mettelus

iZotope's Radius is also the underlying algorithm in some other programs. I believe that was originally a product that ended up getting more royalty fees from being licensed as an algorithm (Cakewalk, Adobe, etc.). I am not sure of the details to that one, but one thing that does come to mind is that I am not sure if all of the parameters (transient sensitivity, pitch coherence, phase coherence, etc. on the bottom of that page) are exposed in SONAR/CbB? There is a nice section about best results/usage also at the bottom. RX5's version doesn't seem to have changed much. The parameters (to any algorithm) - what they do, and how they are applied - is another important factor to consider.

I have had an ultrawide for a couple years now, and this is how I run it. I sit 3 feet away from the monitor, so it is not worth taxing my eyes to use it. I actually searched for one that didn't have higher resolutions, since they are far cheaper and 4K would be wasted for my use (I just wanted more track view length and easier ability to edit documents side-by-side on the same screen). Most editing applications that allow for menus to be right/left docked benefit greatly without upping the resolution (the screen is slightly shorter, but 8" wider than my old 27").

I just downloaded the most recent NVIDIA driver to check the install screen and was pleasantly surprised that they split the install options into "Experience" and "Drivers only." The quickest way to clean that garbage is to do a fresh install, select Custom, and only choose the "Graphics Driver" and "PhysX Software." Also check the "Clean Installation" box at the bottom. Do not install anything else (HD Audio Driver, Experience Stuff, Shadow Play, etc.). Another thing to check on a system is "Services" (type in the search menu) and see which load automatically on boot. Many "automatic start/update" services can be set to "Manual" under properties. All this does is restricts them from launching until you load an app rather than them loading on boot. More and more apps have intrusive phone home and auto update services that load into memory on boot that are unnecessary. You can also disable network connections when doing DAW work to achieve similar, but the Services route will let a machine stay connected without issues.

A quick check before you get too far is to take one of your troublesome projects and "Save As..." to your SSD. Be sure to also check the box that says "Copy all audio with project." I would even close out CbB afterwards, reopen CbB, and load the project you just saved. That will let you see the difference the disc drive makes. Try that first before predetermining a solution. The reason I say this is that there are a number of reasons that a computer will drop out (buffer to small and taxing the CPU for FX, background processes, etc.). Windows "Auto Network Discovery" is one background process that is a menace (#1 offender in my experience), so there are more possible issues than simply the drive speed. As mentioned above, heavy streaming from a drive would be of concern, but I am not sure if that really applies to you. What instruments and effects are you running? Have you checked your system running a project with LatencyMon yet?

Nope. There is definitely a lot more in the fray than just the EQ curve... sensitivity, polar pattern, saturation characteristics, etc. The EQ portion specifically is not as crucial as advertising suggests (even that page seemed to focus on that parameter). I was more surprised by the level of detail in the taps into retailers' sites. Based on the video, I am more inclined to think that they "characterized" mics, then ran the same signal through each "simulation," but I didn't readily see the specifics of what they did. It almost comes across as if they applied the EQ curve to a given signal, then normalized it, so the only difference you are seeing is the EQ curve. I am more curious as to the conditions of their testing TBH, since using a laser to detect sonic parameters requires a different setup than I could tell from that video. I actually presented the distortion on the face of an acoustic guitar years ago (holography - diffused/split 3B laser, optics table, whole nine yards), and those promotional pics look like a laser level from Home Depot, which isn't even close. I just saw a line level across a speaker and was wondering, "And the laser is doing 'what' exactly?" But even without that, the conditions of the test are in question. If (BIG if) this is merely applying the EQ curve of a mic to a stock signal (which you can then EQ back out), the usefulness of the site is a lot more limited than it appears.

Nice video. His comment about "being able to work with any of these" (because he can EQ them any way he wants to) is entirely accurate. As long as the SNR is good, there is no reason to get fixated on the stock EQ curve of any mic (or pickup for that matter), unless post-processing capabilities are limited (in the DAW world, this is never true). Polar patterns, signal strength, and the like are things that post-processing cannot always make up for, but the EQ curve isn't anything to get worked up over.

That would be preferred (if it is an option), but I still do that on all machines. It is amazing how much garbage is loaded into RAM on boot without user intervention.

Very nice, thank you for posting! I didn't even know the newer interfaces can go down to 16 sample buffers... that alone is pretty nutty.

If you disable apps and processes (services) that autoload on boot (basically strip the machine for "DAW mode"), it can give you a bit more room. IIRC, my HP would go to 35%, but rapidly jump to 50% with simple usage. I was manually cleaning RAM on that machine when it hit 75%, then the process would start over. Quite a few programs do not release RAM properly, so it is sometimes simpler to set processes (via services) to "manual" (manual start), and only open them if you intend to use them.

^^^^ AudioSnap will make you want to tear your hair out, and is overkill for a single track. SM/BAN is the only sure-fire way to tempo map a single track (Melodyne will fall on its sword in the absence of transient material fairly frequently - although I have not tested this with the most recent updates).

They really need to revamp the Spitfire Audio App to drill into the LABS banks, since they are free and that is what deploys them. Having to log into their website and click on "Get" (with the SFA app closed... it fails if already open) is becoming a nuisance for me.

Which OS are you running? 4GB of RAM stands out regardless of DAW for me. I had a HP Elitebook with 4GB of RAM (non-expandable), that suffered greatly with PowerPoint if opening several files at once. Win 10 is probably chewing up roughly half of that RAM with nothing running. Win7 should be a bit better. When working with audio recording/playback, CPU/RAM restrictions are pretty light, where they get into problems is smaller buffer size (ups CPU hit) for tracking and the use of VST(i)s (some have massive CPU hits, even with large buffers). You can counter this during tracking by disabling FX (Global Bypass - hotkey "E"), being judicious with FX, and baking in FX/bouncing & archiving soft synth tracks. Since CbB is free, you can easily give it a go, but be aware of the above as you develop your workflow. With an SSD, you will minimize your pagefile visibility when using larger buffers, but you will never get to heavy-handed VST(i) usage.

They are free and do not take up a lot of space. The disadvantage is that some have FX baked into them (either negating raw samples from being used or requiring a bit more post-processing to manipulate them), but "as is" most are creatively designed to give them a unique flavor unto themselves.

iZotope's Vinyl has been around a long time. Be sure to remember you have it if used in projects. It is one of those plugins that inserts noise with the transport stopped. Remembering you have used it will save you some heartache trying to troubleshoot your audio interface in the future!

iZotope's Vinyl is free, and I think that has a decade dial on it but forget for sure. Check a recording that you really like for frequency response. The high end might be rolled off more than you may be anticipating.

This is Motherboard dependent, so be sure to check this information as you build your machine. Typically one of them can be set to higher speeds (X2 or X4) at the expense of SATA III connections on the board (my NVMe X4 drive disables SATA 5/6 connections). I used normal speed for the OS, and the X4 for a larger sample drive. Otherwise, your intended configuration will work, but make sure the motherboard supports the NVMe drives too. Another thing to check is the form factor and keying for NVMe.M2 drives... one is for their length.... make sure what you want to use will also physically fit onto the motherboard.

Filters also attenuate the desire signal, so noise reduction that can capture a noise print (part of the file that has nothing but noise in it) and then removing that from the entire file (as long as it is consistent noise and not transient-based) is the most effective. The two best option for this are Audition and RX, both of which are not free. Audactity and ReaFIR are both free, but do not do quite the same thing. Is there a portion of that file that does not have bass notes in it? Is that hum consistent in the file? If yes to both and you are willing to post the file, someone with Audition or RX can address it (if you can post the link here). If desired, you can also send me a PM with a link and I can do it after the sun goes down today. As an aside, the Capture Noise Print/Noise Reduction is an effective tool for any audio recording, but requires a portion of the file to be "just ambient/environmental." For that reason, I will record lead in/out portions of an audio file if I have environmental noise. As long as you have the tools that can remove it properly, it saves a lot of time in catching ideas on the fly (particularly microphones) and recover them to a condition that can be subsequently used. I kept a file from years ago that had -40 dB of fireplace noise in it that got slightly louder over the 90 second recording (I do not noise gate above -55 dB in most cases). Audition removed that so impressively that my workflow has relied on this since. This has also been an ongoing Feature Request for years to have included in SONAR/CbB.

The only reason that I can think of for boot disc "size" is for imaging the OS, which is the only image I perform. The smaller the image, the faster it can be done/restored. That being said, with NVMe drives, most machines are limited to 2 these days, so maximizing capacity for those is more cost effective. From an image standpoint, that is the only situation I would partition a drive these days. But in doing so, also realize that a physical failure of the drive can lose it all, so backup the data on the secondary as well. Consider the imaging perspective when choosing your OS drive/partition. I image the OS, but use xcopy/robocopy on data drives since it is far quicker. For an OS that small (mine is 256GB), bear in mind that I also riddle my OS drive with junctions. I let software install to default locations (except VST dlls), then end up moving/junctioning anything that gets larger than 4GB. My OS drive is 256GB, and runs around 130-140GB and images less than 100GB. When keeping an OS that small, you also want to be aware of how to clean the rubbish off of it (temp files, downloads, etc.), especially before each image.

+1 to this. The only issue I know of is backwards compatibility with Aux tracks/Patch points. Forward compatibility is not an issue (X1 files open in CbB). You will simply get a warning that "x VST failed to load" without re-installing those missing VSTs, but all MIDI data and audio will be there. Chuck's cleanup advice is definitely advised. Anything you want to pull into SO as a "project," I would also recommend exporting audio as broadcast wav files, one per track, so all audio will line up for you during the import into SO.

Be VERY careful doing physical loopback, they can create feedback loops so quickly that they can damage equipment. Do NOT just attempt this. CbB will record to an armed and muted track which is "typically" the safest way to prevent feedback loops, you do not want to reintroduce output back into inputs. Basically you want monitoring and inputs separate, which needs to be understood before trying this. Also, always start with gains at zero on inputs before introducing signals or making/breaking connections. Before you take off attempting this, understand the above and the details of your gear. I am on a cell now, but wanted to inject some reality before you get into issues that could destroy something. Most physical loopbacks (if not internal) have NO safety features to prevent damage to components.

The OP references not being able to open files because the sample rate is higher than the sound card. I have never heard of this, since CbB automatically streams to the current sound output capacity. You "should" be seeing an error on open with the option to select another output source, as simple as that. That file not opening could be internal corruption or another error (there have been posts similar). I would recommend contacting customer support if you are not seeing the option to choose another output device. Azslow3 created a plugin for Reaper which converts cwp files to Reaper format. If that cwp file will not open, that is an option to recover the MIDI information. Numerous free wav editors will convert the audio, which should reside in the project->audio subfolder.

I didn't even realize this existed and was a little shocked that it is a paid add-on. Every other program that I know of for batch conversion has it baked in. SO lets you drill into the wav folder by opening Windows Explorer, so I am a bit at a loss for the excitement for it. Even the script processing has been out for years (at least for Adobe Audition 4+), but that particular feature is very nice for dialog work or instrument sampling, and not sure if that exists in anything that is free. The price point seems to be a bit high to me for simply being integrated, but the scripting feature may be the real reason for it.

TTS-1 does have the advantage that CbB (with no output MIDI selected in preferences) will launch MIDI files, insert TTS-1, and route it for you. That feature is nice if you are starting with another's MIDI file, then can replace content (shift-drag MIDI) to other VSTis. I'll admit that with the processing power of newer machines, I do not get as excited by multi-tibral as I used to. For some reason I waste more time routing/setting them up than to just pick and choose what works best piecemeal. Then again, it could just be laziness on my part.

Another comment to add to the above. The SI bass is also off by an octave IIRC, requiring you to to manually offset it. Unless you also have notes in the higher octaves, it will appear to be playing, but no sound.

I just ran it again and it came up fine with the Install option for the update. Launched fine afterwards as well (no need to refresh), so I think they were fiddling with things on the back end while this thread has been going.