ProDigit wrote: ↑
Sun Aug 18, 2019 12:41 pm
thagrol wrote: ↑
Sat Aug 17, 2019 1:27 pm
ProDigit wrote: ↑
Sat Aug 17, 2019 1:21 pm
I really can't explain any better, but if you're not Willing to do the research, and just GOOGLE the topic, I really can't do much about that.
Couple of reasons that won't help your (or my) position:
- Traditionally it's your job to support your argument with evidence not mine.
- There's no guarantee google will return the same results for me as it did for you and in the same order. We might not use the same search terms and google does result filtering based on your search history too.
- I don't use google. (Because of thier logging but that's an entirely different discussion)
Not true in case of fact.
Then you'll have to read up on fact. It's not my job to provide you the proof you can provide yourself.
I don't quite follow you here.
Are you claiming it's not true that anyone needs to provide evidence to support thier position? That it's not true that different search terms (you haven't told us which terms you used) provide different results? That google doesn't tailor search results based on your past history? That different search engines don't return diferent results.
As for who has to provide evidence to support their position, look at it this way: it's not the defense's job to provide evidence to support the prosecution's case. That's what you seem to expect us to be doing.
In fact, you'd have to provide the proof that any opposing (counterintuitive) thought could be true.
In my case, there are thousands of documents supporting the delete swap theory.
So provide links to some of them
In your case, there aren't many, if any at all, supporting the idea of swap improving the speed, most of these documents are pre-1990's era, when RAM was 512MB or less.
Have you actually read and understood any of my posts?
claiming that swap increases performance. I've never said that and in one of my earlier post I have stated explictly that swap reduces performance.
What I'm trying to explain, which you are persistantly mis-interpreting is that the presence of a swap file or swap partition does not mean it is being used (see my earlier post re the free command) and that an unused swap file/partition has no negative effect on system performance.
How do you know the system doesn't use swap? Or are you just presuming this?
Is there any evidence on that claim?
and see my earlier posts.
The system may not use the swap, but it also may.
If it does access the swap, it will definitely slow down the whole system!
When deleting the swap, it guarantees it doesn't use the swap, and the only slow downs will happen when one is running out of memory.
No. Without swap when the system runs out of RAM you'll get hangs, crashes and other instabilities.
There's no reason to have a swap file on Raspbian when you're having 1-4GB of RAM (unless you go crazy, like having animated desktop backgrounds, 3D accelerated aero-like themes, etc...).
There's no reason not to either.
I can do 95% of my desktop and internet needs within the confines of the 1GB of Ram my Pi3B+ offers.
The only things I can't do within the constrictions of the RAM, is run large, bloated software, or run a browser with multiple tabs and high definition video (which I barely ever do). For this I'd need a swap file. But I'd rather do these things on a laptop or desktop, or trim my OS to fit the device better.
That 5% is why you should have a swap file.
What on earth make you think your laptop will handle that 5% any better? If it does it's because it has swap enabled (that's the default on most modern OS).
Don't believe me? Try disabling swap on your laptop... (and that experiment would only be even partially valid if both your laptop and Pi have the same amunt of RAM).
To get Raspbian work properly on 512MB of the Pi Zero, it gets tough, and one might need a swap for watching Youtube video, or having more than 3 tabs opened. But then again, the Pi Zero wasn't really made to run Raspbian Desktop os.
Maybe, I wouldn't know. All my Pi are running headless though I did have a desktop running on a 1B rev 1 back in the day (that's 256MB RAM for what it's worth).
This space unintentionally left blank.