7

There is a question about rendering overtype in HTML that's on-topicness has been questioned. In the comments Leo B. asked the following:

Food for thought: would a question about a way to reproduce the bit-manipulated PWM sound of a retro-computer, or a way to reproduce the CRT artifacts of a videoterminal of a mainframe on non-retro machines (PCs, Macs) in non-retro emulators using a non-retro library (say, SDL) be off-topic?

What is our policy on such questions, and should the existing question be closed?

1

3 Answers 3

Reset to default
8

To concur in part with (and to object in part to) the answer by @Chenmunka : IMO, asking how to mathematically model a particular RC feature for the purpose of a faithful reproduction, be it the diaphragm of a speaker, or the raster ray or a CRT (or, say, the wobble of letters on an LP printout due to varying interrupt latency), should be on-topic; how to implement that mathematical model in a particular language/library/OS - off-topic.

Then it follows that my initial question if off-topic, because the modeling aspect of overprinting per se is trivial, and the only on-topic side of it was hinted at by @supercat in mentioning that the ink smear effects due to overprinting can be reproduced and modeled as well.

1
  • 2
    I completely agree with this answer. Note that code in an answer would (most likely) be allowed if it helped to explain the specific behaviour.
    – wizzwizz4 Mod
    Jan 13, 2017 at 7:20
5

To start the discussion: We are into the nebulous fringes of topicality here.

I believe that the question you refer to is better asked somewhere like StackOverflow as it is an HTML / CSS coding question - not really retrocomputing.

Questions like "Can I set up VMWare to run CP/M" would, I believe be off-topic as they are about configuring a modern software package - ask on SuperUser

Questions like "How can I get my Raspberry Pi to talk to my Centronics printer" would be on-topic if they concentrate on the signals required to talk to the hardware. As discussed in this question.

So, to Leo B's points:
Replicating a sound is a coding issue - off-topic.
Replicating CRT artifacts - If you ask about how the artifacts arise and were generated in the first place - on-topic.

I suspect this is a discussion that will run and run. I hope this question will get more comments.

1
  • 1
    And conversely, if you ask about how the sound arises and was generated in the first place, on-topic. If you ask about how to implement the CRT artefacts in SDL, off-topic.
    – wizzwizz4 Mod
    Jan 12, 2017 at 18:58
3

I would tend to suggest that the Retrocomputing site would likely have different kinds of people reading it than a site that was targeted toward modern HTML programming. The latter might have more people who know how to overlay text in HTML, but the former might be more people who know what the effect in question would actually look like when printed on vintage equipment. Further, I would think that the Retrocomputing site might be more likely to have someone who had actually used the HTML or other techniques for the purpose of creating the look of vintage FORTRAN printouts.

IMHO, the most important factor in deciding which site would be appropriate for a question would be which site would most likely have people who would be interested in what the question describes and able to answer it. A question "How can I generate real-time audio in Javascript" might be best answered on Stackoverflow even if the intended purpose would be to emulate an old machine, but a question "Why does my attempt at emulating the audio of an XXY sound terrible" might be more appropriate on Retrocomputing; even if the problem might turn out to be Javascript related, someone else trying to emulate old computer audio in Javascript might benefit from being able to find the question here.

5
  • I see where you're coming from... But would a question about "Why does my attempt at emulating the audio of XXY sound terrible?" might be better asked elsewhere with a copy of the algorithm provided, because the problem is unlikely to be useful to future people interested in Retrocomputing.
    – wizzwizz4 Mod
    Jan 17, 2017 at 7:26
  • I would expect that a fair number of people who visit retrocomputing are going to be interested in writing emulation software or adapting it to different platforms. Good emulation requires balancing a number of different issues--some purely related to understanding the emulated system (how does the timing of Q register writes affect affect the phase of the generated output), some purely related to understanding the target (how can I turn off smoothing when rendering my output), and some related to emulation as a concept...
    – supercat
    Jan 17, 2017 at 12:39
  • ...(how can I efficiently "batch" emulated actions so as to achieve decent performance without affecting emulation accuracy), and requiring an understanding of both the system being emulated and the target.
    – supercat
    Jan 17, 2017 at 12:45
  • But problems are likely to be duplicates on SO or incorrect implementations of an algorithm.
    – wizzwizz4 Mod
    Jan 17, 2017 at 17:25
  • @wizzwizz4: It's not always obvious whether a particular problem will need to be fixed on the "emulated system" side or the "target system" side, and sometimes a fix may require an understanding of both systems [e.g. someone wanting to emulate the C64 filters would have to know how they work, and also know what kinds of filter algorithms the target system could process efficiently].
    – supercat
    Jan 17, 2017 at 17:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .