|
The only time I've had lag in a wireless mouse or trackball is when it's dirty.
|
|
|
|
|
wired mouse and keyboard all the way. Minimum latency. No line of sight, issues. I have USB port on keyboard for mouse to reduce cable clutter. Perixx model TK566 keyboard. Cheap and has 2 USB ports.
"A little time, a little trouble, your better day"
Badfinger
|
|
|
|
|
jmaida wrote: wired mouse and keyboard all the way. Minimum latency. No line of sight, issues.
lol...well if there was a line of sight problem then you should check which universe you woke up in.
|
|
|
|
|
 actually had a line sight issue when the wireless mouse/keyboard USB interface (little USB pimple) on computer was sometimes blocked by a portable USB drive. Mouse and keyboard lost signals from time to time.
I have other issues with wireless mouse/keyboards, but this ended their use for me on my system.
"A little time, a little trouble, your better day"
Badfinger
|
|
|
|
|
jmaida wrote: No line of sight, issues.
But you do have line of USB cable issues 
(I like to retire to my recliner, keyboard in my lap, mouse on the armrest. That would require a USB extension cord, which would too easily be pulled out every time I swing my recliner around to put another log in the open fireplace, or pat the dog, or whatever. I know from experience (although that is long ago, but USB Extension cables haven't changed).)
Religious freedom is the freedom to say that two plus two make five.
|
|
|
|
|
 True, wires are wires of any length
I used Logitech wireless mouse and keyboard for years, but was problematic.
I upgraded but still issues.
I do not use a recliner (Lazy Boy as known in US),
but if I did, I would figure out a way to make wires work.
I may return to the wireless world, but happy right now.
"A little time, a little trouble, your better day"
Badfinger
|
|
|
|
|
I also have not noticed lag w/ my wireless mouse. re/ insistence I insist on softness and low coefficient of friction. I utilize a Teflon cloth atop a soft mouse pad w/ double sided sticky cloth betwixt and between also after market Teflon mouse feet. Much less friction than standard mouse pads. I am surprised the mouse and pad manufacturers do not utilize same.
|
|
|
|
|
Update: Thanks to Daniel asking me just the right question I figured out a way to hack my way around the lack of Span/ReadOnlySpan with Deslang by making a dummy with the same name and some appropriate stub methods. I don't actually use that code directly, but my Deslang engine does when it needs to resolve System.ReadOnlySpan<char> which again, it must do under the .NET Framework where that type doesn't otherwise exist. Deslang just needed to eat. 
So I'm still going to target the DNF. I may also provide a roslyn generator based alternative to using the codedom but it seems it's a bit of a learning curve, and more work given the tools I already have for CodeDOM stuff, like Deslang. Also it may require the end user to have more microsoft fluff installed and running in order to use it, so all that is something I need to run down. Progress!
_____________________________________________________________________________________
I don't know whether to target .NET Framework with my FA library anymore.
The issue is that you don't have spans in DNF, and now my regex stuff uses it to keep up with Microsoft's engine (which does the same thing)
This created a snowball of issues. For starters, I use a tool called Deslang to turn a subset of C# into a generic Abstract Syntax Tree that can be rendered into C#, VB.NET, or something else. I use this tool to prepackage shared code that I want to "generate" for the user in any language they specified.
The issue with building the AST directly is it's very verbose, taking a solid paragraph of code to declare a local variable. Deslang does it for me and produces code that has that AST already cooked into an object graph.
Deslang uses some reflection magic. And that magic is .NET Framework only. However, it cannot see ReadOnlySpan<char> because it's a span, and those don't exist in the DNF. 
Ergo, I cannot generate span code this way.
I'm currently using the codedom and thinking of moving over to the far more modern roslyn services and abandoning DNF altogether.
There's one issue though, and that is that if you were using my code generators to make build tools of your own, you won't be able to make self contained executables anymore because that's a .NET Framework feature only. The other cases have multiple files per assembly which makes their use as build tools a bit uglier. Not insurmountable, but less than ideal.
Still, in order to fully support the .NET framework at this point I'd have to fork my runtime string matcher code and my compiler plus my code generator too, all so I could generate a less performant, .NET Framework friendly alternative. I think this is basically what Microsoft does, but they have staff. I don't.
I really like the .NET Framework. I like Winforms. I like self contained executables. They do run on linux regardless of what microsoft says (except winforms of course) because Mono has supported DNF from the beginning.
I never really understood the need to make several more .NET "kinds", but I guess now that span doesn't work in the DNF I've got some hard choices to make.
Am I just getting old? This all seems like a lot of unnecessary trouble Microsoft made for people.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
modified 14-Jan-24 11:22am.
|
|
|
|
|
I too like .NET as an environment, but as long as Microsoft is the only full implementor of the language and libraries, one is liable to run into the kind of issues that you mention. They are the only ones that specify the CLR, and everyone else must fall in line.
That is one of the reasons that I prefer to work in standardized languages such as C++. I know exactly what I'm getting (and not getting), and if a compiler implementor decides not to support part of the Standard, there are other fish in the sea.
Could you not create a wrapper around the parts of Microsoft's DNF that you need, providing the same functionality where it exists, but your own implementation where it doesn't? That would at least allow the rest of your code to remain unchanged.
Freedom is the freedom to say that two plus two make four. If that is granted, all else follows.
-- 6079 Smith W.
|
|
|
|
|
Daniel Pfeffer wrote: Could you not create a wrapper around the parts of Microsoft's DNF that you need, providing the same functionality where it exists, but your own implementation where it doesn't?
Unfortunately, no. Read only spans kind of work like string views do in C++ except unlike C++ you have to rely on the CLI to support the feature. You can't even hold Span/ReadOnlySpan instances as fields in classes or structs. It will COMPILE error.
Edit: However, I thought about what you asked and it gave me an idea. To that end I created just enough of ReadOnlySpan<t> in C#/Slang that Deslang could find it and use it. I don't actually use the code for the ReadOnlySpan<t> cutout I made - but it's enough that Slang can resolve the methods on it. So that works.
namespace System
{
internal class ReadOnlySpan<T>
{
public override string ToString()
{
return null;
}
public int Length { get { return 0; } }
public ReadOnlySpan<T> Slice(int position, int length)
{
return null;
}
}
}
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
modified 14-Jan-24 5:45am.
|
|
|
|
|
Never worked with spans and never needed them, I think.
So I'm going to abuse this opportunity to ask a programming question in the Lounge.
What do they do, would I need them in an average LOB application, and if so why and how?
I know I can google it, but you can probably answer my question(s) better.
|
|
|
|
|
With the type of development you do, I don't see you using them, at least directly.
They're a bit into the weeds.
Okay, so like normally
string foobar = "foobar"
string bar = foobar.Substring(3,3);
That's fine of course. One can copy strings. However, it's not efficient, and if you needed to do it a million times a second you might be sweating it.
string foobar = "foobar";
ReadOnlySpan<char> sp = (ReadOnlySpan<char>)foobar;
ReadOnlySpan<char> sbar = sp.Slice(3,3);
Spans basically point to another thing's memory directly without copying. They are a useful way to speed up low level operations on strings and arrays when you need a "view" into them without copying the data.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|
|
Thanks, figured it'd be something like that.
I'll stick to the more readable, concise and expected string.Substring
|
|
|
|
|
Right? Generally developing at the level you do you don't have the time or need to get into the gritty details.
Suffice it to say that MS uses Spans under the hood to make your higher level code perform better than it otherwise would. Doing things like XML and JSON processing, or even *gasp* Regex scanning involves a lot of string manipulation, and therefore under normal circumstances, a lot of copying. Spans, used judiciously can dramatically reduce that copying that's going on under the covers, and Microsoft has taken care of that in their core libraries, so you don't need to worry about it.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|
|
honey the codewitch wrote: Deslang uses some reflection magic. And that magic is .NET Framework only.
I am trying to parse that statement and failing.
Is "Deslang" a third party library or your name for something? I did find the first but unclear how you would be using those.
Also not sure what "magic" you are referring to. Perhaps you are referring to the C# language specification only? Because other than that I think reflection is always available via the System.Reflection namespace. Certainly assemblies always contain the metadata and they must because runtime link resolution requires that information to validate and resolve calls.
This latter part could just be because I never memorize what I am stuffing into a C# app. I just throw things at it until I get the parts I need.
|
|
|
|
|
Deslang is a code generator for generating code that generates code.
Deslang: From Code to CodeDOM and Back[^]
Reflection is available in Core and Standard, but not the particular bits I am using. I get a NotSupported exception coming from deep within the bowels of my CodeDomResolver class due to some reflection chicanery not being functional under anything other than Framework. I haven't run down all the details yet since I found another path forward.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|
|
Wordle 939 3/6*
⬛⬛🟨🟩⬛
⬛🟩🟩🟩⬛
🟩🟩🟩🟩🟩
|
|
|
|
|
Wordle 939 3/6
⬜🟨🟩⬜⬜
🟩⬜🟩⬜⬜
🟩🟩🟩🟩🟩
|
|
|
|
|
Wordle 939 3/6
⬛⬛🟨⬛⬛
⬛🟩⬛🟩🟩
🟩🟩🟩🟩🟩
|
|
|
|
|
Wordle 939 3/6*
⬜⬜🟨⬜⬜
⬜🟩🟨🟨⬜
🟩🟩🟩🟩🟩
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Wordle 939 3/6
⬛🟨🟨⬛⬛
🟨🟩⬛🟨⬛
🟩🟩🟩🟩🟩
Ok, I have had my coffee, so you can all come out now!
|
|
|
|
|
boss " ...can you comment your code? "
coder "... my code is self commented "
corollary, not really funny
caller "...I am having a hart attack..."
911 operator " ... everything is peachy fine here..."
modified 13-Jan-24 17:16pm.
|
|
|
|
|
|
Alternate version:
boss: Did you even test this? 
coder: Well, it worked on my computer. 
"Go forth into the source" - Neal Morse
"Hope is contagious"
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
