Any total gurus on machine 'snappyness', extreme 'general desktop use' / browsing

[tjw]

hmmmm I got mine from Newegg and it did but good to know

[0xDE57]

@twin_savage

loose access to these valuable resources when websites go down.
The wayback machine doesn’t capture many of these sites

This is a growing issue. It’s link rot, but with content in general.

A software forum I used to use shut down and the community moved to discord. Its partially archived, but I was looking for a specific thread that wasn’t captured, and I can no longer find that specific resource. It would be handy to have a full local archive of that forum before it went offline.

Since the community migrated to discord, now those helpful threads are now stuck on discord instead of searchable.

Still looking for a good general solution to capture content for offline access.

[Log]

Link rot is definitely an important factor when trying to go back and reference things. I’ve been burned too many times.

I don’t have a great solution currently outside of one and done manual page captures with the SingleFile extension (always set it to open the page afterward to review if it worked), but I’m looking into the viability of setting up ArchiveBox and feeding it links from some kind of cross platform bookmarking tool. If I understand it correctly, this may also help make such pages accessible to others via archive.org as well. (A note about uploading to the internet archive, always use a purpose made alternative email for accounts there, as uploaded collections will contain the email of the uploading account in the metadata.)

For those that wish to investigate the topic further, https://github.com/iipc/awesome-web-archiving is a great place to start. The entire “awesome” list ecosystem is a great reference for finding tools and information on many topics.

[greatnull]

obsidian → works completely offline and note files are human readable
notion,so → easy export to other formats like html, markdown or pdf for archival.

Dont knock till you try it !

[nutral]

Obsidian is a bit tricky with licensing. Apart from personal use or as a single person company. You have to pay for it.

I’ve been really enjoying obsidian, I don’t go deep into the features, but it’s really helpful for mixed media and doing stuff like linking to a specific page in a pdf for example. Livesync is a great plugin that allows you to sync notes over a self hosted couchdb docker container. it works on different platforms.

[bedHedd]

We might want to move this discussion to

The Ultimate Home Server - Component: Knowledge Repository

Knowledge Repository – perhaps the most important aspect of my home server is that it is my brain outside my brain. I need a wiki to store stuff and keep track of things. I need something to store and keep track of my books BUT ALSO my notes in books. – love something that integrates with all my old written notes that have been digitized. Used L…

@greatnull

obsidian → works completely offline and note files are human readable

Not the issue I was referring to, the issue I had was that transferring my notes with the formatting preserved. The ideal is to be as close to plain text as possible and include a formatted preview that easy to share to others.

@greatnull

Dont knock till you try it !

when I tried both notion, obsidian, and one note a year ago, they didn’t work for my bullet workflow. I didn’t like how the render handled indentation when adding new sub bullet points.
*I like to use multi level indentation to

• organize my thoughts
  • add additional notes/comments
• create conversation threads
  • using sub points as replies
  • as well as branching points
• obsidian, notion, and one note
  • frequently broke if I needed to add new sub bullet points
  • did not know which root bullet points to link back to, leading to stray points

Furthermore I found it tedious to share or migrate the files between editors, none of the three translated very well to markdown, a email body, or as a message. All apps would prefer you to use their proprietary syncing solution and lock you into their system.
Obsidian is considerably more open with extensive plugin support, but falls short of vscodium or doom emacs. The EULA/licensing is still a deal breaker for me as getting my company to cover the yearly license is not worth my time to request it.

Some links if you want to dig deeper

How much does not having propriety / non standard Markdown syntax in your...

There are lots of Obsidian plugins that offer extra features or functionality but also introduces proprietary / non-standard markdown syntax. Its a double-edged sword since such plugins and syntax aren’t available in other markdown or text editors.

I was thinking about this recently. I spent countless hours in the past two months working on my vault, having 46 plugins and >1000 lines of my own CSS, at the time I write this. How many plugins do I have that introduce proprietary markdown?

https://ryan.himmelwright.net/post/leaving-notion/

While data security is very important, my biggest concern is that the database data is in a weird format. The page content is in markdown-ish, but the only way to view everything organized properly in the databases (without hassle, as far as I know), is within Notion. There are no raw files I can easily open or convert. They have an exporting system, but every time I tried to use it, it froze. So even while in theory you can export everything to md and csv files… I couldn’t. (Update on this below)

So while taking screenshots, I tried again to export the data. This time, it made it all the way through the process successfully. I was then able to take the output and run it through the Notion-to-Obsidian-Converter script (spoiler for what I’m using now ), and import it into an Obsidian vault. While this is nice to have as an archive, by default it isn’t useful in how it’s formatted and would be a lot of work to truly convert all the data. This is mostly due to how I’ve organized by Notion notes, but still… it is a mess and I don’t even know where to begin in understanding it.

I peeked at the onenote format standard [1] and the obsidian canvas standard. The difference is hilarious. The onenote standard is painfully complex, provided as a .pdf, and binary to boot. Compare to an example obsidian canvas - this is obvious, text-based (I could read it with notepad++) and easy to understand just by reading it:

  {
   "nodes":[
    {"id":"6c711bf8c24c4f5b","x":-226,"y":-62,"width":400,"height":400,"type":"file","file":"testin/2022-10-14.md"},
    {"id":"4dd7d04cdd0b379c","x":-530,"y":-209,"width":250,"height":60,"type":"text","text":"this is a note"}
   ],
   "edges":[
    {"id":"0c589a4d6bbb06aa","fromNode":"4dd7d04cdd0b379c","fromSide":"bottom","toNode":"6c711bf8c24c4f5b","toSide":"left"},
    {"id":"eda9f3edb3ec232a","fromNode":"6c711bf8c24c4f5b","fromSide":"top","toNode":"4dd7d04cdd0b379c","toSide":"right"},
    {"id":"abf404722ba48c3b","fromNode":"4dd7d04cdd0b379c","fromSide":"top","toNode":"6c711bf8c24c4f5b","toSide":"right"}
   ]
  }

If I could have it my way I would want this forum as my notes editor and twitter replacement

[Marandil]

@alkafrazin

It may well be that you just really need your pagefile and browser cache data on something really fast, like an Optane drive. Maybe even a dedicated one specifically for it.

Unrelated to windows and quite anecdotally, going from NAND flash for a swap file in Linux to a larger Optane swap file, I went from the entire system lagging out and sometimes hard crashing whenever Krita started running out of memory, to actually not even noticing when I’m 40GB into swap.

Why do people still use swap/page file with >32GB of RAM? I did not create a single Linux swap since ~2014 and try to remember to remove it from Windows almost every time. Ideally, when you have a workload that tries to allocate more RAM than you have you’d want it to have it reported/crash due to OOM rather than start dumping other processes’ memory to drives.

Browsers already have mechanisms to store the unused tabs on drives if they run out of memory. Splitting this responsibility between browser and operating system is only going to slow things down.

@disk_diddler

So let me give you a short example of how I’d research, here’s some google strings of late.

“7950x3d” “windows” “Desktop” “general use” “benchmarks”
(1 search, middle click at least 3 to 10 hits which look relevant, so they load in the background)

Next search.
“7950x3d” “performance” “Chrome” “tabs” “cache” “helps”
(Another search, middle click 5 to 10 relevant looking links)

Then I’d refine the searches , add words, subtract words etc then MORE middle clicks for more tabs…

I have a habit of doing all the “random” searches in Incognito/Private tabs to:
a) Not pollute search history
b) Limit exposure of search habits to “data hoarders” (e.g. advertisers)
c) Make them volatile and “go away” when I’m done with the topic.
If there is a result that’s actually worth keeping, it can get promoted to a regular, non-incognito tab, or even go into Keep.

@disk_diddler

I’m very, very good at googling!

If you actually were, you’d be fine with the first 3-5 results

In general, there are 2-3 things to optimize:

• (unlikely) usage habits. Even if you need to keep so many tabs open, maybe try segregating them into browser profiles, s.t. they don’t have to be open at the same time.
• RAM capacity s.t. more tabs can be kept in RAM instead of being moved to permanent storage
• Storage speed & latency. Perhaps a striped RAID could help here.

X3D is unlikely to help as more L3 cache is essentially more “low latency” RAM and RAM latency is unlikely an issue here, unless you can benchmark that it actually is.

[alkafrazin]

@Marandil

Why do people still use swap/page file with >32GB of RAM?

To prevent full system crashes.
Literally, I had problems where the system would lock up and Krita would crash, losing whatever it was working on, and potentially corrupting anything that was still saving(which takes an obscene amount of time for no good reason)
Adding swap made it occasionally recover for long enough to save and quit. Adding a large, fast swap made it pretty much unnoticeable to have krita it’s self overflowing into swap.

Before that, I had no swap, and just had to not work on things for a long time.

Also, browser will often cache to disk a lot even with plenty of ram, so having a fast drive for browser caching can still reduce stutter or browser lags.

[Marandil]

@alkafrazin

Also, browser will often cache to disk a lot even with plenty of ram, so having a fast drive for browser caching can still reduce stutter or browser lags.

That’s the thing, since the browser is caching to disk and the system may decide that something is not worth keeping it in RAM, you may be hitting that drive more than it’s needed. Fast drive may help regardless.

@alkafrazin

To prevent full system crashes.

Honestly, I can’t remember the last time I had a system crash due to OOM event. Application - sure, many times, but system? May be due to allocation patterns of my applications not triggering anything catastrophical in the system.

@alkafrazin

Literally, I had problems where the system would lock up and Krita would crash, losing whatever it was working on, and potentially corrupting anything that was still saving(which takes an obscene amount of time for no good reason)

That, honestly, sounds like a very bad coding. I have not seen krita source code, but if you’re getting OOM events on save, sounds like someone’s not utilizing mmaps nearly enough

[alkafrazin]

memory crashing happens on big operations, like changing color space, multiple layer transform, combining layers, etc. Saving just takes a minute or two, and waiting for it to finish is a waste of time, so if the program crashes during that time after a save, it can be bad.

Really, Krita probably isn’t well programmed at all, but since nobody is making a better alternative, what can you do besides download more ram?

[Tanmay]

I was scrolling for so long hoping I don’t recommend it again. and i was not disappointed.

[disk_diddler]

Apologies to all for the late replies, been busy.

This post in particular is worrying. I don’t like “odd weirdness” where machines just lock and lag for no reason.

Generally I see that with say, mapped network drives, faulty / plugged in USB drives, stuff like that.

I’d like to think your situation is a bit of a one off but certainly curious to hear if others have 7800x3d or 7950x3d systems and find them utterly pump in Windows or now and then they kinda do odd stuff.

Again, no games ever for me, not on PC anymore anyhow.

[disk_diddler]

The 5650GE systems I’ve got have had some issues with the latency, I’d be curious if you (and anyone here) could give me very particular tips on really getting a deep dive on the performance monitoring.

I’ve always been ok at it, but expect people here to be smarter than I.

What / how should I filter with process explorer? I’m quite fluent with MMC and adding a variety of metrics, but unsure which ones are the best ones to capture ‘sluggishness’

[nutral]

The best way to capture sluggishness is to use latencymon. Let it run in the background for a bit and throw a pastebin link here of it.

Because of this post I made my pc snappier, by installing a new chipset driver.

[disk_diddler]

Can I get an understanding of optane products? I’m searching for used stuff and see items such as:

"Intel Optane Memory H10 256GB+16GB "

Is that in actual fact, only 16GB of optane and 256GB of normal NVMe SSD behind it?

[greatnull]

Oh yes, you are reading that correctly. Peak intel moment, or how not to market breakthrough technology into ground.

That is how intel envisioned consumer optane for masses → pair cheap but otherwise extremely poorly performing QLC NAND flash with small optane cache device.

Well there was even worse option B , the same standalone optane device as cache for standard hdd.

Both options also required intel software raid on target platform (Intel® RST) to implement transparent caching, otherwise you just ended up with two physiscal drives per module.

No wonder it didn’t succeed . If my memory serves me well, those 16/32 GB optane modules had same asking price as good mainstream 1TB TLC nvme ssd today, so that was 16GB optane for price 128/256GB ssd then?

In short avoidit , only worth optane are 900p and 905p consumer devices or DC variants.

[tjw]

@disk_diddler

Is that in actual fact, only 16GB of optane and 256GB of normal NVMe SSD behind it?

correct

i have used the following things:

P1600X:Intel® Optane™ SSD P1600X Series

905P:Intel® Optane™ SSD 905P Series

[alkafrazin]

These only work correctly on certain intel systems anyway. They’re bad and should be avoided.

[disk_diddler]

I certainly appreciate a post like this with someone trying to help - this is up my alley.

So as I’ve said I’m a Chrome guy, because of the plugins which it supports happen to work flawlessly with my desired flow and functions as I did start advanced browsing with NetCaptor.

NetCaptor

I have some of the following features.

• Open links in background and directly RIGHT of the current tab. (“Tab Position Options”)
• Close tabs, moves left of current tab.
• Find and close duplicate tabs (“Duplicate tabs closer”)
• TabManager.io (extremely good) - can search ALL open tabs and CLOSE all tabs matching search criteria across all windows [feature I requested personally] - this is a fork of Cluster tab manager, I think
• Lastpass (yes, I’ll ditch it eventually)
• Session Buddy (more save sessions just in case tools) also lets me search open tabs
• Tab Menu (another easy switch to tabs tool on the toolbar)
• Dstill web monitor (price alerts for example)
• Downloads router (very handy for saving files matching *.exe to \downloads\exe or *.jpg to \downloads\images etc)

By using 2x 4K displays and a VERY FAST browser, when I’m ‘in a flow’ I can really close a lot of tabs quickly using TabManager.IO - but it’s the responsive browser I need to do it.

I’ll def consider the discard (though doesn’t Chrome do this now?)

[tjw]

@disk_diddler

TabManager.io (

holy shit i had no idea this existed