Bill Roper's Journal

I got up early this morning, because one of Gretchen or I needed to take K off to rehearsal this morning, as we're less than a week from the opening of the fall play over at the high school. I grabbed the short straw without a drawing, because I *also* needed to be up to write the deposit check for the massive upcoming work on the fence. Nothing will happen there until Julie shows up to mark the utilities and I am told that they are not moving quickly at the moment, partly because the request went in on Friday and partly because Monday is a holiday. (Not for me, but for many people, including the kids.) But I am looking forward to a better fence.

Meanwhile, since I was up early, I decided that it was time to swap in the new 4K cable box. I pulled out the old box, getting covered in dust myself, plugged in the new box, and discovered that the new 4K cable box was an HD box, which is functionally identical to the box I had just dustily unplugged. I got back on a chat with an Xfinity rep and suggested that when I request a wired 4K DVR cable box that I actually need a box with all of those attributes. I have been promised that the next box that I receive will be correct and have been given a small discount for the next 12 months in compensation. This would not be nearly so irritating if I hadn't had a similar problem when trying to replace the cable box in the bedroom when we put a 4K TV in there.

Laundry, in the meantime, had been set in progress and Gretchen joined me for a quick lunch over at Texas Roadhouse. When we got back, I headed off to the basement to swap laundry and download the new plugins that I'd picked up with the Ultimate 13 upgrade from Universal Audio. I got everything installed, fired up Cubase, and Cubase refused to talk to the Apollo's Thunderbolt interface, although everything *else* was seeing the Thunderbolt connection.

*sigh* I read assorted webpages, followed the suggestions, uninstalled the software, reinstalled it, rebooted the computer, and eventually got the connection working again. Thunderbolt! It's not just an interface, it's an adventure!

But I finally managed to pull up the scratch track that I've been using to play with lately in Cubase and start auditioning the new plugins. There were some nice amplifier plugs that may be useful in some environment (although not so much with this track). The new Topline Vocal Suite might come in handy some time, but not at the moment. I was ready to button everything up when I remembered that I had the new SSL 4000E channel strip that I should check out.

Oh, my! I started fiddling with this and quickly found that it likes my voice. I mean, it likes my voice *a lot*. I am going to need to get some other people to listen to it to see if they like what it does with my voice nearly as well as I do, because it's always suspect when you are listening to your own vocals, but wow.

Now, I just need to get all of the CD orders placed for OVFF.

And I also now know who my rooting interest is for the rest of the baseball postseason. That would be Cleveland, as the remaining teams are all from New York and Los Angeles. It's been a long time since they won a World Series.

They'll do. :)

One of the things I've learned about complex software is that the best way to describe what it does is that it has behaviors. The behaviors of the software are controlled by the rules that are programmed into it.

If you write the rules correctly and the interactions are nice and tight, then your software will be well-behaved. If you write the rules badly -- or, God forbid, fail to initialize parameters for the rules through one or more species of programming errors! -- then the software will be badly behaved and unpredictable.

All of this turns out to be harder to get people to do than you'd think. :)

It is now 15 minutes until it is time to start drinking the colonoscopy prep solution. The good news is that it is a much better prep than the one that I had to take when I started getting regular colonoscopies.

That's pretty much the limit of the good news.

So far today, I have eaten one serving of lemon ice and drunk a Diet Pepsi. I am now working on a second serving of lemon ice and a glass of real Coke with actual sugar (ok, HFCS, but at this point, I am starving and *really don't care*). And I know it will take me a long time to actually starve, but it has been a good day to spend time down in the basement studio transferring old ADAT tapes.

The transfer game has caused me to really appreciate modern DAWs. There is one ADAT with two songs on it. The first one was fine in one take. The second took multiple takes. *Many* multiple takes. If I were doing this in Cubase today, I would stack up two or maybe even *three* takes of the track and then assemble a final copy. This was not an option back then. Punch-ins *were* an option, but the song is complex and just irregular enough that punch-ins were going to be difficult, so we just did multiple takes.

Listening to that particular marathon, I could only think, "Oh, I'm sorry."

In and around all of the transfers, I also managed to get a load of my shirts done. This resets the laundry clock so that a load of my shirts is no longer due to be washed on eclipse weekend, which will be a good thing.

And now it is time to finish my second meal of the day, because lots of fun will be coming soon!

I spent a couple of days this week popping a new feature into some of our code at work. Today, I was finally in a position to test it.

Naturally, it failed.

But then I went and fixed the null pointer exception that had resulted from trying to retrieve a value from one model using a key that belonged to a different model and now everything appears to be working correctly. At least, the cases that I've run are doing the right thing.

So now I've found the bug that was lurking in the new code and can safely check it in.

Or maybe I'll do some more testing first... :)

So I'm continuing the inventory of tracks for Crosstime Bus. This involves opening up and looking at a lot of very old Cubase projects that I haven't touched in years. This means that they were recorded in a much earlier version of Cubase than the Cubase 13 that I am currently using.

Stupidity abounds.

First, I have completely rebuilt the studio a few times in the intervening years. None of the old routing is worth a darn, but that's a matter of a few mouse clicks to correct. I also need to delete a number of empty tracks that were possibly useful in the old setup, but which are not at all useful now. And it would be good if I actually labeled all of the tracks with what is actually recorded on them, for the sake of my sanity, if nothing else.

All that done, I can get sound out of the speakers, which we count as a victory and also as something that helps greatly as I label these old tracks. But over on the right hand side of my main Cubase display is a giant meter that should be showing me the output levels.

It does *nothing*. It doesn't even *twitch*. What the heck? I *know* I've seen that work before.

A brief Internet search follows and is of absolutely no help, save that someone suggests that I look at the VST Connections window. There *is* no VST Connections window in Cubase 13, of course. There *is* a window labeled Audio Connections that comes up on the same accelerator key as the obsolete VST Connections window, which is a window that I am familiar with, because that's where I go to fix all of the routing errors that exist because of the multiple studio rebuilds since the last time that I opened this project.

When I right-click on the Stereo Out bus, I get a context menu and an option is checked there that says "Set Stereo Out as Main Mix". Well, yes, that's what I want. I see that it is checked. I cannot *uncheck* it. And nothing goes to the meter for the Main Mix.

*grumble*

Let me create a new stereo bus, Test. Now, I go to it, right-click and I can click the option to "Set Test as Main Mix". Ok. That worked. Let's go back to Stereo Out, which is no longer checked, and select the option there. And let's hit the Play button.

Why, look! The signal is now going to the main meter.

I guess I can go delete that Test bus now.

The things you learn...

I am currently fighting with some technology at work that I don't know nearly enough about, so I figured I'd just toss the problem out to my friends and see if someone who knows more about this has the necessary technical incantation to get me around what looks like a configuration problem.

I'm running a VM with Windows Server 2022 that is getting an IP address from a DHCP server. I need to run some additional database software that requires Oracle Linux. Using Hyper-V, we have been able to get the Oracle Linux environment running inside the VM, which is good. The problem is that we have not managed to set up the networking in a way that allows my applications running inside the Windows environment to talk to those databases in the Linux environment.

It seems like the right solution would be to assign a static IP to the Linux environment from somewhere outside of the DHCP range that we're getting addresses from for the Windows VM. That way, the address of my database servers won't be changing. However, I know little enough about this that it's possible that I'm entirely wrong. I also need to allow the Linux server to go out with a browser connection if for no other reason than that I need to pull software in to be able to install it there.

So far, this is defeating me.

Anyone out there know enough about this to give me a few suggestions? I've been applying Google-fu to the problem, but haven't found the right source yet.

Thanks!

I came down this evening and powered up my UA gear (in the correct order) and then fired up Cubase to see if the Capitol Chambers plugin was still loading without error. It is.

I have advised UA tech support of this and wait for their response.

About two weeks ago, I realized that I was having a problem with the loading of my Universal Audio plugins, because loading a copy of the Capitol Chambers reverb immediately results in the plugin disabling itself. This is bad, because although that reverb is a bit of a DSP hog, it is a very nice reverb. I opened a trouble ticket with UA and we have been going back and forth trying to resolve things, which doesn't go as quickly as it might since I don't have time to run down to the studio every day. The ticket has already been closed and reopened once, because I failed to respond in three days.

The tech on the other end has asked for a copy of my system specs (which I've since sent) and suggested that the problem might be that I needed to update my system BIOS. That's not impossible, although I'm pretty sure that this plugin loaded on the Thunderbolt incarnation of the studio computer with an earlier version of the UA software. I checked the ASUS website and a much newer version of the BIOS was available. I downloaded it, I installed it, the computer rebooted --

And hung up without actually starting up.

Well, that sucks.

A bit of research indicated that I might get the computer back by clearing the CMOS memory and then re-entering the appropriate BIOS settings. This required unplugging the computer, opening it up, shorting two pins on the motherboard together with a screwdriver, and rebooting again.

Happily, it worked and the computer reported that I'm running on the new BIOS.

Loading that UA plugin? Nope, not going to happen.

I have sent this information along to the UA tech. We'll see what comes back next.

I was chatting with Smac today and we got into a discussion of trying to remix old albums. This led me to share a blog post with him that I'd found about how they've used the technology Peter Jackson developed for the "Get Back" documentary to take the Beatles' "Revolver" album and decompose it to the original unmixed tracks, which is an amazing feat of technology.

Of course, where there is incredible technology available, there is usually a less-good, but cheaper version available to the masses. Reading the blog post about the blog post about "Revolver", I found that you could do a more basic decomposition of a mix to multiple (if not individual) tracks using a tool that I already owned, Izotope RX 10.

A tool I owned, but not one for which the latest version was installed on my computer, as it turned out when I wandered to the basement later. After a bit of futzing around, I eventually got the two new Izotope versions that I was entitled to due to an earlier upgrade installed on the studio computer. Then it was a question of what to play with.

I grabbed Barry and Sally Childs-Helton's "Lady Snowstar Supernova" from "Tempus Fugitives". RX 10 decomposed it to four tracks, although the bass track was functionally empty, so really I had "Vocals", "Percussion", and "Other Instruments". And then I put the track back together in Cubase, making some adjustments.

It was an interesting process. Not as good as having the raw tracks available, of course, but definitely something that can be worked with.

I had great plans for this weekend to get a great many things done.

The universe seems to have had other plans.

It all started last night when the Internet proceeded to go wonky in our house. This happens periodically and my perception (perhaps inaccurate) of the SLA for the household is "Nothing must break. Ever." I may need to revise this perception, as when I growled at Julie when she asked when the Internet would be back up, she responded, "I was just asking."

But there was no connectivity from the second floor WAP (a TP-Link router configured to be a WAP, actually), which happens periodically. More concerning was that neither my home computer nor the work computer that are hardwired to the router downstairs had connectivity either. That's bad.

The WAP in the basement didn't seem to have connectivity as well, so it was starting to look like the right course of action was to reboot the modem. And that's what I did when I went to the basement. I verified that the computer there which is hardwired to the modem was working and headed back upstairs.

Where I found that nothing was working.

It was now time to make a big mistake. The Ethernet jack was a retrofit that we did nearly 20 years ago when we got a contractor to run conduit to the second floor from the router in the basement. I figured I'd just unplug the cable and plug it back in. It turns out that 20 years of settling and poking around behind the return of my desk had managed to cause the plug to shift. I was able to unplug it, but the jack was far enough off-centered in the hole that plugging it back in wasn't about to happen.

I didn't know at first that was the problem. I found that out later by tugging the very heavy return of the very heavy desk away from the wall so I could stick my phone down there and take a picture. It did not look at all good.

If I could find my stubby screwdriver, I could unscrew the plate and take a good look at it. The screwdriver was, of course, nowhere to be found. There are no other screwdrivers short enough to have a hope of fitting behind the return (the very heavy return) without figuring out how to pull it much farther away from the wall.

The score at this point: the Internet was available by hardwire in the basement and the Wi-fi provided by the cable modem was working, so there was at least some connectivity. It was time to give up and go to bed.

The next morning, I got up, sorted bath towels with Gretchen as we tried to decide what to keep in our bathroom, what to send to the back bathroom, what should go downstairs, and what should go to Goodwill. While considering this and the likelihood that I was going to have to go to Home Depot to buy a new Ethernet outlet jack and then punch in the wiring for the fool thing behind the return (the very heavy return), I remembered that the stubby screwdriver might be on the bookcase in the studio. I sent K to find it and she succeeded, which was the first really good thing that had happened in this particular adventure.

I stuck the end of the screwdriver into the jack and tried to convince it that being slightly better centered would be a good choice. Then I managed to click the Ethernet cable back into place. Unfortunately, this still left me with no connectivity upstairs.

I threw a load of shirts into the washer and the family went out to IHOP for lunch. When we got back, I turned on the Cubs vs. Brewers game, and a few minutes later, headed down to the basement to put the shirts in the dryer and go back to fighting with the Internet.

The modem looked to be ok. I still had connectivity on the hardwired computer there, but the external WAP wasn't working right. I managed to connect to the WAP from the hardwired computer and noticed that it was saying that the Internet connection wasn't there. Looking at the back of the WAP, I realized that the network wire was plugged into a network jack, not into the jack labeled "Internet". Well, that had always worked before, but who knows? Let's try it. I plugged the wire into the Internet jack, the WAP allowed that there was Internet, and my phone was able to connect to the WAP. A small victory!

As part of my messing around with the connections in the basement, I had tried plugging the cable going upstairs directly into a jack on the modem instead of routing it through the small switch in the basement. It hadn't helped. Cables had been rewired in all sorts of ways, finally landing back in the original configuration.

I went to check on the shirts, which were not quite done, but which would go into the wrinkle care tumble mode when done. Fine. They could wait. I would go upstairs to the office and try to sort this out.

Oh, look. The Cubs game is on the TV. I remember turning that on. And I sat down and ended up watching the Cubs beat the Brewers in extra innings, which was way more fun than fighting with the Internet.

And then it was time to go get the shirts, hang them up, and go fight with the Internet in the office.

At least I was able to use my phone to search the Internet for ideas by connecting to the basement WAP. And on one of those searches, I found someone who explained how their computer dock that was connected to a switch had bollixed up *all* of the traffic through the switch until it was unplugged and plugged back in.

I unplugged the dock from the work laptop. I plugged it back in. I now had Internet on both the work laptop and the home desktop. And there was much rejoicing.

Gretchen went out for a drive with Julie while I tried to revive the WAP upstairs. Since moving the connection from the network jack to the labeled Internet jack had worked in the basement, I figured I'd try that here. This turned out to be another mistake. I spent a bit over an hour rebooting, resetting to factory configuration, and reconfiguring the WAP only to be completely unable to get it out of router mode successfully. Happily, I was able to grab the network wire from my computer so I could make a direct connection with the WAP/router or this would have been nigh unto impossible.

I even loaded fresh firmware onto the WAP/router. It did not help.

I finally realized after searching the Internet in vain that *this* variation of WAP/router positively would *not* let you reconfigure it as a WAP if you plugged the network wire into the Internet jack. It *must* go into a network jack.

And now everything is working again.

For a while.

I will try to get some more non-Internet-related things done tomorrow.