Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

24-core CPU and I can’t move my mouse

randomascii.wordpress.com

Post reply on HN
Print view View original

481–490 of 509 posts

Re: 24-core CPU and I can’t move my mouse

#481

Post by TeMPOraL »

mark-r wrote: post #435
TeMPOraL wrote: post #73

Earlier quoted context omitted.

I agree. However at this point, I don't see anything an OS can do to help. I see plenty of typing slowdowns every other day now. But I'm not sure just how many of them are the OS's fault. When your typing seems to lag, there are two places that can be slowing it down - the input side (reacting to hardware events) and the output side (drawing and updating the UI). I suppose keyboard buffers are pretty well isolated, a…

It's hard to believe the OS couldn't help when Windows 10 has the problem but Windows 7 doesn't.

In this particular case, yes. But this subthread was about a more general principle of letting an app exhaust system's performance. What I'm saying is that, when facing a crappily coded app, the OS can at best choose between letting it suck or letting the performance of everything suck.

Re: 24-core CPU and I can’t move my mouse

#482

Post by beagle3 »

StillBored wrote: post #477
ShinTakuya wrote: post #60

Earlier quoted context omitted.

I mean the trackpad issue is generally down to the manufacturer and the drivers they provide, but I take your point otherwise. A pro tip with 7z is to use the two pane interface to extract and not dragging files to Explorer. The latter option extracts the files to a temp directory before copying them whereas the former extracts directly to the destination. The Windows file systems are pretty cruddy in general though…

The Windows file systems are pretty cruddy in general though so decompressing tonnes of small files in general will take longer than on OS X or Linux. I haven't checked this in a while, so it could be out of date... But what you claim is windows being cruddy, might be a symtom of a slightly different mindset in data protection. Here is a test, grab a big tar.gz/zip file and uncompress it in linux and in windows from…

No, NTFS is cruddy. It is eager to flush metadata but not data, and there is no way to switch it to "journal data as well" mode (at least not as far as Win8.1, maybe 10 added this mode).

This is the default behavior on ext3 and ext4 as well, of course, and if you pull the plug on NTFS, ext3 or ext4, you WILL end up with files whose length reflects recent writes but whose contents do not (and I am saying this from analyzing hundreds of cases of file contents that are impossible given program flow -- about 10:1 on NTFS than on non-data-journalled ext4). But at least on linux you can tune2fs -o journal_data, still get very decent performance and much, much better integrity.

Also, I think you are conflating sync(), fsync(), fdatasync() and FlushFileBuffers. calling fsync() or fdatasync() on linux is no worse than FlushFileBuffers - the IO killing behaviour you describe is associated with sync(), which, when you need it, is perfectly justified in killing IO.

Re: 24-core CPU and I can’t move my mouse

#483

Post by gurkendoktor »

dispat0r wrote: post #77
gurkendoktor wrote: post #46

Earlier quoted context omitted.

Staying responsive (under load) is also still a work in progress on Linux: https://www.phoronix.com/scan.php?page=news_item&px=BFQ-Queu...

That is a disk scheduler and mostly for spinning disks. I haven't seen a stuck mouse cursor for a long time in linux.

My only disk is an M2 SSD, and when I run an OpenStreetMap import in the background, my Ubuntu desktop becomes unusable.

I don't remember if the mouse cursor is affected, though, and I don't know if the new IO scheduler really fixes this.

Re: 24-core CPU and I can’t move my mouse

#484

Post by vidarh »

cheez wrote: post #174
lucian1900 wrote: post #93

Earlier quoted context omitted.

Unfortunately that's not feasible for highly performance-sensitive projects like browsers or games.

This would be the end of the discussion if it weren't for stupidity like this: https://groups.google.com/a/chromium.org/forum/#!msg/chromiu... Generally, I find when people crow about performance, the product they're talking about usually has some questionable architectural/design/implentation decisions that dominate the performance issues so I have to do my best not to roll my eyes. Yes, you can write performant C++…

I once cut 30% of page generation times for a commercial CMS in half a day by just skimming through their output generation code and changing std::string method invocations to get rid of unnecessary temporaries.

People very rarely has any clue about this at all.

Re: 24-core CPU and I can’t move my mouse

#485

Post by ash »

andrepd wrote: post #294
ash wrote: post #213

Earlier quoted context omitted.

Why not fast browser and fast compilation at the same time?

Show me how, then.

I didn't say I know how to do it. And I didn't say it's easy. Yes, currently we have fast browser and slow compilation. Just let's not assume situation can't be improved at all.

Re: 24-core CPU and I can’t move my mouse

#486

Post by stinos »

StillBored wrote: post #474
zlynx wrote: post #371

Earlier quoted context omitted.

That used to be true back when the CPU did all of the GUI rendering. But now most all of it is offloaded to the GPU. Any GPU that can render Quake 3 Arena at 120 FPS (and that's ALL of them even Intel IGPs) can wiggle a window around very easily. Not sure about file opens. Simple applications like GVIM don't seem to have seconds of delay for me, but I know what you mean with things like spreadsheet or word processor…

Actually, its not as accelerated as it was back in the winXP days. Here is a random link about it: https://www.youtube.com/watch?v=ay-gqx18UTM . Basically, the GPU card vendors could hook any part of the win32 GDI pre windows vista, and they did. Post vista, only a tiny portion of the GDI is accelerated. In theory you can avoid this by writing your application using a more modern API, but the vast majority of native…

tweaks I've been collecting/researching for the past decade

Do you happen to have this available to the public somewhere?

Re: 24-core CPU and I can’t move my mouse

#487

Post by josteink »

titzer wrote: post #45

Full disclosure: I work for Google on Chrome. A Chrome build is truly a computational load to be reckoned with. Without the distributed build, a from-scratch build of Chrome will take at least 30 minutes on a Macbook Pro--maybe an hour(!). TBH I don't remember toughing out a full build without resorting to goma. Even on a hefty workstation, a full build is a go-for-lunch kind of interruption. It will absolutely own a…

> A Chrome build is truly a computational load to be reckoned with. Without the distributed build, a from-scratch build of Chrome will take at least 30 minutes on a Macbook Pro--maybe an hour

From someone regularly working on multi hour builds (on build servers), this imo sounds like the light kind of build I'd wish I worked with ;)

That said, this article was about a weird behaviour in Windows when destroying processes, not about heavy builds.

Re: 24-core CPU and I can’t move my mouse

#488

Post by thriftwy »

SolarNet wrote: post #478
StillBored wrote: post #475

Earlier quoted context omitted.

Windows defender is by far the "best" malware software when it comes to having a light footprint. Scanning every byte of IO for malware signatures, and the ton of other heavyweight operations are a sure way to eat a ton of processing power. I have a under powered win10 tablet, and I have to remind myself to "disable" windows defender every time it goes to install an update. Otherwise what should be a 2-3 minute opera…

Yea I don't have a problem with it per se, I just wish it implemented any sort of improvement to reduce the hard-drive usage. Say, integrating with the OS to read the same memory as the program will (rather than reading it 2+ times: this kills the drive). Or perhaps storing a list - with hashes - of already scanned files, so it doesn't scan the entire 30+ GB windows dev kit every time I open a large project in visual…

"integrating with the OS to read the same memory as the program will (rather than reading it 2+ times: this kills the drive)."

Is it even possible to read a file twice while ignoring system buffers cache?

Re: 24-core CPU and I can’t move my mouse

#489

Post by DiThi »

jhasse wrote: post #203
DiThi wrote: post #200

Earlier quoted context omitted.

A lot of things did happen. One of them is having the option to check the source which is much faster than compiling. 99% of the time you're compiling code in rust is to run the borrow checker and fix those errors.

That's indeed something where Rust is a lot better than C++. But for many projects (like games or GUI apps) you'll need to iterate with changes which can only be seen in the final product. So you'll need a complete build to run the binary.

Fair enough. However games and GUI apps are still a small percentage of applications developed in rust. I want that to change, of course. I just wanted to point that out.

I'm personally writing a game in Rust but the main logic is written in a compile-to-JS language and uses V8, so the issue doesn't affect me.

Re: 24-core CPU and I can’t move my mouse

#490

Post by DiThi »

ori_b wrote: post #462
DiThi wrote: post #197

Earlier quoted context omitted.

You can have all the files you're working on not concatenated.

That's painfully manual.

It doesn't need to be. I think it shouldn't be hard to make a build script that re-makes the bundles except for files modified in the last hour.
Post reply on HN