76

(117 replies, posted in Sinclair)

Yep, both of my tracks are Ear Shaver EX, and a few more to come sometime soon. I was trying to explore the possibilities of this particular engine and the potential of the whole engine-per-row approach. It is really promising indeed, although I'm thinking to EX it a bit more.

Totally, the strongest compo in years.

77

(117 replies, posted in Sinclair)

Complete results yet to come, as the prizegiving is going right at the moment, but congrats, AtariTufty (2nd with Synthetic Heartbeat) and utz (1st with A Strange Game)! Finally, utz's entry got a praise it deserves! My personal favorite this time was AER's Cosmic Puppy, though.

78

(117 replies, posted in Sinclair)

Yes, online voting is still open, results tommorow: https://events.retroscene.org/dh2023/beeper

79

(117 replies, posted in Sinclair)

Color me impressed, guys! 7 songs, and the humble beeper is still at rise! So many cool sound colors, so many neat finds. The compo is still kinda crowded with my engines, tho. Sorry, was busy, had no time to add the recent utz's developments to the tracker yet, but I believe some of those are the next best picks.

80

(4 replies, posted in Other Platforms)

A regular VIC-20 sound chip sound is heard in some demoscene productions. The chip is actually interesting to explore, as it is capable to produce not just square wave, but bit patterns. I've tried to exploit this trick in the PETSCII Robots driver, but it didn't turn good enough. There is a lot to be explored. A glimpse of this is heard in the video linked by utz.

As for 1-bit stuff on VIC-20. Well, it is 6522 then, so exactly the same thing as the stuff that I did for the PET.

A spoiler: a native PET PeskySound-driven editor has been put into the queue of projects some time back. Should be portable to the whole 6522-equipped Commodore range easily enough.

81

(4 replies, posted in Sinclair)

Great! I thought of eliminating the row noise as an obvious engine evolution step to take, but never really done any engine with that.

82

(117 replies, posted in Sinclair)

DiHalt 2023 is coming up! Jan 5 it is, so there is some time to get your bits together!

A note, the tables for the data structures in the docs are messed up for Pulsatilla and Pindsvin. Not really sure how interpret them at the moment, Pulsatilla's table somehow mention ch5 even.

Just as I thought just a dozen engines left to be added to 1tracker, and we're done, three more gets added. Great sound!

85

(3 replies, posted in Sinclair)

Pretty cool, dynamic tune. Guess it is time to put some work into a Phaser4.

86

(3 replies, posted in Sinclair)

That's Tritone Digi as the file_id.diz says.

This is kind of an answer of the Soviet electronics to the graph calculators such as Ti-8x. To my knowledge, it is a totally unique design, not a copy or derivative of something (like most of other Soviet electronics), yet it is based on the cloned CPU architecture, yeah - the 16-bit LSI-11, which is at heart of the BK line of home computers. It even had a dock station MK-92 with a 4-color plotter and tape interface. First prototyped in 1986, released in 1988, produced till 1992. At the time of release it used to cost nearly as much as a small Soviet car. Not a vapor hardware either, I personally seen it in a shop being a kid.

It was the last mass produced device of such kind in the USSR. There was a similar but 8086-powered MK-98 follow up in the works, a few production quality fully functional prototypes were found, but it was never available to the mass market.

Of all places and possibilities, the (probably) first PCSPE use in a published production that has been done by someone but me, is the Bad Apple port to the USSR designed pocket computer MK-90. https://www.youtube.com/watch?v=1qESkvDjKXA

Just released a new demo for the Commodore PET. Complete with the CB2 1-bit music! This time the music has been created with Reaper and my PETCB2 VST.

https://www.youtube.com/watch?v=3FFPCuDqi_g

90

(2 replies, posted in Other Platforms)

It seems the humble PET finally gets some love lately!

An insane video hack, would never ever think of something like that.

91

(164 replies, posted in Sinclair)

Oh, you probably creating the source block at the very first row. It is a common issue in the code, will fix it ASAP.

Edit: fixed, uploaded.

92

(164 replies, posted in Sinclair)

Alright, here is v0.44:

- A fix for the RowOpt feature to properly handle existing speed changes.
- Reference blocks, block parenting
- Copying/inserting a block by its name

I decided to keep the idea of reference blocks out of the main focus of the project, and implement it in a minimalistic way, just to have it for specific cases, and not make it an integral part of the pattern-less concept. Let's see how useful it actually is.

93

(164 replies, posted in Sinclair)

I was thinking more on the idea with references, and come to the same conclusion, it is better to have it as a manual 'sync' command rather than an automatic update. Besides of the possible unwanted changes, it brings too much trouble handing the insert/delete position (as all data shifts inside all blocks). I'm going to think some more before trying to implement it. At least a simple block copy function is possible and should not hurt anything, like, press something, input an existing block name, and it'll get copied into current position.

I implemented RowOpt to cram down all songs into my 1-bit collection, it worked just fine at the moment, but yes, it may have issues. One thing is that it won't optimize above the slowest speed, which is vary for different engines. I confirm the issue with the example, will look into it.

94

(164 replies, posted in Sinclair)

Thanks! One of reasons why I'm doing updates is to improve my creative throughput, make the process more efficient. I'm also discovering some issues and new ideas as I'm working on more songs. So it is connected, more music releases - more updates to 1tracker, and vice versa.

95

(164 replies, posted in Sinclair)

A minor update to v0.43:

- Main/side display in the header with M/S
- Rows with contents on the other track marked on the left with > < (depending on current track)
- Inactive track gets displayed in the dual mode (F6)
- Marker name copying for the whole-width copy/paste operations
- Module filename and path displayed in the window header

I also got an idea for another big improvement to the pattern-less system, a link for the already defined named blocks, but I'm still not exactly sure how to implement it interface-wise. It certainly would be a front-end side feature, so no changes needed for the engine scripts. Basically, we already have blocks of variable length and with an optional name. That name can be used as a reference for a 'link' that would synchronize contents of the same-sized same-named blocks. So you would be able to edit one block, and it automatically updates all linked blocks.

Edit: actually when I was explaining my idea here, I got a better understanding how to do it. Hopefully it won't turn the tracker into another hard-to-fix glitch fest.

Probably the first MuzCell song in two decades, by FADE: https://www.youtube.com/watch?v=9nXsgoMdg8U

Basically just that, a version of the PCSPE retargeted for the PET. Why limit it to just PeskyTone/PeskySound and 1tracker support, if we can go further. Now only the RAM is the limit.

A simple 6502 player for the VIA log files is included as well.

Download v1.0

98

(8 replies, posted in Sinclair)

The worst part is that it uses the whole/half/quarter note duration system, so adapting it for a tracker isn't too easy. Maybe it is better to separate the sound synthesis core from the rest of the engine, and do our usual pattern driven stuff around it. Will lose some authenticity, though, but I think the sound won't be much different.

99

(43 replies, posted in Atari)

This was a matter of discussion for several years on ZX scene, and the conclusion was that the difference between these volume levels is very subtle (a fraction of a volt) to be noticed by the ear; and this behavior is not replicated on most Amstrad models and clones. To my knowledge, no one ever actually tried it, though.

100

(164 replies, posted in Sinclair)

Ctrl+F6 selects the currently displayed/edited track, and when the dual mode is off, only currently displayed track plays.

Yeah, it need some thoughts to be put in to make it even more useful. Will try to develop it eventually.

Damnit, the AngelScript is a troublemaker sometimes. I don't have these particular issues on my end in the Windows build, but it sometimes drops in error messages about Z80Ass anyways. Feels like some vars/arrays don't get cleared between script calls.