As are mine. I build all of my own machines. Stefan, my point is that every time some issue arises, these articles tell of 'many' people having problems. There is never any way to measure just how many have issues and what hardware they have. Microsoft knows only if people allow their telemetry, and that is if they even hook up their machines to the internet, and many people are scared of doing that because they don't want anyone to know what they do on their computers. Unless you run Windows 10 Enterprise, you cannot permanently block updates. The best you can do is delay them to a certain point. And, if you run the home version, you are plain out of luck.
That's my next project. I've never built a PC before. Always had them built. My main needledrop PC is an old i7-3770 and I keep wondering how much faster a 10th or 11th gen machine would process things.
You can block Windows updates by simply disabling the Service. I did so for about two years with my old Dell laptop. As for the "articles" I wasn't talking about any. When I have a problem, I google to see if someone else has had the same problem. This is how I discovered that quite a few people on various audio forums, Windows 10 forums, etc., all seem to have the same problem with audio after updating to 20H2. I`m just glad I was able to go back to 2004 without having to do a fresh, clean install as I have a lot of software installed. I also use my needledrop PC for music synthesis and sample playback so I have a lot of stuff installed.
I keep image backups of all my machines. For years I have built my own machines. Mostly using less expensive AMD boards. My DAW computer is an i7 and hands down has been my best computer. What I dislike about this installation of w10 I always get messages that my microsoft password has an issue and when I look into it it is always about me not joining the cloud, or something.
MOTU M2 works perfect but I run Windows 7 Professional. I even have the Samsung SSD 970 EVO Plus on C: connected through M2 pci-e which works perfectly and speed up the DAW work. I still don't trust this W10 full of bugs and ads, created by salesmen.
I thought I'd do a ghost rider special and post a denoised clip with the undenoised version after it. Here's a clip from end of the last track on Toto's 1986 Fahrenheit LP with a particularly noisy ending. Sounds awful but cleaned up pretty well thanks to some M/S surgery! Don't Stop Me Now
My MOTU M2 Asio driver makes RX shut down. I use the MME driver instead. It's weird because it used to work flawlessly and now it doesn't. I should mail Izotope support about this.
I'd wait for the Alder Lake generation 12 CPU due out in late summer. The only reason I built a new machine recently is because the 11-year-old Duo Core 2 wasn't cutting it anymore, and the motherboard was too old to support current technology. It will be easy to build a new machine. Just follow the steps on certain YouTube channels. I highly recommend Carey Holzman's build videos. he does a very detailed videos, goes slow, and explains every step, and why. The videos often run two or three hours because he likes to chit-chat with the members. All of his videos are recorded live, warts and all. If something goes wrong, it goes wrong on camera, just as it might with a real builder. But, he's been in the business for 30 years. He does it for a living. he'll show how to spend your money wisely, and how to build a reliable machine that will last years and years. No fads, no water cooling, no obsessive GPU crap that other channels have. He mostly builds for businesses, but will build for anyone, including gamers. Now, I just said that i'd wait for the Alder Lake 12 generation chips, and why I didn't, but you really don't either. You could build a Ryzen platform, but you have to know that Ryzen is picky with RAM, and the chipsets aren't as mature. And, there's a worldwide shortage of CPUs and GPUs. The GPU prices are outrageous! For audio, you can do with the video on the chip, like I got. Works fine for our purposes, and you can still enjoy 4k video. 16 GB RAM is plenty, and will work fine.
I do not recommend doing that, and I don't care what the audio forums say. But, thanks for alerting me to the potential USB issue with the Focusrite cards. Is that for all Focusrite cards or just certain ones?
I didn't hear an issue with the highs. I'm not sure what you mean by "smearing". Can you point out a time references where this is prevalent and what to listen for? Thanks
That wasn't from an audio forum but rather from a Windows 10 website or forum (can't remember which). Anyway, it worked fine for me for a couple of years. As for which Focusrite, I don't recall which ones. I don't pay much attention to them since I tried one a couple of years ago and ended up returning it to Amazon as it had some serious latency issues. I instead picked up a Steinberg/Yamaha UR44 that worked flawlessly with the same PC. For my other PC I had the choice of either an MOTU M2 or Steinberg UR22C. I decided to go with the latter since I've had such good luck with the UR44. Regarding the problem with w0H2, there was one post on a forum that claimed to have resolved the problem by changing a delay setting in the new version of Microsoft Edge, but that didn't work for me. It was really annoying problem because for the first 30-45 minutes, the PC would work fine but suddenly out of nowhere, there'd be distortion, a cluster of stuttering, repeated samples, then back to normal. It reminded me of a problem I had with a previous PC that I eventually tracked down to a shared interrupr between ethernet nd my old E-MU 1212M's PCI slot.
Yeah, I didn't hear any such smearing either. There's a slight loss of stereo definition or ambience in the lower midrange/upper bass noises the sax player makes (when I compare it to my CD version). Because I had to remove all the crap from the Side channel, that part is now in mono, but the highs were pretty much untouched except for the very end.
I didn't hear any damage either. It tough when the LP is so worn. I think it's good to post the before and after. Otherwise there is no basis to make a judgement as far the workflow goes.
When I was working on the needledrop I saved the last 35 seconds of the song as a clip because I realized how noisy it was and thought it would be a good example for this thread. However, when I did the before and after clip using the same section from the raw file, I felt it was too short so since I no longer had the full before file, I took the first 21 seconds from the after version and put it in both clips. So the first 21 seconds of each clip should sound exactly the same because it's the same ones and zeros! Expectation bias perhaps?
I listened again and I still think I hear a difference in the cymbals. My hearing isn't the best so maybe I am expecting to hear something different. All in all, great job cleaning it up.
Thanks. Actually I wrote above that the first 21 seconds of each clip is identical, but now that I think about it, I recall running a quick declick on the entire second clip including the identical part using the RX vinyl preset with the level cut strength cut back to 2.5 (the preset at a strength of 5 definitely softens transients. I would never apply it to a full file unless it was in really bad shape). However, the section on which there was denoising has no cymbals at all so any difference you hear is definitely due to the declicking. My apologies for my statement last night that the data was identical. I forgot about the declick.
I've updated my process with your advice. 1. Record at very low volume 2. Cut & paste S1 & S2 3. EQ to flatten my AT-VM540ML in the highs 4. Adjust total RMS level to -18dBFS 5. Phase to reduce peak levels 6. Declick sens. 2 Freq skew 3 7. Denoise with a setting that gently removes vinyl noise. 8. Manual declick/fade outs/markers/silence between tracks 9. Save file 10. Loudness control to -12LKFS 11. Save as new file 12. Export regions to flac files That's it. Whaddya think?