My student worker, Alex, borrowed a digital oscilloscope and photoresistor from a coworker of mine and we sat down at my workstation to collect some data in an area that's often discussed (vociferously!) but rarely actually tested: latency. Most latency testing is unscientific voodoo ("I can *feel* it") that also suffers from confused terminology (see: the fighting game community's complaints about "lag" and how it makes them drop their combos). In this case, we're specifically examining input latency; that is, the difference in time between pressing a button on the controller and the action taking effect on the screen.
Here's a picture of our test bench, which consisted of a button from my trusty Happ-modded Mad Catz SE wired into the aforementioned oscilloscope:
The input from the button is compared against the voltage running through the photoresistor attached to a battery (and a momentary switch to keep the resistor from just draining the battery):
The photoresistor gets placed against my computer monitor while the button is used to make things happen in the emulators. As the brightness changes underneath the photoresistor, the resistance also changes changes, and the oscilloscope displays the difference in time between the voltage drop in the button and the change in voltage from the resistor/battery circuit, which looks something like this:
We only had the equipment for a day, so I couldn't test as much as I would have liked, but I tried to be as consistent as possible. To that end, we sampled 5 data points for each variable and did all of the testing on the same machine. All SNES comparisons used the Super Famicom Controller Test ROM, while the arcade comparisons used Espgaluda from Cave (in hindsight, probably not the best choice, but it's what I had on-hand). I also didn't have a good way of getting a baseline latency--I'm using a modern, crappy Dell LCD setup rather than a CRT and Windows 7 64-bit, which I chose out of both convenience and the assumption that it would be similar to a typical user's setup--so I was forced to provide data for *total system latency* rather than being able to isolate the latency caused by the individual variables. In attempting to get some sort of baseline, we held up the tester to the built-in gamepad testing applet in Windows, which gave results hovering around 75 ms, which is obviously not accurate, since some of our emulators performed better than that... With that in mind, these results should only be considered relative and not absolute.
Note: my full system info is Intel i7 Sandy Bridge with AMD R7 200 series with all of the GPU control panel crap turned off except for Eyefinity.
Anyway, here are the graphs that illustrate some of the more interesting comparisons:
First off, Aero compositing is bad news for both latency and variance. The increased variance is a real kick in the pants because it makes your performance less predictable. If you want consistent behavior and generally improved latency, stick with a "classic" non-Aero theme. Interestingly, disabling Aero did not seem to help with Higan.
Overall, this graph shows us that exclusive fullscreen is significantly better than windowed for latency, which is expected based on our Aero compositing findings. You'll notice there's no benefit to fullscreen in Higan (it's worse, in fact) because it's not *exclusive* fullscreen. Instead, it's what's known as "windowed" or "borderless" fullscreen. You can also see that ZSNES in exclusive fullscreen is extremely fast; faster than my supposed baseline of 75 ms :O
Higan had the highest latency figures here, even after correcting for the shaders--which I'll talk about more in a sec--with RetroArch about a frame lower (this includes data from both the snes9x and bsnes-compatible cores, which were not significantly different [87 ms vs 92 ms, which is within the variance of USB polling rates]). This also combines both windowed and fullscreen, which hurt ZSNES and ZMZ, the clear winners in exclusive fullscreen mode from our previous graph. Note: when ZSNES and ZMZ went into exclusive fullscreen, they broke Eyefinity, which other testing suggested adds up to ~8ms (or a half-frame) of latency, so keep that in mind when looking at their results.
This one was a dagger in my heart, but I'm posting it here anyway because of SCIENCE. I had always assumed that shaders would never increase latency because, in a worst-case scenario, they would just reduce the framerate (i.e., if the shader takes >16 ms to render). This is obviously not the case, as cgwg's crt-geom increases latency considerably in both Higan and RetroArch, as does crt-lottes. Crt-hyllian, on the other hand, has almost no effect on latency. To explore whether it's just heavy-duty math that causes the latency and whether it's exacerbated by multiple shader passes, I also tested Hyllian's xBR-lvl4-multipass in RetroArch. Shockingly, this one produced lower latency than no shader at all, which I find highly dubious.
I kept this one in here because there's been a contentious debate as to which of these platforms provides the best experience for emulating arcade games. However, there are some serious caveats to keep in mind before drawing too strong of a conclusion: 1.) this used a different test ROM from the SNES emus, 2.) the test ROM I used was selected out of convenience and actually had a lot of potentially confounding noise in the form of enemy bullets passing through my test area, 3.) GroovyMAME and RetroArch are really at their best running in KMS via Linux rather than Win64, so they would likely have more pronounced benefits vs mainline MAME if I could have measured that, and 4.) in initial testing the day before I ran these measurements, mainline MAME performed incredibly badly, with GroovyMAME close behind, which suggests that there may be some other variance involved.
That all said, these data indicate that RetroArch is approximately 1.5 frames slower than GroovyMAME, while the difference between mainline MAME and GroovyMAME is within the variance of USB polling rates. However, in light of the counfounders, I think the strongest conclusion we can draw reliably from the arcade comparison is that RetroArch isn't any *better* in Win64 (i.e., a null finding), so users should go with whichever platform has the features that best suit their needs rather than worrying about slim-to-nonexistent latency differences.
Conclusions
While the testing was not 100% reliable due to multiple confounders in several areas, we can see some trends emerge that can inform our discussions about latency in emulation. Windowed is definitely worse than fullscreen, and enabling Aero compositing is worse than without while also increasing variance and unpredictability. Shaders can actually cause excess latency, sometimes severely so. ZSNES, which has become a bit of a punching bag among SNES emulation scenesters, has outrageously low latency in fullscreen, so if you can stomach the terrible accuracy, there's actually some justification for using it now other than OMGSNOW!1! Alcaro's ZMZ also performed very well and can utilize more accurate emulation cores, so it can be a means to leverage some of ZSNES' latency benefits without being stuck with its poor accuracy.
In the future, I would like repeat these tests with a CRT monitor, which would have a predictable baseline of near 0 ms. I would also like to test latency in other environments, namely Linux+KMS. Finally, it would be very useful to have some comparative figures for original SNES hardware (both via CRT and upscaled via XRGB-Mini) and for RetroArch running via console.
Here is a link to download the raw data in Excel format, in case anyone would like to look at the numbers in more detail and/or perform other comparisons that I didn't think of.
EDIT: I think some people are drawing more conclusions from these data than is really appropriate; specifically, some folks are trying to draw direct comparison between the emulators/frontends tested. These data are simply not extensive enough for that. Furthermore, it's important to keep in mind that I didn't test the quality of sync, which could heavily affect the results. Namely, ZSNES and ZMZ both suffer from frequent audio crackling and frame stutters, which indicate issues with vsync, while RetroArch has none of either. I didn't test RA with vsync disabled (i.e., blocking on audio with video tearing), which could have an effect, and in general gameplay, users need to decide whether improvements in sync are worth minor (potential) increases in relative latency.
Showing posts with label shader. Show all posts
Showing posts with label shader. Show all posts
Thursday, June 25, 2015
Friday, June 20, 2014
True Hq2x Shader Comparison With xBR
Background
I was shocked to learn recently that the shader I and others have long called Hq2x is/was actually a misnamed port of another shader entirely! guest.r originally put out a 2.0 scale shader with a suffix "HqFilter," which stands for a part of the color blending code. Over time, this shader was ported to a million different emulators (including the official Metal Slug PC releases) and, at some point, the name got confused with another popular emulation upscaling algorithm, Hq2x. As CPU filters have been supplanted over time with GPU shaders, no one seemed to notice that no true shader port of the classic Hqnx algorithm--in any scale--existed in any language. EDIT: looks like there was an attempt to port it back in 2005 but it never caught on because it was incomplete and had some bugs, but it's something.
Fast-forward to a few weeks ago, when a shader programmer named Armada brought this up in RetroArch's IRC channel. He shared some pics of the "Hq4x" shader in the common-shaders repo (which itself was based on the old XML shader in bsnes' gitorious repo, which in turn was based on guest.r's ePSXe shader) and an identical pic taken using a CPU-filter version of Hq4x. The differences were obvious and indisputable:
I started digging through old forum posts and it turns out that several people had noticed this over the years, but no one ever posted any comparison pics or were able to backup their suspicion in any way.
Armada and I did find that there was an old DOSBox renderer called OpenGL-Hq that was at least a step in the right direction, being a hardware-accelerated implementation of the Hqnx algorithm, and it was helpful insofar as it demonstrated how the algorithm can use an external lookup texture for the detection. However, it was not particularly applicable to modern shader language, so Armada set out to port directly from the CPU filter's C implementation.
After some intense work, which included creating a program to generate the LUTs that Hqnx bases its calculations on, Armada completed his shader port (also copied into the common-shaders repo) and it works beautifully! Incidentally, the requirement for LUTs means that a true Hqnx port wasn't even possible until fairly recently, as SSNES/RetroArch is/was the first emulator (that I know of, at least) to support LUTs in shaders.
Comparison Shots
If you've read over my previous comparison of Hyllian's xBR vs Hqnx, xBR won by a landslide in pretty much every comparison, which is no surprise because it wasn't really an apples-to-apples comparison. That in mind, here are updated pics that show a true comparison between the two algorithms (2xBR shader first, Hq2x shader right after and Hq2x CPU filter third):
I was shocked to learn recently that the shader I and others have long called Hq2x is/was actually a misnamed port of another shader entirely! guest.r originally put out a 2.0 scale shader with a suffix "HqFilter," which stands for a part of the color blending code. Over time, this shader was ported to a million different emulators (including the official Metal Slug PC releases) and, at some point, the name got confused with another popular emulation upscaling algorithm, Hq2x. As CPU filters have been supplanted over time with GPU shaders, no one seemed to notice that no true shader port of the classic Hqnx algorithm--in any scale--existed in any language. EDIT: looks like there was an attempt to port it back in 2005 but it never caught on because it was incomplete and had some bugs, but it's something.
Fast-forward to a few weeks ago, when a shader programmer named Armada brought this up in RetroArch's IRC channel. He shared some pics of the "Hq4x" shader in the common-shaders repo (which itself was based on the old XML shader in bsnes' gitorious repo, which in turn was based on guest.r's ePSXe shader) and an identical pic taken using a CPU-filter version of Hq4x. The differences were obvious and indisputable:
I started digging through old forum posts and it turns out that several people had noticed this over the years, but no one ever posted any comparison pics or were able to backup their suspicion in any way.
Armada and I did find that there was an old DOSBox renderer called OpenGL-Hq that was at least a step in the right direction, being a hardware-accelerated implementation of the Hqnx algorithm, and it was helpful insofar as it demonstrated how the algorithm can use an external lookup texture for the detection. However, it was not particularly applicable to modern shader language, so Armada set out to port directly from the CPU filter's C implementation.
After some intense work, which included creating a program to generate the LUTs that Hqnx bases its calculations on, Armada completed his shader port (also copied into the common-shaders repo) and it works beautifully! Incidentally, the requirement for LUTs means that a true Hqnx port wasn't even possible until fairly recently, as SSNES/RetroArch is/was the first emulator (that I know of, at least) to support LUTs in shaders.
Comparison Shots
If you've read over my previous comparison of Hyllian's xBR vs Hqnx, xBR won by a landslide in pretty much every comparison, which is no surprise because it wasn't really an apples-to-apples comparison. That in mind, here are updated pics that show a true comparison between the two algorithms (2xBR shader first, Hq2x shader right after and Hq2x CPU filter third):
The first thing you'll notice in those Super Mario World shots is that Hq2x does a great job of killing the jaggies. Much better, in fact, than ScaleHQ from the other comparison, and almost as good as xBR. There are a couple of rough edges (Yoshi's nose is a good example), but Hq2x is also *very* fast, so reasonable tradeoffs here. Hq2x does completely ignore the light texture blobs in the ground, leaving them as hard-edged rectangles, while xBR turns them into ovals. You can also see some slight lingering differences between the Hq2x CPU filter and the Hq2x shader, where the shader actually does a significantly better job of handling various detections.
On these digitized shots from Earthworm Jim 1 and 2, though, the comparison sort of falls apart. xBR is able to spot the jaggies and smooth them out while Hq2x doesn't spot any patterns at all, due to them being outside of its LUT's detection matrix. In fact, Hqnx's inability to work on antialiased images is one of the reasons Hyllian developed xBR in the first place.
It's also worth looking at how the algorithms differ in handling text. For this comparison, I included the two extremes of xBR's corner detection, with the 'a' variant as the most rounded and the 'c' variant as the most square:
In this comparison, Hq2x is essentially indistinguishable from xBR's 'c' variant, insofar as the text is concerned. The xBR 'a' variant is of course substantially more bubbly, which may be desirable for some games.
Conclusion
My previous comparison wasn't really a fair fight, and I apologize to Mr. Steppin for misrepresenting his algorithm. This is a much better comparison of the algorithms, and in ideal conditions, Hq2x is almost identical to xBR in smoothing while running much faster. However, in other cases--particularly digitized artwork--limitations in Hq2x's pattern detection can leave some images completely unsmoothed.
The speed of Hq2x makes it attractive for certain use-cases, such as mobile, where performance is still of the utmost importance and xBR either doesn't hit full speed at all or else drains your battery. xBR, on the other hand, can handle a greater variety of images and is more likely to produce a pleasing image with the digitized artwork that became more common in the PSX era.
Monday, October 14, 2013
RetroArch Overlay Borders
RetroArch has long had support for fun borders/backgrounds that can fill the unused black space on HDTVs using a specially crafted shader combined with an image, as described in a previous post. However, this system can be finicky and it isn't very forgiving of different screen sizes and aspect ratios. There is another way to achieve a similar outcome, though, using RetroArch's built-in overlay system.
![]() |
| Gameboy border pictured with Harlequin's Gameboy shader |
This system was designed to provide onscreen touch controls for the Android and iOS ports of RetroArch, but we can use the same hooks on PC to display our borders. All you have to do is add this to the bottom of your retroarch.cfg:
input_overlay = "/path/to/overlay.cfg"
You can also load overlays directly from the in-game RGUI menu under Settings > Overlay Options.
The overlay.cfg file is extremely simple, containing only 4 short lines (you can either copy these into a blank text file yourself or use the one I include in the download at the bottom of the post):
overlays = 1
overlay0_overlay = border.png
overlay0_descs = 0
overlay0_fullscreen = true
This file tells RetroArch that there will only be 1 overlay, which image to load as the border (in this case, border.png, located in the same directory as the overlay.cfg), that there will be no buttons attached to the overlay (that is, for touchscreen controls) and that it will take up the full screen, including any black space.
In addition to these border effects, you can also use this system to apply semi-transparent overlay effects, such as scanlines, which can be useful in very low-power settings, such as on Raspberry Pis, homebrew-capable Nintendo Wiis or older cell phones that can't handle even lightweight scanline shaders. Even some MAME "rgb" effects can be used with this system with minor modification.
This system will work fine with the borders from my previous post, and here's a bunch more (formatted for 1920x1200 displays) from NeoGaf user richisawesome.
Here is the download (contains border.cfg, two example borders and an integer scale Cg shader, though the shader should not be necessary now that you can force integer scaling through RGUI):
http://www.mediafire.com/download/1fpa407519jg2hu/overlay-borders.zip
UPDATE (06/02/2014): You can also check out RetroArch's overlay repository on github.
UPDATE (06/02/2014): You can also check out RetroArch's overlay repository on github.
Tuesday, January 11, 2011
Bloom Pixel Shader for bsnes
I did some digging around online and came across some bloom filters written in GLSL. The one on this site looked the best to my eyes, so I converted it to work with bsnes and tweaked the settings a bit to achieve the desired effects.
Update (2/2/11):
UPDATE (2/28/11): I made a scanline shader that works with a 3x scale factor, as well. You can download it (and the other GLSL shaders) here.
What I found while playing around with it is:
1. Turning on 'Smooth video' seems to amplify the bloom effect.
2. The same amount of bloom looks different on each game, depending on the overall dark/light-ness of the game.
3. Combining excessive bloom with bsnes' scanline filters can make a pretty decent approximation of a CRT. While this is in no way as accurate or as nice looking as cgwg's CRT shaders, it may be an option for the many people whose video cards don't have enough horsepower to handle that complex shader.
First, I made a simple, fairly subtle (as far as bloom is concerned...) shader, which looks like this:
Looks pretty nice, eh? Unfortunately, the exact same shader settings look awful in Super Mario World:
One big flare-out... All I need is some brown and lens flare and I'll have next-gen graphics.
However, if you add in a scanline filter, things can really get evened out. In fact, you can use the different scanline intensity filters on a game-to-game basis to attenuate the inconsistency of the bloom effect. For example, here's a shot of Super Mario World with even more bloom added and 'smooth video' checked, but with 100% scanlines to chill things out a bit:
As you can see, it kinda gives it a plasticy, oversaturated look, but it's sort of charming, if you ask me.
In contrast, those exact same settings in Chrono Trigger--an altogether darker game--produce a really nice CRT-style effect, like this:
As you can see, even the bright parts don't suffer from excessive flare-out and the color bleed into the scanlines from the bloom provides an effect that is reminiscent of the phosphor glow on a CRT display.
One more shot of Chrono Trigger, this time using heavy bloom, 25% scanlines and smooth video:
I have asked WhateverMan if he would release his shader code under a permissive license. If he does, I will post my simplebloom and heavybloom versions here for download.
Update (1/19/11): WhateverMan was nice enough to allow his work to be distributed/modified under the GPL, so you can download the heavybloom shader here and the simplebloom shader here.
I also made a simple scanline shader that works with or without video smoothing:
For best results, use it along with a scale factor of 4x. Otherwise, the pixels won't line up with the scanlines correctly.
Tuesday, December 28, 2010
CRT Pixel Shader Filter for SNES Emulation
Update 05/18/2011: More screenshots for more new filters in my new post. :-)
CRT.OpenGL.shader and CRT-flat.OpenGL.shader are similar to the shader covered in this post, though they run slightly faster and have no visible artifacts. CRT-simple.OpenGL.shader is a simplified rewrite that should be usable on much older, slower machines.
Here is the original post for informational purposes:
Here is the original post for informational purposes:
This post covers the use of filters to upscale pixel art--specifically as it applies to SNES emulation--with special attention to CRT reproduction. If you just want the pictures and to download the filters, skip to the bottom of the post.
Background:
As everyone who dabbles in old-school emulation knows, artwork that was intended for a 480i CRT television that has been upscaled to an HD resolution looks like absolute garbage on an LCD monitor. The chunky sprites with their often thick, cartoony outlines just weren't designed to be reproduced with sharp edges resulting from nearest-neighbor upscaling.
To get around this ugly upscaling effect, many emulators now include upscaling interpolation filters, which apply complex mathematical algorithms to the original picture to fill in the gaps between things that are impossible to represent in chunky low-res, such as curves and smooth diagonal lines. You're probably familiar with some of the more common and popular interpolating filters, such as SuperEagle, SuperSaI and HQ2x. Unfortunately, none of these filters gets everything quite right, especially numbers and letters, which can look bubbly or overly smoothed (you can learn more about pixel art scaling algorithms here).
Purists have long been turned off by the inaccuracies of interpolating filters and have instead used scanline masks to try and capture the effect of an interlaced display, relying on the human brain's natural ability to recognize patterns and fill in the gaps between lines (you can learn everything you ever wanted to know about scanlines here). However, this too falls short from a true representation of a CRT display, as it ignores the existence of phosphors--the tiny red-, green- and blue-colored lenses that the electron gun in the back of a CRT tube shoots with a beam of electrons to recreate a colored pixel--and the color bleed that naturally occurs in these displays.
Recently, a number of determined individuals have set out to try and capture all of the different effects of a CRT display, warts and all, to truly reproduce classic pixel art the way it was meant to be viewed.
The Comparisons:
(Each of these images is presented as it would be displayed onscreen, at a resolution of approximately 800x600, then again at 400% scale without any interpolation used when scaling; as always, click the thumbnail to embiggen)
First, we should look at the baseline. This was scaled up to size using nearest-neighbor and is otherwise untouched:
Next, we'll add blargg's NTSC filter, which emulates the noise and color bleed of an NTSC video signal (this filter has several presets; I will only be showing the RGB preset, which reproduces the look of an SNES hooked up via RGB connection [not available in the U.S.], and the RF preset, which reproduces the look of the SNES RF modulator attachment, respectively):
As a note, blargg's NTSC filter is so accurate that byuu, the author of bsnes, recommends its use along with bsnes' accuracy profile to achieve proper blending on games that use halftones to simulate transparency (Jurassic Park and Kirby's Dreamland, for example).
Next up, we'll look at cgwg's CRT shader, which includes a phosphor mask and barrel distortion to simulate the screen curvature of a CRT television (just look at those RGB phophors!):
Similarly, there is a version of cgwg's CRT shader, which doesn't include the barrel distortion and represents an idealized flat CRT (actual flat CRTs tended to have slight blurring at the edges where the tube curvature would normally be). Incidentally, this version also has no visible garbage pixels (the occasional black specs that are visible in the curved version):
Pixel Shaders vs. Software Filters
cgwg's CRT shader is a special kind of filter known as a pixel shader. Unlike regular filters, which rely on the CPU to do all of the complex upscaling calculations, pixel shaders draw on the awesome computing power of the video card to do the calculations, thereby leaving the CPU to focus on emulating the SNES. Additionally, since the pixel shader is calculated separately from the filter in bsnes, you can stack blargg's NTSC filter with cgwg's CRT shader:
Finally, for non-purists, we'll look at the combination of cgwg's CRT shader with the popular SuperSaI filter, which creates a pleasing--though not quite as accurate--output:
As amazing as cgwg's CRT shader is already, there is still some room for improvement. For example, the current implementation misses the intensity-based bloom effect on individual phosphors that can be seen in a true CRT. DOLLS (J) [!], one of the contributors to the CRT reproduction effort, intends to write a more complete CRT emulation shader in the future that will incorporate these and other idiosyncracies.
Click here to download cgwg's CRT shader (Also includes the 'flat' version for those who don't like the tube-style curvature; UPDATE: fixed dead link), which is compatible with bsnes and the newest release of snes9x. For more information on CRT emulation, you can check out this highly informative thread on the bsnes forum. For other bsnes-compatible shaders that are not included with the official download, check out my mediafire account.
UPDATE (3/4/2011): Themaister did a rewrite of the flat version of cgwg's CRT shader, moving many of the calculations from fragment to vertex, which provides a substantial ~20% increase in speed (making it usable on many older and less powerful video cards). This rewrite also appears to conform more rigidly to the GLSL shader spec, making it compatible with more cards from different vendors. I have labeled it v4 of cgwg's CRT Flat, and it is available in the aforemented mediafire account.
UPDATE (3/4/2011): Themaister did a rewrite of the flat version of cgwg's CRT shader, moving many of the calculations from fragment to vertex, which provides a substantial ~20% increase in speed (making it usable on many older and less powerful video cards). This rewrite also appears to conform more rigidly to the GLSL shader spec, making it compatible with more cards from different vendors. I have labeled it v4 of cgwg's CRT Flat, and it is available in the aforemented mediafire account.
Subscribe to:
Posts (Atom)
-
Oh gila betul manusia zaman sekarang. Tambah pula polis tu yang langgan pelacor?
-
UPDATE : the comparisons in this post aren't a true representation of hq2x ( more info in this post ). The xBR shots are still valid, bu...
-
Cerita ini di sampaikan oleh seorang bekas banduan tentang pengalaman beliau melihat sendiri ketua pembangkang Anwar Ibrahim di dalam penjar...



















































