Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 14, 2014 17:24:13 GMT -6
|
|
|
|
Post by popmann on Oct 14, 2014 17:25:23 GMT -6
I use Cubase's built in SRC. I have used any number in the past....I'm not picky. I'm aware there are theoretically better ones....but, spitting out what I need from the original file (24/88 or 24/96) always yields better than using 44.1 from the get go....so....I choose what to be picky about. In fact, the only conversion I've ever had be particularly lossy was 48 to 44 on a really old machine--which is what convinced me to move down to 44 for a project--and never did THAT again....
While I'm sure a mastering guy like BobO can speak to what the better/best SRCs are....I've never felt the need to be particularly picky. SRC is not the demon in the room. I've done "uneven math" conversions and never found them to be any different than "even math"....the common result is if the tracking mixing were done at double rate, the end result is better. Regardless of what that end format was.
|
|
|
|
Post by Bob Olhsson on Oct 14, 2014 17:25:35 GMT -6
It's an unfortunate fact that the sweetest combinations of converters seem to be those made by the same manufacturer...
|
|
|
|
Post by Bob Olhsson on Oct 14, 2014 17:27:04 GMT -6
I've tried them all. Weiss Saracon has been the most reliable for me.
|
|
|
|
Post by LesC on Oct 14, 2014 18:39:55 GMT -6
You guys have convinced me to try 96k, instead of 44.1k, so thank you for complicating my life! In order to help uncomplicate my life, can you please tell me the best options for sample rate conversion back down to 44.1k on PC? I know there have been threads about this, but no real consensus. Fear of that last SRC has always held me back from trying other sampling rates. If it helps, for recording I'm using Cubase 7.5, and I have an old Wavelab license that I could probably upgrade fairly inexpensively. Popmann, I think you're a PC guy, what do you use? Or anybody else? FinalCD 0.17. Free. Try the /f2 filter option. Killer SRC. As in...uhm...no namedropping. It is just damn good SRC, period. Measure, hear, believe, whatever. Also has a sharp filter option if there is trouble going on with this setting, depending on the program material. www.sonicillusions.co.uk/downloads.htmThank you! I checked it out on src.infinitewave.ca/ as well, and it seems to measure very well. Several years ago I tried an experiment using Cubase's SRC and I didn't like the results, so I stuck with 44.1k. I didn't realize at the time that all SRC's are not the same. I will give FinalCD a try, and thanks again! |
|
|
|
Post by LesC on Oct 14, 2014 18:43:44 GMT -6
I've tried them all. Weiss Saracon has been the most reliable for me. Bob, have you tried FinalCD? It seems to measure better than Weiss Saracon. I know measurements don't tell the whole story, so I'm curious if you've actually made the comparison? |
|
|
|
Post by matt on Oct 14, 2014 18:56:29 GMT -6
I can't do 96K on my mac mini/Symphony/PT system, so it's 48/24 for me. I think it has to do with needing to track MIDI drums into BFD through PT11; the running processes chew up CPU. Even at 48K, I can only run at 128 samples. It's a good thing my drummer is tolerant of latency.
I see a 12 core Trashcan Mac in my future and then I will move up to 96/32 over T-Bolt2 to my peripherals.
Regardless of the debate over audio quality, it is fundamental that computing processes, assuming equal code quality and correctly designed hardware, benefit from increases in computing power and data throughput (CPU+bus speed, Murphy's Law, blah blah). In my mind, higher sample rates equate with higher data rates over system buses and through software when processing audio. Then, more samples per second = greater throughput = greater fidelity to the original waveform. But ONLY if the code is quality, AND the hardware up to the task, then quality improves. Fidelity improves too, as in the accuracy of reverb algorithms, depth of sound stage, etc, etc. Honestly, my Mini really isn't up to it, and I wish I had bought a 15" Macbook Pro, which has a dedicated video chip. I think that would make a difference to me.
Call me silly, but I think that high sample rates combined with the latest computing badassery and 64 bit software is the way to go. After all, how else are we to emulate the 1950s, 60s, and 70s technology that everyone thinks is grail-worthy?
|
|
|
|
Post by tonycamphd on Oct 14, 2014 19:18:47 GMT -6
I can't do 96K on my mac mini/Symphony/PT system, so it's 48/24 for me. I think it has to do with needing to track MIDI drums into BFD through PT11; the running processes chew up CPU. Even at 48K, I can only run at 128 samples. It's a good thing my drummer is tolerant of latency. I see a 12 core Trashcan Mac in my future and then I will move up to 96/32 over T-Bolt2 to my peripherals. Regardless of the debate over audio quality, it is fundamental that computing processes, assuming equal code quality and correctly designed hardware, benefit from increases in computing power and data throughput (CPU+bus speed, Murphy's Law, blah blah). In my mind, higher sample rates equate with higher data rates over system buses and through software when processing audio. Then, more samples per second = greater throughput = greater fidelity to the original waveform. But ONLY if the code is quality, AND the hardware up to the task, then quality improves. Fidelity improves too, as in the accuracy of reverb algorithms, depth of sound stage, etc, etc. Honestly, my Mini really isn't up to it, and I wish I had bought a 15" Macbook Pro, which has a dedicated video chip. I think that would make a difference to me. Call me silly, but I think that high sample rates combined with the latest computing badassery and 64 bit software is the way to go. After all, how else are we to emulate the 1950s, 60s, and 70s technology that everyone thinks is grail-worthy? Matt, consider a 2009 nahleham mac pro, you can update the firmware to 5,1 and install 12 cores to match and surpass 2014 standards for 1/4 the $, it also assures future proofing to match the 2014 trashcan 8) |
|
|
|
Post by popmann on Oct 14, 2014 19:21:24 GMT -6
You will always lose something going to 44. That should go without saying, no?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 14, 2014 19:42:36 GMT -6
Well, FinalCD measures excellent with the sharp filter.
Weiss Saracon seems to be a long time quasi industry standard. Many mastering engineers like it best. For sure for a very good reason.
Actually a bunch of mastering engineers asked the developer of FinalCD, if it would be possible to implement a medium filter between the very gentle filter and the sharp filter, actually something to come nearer to the Weiss Saracon behaviour. This filter is not yet shown on src.infinitewave.ca ...
Actually, Jaakko, mastering engineer from finland, a very nice gear crazy guy i knew from GDIY for many years, before he started mastering business, posted on facebook a while ago he uses it now with very good results and brought me to try it out...it is an instantly available option if you don't want to invest into the Weiss Saracon or similar highend options....
|
|
|
|
Post by tonycamphd on Oct 14, 2014 20:15:04 GMT -6
Hey small, this is about 1000 miles over my head lol, isn't it used primarily by ME's? How would a dumbass like myself implement FinalCD? |
|
|
|
Post by tonycamphd on Oct 14, 2014 20:32:08 GMT -6
I've been doing some hearing tests lately, is it weird that at 46 years old, and as expected, i'm not hearing a 16khz sine wave at all, BUT i can indeed make out a 20khz sinewave? i'm befuddled by this?.?.. I fully expected to hear nothing above 16k, but 20K! really? Am i hearing a harmonic, or some artifact from my speakers? or am i missing something?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 14, 2014 23:33:46 GMT -6
Hey small, this is about 1000 miles over my head lol, isn't it used primarily by ME's? How would a dumbass like myself implement FinalCD? OK. I wrote 2 simple batch files, attached to this post for a quick start. Download this and the finalcd017 archive from the webpage mentioned. Extract the files in both zips to a same folder whereever you want. Throw a couple 48 or 96kHz PCM waves into this folder as well. Double click 16bitTPDFmedium.bat, or execute with right mouse click, "open"... Watch the files getting converted to 16bit, TPDF dithered, 44.1kHz and drink a cup of coffee. If the program material is polluted with high frequency stuff up to nyquist (;-)), say ugly clipped material, try the brickwall filter version, which is slower in conversion speed: 16bitTPDFsteep.bat You find the converted files in the new folder "converted". This should get you going already. extracttofinalcdfolder.zip (509 B) |
|
|
|
Post by Bob Olhsson on Oct 15, 2014 9:46:12 GMT -6
I played with finalCD for a while but finally decided it was no better, only different and sometimes worse on some program material.
There was a bug I reported in the first version of Saracon that made it sound a bit edgy on some material. They fixed it years ago but the internet mythology persists. Further proof that no matter how esteemed the developer is, they do get it wrong sometimes. The nice thing about analog technology is that when its broken, it's a lot more obvious. Digital signal processing is way more open to human error.
|
|
|
|
Post by matt on Oct 15, 2014 10:19:56 GMT -6
I see a 12 core Trashcan Mac in my future and then I will move up to 96/32 over T-Bolt2 to my peripherals. Matt, consider a 2009 nahleham mac pro, you can update the firmware to 5,1 and install 12 cores to match and surpass 2014 standards for 1/4 the $, it also assures future proofing to match the 2014 trashcan 8) Tony, it is my understanding that the previous Mac Pro will not do T-Bolt. If correct, this is a non-starter for me, unfortunately. |
|
|
|
Post by tonycamphd on Oct 15, 2014 11:16:16 GMT -6
Any peripheral that can be used on a 2014 machine can be used on the 2009 nehalem, its the 5,1 firmware lockout that prevents current standards, that has been dealt with and properly despised of, I'll post the link later if u want  |
|
|
|
Post by popmann on Oct 15, 2014 12:14:24 GMT -6
Matt, the answer is two machines. If you need to use BFD2 for your drummer's Ekit, it needs it own box. And that box needs a PCI or PCIe audio interface.
You simply can't run VIs on the same box as audio and expect either to perform at their best. The answer isn't a $5k MacPro. The answer is a $50 PC....and some kind of PCI/PCIe 44.1 IO box for BFD. I could run BFD2 at 32samples on a 15 year old P4 box I just threw away (because it didn't have resale value). I literally kept it this long thinking I would eventually buy an eKit and that would server as the "brain". Now, I just figure I'd use my current box with it's RME PCI card....as any new machine I won't be able to take it with me (being PCI)...so, pitch it I did...but, you can jump on Craigslist pretty much any day of the week and pick up some old box....then some PCI or PCIe interface....problem solved--set your buffer on your mac to 2048--no reason to care any more--monitor with Maestro. You immediately get 10 times the audio processing AND el nada latency for the drummer. Problem solved--cheaply.
|
|
|
|
Post by matt on Oct 15, 2014 12:22:50 GMT -6
Matt, the answer is two machines. Excellent idea! Thanks! |
|
|
|
Post by Bender on Aug 23, 2023 14:32:17 GMT -6
Bump! About 10 years later, what's the consensus? I'm still recording at 48k 24bit but thinking about jumping up to 96k. What's everyone else tracking at these days? |
|
|
|
Post by notneeson on Aug 23, 2023 14:42:58 GMT -6
I have been jumping around lately and not noticing a meaningful difference in outcomes.
The only exception is 192k, at which rate some plugins don’t load and my SPDIF connection from the D-Box becomes useless.
|
|
|
|
Post by niklas1073 on Aug 23, 2023 15:32:56 GMT -6
Just tried again a project in 96 for the fun of it. Could not feel achieving anything with that besides losing half of my tracking abilities. Ill still stick to 48/24
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 23, 2023 15:36:51 GMT -6
Bump! About 10 years later, what's the consensus? I'm still recording at 48k 24bit but thinking about jumping up to 96k. What's everyone else tracking at these days? The consensus is let this RIP, there's a metric crap ton of tracks recorded 44.1Khz 24 (or less) bit and they've not struggled. |
|
|
|
Post by Bob Olhsson on Aug 23, 2023 15:37:31 GMT -6
I'm still working at 96.
|
|
|
|
Post by chessparov on Aug 23, 2023 16:20:57 GMT -6
Practice vocals at 16/48 on my travel laptop.
24/48 for the very occasional "important".
Would sometimes do 96, if I'm in an excellent sounding rAcoustics. "nice" Project Studio level acoustics.
Chris
|
|
|
|
Post by Martin John Butler on Aug 23, 2023 17:41:56 GMT -6
I read a long time ago that some plug-ins work better at 96k. So, all my personal work is at 96k, but my professional work is at 48k, just for compatibility with friends whose gear is long in the tooth.
|
|
