VOGONS

Common searches


New forum suggestions/Feedback

Topic actions

Reply 220 of 516, by 7F20

User metadata
Rank Member
Rank
Member
Snover wrote on 2020-01-10, 17:22:

Could you please take a photo of this? It’s interesting to me how this is monitor is another one which comes with “flicker-less technology” and yet somehow is a big offender.

I went into the user menu of my monitor and reset it. It seems to have reduced the flickering to unnoticeable levels, but the color is still washed out and weird. Other monitors it looks purple but this one it's basically just grey. Perhaps it's just this LCD being weird. Sorry I didn't get a snap before the reset, but if it was just a malfunction, then no big deal.

Spoiler

7GXgM74.jpg

Last edited by 7F20 on 2020-01-10, 20:38. Edited 2 times in total.

Reply 221 of 516, by Ant_222

User metadata
Rank Oldbie
Rank
Oldbie
Snover wrote on 2020-01-10, 17:22:
Ant_222 wrote on 2020-01-08, 18:00:

I agree about the wasted vertical space. Possible solutions are: remove the greeting line and the two empty lines around it; place the greeting line and banner on the same line:

I like this, except that the width of the display is unknowable, so it’s not possible to reliably position anything to the end like that. I’ll try moving the greeting up and see how weird it feels.

It is safe to assume a line width of a least 72 characters, which is standard in plain-text files. Another option is not to right-aling, but write:

┌───────┐ ────┐
│ V O G O N S │ greets Ant_222!
└─▸ └─────────┘
Snover wrote on 2020-01-10, 17:22:

Or just get rid of it entirely—who needs to be friendly when you’ve got such a cool logo art, you know? ;-)

Sure, do not remove it in any case!

Snover wrote on 2020-01-10, 17:22:
Ant_222 wrote on 2020-01-09, 13:37:

What I dislike about reply notifications is that they force me to fire up the browser and actually log in to vogons.org in order merely to read the new post. It would be great if the full text were included in the notifications and they were not disabled until the next login. That way, we should be able to track discussions in our e-mail clients and log in only when we have something say.

I can understand this desire. Unfortunately the way the notification system works is really intended to deliver only one email per topic and is not really amenable to change. Also, the next thing people would ask for after this is the ability to reply to posts from the email, and now we’ve created a listserv and all the new headaches that come with managing something like that securely, so I am not sure if this is even a good road to go down :-)

This continuous notification system has been used in several forums in which I participated 10-15 years ago, including englishforums.com. Once I set up filters in my e-mail client to put the notifications into separate folders, the whole experience became quite pleasant. Staring up a browser and logging in to read every new post, or only to write something are very different levels comfort. Consider how much motions it saves if I answer every tenth post I read.

Update: I forgot to say that the notifications were properly threaded, so my client displayed them as one thread per forum topic—very convenient.

When I asked the englishforums.com admin why they had changed to the frugal notifications, he said it was to blow up online activity including the clicking on banners. Surely it cannot be the motivation here.

I do understand that posting via e-mail and NNTP is an entirely different business.

Reply 222 of 516, by Snover

User metadata
Rank l33t++
Rank
l33t++
Shagittarius wrote on 2020-01-08, 15:05:

[…]flickering[…]

7F20 wrote on 2020-01-10, 16:22:

[…]flickering[…]

I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedback 😀 Instead, at the moment please start by answering these things:

Using the default (“Reset normal”) settings:

1. On which patterns is there scroll flickering?
2. On pattern 9 and default “Back” colour, what value of the “Fore” slider causes the flickering to stop?
3. On pattern 9 and default “Fore” value, what lower value of the “Back” brightness causes the flickering to stop?

Thanks!

Yes, it’s my fault.

Reply 223 of 516, by 7F20

User metadata
Rank Member
Rank
Member
Snover wrote on 2020-01-10, 19:25:
I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedbac […]
Show full quote
Shagittarius wrote on 2020-01-08, 15:05:

[…]flickering[…]

7F20 wrote on 2020-01-10, 16:22:

[…]flickering[…]

I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedback 😀 Instead, at the moment please start by answering these things:

Using the default (“Reset normal”) settings:

1. On which patterns is there scroll flickering?
2. On pattern 9 and default “Back” colour, what value of the “Fore” slider causes the flickering to stop?
3. On pattern 9 and default “Fore” value, what lower value of the “Back” brightness causes the flickering to stop?

Thanks!

1) Almost all patterns have a very small amount of flicker at default. 12&26 have pretty much zero flicker. 5, 10, 11, 18, 23, 27 are almost imperceptible. 8 is pretty bad, and 4 lights up my monitor like a christmas tree.

2) about 40 on the Fore slider makes it pretty hard to see any flicker to me

3) No setting going darker on the Back value helps. In fact, it seems to get worse as I go darker.

Reply 224 of 516, by bjwil1991

User metadata
Rank l33t
Rank
l33t

Has the monitor done something like that before?

Discord: https://discord.gg/U5dJw7x
Systems from the Compaq Portable 1 to Ryzen 9 5950X
Twitch: https://twitch.tv/retropcuser

Reply 225 of 516, by konc

User metadata
Rank l33t
Rank
l33t
Snover wrote on 2020-01-10, 19:25:

1. On which patterns is there scroll flickering?
2. On pattern 9 and default “Back” colour, what value of the “Fore” slider causes the flickering to stop?
3. On pattern 9 and default “Fore” value, what lower value of the “Back” brightness causes the flickering to stop?

1. All, except 11 and 12. Much less noticeable on 10 but it's there
2. Around 150 it's stops being really annoying and around 100 I cannot see it anymore
3. I'm not sure what I need to change to adjust the brightness from the color panel that pops up. Is it the "Lum" field?

Edit: forgot to write, the above apply for a Dell U2415

Last edited by konc on 2020-01-10, 20:12. Edited 1 time in total.

Reply 226 of 516, by 7F20

User metadata
Rank Member
Rank
Member
bjwil1991 wrote on 2020-01-10, 19:46:

Has the monitor done something like that before?

I never noticed the behavior before. I use this monitor at work pretty often. After reading the last comment, I went to a variety of pages to check and I didn't notice any flickering. I did an image search for color gradients, and they seem to be okay when scrolling as well. Again, this only happens on the one monitor, I think it's led back-lit. The other one I have, and the one at home seems totally fine with this page. (I can't try the special flicker test at home yet though)

Edit: I think I just realized what you are saying. The answer is "no, the monitor has never randomly freaked out or required a reset to make a specific page stop flickering." I was just assuming that this is some image quality processing that I set up making it happen. Unfortunately, when I reset it, I lost the specific color, contrast, brightness and processing combo that was making the Old Skool forum colors go wacky for me. Now, when not scrolling, there isn't a lot of flicker. However, the colors are still super washed-out and there is some scroll flicker, and now I realized I'm getting major trails on the text.

I'll see if I can upload a video.

Video, doesn't have high enough frame rate to capture the trails well so I took a burst shot and picked one in motion.

Spoiler

xztT1EO.jpg

Another Edit: I have found a processing setting (Overdrive) which is enabled by default in the menu. Turning this off has made a big improvement and it has made everything less flickery. It's supposed to reduce ghosting, but it seems to be doing the opposite. (on specific pages like this one. also seems to completely pooch ken rockwells photography page)

Last edited by 7F20 on 2020-01-10, 20:38. Edited 2 times in total.

Reply 227 of 516, by Snover

User metadata
Rank l33t++
Rank
l33t++
Ant_222 wrote on 2020-01-10, 18:42:

It is safe to assume a line width of a least 72 characters, which is standard in plain-text files.

Small phones in portrait mode (e.g. iPhone SE) show a maximum of 33 characters in monospace at a standard size, so no, I am afraid it is not safe to assume a 72 character minimum these days 😀

Ant_222 wrote on 2020-01-10, 18:42:

Another option is not to right-aling, but write:

Yes, it would be something like this.

Ant_222 wrote on 2020-01-09, 13:37:

This continuous notification system has been used in several forums in which I participated 10-15 years ago, including englishforums.com. […] When I asked the englishforums.com admin why they had changed to the frugal notifications, he said it was to blow up online activity including the clicking on banners. Surely it cannot be the motivation here.

There is definitely no motivation to drive traffic so people click on the ads which don’t exist 😀

What I meant is that phpBB’s notification system isn’t designed to work the way you are asking for it to work, and I don’t have the resources right now to change it so that it will do what you are asking.

It tries (and, judging by the related bug reports, fails quite often) to send only one notification per new post/topic.

It does not expose post content to the email templates at all.

It can’t send HTML mail, and the markup model is BBCode instead of something smarter which works naturally in plain text like Markdown, so any post formatting (like “Quote by <user> on <date>”) would be lost. Also, while the parser/formatter is nominally object-oriented, in practice most of this stuff ends up interacting with global objects in unpredictable ways. Also also, even though the formatter is based on XSLT and so in theory should have a raw text output mode, I’ve tried and failed getting it to not require everything that enters and exits to be well-formed XML since it is actually a semi-bespoke XSL engine.

Just like the “why can’t you just expose extra themes” question, the “why can’t you just send emails with more content in them” is a totally rational question and the answer is that phpBB isn’t designed to enable things like this so most of these enhancements end up being fairly hacky (replacing whole components) which then makes it harder to keep up-to-date with upstream versions which are constantly touching things in non-backwards-compatible ways.

Yes, it’s my fault.

Reply 228 of 516, by j^aws

User metadata
Rank Oldbie
Rank
Oldbie

Some positive feedback: Besides Your Posts, New Posts and Active Posts, using the 'bell' icon, you see Your Last Read Post inside your active thread? I don't think I've used that before.

Using the links within takes you to your last post read in those threads. See below for mobile mode in landscape and portrait:

Attachments

  • 20200110_204751.jpg
    Filename
    20200110_204751.jpg
    File size
    1.95 MiB
    Views
    1500 views
    File license
    Fair use/fair dealing exception
  • 20200110_204903.jpg
    Filename
    20200110_204903.jpg
    File size
    1.47 MiB
    Views
    1500 views
    File license
    Fair use/fair dealing exception

Reply 229 of 516, by VileR

User metadata
Rank l33t
Rank
l33t
Snover wrote on 2020-01-10, 19:25:

I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedback 😀 Instead, at the moment please start by answering these things:

Using the default (“Reset normal”) settings:

1. On which patterns is there scroll flickering?

All of them, but it generally seems to be worse for patterns with higher-frequency detail. With 9 it's quite noticeable, with 10 much less so, with 11 & 12 it's pretty much imperceptible. Below 9, they're all quite bad, especially 3 through 7. With the others- the denser the details, the worse it is, especially with 15 through 17.

However, to qualify my observation.. even with the same detail density, it also seems to depend on which pixels are lit, and which ones aren't. For example, patterns 13 and 18 are the inverse of each other - in 13 the larger area is lit, and in 18 it's the smaller; the flickering is much worse in 18. Same goes for 23 & 26, respectively.

I guess some minutiae of LCD technology might explain that, say if pixels take longer to turn on than to turn off... but I dunno.

2. On pattern 9 and default “Back” colour, what value of the “Fore” slider causes the flickering to stop?

It gets pretty much imperceptible at 100 and under.

3. On pattern 9 and default “Fore” value, what lower value of the “Back” brightness causes the flickering to stop?

I'd say 30 (this one seems to use the OS's color-picker, which for me gives a "Lum" slider of 0-255).
Monitor is an HP Z24n, FWIW.

[ WEB ] - [ BLOG ] - [ TUBE ] - [ CODE ]

Reply 230 of 516, by Shagittarius

User metadata
Rank Oldbie
Rank
Oldbie
Snover wrote on 2020-01-10, 19:25:
I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedbac […]
Show full quote
Shagittarius wrote on 2020-01-08, 15:05:

[…]flickering[…]

7F20 wrote on 2020-01-10, 16:22:

[…]flickering[…]

I set up this demonstrator to test different patterns and colours. It’s a big playground so try not to go crazy with the feedback 😀 Instead, at the moment please start by answering these things:

Using the default (“Reset normal”) settings:

1. On which patterns is there scroll flickering?
2. On pattern 9 and default “Back” colour, what value of the “Fore” slider causes the flickering to stop?
3. On pattern 9 and default “Fore” value, what lower value of the “Back” brightness causes the flickering to stop?

Thanks!

I should take this time to mention that I'm running my display in a full 10bit config card > screen @ 98hz. So keep that in mind. I also do not see the reduce flicker option that you mentioned.

1. All patterns have some degree of flickering 9 being the least flickery
2. At 25 I can no longer see flicker.
3. IT looks like the flicker stops somewhere around #1f1b41

Reply 231 of 516, by Snover

User metadata
Rank l33t++
Rank
l33t++

Thanks everyone for taking the time to run some tests for me!

Shagittarius wrote on 2020-01-11, 01:47:

I should take this time to mention that I'm running my display in a full 10bit config card > screen @ 98hz. So keep that in mind. I also do not see the reduce flicker option that you mentioned.

Oh, OK. I guess it is not switchable—I suppose I should have RTFM first. The good news is that this is all in your head since TUV certified the monitor to be flicker free! 😉

Shagittarius wrote on 2020-01-11, 01:47:

1. All patterns have some degree of flickering 9 being the least flickery

Wow, patterns 10–12 are more flickery than pattern 9? That is a surprising and counter-intuitive finding. Of the range 0–12 I would expect things to be worst at pattern 5 and improving as you move away from there.

Shagittarius wrote on 2020-01-11, 01:47:

2. At 25 I can no longer see flicker.
3. IT looks like the flicker stops somewhere around #1f1b41

Can you even see the pattern at those levels? Those are remarkably low contrast levels (1.1–1.29); at that level here it just looks like my monitor has turned into a dithering TN panel since I can’t even tell there’s a pattern unless I stick my nose up to the screen.

Are you running the monitor in sRGB mode under GameVisual function? Are Dark Boost and Auto Black Level turned off? Is Display SDR Input set to sRGB? Do you see similar results with the other 1080p panel you have?

VileR wrote on 2020-01-10, 22:56:

All of them, but it generally seems to be worse for patterns with higher-frequency detail. With 9 it's quite noticeable, with 10 much less so, with 11 & 12 it's pretty much imperceptible. Below 9, they're all quite bad, especially 3 through 7. With the others- the denser the details, the worse it is, especially with 15 through 17.

However, to qualify my observation.. even with the same detail density, it also seems to depend on which pixels are lit, and which ones aren't. For example, patterns 13 and 18 are the inverse of each other - in 13 the larger area is lit, and in 18 it's the smaller; the flickering is much worse in 18. Same goes for 23 & 26, respectively.

I see something similar here where 0–1 are worse than 9–10 even though they are just inversions of the same pattern. I can’t say I have any sort of working mental model of what’s going on yet.

I also can’t figure out why the spec sheet for your monitor says it’s an IPS panel but also that it uses FRC to render 8-bit colour. If true, maybe it explains why there is a bigger problem with this monitor, though I guess I also need to learn if any monitors don’t have this problem and are also using FRC or some other kind of dithering.

Yes, it’s my fault.

Reply 232 of 516, by Shagittarius

User metadata
Rank Oldbie
Rank
Oldbie

"Are you running the monitor in sRGB mode under GameVisual function? Are Dark Boost and Auto Black Level turned off? Is Display SDR Input set to sRGB? Do you see similar results with the other 1080p panel you have?"

Yes, yes, yes.

The test screen looks a little different than last time, this time around it looks like 12 is the only setting that is remotely flicker free. I'm not sure why it loaded so differently the first time. And yes I can see the dithering pattern at the levels I specified above.

I also want to mention compared to the test screen the flickering on Vogons is much less pronounced.

On the cheap monitor same results except the flicker disapears at 50 fore and no matter what i set the back to it still flickers.

I might be overly sensitive to flicker...

Reply 233 of 516, by Dominus

User metadata
Rank DOSBox Moderator
Rank
DOSBox Moderator

I see no flickering 😀 And that doesn't help 😀

Windows 3.1x guide for DOSBox
60 seconds guide to DOSBox
DOSBox SVN snapshot for macOS (10.4-11.x ppc/intel 32/64bit) notarized for gatekeeper

Reply 235 of 516, by VileR

User metadata
Rank l33t
Rank
l33t
Snover wrote on 2020-01-11, 03:22:

I also can’t figure out why the spec sheet for your monitor says it’s an IPS panel but also that it uses FRC to render 8-bit colour. If true, maybe it explains why there is a bigger problem with this monitor, though I guess I also need to learn if any monitors don’t have this problem and are also using FRC or some other kind of dithering.

You got me there... I got burned buying this one - I went for it because I wanted a decent 1920x1200 IPS panel, but ended up with a very TN-like FRC behavior. Sometimes I get flickering even with completely *static* patterns, which seems to come and go depending on the location of the window on the screen. (BTW, which datasheet are you looking at? The ones from the first few hits off the HP website don't mention this -- if they did I would've had advance warning...)

10 years ago I was expecting computer monitors to start ditching LCD tech and move on to something that isn't full of corner-cutting compromises. Still nothing in consumer-land. *sigh*

[ WEB ] - [ BLOG ] - [ TUBE ] - [ CODE ]

Reply 236 of 516, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

I tried messing with the test page and I didn't really see flickering at all, but some of the settings / patterns looked like they were really messing with my eyes.

I wonder if part of what some people are seeing is their brain trying to process the patterns on the screen / optical illusion type stuff.

Yamaha modified setupds and drivers
Yamaha XG repository
YMF7x4 Guide
Aopen AW744L II SB-LINK

Reply 237 of 516, by Dominus

User metadata
Rank DOSBox Moderator
Rank
DOSBox Moderator

Yes, I get that "eye messing" as well. I noticed this in the last years when looking at striped things like close black and white stripes. I'm having a hard time folding shirts that have these. Incidentally that's a reason I gave Snover for some of the patters in early test versions of the new site.
And I got these eye things, too, with the test patterns.

Windows 3.1x guide for DOSBox
60 seconds guide to DOSBox
DOSBox SVN snapshot for macOS (10.4-11.x ppc/intel 32/64bit) notarized for gatekeeper

Reply 238 of 516, by 7F20

User metadata
Rank Member
Rank
Member
cyclone3d wrote on 2020-01-11, 18:03:

I wonder if part of what some people are seeing is their brain trying to process the patterns on the screen / optical illusion type stuff.

If you do the scrolling test, you won't be able to focus on the pattern. The screen will be moving too fast to focus on any "illusions". It definitely is flickering the screen.

Reply 239 of 516, by Snover

User metadata
Rank l33t++
Rank
l33t++
VileR wrote on 2020-01-11, 17:46:
Snover wrote on 2020-01-11, 03:22:

I also can’t figure out why the spec sheet for your monitor says it’s an IPS panel but also that it uses FRC to render 8-bit colour. If true, maybe it explains why there is a bigger problem with this monitor, though I guess I also need to learn if any monitors don’t have this problem and are also using FRC or some other kind of dithering.

You got me there... I got burned buying this one - I went for it because I wanted a decent 1920x1200 IPS panel, but ended up with a very TN-like FRC behavior. Sometimes I get flickering even with completely *static* patterns, which seems to come and go depending on the location of the window on the screen. (BTW, which datasheet are you looking at? The ones from the first few hits off the HP website don't mention this -- if they did I would've had advance warning...)

The spec sheet I found describes FRC in the “Product Features” section. It’s a feature, not a bug! 🤔

Static pattern flicker could be this problem. On this Lagom inversion test, only patterns 2b and 3 to cause two ~5px-wide columns to crawl slightly on my monitor. I don’t know if this inversion flickering is related or not to the scroll flickering. On a high-DPI monitor none of the Lagom patterns should trigger anything since the device pixels are smaller than the test patterns (the page is old and doesn’t know it needs to scale background images—CSS pixels aren’t equal to device pixels on high-DPI monitors).


I’ve created a new test with a possible alternate. Folks who were having trouble, please let me know if this new test works without flickering at the initial settings. If not, tell me at what level the Fore slider needs to be set for it to not flicker, and also if any patterns 6–8 are better. Thanks again in advance!

Yes, it’s my fault.