← Home ← Back to /v/

Thread 713575438

42 posts 6 images /v/
Anonymous No.713575438 >>713575769 >>713576007 >>713576107 >>713577469
>makes gaming fun again
Anonymous No.713575683
>emulation
I'll wait for native ports, thank you.
Anonymous No.713575769 >>713576165
>>713575438 (OP)
Buy an ad, Mister.
Anonymous No.713575794 >>713575962 >>713576171 >>713576367 >>713576589 >>713577164 >>713577380
FPGAs are less accurate than software emulators:
>https://github.com/MiSTer-devel/Atari800_MiSTer/issues
>https://github.com/MiSTer-devel/AtariST_MiSTer/issues
>https://github.com/MiSTer-devel/C64_MiSTer/issues
>https://github.com/MiSTer-devel/CoCo3_MiSTer/issues
>https://github.com/MiSTer-devel/ColecoVision_MiSTer/issues
>https://github.com/MiSTer-devel/GBA_MiSTer/issues
>https://github.com/MiSTer-devel/MegaCD_MiSTer/issues
>https://github.com/MiSTer-devel/MegaDrive_MiSTer/issues
>https://github.com/MiSTer-devel/MSX1_MiSTer/issues
>https://github.com/MiSTer-devel/MSX_MiSTer/issues
>https://github.com/MiSTer-devel/NeoGeo_MiSTer/issues
>https://github.com/MiSTer-devel/NES_MiSTer/issues
>https://github.com/MiSTer-devel/PSX_MiSTer/issues
>https://github.com/MiSTer-devel/S32X_MiSTer/issues
>https://github.com/MiSTer-devel/Saturn_MiSTer/issues
>https://github.com/MiSTer-devel/SMS_MiSTer/issues
>https://github.com/MiSTer-devel/SNES_MiSTer/issues
>https://github.com/MiSTer-devel/WonderSwan_MiSTer/issues
>https://github.com/MiSTer-devel/X68000_MiSTer/issues

FPGAs have more input lag than emulators. FPGAs can only do as low as input lag as real hardware, emulators can as well, but emulators can go even lower thanks to run-ahead:
https://www.youtube.com/watch?v=_qys9sdzJKI

FPGAs have less features and enhancements than emulators.
Anonymous No.713575962 >>713576486 >>713576559
>>713575794
How does run-ahead even work technically? Sounds like black magic shit to me.
Anonymous No.713576007 >>713576143
>>713575438 (OP)
>Open source device with dedicated debug port (unnecessary since you can already run your own code on it and there is a USB port)
>two power only USB A
I don't understand the I/O choices.
Anonymous No.713576107 >>713576201
>>713575438 (OP)
I don't know what that little box is but it doesn't look like it can play my old nintendo tapes
Anonymous No.713576143 >>713576667
>>713576007
>I don't need to debug
>because it runs "my own code"
what the fuck is this nonsense that I'm reading.
I'm 20 IQ points dumber now.
Anonymous No.713576165
>>713575769
why buy an ad when you can buy jeets to shill everywhere just for few cents per post
we just buy traffic
Anonymous No.713576171 >>713576275 >>713576330 >>713576580
>>713575794
Software emulators force me to choose between screen tearing or input lag, that's the main selling point of FPGA
Anonymous No.713576201
>>713576107
You can with addons.
Anonymous No.713576275 >>713576549
>>713576171
You obviously choose screen tearing if you're not a casual normalfaggot
Anonymous No.713576330 >>713576549
>>713576171
>Software emulators force me to choose between screen tearing or input lag
Works on my machine.
Anonymous No.713576367 >>713576960
>>713575794
99% of those are feature requests, using unsafe features that may cause issues with certain games, or are using bad dumps.

MiSTer is as accurate or more than every software emulator with identical input lag to original hardware.

>muh runahead
It's never recommended to use run-ahead as it increases input lag variance for slightly lower raw input lag. It's also not accurate to original hardware, has limited implementation depending on internal game lag and how intensive it is to run, and can cause various issues

The only rare edge case in which it may be worth it (and even then, not really), is games that are easy to run accurately, have significant inherent gameplay lag (5+ frames), and where reactions are more important than timing. In these case, the raw input lag improvement is so large that it could (arguably, but not really) outweigh the increase in input lag variance that it brings.

Also run-ahead is not going to remove any inherent lag from your setup or V-Sync lag, which will usually be 2-3 frames at least. So no, you are almost certainly not getting original hardware/MiSTer CRT level input lag, and you certainly aren't getting that level of input lag consistency either.
Anonymous No.713576486
>>713575962
Y'know how you can fast forward an emulator? Well do that in the background every frame so you can load run the game ahead of time, savestate, and then instead of displaying the next frame of the game when it comes time you load that savestate which is 1 frame ahead of that.
Anonymous No.713576549
>>713576330
No it doesn't.

>>713576275
Yeah that's the lesser of two evils but it's nice to be able to double dip, especially in something like a JAMMA cab where you don't have access to gsync or anything
Anonymous No.713576559
>>713575962
>Emulator is running at frame X
>in this time at least 1 frame of emulation has occured
>without run ahead, inputs on frame X are seen by the emulator and emulated game in frame X+1
Now here's the magic
>With run ahead on and set to Y frames, save state first
>Get input from Windows
>run several frames of logic without video and audio
>stop at frame X + Y, and run logic, video and audio
>send that to your PC's GPU for display out
>finally, load state back to frame X, process logic, move to frame X + 1
Rinse and repeat
Anonymous No.713576580 >>713576727 >>713576748 >>713576969
>>713576171
It's 2025, how do you not have a monitor/tv with freesync/gsync?
Anonymous No.713576589
>>713575794
>Less input lag than real hardware
so its inaccurate and you might as well be playing a remake
Anonymous No.713576667
>>713576143
I didn't say you don't need to debug, but you don't need a dedicated debug port. If you want to debug things you can already upload your own code to communicate over USB.
Anonymous No.713576727 >>713576836
>>713576580
My CRT doesn't have freesync faggotron
Anonymous No.713576748 >>713577146
>>713576580
Not him but I disabled freesync and literally notice no difference in either lag or tearing. I even tried it at a bunch of different framerates and analyzed it with a 240fps slow-mo camera and games with heavy horizontal and vertical scrolling as well as lag testing. Didn't notice any difference, aside from no obnoxious flickering (OLED display). VRR is a total meme, it's just high refresh rate that is important, even if you're only playing 60fps games
Anonymous No.713576836 >>713577489
>>713576727
Maybe get a TV that isn't so old it belongs in the Flintstones?
Anonymous No.713576960 >>713582017
>>713576367
>MiSTer is as accurate or more than every software emulator with identical input lag to original hardware.
No, it isn't.
>It's never recommended to use run-ahead as it increases input lag variance for slightly lower raw input lag. It's also not accurate to original hardware, has limited implementation depending on internal game lag and how intensive it is to run, and can cause various issues
The only rare edge case in which it may be worth it (and even then, not really), is games that are easy to run accurately, have significant inherent gameplay lag (5+ frames), and where reactions are more important than timing. In these case, the raw input lag improvement is so large that it could (arguably, but not really) outweigh the increase in input lag variance that it brings.
100% bullshit. There's no variance. Runahead works perfect when you use equal or lower frames to the games lag, and nearly every game has at least 1 frame of lag so you can get 1 frame free 99% of the time without issue.
>Also run-ahead is not going to remove any inherent lag from your setup or V-Sync lag, which will usually be 2-3 frames at least.
No one who cares about lag is getting 2-3 frames of lag, or using vsync.

Stop talking out your ass.
Anonymous No.713576969 >>713577067 >>713577138 >>713577146
>>713576580
tearing is standard issue regardless of freesync
Anonymous No.713577067
>>713576969
It's really not though. Tearing hasn't been a thing on PC in over a decade. Who knows what dogshit display/configuration you're using to still struggle with screen tear
Anonymous No.713577138
>>713576969
No, VRR inherently cannot tear, as tearing is caused by swapping the swapchain during scanout, but VRR synchronizes when the swap occurs with when the display goes into VBlank
Anonymous No.713577146 >>713577290 >>713577434
>>713576748
probably weren't using exclusive fullscreen and had windows compositor applying vsync automatically
>>713576969
no it's not. that's literally the whole point of it
Anonymous No.713577164
>>713575794
Everytime I see a shill vid on this fucking thing half the video is asterisks.

I'll stick to my "shitty software emulators" thanks.
Anonymous No.713577290
>>713577146
You don't need exclusive full screen for VRR these days and Windows automatically converts it to windowed fullscreen anyways
not sure about linux
Anonymous No.713577380
>>713575794
>enhancements than emulators.
You did NOT play the game
Anonymous No.713577434
>>713577146
Yeah I had V-Sync enabled, if it does add lag, it's so tiny that a multiple 240fps lag tests couldn't even pick it up (a few ms at most). V-Sync really only adds lag for 60Hz monitors with triple buffer
Anonymous No.713577469
>>713575438 (OP)
>This retard is still shilling his faggy buggy hardware.
Anonymous No.713577489 >>713578985
>>713576836
Retard
Anonymous No.713578985 >>713579084
>>713577489
yes you are :)
Anonymous No.713579084
>>713578985
le le le
Anonymous No.713581528
post cleras shitter
Anonymous No.713582017 >>713582553 >>713583020
>>713576960
>No, it isn't.
It is, at least for all the major cores. Saturn core already passed software emulators after only a single year, same with N64, GBA core is one of only a couple that passes aging cartridge test

>No one who cares about lag is getting 2-3 frames of lag, or using vsync.
Pretty much anyone gaming on a 60Hz monitor, or a chinkheld, or their phone, or PS3/Wii emulation is getting multiple additional frames of latency.

>100% bullshit. There's no variance.
Lmao, shut the fuck up retard, I have tested this extensively with hundreds of tests using a 240fps camera using multiple cores. There is 100% significantly more input lag variance when using run-ahead, and if you would see the same result if you tested it.

>Runahead works perfect when you use equal or lower frames to the games lag
no shit..
>and nearly every game has at least 1 frame of lag so you can get 1 frame free 99% of the time without issue.
The real world result is always a half-frame less (well.. more) than expected, so it would only be a half frame improvement, at the cost of significantly more variance (much more important for majority of games, unless you are playing something that is purely reaction time as opposed to timing/consistency).
Anonymous No.713582553 >>713582632
>>713582017
>GBA core is one of only a couple that passes aging cartridge test
Still worse than software.
but at least it's not Anakek
Anonymous No.713582632 >>713582832
>>713582553
That test is like almost a decade out of date you absolute dumbfuck retard
Anonymous No.713582832
>>713582632
https://github.com/mgba-emu/suite
>9 months ago
kill yourself
Anonymous No.713583020
>>713582017
Post some actual evidence of any of your claims then faggot.
Don't worry, I'll wait for you to let the thread fall to page 10 again before you post so your shill thread will stay up as long as possible.