate profile with Gain 25% higher, you have to go back and profile that setting on the Amp.

I agree with the static nature of IRs... and that the Cab section needs to evolve/progress. BTW, The UA Ox Amp Top Box has dynamic Cab sim (not using IRs). It's not cheap... but it's a great recording solution if you have a quality tube amp/s.

FWIW, I don't like the stock Cabs... IMO, TH-U starts to sound a whole lot better if you use quality Cab IRs. What I think is ideal (if you have the Cabs/mics)... is to create your own Cab IRs (your favorite Cab/s, mic'd the way you like). This helps bend the sound (tone) to being more "you".

If you've got an off-the-shelf laptop, it may not expose the BIOS parameters necessary to stop all performance throttling. The cooling in a laptop (in such a tight space) is not ideal for locking a CPU at the highest TurboBoost frequency. With a tower, you can use over-spec'd cooling (much more significant heat dissipation)