English Amiga Board


Go Back   English Amiga Board > Coders > Coders. Asm / Hardware

 
 
Thread Tools
Old 23 September 2023, 14:49   #41
Thomas Richter
Registered User
 
Join Date: Jan 2019
Location: Germany
Posts: 2,844
Quote:
Originally Posted by Karlos View Post
OMG you left out Rust! Rust is the only language that saves the developer from the horrors of memory, concurrency and the hideous dangers of variables that, you know, vary. Or at least that's what the indoctrination session... I mean introduction lesson said

Yes, I did. I learned many nice things about it, though. It's pointer "ownership" concept I consider particularly valuable, even more so as it is close how I work with pointers in C++, so it is pretty close to "what I do".


It's for me on the "observing" side, see how it develops, or whether this is just a "current trend". If it is still around in a couple of years, I will consider.


I need more than just a compiler to be happy - I need the entire ecosystem, and that includes profiler, debugger, coverage tools and so on. C and C++ are very mature in this direction, and so are python, java etc.., for rust all I can say is "I do not yet know". We'll see.
Thomas Richter is offline  
Old 23 September 2023, 14:59   #42
meynaf
son of 68k
 
meynaf's Avatar
 
Join Date: Nov 2007
Location: Lyon / France
Age: 50
Posts: 5,204
Quote:
Originally Posted by Thomas Richter View Post
Assembler? Yes, for almost everything it is the wrong tool. Not for everything. Low-level driver code that interfaces hardware is probably the only exception that comes to my mind. P96 drivers are (to some degree) in assembler. For example, higher level code, like GUIs, are not the domain of assembler. It's just more work, and no benefit at all.
The point was, this makes your "Learn a language, learn a culture" a little bit irrelevant as a comparison...
meynaf is offline  
Old 23 September 2023, 15:32   #43
Jobbo
Registered User
 
Jobbo's Avatar
 
Join Date: Jun 2020
Location: Druidia
Posts: 369
JFC get back to your own thread! I take it back, the quiet was better than this nonsense!
Jobbo is offline  
Old 23 September 2023, 15:39   #44
TCD
HOL/FTP busy bee
 
TCD's Avatar
 
Join Date: Sep 2006
Location: Germany
Age: 45
Posts: 29,798
Play 'interesting' games, win 'interesting' prizes

TCD is offline  
Old 23 September 2023, 16:25   #45
Pyromania
Moderator
 
Pyromania's Avatar
 
Join Date: Jan 2002
Location: Chicago, IL
Posts: 3,270
Quote:
Originally Posted by Thorham View Post
How pedestrian. C# is much better.
You Microsoft fanboy is showing.

Pyromania is offline  
Old 23 September 2023, 17:02   #46
Karlos
Registered User
 
Join Date: Aug 2022
Location: UK
Posts: 2,963
Quote:
Originally Posted by Thomas Richter View Post
Yes, I did. I learned many nice things about it, though. It's pointer "ownership" concept I consider particularly valuable, even more so as it is close how I work with pointers in C++, so it is pretty close to "what I do".
So it's not that I don't appreciate the goals of Rust, but I have to say, I find the implementation and evangelism extremely annoying. There's a conversation that has played out many times. In this conversation, I am in a position similar to meynaf in the conversations we have here:

Rustacean: "C and C++ are inherently unsafe. Rust provides.... long list of usual features trotted out, whereas C and C++ are broken by design, have broken implementations that produce unsafe binaries.... etc."

Rustacean: Proceeds to give examples.

Me: "Your criticisms of C++ in particular seem to be showing me the defects in the 'C with classes' style of C++ that was common back in the 90's. Have you tried modern, idiomatic C++? Say perhaps C++20 ?"

A comparison of language features ensues, resulting in a stalemate where it is apparent that there are no real safety advantages of Rust over idiomatic modern C++.

Rustacean: "Well OK, but nobody writes C++ like that and all C++ out there is older and buggier so the point still stands that it can't be trusted to give safe binaries. Rust can be formally proved and ... insert other frequently stated hyperbole here"

Me: "Does that include LLVM?"

Rustacean: "Of course."

Me: "The same LLVM that rustc depends on to compile binaries?"

Rustacean: "...."

Me: "Then how is it safer? At some point you handed the formal specification of your code to an unsafe compiler binary written in an unsafe language that has bugs and will, by your argument so far, be flawed. Why isn't rust self-compiling anyway?"

Rustacean: "The compiler works fine. We have to fix all these other things. Just look at the Linux kernel! It's begging for the type of safety that Rust provides!"

Me: "By rewriting them in a language that doesn't necessarily compile for all the targets supported by the Linux kernel? How does that safety work when it has to interface with a ton of unsafe code anyway?"

Rustacean: "And all these other terrible, unsafe C++ relics ..."

Me: "As opposed to upgrading their code to more modern C++? I mean, rewriting a ton of stuff in a new language requires potentially replacing your developers with ones not only familiar with Rust but equally able to work with and understand code written in a language that many of them seem to want nothing to do with. Oh, and for a much higher salary expectation. Perhaps upskilling your existing engineers a bit and buying a decent SAST package would cost less."

Rustacean: "Variables are bad."
Karlos is offline  
Old 23 September 2023, 17:50   #47
8bitbubsy
Registered User
 
8bitbubsy's Avatar
 
Join Date: Sep 2009
Location: Norway
Posts: 1,661
Quote:
Originally Posted by meynaf View Post
Hopefully assemblers are allowed to make optimisations. This way we don't need to care about branch sizes anymore, what a relief.
But you can turn them off if you want to. You can even do that locally.
I'd rather optimize my asm by hand so that I know it's equally fast on every assembler.
8bitbubsy is offline  
Old 23 September 2023, 17:58   #48
meynaf
son of 68k
 
meynaf's Avatar
 
Join Date: Nov 2007
Location: Lyon / France
Age: 50
Posts: 5,204
Quote:
Originally Posted by 8bitbubsy View Post
I'd rather optimize my asm by hand so that I know it's equally fast on every assembler.
For branches it may be just not doable, especially when using macros containing branches or conditional assembly making code size to change.
meynaf is offline  
 


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Similar Threads
Thread Thread Starter Forum Replies Last Post
Best settings to keep laptop cool and quiet? MartinW support.FS-UAE 11 20 July 2019 18:11
Quiet in 'ere... meltdown Nostalgia & memories 7 04 July 2019 12:51
Quiet/Low sound on A600 8bitbubsy support.Hardware 9 24 October 2009 10:00
quiet ambient mods Marcuz request.Modules 17 07 December 2008 11:38
wanting a quiet scsi HDD rare_j support.Hardware 8 27 October 2006 13:39

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +2. The time now is 08:41.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2023, vBulletin Solutions Inc.
Page generated in 0.15543 seconds with 14 queries